OCTOBER, 1960

Welaing

Journal

IN THIS ISSUE

FFICIAL PUBLICATION OF THE AMERICAN WELDING society. ©»

i aon / - - ~— . #
 ‘kv

3
LV.)
SSPAS
SR

ry

your most

With Victor’s wide variety of nozzles, tips, and cutting

flexible tools
attachments, and reliable, field-proven regulators,
you can handle all your jobs with ease.

See your Victor dealer’s complete stock of welding

VICTOR

and cutting units for industrial uses ... or if you’re

looking for economy, Victor’s Super- Range unit
Combination (only $99.00, including accessories) is your best buy.
Helpful literature yours for the asking. Call
Welding and your Victor dealer today.

Cutting Units
VicIOR EQUIPMENT COMPANY
844 Folsom St 3821 Santa Fe Avenue
San Francisco 7 Los Angeles 58
1145 E. 76th St., Chicago 19
J. C. Menzies & Co., Wholly-Owned Subsidiary
\VIETOR
for welding
and cutting
70
MFRS. OF HIGH PRESSURE AND LARGE VOLUME GAS REGULATORS; WELDING & CUTTING EQUIPMENT; HARDFACING RODS; BLASTING
NOZZLES; COBALT & TUNGSTEN CASTINGS; STRAIGHT-LINE AND SHAPE CUTTING MACHINES; ROLLER AND IDLER REBUILDING MACHINES
For details, circle No. 1 on Reader Information Card
 alg
am

al
Journ

Technical Papers Designing for Production Welding Gives Industry a Fully Accessible Motor, by A. L. Cooper and W. H. Morse
Recent Developments in Oxy-Fuel-Gas Cutting, by C.C. Anthes.............cceesceeeceees ‘
Items and Machine Welding of a Prepackaged Liquid Rocket Engine, by Ralph L. Hoetger and Walter B. Moen
Reports Stud Welding with a Silicon-rectifier Power Source, by S.Baum
Short-arc Consumable-electrode Welding Applications and Developments, by T. McElrath.

Practical Welder Welded “epithe Center Features Space-saving Economical Design heS. C. Bast, P. Dreier and C .

and Designer

Society Press-Time News. st News of the Industry....

Welding Zones..... . 1006 Personnel
and Related World-Wide Welding News... .-» 1008 Employment Service Bulletin...
Fvents Editorial — Foreign Competition, by Carleton Current Welding Literature
Shugg..
Abstracts of Current Patents
Author's Application Form—1961 National Fall
Meeting. . New Literature...
Society News.. New Products. .
Section News and Events.... . 107 Reader Information Card. .
New Members... 107 Index to Advertisers.

j
Welding Strength
g of Welded
» Aluminum-alloy
n y Box Beams, by y R. J. Brungraber................eeeeees
ungra
Comparative Properties of Aluminum-alloy Weldments, by |. L. Stern, H. V. Cordiano and V. A. DiGiglio. .
Research All-position Welding of HY-80 Steel with the Gas-shielded Process, by C. R. Sibley
Weld Strength and Dimensional Stability of Cold-worked Stainless Steel, by L. Stemann and E. E
Supplement Weismante! ; . . Rea ;
Studies of Methods for Sealing Ends of Reactor Fuel Rods for PWR, by J. J. Vagi and D. C. Martin
Volatilization Phenomena in High-temperature Brazing Filler Alloys, by William Lehrer and Harry
Schwartzbart ‘ rece r ee F is desia abba eae
Proposed Procedure for Testing Shear Strength of Brazed Joints,..
Discussion by W. Lehrer and H. Schwartzbart..
Authors’ Closure

Published for the advancement i i a a as

. Editorial and general offices, 33 West 39th St., New York 18, N.Y. Subst
‘
of the science and art of welding ee nited State 1
ne ee eepossession oreig ee, ntries $10.00.
ie mien te aeme O
Single copies,
ra r 132.122 Copyr right 1960, by the American Weldir
pressed in its publications e
by the American Welding Society ae Gels char tisan of catalan oles Gee a
 interest is divided. These working
PRESS-IIME technical commissions operate on
a continuous basis and are purposely
designed to promote teamwork at
an international level. This policy
is distinct from that of scientific
groups which offer primarily an
annual meeting devoted to the
NEWS
presentation of technical papers.
Each country participating in the
IIW may have one official delegate
... People on a commission. The delegate
may be assisted by experts. A
... Welding chairman is designated who is
responsible for leading his group
and who reports to the several staff
... Products
and ruling bodies.
Membership in the Institute is
not accorded to individuals as
such; rather, individual welding
societies may join and several such
IIW to Hold 1961 Assembly in New York groups from a country may handle
IIW affairs through a_ central
For the first time since 1948, rostrum of AWS technical sessions agency. In the United States
when the group was organized, affords an unprecedented opportu- the agency is the American Council
the Annual Assembly of the In- nity for members to hear the latest of the IIW which represents the
ternational Institute of Welding developments in the field of welding AMERICAN WELDING Society, The
will leave the shores of Europe and outside the United States. The Welding Research Council of the
travel to America. Here, in New men who will present their lectures Engineering Foundation, and the
York City next April, the AMERI- are outstanding figures, in Europe Ship Structures Committee of the
CAN WELDING SOCIETY, supported and Russia, who have contributed National Research Council.
by the American and Canadian heavily to the advancement of The governing body of the I1W
Councils of the IITW, will act as welding in their own _ countries is called the Governing Council in
official hosts to the international and to the _ success of _ the which each member country is
organization. The Assembly will IIW. The arrangement which represented by three delegates and
be held on April 11-19, 1961, at brings them to these shores and has a single vote. This council
the Sheraton Atlantic Hotel. to the AWS Annual Meeting is in functions twice a year —at the begin-
The main purpose of the Annual the highest interest of the develop- ning and end of the Annual Assem-
Assembly of the IIW is to review ment and promotion of welding. bly. During this time an Executive
and pass upon the work of the past Council is appointed to control In-
year, to elect new officers and to set About the ||W stitute affairs for the interim be-
the program of work for the follow- Organized along somewhat dif- tween Assemblies.
ing year. A public session which ferent lines from many international On the administrative side there
is usually held for the purpose of scientific groups, the International is the Secretariat consisting of the
presenting selected technical papers Institute of Welding has for its General Secretariat, which is pri-
will not take place at the next objective “‘to promote the develop- marily concerned with membership
meeting. Instead, a group of ex- ment of welding by all processes.” and finance and is located in Lon-
perts from various national delega- To attain this objective, there were don, England; and a Scientific and
tions will appear at sessions of the formed a number of technical com- Technical Secretariat, which, among
AWS Annual Meeting that is to missions that constitute the main other duties, coordinates the work of
be held on April 17-21 at the Com- productive core around which the the commissions among themselves
modore Hotel in New York City. organization is built. There are at and the work of the IIW with other
The appearance of distinguished present 15 such commissions among groups. The latter Secretariat is
European welding experts on the which the work in the present field of located in Paris, France.

GEweea
seceevamat _ GOVERNING COUNCIL — exec TIVE COUNCIL VICE CHAIRMAN Tee +=Seenerany

I i 1
AMERICAN COUNCIL MATIONAL COUNCIL AMERICANJOCIETYWELDING SHCOMMITTEE
STRUCTURE WELOINGCoumcit
RESEARCH

! |
1S TECHAICAL COMMISSIONS 1S OELEGATES
TECHNICAL &COMMISSIONS
EXPERTS

Charts showing organization of the IIW proper (left) and of the American Council (right)

1004 | OCTOBER 1960

 having
AVil seven welders in one!

With this new HOBART ADI model, you can weld

ALUMINUM, STAINLESS STEEL, COPPER,

ZIRCALLOY, TITANIUM, and other non-ferrous

metals as well as mild and low alloy steels

Meet the demands 1. AC Inert Gas Manual Welding

of every job
. weld with the right 2. AC Inert Gas Automatic Welding

a ae proces 3. DC Inert Gas Manual Welding

4. DC Inert Gas Automatic Welding

5. DC Inert Gas Spot Welding

6. DC Metal Arc Welding

7. AC Metal Arc Welding

COMPARE the extra features at no extra cost.

| Other make welders

FEATURES HOBART; #1 if3 #3 *4

Interior protected from valve NO

leak wetting

Water strainer included NTe) iNTe)

Type stabilizer (standard) NONE GUAR NONE _

duty =P AC-Tig welding 60 35-50* BS)

Separate 4g torch” terminal XTe) NO | NO

Pneumatic timers Te)

Current overload protection STD.

Gas & water pilot light STD. ixTe)

Remote hand rheostat STD. Sa; meeer me STD. Ea asc.

*None standard—Battery extra price. Duty cycle is 35% without battery,

50% with battery.
The versatility of this unit and the many advantages by comparison, make
it one of the biggest welder values on the market today. HOBART BROTHERS
COMPANY, BOX W4J-100, Troy, Ohio, Ph. FE 2-1223. ‘Manufacturers of the
world’s most complete line of arc welding equipment

HOBART BROTHERS COMPANY, BOX WJ-100, TROY, OHIO

] Would like complete catalog describing all mode! welders
— Send information on complete ADI model welder
C1) Same AC/DC unit without inert gas facilities
Name__ = iinet
Address___ = —_ ————— cnitiinanicaimanail
— —— Zone State___

For details, circle No. 2 on Reader information Card

 WELDING ZONES

. -
Three of 12 stainless-steel storable propellant storage vessels welded by Standard Stee! Corp.,
at its Los Angeles, Calif., plant for Titan !| missile program

Bank of 300 silver-soldered condensers yields a

surge of 100 billion watts to imitate the atomic-
fusion energy of the sun. Welding ground clamps a
short the units when idle. A project of Texas
Atomic Energy Research Foundation. (Courtesy
Lincoln Electric Co.)

All-welded heavy-duty 20-ton capacity power-rotated jib

crane and hoist handles heavy freight on rail siding.
(Courtesy R. G. LeTourneau, Inc.)

*
=>_ —
\

rt
re ~~
Te,
tbae
a,

2.
7

ie
aaEee
 Every Hour M. S. Little Brass Goods Company

Makes 650 Appliance Fittings Better

With HANDY & HARMAN SILVER BRAZING

Rotating jig showing mounted assemblies entering and leaving gas-air furnace.
This Hartford, Connecticut, company makes—in volume—an assembly that goes into
the overflow system of household appliances. The assembly consists of brass tubing and
a machined brass casting. The two components are joined by a preplaced ring of
Handy & Harman Easy-FLo 45 silver brazing alloy and HANDy FLux. Heating is auto-
matic gas-air; parts are placed on a rotating turntable to pass through the gas furnace.
every 60 working minutes, 650 assemblies are completed.
The advantages here are that the manufacturer can use thin-walled tubing with
heavier, threaded components at no sacrifice in strength. Because of Easy-FLo’s
penetrating qualities, the entire shear area is fully as strong as the solid parts of the
assembly, yet considerably lighter. And, casting and machining the components for this
assembly have been greatly simplified.
“ Are you in pursuit of a metal-joining method which offers—among other advantages
—high, uninterrupted production at low capital investment? You may easily find the
yy answer in Handy & Harman silver brazing. Hundreds of manufacturers and fabricators
of as many different products, parts and components are right now enjoying the speed,
economy, strength and flawlessness of brazing. You can too. Just ask Handy & Harman,
Left— brazed assembly. Right— components 82 Fulton Street, New York 38, N. Y.
with preform ring of Easy-FLo 45,
COMING IN NOVEMBER!
New Handy & Harman Brazing Correspondence Course. Simple self-study data on all
phases of brazing. Send for details to Dept. BC.
FOR A GOOD START: Your No.1! Source of Supply and Authority on Brazing Alloys Offices and Plants
Bridgeport, Conn
BULLETIN 20 a4 | B&B Chicago, Ill.
This informative booklet gives a Dallas Texae
good picture of silver brazing and = Deira, Mich ;
its benefits ... includes details on Y New Yok NY.
alloys, heating methods, joint de- : HAN D & HARMAN Providence, R , ™
sign and production techniques. 4 General Offices: 82 Fulton St., New York 38, N. Y. aie teas .
Write for your copy. DISTRIBUTORS IN PRINCIPAL CITIES Toronto, Canada
For details, circle No. 3 on Reader information Card
WELDING JOURNAL | 1007
 WORLD-WIDE WELDING NEWS

By Gerard E. Claussen

AUSTRIA axle. Most of the equipment is for tained four spot welds or rivets in a
d-c welding. width of 2 in. The specimen was
A patented boron-treated bronze, 3. Eight pages are devoted to patterned after a highly stressed
described in the September 1959 safety, particularly fire protection, section of the Caravelle. The fa-
issue of the Austrian welding maga- compressed gases and repair of tigue strength based on full sheet
zine Schweisstechnik, contains 12-— torches. cross section was, at 10° cycles,
14% Sn, 0.5% Mn max, 0.3% Si 4. Work at the Magdeburg Weld- 31,300 and 11,200 psi for the welded
max, 0.05°% P max, 0.02% S max, re- ing Institute showed that it was specimens and 27,000 and 9,700
mainder copper. It is used as bare important to secure laminar flow psi for the riveted specimens.
wire for inert-gas welding, metal of CO, through the nozzle of the In November 1959 a group of six
spraying, submerged-arc welding and welding gun. For this reason the French welding engineers visited
oxyacetylene welding, and as core Russian automobile factories, ship-
CO. inlet should be 4 in. above
wire for covered electrodes. Welds the exit of the nozzle. The con- yards, tube mills and welding in-
have unusually high resistance to vergent nozzle constricts the gas stitutes under a _ Franco-Russian
superheated steam and acid mine stream to a diameter of 0.67 in. cultural and scientific exchange.
water, and exhibit a tensile strength In December five Russian welding
at the exit. With » to 17/%
of 36,000 to 50,000 psi. in. nozzle-to-work distance and a engineers visited France. The Rus-
EAST GERMANY CO, flow rate of 17 cfh, the nozzle sian group included Prof. Rykalin;
can be advanced at a reasonable Mr. Kotchanovski, Scientific Vice-
The East German welding mag- President of the Leningrad Welding
travel speed without causing atmos-
azine Schweisstechnik for January pheric contamination. Wind veloc- Institute; Mr. Tarkov, electrode
1960 contains the following ar- expert; Mr. Dudko, Engineer at
ities of 4.5 mph or more must be
ticles: the Paton Institute of Welding in
avoided.
1. The role of CO evolution in Kiev; and Mr. Baranov, an auto-
5. In September 1950, the Czech
affecting transfer of molten globules matic welding expert. Prof.
Welding Institute in Bratislava
from bare steel electrodes */;, in. Rykalin spent two days lecturing
celebrated its tenth anniversary.
diam at 160 amp was studied. at meetings of the French Society
From the papers presented on the
High-speed color motion pictures of Welding Engineers.
occasion the following interesting
750 frames per second) of metal
points were taken.
transfer from 0.93% C steel were
(a) Condensers in parallel with NETHERLANDS
made in argon and air. In argon
the rectifier power supply some-
a few gas explosions were observed A detailed description of the
times are advantageous in CO,
in the globules. In air the globules manual welding of a supersonic
welding.
rapidly expanded and burst. The wind tunnel for the Netherlands
6b) In 16Cr-—13Ni _heat-resist-
gas causing internal inflation of the Aircraft Laboratory is given in
ing steel, 3° tungsten causes the
globule was CO. It was evolved the February 1960 issue of Lastech-
appearance of a special carbide,
in the globule close to the elec- niek. The tunnel consists of 62
called W phase, which improves
trode-globule junction. Chemical cylindrical sections 9 to 39 ft
weldability and is stabilized by the
analysis of gas evolved from steel diam, ' .- to 1' ,-in. wall. All im-
addition of 0.004% boron.
electrodes deposited under water portant welds were radiographed,
(c) A semiautomatic gun for
revealed some CO, which con- less than 4° being defective. All
argon welding aluminum was de-
firmed the presence of CO in the vertical welding was done with
veloped for use with a generator
globules. By bursting the globules, | .-in. electrodes at 135 amp.
having a slightly drooping char-
CO increased the number of globules
acteristic. Current densities from
per second from 1 in argon to 10 RUSSIA
30,000 to 75,000 amp per square
in air.
in. of electrode cross section are The December 1959 issue of
2. Components of equipment for
submerged-arc welding are dis- used. Avtomaticheskaya Svarka, the
cussed by two engineers of the Russian automatic welding mag-
FRANCE azine, contains the following ar-
Central Welding Institute. Four
types of current pickup are used. The French Welding Institute’s ticles:
Brass contact tubes wear rapidly, magazine Soudage et Techniques 1. Low dilution in surfacing high-
but a pair of brass contact rollers Connexes for November-December pressure steam valves with Type
provides three months of two- 1959 contains an article that com- 430 stainless and stellite by sub-
shift operation without replace- pares the tensile fatigue strength merged-arc welding is achieved by
ment. The live axle of the travel of spot-welded 0.032-in. Alclad dur- feeding two electrodes into the
carriage is hollow; the wheel axle alumin specimens with similar riv- same arc zone. As the speed of
fits inside and is keyed to the live eted specimens. Two thicknesses feeding the second cold electrode
of Alclad duralumin were joined was increased, dilution decreased.
Dr. GERARD E. CLAUSEN is associated with For example, with *’,,-in. electrodes
Arcrods Corporation, Sparrows Point, Md. to the through sheet which con-

1008 | OCTOBER 1960

 The correct selection of flux can offer unexpected Silvaloy fluxes are packaged in 65-lb. and
help in speeding and simplifying production, mini- 30-Ib. drums, 5-lb. wide mouth jars (5 to a
mizing rejects and lowering costs in low temperature carton), I-lb. and '2-lb. jars. The wide
silver brazing operations. The advantages to be opening of the 5-lb. package makes it a
gained by “selective fluxing” are sufficiently impor- most practical, time saving dispenser that
tant to warrant careful, thorough study! also enables the operator to make use of
Silvaloy offers the most advanced flux develop- every bit of flux in the jar.
ments in this specialized field. Here, is a complete
line of fluxes . . . each providing outstanding per-
formance, enabling you to select the correct flux for
every possible low temperature brazing operation.
The extra efficiency of Silvaloy “Selective Fluxing”
is being proved daily on the brazing production lines
of the country’s leading manufacturers.
Call the Silvaloy distributor in your area for con-
sultation and detailed information or, send for our
booklet “A Complete Guide to Selective Fluxing for
Low Temperature Silver Brazing.”

SEND FOR THIS COMPLETE GUIDE TO FLUXING e

1 N > = Ss eS rm 8 @ Ss, ! N c.
AMERICAN PLATINUM & SILVER DIVISION
231 NEW JERSEY RAILROAD AVE. + NEWARK 5S, NEW JERSEY

SALES OFFICES: SAN FRANCISCO * LOS ANGELES * CHICAGO» NEW YORK» MIAMI ORLANDO » DALLAS

Sc TRIBUTORS' a2.8.C. METALS CORPORATION * DENVER * AUSTIN-HASTINGS COMPANY, INC. * CAMBRIDGE * WORCESTER+ HARTFORD. BURDETT
OXYGEN COMPANY CLEVELANO INCINNAT COLUMBUS AKRON © DAYTON YOUNGSTOWN MANSFIELD © FINDLAY DELTA OXYGEN COMPANY. INC,
MEMPHIS EAGLE METALS COMPANY SEATTLE PORTLAND SPOKANE NOTTINGHAM STEEL & ALUMINUM DIV. A. M. CASTLE & COMPANY CLEVELANO
OLIVER H. VAN HORN CO inc NEW OP LEANS FORT WORTH “OC STON PACIFIC METALS COMPANY LTD SAN FRAN sco SALT LAKE CiTy
LOS ANGELES SAN DIEGO PHOENIX STEEL SALES CORPORATION ~ acc MINNEAPOLIS * INDIANAPOLIS KANSAS CITY GRAND RAPIOS
DETROIT st. Lours MILWAUKEE LICENSED CANADIAN MANUFACTURER ENGELHARD INOUSTRIES OF CANADA, LTO. TORONTO MONTREAL
For details, circle No. 4 on Reader information Card
WELDING JOURNAL | 1009
 at 540 amp, 38 v, dilution decreased 4% tin, 3% zine, provided the The repair of manganese bronze
from 40 to 9% as the speed of lead content of the base metal and aluminum bronze by metal
feeding the dead electrode increased did not exceed 3%. inert-gas welding is described
from 0 to 30 ipm. 7. The conditions are analyzed in the January 1960 issue of Svetsen,
2. A three-phase system for re- for supercritical flow of oxygen from the magazine of the Swedish Weld-
sistance butt welding railroad rails the orifice of a cutting torch. A ing Society. Good results were
utilizes 10- and 50-cycle power at graph shows that beyond a pres- secured with local preheat at 570
a power factor of 0.7 to 0.8. The sure of 27 psi, the density of the F and '/,y-in. aluminum bronze
welding time was lower for the 10- oxygen is directly proportional to electrode. The weld metal had a
than for the 50-cycle power. pressure. Steel masses 5 ft thick tensile strength of 72,000 to 79,000
3. A mathematical analysis re- were cut at an oxygen pressure of psi.
sults in an equation relating the 21 to 29 psi. The penetration of a weld bead
angle of ignition of an ignitron in 8. The building up of worn steel beneath the surface of the plate
a spot-welding circuit to the ef- rolling-mill rolls 10 to 50 in. diam by is expressed by Gunnert as p
fective welding current and voltage. submerged-arc welding is described. yVI‘/VE*, where p is_penetra-
4. Empirical formulas for spot Flux AN 20 was used with an elec- tion, mm, / = current in amp, V
welding mild steel show that the trode containing 0.35 C, 2.5 Cr, travel speed, cm/min, E = arc
diameter in mm of the effective 8 W, 04 V. The deposit was voltage. The constant y is 0.04
working area of the electrode is 460 to 600 Brinell. for submerged-arc welding and 0.055
5.5T, where T = sheet thick- 9. A condenser-discharge weld- for CO, welding of steel. The tests
ness in mm. The compressive ing machine for watchmaking is on which the CO, constant is based
force in kg/mm? on the electrode described. included travel speeds from 15 to
is 2007. The relation between 10. An instrument has _ been 45 cm/min (6 to 18 ipm), 280 to
current / in amperes and welding devised for the nondestructive 360 amp, 27 to 35 v, with ', «-
time S in seconds is /*°V7S/10° = metallographic examination of in. electrode.
22T? + 10. welds. The Swedish Welding Society
5. The effect of potassium in had 2623 members on Jan. 1, 1960,
increasing the stability of a steel SWEDEN distributed among 12 local sec-
welding arc in alternating current The Swedish magazine Svetsaren tions.
is expressed mathematically in terms No. 3, 1959) describes the welding
of the decrease in temperature of the of traveling cranes of lattice-truss SWITZERLAND
arc upon switching off the current. and plate-girder design. Rotating The first of a series of articles
A small amount of potassium in positioners and radiographic con- “From the Designer’s Portfolio”
the arc lowers the arc temperature, trol were used. is featured in the November 1959
but greatly reduces the rate of cool- The “Delta” bridge, a new de- issue of the Swiss magazine Oe6cr-
ing of the arc during passage through parture in truss bridge design for likon Schweiss-Mitteilungen. Sam-
the zero-current part of the a-c highways, is described in another ple calculations are given for six
cycle. The melting rate of the article. A cross section of the pressure-vessel designs: (1) shell,
electrode was not changed by the bridge has the shape of the Greek 2) shell with unreinforced holes,
addition of potassium. letter delta. All members, in- (3) thick-walled shell, 5'/> in. thick,
6. Leaded brasses were sub- cluding the single top chord, are (4) shell with external pressure,
merged-arc welded without cracks of welded lattice design. The de- 5) jacketed shell, and (6) checking
or porosity with AN 20 flux and a sign is patented by Hans Schroder, for yield under triaxial stress.
copper-alloy electrode containing AB, Lidingo, Sweden. Another article discusses the
straightening of welded structures.
The relatively small distortions in
JAPAN the tension zones of welded struc-
tures do no harm and should not
be straightened. If straightening
is necessary in compressive zones,
as in web plates, the section should
be heated uniformly just below
Ac, and straightened mechanically.
Internal stresses created by straight-
ening may do more harm than
good.
WEST GERMANY
The basic philosophy of eco-
nomical manufacture of welded
pressure vessels, according to a
writer in the West German maga-
zine Schweissen und Schneiden for
November 1959, is ‘“‘Not as good as
possible but as poor as allowable.”
This principle is illustrated with the
aid of probability curves of bursting
pressures. The shape of _ these
curves is shown to depend on weld
defects and other factors not capable
of inclusion in the usual design for-
Welded construction of Kawasaki power plant. (Courtesy IIW.) mulas.

1010 | OCTOBER 1960

 =
Light and easily carried, requiring no outside
power supply, the Model 30 uses iridium 192. —
it can take internal and panoramic exposures not
possible with standard equipment.

Remote control of iridium... with —,-44 Iriditron units

® Especially designed for remote handling of iridium much as 30 curies of iridium 192, approximately
192 equivalent to a 260 KV X-Ray machine. This adds
e@ Excellent for panoramic and internal exposure up to significant reductions in exposure times.
© Permits exposure of iridium sources by remote con-
trol as far as 50 ft. away. And you get complete service from Budd— including
radioactive source supply and encapsulation, source
oa™ - Skid mounted or portable, this Budd replacement and disposal, training for your person-
) Iriditron unit provides all the versatility nel (at no charge) and aid in setting up complete
of the traditional unshielded source for radioactive facilities.
pipe weld radiography, panoramic ex-
posure of multiple specimens, circumferential and Write or call Budd Instruments Division for our
longitudinal welds in boilers and pressure vessels. Gamma Radiography Bulletin . . . or for a consulta-
The unit can hold extremely strong sources... as tion on your requirements.

INSTRUMENTS aD iA yA

MA DEMEMEM oision
THE BUDD COMPANY - P.O. Box 245 + Phoenixville, Pa.
Consult your phone hook for sales offices in: Atlanta, Ga., Oak Park, Ill., Dallas, Tex.,
Los Angeles, Calif.
In Canada
Budd Instruments, Ltd., 170 Donway West, Don Mills, Ont.
For details, circle No. 5 on Reader Information Card
WELDING JOURNAL | 1011
EASY-TO-USE Inco-Hard “1” Electrode proved an inexpensive way to
extend the service life of mild steel box used for blast cleaning. No
special welding techniques needed.

Inco-Hard “1”

overlays extend

service life of

blast cleaning box

The surfaces of this steel box are exposed good for long, trouble-free service. excellent control over bead contour and
to the highly abrasive action of blasted It’s easy to see why. As applied to size ...and the deposit is usually smooth
steel shot for 5 hours a day in the clean- steel, a one-layer deposit of Inco-Hard enough to let you skip grinding.
ing of castings and forgings. “1” gives a 500 Brinell surface hardness; Want the full story? Write for our 8-
The steel cover-plate was wearing two layers give 600-700 Brinell. page booklet, “New Inco-Hard “1” Elec-
out in 6 or 8 weeks, the liner in 2 weeks. You can do it, too. There are no spe- trode for Hard-Surfacing . . .” It’s free
Overlay welds made with ordinary mild cial techniques to learn. With either AC on request. *! trademark
steel welding rod didn’t extend service or DC, you get a stable, spray-type arc
life any measurable amount. in all positions. There’s a minimum of HUNTINGTON ALLOY PRODUCTS DIVISION
spatter ... excellent “wash”... good slag The International Nickel Company, Inc.
But now things are different. Since removal. Deposition rate is high, with Huntington 17, West Virginia
Smith System Manufacturing Co., Min-
neapolis, Minnesota switched to Inco-
HKard* “1” Electrode, the box has gone 4N~.
10 weeks—with so little wear on the INCO WELDING PRODUCTS
overlays that no repair or rebuilding
has been needed, and appears to be TRADE MARK electrodes * wires * fluxes
For details, circle No. 6 on Reader Information Card
1012 | OCTOBER 1960
 Foreign Competition

Fifteen years after the close of a war which was standards of living, are far lower. It is time
won largely by American control of the seas, the for the American welding industry to take
American shipbuilding industry, whose superb a hard look at its present practices. The answer
production effort was in large part responsible for lies in increased productivity through the de-
our gaining that control, finds itself priced velopment of new engineering and welding pro-
out of the world market. An industry that cedures. Mechanization of production welding is
turned out tankers, liberty and victory ships, imperative, and this must be preceded by in-
landing craft and combatant vessels by the creased research and development in metal join-
hundreds is reduced to the production of specialty ing techniques. Many production techniques,
items-—highly complex Naval vessels and prestige loosely called “‘automation”’ in other industries,
passenger liners--and what little merchant ton- can and should be applied to shipbuilding.
nage that can be sustained by a meager subsidy. To accomplish all of this will require a con-
Despite the advantages of a 10-year competition- certed effort on the part of the design engineer,
free crack at world markets hungry for our prod- the welding engineer, the research scientist, the
ucts, we have been unable to maintain leadership. educator, industry and government.
American ship operators are building tankers, The Navy’s Bureau of Ships set a good ex-
and other merchant ships on the building ways ample, recently, in the formation of the Sub-
of Europe and Japan. marine Structural Advisory Panel, to help im-
One of the more important elements in high prove construction practices in submarines con-
shipbuilding costs is the cost of welding. The structed of HY-80 steel. More action of this
plain fact is that our foreign competitors today kind, on an across-the-board scale, is essential if
rival us in know-how, and in_ productivity, we are to maintain a healthy shipbuilding indus-
while their wage rates, geared to lower try.

Carleton Shugg

PRESIDENT
ELECTRIC BOAT DIVISION

Welding

Journal
 Model M-295 a-c welder

Model H-3-SR d-c welder

Wide range Miller M-295 a-c welder packs a lot Star SRH) but is less weather-resistant. If your
of versatility in its compact 25” x 19” x 28” case. d-c welding is done under a roof, then you'll
What’s more, it’s a genuinely low-cost 40 volt want to know more about the Miller Silver Star,
200 ampere industrial welder that doesn’t pre- and the economies it offers. Two sizes avail-
tend to be all things to all men. Rather, it offers able: H-3-SR with amperage ranges of 40-295
typical Miller sturdiness and reliability in a 60% and 75-450; and H-4-SR with ratings of 50-350
duty cycle machine that’s ideal for general main- and 100-600.
tenance and light production welding. Welding
amperage ranges are 25-115 and 80-295. Power Complete information will be sent promptly upon re-
factor correction available as option. quest. Please specify model in which you're interested.

SILVER STAR MODEL

Outsized performance has been built into this miller
“inside” welder. Designed expressly to deliver ELECTRIC MANUFACTURING COMPANY
the best in d-c welding at the lowest possible APPLETON, WISCONSIN
cost, the brand new Miller Silver Star performs
on a par with the world’s best (Miller’s Gold EXPORT OFFICE: 250 West 57th St., New York 19, N.Y.
Distributed in Canada by Canadian Liquid Air Co., Ltd., Montreal
For details, circle No. 7 on Reader information Card
1014 | OCTOBER 1960
 Emphasis on welding—fully accessible motor components

Designing for

Production Welding

Gives Industry a Fully Accessible Motor

that simplifies inspection and maintenance, has interchangeable parts, operates at

reduced noise levels and is produced by production-line techniques

BY A. L. COOPER AND W. H. MORSE

ABSTRACT. A new concept of welding design has produced design were drawn from experience developed in
large electric motors which are fully accessible for inspec- producing large electric motors hereafter referred
tion and maintenance, have interchangeable parts, oper-
ate at reduced noise levels and are produced by produc- to as conventional machines. While conventional
tion-line techniques. Utilization of weld assembly machines represent a design that utilizes weld
fixtures with manual metal-arc, automatic submerged- fabrication and a minimum of material, they pos-
arc and inert-gas-shielded arc processes to produce the sessed disadvantages in manufacturing and servic-
motor components is described. Production-line tech-
niques for producing stator cores without machining were ing after installation. With this experience, the
made possible through application of a new type welded F,/A motor objectives were established as follows:
locator joint coupled with functionalized welded-motor 1) shall be capable of fixtured weld manufacture;
components. 2) shall provide interchangeable parts; (3) shall
Introduction minimize replacement spares for the user customer;
4) shall permit quick disassembly and reassembly
In 1957, a design concept for a fully accessible motor without requiring realignment of the parts; (5) shall
(F/A) was developed for 200- to 7000-hp induction operate at reduced noise levels; (6) shall be a more
and synchronous electric motors. Objectives of the readily manufacturable design adaptable to shorter
delivery dates; and (7) shall possess improved ap-
A. L. COOPER is Supervisor, Mfg. Engineering and W. H. MORSE is
Manager, AC Motor & Generator Engineering, L. R. A. Dept., Westing pearance. These objectives were applied to the
house Electric Corp., East Pittsburgh, Pa drip-proof, splash-proof and the weather-protected
Paper presented at AWS 4ist Annual Meeting held in Los Angeles
Calif., Apr. 25-29, 1960 motors more commonly found in the conventional-

WELDING JOURNAL | 1015

 WEATHER
PROTECTED

SPLASH
PROOF

Three types of fully accessible machines in final assembly

type machines. Figure 1 shows three of the types

of fully accessible machines that evolved from the
engineering and manufacturing program.
The FA motor replaces the standard conventional
cylinder-type drip- and splash-proof motors such
as shown in Fig. 2. The conventional machine is of
a weld-fabricated design; however, it does not adapt
to customer and manufacturing requirements as
previously outlined. First, it was necessary to
fabricate a frame before any other work could be
done. The next major operation was stacking the
electrical sheet stator core inside of the fabricated
frame (see Fig. 3, left). It is apparent that the
stacking operation is beset with handling and ac-
cessibility problems. Obtaining uniform rotor-to-
stator air gap required accurate frame machining
after stacking (see Fig. 4).
Fig. 2—Conventional fabricated cylinder-type drip-proof Stator-core winding and connecting was laborious
and splash-proof machine with end cover removed and difficult because of the inaccessibility of the

Fig. 3—Stacking electrical sheet stator core. Left, conventional machine; right, new fully accessible machine
 Fig. 5—Stator-coil winding. Left, conventional
machine; right, new fully accessible machine

Fig. 4—Machining conventional frame for accurate

positioning of bearing brackets

outside of the coil ends where roping of spacers and

coil supports had to be done (see Fig. 5, left
Welded Fabrications Functionalized
Manufacturability and design objectives de-
manded functionalized components. Figure 6 is an
exploded view of the F A motor components which
may be produced simultaneously on parallel pro-
duction lines. The majority of these components
are produced by weld fabrication and were adaptable
to existing manufacturing lines with the exception
of the top enclosure and the base assembly which
would require fixturing for economical fabrication BEARING BRACKET
and a separate production line. Ala SHIZLDs

Oxygen Cutting Detail Parts Fig. 6—Exploded view of fully accessible motor components
Cutting with natural gas and oxygen was used to
produce detail parts from material in. and greater
in thickness. Tolerances of plus nothing, minus
in. are maintained on critical dimensions and con-
tours. Figure 7 shows an electronic line-scanning de-
vice tracing from a paper template and cutting a top-
enclosure detail. Maintenance of the oxygen-cut-
ting tolerance requires: (1) preparation of the paper
template from a master metal template; (2) a 4 hr
maximum time limitation on the use of the ‘paper
template; (3) single-torch operation; and (4) use of
wedges at preselected spots to prevent movement of
the detail part within the oxygen-cut kerf during the
cutting operation. Oxygen-cutting data for in.
SAE 1015 steel areshownin Fig. 7. In addition to the
wedges, gas savers are utilized to minimize distortion
of the part during the cutting operation. The
gas-saver equipment is shown immediately above the
torch and to the right of the control console. High
readings for the preheat gases are used to pierce the
starting hole for the cut and then immediately
switched to a preset low setting for cutting. The
* i/e We
oxygen-cut detail parts are then drilled and tapped iam High < 35 pei
Preheat ¢ Metural High - 1 of.
as required prior to the next manufacturing opera- ae \ Ges
tion. fe.

Welding-fabrication Area
Fig. 7—Natural gas—oxygen cutting a top enclosure detail
Effective manufacture of the base assembly and from in., SAE 1015. Note wedges inserted in kerf
top enclosure required fixtures and a basic production to prevent movement in cutting operation

WELDING JOURNAL | 1017

 ra
OaINSPECTION |

BASE FINAL ASSEMBLY

BASE DETAIL FABRICATION

f

4
FIXTURE STORAGE

RAW MATERIAL STORAGE

TOP ENCLOSURE DETAIL & FINAL ASSEMBLY

Fig. 8—Three-dimensional layout of welding-fabrication area

Fig. 9—Welding base-corner assembly in fixture. A line. Figure 8 is a three-dimensional layout of this
leted corn ly h h ‘ ,
—, corner assembly hangs from chain area. Raw material and detail parts are stored at
attic Hse one end of the aisle. The details used in the base
tt) fabrication are fabricated into subassemblies that
thet | : move to the right into base final assembly.
In contrast to the base fabrication, only one fix-
ture is required to fabricate the top enclosure.
Detail parts are moved from the raw-material
storage to a point next to the fixture so that the parts
may be loaded into the fixture with an overhead jib
crane.
The inspection area is located between the base
and top-enclosure final-assembly areas. Fixture
storage in special racks is shown at the far right.
Base Fabrication
Adaptation of the machine base to production-line
techniques required fabrication of two subassemblies.
Four corner subassemblies are required for each
completed base. Figure 9 shows the welding of the
base-corner assembly in a _ trunion-type fixture.
This fixture locates and clamps the various detail
parts for welding and may be positioned for hori-
zontal and positioned fillet welding in accordance
with the arc-welding process procedure described in
Fig. 10. Intermittent welds made with E6016
or E7018 electrode are used. The procedure sheet
 — ss
Weertagtenna Ther erperestem Weorimghewss Ehertrh: ( orperetion Becuk ee Sera
oo _1™oms 1 oes ca om __17un055 con 11-08-08
vm __SSE SUS ASSET *L" AnD +e wn S458 SUB ASSEMBLY *\" Amo oR ome SASE_ SUB ASSEMBLY *L* AND *8*
owe /. 2 3. 5, Om 8 wen_', 2, 9, %, mo 9 ve ' 2.3.4 am 5
err oom onevccsremar USING TRUNNION TYPE FinTURE sorter eqeee one csrem er USING TRUBNIOW TYPE Fe TURE marten qowee enevecstamov USING TRUMHION TYPE Fin TURE
(. PLACE (TEM | 18 PORTURE amo Cae ©. MOTATE sus assemy
PLACE ITEM © 18 FIRTURE, CeGAGE TOF Claws. 10. WELO ITEM | TO (TE &, «(CENTER F)RBT - ERDS cesT) 19. ROTATE ASSEMBLY
- PLACE ITEM6 OH F)KTURE AmD Clase | (1. ROTATE sue assemmcy ie, WELD ITEM2 TO ITEMS
. Fieree Clare | TEs a. 12. WELO (TO © TO ITE S. «(CENTER FIRST . OwOS Last) 15. WELO (TER 3 TO (TEx (i-TaCK (8S1GE Eace Emo FIRST)
- PLACE (TEM 2 Om FORTURE an Coe
. PLACE (TEM & 18 FORTURE 4m0 Clee t2-t 3a)
6-4) CXTERNAL Clam & TO& ThGHT If mtCESSaeY 3-1, wewens PLUG GEFORE WEL DUS
. Taek WELD ITEM) TO iTe & ‘ A.
ELD ITEM & TO (TEMS. (CENTER 2 FIRST. CMOS LasT) “ f~. a ~{ D0 NOT WELD To
- v ~~ 1 en]
Te ENO OF IT
(a)
> “a : yy, Ce ae, ASG
TZ tT) see wus re 7) oy oe 5)
22 igwees
i, J . | J 4 y POSITION _OF
— a= acne }
pf oF |
PQerON fon sy? © POSITION
FOR STEP I2
st

COMPLETED SUS ASSEmeLy

cman 268° fh jaesecwanat 287 quer OCP? mee,
cxé aetna os -_ = mn ne - - alee . - — _J
q wee TO Serr —_ " weeo TO SUIT wo wees 10 Su =
i ecremne 5/22" O)AMETER cores, LOH OF WHIT 8 equananass panes AMET EMagtenies LOK OF WHIT 18 5/32" DIAMETER rem, (OF OR WHIZ 16
mma, TN 08e amuay A a = ine eo 2 4% MANUAL METAL ARC
ovvrorm 6-8-8 am 8 coves 4 ©. Mrin ington eretom bP = | om debe WORTHINGTON

Fig. 10—Arc-welding process procedure for base-corner assembly

describes the loading of the fixture and the sequence

of welding to be followed to produce a completed
assembly. Such procedures are used to control the
welding of all parts for the FA motor.
Figure 11 (top) shows the base side-frame assem-
bly fixture and the subassembly welded therein.
Two corner subassemblies, the side-frame machine
foot, the stator support rib and two gussets are
assembled in this fixture. Clamping is utilized to
position the parts and minimize weld distortion.
Since vertical welding was limited to the inter-
mittent welds joining the corner assembly to the
gusset and the stator support rib, it was not economi-
cal to use positioning facilities. Manual metal-arc
welds using E6016 and E7018 electrodes are used.
Figure 11 (bottom) shows the base final-assembly
fixture wherein two _ side-frame assemblies are
positioned and clamped in position with the air-
shield support to form the rectangular base. The
air-shield supports are clamped in position with the
side-frame assembly with air-cylinder clamping
devices. The floor-pan assembly is placed in
position and the unit is welded with gas metal-arc
fillet spots and manual metal-arc intermittent
vertical and horizontal fillets. The small blocks
on top of the corner subassemblies are positioned
with bayonet type locators and welded complete.
These blocks form mounting pads for the top en-
closure at assembly. Each base final-assembly
fixture was designed to receive two bases of different
lengths. The fixture shown in Fig. 11 is presently
extended to its maximum position for its largest Fig. 1l—Top: base side-frame assembly fixture
base. Bottom:
The as-welded base assembly is machined in a Base final-assembly fixture. Mechanical quick-acting and
special single-purpose automatic machine. This air-cylinder clamps are used to hold parts

WELDING JOURNAL |} 1019

 welding. The rings used to hold the arms in posi-
tion on the shaft are SAE 1015 steel and may be
welded to the rotor arm prior to preheat. After
tack welding, the shaft is placed on roller supports
for the automatic submerged-arc operation. The
first pass is made in the groove with the arm rotated
approximately 10 deg above horizontal. The second
pass which forms a fillet between the rotor arm and
shaft is made with the arm approximately 40 deg
above the horizontal. The first pass is made on all
arms on both sides prior to starting the second pass.
After welding, the rotor-shaft assembly is stress
relieved at 1175° F.
Assembly of Basic Machine
Figure 14 shows the assembly of the rotor, stator
and base by welding an adjustable ball-and-socket
locator assembly to the stator and to the base stator-
% support rib. This operation is performed during
Fig. 12—Electro-hydraulic-controlled preset-automatic- final assembly of the machine. First, the bearing
cycle machining of welded-base assembly brackets are attached to the base. Next, the rotor
assembly is positioned within the stator and the
two are lifted and placed with the rotor shaft resting
on its bearings. Air gap shims, shown protruding
machining requires weld-fabrication tolerances that on upper left, Fig. 14, are inserted between the
can only be produced by welding fixtures. The rotor and the stator, and the weight of the stator is
base is inverted and rests upon the top enclosure then supported by the rotor. This operation auto-
mounting pads. The base mounting foot is milled matically positions the stator with respect to the
with a carbide tool in a single pass, while the bear- base. The locator-joint assembly is then placed in
ing bracket fit is faced, bored, drilled and tapped. position for welding to the base and to the stator
Figure 12 shows these operations: the upper head with slot and fillet welds with manual metal-arc
machines the mounting foot, while the lower head is and E7018 electrodes. After welding the four
facing, boring, drilling and tapping the bearing- locator joints, the weight of the stator is picked up
bracket fit. There are two lower heads on the by turning the threaded insert in the locator-joint
machine which work simultaneously. Upon the
completion of the machining on one side of the base,
the heads position themselves to the opposite side Fig. 13—Automatic submerged-arc welding of
and the sequence is repeated. rotor-shaft assembly
-
Stator-core Assembly
Figure 3, right, shows the stacking operation for
the F/A machine. Oxygen-cut and predrilled end
rings are spaced apart on the building bolts. By
comparison of the conventional machine on the
left with the F/A machine on the right, it is apparent
that the functionalized design provides ready ac-
cessibility for the stacking of the electrical sheet
punchings. Each punching contains a_ special
locking ear which simplifies the stacking operation
and gives stability to the core after assembly.
Figure 5 shows the winding and connecting of the
conventional and the F/A machine. It is apparent
again that functionalized design has simplified the
work through improved accessibility for the winder.

Rotor Assembly
For certain sizes, forged rotor shafts were rede-
signed to mild-steel (SAE 1015) arms welded to an
SAE 1035 carbon-steel shaft. Figure 13 shows the
automatic submerged-arc welding, the preheating
operation and the welding data. The shaft and
arm are preheated to 400 to 500° F prior to tack

1020 | OCTOBER 1960

 already been described and is shown in Fig. 14.
Figure 6, an exploded view of the F A motor, shows
all of the components of the machine. Following
assembly of the base, stator and rotor, the top en-
closure is fastened to the base assembly with four
in. bolts. Next, the air shields, end covers and
side covers are bolted in place. These small sheet-
metal components are fabricated in the sheet-metal
department and contain a minimum of welding.
Figure 1 shows the three types of machines after
final assembly.
Summary
Through the integrated efforts of the engineering
design and manufacturing engineering personnel and
Fig. 14—Assembly of rotor-stator to base the functionalized concept of designing for produc-
with locator joint. Insert, lower left: exploded tion welding, industry now has a fully accessible
view of locator joint
motor. The functionalized concept allows simul-
taneous production of all motor components, result-
ing in a decrease in elapsed manufacturing time
assembly and a final micrometer adjustment of the which allows short delivery dates to be met. Utili-
air gapismade. The air gap shims are removed and zation of welding fixtures to fabricate the top en-
the threaded insert is tack welded to the locator closure and base assembly from predrilled and tapped
joint base shown at the bottom left of the lower left detail parts has utilized weld-fabrication techniques
insert, Fig. 14. that are ordinarily reserved for manufacture of
smaller apparatus. Detail-part tolerances of plus
Top-enclosure Fabrication
nothing, minus in. were required on all mating
Figure 15 shows the top-enclosure fabrication- parts assembled and welded in fixtures. These
assembly fixture in the foreground, and the I-beam tolerances were developed for electronic line-tracing
subassembly fixture in the upper left background. oxygen cutting of shaped detail parts.
The sheared and press-formed, ' ;-in. thick steel
covers are first placed in the fixture. The next item Acknowledgment
to be positioned is the I-beam unit shown in the The authors gratefully acknowledge the assist-
lower center of the top enclosure. The air-shield ance and design and manufacturing efforts of A. S.
support plates are then positioned against the louvres DePaul and M. M. Schnaubelt, Engineering Design,
in the cover. The remaining details are placed in and F. H. Frantz and J. C. Worthington, Manu-
position along with the end-frame ring. The air- facturing Engineering.
shield support and the end-frame ring are oxygen
cut to shape, and drilled and tapped prior to as-
sembly. All pieces are then joined by inert-gas con- Fig. 15—Top-enclosure fabrication assembly fixture and port-
able consumable inert-gas-shielded metal-arc welding unit
sumable-electrode metal-arc welding with fillet Upper left: assembly and weld fixture for top-enclosure
spot welds and using the weld process data shown l-beam unit
at the bottom of Fig. 15. <A portable wire-feed unit
is used in conjunction with the welding machine to
improve accessibility to the work with a minimum
of effort in moving equipment. The portable unit
is placed at each end of thé enclosure fixture, as
shown in Fig. 15. One-halfof the fillet spot welds are
made with the drive unit positioned in each of its
positions. Final fillet welds are made between the
-in. cover and the end-frame ring and the lifting
support bracket with manual metal-arc process,
according to the weld data shown in Fig. 15.
After welding and after removal of the manual
metal-are slag, the inside surfaces of the top en-
closure are sprayed with an asbestos fiber and
powdered-cork gilsonite asphalt-base coating. This
treatment provides additional sound muffling for
quieter operation.
Final Assembly of Motor
The assembly of the base, stator and rotor has
 Shape cutting of heavy steel plate with multiple-torch machine that can operate with natural gas as well as with acetylene

The proper use of propane or natural gas and the selection of suitable

equipment are covered in this discussion of the

Recent Developments in

Oxy-Fuel-Gas Cutting

BY C. C. ANTHES

In recent years there has been an increasing use of paper is to discuss the proper use of propane or
fuel gases other than acetylene in the oxygen cutting natural gas and the selection of suitable equipment.
of steel. In some types of cutting, the use of either Operators, who have used acetylene as the pre-
oxypropane or oxy-natural-gas preheat produces heat fuel gas in cutting, often find it difficult to
satisfactory results with over-all cost savings. switch to other fuel gases and obtain satisfactory
Other types of cutting can best be performed with results. Fuel gases such as propane and natural
the high-temperature highly concentrated flame of gas call for torch-valve adjustments and preheat-gas
the oxyacetylene preheat. The purpose of this pressure settings that are different from those for
acetylene. In addition to different flame settings,
C. C. ANTHES is Project Engineer at the Development Laboratory of torches designed for acetylene service are not
the Linde Co., Div. of Union Carbide Corp., Newark, N. J
entirely suitable with other fuel gases, particularly
Paper presented at AWS 41st Annual Meeting held in Los Angeles, Calif.,
Apr. 25-29, 1960 natural gas.

1022 OCTOBER 1960

Problems of Propane and Natural Gas orifices (Q/A value). It will be noted that the
The primary problems in using oxypropane or curve is for natural gas and for a flow of 65 cfh ata
oxy-natural-gas preheat are (1) excessively long ratio of preheat oxygen to natural gas of 2 to 1.
preheat time, (2) low cutting speeds and (3) poor The flow of 65 cfh was chosen as a reasonable flow
cut quality. These three problems are interrelated, for fast pierce starts. The 2 to 1 ratio of preheat
and their causes can be multiple in nature. This, oxygen to natural gas was chosen because this ratio
in turn, means that multiple remedial steps are gives the highest flame temperature and greatest
frequently necessary and a full understanding of concentration of heat per unit area of flame front.
preheat-flame requirements and adjustment are The very tips of the preheat-flame inner cones were
needed to obtain good results. just barely touching the plate surface at all times.
Oxy-natural-gas preheat offers perhaps more Study of Fig. 1 will illustrate the important part
problems than oxypropane preheat in that the oxy- preheat-gas velocity plays in heating effectiveness.
natural-gas flames provide less concentration of heat A gas velocity of 1000 fps gives a preheat start
per volume of fuel gas. Heat output, flame tem- time of 5.3 sec compared to 10 sec for a velocity of
perature and concentration of heat play a vital 450 fps. The velocity of 450 fps is comparable to
part in the effectiveness of preheat, especially during that normally obtained with oxyacetylene preheat
the preheat cycle prior to cutting. In heating
effectiveness, acetylene is the top performer by a
wide margin, followed by propane and then natural
gas. However, proper use of propane and natural
gas reduces the differential to an acceptable level for
the actual preheat starts.
Fundamental Elements of Heat Transfer
The transfer of heat from a flame to metal is
accomplished by radiation in combination with ra)
x)a@
impingement of the hot-gas molecules on the surface
of the metal. The quantity of heat transfer and the
>
temperature of the metal attained are governed by
many factors. Some of these factors are: flame PREHEAT
START
TIME
SECONDS
—
temperature, concentration of heat units per unit
area of flame front, velocity of impingement of the
hot-gas molecules on the metal surface, temperature 4 6 8 10 12 4 16
gradient between the flame and the metal and PREHEAT GAS VELOCITY ~ FT. PER. SEC
through the mass of metal and chemical reaction Fig. 1—Preheat start time vs. velocity of preheat gases
of the hot gas on or with the metal. issuing from flame ports (natural-gas flow—65 cfh, preheat
The transfer of heat by radiation is, for all practical oxygen to natural-gas ratio—2 to 1, center of plate preheat—
purposes, fixed for a given flame temperature and 2-in. material, flame inner cones just touching metal sur-
face)
area of flame front facing the metal. The transfer
of heat by the impingement of hot-gas molecules on
the metal surface is proportional to the velocity of
impingement, assuming a fixed temperature gradient
between the flame and the metal.
Some heating applications require a relatively
slow rate of heat transfer to obtain uniform heating
of the metal and a low heat gradient through
the mass of metal. The “soak’’ preheat of castings is
one example of such an application. Other heating
applications require a high rate of heat transfer to
minimize heat conduction through the metal and to
permit short surface-heating time. Gas welding,
flame hardening, scarfing of metal and oxygen
cutting fall into this category. -PREHEAT
START
TIME
SECONDS

Preheat-gas Velocity 44
The effect of the velocity of the preheat gases oly 5 1.6 1.7 8 1.9 2.0 2.1 2.2 2.3
Q/A value) leaving the flame ports is important PREHEAT GAS RATIO- VOLUMES OF PREHEAT OXYGEN
PER VOLUME OF NATURAL GAS
relative to the effectiveness of preheat, especially
for pierce starts in mechanized cutting. Figure 1 Fig. 2—Preheat start time vs. preheat-gas ratio (natural gas
flow—65 cfh, velocity of preheat gases issuing from flame
shows the preheat time in seconds for pierce starts on ports—1000 fps with 2 to 1 preheat-gas ratio, center of plate
a 2-in. thick plate in relation to preheat-gas velocity, preheat—2-in. material, flame inner cones just touching
in feet per second, as the gases leave the flame port metal surface)

WELDING JOURNAL | 1023

 and is frequently used with fuel gases such as propane Preheat-time Relationships
and natural gas. The use of low velocity is one of Figure 3 shows the relation between pierce-
the contributing factors to excessively long preheat start preheat time and preheat-gas ratio when
time, especially when using natural gas. Low- using propane as the fuel gas. The most effective
velocity preheat can also cause reduced cutting ratio is approximately 5 volumes preheat oxygen to 1
speed, especially if the metal surface is heavily volume of propane. As with natural gas, operators
scaled or caked with rust. have a tendency to use a lower gas ratio (approx-
Importance of High Preheat-gas Ratios imately 4 to 1) with attendant losses in preheat
effectiveness. A propane flow of 30 cfh was chosen
Preheat-gas ratio plays a large part in effective
as a reasonable flow for fast pierce starts. The
heat transfer from preheat flames to plate for fast
flame inner cones were just touching the plate sur-
cut starts. Figure 2 shows the relation between
face. This flame cone-to-plate condition is the
preheat time in seconds for pierce starts in the
most effective for maximum heat transfer from flame
center of a 2-in. thick plate and the volume ratio of
to plate.
preheat oxygen to natural gas. A flow of 65 cfh
The rate of fuel-gas flow plays a part in the pre-
of natural gas was maintained. This is the same
heat effectiveness of preheat flames. However, the
flow used in establishing the effect of velocity.
flow rate does not play nearly as large a part as
The velocity of the preheat gases as they issued from
either preheat-gas velocity or ratio of preheat
the flame ports was also a variable. However, this
oxygen to fuel gas. There should be sufficient flow
variable had less effect in establishing the gas ratio
for attaining the high velocity and ratio required
relation than would have resulted from altering
and also for liberating the volume of heat necessary
the flame ports to maintain a constant velocity.
for reasonable heating effectiveness.
As in the case of the velocity graph, the preheat-
Figure 4 shows the relation between preheat time
flame inner cones were just touching the plate
for pierce start and natural-gas flow in cubic feet
surface.
per hour. The ratio of preheat oxygen to natural
Study of Fig. 2 illustrates the value of using a
gas was maintained at 2 to 1, and the velocity of the
high preheat-gas ratio. A ratio of two volumes of
preheat gases issuing from the flame ports was held
preheat oxygen to one volume of natural gas results
at 1000 fps—optimum values for both variables.
in the shortest pierce start time. At this ratio,
It will be noted that the slope of the curve is more
the flame temperature is at maximum, and the heat
gentle than the previous curves. This indicates that,
concentration per unit area of flame front is at a
while fuel-gas flow rate has an effect on preheat time,
maximum. The average ratio of preheat oxygen to
the magnitude of change in preheat time is not as
natural gas used by most operators is approximately
great as when changing either gas velocity or
1.7to1l. At this ratio, the flame appears to have the
ratio.
most mass and heat. Again, it is a case of long
Figure 5 shows the relation between preheat time
experience with the use of oxyacetylene preheat
for pierce start and propane flow in cubic feet per
flames and the effort of the operator to simulate
hour. As in the case of natural-gas preheat, the
these flames when using oxy-natural gas. It can
preheat-gas velocity and ratio were maintained
readily be seen that use of this lower ratio results in
respectively at the optimum values of 1000 fps and
double the preheat start time obtained when using
5 to 1. The trend and indications shown here are
the 2 to 1 ratio.

ro)
3@

o
mo
+
n
PREHEAT
TIME
START
-SECONDS
- nN
PREHEAT
TIME
START
SECONDS
= 40 50 60 70 80
i} 30 3.5 40 45 5.0 5.5 6.0
PREHEAT GAS RATIO-VOLUMES OF PREHEAT OXYGEN PER VOLUME OF PROPANE NATURAL GAS FLOW ~- CFH
Fig. 3—Preheat start time vs. preheat-gas ratio (propane Fig. 4—Preheat start time vs. natural-gas flow (preheat
flow—30 cfh, velocity of preheat gases issuing from flame oxygen to natural-gas ratio—2 to 1, velocity of preheat gases
ports—1000 fps with 5 to 1 preheat-gas ratio, center of plate issuing from flame ports—1000 fps, center of plate preheat—
preheat—2-in. material, flame inner cones just touching 2-in. material, flame inner cones just touching metal sur-
metal surface) face)

1024 | OCTOBER 1960

similar to those previously shown for oxy-natural- propane. These start times are independent of
gas preheat. plate thickness on all plate over approximately
Figure 6 illustrates the relation between pre- s in. thick. In other words, it will take no longer
heat time for pierce start and preheat-flame inner- to pierce start on 6-in. thick plate than it will on
cone distance from the surface of the plate. The -in. plate. Pierce start refers to the start of
natural-gas flow and oxy-—fuel-gas ratio were 65 cfh rapid reaction of the oxygen with the steel and does
and 2 to 1, respectively. The velocity of the pre- not take into consideration the time to pierce
heat gases issuing from the flame ports was 1000 through a plate. Edge starts will be considerably
fps. The graph illustrates the rather critical set- faster——practically instantaneous on sharp edges.
ting of the flame in respect to the plate surface. Preheat-start requirements have been given heavy
The optimum setting, where preheat start time is at a emphasis since most difficulty has been experienced
minimum, is where the preheat-flame inner cones in this phase of cutting when changing over from
just touch the surface of the plate. The flame acetylene to other fuel gases. In some cases
cones either impinging on the plate by as little as operators have been struggling along with preheat
in. or off the surface of the plate by the same times up to and even exceeding one minute. Barring
amount results in a marked increase in preheat start torch and nozzle shortcomings for the moment,
time. these long preheat times are the result of lack of
knowledge as to proper preheat-flame adjustment or
Pierce Starts
as to an inability to use the more effective preheat
General Requirements flames because of excessive heat causing kerf-edge
Reviewing what has been covered up to this point, roll-over in machine cutting.
the fastest possible preheat time for a pierce start
will result if: Preheat-flame Requirements for Cutting
1. The velocity of the preheat gases as they leave
Preheat Cutback
the flame ports is high, i.e., 1000 fps (QA value
or better. The preheat-flame requirements for the cutting
2. The ratio of the preheat oxygen to fuel gas is at cycles are frequently quite different from those for
the point where maximum flame temperature is at- preheat starts. In the case of hand cutting
2 to 1 for oxy-natural gas and 5 to 1 for riser cutting, scrap cutting, rough sizing cuts, etc.,
tained
oxypropane. where kerf-edge squareness is relatively unimportant
3. The flow of fuel gas is at a reasonable rate the full preheat as used for starts is satisfactory
whereby the heat liberated will be of sufficient and often preferred. However, in most mechanized
volume to provide effective heat build-up in the cutting it is desirable to have cuts of top quality with
plate. Approximately 65 cfh of natural gas and sharp kerf-top edges, parallel-cut faces and no slag
30 cfh of propane are reasonable rates of flow. adherence on the bottom of the kerf. In these
cases it is necessary to reduce the size of the pre-
Results of Proper Preheat heat flames and also to reduce the ratio of preheat
Figures 1 to 6 indicate that if the above conditions oxygen to fuel gas. This is best accomplished with a
are met, it is possible to make pierce starts on plate preheat controller that will automatically cut
normal hot-rolled mild steel) in approximately 5 back the flow of preheat oxygen and fuel gas by a
sec when using natural gas, and 3.5 sec when using predetermined amount when the cut is started.

>nyfFDaN@Ww a
-/’4:
4
>
PREHEAT
TIME~-SECONDS
START
So= > °s
PREHEAT
START
TIME-
SECONDS
or= 30 35 40 45 Yea "ie ~hea -Y2 ~Yea ©
PROPANE FLOW — CFH FLAME INNER CONE DISTANCE FROM PLATE SURFACE — INCH
Fig. 5—Preheat start time vs. propane flow (preheat oxygen Fig. 6—Preheat start time vs. flame inner cone distance from
to propane ratio—5 to 1, velocity of preheat gases issuing plate (natural-gas flow—65 cfh, preheat oxygen to natural
from flame ports—1000 fps, center of plate preheat—2-in. gas ratio—2 to 1, velocity of preheat gases issuing from
material, flame inner cones just touching metal surface) flame ports—1000 fps, center of plate preheat—2-in. mate
rial)

WELDING JOURNAL | 1025

 START PREHEAT CUTTING PREHEAT
CUTTING
OXYGEN

PREHEAT
—— OxYGEN >

—>

Suen GAS aS

Fig. 7—Action of preheat controller

The degree of preheat cutback is governed by and thereby greatly shortened preheat start times.
several factors. One factor is the condition of the One advantage is the greater stability of such
plate surface. If the plate surface is coated with flames. High-velocity flames are resistant to dis-
heavy rust and scale, the quantity of preheat re- turbance by flying scale or dirt from the plate surface
quired may very well approach the full preheat that would normally cause lower-velocity flames to
flow used for starting. Another factor is the type of pop and possibly result in losing the cut. Another
steel being cut. Some steels, especially the high- advantage is that the high-velocity flames reduce
alloy steels, require considerable preheat volume nozzle fouling with slag to a minimum. This is
even during the cutting cycle. A third factor also especially important in the case of mechanized
is the type of cutting being done. Cutting bevels cutting when using multiple torches in _ pierce
for plate preparation for welding is one such type starting. In this type of cutting, it is customary to
requiring considerable preheat during the cutting reduce further the chances of nozzle fouling by
cycle in order to maintain a reasonable cutting raising the preheat flames off the surface of the
speed. Preheat cutback for the cutting cycle when work for a distance. This is done even though it
cutting relatively clean low-carbon steel with the means lengthening the preheat start by a marked
average light coating of mill scale will be as much as degree.
approximately 75°% for relatively light plate (*/s; to Torch Selection
2 in.), tapering off to 10 to 25% cutback for the
Having established preheat requirements, the
heavier plate (3 to 8 or 10 in.). In all cases, the
next step is to select torches that will permit use of
preheat cutback should be just sufficient to eliminate
the correct preheat. This is often difficult, since
kerf top-edge roll-over. Any preheat reduction
operators are not fully familiar with torch perform-
beyond this point will reduce the cutting speed.
ance on the various fuel gases. The cutting torch
In fact, many operators complaining of reduced
plays a large part in the effective use of fuel gases.
cutting speed when using fuel gases other than
Choice of the type of torch (low- or medium-pres-
acetylene are, in most cases, using an insufficient
sure) must be governed by the fuel-gas supply
volume of preheat.
pressure. An aspirating injector-type torch must
Oxy-Fuel-gas Ratio Changes be used if the fuel-gas supply pressure is 3 psi or less.
It is desirable to also reduce the ratio of preheat If the fuel-gas supply pressure is above 3 psi, a
oxygen to fuel gas once the cut is started (see Fig. 7) torch having a medium-pressure injector or mixer is
In the case of natural gas, the ratio should be reduced preferable if full preheat effectiveness is to be
from 2 to 1 to approximately 1.5 to 1. Propane realized.
requires a reduction in ratio from 5 to 1 to approx- Aspirating injector-type torches as illustrated in
imately 3.5 to 1. The softer flames of the lower Fig. 8 for use on low-pressure fuel-gas supplies have
ratio ensure maximum cutting speed coupled with limitations that should be understood. In order to
top-quality cuts with no slag adherence. Very aspirate sufficient fuel gas for effective preheat, it is
heavy cutting (10 in. on up) will benefit from use of necessary to utilize a relatively high preheat-oxygen
still softer flames—that is, flames obtained with a pressure. In the case of a two-hose torch, such as a
ratio as low as 1 to 1 for natural gas, and 2.5 to 1 for hand-cutting torch or a so-called two-hose machine
propane. cutting torch, the cutting-oxygen pressure required
The use of high-velocity high-ratio preheat has for a given thickness of cut establishes the maximum
advantages other than more effective heat transfer preheat-oxygen pressure that can be obtained.

1026 | OCTOBER 1960

Therefore, the preheat-flow range must be restricted permit use of maximum volume and velocity of pre-
to a large extent if most effective aspiration of low- heat for the very short preheat-start times. This
pressure fuel gases is to be realized. type will also extend the range of a torch by pro-
It is very desirable to furnish at least two aspirat- viding very high rates of preheat flow required
ing injector sizes for torch operating on fuel-gas for such operations as extremely heavy cutting,
supply pressures under 3 psi. One injector should be cutting through very dirty or heavily encrusted
of such size that it will aspirate the maximum metal, gouging, rivet cutting, fin washing and pad
possible quantity of fuel gas when the fuel-gas removal.
supply pressure, as measured at the torch inlet Torches designed specifically for oxyacetylene
connection, is '/, psi or lower. This injector will be preheat can be used for other fuel gases. This is
somewhat limited as to the quantity and velocity of especially true if the torch has a medium-pressure
preheat that can be obtained but will be able to injector or mixer and, of course, is to be used on
perform at the very low fuel-gas pressures. The medium-pressure fuel gases. An oxyacetylene torch
second injector should be of larger capacity, having an aspirating-type injector will have very
capable of furnishing greater volumes of preheat at a limited use on other fuel gases. It is necessary to
higher, more effective velocity through the nozzle replace the injector with one that is designed for
preheat ports. This injector will not be able to fuel gases. Once the injector is replaced, the
aspirate as effectively as the smaller injector and performance and capacity of the torch will be fully
will, therefore, require the higher fuel-gas pressure realized.
from '/, to 3 psi.
A torch that is to be operated on a fuel-gas Nozzle Requirements
supply line that will ensure a fuel-gas pressure of 3 Nozzles for oxy-fuel-gas cutting must meet cer-
psi or more (as measured at the torch inlet con- tain requirements. They must be able to support
nection) should have a medium-pressure injector the full flow of fuel gas for quick starts without the
or mixer (Fig. 9). This type injector or mixer will flames blowing_off or the nozzle overheating. They
should be able to provide the concentrated high-
velocity high-ratio preheat flames necessary for
effective cutting of heavily rusted and scaled metal,
and pierce starts on heavy metal in preheat times of
3 to6sec. It should be possible to cut back the pre-
heat without the preheat flames either overheating
the nozzle or becoming unstable. Fuel-gas nozzles
should be of rugged construction in order to stand
up under such operations as plate-edge bevel
operations, piercing and cutting of very dirty mate-
rial.
FUEL GAS
(-2 TO+3 PS.I.)
PREHEAT OXYGEN Summary
\ J (15 T0 75 P.S.1.) The use of fuel gases other than acetylene is
\ frequently accompanied by excessively long pre-
FUEL GAS & PREHEAT OXYGEN heat times, low cutting speeds and poor cut
quality. These difficulties can be resolved through
Fig. 8—Aspirating injector-type torch
the use of:
1. Torches with aspirating-type injectors when
fuel-gas supply pressures are 3 psi or less—and
torches with medium-pressure injectors or mixers
when fuel-gas supply pressures exceed 3 psi.
2. High preheat-gas velocities and preheat oxygen-
to-fuel gas ratios of 2 to 1 for oxygen-natural gas
and 5 to 1 for oxygen-propane when making pierce
starts.
3. Preheat controllers during mechanized cutting
FUEL GAS to cut back the flow of preheat oxygen and fuel
(3. TO 15 PS.) after pierce starting.
Under the proper conditions, it becomes possible
PREHEAT OXYGEN
(3 TO 30 PS.L) to make pierce starts in approximately 5 sec with
natural gas and 3-4 sec with propane compared to
the minute or more which has sometimes been re-
FUEL GAS & PREHEAT OXYGEN quired. Subsequent cutting may then be carried
Fig. 9—Mixer or medium-pressure injector-type torch out with entirely satisfactory results.

WELDING JOURNAL | 1027

 F4H-1 launching Sparrow III with Guardian | engine

Machine Welding of a Prepackaged

Liquid Rocket Engine

contributes to successful quality output of new family of thrust units

BY RALPH L. HOETGER AND WALTER B. MOEN

ABSTRACT. Many factors enter into the selection of a instant use. During the course of propellant research,
rocket power plant for missile systems designed for the engineering development of a rocket engine to
specific missions. One of the most important factors in utilize the propellants was conducted. This develop-
the selection is a consideration of the propellant system. ment was concerned not only with engine design to
A significant addition to the arsenal of missiles is the first meet performance predictions, but also with the physical
production of a prepackaged storable liquid rocket design for manufacture in production quantities. One
engine. This rocket-engine development is the result of of the most important design considerations was the
propellant research that has culminated in a hypergolic use of welding to assemble the forged and machined
bipropellant combination. The propellants can be components so that the assembly would be capable of
loaded into the tanks at the point of missile manufacture withstanding the high pressure developed in firing and,
and then be stored over extended periods ready for during storage, would resist the corrosive attack of the
propellant.
RALPH L. HOETGER is Plant Superintendent, Thiokol Chemical The type of liquid engine described consists of a com-
Corp., Reaction Motors Div., Bristol, Pa., and WALTER B. MOEN is
Engineering Manager. Special Products Department, Air Reduction bustion chamber and exhaust nozzle completely sur-
Sales Co., Union, N. J rounded by a propellant tank. Forward of this com-
Paper presented at AWS 4ist Annual Meeting held in Los Angeles, bination, but integral with it, is another annular-pro-
Calif., Apr. 25-29, 1960 pellant tank surrounding a solid-propellant grain used

1028 | OCTOBER 1960

 for the liquid-propellant pressurization. At the time of hindered by difficulties in mechanical achievement
firing, the solid cartridge is ignited, thereby liberating and in basic rocket-engine technology. The ad-
gases for the pressurization of the propellant system.
Simultaneously, the gas pressure is exerted against a vantages and inherent simplicity of this approach
piston which moves to uncover ports that permit the were foreseen by German engineers at the end of
propellant to enter the combustion chamber. Upon World War II, and work continued in this country in
mixing, the propellants spontaneously react to put the the period immediately following the war years.
engine into operation. One can visualize that the
assembly is a system of concentric pressure vessels In the early 1950’s, the potential of the packaged
requiring resistance to both high shock and steady-state liquid-rocket power plant was recognized, provided
loading. solutions to problems in three specific areas could be
High-quality weldments with rigid ‘quality-control made. These problem areas and the methods of
standards” allow little variation in the welding-process solution arrived at are:
parameters. For this reason, wherever possible, ma-
chine-welding techniques have been employed. In addi- 1. The selection of a high performance bipropel-
tion, the engine requires accurate dimensions so that the lant combination and tankage construction material
proper amounts of propellants can be charged. suitable for long-term storage. Following detailed
Both gas metal-arc welding and gas tungsten-ari
welding processes were selected for joining of the virtually studies, a basic engine concept was evolved utilizing
all-aluminum assembly of various alloy compositions inhibited red fuming nitric acid as the oxidizer, and
under different conditions of heat treatment. Because of an alcohol-base or amine-base fuel stored in a welded-
dimensional limitations and the configuration of the aluminum tank shell.
parts, machines specifically designed for the joining
processes were constructed. Incorporated into these 2. The choice of a compact energy source to
machines are precision controls for speed of rotation, pressurize and expel the propellants from the tanks
welding current, filler-wire feed, wire-feed positioning, into a combustion chamber. A solid propellant
plus programmed initiation, welding cycle and shutdown. charge was chosen as the pressurizing means rather
These controls and welding parameters to achieve the
end result will be described in detail. than compressed-inert-gas or mechanical pressurizing
In addition to the actual welding conditions, close systems.
control must be exercised over the preparation of the 3. The choice of an injector and initiating device
components before welding. Rigorous dimensional con- both robust and readily producible, would enable
trol to obtain accurate fit-up of the parts and thorough
cleaning to remove all contaminants are among the fac high performance to be achieved. A simple injector
tors considered. and initiating means was devised to meet these re-
Because of the tight controls and established prepara quirements.
tion and welding process factors, it has become possible The engine concept arrived at in these early studies
to use operators with little or no welding experience
The assemblies produced under these conditions have is essentially the engine being produced today. The
resulted in complete engines that have passed rigorous liquid propellants are completely sealed in high-
quality-control standards. strength welded-aiuminum tankage which is rugged
As the missile era enters high-production stages, it and able to take a considerable amount of abuse.
becomes evident that machine welding to close controls The solid-propellant pressurizing charge is stored in
is necessary to achieve quality missile systems at reduced
cost. This paper will completely describe the methods the unit and a “last minute’”’ initiator cartridge is all
and conditions used to meet these requirements. that is required to make the thrust unit operational.
The unit is handled as a single round in precisely the
Introduction same manner as a solid-propellant motor and em-
A packaged liquid-rocket thrust unit is one in which bodies the basic missile-installation features such as
liquid propellants and the pressurizing medium are the attachment threads, mount lugs and fin attach-
sealed into a tankage shell which is integral with the ments. Firing of the unit is accomplished simply by
rocket thrust chamber. While the possibility of the provision of electrical energy to the initiator and
incorporating these elements into one round had been there is no countdown or other preparation period as
recognized for many years, this development was is conventionally associated with liquid-propellant
rocket engines.
Currently, two engines identical in principle and
differing only in dimensions are being produced.
One thrust unit, the “Guardian I,” is used for the
Sparrow III air-to-air missile; and the other,
“Guardian II,” is employed as the power plant in an
air-to-surface missile, the Bullpup. It is interesting
to note that this concept of packaged liquid rocket
engines is not limited to the sizes of Guardian I and
Guardian II. A much larger unit in the 50,000 lb
thrust class already has been fired.
WELDED A schematic diagram of the Guardian I engine, in-
JOINT dicating the machine weldments and major compo-
MOUNTING
CONE nents, is shown in Fig. 1. A cut-away photograph of
Fig. 1—Typical construction of prepackaged liquid the Guardian II is shown in Fig. 2.
rocket engine—machine welded While many problems were encountered in the

WELDING JOURNAL | 1029

 development of the engine, one of the principal
achievements in reducing the prototype to produc-
tion was the application of sound state-of-the-art
welding techniques fully integrated with the many
other production processes. This growth of welding
to a fully integrated operation with other production
processes in the production of advanced missile
power plants is the subject of the remaining portions
of this paper.
Thrust-unit Production
An insight into the over-all production pattern of
the packaged thrust unit will give a better under-
Fig. 2—Partial cutaway of typical prepackaged standing of where the various welding steps fit. A
liquid rocket engine
simplified sequence of production processes is shown
in Fig. 3, with the welding operations clearly indi-
cated.
The unit is constructed almost in its entirety of
WEAOER 6066 aluminum alloy in the T4 condition. This
ZER
TANK ; material, in both the forged and extruded forms, has
EXTRUSIONnIDITER
OxI0LiweR BAND THRUST
BURST CHAMBER
FORGING CENTRAL
FORGING THRUST
CHAMBER
PaRTS PURCHASED a 50,000 psi ultimate tensile strength with machin-
GAFFLE
ability and weldability comparable to 6061-T4.
rH The only exceptions of metallic parts to the 6066 al-
loy are the use of some other aluminum alloys,
OHFC copper for an exhaust nozzle and steel for the
pte piston. Welding is not required on these materials.
These items and purchased items such as propellants,
ceramic rods, solid-propellant grain, etc., are all part
of the final assembly.
‘a The majority of parts to be welded, whether ex-
(" WER- ass'y
BURST | truded or forged, require initial machining, both to
[weaoen asst]
|

BAND bring them down reasonably close to final dimen-

J sions and also to prepare the abutting edges for weld-
ing.
~ A cleaning process for the parts to be welded is per-
formed under closely controlled conditions and is
waMeer|
[neaoen assy
—
war
wer
c__- CHameEr—
Tanne 6assy

“al
{ PRessune
Tesr a
[ctwawic
© COATING]
coarim
* CHAMBER
fimau assy
FALLINGB LOADING
M-MACHINE
W-WELDING PAINTING
1- INSPECTION a

Fig. 3—Simplified sequence of production processes Fig. 4—Header—metering orifice welding machine

1030 | OCTOBER 1960

handled with almost antiseptic care. It is worth sideration, three different weld-quality categories
while here to outline the cleaning sequence: were established as follows:
1. Welds which are used for primary structural
Immersion in alkaline detergent at 4 oz/gal
application: any weld, the failure of which would
solution at 120—140° F for 5 min.
cause loss of power, thrust, control or structural col-
Cold-water rinse—air agitated. lapse. These welds are X-rayed where possible or
Deoxidation in chromic acid, sodium acid provisions made for hydrostatic or static proof test-
sulfate at 14 oz/gal solution. ing. A special subcategory was established for
Cold-water rinse—air agitated. machine welds wherein minimum weld properties
5. Hot-water rinse at 140—-160° F. and quality and methods for evaluation were set
6. Hot-air dry. forth.
2. Welds which are used for secondary structural
After each welded subassembly is completed, rigid
applications and seal welds in pressure components:
inspection requirements must be met. These re-
any weld, the failure of which in conjunction with
quirements are detailed later in this description.
the failure of another weld would result in failures
Also, after some of the subassemblies are made,
of the kind described under paragraph 1 above.
heat treating and subassembly machining operations
3. Welds for low and/or unstressed nonstructural
are performed either to bring the parts down to final
applications have to be of good workmanship but are
dimension in cases where further machining is
not subject to the previously mentioned rigid re-
difficult because of inaccessibility, or where such is
quirements.
not the case, to approach final dimensions and pre-
Each welded joint was designed initially so that it
pare additional edges for welding. Here again, rigid
would meet the performance requirements and desig-
dimensional checks are made so that the required
nated as to the category in which it fell. Then,
close tolerances can be maintained.
depending upon the category, inspection require-
It is interesting to note the large number of dif-
ments were established. For all weldments of 6066
ferent types of operations that must be performed
aluminum alloy, one or more of the following inspec-
after the last weldment is completed and prior to
tion methods were delineated:
shipment. Such items as final cleaning, painting,
1. Dye-penetrant Inspection—aAll welds where
filling, crating, etc., also are indicated in Fig. 3.
possible are 100% inspected per MIL-I-6866; cracks
One of the steps of interest to engineers is the special
are cause for rejection.
ceramic coating sprayed on the interior of the com-
2. Radiographic Inspection—Where possible,
bustion chamber to act as a thermal barrier. It isa
joints are 100% radiographically inspected per
process applied with an oxyacetylene flame in a
MIL-I-6865; and to pass inspection must be at least
mechanized operation.
as good as the radiographs of qualifying test welds.
All of the above operations are performed in a
Scattered porosity and tungsten inclusions are judged
rehabilitated plant with a manufacturing area of
from Fig. Cl-1 of MIL-R-11468. The qualifying
approximately 100,000 sq ft. The area is utilized
test welds referred to herein include X-ray, macro-
in a manner to insure efficient flow of the work from
examination and tensile testing of at least two com-
raw materials to finished engine storage. All of the
plete joints made under identical conditions by
welding operations are not located in one area as is
qualified welders under a qualification procedure.
common in many plants, but are spotted in preferred
3. Hydrostatic Testing—-Where necessary, specific
locations in the product process. To accommodate
test methods adaptable to the thrust-unit compo-
the welding equipment, water, power, air and inert
nents were developed. Evidence of leakage or
gas have been piped from a central supply to the
permanent set is cause for rejection.
welding stations. The inert gases, helium and argon,
4. Visual Inspection—All welds examined for ap-
are supplied from manifolded ten-cylinder cradles
pearance and surface quality to pass inspection must
and distributed to the service drops, where individual
meet these requirements:
pressure and flow regulation is achieved for use, either
neat or in admixture. a) Base metal to be free of undercutting.
b) Weld surface to be free of overlap, excessive
Quality Demands roughness, surface cracks and porosity.
Severe and rigid requirements were set up for all Groove and fillet welds to be slightly convex.
welded joints. Two primary dictations were that all
weldments had to be of aircraft quality so that all Machine welders must adhere to the process con-
structural loads could be taken; also, that the di- ditions established during the time of the weld-
mensions of the subassemblies and assembly would qualification test. For each machine weldment, a
be within very close limits. Changes in tankage di- specification sheet has been established which sets
mensions will affect the propellant capacity; pro- forth all process conditions.
pellants are loaded into the thrust unit to an ac- All manual welders have to pass the tests described
curacy of +'/,oz. Any lack of control here will re- under MIL-T-5021 for Aircraft Welding Apparatus
sult in off-required performance and range. Certification. Repair welders also fall into this
To set up reasonable requirements for the first con- category. Repairs can be made to both hand and

WELDING JOURNAL 1031

 machine weldments under strict limitations.
Dimensional checks are made not only of ma-
chined parts but also of welded assemblies. Ec-
centricity, diameter, mismatch, part growth during
welding, etc., all are determined where applicable to
specific subassemblies to insure that subassemblies,
when mated, will be within tolerance requirements.
Welding Operations
In order to describe the welding operations, one of
the thrust units, the Guardian II, will be dissected
and the requirements of each joint, the process,
tooling and equipment used will be given in some de-
tail in the sequence in which weldments are made.
Metering Orifices
The first welding operation is to join a series of
metering orifices in a subassembly to the central Fig. 5—Burst-band welding machine
header. Because of the severe space limitations, it
was thought that this operation would have to be
performed by hand. However, a machine was de-
vised that is capable of making joints that meet a
zero-leakage test requirement. Figure 4 is a photo-
graph of one of the two vertical-type welding ma-
chines used for the job.
The space limitations imposed by the header
flange and the supporting ribs prevented the use of a
commercially available tungsten-arc torch. Many
unsuccessful attempts were made to utilize filler wire
until the joint design was changed to accommodate
preplaced rings. The use of a special tungsten-arc
torch holder also caused many arrangements to be
tried before adequate concentric inert-gas shielding
was obtained.
As the machine is now used, complete control at
the front permits the operator to move through the
entire sequence to make all the weldments without
any change in process conditions. The header which
has been previously assembled is now held on the
machine with an air-actuated indexing chuck. The
prepositioned special gas tungsten-arc welding torch
is swung against a stop and the welding operation
can be started. Inert gas, cooling water, up and Fig. 6—Oxidizer-tank and ring-assembly welding machine
down slopes, welding current and speed are all auto-
matically controlled and sequenced on a panel at-
tached to a commercially available balanced-wave Fig. 7—Fuel-tank-assembly welding machine
300 amp a-c power supply.
Burst Bands
To initiate the propellant’s flow from the storage
tanks to the combustion chamber, a pressurizing solid-
propellant grain is energized. The pressure causes
the propellants to break soft-aluminum parts acting
as rupture diaphragms. The latest machine to se-
cure these burst bands to the engine structural parts
is shown in Fig. 5.
This is a two-spindle vertical machine, each spindle
being separately powered and controlled. Each
spindle has tooling designed to hold the parts to be
welded. In the event that different design parts are
forthcoming, tooling can be easily changed. Because
of the light section of the bands, a copper chill is
used.

1032 | OCTOBER 1960

 Miniature filler-wire feeders were developed and Contained in the head-stock end are the variable-
are attached to commercially available gas tungsten- speed rotation drive, a position-sensing overlap de-
arc holders. The one severe process problem that vice and most of the controls for actuating the opera-
was overcome was the difficulty of the reliable feeding tion sequences. An air-cylinder-operated live-center
of the filler wire into the weld puddle, with tolerable tailstock is used to clamp the parts in axial align-
deviation in placement. This was achieved by ment; no radial indexing is required. To assist in
utilizing wire guides with hypodermic stainless-steel loading the parts, a retractable support is supplied on
inserts and nylon liners. This assembly has the the bed section. Rotation cannot be started until
ability to direct accurately the wire, eliminate galling the support is released.
and adequately support the dead-soft filler wire after A standard gas metal-arc welding head with a 400-
it leaves the drive-roll stand. amp power supply completes the welding package.
Remote control of current from the two 300-amp A water-cooled chill ring on the tank head assists in
balanced-wave power supplies, other welding-process balancing the heat input to each side of the joint.
controls and all machine functions are available This chill ring is of two halves hinged at one point and
from the front of the machine, as can be seen from clamped diametrically opposite. To prevent the
Fig. 5. yo-yo effect of water hoses wrapping around the
Oxidizer Tank and Ring Assembly perimeter, a novel slip-ring assembly to prevent water
leakage was designed and built.
This weldment of a tank section and ring are gas
metal-arc welded on a specially built machine for the Fuel-tank Assembly
purpose (see Fig. 6). A lathe-type machine with The basic automatic machine, Fig. 7, used for
head-stock components, similar to other lathe ma- making this weldment is similar to that described for
chines for this job, was established as standard. the oxidizer assembly. Mechanically, the basic dif-
ference is that an expanding internal mandrel of a
unique design is actuated through the hollow head-
stock spindle. Since the mandrel is of a new de-
sign and may find many other applications, it is
worth while to describe in detail. Figure 8, parts a
and 6, represents schematics of the mandrel con-
struction and installation. This mandrel performs
a threefold function as follows: (1 It assists in the
loading and alignment of the piece parts. Because
the tank section has wide and thick axial ribs about a
portion of its internal periphery, the mandrel cannot
be expanded completely in one step. In use, the
tank section is loaded into the head-stock tooling, the
mandrel is partially expanded automatically, the
tank-head section set in place, the mandrel expanded
the rest of the way, and finally the tailstock air
Fig. 8—Expanding-mandrel design cylinder is actuated to axially clamp the parts.
(a) (Above) isometric view 2) It acts as a controlled chill on both the tank and
(b) (Below) schematic of installation on fuel-tank head sections. (3) It acts to help maintain a round
welding machine concentric assembly.

WELDING JOURNAL | 1033

 A double V-block affixed to the air-cylinder piston sign of the machine is similar to the other lathe-type
rod extends through the head-stock spindle. As machines previously described. However, the tool-
the piston rod is drawn toward the head end, the ing is different in that a steady-rest is employed to
V-block causes four copper segments to move both restrain the header rim so that concentricity is main-
axially and radially, the final position being in line tained. To facilitate movement of the automatic
with the joint, and the diameter matching the part. tungsten-arc head, a hand-powered side beam and
To insure that the sectors will always be in contact carriage is mounted between head and _tailstock
with the piece parts, each sector is separately spring ends of the machine.
loaded. As the parts grow with the welding heat The internal mandrel used for one of the joints is
input, each sector grows radially, but does not move similar to that previously mentioned except that it is
axially. appreciably smaller, and does not have spring-ac-
As on other machines utilizing gas tungsten-arc tuated radial motion since piece-part growth is not a
welding, a 300-amp balanced-wave power supply is problem. It is retractable through the head-stock
used in conjuction with an automatic tungsten-arc spindle so that parts can be easily loaded and un-
head designed to maintain constant arc voltage and a loaded.
filler-wire feeder. The controls for all welding and
mechanical operations are located convenient to the Baffle Subassemblies
operator. As mentioned earlier, the propellant is used as a
combustion chamber regenerative coolant. So that
Internal Subassembly uniformity of distribution is obtained, the baffle sur-
The internal subassembly consists of the header as- rounding the chamber is of a general corrugated-
sembly previously mentioned, the combustion cham- aluminum sheet-metal construction and is a light-
ber and a liner that eventually retains the solid- press fit over the chamber. It is welded to the
propellant pressurizing grain. The production re- headers manually in a rotating fixture similar to the
quirement is such that the two joints can be made lathes already described.
successively on one machine, and one automatic
welding head is shifted easily from one position to Center-closure Final Assembly
the other as the need requires. The subassemblies of fuel, oxidizer and internals
It will be noted from Fig. 9 that the mechanical de- are joined on this machine, Fig. 10, to make another

Fig. 9—Internal-subassembly welding machine

Fig. 10—Center-closure welding machine Fig. 1l—End-closure welding machine

subassembly which can be recognized as the final main part of the engine assembly. When the parts
engine envelope. Here again, this lathe utilizes a are loaded into the machine with the aid of loading
semistandard head-stock design. However, to per- supports, the parts must be indexed radially with re-
mit the use of the machine for a variety of present spect to each other. Radial alignment is obtained by
and future uses, the tail stock is both powered and the use of special head and tailstock tooling that ro-
adjustable along the bed ways. Power is applied to tate in unison. Backlash is held to an absolute
the tail-stock spindle through a chain-and-sprocket minimum.
arrangement. A tungsten-arc holder and filler-wire
feeder are mounted on a hand-powered side-beam End-closure Final Assembly
carriage. A 300-amp balanced-wave power supply The end closures are made on the same type of
furnishes welding power. basic machine except that the rotation axis is vertical
Prior to loading into the machine, the parts are and the gas metal-arc welding-head mount is on top
assembled with stainless-steel backing rings that re- of the machine, Fig. 11. Torch arrangement for

Table 1 Welding-process Conditions

Thickness Gas- Shielding- Welding Weld
Materials at joint, shielded No Shielding gas flow, current, speed, Weld
Weldment welded in. process Filler passes gas scfh amp ipm category
Header— 6066/1100 0.048 Tungsten-arc Preplaced 1 Argon 105
metering A43S Al
orifice ring
Burst band 6066/1100 Tungsten-arc A43S Argon
Oxidizer- 6066/2014 Metal-arc A43S 5 argon
tank 25 helium
assembly
Fuel-tank 6066 /6066 0.288/0.223 Tungsten-arc Argon 290/330
assembly
Internal 6066/6066 A —0.18 0 #Tungsten-arc Argon A—275
Sub- B —0.12 9 B—250
assembly
Center 6066 /6066 0.370 Tungsten-arc Argon 300/320
closure
End closure 6066/6066 Aft 0.509 Metal-arc Aft 3 75 argon 215
Fwd 0.540 Fwd 4 25 helium

WELDING JOURNAL | 1035

 Metering orifice to header Burst band to header Oxidizer tank to ring assembly

Fuel tank to head assembly Internal subassembly chamber to header Internal subassembly liner to header

Center closure End-closure chamber to fue! tank closure liner to oxidizer tank
Fig. 12—Section of welds obtained on Guardian I! engine

each end closure is provided in the head mount by Because of advances in welding that have per-
having two degrees of rotational freedom and two mitted applications to high production under closely
degrees of linear freedom. Special tooling, roughly controlled conditions, high-quality rocket engines
in the center of the vertical support column, con- with minimum rejects are being produced. It is
sists of a fixture that permits the entire assembly to significant that production management now regards
be inverted. welding as another step in the fabrication process
The joint design of the piece parts welded on the and have completely integrated it into the production
aforementioned machine, the process data for each cycle.
joint and other pertinent facts were established after The ability to produce a reliable packaged liquid
a prove-out period. This information is summarized thrust unit is a testimony to the intelligent coopera-
in Table 1. Sections of the weld are shown in Fig. tion that exists between the design and production
12. engineers of the fabricator and the supplier of the
integrated package consisting of welding equipment
Conclusion
and tooling.
While most of the public interest is centered on the
development of large ballistic missiles, the smaller, Acknowledgment
shorter range, work-horse guided missiles occupy an The authors wish to acknowledge the efforts of
equal if not greater importance to the defense effort. many individuals in both organizations who have
Significant developments in the propellant and contributed to the satisfactory solution of many of
thrust-unit field have opened up a new era in rocket the production-welding problems and to their col-
engines, employing storable-propellant systems. leagues who offered constructive criticism in the
The first of these systems is now in production. preparation of this presentation.

1036 | OCTOBER 1950

 —

Fig. 1—Welding '/, in. diam solid-fluxed hook studs in a compartment aboard the N.S.S. Savannah,
using two paralleled silicon-rectifier power sources stationed 200 ft away. Inset shows power sources on
main deck. (Courtesy of New York Shipbuilding Corp.)

ABSTRACT. A silicon-rectifier power source has been

Large-diameter studs can be developed specifically for stud welding. The unit is de-
signed for a maximum current demand of 1500 amp and
used satisfactorily, employing as much as can be paralleled quickly and easily for greater require-
ments. Large-diameter studs can be welded satisfac-
200 ft of welding cable by torily using as much as 200 ft of welding cable. Operat-
ing performance and reliability are excellent. In addi-
tion, the unit is small, compact and relatively lightweight.

Stud Welding Introduction

Historically, developments in stud welding have
been confined to the stud-welding gun, the stud and
with a Silicon-Rectifier controls. Little consideration was given to the
power source; a standard variable-voltage arc-

Power Source S. BAUM was formerly associated with the Philadelphia Naval Shipyard
He is now Consultant Welding Engineer, New York, N. Y
Paper presented at the AWS National Fall Meeting held in Pittsburgh,
BY S. BAUM Pa., Sept. 26-29, 1960

WELDING JOURNAL i 1037

 tests that were conducted to develop and determine
the design, performance and reliability of the sili-
con-rectifier stud-welding power source under ex-
amination.
Description of Stud-welding Power Source
This equipment is a transformer-rectifier machine
bag 0 gts Nyege be gst tigate’ ih Ne alee”) converting three-phase alternating current to direct
SSTERMV. current. It consists essentially of a three-phase
transformer, resistor-selector bank, three-phase full-
wave silicon-rectifier bridge, circuit breaker and
ventilating fan. It has an input voltage rating of
TIME DIRECTION 230/460 v, 3 phase, 60 cycles and a secondary
potential of 100 v, direct current, open circuit.
The unit is nominally rated 1000 amp at 10% duty
cycle. The machine has a maximum output of 1500
Fig. 2—Oscillogram of electrical characteristics taken in amp. Current selection is made by appropriately
welding */,-in. diam studs on 10.2 Ib mild-steel plate in the setting the tap switch. The physical measure-
vertical position with 200 feet of stud-welding-gun cable ments are 23- by 23-in. base, 40-in. height and a
weight of about 450 lb. A view of the machine can
be seen in Fig. 1.
welding machine being commonly utilized. How
ever, the increasing use of large-diameter studs General Test Procedure
in. diam and greater) in numerous critical ap- Stud welding was performed in the vertical posi-
plications requiring welds of consistently high quality tion (workpiece in vertical plane—stud horizontal
and reliability, focused attention on the need for a with the exception of the 1',,-in. diam stud, which
more satisfactory power source. As a consequence, was welded in the flat position. The vertical posi-
a silicon-rectifier machine has been designed par- tion was selected in conducting the tests since it was
ticularly for stud welding. This article deals with considered to represent the most difficult condition.
the investigation of the electrical characteristics Straight-shank carbon-steel studs, with a solid
under various operating conditions and the welding aluminum-flux insert, were used. The nominal

MACH. NO. 1587 MACH.NO. 11573

Fig. 3—Oscillograms of electrical characteristics of two paralleled units taken in
welding in. diam studs on 10.2 Ib mild-steel plate in the vertical
position with 200 ft of stud-welding-gun cable

10388 '| OCTOBER 1960

diameters ranged from '/, to */, in., inclusive, and Paralleling Characteristics
1'/, in. (*/;¢- to °/s- and 1-in. weld ends). The arc Recent developments in stud-welding technology
shields were of the parallel groove-vent serration have resulted in the application of ever increasing
design. Both studs and arc shields conformed to stud sizes. The power requirements involved in
the requirements of Military Specification MIL-S- welding large-diameter studs are considerable; cur-
18109C. The test plates were 10.2 lb mild steel and rents in excess of 2000 amp are not unusual. Thus,
40.8 lb high-tensile steel with the surface ground there exists justification for the design and manu-
clean in the location of welding. facture of ultra-large-capacity machines. Econom-
The testing program was designed to develop com- ically, the utility of such equipment would be
plete information, within practical limits, of the rather restricted since the dollar cost per ampere
operation and performance of the _ transformer- would be very high when operating in the lower
rectifier stud-welding machine. To effect this ob- ranges. It is, therefore, evident that the feasibility
jective, the following tests and inspections were of welding large-diameter studs in many cases, par-
accomplished: ticularly where the number is very limited, requires
1. Establishment of the dynamic-electrical char- the availability of comparatively low-cost standard
acteristics over the load range. equipment of specific design which can be paralleled
2. Determination of the load-distribution pattern readily and satisfactorily to supply the necessary
when operating machines in parallel. power.
3. Evaluation of the welding performance. The paralleling characteristics of the subject
4. Determination of the influence of varying weld- transformer rectifier machine were investigated by
ing-cable length upon electrical characteristics, operating two units in parallel and welding */;-in.
power requirements and welding performance. studs in the manner described previously. The
5. Job testing of the machine on indoor and out- output voltage and current of each machine
door locations. were recorded by switching the oscillograph from
the terminals of one to the other during weld-
Dynamic-electrical Characteristics ing. The oscillographic traces are shown in Fig. 3.
In order to develop and define the dynamic elec- As can be observed from the oscillograms, the load
trical characteristics over the designed range, the division between the two machines is practically
machine was loaded by welding various size studs, equal. The character of the oscillogram taken of
; to */, in. diam, inclusive. This work was ac- each machine for the complete welding cycle, did
complished on vertically positioned 10.2 lb mild- not appear to be at all affected by the parallel
steel plates using 200 ft of welding cable (50 ft of operation. Excellent-quality stud welds were pro-
No. 1 and 150 ft of 2/0 cable). A continuous record duced, as noted subsequently.
was made with a film-recording oscillograph of cur-
rent and voltage during the welding of each size Welding Performance
stud. The evaluation of the welding performance of
Figure 2 is the oscillogram taken in the welding the machine, for both single and parallel operation,
of a in. diam stud. This oscillogram is con- was made by end welding !/,- to */,;-in. diam studs,
sidered representative of the series made. An inclusive, on 40.8-lb high-tensile steel plates posi-
analysis of the current and voltage traces indicates tioned vertically. This welding was done with 200
the following interesting and unique features of the ft of welding cable (50 ft of No. 1 and 150 ft of 2/0
design: cable) and straight polarity, direct current. The
welds were examined both destructively and non-
a) Response to total demand is almost instanta-
destructively as follows:
neous current reaches maximum in about one
a) Appearance. The visual inspection of all
cycle.
the stud welds made for the accomplishment of the
The inherent reactance inhibits surges and
destructive tests showed the fillets to be complete,
assists in stabilizing both voltage and current
free of undercut and of good contour. Very little
in minimum time.
to a moderate amount of spatter was observed.
Voltage and current are practically constant
b) Bend Test. The test assemblies were pre-
throughout the welding cycle. —
pared by welding 10 studs, in each of the above
The current falls sharply to zero upon in-
stated sizes, on a 12-in. square piece of 40.8-lb
terruption.
high-tensile steel plate. Figure 4 shows the arrange-
Theoretically, the shape of the electrical curves ment of the studs; the outside studs are located
exhibited in the oscillogram is ideal for making sound 2 in. in from the plate edge. The sequence and the
stud welds since it is conducive to the production and time interval between welds were determined so as
maintenance of arc stability. This was confirmed to control and minimize any increase in over-all
by taking high-speed motion pictures of the arc at plate temperature.
approximately 4000 frames per second and the re- The welded studs were hammered flat against
sults of the examination of the stud welds made in the plate, as can be seen in Fig. 4. Each of the studs
the performance test. was bent in a different direction, thereby loading

WELDING JOURNAL | 1039

 */-in. diam stud; single machine 3/,-in. diam stud; two paralleled machines
Fig. 4—Bend specimens of studs welded on 40.8 |b high-tensile steel plate in the
vertical position using 200 ft of stud-weiding-gun cable

and testing the weld completely and uniformly

around its perimeter. All of the welded studs bent
the full 90 deg without fracture.
(c) Macroinspection. The test assemblies were
prepared by welding three studs, in each of the given
sizes, on a 3- x 12-in. piece of 40.8-lb high-tensile
steel plate. The studs were located on the center-
line of the plate, with one in the middle and the
other two spaced 4 in. away, on each side. The
plate temperature was maintained at ambient dur-
ing welding.
The welded studs were sectioned, polished and
/s-in. diam etched. Representative macroetched specimens of
the larger size studs are shown in Fig. 5. The macro-
inspection of the welds indicated complete fusion
and the absence of any physical discontinuities,
such as cracks or undercutting. Porosity was ob-
served in only one instance—the *,,-in. diam stud
weld made with one power source.
In addition to the preceding, about fifty 1! ,-in.
diam studs were welded in the flat position
with two paralleled machines. This welding
was carried out under shop production con-
ditions. Figure 6 shows one of the welded studs
together with an etched section of the weld. As can
be noted, the quality of the weld appears to be quite
*/,-in. diam; 3/,-in. diam; good, with only some scattered porosity present.
single machine two paralleled machines (d) Tensile Test. The test assemblies were pre-
Fig. 5—Macroetched specimens of welds made with various pared by welding on the center of both faces of a 3-in.
size studs on 40.8 Ib high-tensile steel plate in the vertical square piece of 40.8 lb high-tensile steel plate.
position with 200 ft of stud-welding-gun cable Three assemblies were welded for each of the
various-size studs used.
The welded assemblies were placed in a tensile-
test machine with the grips fastened to the studs and
loaded to failure. The nature and location of the

1040 | OCTOBER 1960

 \\\\
\\ \
’ _ \\

mSek
i.
iivinnnanetinny:

\"
\\
\\
\"
\ vveedidenentttite
Fig. 7—Tensile specimens—type and location of fracture of
-in. diam studs welded on 40.8 Ib high-tensile steel plate
in the vertical position using two paralleled machines

fractures are shown, typically, in Fig. 7. It

was found that the fracture invariably occurred in
the reduced section of the stud, some distance away
Fig. 6—1'/,-in. diam stud welded on 40.8 Ib high-tensile steel from the weld, and in a ductile manner.
plate in the flat position using two paralleled machines
Effect of Welding-cable Length
The resistance of the welding circuit is a critical
factor in the production of acceptable stud welds.

50 ft of stud-welding-gun cable 100 ft of stud-welding-gun c

150 ft of stud-welding-gun cable 200 ft of stud-welding-gun cable

Fig. 8—Oscillograms of electrical characteristics taken in welding
5/.-in. diam studs on 40.8 Ib high-tensile steel plate in the vertical position,
using various lengths of stud-welding-gun cable

WELDING JOURNAL 1041

 Any resistance that causes a significant decrease in instance, show no variation under load except for
the current and an unduly large line drop, all other magnitude. The arc time had to be increased, as
conditions being equal, will make stud welding dif- increments of welding cable were added, in order to
ficult if not impossible. Since it is common practice obtain satisfactory stud welds. The current was
to add cable as required, the length of the welding increasingly affected by increased length of welding
cable may be considerable. The capability of the cable; the percent decrease being doubled with each
power source to compensate for increased external addition of 50 ft of 2/0 cable. However, it is sig-
resistance, in order to assure satisfactory welding nificant that the terminal volts increased proportion-
performance, is significant and must, of necessity, be ately more than the decrease in welding current
carefully evaluated. (2.33 times as much). The energy required to weld
The effect of the resistance of the welding curcuit the °/s-in. diam stud with 200 ft of welding cable was
upon machine performance was studied by welding almost twice that required with 50 ft.
first with 50 ft of No. 1 stud-welding-gun cable The stud welds made in obtaining the oscillograms
furnished with the machine) and then successively in Fig. 8 were sectioned, polished and etched.
adding 50 ft of 2/0 cable until the total length was Photomacrographs of the sections are shown in
250 ft. The welding was done with °/;-in. diam Fig. 9. The weld quality (good fillet formation,
studs on 40.8-lb high-tensile steel plate positioned good fusion, no unacceptable defects) exhibited is
vertically. The machine setting as well as the other satisfactory regardless of the length of welding cable;
welding conditions, with the exception of arc time, in fact, the quality of the weld made with 200 ft
were identical for the different lengths of welding of welding cable is comparable to that obtained with
cable. As it was found that welds of acceptable 50 ft.
quality could not be made: consistently with 250 ft
of welding cable, the oscillographic study was limited Job Testing
to 200 ft. The final phase of the evaluation was to turn the
The electrical characteristics of the machine, machine over to shop personnel for on-the-job opera-
when welding with each of the four different lengths tion. This covered a period of four months. The
of stud-welding-gun cable, were recorded with the machine was located both indoors and out. It was
oscillograph and are shown in Fig. 8. It should be operated under extreme weather conditions, such as
noted that the electrical characteristics, in every the hot summer sun and rainstorms. A wide range
of stud sizes and cable lengths was used as, for ex-
NN ample, '/,-in. diam studs were welded to a steel-
maeena, ated>
+ framed structure, for attaching siding, with welding-
cable lengths of up to 250 ft; °/s-in. diam studs were
welded to manhole rings; * ;-in. diam studs were
welded to tank tops (floating drydock), for securing
the covers over the openings, using two paralleled
machines.
Throughout this period a continual check was
kept on the operation of the equipment and the
quality of the welds that were made. The equip-
ment functioned smoothly and reliably. Inspec-
tion of the components, at the conclusion of the field
test, did not reveal any discernible deterioration.
50 ft of cable 100 ft of cable
The welds were found to be consistently acceptable.

Conclusion
The performance of the _ transformer-silicon-
rectifier power source for stud welding has resulted
in wide acceptance of this specialized machine.
In turn, this has generated improvements and re-
finements in the design leading to a reduction in the
ripple in the secondary characteristics and an in-
crease in the number of current selections. Addi-
tionally, circuits have been modified for maximum
equipment reliability.
Stud welding has been finding an ever increasing
number of applications in recent years. The avail-
150 ft of cable 200 ft of cable
ability of a high-performance compact power
Fig. 3—Macroetched specimens of welds made with 5/,-in. source should spur the rate of growth of stud welding,
diam studs on 40.8 Ib high-tensile steel plate in the vertical particularly in critical applications where failures
position using various lengths of stud-weiding-gun cable
cannot be tolerated.

10422 | OCTOBER 1960

Tops in the Industry!

The Welding Journal is the

world’s most authoritative

welding and allied process

magazine; it has fully 2'/

times the editorial content

of any competing magazine;

it.is unequalled in coverage

of welding engineering,

research and application.

If you have a product

for the welding or allied

industries, the Journal's

pages will reach your market!

WELDING JOURNAL | 1043

 Welding of clips to generator coils by means of short-arc process

Acceptance in automotive industry and in field of general metal fabrication

points to the growing importance of

Short-arc Consumable-Electrode Welding

Applications and New Developments

oY t. McELRATH

The short-arc consumable-electrode process-—or more At present, short-arc welding is finding unusual
simply, short-arc welding—-has emerged from the acceptance in the automotive industry because of its
haze that always surrounds new developments. ability to meet high-speed operating requirements
Its elements are now understood. As a result, use economically. Secondly, the process is being utilized
of the process has been accompanied by remarkable in the field of general fabrication for a variety of
successes. In some instances, new fields for electric products. Finally, short-arc welding is proving to
welding have opened up in that the short-arc proc- be unusually useful for the welding of missile and
ess has proved to be the only electric-welding method other special materials.
that could be applied. In other instances, the intro-
Development of Short-arc Welding
duction of short-arc welding has led to substantial
operating economies. Ever since its introduction to industry in 1948,
investigators have sought to use the conventional
lr. McELRATH is Special Project Engineer at the Development Labora- inert-gas consumable-electrode process with low-
tory of the Linde Co., Division of Union Carbide Corp., Newark, N. J
heat input suitable for joining light-gage materials in
Paper presented at AWS 4lst Annual Meeting held in Los Angeles,
Calif., April 25-29, 1960 all positions and for the vertical and overhead weld-

1044 | OCTOBER 1960

 ing of heavier sections. Invariably, however, weld- mately, than carbon dioxide in many instances.
ing torches have been too heavy while welding-wire Third and finally, the very important ingredient
diameters have been too large and available shield- of power-supply control became recognized and sup-
ing-gas mixtures have produced unsightly beads or plied the last link in the chain of developments lead-
excessive spatter as well as low quality welds. ing to short-arc welding as known today. Without
Finally, power supplies have been harsh and un- probing the depths of power-supply design, it is
controllable, thereby creating either too much cur- sufficient to say that the amount and rate of current
rent surge or virtually none at all. surge are the keys to successful operation. Once
All three of these difficulties were corrected during they are properly selected for the application, the
the development of short-arc welding. First came a arc is “‘tuned”’ to stability and operates with un-
lightweight, 1-lb curved torch coupled with a per- usual consistency.
fected drive system for small wires; this system pro- These three important factors can be easily re-
vides the ultimate in operating ease and convenience membered if we borrow a section of the alphabet and
for reaching the joint. Next, an intensive investiga- refer to the E, F, G of short-arc welding. ‘‘E’’ is
tion of all usable gases and mixtures led to the de- for the electrical ““know-how” required to obtain the
velopment and availability of a mixture consisting of necessary arc with the appropriate power supply;
75% argon and 25% carbon dioxide. This unique “F”’ is for the wire feeding and torch equipment;
combination, tailor-made for short-arc welding, com- and “‘G”’ is for the shielding gas mixture.
bines the advantages of both gases and eliminates
their most undesirable characteristics. While not Short-arc-process Characteristics
usable in all cases, this mixture is a ‘“‘must”’ for car-
Arc, Current and Power Supply
bun-steel applications, being no more expensive, ulti-
Short-arec welding employs low currents ranging
from 20 to 175 amp, low voltages of 12 to 20 v, and
small-diameter wires with 0.030 in. diam being a
popular size. Its outstanding characteristic is the
frequent shorting of the wire to the work. All ma-
terial transfer takes place at arc outages which occur
at a steady rate and which can vary between 20 and
200 times a second. The net result is a very stable
arc of low energy and heat input ideally suited for
welding light materials in all positions and heavier
materials in the vertical and overhead positions,
filling large gaps where encountered. The low heat
VOLTAGE REFERENCE input minimizes distortion and metallurgical effects.
A typical oscillograph tells the story of arc out-
| ages. Figure 1 illustrates arc voltage and arc cur-
60 rent during a typical welding cycle. The voltage
SEC.
averages 16 v but drops to zero about 65 times per
Fig. 1—Oscillograph showing short-arc voltage and second as measured by the 60-cycle time lines. Each
current characteristics
outage should produce a predictable and controlled
current surge sufficient to recreate the arc without
SHORT CIRCUIT PERIOD— an undesirable high surge or blast. Figure 2
NSUMABLE [4 illustrates a complete cycle, showing short circuit and
ELECTRODE | )
f3 metal transfer followed by reestablishment of the arc.
> -~ae | y The cycle is completed as the wire touches the puddle
‘ ( ‘
Le | ) D again.
The short-circuiting characteristic of short-arc
welding cannot be obtained for the full range of
oe ~ARCING PERIOD operation with ordinary power-supply units. Newer
| fr
} ARC | 4 a) machines of _ several
IK Lina? vale). | = ARC MATERIAL makes contain slope and
— s Lf } ?
reactance adjustments
< , —_ - — ; $—— suitable for producing
the predictable and con-
IMMEDIATELY AFTER trolled current surges
EXTINCTION PERIOD
at ARC needed to implement
NOTE- } \ EXTINCTION short-arc welding over
DEALIZED SKETCHES i yy
NOT TO SCALE a portions of the range in
which the process is now
being applied. Figure
Fig. 2—Complete short-arc cycle 3 illustrates the newest

Fig. 3—Power supply for

short-arc welding

WELDING JOURNAL | 1045

 S
Me COST VS. METAL THICKNESS
°o | = |
- 0 T
S WIRE, GAS
AN
EY hei ol |
gi MIXTURE—.
a ‘ -_
2 / q--""
z 6Fr
WwW + / + + + + +
© <CO2 ENDS AT Q050 IN. CO2—
Ls
So
8 025 050 075 JOO 25 4.50
METAL THICKNESS (IN)
CARBON STEEL
Fig. 4—Cost of short-arc welding

machine for this purpose, and the unit shown will

cover the entire range of short-arc welding. Thus
as may be seen, the “E”’ or electrical “‘“know-how”
as built into the power supply is the first essential in-
gredient of the process.
The next or second consideration is the torch and
wire-feeding equipment, referred to as the “F” of
short-arc welding. It is only necessary at this point Fle, §~Manual shert-ese webdine: eotup ter
to refer to the application illustrations to indicate rebuilding automotive crankshafts
the proper type of equipment for welding thin ma-
terials. Finally, the third factor or the shielding
Fig. 6—Oil holes in crankshaft welded right to the edge
gas referred to as the “G”’ of short-arc welding is without the need for filling hole with carbon
discussed immediately below.
Shielding Gas
All shielding gases reacted favorably to controlled
power supplies for short-arc welding. Pure argon
and helium, or mixtures thereof, are used on thin
aluminum, and argon-oxygen mixtures are employed
for certain alloys. For carbon and low-alloy steels,
however, the 25°% carbon dioxide and 75% argon
mixture has become unique. It provides better
wetting action than argon-oxygen mixtures and pro-
vides a smoother arc with less spatter than carbon
dioxide. High-speed films show the basic reason for
the superiority of this mixture over carbon dioxide
used alone. They show that the arc is always re-
created when the mixture is used. A molten elec-
trode end advances toward the puddle, the short
circuits are soft, and metal transfer is easy and com-
plete. With carbon dioxide alone, however, the arc
is seldom recreated after an arc outage, allowing a
relatively cold, unheated electrode to advance and
stub into the work. This creates the spatter and arc
instability which is observed in this type of welding.
In addition, the mixture as applied in sheet-metal
welding eliminates the economic advantage of carbon
dioxide. Figure 4 shows the cost of consumables to
be no higher for the mixture as compared to carbon
dioxide for thicknesses up to and including */;2 in.
This fact stems from the low deposition of filler wire
per foot of weld which just about offsets the dif-
ference in the cost per foot of gas.
Automotive Applications
Automotive applications are only one area where
short-are welding is being used successfully. Other

1046 | OCTOBER 1960

Fig. 7—Parts of engine heater prior to assembly

areas are in general fabrication and for missiles and

special materials. In each instance, the three factors
E, F and G) of short-arc welding as previously dis-
cussed have constituted the basis for its successful
use.
The first application of concern here is the fabrica-
tion of automotive window frames requiring 2500
welds per hour. Each weld is in. long, and three
are required for each frame. Oxyacetylene welding
previously distorted the frame, while covered-elec-
trode welding required subsequent clean-up and
frequently burned through the 16- and 18-gage
frame members. Short-arc welding was adopted as
a production-welding method over a year ago. With
this process, skilled women operators have since
averaged 30 welds per minute. A weld timer en-
sures a uniform length of weld. The operation is
carried out without any waste of electrodes (no stub
ends) and without smoke or fumes.
Interest is very high in the short-arc process as a
replacement for metal spraying in the rebuilding of
automotive crank shafts. The short-arc torch can
either be hand-held or machine-mounted, and weld-
ing proceeds without prior grinding. Each journal
is completed in 3 min. Figure 5 shows a manual
setup, which allows the operator to manipulate the Fig. 9—Mechanized short-arc welding of
torch so as to avoid the unfilled oil holes shown in lap joint on 20-gage engine heater
Fig. 6. Formerly, the shaft had to be moved from
the spray machine to the grinder. Now welding and
Fig. 10—Short-arc welding of nose cone for practice bomb
grinding are completed on the same machine, thereby
eliminating costly handling.
The assembly of automotive and truck-engine
heaters requires two girth welds between a 20-gage
shell and 18-gage caps. Figure 7 shows the parts
prior to assembly, and Fig. 8 illustrates the turning
fixture. The first weld is being made at a speed
of 25 ipm using 65 amp and 11 cfh of the
25% carbon dioxide and 75°, argon mixture.
Figure 9 shows the final weld being applied. The
assembly is painted without the need of a cleanup
operation. Oxyacetylene welding and silver solder-
ing were formerly employed. Both were slow, while
conventional arc-welding processes had always
proved unsatisfactory until the efficient low-heat
short-arc method was introduced.
 argon mixture. Prior to the introduction of short-
arc welding, the tungsten-arc process was used and
resulted in welds which sometimes cracked in the heat-
affected zone during forming. With the short-arc
process, rejects have been less than 1°% and welding
speeds are three times faster. At present, other oper-
ations in the plant where the above application is in
effect are being converted over to short-arc welding.
Steam generators, commonly called flash boilers,
contain pancake-type coils made by winding 16-gage-
thick steel tubes into a flat-based jig. Figure 12
shows such a unit mounted in position for fastening
of the straps. Here can be seen a close-up of the
strap and of the curved torch. This photograph was
taken between welds. The light created by the
preceding weld enables the operator to make a sure
start on the next joint without raising his helmet.
Both oxyacetylene and tungsten-arc welding were
replaced on this application by the short-arc process
Clips are next welded to hold a bank of coils in posi-
tion, as shown in the lead photograph. The welding
current for both welds is 70 amp at 17v. Mixing of
Fig. 1l—Welded nose cone the 25° carbon dioxide and 75°; argon is done at
each station. Both former processes required a high
General Fabrication degree of skill, and leaks and burn-throughs were
Although short-are welding is important for auto- frequent. Short-arc welding is twice as fast and pro-
motive applications, it has been utilized to a greater duces both better-appearing and stronger welds.
extent for other fabrication. Figure 10 shows the nose In this particular application, four machines have
cone of a practice bomb made from 20-gage steel. A saved $32,000 per year per 8-hr shift, and the plant
straight butt joint is welded at 16 ipm with 60 amp, has doubled its production with no increase in space.
17 v. Figure 11 shows the completed nose after the Blower housings made of 0.100-in. thick steel re-
press-forming operation resulting in 18% elonga- quire fillet welds for interior attachments. In this
tion. Shielding for this application was affected application, a curved torch is used to advantage in
through the use of 25°; carbon dioxide and 75“; reaching into the assembly to make the weld.

Fig. 12—Short-arc welding applied to pancake-type coils for steam generator

Short-arc welding is twice as fast as covered-elec-
trode welding due to the elimination of flux removal
and cleaning. A fan housing of 16-gage steel re-
quires corner welds which are also easily made with
the short-arc process.
Application of short-arc welding to pipe welding
has opened up an entirely new field for this new proc-
ess. Succeeding first on ordinary butt welds as
shown in Fig. 13, the use of short-arc welding has
spread to the welding of flanges, saddles, reducers
and elbows. The short-arc welding torch is easily
handled in position for pipe-line work, and it elimi-
nates costly handling in shops. Short-arc welding is
applied in one-fourth to one-half the time required
for covered-electrode welding and provides a perfect
inner bead with standard edge preparations. As a
further dividend, welded saddles as well as flanges
lack the distortion usually encountered with covered-
electrode welding, thus eliminating a great deal of
straightening. Fig. 14—Short-arc welding
of aluminum shipping container
Aluminum as well as steel responds to the short-
arc process, although no carbon dioxide is employed
Power-supply control is important, and proper appli-
cation allows welds in 0.030-in. thick aluminum with
currents as low as 25amp. In Fig. 14, shipping con-
tainers of 0.090-in. thick 6061 aluminum are being
joined with 0.030-in. wire at 100 amp. Manual
tungsten-arc welding was replaced in these opera-
tions, and production time has been reduced by al-
most 70°;; welding time was reduced from 20 to 6
min.
Missile and Special Materials
The final group of applications involves missiles
and special materials. One application involves the
welding of a ;-in. thick carbon-steel missile-han-
Fig. 15—Sergeant missile as fabricated with short-arc process
*.

dling truck frame. The assembly, requiring butt and

lap welds, is fixtured and tacked with the short-arc
process, removed from the fixture, and completely
welded with short arc. The current of 165 amp
represents the upper limit for 0.030-in. wire. Short-
arc welding on this application showed a 50% saving
over covered-electrode welding.
An outstanding application of short-arc welding is
the production of the Sergeant missile made of
0.109-in. thick 4130 steel. This application brought
to light an unexpected advantage of short arc.
Originally, a spray are (conventional inert-gas
consumable-electrode welding) was utilized with this
fixture but never produced consistent penetration
due to arc blow. With a short arc, however, the
very first weld was uniformly penetrated and
production was soon under way. Figure 15 shows
the over-all make-up of the missile consisting of three
longitudinal sections welded with the short-arc
process and two head assemblies requiring four
girth welds made with an “‘inside-out”’ technique.
Fig. 13—Short-arc weld in pipe In this application, after preheating at 450° F,

WELDING JOURNAL | 1049

 welding progresses at 25 ipm using */,-in. diam plant for various stainless-steel assemblies on a
wire, 150 amp, 16.5 v, the 25% carbon dioxide and production basis. Using a 2% oxygen and 98%
75° argon mixture for shielding, and pure argon argon mixture with 0.030-in. diam wire at currents
for backing. As shown in Fig. 16, a longitudinal ranging from 100 to 165 amp, injector rings, mani-
welding machine provides unusually accurate and folds and flanges have been welded to engine cham-
firm clamping as welding progresses on the inside of bers. The low heat input of the short-arc process
the cylinder. Girth welds are also made from the has cut welding down on various jobs as follows:
interior, using the new “inside-out” technique.
Every weld exhibits unusually uniform penetration Tungsten-arc Short-arc
as shown in Fig. 17. Figure 18 illustrates the cross hours hours’
section of the weld produced in a joint prepared 7 0.5
with a 100-deg included angle and * /,,-in. nose. 12 3
12 a
Short-arc welding has been evaluated at another 10 ol.
3.0' 5.
* Inert-gas consumable-electrode spray arc.
+ Average 86°% less time
The average time for short-arc welding is only 14%
of that formerly required due primarily to the
elimination of cooling periods. In addition, welds
have been sounder, with less distortion.
Another aircraft company has found that the
quality of short-arc welds in 16-gage PH15~-7
Mo stainless steel made with 0.030-in. alloy wire and
a shielding gas consisting of 99°% argon and 1°;
a oxygen is superior to that of tungsten-arc welds.
Special attention was required to provide the
proper power-supply response and backing gas flow.
Welding conditions, however, were not critical using
60 to 80 amp and 16 to 19 v. For this company, the
great importance of short-arc welding has been the
i

Fig. 16—Use of longitudinal welding machine to fabricate reduction of distortion by a factor of 2 and the
missile with short-arc process width of the heat-affected zone by a factor of 3.
Laboratory work is continuing to establish the
range and scope of short-arc welding. Larger wire
sizes, different wire and base materials and new
power-supply designs are being evaluated in many
locations. Notable among these new developments
has been the success in welding age-hardenable
nickel-base alloys.
Summary
Short-arc welding utilizes a new type of power-
supply tunable to the new short-circuiting arc, a
lightweight curved torch and a wire feeder perfected
for small wires, and a 75°% argon and 25° carbon
dioxide mixture. The unique combination of these
-7 4 three factors—the E, F, G of short-arc welding
produces low heat input and a small, easily controlled
Fig. 17—Weld penetration is uniform in missile weld puddle.
case made from 0.109-in. thick 4130 steel Use of the short-arc process has improved produc-
tion of automotive parts by eliminating cleaning,
smoke and spatter and stub-end loss. It has
produced welds that stretch better, are stronger and
do not burn through. It drastically reduces distor-
tion in pipe welding and aircraft components and
eliminates long cooling periods. The process
operates with a minimum of arc blow.
Now that it is possible to control a consumable-
electrode arc at low currents, a great number of
Fig. 18—Cross section of short-arc weld in advantages have become apparent and industry is
joint prepared with 100-deg included angle, */,,-in. nose rapidly finding production uses for this new tool.

1050 | OCTOBER 1960

 33 WEST 39TH STREET NEW YORK 18, NEW YORK

A NON-PROFIT ORGANIZATION FOUNDED IN 1919 FOR ADVANCING THE SCIENCE AND THE ART OF WELDING

FRED L. PLUMMER PENNSYLVANIA

NATIONAL SECRETARY
6-9220

October 1, 1960

AN INVITATION TO AUTHORS
—to participate in the American Welding Society’s
1961 AWS National Fall Meeting
Dallas, Tex., September 25-28, 1961
Gentlemen:

The American Welding Society will hold its 1961 National Fall Meeting in Dallas, Tex., on September 25-28, 1961.
Each year our Society offers opportunities to Authors for bringing their outstanding work, development and research
to the attention of our Membership and the welding and metals industry, by having previously unpresented and unpub-
lished papers presented at its national meetings.
The Society's Technical Papers Committee will be happy to receive your application for entry in the 1961 National Fall
Meeting activity. All applications, abstracts and manuscripts will be screened by the Committee, and Authors will be
notified sometime in March 1961 regarding acceptance.
Each abstract should be sufficiently descriptive to give the Committee a clear idea of the content of the proposed paper.
In any case, it must contain not less than 500—but preferably not more than 1000—words. Also, in order to place the
Committee in the best possible position to evaluate these papers, it is suggested that each abstract be accompanied by a
complete manuscript.
The Committee reserves the right to consider all applications on the basis of acceptance for placing on the 1961 National
Fall Meeting program, or consideration for placing on the 43rd Annual Meeting program, the next national meeting of
the Society. Papers may be considered for publication in the Welding Journal regardless of acceptance for presenta-
tion at either Meeting.
Papers dealing with latest developments in (1) equipment and pipe lines used in petroleum industry, (2) pressure vessels
and storage tanks, (3) design and fabrication of all types of weldments, including machinery, (4) welding of aircraft
and rockets, (5) fabrication and maintenance of equipment used for radioactive applications, (6) structures, (7) auto-
mation as applied to welding processes, (8) resistance spot, seam and projection welding, (9) gas-shielded arc welding,
(10) new processes, (11) welding of castings and composite structures, (12) welding of “new” alloys, (13) weldability of
high-strength steels, (14) welding of aluminum, magnesium, zirconium, titanium, molybdenum and like metals, (15) brazing,
(16) maintenance, (17) surfacing, (18) soft soldering, (19) adhesive bonding, (20) welding of plastics, and (21) prac-
tical applications or “how-to-do” topics are deemed to be of particular interest at the Dallas meeting. However,
any papers dealing with the educational and informative categories of welding production, engineering, research and
metallurgy, are welcomed as long as the subject falls within the field of our Society's activities.
Please fill out the Author's Application Form on reverse side of this letter, attach abstract thereto, and return to AWS.
To assure consideration for the 1961 National Fall Meeting program, your abstract must reach AWS not later than
January 15, 1961.
Sincerely,

Jeet
0 Fathom

Fred L. Plummer
Notional Secretary

IMPORTANT: ABSTRACTS CONTAINING LESS THAN 500 WORDS AND THOSE POSTMARKED LATER THAN
JANUARY 15, 1961 WILL NOT BE CONSIDERED.
 AUTHOR’S APPLICATION FORM
DEADLINE
FOR
JANUARY 15
1961 AWS NATIONAL FALL MEETING
1961
SEPTEMBER 25-28, 1961
Complete in Full and Return to American Welding Society, Inc., 33 West 39th Street, New York 18, N. Y.

Date Mailed
Author's Name...
Company or Organization.
Title or Position..
Mailing Address....
If there are to be Joint Name..
Address
Authorships, give name(s)
Name...
of other author(s) Adideais.

PROPOSED TITLE (less than 10 words):—___

SUBJECT CLASSIFICATION:
e Classify your paper by placing check mark in appropriate box:
Industrial Application [_] Engineering or Design Data [_} Process Development [_} Shop Practice [_]
Original Research [} Education [] Other [} an a - -
(state which)
ABSTRACT:
e Type, single spaced, an abstract of not less than 500—but preferably not more than 1000—words on separate sheet
and attach to this form. Be sure to give sufficient information to enable Technical Papers Committee to obtain a
clear idea of content of proposed paper.
@ If complete manuscript is available, in addition to abstract, please attach copy hereto.
¢ Application Form and Abstract must reach AWS Headquarters not later than January 15, 1961, to assure considera-
tion for 1961 National Fall Meeting.

MANUSCRIPT DEADLINES:
@ All manuscripts
must be in the hands of the Technical Papers Committee not later than July 15, 1961. If received
prior to May 15, 1961, every effort will be made to publish them in advance of meeting.
@ It is expected that the Committee’s selections will be announced sometime in March, 1961.
e If your paper is made a part of the program, which of the following manuscript deadlines will you be able to meet?
May 15, 1961 [_] June 15, 1961 [] July 15, 1961 []
e If manuscript is attached hereto, please check here [_]

PRESENTATION AND PUBLICATION OF PAPERS:

e Has material in this paper been previously presented in meeting or published?
Yes [ } No [_] When? Where? - cineccmaneiatademiin
¢ Following presentation at the Society's 1961 National Fall Meeting, would you accept invitation to present this paper
before AWS Sections? Yes [ ] No []
© Papers accepted for presentation become the property of the Society, with original publication rights assigned to
the Welding Journal.

RETURN TO AWS HEADQUARTERS. MUST BE RECEIVED

NOT LATER THAN JANUARY 15, 1961, TO ASSURE
CONSIDERATION.
aaa” MUST CONTAIN NOT LESS THAN 500 Author's Signature

FOOSE
EOS
SEES
SESSE
SSE
EEEEE
EEE
SEE
SEEE
ESE
EEE
OOO
Ee e
 Practical Welder

and Designer

Welded Engineering Center

Features Space-Saving,

Economical Design

BY SAMUEL C. BAST, PHILIP DREIER

AND CHARLES M. DICK,
Jr. —_ =
Se ee « >»
a“
mr: | —=
Welded construction was the key to economically |, Le /
maximizing usable floor space while achieving ah: reas
attractive design in the United Engineering Center a
now under construction in New York. : Wii |
rT . . e ,
Unlike many welded structures which were orig- ie) i» i Ri
se ; — a beng got or Pm = a ips os
struction, the Center was initially designed for a
. . oT
welding, taking every advantage of the method’s : ap
—
versatility to secure some unusual structural — | ~~ i Ti me waeA
aspects of the building. In some instances, welding : 1) wis > mle a
was the only joining method which could satisfy : rm gts
design considerations. fe > Wile
F
ib
: 2
eu tt meee
Design Considerations
Every effort was made to obtain maximum usable a aie
space. To this end, the narrow tower section was ; ~ailig .* Ve W)
framed with only one set of internal columns. And :
the east end of the tower was cantilevered 11 ft
past the external column line. This minimum
column design maximized space while enhancing
attractiveness.
By choosing a narrow tower section, measuring Topping-out column being set atop the
139 x 66 ft, the architects were committed to par- United Engineering Center, July 26, 1960.
ticularly high stresses due to wind loads. In ad- United Nations tower rises in background
dition, the width of the columns was restricted by
the architects, consistent with the space-saving de-
sign concept. where necessary. Cover plates. 3° , x 14 in. were
To execute this design, standard H-column sec- welded on each side of 14 WF 426 sections in the
tions were used with additional strength added cantilevered area where loads reached a_ peak.
Lighter cover plates were used in other areas.
SAMUEL C. BAST is an Associate with Seelye, Stevenson, Value & 4 : ; ;
Knecht, New York City; PHILIP DREIER is Chief Engineer for Dreier Another design consideration was aimed at lower-
Structural Steel Co., Inc., Long Island City, N. Y.; CHARLES DICK ing ceiling height and involved running duct work
is Electrode Product Manager for Metal & Thermit Corp., Rahway ae ci sae
N.J and piping through girder webs. Stiffeners were

WELDING JOURNAL | 1053

 Fig. 1—Detail of butt-welded joints at the flange- plate
splices and of the fillet-welded stiffeners at cut-out

TT po —>-
[Top R Fig. 3—A 22-ton 50-ft truss awaits shipment while a
vA 41-ft 6-in. girder is being welded
7
Stiff Ps / Width of top fP+7”
where req'd — ay
-_ length of the plate. This practice allows the beam
< Girder
Ze to rotate slightly under stress. In addition, by allow-
ing the plate to yield before the weld, the design
tends to lower fatigue stresses in the weld.
— Box Column
Bottom f A box column located between elevator shafts
/. fh —
Col. ange Seat ang/e for shear represented another instance where welded con-
struction was a necessity. Unable to use a stand-
\ ard H-column because of space limitations, the
Fig. 2—Typical detail of girder-to-column connection. structural engineers substituted a 18- x 8-in. box
Column coverplates were added where wind loads are column fabricated from four plates. Plate thick-
heaviest ness ranged from 1 and 1°), in. at lower levels
to '/, in. at the upper floors. Fabrication of these
two-story sections would have been virtually im-
possible by riveting or bolting.
welded around the openings where additional
strength was required, as shown in Fig. 1. Designed for Expansion
Wind connections also reflected the over-all Future plans for adding to the three-story sec-
emphasis on strong, compact units. Welding was by tion adjoining the tower played an important role
far the simplest, most economical method of making in fixing design considerations for that area. Three
the beam-column connections. Other methods 50-ft trusses and 4 plate girders, 41 ft 6 in. long,
would have produced units bulkier than the design span the reading room and the library stacks re-
allowed. Savings through reduced material and spectively, providing unobstructed space.
erection costs were roughly estimated at about 5% Eventually, eight stories will be added over the
of total framing for these connections alone. reading room. The three 50-ft trusses, 10 ft 2 in.
A typical wind connection between a beam and deep, are now enclosed between the third floor and
column is shown in Fig. 2. Two plates, ranging in the roof at the fourth floor level. One of the trusses
thickness from '/, to 1'/, in. depending on location, is shown in Fig. 3 before being shipped to the build-
were welded to the beam flanges and butt welded ing site. Figure 4 shows the three trusses in place.
to the column. The butt weld is a single V with a By fabricating the 22-ton units from 14 WF sec-
45 deg included angle. The fillet weld joining the tions rather than angles, a substantial saving in
top plate to the girder flange does not extend the full weight was realized. The sections weigh up to

1054 | OCTOBER 1960

 Welding Practice
As is generally the case, as much shop fabrica-
tion as possible was done. Where possible, all
plates for wind connections were welded in the shop.
During the erection of steel, which started in May
and was completed in late July, an average of five
d-c rectifiers powered by a diesel generator were on
the job daily at the building site. Progress was
steady. Welders were never more than two stories
below the erectors—a factor important in maintain-
ing stiffness in the slim tower while erection was
under way.
Welding at Dreier’s Long Island City plant was
handled by both a-c and d-c units. Edge prepara-
tion involved milling, saw cutting and flame cutting.
Fig. 4—The three trusses shown in position. Electrodes
The girder connections are shown on the right
General practice on most weldments called for a
root pass with low hydrogen E 7018 electrodes rang-
370 lb/ft. Heavy gusset plates and angles required ing in diameter from °/;. to '/,;in. Additional passes
by other methods of construction were eliminated. were made with general purpose E 6014 electrodes.
The completed trusses, designed to carry future Exceptions to this practice included such heavy
column loads of 916 kips at midspan, were shop sections as the trusses and girders where low hy-
fabricated and shipped to site ready for installa- drogen electrodes were used exclusively. Welds
tion. were subjected to visual inspection and no problems
The four 41-ft 6-in. roof girders, located in the bay were encountered.
adjacent to the trusses, were designed to carry
future column loads of 363 kips near midspan. Conclusion
Fabricated from 40-in. deep by °*/;-in. thick web As has been pointed out before, the desire to save
plates and 20-in. wide flange plates that vary from space through limiting column and girder sizes
2 to 3 in. thick along the girders’ length (see Figs. was a large factor leading to welded construction.
1, 3 and 5), the units were also completed in Dreier’s Important savings were realized through reductions
Long Island City shop and shipped to the site ready in both material costs and erection time. Welded
to be placed. wind connections alone resulted in an estimated 5%
Two welders, working opposite each other, were saving in material costs. The virtually rigid frame
used as a team on each girder. Welding permitted produced by welding enabled the designer to use a
elimination of filler plates in back of stiffeners used moment factor of '/;, instead of the conventional
to strengthen the web area (see Figs. 1 and 5). '/s, greatly reducing the steel tonnage in the north-
Filler plates would have been a necessity if riveted south girders taking wind loads.
construction had been used. In addition, there was The end result was an attractive structure which
no need to use additional angles in fabricating the featured a maximum of space achieved economically
stiffeners. through welding.

Fig. 5—Welding a splice in the girder flange plates. View shows reinforcing plates around web cut-out

own

zg
 - PATENT No. NT NO. 1,338,335 moat er
344,617 U.S. 510 U.S. PATENT NO. l, f PATENT NO. 1,547,18
ENT NO. 957,35 0. 1,3557,3 8 U.S. PATEL 9,001 U.S. PATENT NO.
1 U.S. F ; Fate PATEN™ 1,373,918 ll U.S.
249 | 1,407,900 ATENT NO. 3 S 0. 1,409
OO U.S. P x wanna
NO. 1,454
U.S. PATE
NO. é NO. "1,476,
J.S U.S. PATENT
1,504,081 D
0,46 PATENT NO.
9, PATENT NO. ‘ENT NO.. 1 410 U.S.
TENT NO. »~996,630 U.S. PATENT WN’ 472 U.S. tb. NT NO. 1,6
x7 U PATENT NO. 1,630 . 1,650,045 U.S ). 1,654,914 U.S. PATEN’
at +i
‘
Ava . PATENT NO. ‘ PATENT NO. 1,645,25
4 Ar? ot a? .¢ ] wT .
64 Hy. ny z = ns Em : > 4 “hI OD
=
1,67

creare an

THE BEGINNING, 1917 — With the world at war, it became

impossible for our country to import the asbestos-covered
electrodes that were, at that time, considered essential in
the welding of aerial bombs. Working furiously in the face
of this emergency, A. O. Smith came up with an answer —
electrodes wrapped in sodium silicate treated paper. Not
only did+this solve the immediate problem, but it led to
O. Smith’s invention of the extruded electrode which
industry

ushered in the modern era of mass-production welding.

HE ART of electric welding, as it was practiced in
pre-World War I days, would completely frustrate to-
day’s production planner. Costly, asbestos-covered
electrodes had to be used to weld joints that, all too
often, were erratic and of sub-standard quality.
THEN AN IDEA from A. O. Smith turned and
shaped this temperamental technique into what is to-
day the world’s most useful metalworking method.
Essentially a simple idea this development took
the vagaries and the high cost out of electric welding.
It turned this art into a wide-reaching industry.
THE PICTURE-STORY at right tells how it hap-
pened and how A. O. Smith progress is continuing
to make things happen in welding for the future.

Through research& a better way THE PROGRESS — Out of A. O. Smith’s continued develop-
ment of a truly practical arc-welding electrode has grown an
industry impossible to estimate in cash values. Wide-span-
ning bridges, tall-reaching buildings and low-contoured
erro 8 & cars all reflect this progress, with welded components pro-
WELDING PRODUCTS Seven viding strength where it is needed. And in terms of total
Milwaukee 1, Wisconsin technology, A. O. Smith has remained the leader, pioneer
A. 0. Smith INTERNATIONAL S.A., ing such advances as the CO. welding process shown above,
Milwaukee 1, Wisconsin, U. S. A. for stronger, more accurate and more economical work.
For details, circle No. 38 on Reader Information Card
 U.

NO. 1,368, U.S. PAT

U.S. PATEN1 388,1:
1,416,183 U -~ATENT NO
PATENT NO. 1 {%,161 U.S
,452,224 U.S. ENT No.
; NT NO. 1,4 |
S. PATENT NO. 1,477,778
1,491,565 U.S. PATENT NO.
PATENT NO. 1,505,174 U. S. Pé
532,856 U.S. PATENT N O. 1,548
ENT NO. £593,783 3. PATENT
909 U.S.WRATENT NO

WHAT’S TO COME — NUCLEAR-POWERED WELDING - ing-voltage requirements. The proposed f

Already at work on the welding equipment and processes sotope of hydr
of 25 years from now, the scientists of A. O. Smith have visual he tremendous energy re
ized the STELLARWELD. This spectacular wer source
could conceivably utilize fusion power. With science on the ty, yn ng down this
threshold of establishing a self-sustained fusion reaction more handled in weld
produce electrical power, the STELLARWELD becomes
logical next step. This self-contained power source could, it This future-thinking creativity is at work throughout the
is estimated, easily supply 20,000 kw (sufficient for as many A. 0. Smith Corporation, finding the “better way” in scores
as 2500 arcs), as well as its own reaction current and start of products for industry, government, farm and home.
For details, circle No. 38 on Reader information Card
 Yewee Py.Bit tom

Fig. 1—These are the components of the expandable-segmented backing bar

used for ‘‘perfection plus’’ gas metal-arc welding of rockets

“Expanding Bracelet” Backing Bar

Simplifies Welding of Missile Shells

BY C. M. JENKINS

To meet man-in-space “perfection plus’ welding in addition to close dimensional tolerances, these
requirements, a novel “‘expanding bracelet’’ welding welds are required to meet ultra-rigid specifications
backing bar has been developed at the Reynolds for strength and X-ray requirements.
Metals Co. Missile Plant in Sheffield, Ala. (see Fig. Welding of the end-to-end aluminum shells was
1). done on a gantry welding machine on which the work
The new type bar has 225 individual stainless- is rotated under a fixed welding head to form a
steel segments held firmly and evenly by air pres- continuous weld bead of a specified reinforcement
sure against the edges of the aluminum shells to be and penetration.
welded by the inert-gas-shielded metal-arc process. Forty-five grooved backing-bar segments are
It was developed to meet ultra-stringent require- mounted on a curved aluminum frame. It takes
ments for welding of the modified U. S. Army Red- five of these 45-segment units to fit inside the 70-
stone boosters to be used in early stages of the in. diam shells.
*“Mercury” manned satellite program. The grooved stainless-steel segments are bonded
The ‘“Mercury”’ Redstone rockets are scheduled to and held in precise alignment on rubberized canvas
to carry U. S. “astronauts” on their try-out space strips. Each of the five 45-segment units also con-
journeys 100 miles or more above the earth. So, tains an expandable rubber diaphragm with an air
fitting.
C. M. JENKINS is Production Superintendent, Reynolds Metals Co.
Missile Plant, Sheffield, Ala The five frame units are tongue and groove con-

10598 | OCTOBER 1960

 nected. In place, the complete backing bar has
225 segments, each separated by an 0.010-in. air
gap.
As the work is rotated under the welding torch,
each segment is heated for about two seconds.
Because of the minute air gaps, residual heat is not
transmitted to the next segment so there is no con-
tinuous build-up of heat as is found in the usual
one-piece backing bar. At the same time, heating
of the backing bar is held to an optimum localized
level, resulting in maximum stability of welding cur-
rent and voltage control. ee
The shape and size of each of these segments were
determined after a careful analysis of the many \
functions these parts must perform. The individual
segment mass was fixed by the rate of heat transfer
needed to produce a weld of the required tensile Fig. 2—About to be placed in position is the final section of
the expandable-segmented welding backing bar developed
strength and the face surface of each segment was for high-precision aluminum welding. Four of the sections
machined to fit the inside of the shell. are already locked in place and their individual diaphragms
The shape of the 52-deg, 0.160-in. groove was de- connected to the air manifold
signed to control the penetration to dimensional
specifications. In addition, the molten metal first
contacts the sides of the ““V”’ and cooling starts at joint between the shells, and the diaphragms are
this point. Surface tension forms the remainder inflated by means of a detachable manifold. Five
of the molten metal into a rounded bead before chill- individual valves permit the inflation of one or more
ing is completed. sections as needed by varying air pressure to force
This leaves a space between the chilled bead and each of the 225 segments into firm contact with the
the bottom of the “‘V,”’ providing an escape tunnel work.
for the constantly flowing inert gas used in the weld- To permit the diaphragm to slide easily against
ing process. the rubberized canvas backing, a layer of aluminum
A bonus value in the tight-fitting segmented as- foil is used on the back of the canvas.
semblies is the freedom from pitting of the backing- The weight of the assembled bar is held to a
bar surface. minimum by the use of 7075-T6 aluminum alloy for
In setting up a weld with the expandable backing the curved aluminum frames which hold the actual
bar, the shells to be welded are brought together and backing-bar segments and related items. The
securely clamped. Then two of the five unit as- complete 225-segment bar weighs approximately
semblies are put in at the six o’clock position, fol- 350 lb, yet the frame shows no distortion when the
lowed by the next two at three and nine o’clock, diaphragms are expanded under 100 psi. This is
respectively. A spreader bar is then locked into equivalent to a compressive load of more than 60,000
place to secure the latter two sections and the shells lb.
are rotated 180 deg. The fifth, or closing, section To prevent this pressure from distorting the outside
is then lowered into position and the whole assembly of the aluminum shells being welded, a stainless-
is bolted together by draw bolts at each of the five steel retaining band is used on each shell, serving a
junctures (see Fig. 2). double purpose as a tracking ring for the welding
Next, the groove is carefully aligned under the torch.

Fig. 1—-Welding of lattice pole

Welded Aluminum Pole

"am =e Withstands Hurricane Forces

An all-aluminum electrical transmission tower that

exceeds standard strength requirements, yet weighs
only one-third as much as comparable steel towers,
has been developed by the Handley-Brown Co.,
Jackson, Mich., for Line Material Industries,
Milwaukee, Wis.
Engineering tests have proved that the free-

Based on a story by Line Material Industries, Milwaukee, Wis

F
DC Bumblebee

WELDING JOURNAL | 1059

 a
Ahoy
y
standing lattice pole, with the one-piece tower is to connect it with the con-
h\ aluminum crossarms sup- crete foundation base. These bases, of course,
as S\V/\ porting ACSR conductors, can be poured in advance without the necessity of
can withstand load stresses having the towers on hand.
\ equivalent to winds of more Erected, the towers are free-standing—that is,
[ than 140 mph. the only ones that need to be guyed are those at
—, The pole, fabricated from corners where the transmission line ends or changes
aluminum extrusions, has direction.
been designed to carry 69-kv
transmission lines through Load Capacities Tested
TPIT
<7 Structural tests showed the new towers were
residential, commercial and
industrial areas where at- capable of sustaining loads equivalent to winds of
tractive tower appearance is 140 mph while supporting 3 conductors on 3 cross-
essential. arms plus static wire, with the poles spaced 600 ft
apart.
Competitive in Cost Under loads equivalent to 140 mph wind, the
The welded truss-type pole deflected 84 in. from the horizontal without
tower is generally competi- failure of a single member or weldment, then re-
tive in installed cost with turned virtually to horizontal when the loads rep-
its steel counterparts. Its resenting lateral stress were removed.
light weight results in sub-
stantial savings in handling Poles Can Be Reclaimed
and erection costs. The lattice poles are fabricated of 6061 T-6 heat-
In addition, since alumi- treated aluminum-alloy extrusions. A _ protective
num cannot rust and is coating of aluminum oxide forms naturally on all
highly resistant to corrosion surfaces of the towers exposed to the air, giving
even in unusually corrosive the structure a uniform gray satin-like finish.
atmospheres, the new pole Also, the towers can be reclaimed as scrap and
needs no subsequent main- sold at about 25% of their original cost, whereas
Fig. 2—An 85-ft tenance such as_ periodic their galvanized steel counterparts have almost no
lattice pole scrap value because the galvanizing has to be re-
painting.
The lattice poles are 85 ft long and 24in. square. moved before salvage.
Basically, they are truss assemblies comprised of
specially shaped extruded corner chord angles, laced Special Chord Design
with T-channels. The complete pole, including The lattice pole employs a specially designed ex-
three crossarms each nearly seven feet long weighs trusion shape for the chord angles at the corners of
1400 lb, not including the concrete foundation base. the trusses. The edges of these chord angles are
enlarged with a bulb that has four purposes:
Built Like Good Fishpoles 1. It strengthens the edge where failure usually
A significant structural feature of the towers is begins in standard shapes of both aluminum and
that although they do not taper in size toward the steel.
top of the pole, they do “taper” in weight. In 2. It provides room on the chord angle edge for
other words, they are built like well-balanced fish- a longitudinal groove into which the lacings are
poles: heaviest near the base where strength is welded. This expedites the positioning of the lac-
most needed, and tapering off to the lightest chords ing during the welding process utilizing a large
at the top where excess weight would be undesir- moving fixture.
able. 3. The groove or track in the bulb permits an
This tapering is achieved by using three different exceptionally strong weld—essentially a combina-
sizes of chord angles and by varying the weight and tion of butt weld and lap weld since the welding
number of aluminum stiffener plates welded inside material flows down the end and up the side of the
the chords. The largest chords (heavy 3- x 6-in. lace. This also eliminates the need of welding from
extrusions) rise 32 ft from the base. Spliced to inside the truss.
these by welding are smaller 23-ft-long chord angles, 4. And most important, the bulb in this applica-
heavy extrusions measuring 3 x 3 in. and atop these tion acts to carry off the heat of the welding before
are spliced lighter (thinner) 3- x 3-in. chord angles it can affect the weld area, reducing the heat effect
that complete the pole. to a minimum and leaving the area of the weld
Since the pole is welded throughout— including all strong as the rest of the structure.
chord angle splices, lacing and cross-beam con- During welding, the weldments are constantly
nections—the slippage that often accompanies bolted checked to make certain they exceed by better than
construction in conventional towers is eliminated. 10% the standard strength requirements of the
The only field bolting required during erection of National Electrical Manufacturers Association.

1060 | OCTOBER 1960

 Society News

United Engineering Center—The New AWS National Headquarters

Construction of the United En- ance of an invitation to establish The Center will serve as the
gineering Center has been under headquarters in the new United headquarters business office for
way for one year and all phases of Engineering Center would be most some 600 staff members of the par-
the building schedule are on time economical, would result in most ticipating groups; it will accom-
or a little ahead. Since the drive efficient operation and would justify modate the editorial staffs con-
for funds has now entered the crucial unprecedented pride and prestige as a cerned with the preparation and
stages, it is timely to review the result of the associated societies and publication of periodicals, papers,
background and status of this most the fine facilities. books, manuals, codes and stand-
important undertaking as it af- This Somety will enjoy exactly ards; it will permit mechaniza-
fects this SocrIery. the same privileges and responsi- tion with modern equipment and
The 54 year old Engineering bilities as all other groups occupy- thereby improve certain services
Societies Building, in which AWS ing the building. Meeting and ex- to members; it will house the En-
Headquarters has been located for hibit rooms, dining facilities and gineering Society Library equipped
many years, has been sold and must one of the world’s most complete to serve the library search needs
be vacated during 1961. engineering libraries will be housed of the 250,000 members; it will
The new United Engineering in this building. stand as a world-wide symbol of the
Center is being constructed adjacent However the Center is not to importance of engineers.
to the United Nations Building in be a gathering place for engineers ASME President Glenn B. War-
New York City. The twenty story, such as those found in many other ren has written “that this United
welded-steel frame for the new cities—where the buildings serve Engineering Center, carefully
center has been completed and the as a focus for numerous social and located in New York for numerous
glass and stainless-steel walls are professional meetings and assume sound reasons as a result of a study
now being installed. The building the character of an engineer’s club. by a _ top-level committee with
should be ready for occupancy by It is not designed to handle the representation . . . from many parts
the third quarter of 1961. major meetings of the societies. of the Country, is established to
Cost of the project is estimated
to be $12,560,000. Sale of present
building and site, and deprecia-
tion funds are expected to yield AWS DIRECTORS-AT-LARGE
$3,360,000. Industry has already
contributed $4,951,000 of a total Term Expires 196] 1962 1963
goal of $5,400,000. Members of A. A. Holzbaur Jay Bland R. B. McCauley
the Societies which will establish D. B. Howard F. G. Singleton John Mikulak
headquarters in the Center have J. L. York C. B. Smith E. F. Nippes
subscribed $3,252,000 of their total
goal of $3,800,000 (86% W. H. Hobart, Jr J. R. Stitt R. D. Stout
With the cost of the project
fully paid, and since all groups oc-
cupying the Center will be non- AWS DISTRICT DIRECTORS
profit organizations, tax and in- District No. leNew England G. W. Kirkley District No. 6eCentral
terest expenses will be largely elim- District No. 2eMiddle Eastern E. E. Goehringer District No. 7eWest Central
inated. Charges for occupancy,
much less than for comparable District No. 3eNorth Central J. W. Kehoe District No. 8eMidwest
commercial space, will be based District No. 4eSoutheast J. M. Shilstone District No. 9eSouthwest oss, II]
on operating and maintenance costs District No. 5eEast Central H. E. Schulz District No. 10eWestern Connor
plus depreciation allowance. District No. 1leNorthwest obinson
During 1957 AWS officers and
directors initiated careful and ex-
tensive studies of the relative ad- AWS PAST-PRESIDENT DIRECTORS
vantages of alternate plans and
locations for national headquarters. C. |. MacGuffie G. 0. Hoglund
It was finally agreed that accept-

WELDING JOURNAL | 1061

 modernize and bring up to date
UNITED ENGINEERING CENTER the headquarters of the principal
Honor Sections engineering societies in this country,
Section Goal, % Section and is so organized and planned as to
Oklahoma City 125 Dayton serve the best interests of all of the
Mahoning Valley 113 Holston Valley members of these societies irre-
Hartford 104 Louisville spective of where they are located.”’
Kansas City 102 —_N. Central Ohio The Board of Directors of AWS
Northeastern Tennessee 102 Olean-Bradford has agreed to seek a total of $60,000
Puget Sound 102 Pascagoula (less than '/,% of the project cost
Colorado although AWS will occupy about
101 Sangamon Valley
Baton Rouge 101 5% of the office space) from mem-
Niagara Frontier 101 Santa Clara Valley bers and sections. As of August
Providence 101 Syracuse lst, fund drive records show the
Rochester 101 Toledo following percents of indicated goals
St. Louis 101 Western Mass. subscribed by members of major
Boston 100 Wichita societies:
Bridgeport 100 Worcester
Pledges Needed to Meet Goal
Society Pledged, %
Section Needed Section Needed
ASCE 99
Detroit $ 30 South Florida $ 395 AIME 82
Philadelphia 82 Madison 400 ASME 88
Northern N. Y. 85 Richmond 400 AIEE 94
Mohawk Valley 100 San Antonio 400 AIChE 106
Tri-Cities 100 Western Michigan AlindE 105
145 400 AWS 51
Nashville 435
150 York-Central Pa. ASHRAE 10
Albuquerque Cincinnati 438 IES 17
Anthony Wayne 150
Cacinen 190 Saginaw Valley 466
Shreveport 195 Salt Lake City 480
Arizona 200 Portland 485 Twenty-seven AWS Sections had
Eastern Illinois 200 Indiana 490 pledged amounts equal to or ex-
Nebraska 200 Long Beach 490 ceeding their goals prior to August
New Hampshire 200 Stark Central 495 lst. Forty-four other Sections need
Maryland 215 Fox Valley 500 to pledge $500 or less to meet their
New York 218 Washington, D. C. 500 goals and only thirteen sections
Carolina 245 North Texas 585 must pledge more than $500 addi-
Northwestern Pa. 280 Lehigh Valley 725 tional to meet their goals. Check
J. A. K. 295 Columbus 790 your section standing in the ac-
Tulsa 295 Long Island 800 companying list.
lowa 300 Northwest 822 The 1960-61 campaign team is
Michiana 300 Milwaukee 997 headed by Vice-president John
Mobile 300 San Francisco 1154 Blankenbuehler and eleven Dis-
Birmingham 355 New Jersey 1460 trict Directors—G. W. Kirkley,
“Peoria 365 Houston (and Sabine) 1575 E. E. Goehringer, J. W. Kehoe,
San Diego 365 Chicago 1640 J. H. Shilstone, H. E. Schultz,
lowa-lilinois 370 Pittsburgh 2168 R. H. Hoefler, L. L. Baugh, G. O.
Susquehanna Valley 370 Cleveland 2355 Bland, C. L. Moss, III, D. P.
New Orleans 390 Los Angeles 2825 O’Connor and C. B. Robinson.
Many AWS Sections have met
their goal with treasury funds or
with funds from educational or
MY CONTRIBUTION TO THE NEW HOME OF AWS other special projects. Some have
conducted individual member fund
drives.
In consideration of the gifts of others, I intend to give to
O UNITED ENGINEERING CENTER BUILDING FUND
_Dollars $
Welcome
Paid herewith $ Balance will be paid as follows
e Supporting Company
Credit my gift to: Effective Sept. 1, 1960:
Oo AWS 0 Herbst Bros.
Signed 1730 Harvard Ave.
Anaheim, Calif.
Print Name
Omark Industries, Inc.
9701 S. E. McLoughlin Bvd.
Portland 22, Ore.

1062 | OCTOBER 1960

 KEEPING YOU POSTED

by Fred L. Plummer

@ On October ist Bonney Rossi in the June issue of the Journal of Mr. Mooney is chairman of this
who has so effectively guided the Engineering Education. group.
development of the WELDING JouR-
NAL will assume greater respon- e During July a network program @ On July 18th President and Mrs.
sibilities as Executive Secretary of on trades with about 70% of the Thomas and their daughter were
the Society for Experiment Stress time devoted to welding operations guests of your Secretary and Mrs.
Analysis, establishing new national disclosed active AWS Member Plummer at their home in Stam-
headquarters for that group in H. A. Sosnin demonstrating welding ford, Conn.
Westport, Conn. However, he will to a group of students. It was one
of the finest pieces of educational @ Leaders in the American Council
continue as Consulting Editor for of the International Institute of
a few months. T. P. Schoonmaker publicity the welding industry has
had ina very long time. Welding met in your Secretary’s
is our new WELDING JOURNAL office on July 27th to complete
Editor. He comes to the SociETy @ Former AWS Technical Secre- plans for the 1961 IIW Assembly
from Linde Co., Division of Union tary Simon A. Greenberg, now an which will be held in New York
Carbide Corp., where he has served Industrial Consultant, has opened next April. Ed Dato who is Chair-
as Technical Information Admin- an office in Flushing, N. Y., ‘“‘to man of the AWS Convention Com-
istrator. Formal announcement will serve industry in cost reductions, mittee has agreed to serve as Chair-
appear in the November issue. consistently uniform good quality man of the New York Arrange-
@ On July 27th Neale E. Orrok and efficient and profitable produc- ments Committee for both the
joined the AWS headquarters staff tion.”’ IIW Assembly and the AWS Annual
as Assistant to Technical Secretary e@ Foreign visitors at headquarters Meeting and Welding Exposition.
Fenton replacing one of the two during the summer include E. H. He will be assisted by members of
assistants who have left this de- Lee of Manchester, England, the Long Island, New York and
partment to accept positions with who is active in the British In- New Jersey AWS Sections. IITW
industry. Past-president Howard Biers, Ad-
stitute of Welding; I. D. Baring,
@ Nathan R. Presser joined the Welding Engineer, active in the miral E. H. Thiele, who is Chair-
staff on August 8th to work with Australian Welding Institute and man of the Ship Structure Com-
Ed Krisman and Frank Mooney Managing Director of I. D. Baring mittee, Director Ken Koopman of
and make possible new member- Pty. Ltd. of Melbourne; Kan the Welding Research Council, AWS
ship promotion activities and addi- Okada of Yawata Welding Elec- President R. D. Thomas, Jr., and
tional support of Section programs. trode Co. and Kozo Uechi of Yawata their associates together with AWS
These men will assist Membership Iron and Steel Co., both from Japan; staff members are planning these
Chairman H. E. Miller and Vice- Y. Miyazaki and M. Fukui of important programs.
chairman Andy Axtell in their Osaka Denki Co. in Japan and e AWS Past-president J. H. Hum-
work with Section membership com- K. Hokari of Toyo Menka, now berstone heads an industry sponsor
mittees. stationed in New York. committee and is obtaining sup-
@ Former teacher of English in @ On June 29th members of the port of fabricators, manufacturers
engineering colleges, Howard G. AWS Publicity Committee met and distributors. Committees re-
Zettler joined Art Phillips and Bill at headquarters with Chairman sponsible for special events, ladies’
Hall on August 15th to expand and R. E. Lawson and Staff Member programs, plant tours, breakfasts,
strengthen SociETy educational and A. L. Phillips to plan an active lunches, receptions, banquet, enter-
information programs and to assist program including new _ support tainments, visits to UN-Museums-
in editing the WELDING HANDBOOK. for Section publicity activities. Empire State-RCA-MusiceHall-etc.,
e@ During October President R. D. are active and have held meetings
@ A special committee with Past- with the AWS staff, the New York
Thomas, Jr., and your Secretary president O. B. J. Fraser as Chair-
will attend the AEC Welding Forum Convention Bureau and various spe-
man, Past-president H. W. Pierce, cial representatives.
in San Antonio on the 4th and President R. D. Thomas, Jr., and
5th and then meet with AWS Treasurer H. E. Rockefeller as @eOn July 19th your Secretary
Sections as follows: Baton Rouge members, met with your Secretary joined Vice-president C. M. Parker
and New Orleans on October 6th on July 22nd to study pension plans of the American Iron and Steel In-
and 7th; San Antonio, Dallas, which might be made available to stitute for lunch at the Pinnacle
Houston, Sabine and Tulsa on AWS staff members. Club atop the Socony Mobil Build-
October 10th to 14th. They will On this same day Assistant Secre- ing in New York for a discussion of
also visit the Buffalo Section on the keynote address “Steel: Avant
tary Frank Mooney held a well
October 27th. attended luncheon and panel dis- Garde”’ which he will present at the
e@ An article “Welding Education cussion meeting for members of the AWS National Fall Meeting in the
in the Engineering Curricula’ by National Association of Exhibit Penn-Sheraton Hotel in Pittsburgh
President Thomas was published Managers in the New York area. on September 26th.

WELDING JOURNAL | 1063

 For tips on finding the Great Horned Owl...

call an ORNITHOLOGIST
(specialist on birds)

For details, circle No. 8 on Reader information Card

1064 | OCTOBER 1960

for tips on welding stop-and-go jobs...

call in LINCOLN

(specialists in are welding)

gesacone MANUFACTURER OF MOBILE HOMES doubled the welding speed on his

undercarriages by simply changing electrodes—and in addition, saved over $8000
in the first year.
Manufacturing cost on undercarriage fabrication was prohibitive. Thirteen gauge cross
members were welded to twelve gauge channels by welds made in both vertical and
flat positions. These short welds on steel having some scale and oil slowed down
production.
Finally they called in their LINCOLN Field Engineer. Painstaking tests, made by the
LINCOLN man with the welding foreman and plant superintendent, proved LINCOLN’s
Fleetweld 37 electrodes far better for this application.
RESULTS: lower costs . . . welding speed doubled . . . cleaning time cut in half
That's why we Say it’s a good idea to do business with LINCOLN where arc welding is
a specialty and cost reduction comes to you as a “‘plus’’ at no charge.

THE LINCOLN ELECTRIC COMPANY

Dept. 1580 e Cieveiand17,
Ohio WELDERS

For details, circle No. 8 on Reader information Card

WELDING JOURNAL | 1065

@ Chairman J. E. Norcross presided
at a meeting of the Exposition Com-
mittee held in your Secretary’s
office July 28th. Financial reports
covering the successful Los Angeles Welding Engineering Course at University of Illinois
Exposition held last April were
The first biennial short course on welding engineering has been
reviewed and a final budget cover-
scheduled by the University of Illinois, University Extension Division,
ing the April 1961 Welding Ex-
Urbana, Ill., during the week of Dec. 5-8, 1960. Fifteen experts from
position to be held in the New York
industrial and research organizations will team up with 12 university
Coliseum was adopted.
professors in a joint effort with the AMERICAN WELDING SOcIETY.
e During July and August your Registration will be limited to sixty. Professor W. H. Bruckner of the
Secretary interviewed many in- Department of Mining and Metallurgical Engineering is general chair-
dividuals interested in becoming man. Co-chairmen are Professor W. H. Munse, Department of Civil
associated with your headquarters Engineering, and M. B. Singer, Department of Mechanical Engineer-
staff as editor of the WELDING ing. The program is as follows:
JOURNAL. He was assisted and
guided on frequent occasions by
Past-president C. I. MacGuffie, Monday, December 5th
President R. D. Thomas, Jr., Treas-
urer H. E. Rockefeller, Publica- 8:30 9:30 Registration—Rm. 314 South, Illini Union Bldg.
tions and Promotions Council Chair- 9:30 10:00 Welcoming address by Prof. R. J. Martin
man C. E. Jackson, Welding Jour- 10:00 12:00 Review of Welding Methods: (a) Metallic-arc Methods
J. Revelt; (6) Gas Tungsten-arc and Gas Metal-arc Meth-
nal Committee Chairman J. E. ods—W. G. Schumacher; (c) CO, Welding—R. W. Tuthill
Norcross and WELDING JOURNAL 3:00 Welding Machines—-Generators, Transformers, Rectifiers:
Editor B. E. Rossi. (a) From the User’s Point of View—J. D. Brown and D. H.
e@ The Executive Group of EAC Hawes; (6) From the Manufacturer’s Point of View—G. K.
met in Philadelphia August 11th, Willecke
3:00 4:30 Metal Cutting with Gas and Arc Methods——C. F. Alexander
with Chairman E. C. Miller and 7:30 9:30 Smoker—General Lounge, Illini Union Bldg.
Education Secretary A. L. Phillips
to plan courses to be given during
the 1960-61 fiscal year by the AWS Tuesday, December 6th
School of Welding Technology and
other activities. 8:30 11:50 Welding Metallurgy: (a) General Survey—-W. H. Bruckner;
(6) Practical Metallurgy—-W. M. Norton; (c) Special
e@ Your Secretary met with Steve Problems, High Alloys—-G. E. Linnert; (d) Weld Metal
Marras, Secretary of Engineering and Weldability Problems—-G. E. Claussen
Foundation, on several occasions 1:00 1:15 Group Picture—Rm. 314 South, Illini Union Bldg.
during August to discuss coopera- 1:15 2:45 Introduction to Heat-flow Theory—B. T. Chao
tion on future projects of mutual 2:45 3:10 Discussion of Heat-flow Problems
3:10 3:40 Electrodes for Fusion Welding and Filler Metals—O. T.
interest. Barnett
@ Elsewhere in this issue of the 3:45 5:30 Visits to welding activity in Department of Civil, Mechani-
JOURNAL you will find information cal and Metallurgical Engineering
about the new United Engineering 8:00 10:00 Motion pictures of resistance and fusion-welding applica-
Center which is being erected near tions
the United Nations building and
already provides a new point of in- Wednesday, December 7th
terest in the East River New York
skyline. Vice-president John 8:30 10:00 Resistance Welding, Principles and Methods—-H. A. James
Blankenbuehler and your eleven 10:00 11:00 Brazing, Principles and Methods—G. S. Hoppin, III
District Directors solicit your co- 11:00—-11:50 Distortion in Welding—J. R. Stitt
operation in helping your Section 1:15 2:10 Residual Stresses and Relief—R. E. Lorentz, Jr.
2:15 5:15 Round Table Discussions Topics: Welding Design; Dis-
join the select list of about thirty tortion and Stresses; Welding Electrodes and Filler Rods;
which have subscribed all or more Metallurgical Problems
than their share of our SOCIETY’s 6:45 10:00 Banquet at Urbana-Lincoln Hotel——T. B. Jefferson
goal.
Thursday, December 8th
Free Information :30 9:25 Mechanical Properties of Welds and Weldments
e Munse
Free Literature 10:25 Design and Welding for Buildings E. H. Gaylord
11:25 Design and Welding for Bridges J. E. Stallmeyer
& 11:50 Design and Properties Discussion
“YOURS FOR 2:00 1:00 Luncheon and Presentation of Certificates
THE ASKING” 715 2:20 Design and Economics of Machine-teol Welding
Singer
:30 3:30 Quality Control—J. A. Henry
Use Reader Information 3:30 4:30 General Round Table
Card :30 Adjournment
t
Page 1103

1066 | OCTOBER 1960

 -CDUCATIONAL ACTIVITIES

Welding Education at the University of Wisconsin

BY Norman R. Braton

Demand for welding education is is designed as a survey of welding shortened version of this course’
growing. Practical courses in vo- theory and practice. Sixteen busy for which one credit is offered, is
cational high schools are constantly weeks, each consisting of a one- required of all chemical-engi-
requested, and vocational school hour lecture and a three-hour lab- neering students. It has been spe-
administrators report that classes oratory period, are spent in gaining cially designed to cover problems
are filled to capacity. In addition, a basic understanding of this vital which may be met in this engi-
adult education curricula are being field. Two credits are offered to- neering area. In addition, a prac-
expanded to include practical weld- ward the fulfillment of graduation tical course is offered for the as-
ing courses. In one eastern com- requirements. sistance of agricultural students
munity where rising costs made The course covers current weld- who come to Madison from all parts
reduction of the adult education ing practice, metallurgical data of Wisconsin.
program necessary, the welding and welding theory. All industrial At the University, the welding
course was considered nonexpend- welding processes are taught, from staff is headed by Norman R.
able. oxyacetylene welding to the latest Braton, assisted by Gerald A.
Practical courses, however, are developments in gas-shielded arc- Duchon and Charles F. Peters
not sufficient, as the overwhelming welding techniques. Stress in the who has taught welding since the
demand for admission to the AWS laboratory is placed on practical Wisconsin program was instituted
School of Welding Technology in- exercises involving both new and and who has been a member of the
dicated (WELDING JOURNAL, April popular methods so that students University for 45 years. Each is
1960, p. 362). The continuing ad- will be familiar with the industrial an active member of the AMERICAN
vancement of welding as a science methods commonly in use. Met- WELDING SOCIETY.
and an art depends strongly on the allurgical data presented in the Programs similar to Wisconsin’s
dissemination of the theory under- course includes the study of the are feasible for the majority of tech-
lying welding processes and of a weldability of various metals and nological colleges. Fitting courses
basic knowledge of welding metal- their alloys, including aluminum concerning welding technology into
lurgy, areas which practical courses and stainless steel. In addition, already heavy schedules is not an
can touch only lightly. Refine- students are exposed to problems insoluble problem. Moreover, the
ments in standard processes and involving welding costs, design, increasing use of welding applica-
advancements derived from com- inspection and quality control. tions in industry, the rapid advances
plex metallurgical research can con- Quizzes and hour examinations are made in metallurgical knowledge
tinue to be achieved only if insti- frequent and demanding. and the burgeoning complexity of
tutions of higher learning respond M. E. 37 is, however, only part welding technology make the estab-
to the demand for a more inclusive of the welding program offered by lishment of such programs almost
welding education. the University of Wisconsin. A mandatory.
The fact that, each year, over
350 students enroll in the welding
program in the College of Engi- WELDING AT UNIVERSITY OF WISCONSIN
neering of the University of Wis-
consin indicates that institutions
of higher learning are meeting the
demand. At Wisconsin, in connec-
tion with the other branches of
technology necessary to continued
welding advancement, welding tech-
nology—as well as welding proc-
esses—is stressed as necessary to a
well-rounded technological educa-
tion.
Students learn gas-welding methods
Mechanical-engineering course 37
is the most ambitious of the r
courses offered by the University.
Required of all mechanical-engi-
neering students, but open by
election to students of such other
engineering disciplines as electrical
engineering, metallurgical engi-
neering, civil engineering and ag-
ricultural engineering, the course
NORMAN R. BRATON is Assistant Professor
of Welding in the Department of Mechanical
Engineering, University of Wisconsin, Madison C. F. Peters makes submerged-arc weld Spot-welding machine aids study

WELDING JOURNAL | 1067

ELIMINATE TEN vol
CHANGES

You save time, space and money with

PAY-OFF-PAK drums of Lukens Plate-Mate
Welding Wire.
You save 42 cent a pound by purchasing
welding wire in 700-pound PAY-OFF-PAKS.
You save 43% storage space compared with
coils. You save downtime by eliminating ten
coil changes... 2% to 5% hours that could
be put to profitable production. You save
power —with accessories, PAY-OFF-PAKS can
be turned with no more power than it takes
to turn a 60-pound coil. And PAY-OFF-PAKS
protect the wire from dust, dirt and damage.
Available for all diameters up to 7/32”.

PAY-OFF-PAK ACCESSORIES
HAND TRUCK~— Where an overhead crane
is not available, a Plate-Mate hand truck will
speed stock arrangement and replacement.
GRAB-—Handy drum grab quickly and re-
liably grips the strong fiber drum, permits
sure, fast movement by crane.
TURNTABLE~—Plate-Mate turntable feeds the
continuous wire dependably and easily
through a stationary eyelet. Ball-bearing
base.

LUKENS STEEL COMPANY, Coatesville, Pa.

Aida] PLATE MATE

ak we

For details, circle No. 9 on Reader information Card

WELDING JOURNAL | 1069
 SECTION NEWS AND EVENTS

As reported to Catherine M. O'Leary

Orange County Section Receives Charter

A new section was welcomed into chapter for the State of California.
the ranks of AWS in August, as Officers reported for the present
Orange County, Calif., successfully term are as follows: chairman,
petitioned for official recognition. D. Larson; Ist vice-chairman, AWARDS
The action of the Board of Direc- A. R. Dietrich; 2nd vice-chairman,
tors, which became effective Aug. H. D. VanCleve; secretary, W. C.
1, 1960, upon fulfillment of the By- Smithey; treasurer, J. W. Boehles; Long Beach—On June 18th the
law requirements, brings to 86 membership, D. L. Lohoff; tech- Long Beach Section held its last
the number of active sections in the nical representative, J. A. Regan; meeting for the 1959-60 season.
growing roster of AWS. program, C. L. Driscoll; educa- After dinner, Chairman Ralph
The formation of the Orange tional, J. T. Parker; publicity, Hogate turned the meeting over
County Section makes the 6th active J. Parks. to Installation Officer Charles
Breese, a member of the Past
Chairman’s Club. Mr. Breese then
introduced the new officers, giving
SECTION MEETING CALENDAR a little information about each
one of them and what jobs they
performed for the AWS.
OCTOBER 5 DETROIT Section. The Engineering Society of Following the introduction of new
SUSQUEHANNA VALLEY Section. Foot Hills Detroit (Rackam Bidg.). 8:00 P.M. ‘“‘Electron- officers, the meeting was turned over
Manor, Shickshinny, Pa. Dinner 6:45P.M. “CO, beam Welding,” William J. Farrell, Sciaky Bros., to George Gorich, Educational Ac-
Welding,” Robert Mann, Air Reduction Sales Co. Inc. tivities chairman, who announced
TULSA Section. Danner’s Cafeteria, Tulsa,
OCTOBER 11 Okla. Social 6:30 P.M. Dinner 7:00. Technical the winners of the trophy and cash
BIRMINGHAM Section. Salem's Restaurant meeting 8:00. ‘Some Dramatic Applications of awards. These awards are given
Number Two, Birmingham, Ala. “Gamma Radi- Stainless and Low-alloy Steel Welding,” R. D. to students who compete from six
ography,” A. V. Kelley, Budd Co., Instruments Div. Thomas, Arcos Corp. Report by Secretary Fred L. junior colleges in the Long Beach
NORTHEAST TENNESSEE Section. Speaker, Plummer. Section. Trophies with the stu-
John H. Berryman, Air Reduction Sales Co. OCTOBER 17 dents’ names inscribed on them are
SANGAMON VALLEY Section. Springfield, lil. PHILADELPHIA Section. Engineers Club, Phila- awarded to their schools. Cash
“CO. Welding,” Harley Orr, Hobart Brothers. awards are given to the students.
WESTERN MASSACHUSETTS Section. Oaks delphia, Pa. Dinner meeting. Coffee speaker
Inn, Springfield, Mass. Dinner 6:30 P.M. Meet- H. K. Hiestand, Reynolds & Co. “Planning Your In- A special award, the Hank Tullis
ing 7:45 P.M. “100 Years of Welding,” Harry vestments.” “‘Electron-beam Welding—its De- Foundation, was given for the best
Schwartzbart, Armour Research Foundation. scription and Application,”’ David N. Reese, N.R.C. combination welder. This is a
Corp. $100 cash award. There were four
OCTOBER 12 ROCHESTER Section. Plant trip to Delco Ap- awards given in all, as follows:
CLEVELAND Section. Cleveland Engineering pliance Div., General Motors. Dinner in Delco
and Scientific Center, Cleveland, Ohio. Cocktail Cafeteria, 7:00 P.M. Guided Tour 8:00-9:30 P.M. Arc—Robert Saunders; Acety-
6:30 P.M. Dinner 7:00. Meeting 8:00. OCTOBER 18 lene—Richard Ediness; Com-
HOUSTON Section. “Report from the Secre bination—Hank Tullis Award,
tary,” Fred L. Plummer. “Fabrication Applica MOBILE Section. Korbet’s Restaurant. Social
Hour 6:30 P.M. Dinner 7:15. Technical Session Hudson Remel.
tions,” R. D.Thomas, President.
NORTH TEXAS Section. Howard Johnson's 8:00. This year recognition was given to
Turnpike Resturant. Dinner and Technical Session OCTOBER 19 Instructor James Craig of Long
6:30 P.M. PEORIA Section. Ladies Night. Vonachen’s Beach City College whose students
OCTOBER 13 Junction. Speaker Charles Dancey, Editor, Peoria won all four awards. All winners
Journal Star. were present to receive the awards
ALBUQUERQUE Section. Leonard's Restau
rant, Albuquerque, N. Mex. Dinner meeting 7:00 OCTOBER 20 and show specimens of their work.
P.M. “Welding in Missile Fabrication,” Don WASHINGTON Section. Washington Gas Light The newly elected officers for the
Howard, ACF Industries. Co. Auditorium, Washington, D. C. “Welding Ac 1960-61 season are as follows:
BEAUMONT Division HOUSTON Section, Beau- tivities in the Bureau of Ships,” Comdr. Earl C. chairman, Gerald Garfield; vice-
mont, Tex. ‘Fabrication Applications,” R. D. Vicors, U.S.N. chairman, Wm. DeRouchey; pro-
Thomas, Arcos Corp. “Report from the Secre gram chairman, Ted Strangman;
tary,”’ Fred L.Plummer. OCTOBER 21
MARYLAND Section. “Mechanized Hard Sur membership chairman, Hubert
IOWA-ILLINOIS Section Plant Tour. Rexroat; comptroller, Gail Beck-
facing,’’ Eldon C. Hurt, Haynes Stellite Co.
OCTOBER 14 MILWAUKEE Section. Ambassador Hotel. Buf- strand; treasurer, R. S. Miles, Jr.;
COLUMBUS Section. Columbus, Ohio. ‘Meth- fet Dinner 6:30 P.M. Technical meeting 8:00. secretary, Lee Champieux; _re-
ods Engineering as Applied to Welding,”’ Donald E. “Power Sources for Manual and Automatic Weld- cording secretary, Gordon Williams;
Cox ing,” J. W. Pomazal, Harnischfeger Corp. publicity chairman, Timothy
Crosby; codes chairman, Wayne
Editor's Note Notices for January 1961 meetings must reach JOURNAL office prior to October 20th DeMeritt; sergeant-at-arms, Glen
so that they may be published in the December Calendar. Give full information concerning time, place, G. Schuyler; entertainment
topic and speaker for each meeting.
chairmen, Dick Tullis and Bud

1070 | OCTOBER 1960

ACHIEVEMENT AWARDS GIVEN BY LONG BEACH SECTION

go VELING Dp
v,
i> \ a, tp =
We

Welding contests were recently spon- Mr. Remel, left, is shown above accept- Incoming Section Chairman Jerry Gar-
sored by Long Beach Section. Shown ing the ‘‘Hank Tullis’’ award from Mr. field is shown accepting group records
above, left to right, are R. Sanders, winner Gorich and Section Chairman Ralph from Mr. Hogate. C. Breese, instal-
of arc-welding problem; Instructor J. Hogate lation officer, smiles approvingly
Craig; H. Remel, winner of gas and arc-
welding contest; G. Gorich and
R. Ediness, winner of gas problem

Tullis; educational activity

Chairman, George Gorich; ad-
vertising chairman, Don Christo-
pherson; aircraft advisory chair-
man, Wayne DeMeritt.

SEMINAR
Decatur—The first annual Weld-
ing and Fabrication Seminar spon- pe
sored by Millikin University and A number of technical sessions were held at the March 18th and 19th seminar on
the Sangamon Valley Section was welding and fabrication sponsored by the Sangamon Valley Section. One of the
held at the University in Decatur typical groups is shown above
on March 18th and 19th. Registra-
tion started at 10 A.M. At 1:30
P.M., a plant tour was conducted of
four of the manufacturing plants
in Decatur. York Division of Borg
Warner and Caterpillar Tractor Co.
were visited by one group.
The other group toured the Mis-
sissippi Valley Structural Steel Co.
and Leader Iron Works, Division
of Standard Steel Co.
There were 43 who attended these
tours and were transported by
buses.
A technical session was held at Plant tours attracted many of those in attendance at seminar. This group is waiting
1:30 P.M. on the first day at for bus to take them to destination
Millikin University.
A banquet was held at 7:00
P.M. at the Orlando Hotel, with
an attendance of 143. The principal
speaker was Harry A. Stuhldreher,
assistant vice president, personnel,
United States Steel Corp., and one
of the famous ‘‘Four Horsemen”’ of
Notre Dame. Mr. Stuhidreher
gave a very interesting talk on
“Every Day Living.”
On Saturday, a breakfast was
held for the speakers. There were
eight sessions on this day and they
were all well attended. The total Guest speaker at banquet was Harry A. Stuhldreher, one of Notre Dame's famed
registration for the technical ses- “Four Horsemen."’ He is shown above, third from left, at speaker's table.
sions was 163. Director Lester L. Baugh is second from right

WELDING JOURNAL |} 1071

 ATTEND WELDING-DESIGN SEMINAR

ron CONTROLLED

ATMOSPHERE

zy

Zi’

BLICKMAN

VACUUM DRY BOX A 2-day seminar on welding design was co-sponsored by the Olean-Bradford Section
and Clark Bros. Co. on July 21st and 22nd. Among the speakers were left to right,
C. M. Richardson, O. Blodgett, D. F. O'Donnell, J. Matheny, G. Goetschius,
Designs and specifications are avail- B. Nick and R. F. Dailzell
able for a variety of welding enclo-
sures for research and production
welding, and for work in the fields of
metallurgy and physical chemistry.
These enclosures can be fully evacu-
ated and then be filled with an inert
gas for welding in an inert atmosphere. by Clark Bros. Co., an operating
Write for Technical Bulletins on vari- division of Dresser Industries, with
ous types of welding enclosures: key men of the seminar being pro-
S. Blickman, Inc., 3010 Gregory Ave- vided by Lincoln Electric, Cleve-
nue, Weehawken, N. J. Olean—Sixty-eight design en- land, Ohio. In attendance were
BLICKMAN gineers, supervisors and inspectors industry men from Clark, Dresser
LABORATORY EQUIPMENT of four Olean, N. Y., and area in- Mfg. Co., Bradford, Pa.; Air Pre-
dustries attended a two-day seminar heater, Wellsville, N. Y.; and the
Look tor this symbo! of quality Bille urns rua on Welding Design on July 21st and Bovaird & Seyfang affiliate of
22nd. The seminar was sponsored Clark Bros. Co. in Bradford, Pa.
For details, circle No. 10 on Reader Information Card by the Olean-Bradford Section and Omer Blodgett, a design welding
engineer from Lincoln, served as
the main speaker and discussion
leader in the main topic area en-
21ST ANNUAL WELDING SYMPOSIUM HELD IN CLEVELAND titled, ““The Use of Welding in
Machine Design.” The main topic
was divided into five subtopics,
namely (1) “Designing for Impact
and Repeated Loads,” (2) “How
to Build Up Sections Having Re-
quired Properties,’ (3) “Efficient
Use of Steel Sections,” (4) ‘““Design
Formulas for Welded Construction”
and (5) “Best Method of Fabrica-
tion of Those Items Encountered
Technical speakers T. McElrath, W. H. Otto J. Bornemann, extreme left, is in Design.”
Wooding and R. A. Wilson and Chair- applauded by Section Chairman T. L. Other Lincoln Co. speakers were
man W. Romance are introduced by Dempsey (second from left) and others C. M. Richardson, district sales
Technical Chairman C. B. Herrick (left at dinner meeting. Mr. Bornemann was manager for Northwestern Penn-
to right) at the recent symposium Symposium Dedicatee sylvania and R. F. Dalzell, Re-
sponsored by Cleveland Section gional sales manager. James H.
Matheny, welding engineer at Clark,
spoke on “Present Welding Opera-
tions at Clark.” D. F. O’Donnell,
Clark’s supervisor of education,
served as the seminar moderator.
The seminar is the first of a busy
educational activity scheduled by
Clark Bros. for the coming year.
The present training program has
been directed to draftsmen, in-
spectors, shop supervisors, etc.,
whereas the above seminar was
for the benefit of designers and
management level personnel to pro-
Over five hundred people attended the evening dinner meeting mote the use of welding.

1072 | OCTOBER 1960

 COLUMBUS SECTION
OFFICERS
O Lit

WELDING SYMPOSIUM

Cleveland—The 2ist Annual

Welding Symposium success-
fully climaxed an outstanding series
ESISTANGE

of welding meetings held by the

Cleveland Section during the 1959
60 season. Afternoon and evening (Ldn

meetings were held at the Hotel

Manger in Cleveland, Ohio. An The Columbus Section has elected — competitively
invitation—extended to engineer- R. S. Ryan, left, and W. L. Green, chair-
man and vice chairman, respectively, priced!
ing, management and supervisory
for the 1960-61 fiscal year
personnel of local industry to be-
come acquainted with the latest
developments in welding processes
and their application—turned out an
attendance in excess of 500 people. ELECTION OF OFFICERS
The afternoon-evening program Columbus—The Columbus Sec-
began at 2 : 00 P.M. with a technical tion announces the election of
session chairmanned by Clayton B. officers for the 1960-61 season as
Herrick. Subject matter concerning follows: chairman, Robert S. Ryan,
production fabricating included dis- Columbia Gas System Service Corp.;
cussion of the electroslag-welding vice-chairman, William L. Green,
process by Walter H. Wooding of Ohio State University; secretary,
Arcos Corp., Philadelphia, Pa., and Charles J. Meinhart, North Amer-
“Short-arc Welding’ by Thomas ican Aviation; treasurer, Milton
McElrath of Linde Co.’s Develop- D. Randall, Battelle Memorial In-
ment Laboratory, Newark, N. J. stitute; directors, Francis E.
Other featured speakers included Adams, William J. Coughlin,
Kenneth J. Altorfer of Air Reduc- Glenn E. Faulkner, E. G. Hirsch,
tion Research Laboratory of New John L. Purdy and Frank D.
York City discussing “‘Automation Smith; committee chairmen— ar-
Unlimited in Machine Flame Cut- rangements, Robert R. Wright,
ting’ and Robert A. Wilson of Columbia Gas System Service
The Lincoln Electric Co., Cleveland, Corp.; educational, Glenn E. Faulk-
Ohio, describing “How to Deposit ner, Battelle Memorial Institute;
Weld Metal at 200 Pounds Per membership, Howard A. _ Bal-
Hour.” thasar, Welding & Mechanical Sup-
After the technical talks, the plies, Inc.; publicity, Frank D.
assembled group migrated to a Smith, Welding Supply Service;
booth area where men expert in the technical representative, William
various welding fields offered in- L. Green, Ohio State University; THOMSON manufacturing capabili-
formal consultation service on spe- CTC representative, John T. Nie- ties have resulted in reduced costs
cific welding problems. mann, North American Aviation. on their resistance welding machines
The symposium program, chair- of long recognized quality...
manned by Wasil Romance, 2nd
vice chairman, continued into the SPOT PRESS
evening with a well-attended dinner. MEET IN PHILADELPHIA PROJECTION FLASH-BUTT
Introduction of the new officers SEAM HIGH-FREQUENCY
for the coming year and presenta-
tion of various awards were followed
For all AUTOMATED and HIGH-
by a talk by Dr. Carl C. Byers on
PRODUCTION requirements —
the subject ““Get Off Your Launch-
either special or standard design
ing Pad.”
— THOMSON QUALITY is within
The evening was highlighted by your budget. A quote from
the recognition of the Symposium THOMSON will show you how.
Dedicatee, Otto J. Bornemann,
through retiring Chairman T. L.
Dempsey. Mr. Bornemann, active Look for our nearest representative
in the SocrETy since 1923, is a in the yellow pages or contact —
maintenance welder in a local in-
dustrial concern and operates his Lorin K. Poole and H. J. Stetina, newly
own welding shop in his spare time. appointed members of Philadelphia Sec-
Mr. Bornemann typifies the man tion Executive Board, talking shop with
who has been the foundation of the Vice Chairman Dave Buerkel at August HOMSON ELECTRIC WELDER COMPANY
welding industry’s growth. 8th meeting 161 PLEASANT STREET, LYNN, MASSACHUSETTS
€Ynx 2-7710
For details, circle No. 11 on Reader information Card
WELDING JOURNAL | 1073
 owe |, wwe FT Veen UCU

New Members EFFECTIVE AUGUST 1, 1960

MEMBERSHIP CLASSIFICATION
A—Sustaining Member D—Student Member
B—Member E—Honorary Member
C—Associate Member F—Life Member
TOTAL NATIONAL MEMBERSHIP
Sustaining Members
Members
Associate Members
Students
Honorary Members
Life Members

AWS Builds Men of Welding

BATON ROUGE MILWAUKEE PHILADELPHIA MEMBERS NOT IN SECTIONS

Doran, Hugh T. (B) Leinenkugel, James M. (C) Dale, Bjarne J. (B Greenfogel, Paul R. (A)
Stanyon, Harold (C Henes, Charles J. (B Shelley, John William (B)
BOSTON Vignola, Mario (B)
Myerow, Sumner W. (C) MOBILE PITTSBURGH
Lehman, Robert (B Members Reclassified
Hickman, T. K. (B)
CANADA During August 1960
Culumovic, Stephen (B) NEW JERSEY PUGET SOUND
ATLANTA
Forgie, John D. (B Otteau, Henri J., Jr. (B Judson, C. J. (B
Tyrrell, W. A. (B Hrisinko, George (B to C
NEW YORK SABINE INDIANA
CHICAGO
Nichols, Herbert L., Jr. (B) Faucett, W. T., Jr. (B Hurcomb, F. A. (C to B
Bishop, Charles A., Jr. (B)
Bowers, Fred M. (C SAGINAW VALLEY LONG ISLAND
Chamberlain, Ron (C NIAGARA FRONTIER
Chessman, L. F. (C) Brown, F. D. (A Browning, Roy T. (C Payne, Kenneth W., Jr. (D to
Freisinger, Anton (B C)
Hess, Wendell T. (B NORTH TEXAS SALT LAKE CITY
Kostyal, Stephen Peter (B) NEW JERSEY
Goodfellow, William J. (B) Myers, Fred J. (B)
Owen, R. L. (C) Bell, Richard S. (D to B
CLEVELAND
NORTHWEST NORTHERN NEW YORK
Cheng-Teh, Shen (C) SAN FRANCISCO
Rosys, Stan (C Nippes, E. F. (C to B
Esplin, Randy (C)
COLORADO Greb, R. W. (B)
ORANGE COUNTY PHILADELPHIA
Radcliff, Willis M. (B Mahoney, John W. (B
Dietz, Richard C. (B) Betz, Irving G. (C to B)
COLUMBUS Gineyko, Nicholas A. (C TULSA SAN FRANCISCO
Conklin, Delbert E. (D) Larson, Dan E. (B)
Parks, Jack (B) Coombs, Leo E. (B Coops, William J. (D to B)
DAYTON Regan, George E. (B Schaub, Paul (B) Gallaher, Morgan (C to B)
Regan, John A. (B
McNeeley, Enos O. (B Royer, Chas. L. (C WASHINGTON, D. C. SANTA CLARA VALLEY
HARTFORD Van Cleave, H. D. (B) Simpson, Ross E. (C Weare, Norman E. (C to B)
Nessler, Charles G. (B
J. A. K.
Stefanich, Ronald J. (C MEMBERSHIP IN THE AMERICAN WELDING SOCIETY
LEHIGH VALLEY helps you improve your product, increase your production and
Lee, Warren S. (B lower your welding costs. You'll have for your own use latest
LONG BEACH available welding ‘‘know-how,”’ including the Society’s Welding
Dendy, John W. (B Journal and Welding Handbook. How you can join the Society
LONG ISLAND and take advantage of its many benefits is explained in descrip-
Platz, R. M. (B) tive literature available.

LOS ANGELES For further details write to:

Ellis, Herbert B. (C AMERICAN WELDING SOCIETY
MADISON 33 West 39 Street, New York 18, N. Y.
Bruschi, Daniel J. (C)

1074 | OCTOBER 1960

 NEW WEAR-O-MATIC WH Produces High Strength

Manganese Attachment Welds Equal to 308 Stainless

For PAGE ENGINEERING COMPANY

The Page Engineering Co.

was founded in 1903 by
Mr. John W. Page who
originated the dragline
method of digging. Today
Page Engineering is the
nation's leading manu-
facturer of dragline
buckets, walking drag-
lines and other
equipment used in
dragline operations

High strength properties and crack free deposits are the two major require-
ments for welding alloys used to attach manganese parts to PAGE AUTOMATIC
BUCKETS made of high strength alloy steel. Until recently, only type 308 stain-
less manual electrodes satisfactorily met both requirements. Other alloys have
been tested from time to time but failed one or both tests leaving Page no choice
but to continue using 308 stainless electrodes for critical joints.
Now, new Wear-O-Matic WH semi-automatic wire developed by Alloy
Rods Company fills both the requirements by producing high strength properties
identical to those of 308 stainless electrodes while depositing absolutely crack-
free weld metal. The Page Engineering Co. has replaced most 308 stainless weld-
ing in their dragline bucket production with new Wear-O-Matic WH alloy,
gaining the economies of semi-automatic open are welding and retaining the same
high strength requirements for manganese attachment welding. Page engineers
report Wear-O-Matiec WH gives them the additional advantages of faster deposi-
tion and smoother are characteristics—better than any semi-automatic wire they
have ever used in the production fabrication of dragline buckets.
You too, can benefit from the experiences of Page Engineering. For any
application involving the high strength attachment welding of manganese, whether
it be production fabrication of manganese parts or field maintenance, Wear-O-
Matic WH will do the job better than any alloy you are now using and do it
semi-automatically. Contact your Alloy Rods Company distributor or Hard
Surfacing Specialist for specific application information. Or write direct to Alloy
tods Company, P. O. Box 1828, York 3, Pennsylvania.

ALLOY RODS COMPANY

YORK, PENNSYLVANIA
SALES OFFICES & WAREHOUSES; BOSTON, NEWARK, PHILADELPHIA, PITTSBURGH, BIRMINGHAM, CHICAGO, SAN FRANCISCO & EL SEGUNDO, CALIF.—DISTRIBUTORS IN ALL OTHER PRINCIPAL CITIES

TWENTY YEARS OF LEADERSHIP IN THE DEVELOPMENT OF QUALITY ALLOY ARC WELDING ELECTRODES
For details, circle No. 12 on Reader information Card
Airco Expands Plant For the first seven months of Harper Trucks Opens Warehouse
1960, new business and shipments
A million dollar expansion of Air Harper Trucks, Inc., Harper,
are 14 and 21%, respectively, ahead
Reduction Pacific Co.’s new air Kans., announces the warehousing of
of 1959.
separation plant at Richmond, their complete line of welding cyl-
Backlogs of more than $11'/,
Calif., was announced recently. inder trucks by their companion
million were reported at the end of company, Tweco Products, Inc.,
The expansion program will boost July.
the rated capacity of the Richmond at their Eastern Division Office and
installation to more than 65 tons Warehouse in West Caldwell, N. J.
per day of high purity liquid oxygen, Owen Firm Expands in Canada This warehouse stock is for the
nitrogen and argon. The new unit convenience of their distributor
will produce 32 tons per day of the Rubery Owen, Canada Ltd., an- organization and to make possible
gases and is scheduled for completion nounces the introduction of “‘Ro- better service to consumers in the
early in 1961. wen-Arc’”’ automatic welding equip- eastern seaboard states plus Ver-
ment and continuous electrode wire mont, Pennsylvania and West Vir-
Metal & Thermit Declares in Canada and plans to appoint ginia.
Dividends distributors in each Province.
““Rowen-Arc”’ equipment is manu-
Directors of Metal & Thermit factured in England under license
Corp., at their regular meeting, from the Westinghouse Electric
declared a dividend of 30 cents per Corp. and is identical with ‘““West-
share on the common stock payable ing-arc’’ equipment which is made COMING
Sept. 12, 1960, to shareholders of in the U. S. Rowen Arc will carry
record at the close of business on the same warranty as West-ing-
Sept. 2, 1960. arc and identical parts numbers. EVENTS
The directors also declared the All parts will be interchangeable.
regular quarterly dividend of 87'/,. Initially the equipment will cover
cents per share on the preferred A Calendar of Welding Activity
200 to 500 amp power sources, el
stock, payable Sept. 26, 1960, to 200 to 600 amp dynamic reactors,
stockholders of record at the close automatic, semiautomatic and man- AWS
of business on Sept. 16, 1960. ual guns, automatic controls and Oct. 13-15. Third Western Weld-
a full range of bare, copper flashed ing Show, Exposition Hall, Santa
Westinghouse Announces New and coated continuous electrodes. Clara County Fair Grounds, San
Department Name Jose, Calif. Western Welding
The Westinghouse Electric Corp. Technical Conference, St. Claire
A. 0. Smith Signs Seven Hotel.
has announced that the Welding
Department and its products will Exclusive Distributors 1961 Annual Meeting & Exposi-
now be designated by the name tion: April 17-21. Hotel Commo-
Seven welding equipment dis-
West-ing-arc, in recognition of the dore, New York, N. Y.
tributors in West Texas and New
fact that its products are based on Mexico recently broke with their
the principle of the electric arc. AEC
long standing policies of multiline
Formerly West-ing-arc applied Oct. 4-6. Southwest Research
sales and contracted to sell A. O. Institute—Welding Forum, Hil-
largely to the process of gas-shielded Smith Corp. welding products on an
welding developed by Westing- ton Hotel, San Antonio, Texas.
exclusive basis according to R. W.
house. Raney, general sales manager. ASM
The new distributors are: Plains Oct. 17-21. 42nd National Metal
RWMA Reports Excellent July
Welding Supply, Plainview, Tex.; Congress and Exposition. Con-
Resistance Welding business dur- Western Air Products Co., Lubbock vention Hall and Bellevue-Strat-
ing July climbed well over the $3 Tex.; West Texas Welders Co., ford Hotel, Philadelphia, Pa.
million mark, according to the Midland, Tex.; Western Oxygen
monthly statistics compiled by the Co., Odessa, Tex.; Pecos Welding NWSA
Resistance Welder Manufacturers’ Supply, Pecos, Tex.; and Western Nov. 28, 29. Southeastern Zone
Association. The new business Oxygen of New Mexico with sep- Meeting. Hillsboro Hotel,
represented a 47% increase over arate headquarters in Hobbs and Tampa, Fla.
June. Roswell, N. M.

1076 | OCTOBER 1960

 This is the first industrial X-ray unit
NEW FROM WESTINGHOUSE
designed specifically to meet all radiographic
inspection requirements, from thin
aluminum up to 4” thick steel. Simple to
BALTOGRAPH 300
operate ... requiring only a few
hours instruction.
PENETRATES
For a free demonstration—at your plant or

UP TO 4° OF STEEL job site—just call, wire, or write to

Mr. Dan Maus, X-ray Department,
permits 360° radiography and can
Westinghouse Electric Corporation, 2519
be coned down to a single beam... Wilkens Avenue, Baltimore 3, Md.—You can
be sure...if it’s Westinghouse.
yet weighs only 136 lbs.

SPECIFICATIONS

X-ray Head

X-ray tube, lead shielding, HT trans-

former, filament heating transformer,
and automatic thermal cutoff device are
all mounted in a single lightweight
housing.
X-ray portal: Total X-ray beam is 360
circumferential
Focus size: 3x 3 mm
Inherent Filtration: Low
Shielding: 5 mm of lead
Power rating: Self-rectified 300 KVP-
3MA
Weight: 136 lbs.
Dimensions: 4614 long, 1214 diameter

Control Panel

Hermetically-sealed penetration meter,

penetration selector, MA control, and
automatic push button reset timer
mounted in rugged metal cabinet.
Power Supply: 115 or 230 volts, 60 cycles,
single phase
Weight: 41 lbs.
wha
Dimensions: 15” x 12” x ‘
Also included: 25’ and 60’ connecting
cables and individual carrying cases.

J-08386

Westinghouse

For details, circle No. 13 on Reader information Card

WELDING JOURNAL | 1077
 SUPPORT YOUR SOCIETY DEDICATE HEADQUARTERS
SYVTRON

SELENIUM
Weldmatic Selects Representative
Weldmatic Division of Unitek
Corp., Monrovia, Calif., has ap-
pointed Tech-Ser, Inc., to represent
WELDING

them in California, Nevada and

Arizona. Tech-Ser will handle
Weldmatic’s entire line of precision
stored-energy welding equipment for
applications such as: electronic
STACKS

component packaging and vacuum

At the dedication of the Salt Lake City
tube assembly; welding thermo- headquarters of C & C Welding Supply
couple and fine wire junctions and Co., Governor G. D. Clyde of Utah pre-
joining extremely thin foils and pares to cut the chain as Francine Felt,
screens. ‘Miss Utah,’ and |. G. Kepner, Linde re-
gional manager, and H. Coulfield of C&C
Ransome Distributors Announced stand by

The appointment of three new

distributors has been announced Coyne Cylinder Merger
by the Ransome Co., Scotch Plains,
N. J. John P. Coyne, president of
Bakers Welding Supply Co., Coyne Cylinder Co., 224 Ryan
Bakersfield and Fresno, Calif., has Way S., San Francisco, announces
been appointed distributor for the the consolidation by exchange of
San Joaquin Valley area; the Weld stock wtih American Cryogenics,
Tooling Corp. of Pittsburgh, Pa., Inc., of Acetylene Cylinder Corp.,
has been given the Western Penn- Farwest Equipment Corp. and
sylvania territory; and Upstate Coyne Cylinder Co., all of San
New York has been assigned to the Francisco, and the Sierra Oxygen
Charles S. Freeman Co., Buffalo, Co. of Reno, Nev.; also, the
M..z. acquisition of McKesson Appliance
Co. of Toledo, Ohio.
Distributor Advisory Council
Welded Abstraction
A distributor advisory council
has been established by Linde Co. A 6-ft high abstraction made of
The 15-man council plans to meet welded nuts and bolts is the creation
--the industry of Professor Egon Weiner of the
twice a year to discuss such subjects
standard for as product development, promotion Chicago Art Institute.
and policies affecting the welding An example of the use of welding
more than in achieving new and unusual effects,
supply business. The first meeting
seven years was held at Linde’s Home Office
in the Union Carbide Building, 270
Park Ave., New York, N. Y., on
Sept. 21-22, 1960. The 1961 Linde
council consists of nine distributors
and six top managers from the
Linde Co. Distributor Products
Department.
Experience by en major manu-
facturer ince 1952 shows less Allbright’s Forms Division
than 2% have failed in field service
SYNTRON Welding Rectifier Allbright’s, a welding supply
-Speciali ts are reaay| to serve you distributor serving Riverside and
San Bernardino counties, Callif.,
Write for information and specifications has changed its name to Welding and
Industrial Sales, Division of All-
SYNTRON RECTIFIER DIVISION bright’s, it was announced recently.
Headquarters for the firm’s 20
258 Lexington Ave Homer City, Penna
stores will be a new 50- x 150-ft
Sales Engineers in; New York, Chicago, Los An- building, located at Eighth St.
geles and Canada and Fairmount Blvd., Riverside. the work symbolizes the biblical
Canadian Manufacturing Plant: Syntron (Canada) Originating as an automotive “Burning Bush”’ and will be used to
Ltd., Stoney Creek, Ontario
Export Representative: Dage Corporation, 219 E parts company in 1921, Allbright’s represent the vital way in which
44th Street, New York, N. Y became a distributor for the Linde nuts and bolts support modern in-
Sales and Engineering Representative: Robert O dustry. The work was commis-
Whitsell and Associates, 6620 East Washington Co., in 1931, handling oxygen,
street, Indianapolis 15, Indiana Offices in Cleve- acetylene and welding apparatus sioned by Heads and Threads, Inc.,
land, Dayton and Cincinnat
For details, circle No. 14 on Reader Information Card and supplies. Chicago, II.

1078 | OCTOBER 1960

 Metco Inc. Is New Name
Expansion of the company’s ac- Profit through
tivities into a wider range of
flame spraying processes has dic-
tated a change in name from E.A.I. Research
Metallizing Engineering Co., Inc.
to Metco Inc. Effective Sept.

Uniform, through-and-through heating of the

Ist, the company expects its new thickest sections of the big stainless steel
name to be more descriptive of its valves was provided by the Smith-Dolan
products. These include the flame method, with precision control of temperatures
spraying of refractories, special (2-hour dwell at 1900°F.). Drawing shows
cross-sectional! dimensions.
hard-facing alloys and _ tungsten
carbide in powder form and plasma-
flame spray equipment, in addition
to high-speed wire metallizing guns. Standard Smith-Dolan units come equipped
with eye hooks and wheels for easy portability,
Welding Supply House Expands and with instrumentation for recording per-
manent charts of the heat run.
The Welding Supply Division of
H. M. Parker & Son, Glendale,
Calif., has moved into new head-
quarters at 11245 Van Owen St.,
North Hollywood. SMITH-DOLAN
The $300,000 headquarters stocks
the latest equipment in the welding
supply industry, including modern On-the-Job HEAT TREATMENT
facilities for handling both liquid
and gaseous oxygen, in addition to Solution to Weld Embrittlement Problem
acetylene, nitrogen, hydrogen and
argon. at Consolidated Edison
The building site totals 86,000 Public utilities are among the largest users of the Smith-Dolan System
sq ft and includes a warehouse, of Low-Frequency Induction Heating. They want to keep their service
sales and administrative offices, uninterrupted, their costs low, and they find that this portable system
modern showroom and off-street simplifies the problem of pre-heating and post-annealing in producing
parking. Complete facilities to sound welds, and in making dependable repairs.
demonstrate advanced welding proc-
esses are an important feature of A typical application of the system by a public utility was described in a
the new headquarters. An exposi- recent article in Welding Design & Fabrication.* At Consolidated Edison's
tion and open house were held Sept. No. 10 Turbine Generating Station in New York City, two large steam
14-15, 1960. pressure valves made of Type 347 stainless had to be repaired because of
weld embrittlement. Removal to the shop for heat treatment would have
Aurora Welding Service, Inc., proved costly and time taking. With the Smith-Dolan portable units, the
Purchased work was completed on the job within 12 hours.
Purchase of Aurora Welding Learn about the savings and other benefits your company can obtain from
Service, Inc., Aurora, Ill., was an- the Smith-Dolan System. It is today’s most versatile heat treating method,
nounced recently by H. B. Van simplifies and upgrades work quality in numerous applications. Write for
Cleve, Jr., West Chicago, and D. brochure, “Smith-Dolan System of Low-Frequency Induction Heating.”
Stephens, LaGrange, who acquired *Reprints available without charge
the firm located at 313 Gale St.,
from G. E. Lidecka, former owner. Rental Against Purchase — Get first-hand proof in
Mr. Van Cleve will serve as presi- your own plant without capital investment.
dent and general manager and Mr.
Stephens as vice president and
general sales manager. DEMAND THE
Aurora Welding has also been BRAND MADE
named authorized distributor for BY PEOPLE
Air Reduction Sales Corp., Van WHO BUY YOUR 152-10 Jelliff Ave.
Cleve said, and will supply every- PRODUCTS Newark 8, New Jersey
thing?required by welders, including
For detaiis, circle No. 15 on Reader information Card
WELDING JOURNAL | 1079
bottled gases, welding rods, elec- RAYNO OPENS FOURTH PLANT
trodes, goggles, helmets, gloves,
electrode holders and other welding
equipment.

Norelco Completes First

Hydrogen Generator
Completion of the first Norelco
hydrogen generator which will pro-
duce 1000 cfh of ultra-pure hydrogen

(less than 50 ppm of impurities)

has been announced by Cryo-
generators, Inc., Ashton, R. I.
The machine was built for in-
stallation in the plant of Com- A three-day welding show marked the opening of Rayno Distributors, Inc., fourth
mercial Steel Treat at Detroit, sales office, plant and warehouse at 1002 Alabama Ave., Brooklyn, N. Y. Shown
Mich. is automatic shape cutter.

HARD FACING PROBLEMS? Authors...

FIELD TESTS PROVE
WELDING TIME REDUCED please note!
MORE THAN 50%
All authors interested in presenting papers at the
WITH MIR-O-COL M-700 1961 AWS National Fall Meeting to be held in Dallas,
The Semi-Automatic Open Arc wire Texas, on September 25-28, 1961, are advised that
feed unit is designed for the applica- the usual forms, “An Invitation to Authors” and
tion of specially fabricated alloy ‘‘Author’s Application Form,” are printed as a
wire by the open arc process. It op-
erates on AC and DC or CV detachable insert on pages 1051 and 1052 of this
welders at 175-600 amperage range. issue of the Welding Journal instead of being sent
Deposits 12-20 pounds of 7/64” wire through the mails.
in one hour at average 265 amp.
Small, compact, weighs but 55 pounds.

NEW AUTO-VIEW WELDING HELMET

MIR-O-COL HARD FACING all of the best features of conventional helmets
plus . . . slight jaw motion raises filter lenses.
ELECTRODES SAFETY Eye protection during chip-
ping. Hands free at all times for support. No
Hard Facing Alloys for Manual loss of balance from hand lifting.
And Automatic Applications. Special
WORK QUALITY Bead may be
Alloy Castings. Automatic Positioners | started precisely where desired. Eliminate
And Welding Heads. scoring and drag beads. Work is continued
easily after inspection, without reorientation.
For complete information write: TIME SAVING No more awkward
ae hand lifting. Increased production with less
adjustment effort, especially in tight places.
MUR-D-COL ~~ AUTO-VIEW Welding Helmet Co.
replaceable AV P.O. Box 917. Santa Monica, Calif.
lenses
For details, circle No. 16 on Reader Information Card For details, circle No. 19 on Reader Information Card

1080 | OCTOBER 1960

GOVERNOR CUTS SILVER RIBBON
BATEMAN

BANTAM
IRON WORKER

THE ONLY IRON WORKER OF ITS

KIND ON THE MARKET TODAY
No Grinding Neces-
sary After Cut. One
Stroke Cycle Clutch
Operated by Hand
or Foot.
The Bateman “‘Ban-
tam” cuts 2” x 2” x
4” angles and 4” x
4” flacs. Standard
punches will fit this
machine. The Coper
will cope 1%
through 4%” material
Ie will punch 2”
hole through 4” ma-
terial. With che clutch
open, the Bantam will
Governor Freeman of Minnesota cuts a silver-alloy ribbon to mark the make 44 strokes per minute. It is made of
opening of a two-day welding show held at the new Minneapolis center of high-grade cast iron, with the clutch, pin
Eutectic Welding Alloys Corp. and dog made of hardened steel. The blades
are made with tool steel. It is powered with
a fly wheel and gear drive, and uses a smal!
% hp motor, 1750 rpm
Bateman Bantam with punch ._$575.00
Shear only . $495.00
Shipping wt. 750 ibs.

BATEMAN FOUNDRY & MACHINE

WESTERN WELDING SHOW MINERAL WELLS, TEXAS
The Third Western Welding Show while the Welding Show will be at n
and Technical Conference, to be the Santa Clara County Fair- ere ee
held in San Jose, Calif., Oct. 13 grounds. Complete program details
15, will feature a 2'/, day technical of the technical session were given
papers session concurrent with a in the September issue of the Jour-
welding exposition. NAL. To date 26 firms have con-
The technical conference will tracted for floor space at the Weld-
be held at the St. Claire Hotel ing Show. g Z Tops in the

Industry!
atin

| ATTEND sitih
i
ity a
{/ Ai‘
WESTERN WELDING SHOW
STERN WELDING E
W TECHNICAL CONFERENC
4.

The Welding Journal is the

world’s most authoritative welding
and allied process magazine; it
has fully 21% times the editorial
content of any compeling maga-
zine; it is unequalled in coverage
of welding engineering, research
and application. If you have a
product for the welding or allied
Charting exhibit areas for 3rd Western Welding Show in San Jose, Calif., industries, the Journal's pages
Oct. 13-15, are (left to right) Don Powell, Linde Co.; Arnold Otto, will reach your market!
Hipp Welding Co.; and Harry Gerin, Gerin Welding Sales. .

WELDING JOURNAL | 1081

 a degree of Mechanical Engineering Mr. Migel received his B.A. degree
in 1927. Prior to 1935, when he in Physics from Colgate University
joined Airco, he was engaged in the in 1934 and received a B.S. degree in
design and engineering of a wide Mechanical Engineering from MIT
range of heavy industrial welding- in 1936. He joined Magnaflux
fabricated products. With Air Re- Corp. in 1940 and has worked in the
duction Mr. Kobel was engaged in field of nondestructive testing since
the various engineering aspects of that time.
electrode manufacturing. In 1942
he was appointed over-all factory Moore, J. C. and G. R. Grant Advance
manager. He is member of the Thomson Electric Welder Co.,
Canadian Welding Society and of Lynn, Mass., specialists in resist-
ASM, having held executive offices ance welding equipment, recently
in each society. announced the promotion of C. D.
Moore, J. C. Grant, Jr., and G. R.
Nyberg To Represent Grant to executive sales positions.
Nottingham
Nyberg Equipment Co., headed
by Ned Nyberg, who has been ac-
tive in the welding industry since
Harrington Named Manager 1936, has been appointed by J. B.
B. G. Harrington 3 has been Nottingham & Co., Inc., to repre-
appointed manager of Air Reduc- sent its multiarc welding and strip-
tion’s Kansas City district, it was heater preheat equipment and sys-
announced recently. As _ district tems.
manager, Mr. Harrington is re- With offices at 1229 S. Laclede
sponsible for the sale and distribu- Station Rd., St. Louis 19, Mo.,
tion of all Airco products marketed the Nyberg Co., will serve that city
through the Kansas City facilities and adjacent industrial centers in
located at 2701 Warwick Traific- Missouri, as well as western Illinois
way, Kansas City 8, Mo. Mr. and northwestern Kentucky. The
Harrington succeeds J. O. McElli- firm is sales representative for sev-
gatt who retired on July 1, 1960. eral lines of welding and cutting
Mr. Harrington, a graduate of equipment.
Mr. Nyberg is now chairman of N. Nyberg
Franklin and Marshall College,
Lancaster, Pa., has been employed the St. Louis section, AWS.
with Air Reduction Sales Co. since
1941 in various posts. Most re- Migel Made Vice President
cently he was assistant manager H. Migel 3 has been appointed
sales in Kansas City, a position he vice president for new products and
has held since 1953. product development for the
Active in civic and social affairs, Magnaflux Corp., Chicago, accord-
Mr. Harrington is a past chairman ing to a recent report.
of the Kansas City section of AWS. Migel’s appointment was one of
five new posts created by the com-
Kobel Appointed Vice President pany. J. E. Heath was named
The appointment of O. Kobel vice president of manufacturing;
to vice president of Air Reduction D. T. O’Connor, director of re-
Canada, Ltd., was recently an- search and development; A. E.
nounced. Christensen, manager of engineer-
~ Mr. Kobel was born and edu- ing, and H. N. Nerwin, chief en-
cated in Europe where he receiyed gineer-electronics.

B. G. Harrington O. Kobel C. D. Moore

1082 | OCTOBER 1960

 }
‘COSTS LESS BECAUSE IT #t

gets hard and stays tougher under ;

MANGANAL

Fen
’ ATOR BARS

WELDING ELECTRODES Workirdons to a high degree.

@ FLO-KOTE Se Toughest
| steel
; commercially produced.

@ SPECIAL TITE-KOTE Easy application . .. cuts down time.

@ BARE Costs less than new parts.
Keeps equipment producing longer.

NEAREST DISTRIBUTOR
UPON REQUEST
WRITE FOR
COMPLETE DETAILS
STULZ-SICKLES CO.
929 Julia Street e@ Elizabeth, New Jersey

Fer details, circle No. 20 on Reader Information Card

WELDING JOURNAL
 Mr. Moore who advanced to sales returned from a European trip
manager, is a designer, a holder of where, as a U. S. delegate, he at- SUPPORT YOUR SOCIETY....
several patents, and a specialist in tended two international meetings:
aircraft welding. He assumes di- the Annual Assembly of Interna- BE ACTIVE!
rect responsibility for domestic and tional Institute of Welding in Liege,
foreign sales, reporting to the presi- Belgium, and the Third Interna-
dent and to the general manager. tional Congress of Precast Concrete regional sales manager, has been
A resident of Byfield, Mass., he is Industry in Stockholm, Sweden. promoted to national sales manager
an active member of the AWS and Subsequent to these meetings, he for three Erico Divisions—The Cad-
frequently serves on technical com- went to Austria and received the weld Electrical Connection Div.,
mittees within this society. degree of Doctor of Technical the Caddy Arc Welding Accessory
J. C. Grant, Jr., who became as- Sciences from the Technical Insti- Div. and the Caddy Toggle Clamp
sistant sales manager, will direct tute (Technische Hochschule) of Div.
customer relations and maintain Vienna. Mr. Langhenry is a graduate of
close field representation for dealers Marquette University where he
and engineers. A senior member of Fullen, Collier, Bauer Promoted
received a Bachelor of Electrical
the organization and a resident of The Lincoln Electric Co., Cleve- Engineering Degree. Married and
Lynn, Mass., he has had many land, Ohio, announces the transfer the father of four children, he started
years experience in production and of D. J. Fullen from Philadelphia his engineering sales career with
engineering. to the York, Pa., sub-office. Fullen Westinghouse Electric Corp. fol-
G. R. Grant, who became factory replaces G. T. Collier who was lowing a tour of duty specializing
sales engineer, will coordinate adver- transferred to Cincinnati, Ohio. in electronics in the U. S. Navy.
tising and promotion programs and T. Bauer, a recent appointment Mr. Langhenry joined Erico Prod-
will be responsible for production to Lincoln’s staff of field welding ucts, Inc., in 1954 as a Chicago
quotations and for disseminating en- engineers after completing a one sales engineer and in 1956 was
gineering data to the field staff and year training course at the factory, appointed Chicago sales manager.
to customers. has been assigned to Philadelphia. In 1958 he was promoted to the
position of Midwestern regional
Amirikian Honored Abroad Langhenry in Top Sales Post sales manager.
Arsham Amirikian 53, special Erico Products, Inc., Cleveland,
structures consultant of Navy’s Ohio, recently announced that E. Wylie Named Manager
Bureau of Yards and Docks, has Langhenry, formerly Midwestern
R. D. Wylie @3 has been named
manager of quality contro! for the
manufacturing department of The
Babcock & Wilcox Co. Boiler Divi-
sion, it was announced recently.
Mr. Wylie received a B.S. degree
in Metallurgical Engineering from
the University of Michigan. He
joined B&W’s quality control labora-
“We sell tory, Barberton, as a metallurgist in
1948. In 1957 he became chief
metallurgist and in 1959 was named
assistant manager of quality control.
not
Omdahl Made Controller
overhead Lyle N. Omdahl has been ap-
pointed controller for Omark In-
dustries, Inc., according to a recent
announcement.
Mr. Omdahl was formerly con-
troller for the W. A. Sheaffer Pen
Co. He is a graduate of the
University of North Dakota.

Coyne Appointed Vice President

The Board of Directors, Victor
"That's why we chose COYNE CYLINDERS—the acetylene cylinders that Equipment Co., announces the ap-
hold more gas. COYNE’S extra gas capacity means we sell more gas pointment of E. J. Coyne as vice
in every cylinder delivered to our customers. At the same time we also cut president, Gas Division. He will
our handling, trucking and administrative costs because it takes fewer be located at the San Francisco
cylinders to deliver the gas.” executive address of the company.
Check with COYNE — let them send you the dollar and cents facts on how Victor manufactures and distrib-
much profit you get from this extra gas footage. utes industrial and medical gases
in the State of California, and main-
224 RYAN WAY, SOUTH SAN FRANCISCO, CALIFORNIA, PLAZA 6-6910 tains gas manufacturing plants in
COYNE 155 WEST BODLEY AVENUE, MEMPHIS, TENNESSEE, WHITEHALL 8-7789 San Diego, La Habra, and Sacra-
cylinder company 3800 SPRINGDALE AVENUE, GLENVIEW, ILLINOIS, PARK 4-3828 mento, California; and direct dis-
24 COMMERCE STREET, NEWARK, NEW JERSEY, MITCHELL 2-1618 tributing outlets in 12 locations
For details, circle No. 21 on Reader information Card throughout the state.

1084 | OCTOBER 1960

 a KAA | Ai rt Vy
min Hi
My y
ii!
i Higa Hi )
AX)

H
E
A
D

Gassy molybdenum being welded in NRC

Model 2405. NRC Electron Beam Welders are
available with a variety of chamber sizes
and with single or multiple guns. They are
surprisingly inexpensive and easy to use.

Can electron beam solve your

What’s So Wonderful About tough welding problems?

Electron Beam Welding?
Freedom from contamination; Mini-
mum heat affected zone; Unlimited
temperatures; Depth/width ratios to NRC Small Lot Welding Service
4/1; Precise control of spot size and Gives Low Cost Answer
position; Operation in high vacuum.
These advantages are daily solving Here’s a low cost, no risk way to find out if electron beam welding
problems posed by reactive metals,
dissimilar materials, heat sensitivity, can solve your problems as successfully as it has those of others.
and unusual geometry. The NRC Small-Lot Welding Service was set up to prove that
the Model 2405 Electron Beam Welder can indeed often “‘weld the
unweldable.”’ It gives you the chance to see exactly what the com-
bination of superior equipment and experience can do with your
problem welds. And, if you like, you can be your own operator!
The cost? NRC charges only for time and materials. Our prime
interest is to help you discover whether electron beam welding is
practical for your specific application.

Typical Solutions
Try The NRC Small-Lot Welding Service
@ Inaccessible joints made by ‘‘continu- The solution to your problem may be waiting for you right now
ous spike weld” (illustrated). at NRC Equipment Corporation. The sooner we hear about your
@ Seam welded tungsten and tantalum
tubes. application, the sooner we can
Delicate instruments evacuated and tell you whether electron beam
sealed. welding sounds practical, and,
Sintered plate welded without cracking. if appropriate, add you to our pe ru <a
Finish machined parts joined without work schedule. Or, send for SI
distortion.
Precision bead at bottom of 1/16” slot. general information on Electron sg RC
Titanium insert edge-welded to copper Beam welding and details on
sleeve. Model 2405, available for early Ps dl
Thin and thick sections joined easily. delivery at less than $17,000.00.
Crackless welding of glass-sealed feed-
thru to sleeve.
A Subsidiary of National Research Corporation
Dept. J-9, 160 Charlemont St., Newton 61, Mass.
See Model 2405 in action at the Metal Show, Booth No. 1301. DEcatur 2-5800
For details, circle No. 22 on Reader information Card
WELDING JOURNAL | 1085
 BIG REASONS Services Available
EMPLOYMENT
WHY A-739. Welding Superintendent
SERVICE Specialist. Twenty years experience
SO MANY in fabrication—all welding processes
WELDERS HAVE BULLETIN using ferrous and nonferrous metals.
Can train operators, and set up de-
SWITCHED TO partment. Capable of producing qual-
ity weldments. Experience includes
Positions Vacant 3 years working in metallurgical lab-
HI-AMP. The Board of U. S. Civil Service oratory division on research—quality
ELECTRODE HOLDERS Examiners, Detroit Arsenal, 28251 control, liaison work on subcontracts,
Van Dyke Road, Center Line, Mich., and consultant to all engineering
Wrap cround Glass Fibre Tip In- has openings for two welding engineers, divisions from design to final acceptance
sulation—30% more heat resistant GS-11 positions with a starting salary of pilot model and field tests. Desires
than any other make. of $7560 per year. To conduct labora- contact with medium to small pro-
Brilliont Red Tips and Trigger- tory research investigations and ex- gressive company with emphasis on
Bright Yellow Handle—all Glass perimental welding projects relative future. Can furnish resume and
Fibre, an outstanding Safety Fea- to the development of new or improved references on request.
ture.
Body completely insulated—no bare welding, flame-hardening and flame- A-740. Manufacturing Engineer of
spots. cutting techniques and _ procedures Welding or Applications Engineer.
for use in the fabrication, maintenance Twenty-one years of diversified ex-
and repair of military tracked and perience in fusion, resistance, brazing.
wheeled vehicles, components and Some metallurgy experience and school-
Ss parts; to analyze service failures; ing. Good background—alloy, stain-
to set up test procedures and to prepare less, magnesium, aluminum, automatic
evaluation reports and recommenda- and hand. Resume on request.
tions on tests conducted. Any gradu- A-741. Welding Technician. Seeks
Just ask your Welding Sup- ate engineer of a recognized school position in the United States or
ply Dealer for PROOF of who has a background in welding Canada. Experience includes five
the above statements. engineering is encouraged to file an years apprenticeship as a_ welder.
SF-57 (application for Federal em- Welding Engineer (U. K. status) on
ployment) obtainable at most post oil depot, refinery and oil and natural
offices, or from the above office upon gas pipe lines. Technical sales repre-
LENCO, Inc. Mir} request. Interested applicants mail sentative (18 months). Chief Weld-
JACKSON, MISSOURI H!-AMP their applications to the attention of ing Inspector for two gas companies.
Mr. Leo Gordon. Age 35.
For details, circle No. 23 on Reader Information Card

CUT BEVELING TIME BY 85%

For Accutllo WELDING HEATS...

on Small Weldments

USE THIS PULLMAX

= 3 "4 3

The PULLMAX Beveller

can be used to bevel cut
large size plates with
WELD-CHEK proper table support and
feeding arrangement.
THERMOMETER But where the PULLMAX
MODEL 573-FM. ACTUAL SIZE Beveller really pays
Indicates up to off is in small segments
difficult to bevel cut by
1000°F. burning and grinding.
Accuracy +2% Feeateo SURFAC Bevel small
plate in both
Straight and
Now you can check these and many other surface temperatures mild radius
quickly, accurately . . . shapes at
ALL STAINLESS STEEL 50°F speeds up to
to 1,000°F. Stability within 60 @ HIGH CARBON STEELS 11 ft. per min.
seconds easy to read. May be and other metals
left on material or area to be tested @ MANGANESE
during pre-heating period ... need @ PRE-HEATED MAGNESIUM ’
not be removed between readings 1 Cut Mild Steel
User can reset, if necessary. Small @ HARD FACING
magnetic clamp holds thermome- @ FURNACE WELDING
ter in place on vertical or slanting Model 573-FM sqecc LL a 2 Cuts Stainless Steel
ferrous surfaces. Diameter 1%", with Magnet Clamp
weight % oz Model 575-FM, same but with maximum- Write today for literature and information
Your order will be filled minimum temperature indicators, $4.75
through your nearest dealer Phone GRanite 8-1134 or write Dept WJ1060 AMERICAN PULLMAX CO., INC. iagaroeeaas
PACIFIC TRANSDUCER CORP. [1g in eee See us, Booth 1756, Metal Show, P.O. Box 460 S. Service Road
Philadelphia, Oct. 17-21 Oakville, Ontario
For details, circle No. 24 on Reader information Card For details, circle No. 25 on Reader Information Card

1086 | OCTOBER 1960

Tempilstik: . ~~
2petaiire tndiialing cwayone

TEMPILSTIKS® provide a simple and accurate means of determining

preheating and stress relieving temperatures in welding operations)
Tempilstiks® are widely used as a standard method of checking temperatures
in all heat treating—as well as in hundreds of other heat-dependent processes
in industry. Available in 80 different temperature ratings..... $2.00 each.
Aost leading welding supply houses carry Tempilstiks®. If yours is an exception,
then write direct to us for further information.
#70
Tempil CORPORATION © 132 West 22nd St., New York 11, NL. Y.
Visit us at Bocth 1327-——A.S.M. Show——Philadelphia Trade & Convention Center—Oct. 17-21.
For details, circle No. 26 on Reader information Card
WELDING JOURNAL 1087
 Buildings Engr., vol. 45, no. 4 (Apr. 1960), pp.
Open-Web Steel Frames Cut Building 94, 96, 98.
Costs, F. E. Théiler and A. H. Marks. Repairing Deep Pits in Cold Expanded
Welding Engr., vol. 45, no. 4 (Apr. Pipe by Welding, T. A. Ferguson.
1960), pp. 76, 78, 80. Pipe Line Industry, vol. 11, no. 6
Business-machines Manufacture (Dec. 1959), pp. 22-26.
Current Welding Manufacture of Everest Office Ma- Gear Manufacture, Heat Treatment
chines, R. E. Green. Machy. (Lond.), Flame Hardening of Gears, J.T. Howat.
vol. 96, no. 2481 (June 1, 1960), Metal Progress, vol. 77, no. 4 (Apr.
pp. 1226-1235, no. 2482 (June 8), 1960), pp. 76-78.
LITERATURE pp. 1268-1274.
Germanium
Copper
Plastic Deformation of Germanium by
Practical Aspects of Welding of Copper Alloying, W. J. Feuerstein. Met. Soc.
by Gas-Shielded Bare Wire Process, of Al ME—Trans., vol. 218, no. 2 (Apr.
B. H. Baker and A. Jobber. Welding 1960), pp. 251-254.
& Metal Fabrication, vol. 28, no. 5
(May 1960), pp. 185-192. Hard Surfacing
Copper Alloys Hard-Facing with “‘Stellite’’ Powders,
Aluminum Bronze Welded Fabrica- J. R. Gault. Brit. Welding J., vol.
For copies of articles write directly to tions, M. Birkhead and C. V. Wilson. 7, no. 5 (May 1960), pp. 318-323.
publications in which they appear. AA list of Welding & Metal Fabrication, vol. 28,
addresses is available on request. Improving Wear Resistance by Hard-
no. 4 (Apr. 1960), pp. 162-168; (May), facing by Welding, E. N. Gregory.
pp. 198-205. Eng. Materials & Design, vol. 2, no.
Die Repair 8-9 (Aug.-Sept. 1959), pp. 425-429.
Aircraft Manufacture Die Repair Time Reduced by New Practical Guide to Hard Facing, C. W.
Welding Technique. Steel, vol. 146, Rogers. Western Construction, vol. 35,
North American Ready to Build B-60 no. 3 (Jan. 18, 1960), p. 89.
Prototypes, C. O. Herb. Machy. no. 1-A (Jan. 1960), pp. 46—49.
(N. Y.), vol. 66, no. 10 (June 1960), Diesel-engine Manufacture
Welding, Inert-gas
pp. 138-143. Electromolding Technique Trims Cost
of Parts. Steel, vol. 145, no. 24 New Consumable Insert Ring Extends
Aircraft Manufacture (Dec. 14, 1959), pp. 96-97. Use of Inert-Gas Tungsten-Arc Pipe
Puddie-Welding. Aircraft Production, Welding, H. Thielsch. Heating, Piping
Earthmoving Machinery & Air Conditioning, vol. 31, no. 11
vol. 22, No. 4 (Apr. 1960), pp. 145-148. Photoelectric Control Aids Weld Uni- (Nov. 1959), pp. 132-135.
Aircraft Materials formity, J. Melton. Welding Engr.,
vol. 45, no. 5 (May 1960), pp. 46—47. Light Metals
Welding of High-Strength Steels for Air-
craft and Missile Applications, H. W. Electric Appliances New Ways to Join Magnesium, T. L.
Mishler, R. E. Monroe, P. J. Riep- Patton. Modern Metals, vol. 16, no. 3
Assembling Washer Tubs, C. Reining. (Apr. 1960), pp. 46, 48.
pel. Battelle Memorial Inst.—DMIC Automation, vol. 6, no. 12 (Dec. 1959),
Report 118 (Oct. 12, 1959) 90 pp. pp. 53-56. Machine Design
Aluminum Sheet Electric Conductors Try Your Ideas with Plastic Models,
Giant Plate Stretcher Improves Proper- Effect of Elevated Temperature on A. J. Purcell and T. P. Theodores.
ties of Aluminum Alloys. Jron Age, Flash - Welded Aluminum - Copper Tooling & Production, vol. 26, no. 2
vol. 184, no. 27 (Dec. 31, 1959), Joints, C. R. Dixon aud F. G. Nelson. (May 1960), pp. 51-53.
pp. 45-46. AIEE—Trans., vol. 78, pt. 2 (Applica-
tions & Industry), no. 46, pp. 491-496. Metallizing
Automobile Manufacture See also Elec. Eng., vol. 78, no. 12 Hard Facing with Plasma Spray Guns,
Motor Car Production in Japan, (Dec. 1959), pp. 1190-1194. E. A. Gerhold. Brit. Welding J., vol.
R. E. Green. Machy. (Lond.), vol. 7, no. 5 (May 1960), pp. 327-330.
Electrodes
96, no. 2483 (June 15, 1960), pp. Sprayed Metal Coatings for Abrasion,
1426-1436, no. 2484 (Jume 22), pp. New Electrode Developed by Navy for
Welding Dynamically-Loaded T-1, Corrosion and Oxidation Resistance,
1552-1562, no. 2485 (Jume 29), pp. G. R. Bell. Brit. Welding J., vol. 7,
1640-1647. J. S. Kobler, Welding Engr., vol. 45,
no. 4 (Apr. 1960), pp. 44—45. no. 5 (May 1960), pp. 305-311.
Welding—Primary Production Method Reclamation by Metal Spraying, W. E.
for Motor Vehicles, A. W. Shearer, Feedwater Heaters
Ballard. Brit. Welding J., vol. 7, no. 4
N. M. Lloyd. Automotive Industries, Why Welded Feedwater Heaters? C. (Apr. 1960), pp. 223-230.
vol. 121, no. 11 (Dec. 1, 1959), pp. O’Connor. Plant & Power Services
43-60, vol. 122, no. 3 (Feb. 1, 1960), Engr., vol. 1, no. 7 (Nov. 1959), pp. Metals Cutting
pp. 35-37, 74, 78, 82, 86-88, no. 5 26-27. Theory and Application of Plasma Arc,
(Mar. 1), pp. 47-50, 58, no. 7 (Apr. 1), R. M. Gage. SAE—Paper no. T41
pp. 39-43, 86, no. 9 (May 1), pp. Flame Hardening
Modern Practice of Flame Hardening, for meeting Apr. 5-8, 1960), 9 pp.
55-56, 65, 76, 78.
M. R. Scott. Metal Progress, vol. 77, Metals Cutting, Plasma
Brazing no. 4 (Apr. 1960), pp. 69-73.
Plasma Flame Speeds Metalworking,
Induction Brazing Slashes Manifold- Foundry Practice J. A. Browning. Tool Engr., vol. 44,
Making Time, E. Altholz. Machy. no. 4 (Apr. 1960), pp. 105-108.
(N. Y.), vol. 66, no. 10 (June 1960), Operating Practice with Arcair Torch,
pp. 160-165. N.J. Krumm. Foundry, vol. 88, no. 4
(Apr. 1960), pp. 164, 167, 170, 172. Missiles
Brazing Nickel-Containing Alloys and Missile Optical Systems Housed in
Stainless Steels, J. Hinde and E. R. Gas Pipe Lines Mig Welded Aluminum Cones. Weld-
Perry. Welding & Metal Fabrication, Aluminum Piping Mig, Tig Welded for ing Engr., vol. 45, no. 5 (May 1960),
vol. 28, no. 4 (Apr. 1960), pp. 145-154. Natural Gas Distribution. Welding pp. 44-45.

1088 | OCTOBER 1960

 NEW All-Aluminum Offset Holder

You'll pay less for this new TUFFALOyY offset holder—

and get more benefits too. Only a complete departure
from previous manufacturing methods could come up
with this low-cost, high quality holder (each holder is
a ‘slice’ of a giant aluminum extrusion).
These light weight, corrosion resistant holders carry
threaded tip adapters that take the brunt of the wear,
always encountered in the tip socket. To change tip
size from a 4 RW taper to a 5 RW taper (Nos. 1 & 2
M.T.), you change adapters, not holders. Standard
AA ery) 2- and 4-inch offsets, with shank diameters of %4, %,
' ~ 1,1%,&1%-inches.
a —

Nall LOW-COST OFFSET HOLDERS

SPOT WELDING TIP 4.

NEW Simplified

Paddle-Type Holder
Simplified design makes this new TUFFALOY
paddle-type holder both lower in cost to use (no
tee connector is needed) and more rugged and
long-lived. An outstanding holder for spot weld-
ing in restricted areas. Standard 4-inch offset,
with shank diameters of %4, %, 1 & 1%4-inches.
It uses TUFFALOY Socket-Type Tips, available
in four nose types. They can be inserted on either
side of the paddle.

New this year are these Tuffaloy straight holders:

The Goldcrown ejector type (top), and the Gold-
spot non-ejector type (bottom). The ejector holder
features an entirely different method of ejecting
tips. The sturdy brass head is moved forward, forc-
ing a stainless steel ejector tube against the base of
the tip. This mechanism does not cause leakage,
usually associated with ejector holders.
Threaded adapters eliminate the need for re-
working worn tip sockets, permit changing tip size e
without replacing the holder. In 8- and 12-inch
lengths and in diameters of %, 1, 1%, & 1%- TUFFALOY
R
inches. For Nos. 4, 5, 6, & 7 RW taper tips.

Request the new TUFFALOY Catalog. It describes for the first time the RW Taper numbering system for spot welding electrodes.

On the west coast—

Air Reduction Pacific Company
AIR REDUCTION SALES COMPANY L_ titernationaity-
Airco Company International
A division of Air Reduction Company, incorporated In Cuba—
Cuban Air Products Corporation
150 East 42nd Street, New York 17, N. Y. in Canada—
Air Reduction Canada Limited
More than 700 Authorized Airco Distributors Coast to Coast All divisions or subsidiaries
of Air Reduction Company, Inc.
For details, circle No. 27 on Reader information Card
WELDING JOURNAL | 1089
 2,929,910--PowDER ENTRAINED GAs- the flux path. Features of the transformer in- wormwheel to rotate therewith Other mem-
SHIELDED METAL-ARC WELDING clude the fact that a rod is reciprocally mounted bers complete the apparatus and rotate the worm-
on the loop with at least a portion thereof extend- wheel to provide circular movement of the guide
Harry E. Kennedy, Berkeley, Calif., ing into the water passageway, and an electrode nozzle and welding wire to obtain the circular
assignor to Union Carbide Corp., a tip is mounted on the rod and is electrically seam
corporation of New York. connected to one of the terminals of the secondary
By the present process, a metered amount of the loop. 2,930,884-—-ARC-WELDING ELECTRODE
desired powder is aspirated into a stream of shield- Ho.pEerR—-Julian A. Monax, Pasadena,
ing gas so as to entrain the powder therein, which 2,929,916—-ELEcCTRODE WHEEL AND
powder and gas stream flow along and outside of HoLpER ASSEMBLY FOR RESISTANCE Calif.
the electrode wire. The powder is deflected in- SPOT-WELDING MACHINE—Julius V. In this new electrode holder, a casing of insulat-
ward toward the wire to adhere at least in part to Bergfeldt, New Britain, Conn. ing material is prov.ded and it has a fixed jaw
it and to be carried along thereby while the carrier of conducting material therein. A movable jaw
gas passes on bo shield the weld arc and weld Bergfeldt’s patent relates to an electrode wheel is provided in the electrode holder and a lever
puddie and holder assembly. In the assembly, a con- member is present to control the position of the
ductive electrode wheel is rotatably supported on movable jaw and force it to electrode engaging
2,929,911-NozzLE FoR GAS-SHIELDED a body of electrically conductive material, and position against action of a compression spring
Arc WELDING AND METHOD oF USING the patented assembly provides a special means in the electrode holder. The spring moves the
for supply ofa fluid coolant toa plurality of spaced movable jaw to electrode releasing position when
Ir—Lester T. Bowers, Oreland, Pa., apart electrodes detachably mounted on the the lever is released.
assignor to The Budd Co., Phila- wheel
delphia, Pa., a corporation of Penn- 2,931,885-H1IGH-FREQUENCY ELEc-
sylvania. 2,929,917—-ELrEctric INSTALLATION IN
CIRCUMFERENTIAL SEAM-WELDING AP- TRIC WELDING WITHOUT ROUGH AND
This patent is on a welding head for gas shielded EXcESsSIvVE INSIDE FLasH—George W.
arc welding and the welding head includes a nozzle PARATUS—Vincenc Kruml, Ales Bra-
positioned concentrically to surround the benec and Vladimir Sulc, all of Chote- Underwood and Harry La _ Tour,
electrode. ‘The outer edge of the nozzle is slanted bor, Czechoslovakia, assignors to CKD Middletown, Ohio, assignors to Armco
relative to the nozzle axis and the electrode pro- Ceska Lipa, Ndrodni Podnik, Cesk4é Steel Corp., Middletown, Ohio, a
Lipa, Czechoslovakia. corporation of Ohio.
This patent is on an electric arc-welding ap- The process of the present invention is for the
paratus for welding cylindrical elements such as production of a welded seam free from flash on
stay bolts or tubes to a workpiece such as boiler one side. In this high frequency welding action,
walls or other vessels. ‘The apparatus includes an perpendicular edges to be welded together are
electric motor comprising a feeding apparatus for brought into contact on that side which is to be
the welding wire and a displacing mechanism for free from flash and so as to leave a V-shaped notch
wire positioning nozzles. The apparatus in- on the other side. The edges are heated to weld-
cludes special controls for the motor circuit and ing temperature by a high frequency electric
welding circuits used. current and the edges are angularly displaced to
close the notch and force the edges together.
2,930,117—-METHOD oF SEAM-BRAZING
PLASTIC-COATED FOURDRINIER WIRE 2,931,886—APPARATUS FOR CLADDING
Abstracts of Current PATENTS William M. Wilson, Allwood, N. J., Harold B. Nunnelee, Waukesha, and
assignor to Eastwood-Nealley Corp., Willerd A. Schumbacker, West Allis,
Belleville, N. J., a corporation of New Wis., assignors to Allis-Chalmers
Jersey. Manufacturing Co., Milwaukee, Wis.
The patented process relates to seam brazing The present patent relates to apparatus for
woven wire fabric having a plastic coating thereon depositing a wire electrode and an additional feed
which melts when subjected to the brazing heat wire as an overlay upon a metallic surface. Both
In the process, a liquid is applied to the fabric of the wires to be deposited on the metal surface
throughout the brazing zone, which liquid is include shank and tip portions and guide means
substantially stable at the brazing heat tempera- are provided for the individual wires so that, as
ture and has a viscosity causing the liquid to flow relative movement is provided between the wire
into and remain in the mesh openings of the fabric and its guiding means and the metallic surface,
Then the seam is brazed by application of heat the electrode wire is above the feed wire and with
prepared by Vern L. Oldham from a gas torch while the liquid is applied to the the electrode shank being positioned at a vertical
fabric angle to lead the electrode tip in a direction of
Printed copies of patents relative movement of the control head and
may be obtained for 25¢ from the 2,930,882--METHOD AND APPARATUS metallic surface
Commissioner of Patents, Washington, D. C FOR FABRICATING STRUCTURAL PANEL
AND CoRE ‘THEREFOR—James_ R. 2,931,888 WELDING Joseph P
Campbell, Laguna Beach, Calif. Thome, Elyria, Ohio, assignor to
jects beyond a plane extending perpendicularly to Campbell's patent is on a method of fabricating Bendix-Westinghouse Automotive Air
the axis arid through the outermost of this slanted a structural panel and includes the steps of dis-
edge of the nozzle. Brake Company, Elyria, Ohio, a cor-
posing an electrically conductive honeycomb core poration of Delaware.
2,929,912—-GAS-SHIELDED Arc WELD- element in the space between two electrically Thome’s new method is for welding a hollow
ING-—Alexander Lesnewich, New Provi- conductive surface sheets positioned in spaced body member in pressure-tight relationship to an
relationship. A supporting electrode of a length apertured plate. In the method, a center plug
dence, N. J., assignor to Air Reduction at least equivalent to that of the core element is
Co., Inc., New York, N. Y., a corpora- positioned between the sheets in contiguity to is positioned to extend through the hollow body
tion of New York. the core element and a welding potential is im- member and the apertured plate to form an an-
pressed across the welding head means. Such nular space for the reception of molten metal
Lesnewich’s patent relates to a gas shielded velding head means are disposed in contact Welding current is supplied to the hollow body
are welding process wherein the electrode is member and apertured plate by means of upper
selected from the group comprising copper and with the external surfaces of the sheets and thus and lower electrodes engaged therewith to pro-
copper alloys. The are is surrounded with a across the electrode and the core element. The duce a continuous annular joint bet ween the body
shielding gas stream comprising essentially inert velding head means is continuously movable along
nonatomic gas and the arc is confined by a sur- the external surfaces of the base sheets in paths member and plate
rounding stream of gas comprising essentially parallel to the longitudinal axis of the electrode to
nitrogen provide the desired welding action 2,931,889-—-APPARATUS FOR ARC WELD-
2,930,883-— APPARATUS FOR AUTOMATIC iInc—John W. Lingafelter, Richland,
2,929,913--PoRTABLE WELDING TOooL Wash., assignor to the United States
Victor F. Miller, Plandome, N. Y., ARC-WELDING CIRCULAR SEAMS Miro- of America as represented by the
assignor to Bell Telephone Labora- slay Adamec, Termesivy, and Zdenek United States Atomic Energy Com-
tories, Inc., New York, N. Y., a cor- Duben and Miloslav Pavlasek, Chote- mission.
poration of New York. bor, Czechoslovakia, assignors to CKD The new welding machine includes an an-
This patent is on a specialized welding tool for Cesk4 Lipa Nérodn{ Podnik, Cesk4é nular electrode, a magnetic core secured at one
attaching a piece of contact metal to a spring Lipa, Czechoslovakia, a corporation of end to the electrode in electrical connection there-
a on relays, crossbar switches and the Czechoslovakia. with and extending from one side of the electrode
ike This patent relates to apparatus to provide arc- aligned therewith and other means are present
velded circular seams. In the frame of the appa- to create an arc between the electrode and the
2,929,915--WELDING - TRANSFORMER ratus, a wormwheel is rotatably supported upon periphery of the workpiece of the same size and
APpPARATUS——Roy M. Taylor, Belmont, the frame and is positioned coaxially to the circu- shape as the electrode. The workpiece is posi-
and Preston E. Girton, Grand Rapids, lar seam to be formed. A supply of welding wire tioned on the opposite side of the electrode from
Mich., assignors to Kirkhof Manufac- is stored upon the wormwheel to rotate therewith the core and a coil is also present to create elec-
turing Corp., Grand Rapids, Mich., a and a guide nozzle for the welding wire is carried tromagnetic flux passing through the core and the
corporation of Michigan. by the wormwheel to rotate therewith and guide electrode axially of the electrode and between the
the end of the welding wire toward the circular electrode and the workpiece radially of the elec-
The patented transformer has a primary coil velding seam. Feeding means are present to trode across the arc for rotating the arc about the
stacked adjacent a water cooled secondary loop continuously feed weld wire into and through the electrode. The coil is wound upon the core and
that has a water passageway and a core providing nozzle and such feed means are carried by the is electrically connected thereto

1090 | OCTOBER 1960

 For jobs where only the best will do—

NEW GLENN Verret:

LOPE

CONSTANT VOLTAGE

POWER SUPPLIES

for all consummable-electrode automatic and

semi-automatic welding processes, including:

MIG - SHORT ARC

SUBMERGED ARC - MAGNETIC FLUX

New GLENN V/S power supplies are totally new from basic trans-
former design to their exclusive new linear slope control. They can
be precisely “tuned” to the exact arc characteristics needed to
produce any type of metal transfer and deposition desired. The
result is improved weld quality, appearance, speed and economy

Baaic featwres

NEW TRANSFORMER DESIGN increases efficiency,

improves voltage regulation and minimizes effects
of line voltage fluctuations.

STEPLESS VERNIER SLOPE CONTROL provides con-

tinuous linear slope control from flat no-slope to
maximum slope; in two overlapping ranges.

EXTENDED OUTPUT VERNIER VOLTAGE RANGE pro-

vides effective de welding (loaded) voltages from
approximately 8 to 42 volts, depending on slope
settings.

GLENN PACIFIC
WHEN SPECIFICATIONS ARE CRITICAL and weld POWER SUPPLY CORPORATION
quality, uniformity and appeorance must be kept
high, you'll be hours and dollars ahead to specify 7O3-37th Avenue + Oakiand 1, California
GLENN V/S power supplies. Why not get the Originators of CV Power Supplies
facts? Write us today for literature, specifications; Eastern Office Midwestern Office
please address Dept. 138 221 Dukes Rd., Rahway, N. J. 640 So. York, Elmhurst, til.

* GLENN Balanced Wave Power Supply for TIG Welding > GLENN Manual )
and Stud Welder Power Supply > GLENN Arc Gouger Power Supply * GLENN ‘
Constant Potential Welder for “Gang” Manual Welding % GLENN Industrial
Power Supplies and Heavy Duty Variable Voltage Transformers ey

For details, circle No. 28 on Reader Information Card

WELDING JOURNAL | 1091
 New Literature Electric Corp., 5 New St., Stamford,
Conn.: (1) The Use of Portable
X-Ray Equipment in Aircraft Main-
tenance Operations; (2) The Use
of Portable X-Ray Equipment in
Storage Tank Fabrication; (3) A
Comprehensive Listing of All Popular
Nondestructive Testing Methods and
the Applications Best Suited to
Each.
For your free copy, circle No. 55
AWS Welding Show Attendance Analysis Available on Reader Information Card.

The AMERICAN WELDING SOCIETY production, purchasing, instructors

and technicians and government. Precision-welding Catalog
announces the availability of the
Attendance Analysis of the 1960 This breakdown is particularly in- A 24-page catalog offered by
Welding Show, held in Los Angeles, teresting to manufacturers who Weldmatic Division Unitek Corp.,
Calif. The analysis is the result of wish to evaluate potential sales by 950 Royal Oaks Drive, Monrovia,
an independent survey. It gives a the quality of attendees. Prac- Calif., shows complete line of stored-
regional breakdown and an occupa- tically 90% of those attending the energy, capacitor-discharge power
tional breakdown of the registered Los Angeles show had the authority supplies, precision welding heads
attendance. The figures are very to recommend, specify or place and handpieces, and welding ac-
revealing. They indicate’ the orders. This unusually high cessories. A_ special section pre-
percentage of attendance drawn percentage accounts for the fact sents advantages of resistance weld-
from the areas surrounding Los that 70% of exhibitors have ex- ing, compares stored-energy weld-
Angeles and the percentage from hibited at the AWS Welding Show ing to a-c welding and outlines
more remote areas. This is valuable important factors in selecting stored-
for the last three years, and 58%
information for all prospective ex- energy equipment. Photos of
hibitors, especially those who are have never missed a show. Copies
of the Attendance Analysis may be typical precision-welding applica-
primarily interested in specific areas. tions are included.
The occupational breakdown lists obtained without charge from The For your free copy, circle No. 56
percentages from executive manage- AMERICAN WELDING SOCIETY, 33 on Reader Information Card.
ment, engineers, general welding, W. 39th St., New York 18, N. Y.

Welding News
“The Stabilizer’ consists of 20
Welding Training Aids including new application and design pages of news on all aspects of
data, has been published by Eutectic welding which are of interest to
Fifty helpful welding training Welding Alloys Corp., Flushing, men who weld and who like to read
aids are listed and described in a N. Y. The 148-page illustrated and talk about it. The pages con-
new folder issued by Hobart Weld- publication is the seventh edition tain welding ideas ranging from
ing School, Troy, Ohio. The list- of the Data Book. In addition to simple gadgets to success stories
ing includes books on a wide variety providing a guide, indexed by about the men who went out on
of welding subjects, welding charts, application, to nearly 200 different their own and “rang the bell.”
welding educational films, welding Eutectic welding rods, electrodes Published by Lincoln Electric Co.,
training aids, welding codes and and chemical aids, the book illus- 22787 St. Clair Ave., Cleveland
welding specifications. trates and explains general weld- 17, Ohio.
For your free copy, circle No. 51 ing techniques and joint design
on Reader Information Card. hints.
For your free copy, circle No. 53 Electronic Welding
Silicon-rectifier Machines on Reader Information Card. Weldmatic Division Unitek Corp.,
A data sheet describing Murex 950 Royal Oaks Drive, Monrovia,
silicon rectifier d-c welding machines Calif., has published the first issue
Bronze Alloys
covering current ranges up to 625 of its 8-page magazine, “Electronic
amp is available from Metal & “Ampco Welding News,” No. Welding.”” Issued quarterly, it is
Thermit Corp., Rahway, N. J. 83, a 4-page news leaflet, carries intended to provide the electrical,
These machines, models M20R, articles on maintenance and re- electronic and metalworking in-
M30R and M4OR, are described pair welding of steel mill equipment, dustries with authoritative, up-to-
as power sources for d-c welding, press, die and wearing parts of in- date information on precision metal
maintaining output efficiency of dustrial machinery—all by the use joining applications, such as as-
98% over years of operation. Non- of the company’s bronze alloy sembly of electronic packaging,
aging silicon diodes are hermetically electrodes and rods. Published by fine wire leads, thin and hard foils
sealed. The units conform to Ampco Metal, Inc., Dept. AWN, and screens, and miniature circuit
NEMA standards. Milwaukee, Wis. modules. Notes on welding tech-
For your free copy, circle No. 52 For your free copy, circle No. 54 nology written by Weldmatic en-
on Reader Information Card. on Reader Information Card. gineers, new materials and new
developments in welding equipment
Welding Data Book Nondestructive Testing will be other regular features.
A new edition of its pocket-size Three new technical bulletins For your free copy, circle No. 58
Maintenance Welding Data Book, are available from the Balteau on Reader Information Card.

10992 | OCTOBER 1960

Magnifying Glasses Brazing rods
Bausch & Lomb, Inc., Rochester
2, N. Y., has revised the descriptive to meet AWS-ASTM specifications
catalog (1-103) which covers their
line of readers and magnifiers.
New items have been added to the
14-page illustrated booklet, which
lists over 65 individual models
(powers from 2x to 20x). Speci- AMPCO-BRaZ

fications and prices are included

for an extensive selection of round
and rectangular readers, folding Produced from start to finish Packaged for easy
pocket magnifiers, watchmaker’s by one reliable company handling and stocking
loupes, surface comparators, en-
larging focusing magnifiers, etc.
For your free copy, circle No. 59
on Reader Information Card. AMPCO-BRaZ No. I AMPCO-BRaZ No. 4
(AWS specifications, RB Cu Zn-D) (AWS specifications, R Cu Zn-B)
Low-fuming, high-strength, high Low-fuming, manganese bronze
nickel rod for joining and overlay rod with nickel added for higher
with the oxy-acetylene process strength and ductility. For brazing
oaasaee Bonds readily to copper.-, iron-, and braze-welding high-strength
and nickel-base metals and alloys. bronzes and brasses, steel, cast
iron, and malleable iron with the
oxy-acetylene process.

REMIEWS
OF NEW BOOKS

AMPCO-BRaZ No. 2 AMPCO-

Case Histories on Statistical Meth- (AWS specifications, R Cu Zn-C)
BRaZ FLUX No. 2
ods for Quality Control (Regression, Low-fuming, high-strength, man
Correlation and Association) Series ganese bronze rod for brazing and Saves time on general
_braze-welding steel, cast iron welding, cast-iron
II. Prepared by Metals Tech- malleable iron, and copper and welding, and brazing
nical Committee of American its alloys with the oxy-acetylene Has many other
Society for Quality Control. 94 process. « advantages.
pp., tables, 8'/, x 11 paperbound.
Available from ASQC, 161 W.
Wisconsin Ave., Milwaukee 3, Wis.,
at $1.00 each.
A single set of data is taken from
steel process metallurgy and a
number of different statistical meth-
ods are applied to the solution of
a single problem. No new methods
are introduced but the presenta-
tion itself is said to be unique.
By keeping the data constant, the
opportunity is offered to study, by
comparison, a group of known statis-
tical methods. This idea is well
AMPCO METAL INC.
supported by the use of an ordered
outline of procedure identically ap- MILWAUKEE 46, WISCONSIN
plied to each case. Calculations West Coast: Huntington Park, Calif
Southwest: Garland (Dallas County), Tex.
are given in detail and are discussed.
The book is intended for the non- Se SS SS SS SSS SSSSSSSSS8 S828 8828888888895
professional who must deal with AMPCO METAL, INC.
problems of variation and who needs Bulletin W-17 Dept. WJ-10, Milwaukee 46, Wis.
to understand more about the gives detailed facts 0 Send me Bulletin W-17.
selection of a statistical method. 0) | am interested in becoming an
on the complete AMPCO-BRazZ distributor.
The publication was sponsored by
the American Iron and Steel In- Ampco Welding Name
stitute and is designed to make line. Send for your Company
people aware of the possibilities copy. Tear coupon
offered by the use of statistical Address
and mail today.
methods in the field of metallurgy. City Zone State
The methods are applicable to any Trtrrrrrreffttttttttttttttttttttttt
Peseeeee2e22882828288295
suitable field. w-1is3
For details, circle No. 29 on Reader information Card
WELDING JOURNAL | 1093
 New Products

Flux-covered Brazing Rods It is recommended by the manu-

facturer for welding with fine
Two types of bronze brazing diameter wires using the Airco-
rods with extruded flux coverings tions in wave form produced by the
matic welding processes including
are being produced by All-State use of phase control. The accuracy
the CO, dip-transfer process.
Welding Alloys Co., Inc., White is givenas +3% of fullscale. Seven
The 500 amp d-c rectifier machine
Plains, N. Y.—a thin-covered rod scale ranges can be set between
of horizontal design can be used
for high-speed production lines and 2500 and 250,000 amp.
for welding with all diameter wires
an all-round heavier-covered main- The 6 x 10 x 4.5-in. unit weighs
where conventional constant arc
tenance rod for general use. 10.5 lb with batteries and toroid
voltage or down slope is required.
The flux-covered bronze rods and can be completely enclosed for
The machine can be used with the
offer the user just the right amount carrying. The toroid and cable
submerged-arc welding process. It
of flux—and no more—for all braz- are stored in a rear compartment.
is also stated that fourteen down-
ing operations. The brazing rods For details, circle No. 104 on
slope adjustments are obtainable
Reader Information Card.
by reconnection of leads accessible
through a panel on the side of the Projection-welding Machine
welder.
For details, circle No. 102 on A machine to encapsulate semi-
Reader Information Card. conductor and transistor compo-
nents by resistance welding over a
Electronic Solder Flux wide range of sizes (diameters
ranging from 0.3 to 1.25 in.) is
A mildly activated rosin-type
announced by the Thomson Elec-
flux called the Fusion RU series has
tric Welder Co., 161 Pleasant St.,
been developed by Fusion Engineer-
Lynn, Mass., as_ their model
ing, 17921 Roseland Ave., Cleveland
2400 projection welding machine.
12, Ohio. This flux, though mildly
activated, has been tested neutral
are covered with a special non- both before and after soldering mak-
fuming, noncharring flux that will ing it acceptable for electrical and
not crack off the rod even when the electronic work where nonconduc-
rod is bent over 360 deg. A further tive, noncorrosive flux residues are
extra is negligible clean-up if proper required.
brazing techniques are followed. The Fusion RU series is said to
The nonhygroscopic flux cover- make possible much more rapid
ing allows unlimited storage of soldering than had previously been
brazing rods. Packaged in tough possible with neutral fluxes—as
10-lb tubes. The rods are available much as 100% in some cases.
in the following sizes: * 3, '/s. Humidity chamber tests have
‘/1, and '/, in. diam by 36 in. long. definitely determined that the flux
For details, circle No. 101 on is noncorrosive after soldering and
Reader Information Card. thus suitable for fine electronic
component soldering, according to
Metal-arc Power Sources the manufacturer. This capacity range is realized by
Air Reduction Sales Co., 150 For details, circle No. 103 on using twin 75 kva_ transformers
E. 42nd St., New York 17, N. Y.., Reader Information Card. coupled to the welding head through
has recently marketed two welding balanced secondary circuits and a
Resistance-welding Current Meter new welding head design.
power sources for use with manual
and automatic gas-shielded metal- A toroidal current meter for The welding head features two
arc (Aircomatic) welding equipment. measuring peak or rms secondary column construction to eliminate
These units are described as capable current in resistance welding ma- deflection plus a low inertia force
of providing constant arc voltage chines of 60-cycle single-phase in- system utilizing the latest develop-
output, or varying degrees of down put has been introduced by Duffers ments in antifriction bearings, light-
slope. The machines are rated at Associates, Inc., Box 296, Troy, weight diaphragms and precision
300 amp, 100% duty cycle and 500 N. Y. air regulators. The system has a
amp, 100% duty cycle. An outstanding feature claimed special sequence to provide con-
The 300 amp d-c rectifier machine for the meter is that its accuracy trolled approach that eliminates
is of vertical design and provides is unaffected by the heat-control electrode hammer on the work.
continuous infinite down-slope ad- setting. The instrument is designed For details, circle No. 105 on
justment through a hand wheel. to respond to high-frequency varia- Reader Information Card.

1094 | OCTOBER 1960

 . get

>
CONTROL
8
STRENGTH,
WELD
LB.

Test results compare consistency of welds made with conventional

control and Monautronic V-2 control over an eight-hour production
period. Notice the striking uniformity of Monautronic-controlled welds.*

Ar avy

MONAUTRONIC
CONTROL

WELD
STRENGTH,
LB.

NOON
12

new feedback control gives you consistently

high quality welds... automatically

The new Monautronic V-2 welding control makes use

/ ‘ ; ,
é SJ y of the latest advances in electronic computing to over-
come automatically such obstacles to weld quality as
line voltage fluctuation, electrode wear, variations in electrode
tip force, surface finish and shunting.
The control compensates for undesirable variations usually en-
countered in resistance welding by maintaining voltage across a
weld at a constant value. This constraint of voltage amounts to
constraint of final weld temperatures, and such temperature
control assures uniform production of high quality welds.
Any metal that can be resistance welded can be welded better with
the Monautronic V-2 than with any other control on the market.
For complete details, contact THE BUDD COMPANY, Elec-
tronic Controls Section, Philadelphia 32, Pa., or one of our
Monautronic V-2 welding control has fully regional offices. *Case study upon request.
automatic sequencing with all provisions
for single spot, roll spot and seam welding.

cecronc MD At
2450 Hunting Park Ave. 12141 Charlevoix Ave. 3050 East 11th St.
Philadelphia 32, Pa. * Detroit 14, Mich. * Los Angeles 23, Cal, ADEs cons
For details, circle No. 30 on Reader Information Card
WELDING JOURNAL | 1095
 USE
READER
INFORMATION CARD

Silver-brazing Rods
Recent improvements in the sur-
face finish of Silvaloy 15 and
Silvaloy 5 silver brazing rods have
resulted in further enhancing the
self-fluxing properties of these alloys,
according to the American Platinum
& Silver Division of Engelhard
Industries, Inc., 75 Austin St.,
Newark 2, N. J.
While paste fluxes may be used
with these alloys, their self-fluxing
=A)= a od= 1) Ol83 oO) 1 OORT 0) ad od On A ability is said to make them particu-
larly useful in silver brazing of joints
where complete flux removal is
both difficult and important, as in
electrical and refrigeration equip-
ment. In such work, the brighter,
cleaner, more consistent finishes
now obtained are doubly important
Job report courtesy of in helping to eliminate foreign
Conch Methane Co., Ltd. matter and contaminants from the
joint. The new rods are easily
when pipe joints must be com- distinguished by their bright silvery
appearance.
pletely dependable at minus 258°F For details, circle No. 112 on
Reader Information Card.
The engineers who successfully “broke through”’ the problem of handling liquefied
methane gas on and off the “METHANE PIONEER” at minus 258°F had to Misfire Detector
“play it safe’’ in every respect. To provide unquestionably safe stainless pipe joints, for Resistance Welding
Arcos EB Consumable Weld Inserts were used to make the important root passes.
A misfire detector for resistance-
EB Inserts permit welding to be done from one side only. They eliminate back-up welding control equipment has been
ring obstruction and produce a smooth inside contour to expedite gas flow. If you developed by General Electric,
are not familiar with the advantages of EB Weld Insert for “‘tough” pipe joining Schenectady 5, N. Y.
problems, write ARCOS. *Trademark of General Dynamics Corp. Any misfire of the control’s igni-
tron contactors, which operate on
alternate half-cycles to switch cur-
rent into the welding transformer
primary, trips a relay and gives an
indication to the operator.
WELD d
WI THs DC
fi
G}, OS Should one tube fail to fire, the
welder skips and the second tube
fires twice in succession. This may
EB* WELD INSERT saturate the transformer and raise
exciting current to a high value that
could damage the other ignitron.
The detector is said to give an in-
stant indication of improper welds,
useful in automatic seam-welding
of materials with a narrow plastic
temperature range. It helps pre-
vent overloading an ignitron and
indicates when tubes need replace-
ment, according to the report.
The detector may be used to
D * FLUKES operate a signal light or horn,
Thi)
counter, marking device, reject
mari A D AULQMA iC EQUIPMENTFOR WELDING& mechanism, shutdown circuit or
3 SS 9 LOW ALLO ALUMINU a: LDSTEEL any combination of these.
For details, circle No. 113 on
ARCOS CORPORATION, 1500 South 50th Street, Philadelphia 43, Pa. Reader Information Card.
For details, circle No. 31 on Reader Information Card
1096 | OCTOBER 1960
Temperature Indicator Aerosol
Packaged

The Tempil® Corp., 132 W.

22nd St., New York 11, N. Y.,
has announced the packaging of
temperature-indicating coatings in
aerosol containers. Forty tempera-
ture ratings from 100 to 650° F
inclusive are now available in the
spray-type containers. Develop-
ment work is under way to similarly ig _
UE
package forty ratings from 700
to 2500° F. loettttl “tim.
The spray-can package is in-
tended to supplement the standard
glass-bottled line and is offered as
a convenient way of coating larger
surfaces.
For details, circle No. 114 on ul
Reader Information Card.

Zirconia Coating Available

Zirconia (ZrO.) coating is now

available for application by Linde’s
Plasmarc plating process, according
to an announcement by Linde Co.,
Division of Union Carbide Corp.,
270 Park Ave., New York 17, Job Report Courtesy of
N. Y. The coating will be used Southwest Welding & Mfg. Division,
Yubo Consolida.ed Industries, Inc., Alhambra, Calif.
primarily for insulating purposes.
Linde’s Plasmarc process is a
method for applying coatings and When strong welds are needed
for fabricating shapes, using ma- to resist impact at minus 320°F
terials having ultra-high melting
points. In addition to the new
zirconia coating, tungsten and other
refractory metals have been ap-
WELD WITH SFecos a,
plied to a number of different base
materials including graphite, steel,
aluminum, brass and molybdenum.
For details, circle No. 115 on
Reader Information Card. STAINLESS ELECTRODES
This six-section spherical liquid oxygen container is made of
Type 32] stainless steel. Walls 414” thick were required to with-
lron-powder Electrode stand 3000 psi pressure. To assure crack-free welds to meet
ASME impact properties at —320°F, it was welded with Arcos
The addition of iron powder to Chromend 19/9Cb-LC electrodes with controlled ferrite. The
the standard E-6012 electrode is said first time —and every time you use Arcos electrodes —you get
to add several advantages to this top performance, save time, and money. ARCOS CORPORATION,
type of West-ing-arc rod. Espe- 1500 South SOth St., Philadelphia 43, Pa.
cially designed for all types of struc-
tural and plate welding, it is ideal
for wrought iron, low and medium-
carbon steels, low-alloy steels, cop-
per-bearing steel and low or
medium-carbon cast steel, ac-
cording to the West-ing-arc Dept.,
P.O. Box 2025, Buffalo, N. Y.
Known as Zip-12, the electrode
is said to provide better arc action,
finer metal drop transfer and mark-
edly superior slag characteristics.
Faster metal wash and slag re-
moval is obtained. Thus faster
welding speeds on lap and fillet
welds are possible.
For details, circle No. 116 on
Reader Information Card.
For details, circle No. 32 on Reader Information Card
WELDING JOURNAL | 1097
Fluxed Silver-brazing Rod plication temperature and high fluid-
ity, strength and high electrical
A flux-coated silver-brazing-type conductivity. Deposits are cad-
alloy, which is said to cut brazing mium-free and thus the product
time by two-thirds through elimina- may be used safely on food vessels
tion of separate flux, has been an- and processing equipment, accord-
nounced by Eutectic Welding Alloys ing to the manufacturer.
Corp., Flushing, N. Y. EutecSil For details, circle No. 117 on
1020FC alloy offers the advantages Reader Information Card.
of conventional silver-brazing-type
£6013 Electrode
Hobart Brothers Co., Troy, Ohio,
has developed Hobart 447A, a tings are now available in 45 deg
E6013 electrode with a small amount angles. Previously their produc-
of iron powder added to the coating tion was mostly in 90 deg saddle
to improve arc stability on out of fittings.
position welding. It can be used on Manufactured of high-strength
ac or dc, straight or reverse polarity. seamless tubing in three weight
Available in */;. in. diam in 12 in. schedules, Allied Fittings come in
lengths; '/s, °/3, */16 in 14 in. A-53 for most piping requirements
lengths and 7/3, '/, in 18 in. and A-106 Grade B for higher pres-
lengths. Standard package is 50 sure applications. Grade A-106
rod while eliminating need for lb, '/s in. and larger, 25 |b for */3 can be delivered with mill test
flux mixing and application and in. certificate if requested. Sizes range
reducing need for precleaning and For details, circle No. 118 on from 1 through 12 in. for welding
preheating. Reader Information Card. and 1 through 4 in. for threaded
EutecSil 1020FC alloy is available filling. All are beveled for welding.
in rods of '/\, and */;. in. diam, and Saddle Fittings Special fittings 1 through 36 in.
is suitable for ferrous and nonferrous aluminum, alloy or stainless steel
metals and all types of joints in- Allied Piping Products Co., Inc., are available in any desired angle.
cluding tee, butt, fillet and lap or P.O. Box 311, Salem, Ohio, an- For details, circle No. 119 on
sleeve. Features include low ap- nounces that preshaped saddle fit- Reader Information Card.

WELDING JOURNAL
General Advertising Rates
WANT TO MEASURE Effective January 1, 1960
3 times
$410.00
295.00
RESISTANCE WELDING 230.00
165.00
.00
.00
SECONDARY CURRENT? 90.00 80.00
*24 times rate, for full page only: $330
*36 times rate, for full page only: $300
PREFERRED POSITIONS:
490.00
Inside front cover
Inside back cover
Outside back cover
COLORS:
Inside pages, standard AAAA color, extra
Inside pages, special color, extra......
Color on cover, extra 888
8

METALLURGICAL ENGINEER
The Chemstrand Corporation’s nylon manufacturing plant at Pensa-
cola, Florida is seeking a graduate engineer to assist the chief metal-
Battery powered, transistorized portable Model 260 meter lurgist in shop problems related to materials selection, welding, and
measures true RMS value of secondary current for single heat treatment. Experience with chemical processing equipment
phase welders used with or without heat control. Seven
desired.
ranges to 250,000 Amps, +3°% accuracy, down to one
cycle weld time, uses special air core toroid requiring — Excellent working conditions and benefit programs. Attractive
no direct connection to welder. community life and living conditions in Gulf Coast location.
Send resume of academic training and experience to:
DUFFERS ASSOCIATES, INC.
Manager, Box P-3
P. O. Box 296, Troy, N. Y. Employment- THE CHEMSTRAND CORPORATION
Recruitment Decatur, Alabama
For details, circle No. 33 on Reader information Card

loss | OCTOBER 1960

Fiberglass Safety Hats and caps is now being manufactured
by The Fibre-Metal Products Co.,
High-pressure-molded fiberglass 5th and Tilghman Sts., Chester,
hats and caps that combine high
strength with lightweight comfort
are offered by Welsh Manufacturing
Co., 79 Magnolia St., Providence,

Low-voltage-insulated links are

manufactured by E. D. Bullard
Co., 2680 Bridgeway, Sausalito,
Calif., in sizes '/, to 5 ton, 1 kv.
Steel inserts separated with lam-
Bm. 1. They feature nylon suspen- inated insulating core are wound
sion with snap-in design. Headband with glass fiber roving and bonded
adjusts from 6'/, to 7°/s sizes. together with epoxy resin. When
Winter liners are available. Shell completed, Bullard insulated safety
is made in wide variety of colors. links are said to be easily con-
For details, circle No. 120 on Pa. The full line includes, in
nected by means of a bail or safety addition to models of standard
Reader Information Card. hoist hook. Each link has a 4 to 1 construction, neoprene-coated
safety factor and a 2 to 1 dielectric liners without metal parts for use
Portable Blast Cleaner safety factor. with electrical hats and caps, knitted
Bren Weld Corp., 5114 Third For details, circle No. 122 on liners for universal use and dispos-
Ave., Brooklyn, N. Y., announces Reader Information Card. able liners for perspiration absorp-
the addition of their Model BC-1 tion and for use as a “‘visitor’s”’
Portable Blast Cleaner. Helmet Liners sanitary liner.
This unit features a blasting gun A wide variety of winter liners For details, circle No. 125 on
providing trigger-operated control. designed for use in all safety hats Reader Information Card.

no matter what the two metals are. Be

yon
LD Elon

Joining copper pipes in extremely

hard water area. Lead solders made
porous joints. Heat required for
high temperature solders burned
copper pipes.
SOLUTION:
Weighing 25 lb, 20 in. high and All-State’s +430, a silver-bearing
with a capacity of 50 lb of sand, the solder which flows at 430F, made
BC-1 comes complete with hood, perfect, permanent joints, solved
blasting gun, adjustable feed elbow, the problem.
screen and 10 ft blast hose and
connection. ALL-STATE has a solder for joining any commercial metal or alloy to any
other... in one or more temperature ranges. For typical examples, see
For details, circle No. 121 on tables below:
Reader Information Card. 400°F-500°F 700°F-800°F
Al. Copper Brass S.S. Nickel Al. Copper Brass $.S. Nickel
Link Insulates Hoist 107 Aluminum 105
430 Copper
A low-voltage insulated link was 430
430
designed to insulate a_ hoisting 430 Stain. Steel
430 Nickel
mechanism from the weldment. Reference numbers above indicate All-State solder to be used for joining metals.
When a weldment is not properly
grounded, the current flows back A set of four complete tables, covering temperature ranges from 400F to
through the chain into the hoist. 800F, is yours for the asking. Send for free Instruction Manual, too.
Frequently electric hoists are burned Distributor-Stocked, convenient to buy. Economical to use.
out, and the temper of expensive y-O7y ALL-STATE WELDING ALLOYS CO., INC., White Plains, N. Y.
alloy chains is lost because of this Call WHite Plains 8-4646 or write for nearest distributor
current flow from weldment to
hoist.
For details, circle No. 17 on Reader Information Card
WELDING JOURNAL | 1099
Magnetic-particle Film Welding Machine for AC —60 to +700° F and are said to
have accuracies of better than 2%.
Recordaflux, a magnetic particle The Westinghouse Electric Corp., They are held to steel and other
inspection system, has been put on has introduced an a-c welder known ferrous materials by permanent
the market by Instruments Division as the TC. The basic unit is a Al-Ni-Co magnets.
of The Budd Co., Box 245, Phoenix- single-phase transformer-type a-c For details, circle No. 128 on
ville, Pa. welding power source. It is built Reader Information Card.
Recordaflux, formerly known as to operate from 230/460 v primary
Permaweld, is used to detect defec- and is NEMA rated as a 500 amp Heavy-duty Hose Reel
tive welds, cracks, inclusions and
other flaws in such materials as A heavy-duty automatic reel for
are currently inspected by magnetic industrial hose that avoids leaving
particle systems. The system has hose in a tangle on the shop floor,
and holds it ready for instant use
when needed, is announced by
United Specialties, Inc., P. O. Box
698, El Dorado, Ark. The Model
A-4 has capacity for 50 ft of */,

been used by The Budd Co. to machine, according to West-ing-arc

determine nondestructively the di- Dept., P.O. Box 2025, Buffalo,
ameter of spot welds in stainless os
steel, such as cold rolled Types The West-ing-arc type TC is
301 and 201 and in precipitation said to be designed particularly for
hardening grades such as PH15- use in shipyards, but has broad ap-
7MO and AM355. Since there is a plication for a-c welding in the
direct correlation between the di- transportation industry, construc- he ~ ~ -—**
ameter of a resistance spot weld and tion trade and for general heavy in. ID, 2-braid, single hose. Con-
its tensile shear strength, Recorda- fabrication. nections furnished are: swivel, *
flux is said to supply very reliable For details, circle No. 127 on in. npt, female thread; shaft, 1-
weld-strength information. Re- Reader Information Card. in. npt, female thread. Rec-
search is proceeding for use of the ommended working pressure with
spot-weld technique in low-carbon Dial Thermometers standard 2-braid hose, 250 psi.
alloy steels. For details, circle No. 129 on
A line of magnetically attaching
For details, circle No. 126 on Reader Information Card.
bi-metal dial thermometers has been
Reader Information Card.
released by Industrial Service Co.,
2 Richmond St., Providence 3, Arc-spot Power Supply
R. L., for temperature measure- Bren Weld Corp., 5114 Third
Now you can check ment on ferrous materials. Ave., Brooklyn 20, N. Y., an-
Three styles of thermometers are nounces the addition of the new
WELD TIME with available: (1) thermometers for Model #150 power supply for their
temperature measurement only, (2) Arc Spot weld gun.
new PORTA-COUNT
thermometers with maximum reset
cycle counter hand and (3) thermometers with
electric contactor to permit “on-
off’ temperature control with elec-
tric resistance or induction heating.
The thermometers are available
in various temperature ranges from
only $282
Counts actual weld time during firing.
PORTA-COUNT is light and small enough
to carry easily—keep right on the job.
Completely portable and self-contained
with long life (up to 2 years) batteries.
Counts single phase.
The unit is provided with 5 heat-
No Plug In No Clips No Clamps stage receptacles and male plugs to
PORTA-COUNT uses an inductive pick up insure fast, positive connection.
to count, sensing magnetic field around Used with the Model H Arc Spot
power line, welding buss, transformer or
electrodes. gun, the Model #150 power supply
is said to weld 14 gage or lighter to
INSTRUMENT CONTROL CO.
1554 Nicollet Avene any thickness (minimum 2 pieces
Mi. ad i 3, NNMi 26 gage). Weight is 41 lb.
For details, circle No. 130 on
For details, circle No. 34 on Reader information Card Reader Information Card.

1100 | OCTOBER 1960 For details, circle Ne. 35 on Reader information Card —>
_ “as
UNION
CARBIDE

See the next page for details.

mpeeeeeeeeeeeeeeeeeeeee eee

rl 4h

UNION
CARBIDE
 October, 1960
Cerd valid until December 31, 1960
READER INFORMATION CARD Please send me, without cost or obligation, further
information and literature on items circled below:
76 101
77 192
78 103
79 104
80
81 Ue
82 107
83 108
84 109
WY
85
86
8&7
&8
89
90
91
92
93
94
95
96
97 122
96 123
99 124
100 125

a list of AWS publications

Mail me { information on AWS Membership [J
information on Journal Subscription [J

NAME

Anoeees i
FREE

pak S.. .. OOR

Information

and
| — j

Literature... Y
N.

PERMIT
9286NEW
NO.
FIRST
CLAS YORK,

Now, with just one postcard, you can gather

all the literature offered in the Journal. |
Just follow these simple steps for quick action:

1. Note reference number on text pages

and advertisements.

2. Circle the Corresponding number on one

of the facing reply cards. (Keep other
card for future use if desired.)

3. Please print name and address legibly.

Welding
Journal
33 18,
York
New
Y.
N.
Street
39th
West
States
United
mailed
[f
the
in
No
postage
necessary
Card
Business
Reply 4¢
POSTAGE
BY
PAID
BE
WILL
 :
:
.
:
:
'
:
'
:
:
’
:
'
.
.
:
.
’
PERMIT
NO.
9286NEW
FIRST
CLASS YORK,
N.
Y. ’
: Air Reduction Sales Co.
:
. All-State Welding Alloys Co., Inc.
:
:
: Alloy Rods Company
.
:
. The American Brass Co.
’
. American Pulimax Co. Inc.
.
’
: Ampco Metal, inc.
’
.
: Arcos Corporation
:
. Auto-View Welding Helmet Co.
:
:
’ Bateman Foundry & Machine
:
:
’ S. Blickman, Inc.
’
: The Budd Company
:
'
: The Chemstrand Corp.
.
!
: Coyne Cylinder Company
:
. Duffers Associates, Inc.
:
:
Welding
Journal33 New
West York
N.
18,
Y.
Street
39th ’ Electric Arc, Inc.
:
States 4¢
United
the
in
mailed
if
necessary
postage
No :
Business
Card
Reply POSTAGE
WILL
PAID
BE
BY . ee industries, Inc.
: n Platinum & Silver Div.
.
.
: Glenn Pacific Power Supply Corp.
’
'
: Handy & Harman
. Hobart Brothers Company
Octoher, 1960
Card valid until December 31, 1960 Instrument Control Co.
Piease send me, without cost or obligation, further International Nickel Co., Inc.
information and literature on items circled below:
Lenco, Inc.
101
The Lincoln Electric Company
Linde Company, Division of
Union Carbide Corporation
Lukens Steel Company
Miller Electric Mfg. Co., Inc.
Mir-O-Col Alloy Co., Inc.
NRC Equipment Corporation
Pacific Transducer Corp.
Pacific Welding Alloys
Mir-O-Col Alloy Company, Inc.
A. 0. Smith Corp.
Stulz-Sickles Co.
Syntron Rectifier Division
MS
SRSRESSSVERTSS
eenenanens
SSSQsgtosrsseeeeeveresery
wNBNVysSSsaseeseesssusersas
SSseeeessessecesessesyyyy
Tempil® Corporation
a list
of AWS
Mail me { information Thomson Electric Welder Co.
on AWS Membership 0
information
on Journal Subscription [J Tuffaloy Plant
Air Reduction Sales Co.
NAME
Union Carbide Corporation
TITLe Linde Company
COMPANY Victor Equipment Company
1077 Westinghouse Electric Corp.
X-Ray Department
 Welding

esearch Sponsored by the Welding Research Council

of the Engineering Foundation

SUPPLEMENT TO THE WELDING JOURNAL, OCTOBER 1960

Strength of Welded Aluminum-Alloy Box Beams

Bending tests demonstrate that welded aluminum-alloy

box beams have good strength and ductility. Proposed design

methods are shown to give satisfactory values of

bending ultimate and yield strengths

BY R. J. BRUNGRABER

SYNOPSIS. Welded box beams fabri fabricated by welding, the welded and 0.25% chromium and, in the
cated from plates of aluminum alloys beams being considerably stronger —H34 temper (which is the half-
5456-H321, 5154-H34 and 6061-T6 than similar extruded beams of hard condition achieved by roll-
not heat treated after welding) dis annealed material, the majority of ing), has a guaranteed yield strength
played good strength and ductility in the failures were due to lateral buck-
bending tests. Heat treatment of of 29 ksi and tensile strength of
alloy 6061-T6 beams after welding in ling; thus, the bending moment nec- 39 ksi. 5456 is the highest-strength
creased the strength but caused a essary to cause fractures in the commercial aluminum-magnesium
marked decrease in ductility. Test beams could not be established. In alloy. It contains about 0.20 5 %
specimens included beams with trans order to better study the ultimate magnesium, 0.8° manganese and
verse butt welds near the center as well carrying capacity of welded beams, 0.10% chromium and, in the —H-
as beams with longitudinal fillet welds a series of welded box beams, which 321 temper (which is about the
only. The bending ultimate and yield would have sufficient lateral stiffness quarter-hard condition achieved by
strengths were in satisfactory agree- so that lateral buckling would not rolling), has a guaranteed yield
ment with values calculated according occur, was tested. In this paper,
to proposed design methods. strength of 33 ksi and a tensile
the results of these tests of box strength of 46 ksi for thicknesses
Introduction beams are reported and compared up to 1'/, in. Alloy 6061 contains
with proposed design methods. about 0.25% copper, 0.6% silicon,
In connection with the development
of design rules for welded aluminum- 1.0% magnesium and 0.25% chro-
Material mium and, in the -T6 temper
alloy structures, an evaluation of
The three weldable aluminum which is achieved by heating to
the behavior of some typical welded
structural members was judged to alloys considered in this investiga- 970° F, quenching in cold water
be necessary. At the suggestion of tion were 5154-H34, 5456-H321 and artificially aging at 320-350
and 6061-T6. The first two are F), has a guaranteed minimum
the U. S. Navy, Bureau of Ships,
examples of the aluminum-mag- yield strength of 35.0 ksi and a
the Alcoa Research Laboratories
nesium, or 5000 series alloys, in tensile strength of 42.0 ksi.
tested a series of welded aluminum-
which desirable mechanical proper- In the process of welding, the
alloy I-beams.' Although the re-
sults of these tests demonstrated ties are obtained by alloying and mechanical properties of all of
cold working. Alloy 6061-T6 is the these alloys are reduced owing to
that satisfactory beams can be
most popular of the heat-treatable the heat of welding. In the case
aluminum alloys and, prior to the in- of the 5000 series alloys, the re-
R. J. BRUNGRABER is Research Engineer troduction of the 5000 series alloys, sulting properties in the immediate
Alcoa Research Laboratories, Aluminum Com was the alloy most commonly used vicinity of the welds approach
pany of America, New Kensington, Pa
in welded applications. 5154 is an those of annealed material, while
Paper presented at the AWS National Fall intermediate-strength alloy which only a short distance from the
Meeting held in Pittsburgh, Pa., Sept. 26-29
1960 contains about 3.5% magnesium weld, generally less than 3 in.,

WELDING RESEARCH SUPPLEMENT 4l?-s

 TEST SPECIMENS of the specimens. The values for
10° af all specimens are within +4 and
—3% of the average.
] The two types of specimens
which were tested are illustrated in
AVERAGE SECTION
- PROPERTIES FOR Fig. 1. The specimens of Type A
ALL BEAMS were fabricated from four pieces
A+ 12.860 in? of sawed plate by fillet welding the
Inn * 56.7 INA two flange plates to the two web
S, » 143 1nN5 plates. The specimens of Type B
were similar to those of Type A
with the exception that each of
the four plates was sawed apart
ute near the center ao le *=
joined by a butt weld. e plates
a et arte ata were an assembled into the box
beam so that the splices in the web
plates were displaced 4'/, in. to one
side of the center line, and the splices
— an yy in the flange plates were 4'/; in. to
SPECIMEN the other side of the center line.
Thus, at no section of the beam
was there a complete butt weld.
. The tensile properties of the
a ; materials were determined using '/s-
cook in. diam specimens in accordance
FILLET WELOS< j 4 TYPE 8 with ASTM Standard Methods’
SPECIMEN
and are recorded in Table 1.
+ BuTT weiss — e The values given for the mechanical
REACTIONS properties of the base metal are the
Fig. 1—Method of loading averages for the number of tests
shown in Table 1. In no case
did an individual test result differ
the original base metal properties Specimens from the average by more than
remain unaffected. The same is +2%. All the values shown for
true for alloy 6061-T6 except that The dimensions and section ele- base metal meet ASTM?® require-
the properties in the immediate ments of the 6-in. unsymmetrical ments and agree closely with pub-
vicinity of the welds are interme- box section selected for the beams lished typical values.‘ The values
diate between those of annealed tested in this program are shown in for the mechanical properties of
material and unaffected base metal. Fig. 1. The beams were tested the most highly heat-affected ma-
It is this variation of mechanical with the larger flange in compression terial in the vicinity of a weld were
properties throughout an aluminum to avoid local and lateral buckling. taken from the results of tensile-
weldment which makes the design The section properties given in property surveys run on the tension
of such members differ from that Fig. 1 are average values deter- flanges of the unspliced beams.
used for other metal members. mined from the actual dimensions All the welding was done using

Table 1—Results of Tensile Tests

Alloy Tensile Yield strength Modulus of
and strength, (0.2% offset), Elongation Thickness, elasticity,
temper Location psi psi in 4D, % in. psi
Properties of base metal
fuo _
5154-H34 Top flenge 43,300 16.0
Webs 43,500 18.0
Bottom flange 45,300 16.0 a ooo
5456-H321 Top flange 54,000 17.0 ~
Webs 52,500 20.0
Bottom flange 54,700 16.0
6061-T6 Top flange 46 ,400 18.0
Webs 45,300 SESERLSS
ssssssess:
20.0
Bottom flange 45,800 z 18.0 iooooc.;c
383888388
sS8S88858
Properties of most highly heat-affected metal in vicinity of welds on tension flange parallel to axis of weld
fur fyn
5154-H34 31,500 15,500
5456-H321 48 , 300 20 ,200
6061-T6 As-welded — 28 ,800 19,400
6061-T6 Reheat-treated and 45,800 42,000
aged after welding
@ Lowest value determined in either of two tensile property surveys.

418s | OCTOBER 1960

 the semiautomatic inert-gas shielded zone is designated as 6,. The l-in. diam rounds. Four of the
consumable-electrode process. Pri- results of the mechanical-property specimens—A-1, A-3, B-1 and B-3
or to making the welds, the parts surveys are shown as Figs. 2-5. were tested in a setup in which the
were cleaned by degreasing. All No tensile tests were made on reactions were introduced through
welding was done in the down-hand Specimen A-4 since the hardness 2-in. diam rounds which were re-
position. After each pass, the sur- survey demonstrated that reheat strained against movement in the
face of the deposited bead was treatment had essentially restored axial direction of the beam. ‘Thus,
cleaned by manual wire brushing the original strength in the heat- as the beam deflected, the friction
prior to depositing the next suc- affected zone. between the beam and the supports
cessive bead. The pieces were not introduced axial forces in the beam.
preheated and were permitted to Test Procedure This setup was later modified by
cool to a temperature tolerable to The beam tests were made in a putting the supports on roller nests
the touch before starting a subse- 300,000-lb hydraulic-type testing to permit free axial movement, and
quent pass. machine using the 100,000-lb load the modified setup was used to test
Each of the flange-to-web fillet range. Loads were applied through Specimens A-2, A-4, B-2 and B-4.
welds was deposited in two passes
without prior edge preparation.
In preparation for making the
transverse butt welds, the edges
were beveled to form a 60 deg vee
with a '/,-in. abutting lip. For
the */,-in. flanges, four passes were
used to fill the vee, the underside
of which was then back-chipped, Si
and a fifth pass filled the resulting }
groove. The '/,-in. flanges and the *HARONESS,
ROCKWELL
iin. webs were welded in the if
same manner except that only two =
passes were needed to fill the vee BABVBBeeqeaaae = {A
and then one additional pass was
used to fill the back-chipped groove. M0000
a,
The above procedure represents
typical practice for making such
weldments. >>
b>
>_>»
_»_
>»
——————
OOQAP
ASS2»>_>»
JOOOO2> 60
Specially prepared wooden blocks 80 60 Py @°
ULTIMATE AND YIELD STRENGTH HARDNESS, ROCKWELL “F
were used to maintain alignment MARONESS, ROCKWELL“F
of the parts during welding. The ECTIONW |S SYMMET THE CENTER NE AND ORAWN TO SCALE
welding caused a slight warping of PERTY DETERMINATIONS LOCATIONS ALONG THE PLATE CENTER LIWES
the flanges and some bowing of ExTENT OF PEDUCE S SHOWN FOR THE REGIONS AROUND THE WELD
the entire beam due, no doubt, to TENSION FLANGE
THE TENSILE ANC THE SOLID NE REPRESENTS AN AVERAGE OF
to the large difference in the sizes Tw eTeEawinar
of the top and bottom flanges. Fig. 2 Mechanical-property surveys on welded-box-beam
This bowing resulted in an initial cross section. (Specimen A-1, 5154-H34)
upward center deflection of about
'/,in., which is not believed to have
affected the test results.
For each of the Type A specimens,
the extent of the heat-affected and
reduced-strength zones around the
welds in the tension area was de-
termined by means of hardness and “F"
ROCKWELL
tensile-strength surveys made on
sections sawed from the ends
of the beams. As defined in
Reference 5, the heat-affected zone HARONESS
is the region in the vicinity of a
weld in which the heat of welding
has reduced the mechanical prop-
erties by a significant amount,
such as 5%; and the reduced-
strength zone is the region in the
vicinity of a weld such that, if the SPECIMEN A-2 }
5456-32 | |
properties within this region are é 1
considered to be those of the softest 80 60 40 60 80 00
HARONESS, ROCKWELL“F“ MARONESS, ROCKWELL ‘F
material in or adjacent to the weld TENSILE ULTIMATE AND YIELD STRENGTH, KS
and the properties outside this
region are considered to be those of worTes THE CROSS SECTION 1S SYMMETRICAL ABOUT THE CENTER LINE AND ORAWN TO SCALE
ALL PROPERTY DETERMINATIONS WERE MADE AT LOCATIONS ALONG THE PLATE CENTER LINES
unaffected base metal, the resulting THE EXTENT OF REDUCED-STRENGTH ZONE, by, 'S SHOWN FOR THE REGIONS AROUND THE WELD
weighted average properties will ON THE TENSION FLANGE
FOR THE TENSILE AND YIELO STRENGTH SURVEY THE SOLIO LIWE REPRESENTS AN AVERAGE OF TWO
satisfactorily represent the actual OE TERMINATIONS
strength of the welded member. Fig. 3—Mechanical-property surveys on welded-box-
The extent of the reduced-strength beam cross section. (Specimen A-2, 5456-H321)

WELDING RESEARCH SUPPLEMENT 419-s

The modified setup is pictured in A study of the support movements setup are from 5 to 10% high.
Fig. 6. indicated that, for deflections less In all tests, the load was applied
Specimen A-1, which had not been than about 1 in., both test setups in increments until failure occurred
loaded to failure in the initial test would have given essentially the or until the deflection limit of the
setup, was later loaded further in same applied load. This more than test setup was reached. For each
the modified test setup. Compari- covers the elastic range of all the increment, deflection measurements
son of the stress-deflection curve beams. However, for larger de- using Wissler dials were made at
for this specimen, as determined in flections, the results of Fig. 7 the midspan and at one of the load
the two different test setups (Fig. indicate that the original test setup points. When the setup of Fig. 6
7), and study of the support move- would result in loads up to 10% was used, the movement of the
ments in the tests of A-2 and A-4 higher than would the modified supports was determined with scales
and B-2 and B-4 permitted an setup. No doubt, the latter setup graduated in '/1 in. so that the
evaluation of the additional con- gave the best measure of the load variation of the moment arm could
straint offered by the initial test capacity of a beam and the ultimate be determined. In the case of the
setup. loads determined with the original original test setup, the moment
arm was considered to remain con-
stant at 24 in. During the tests
of specimens A-3 and B-3, strains
were measured near the center line
of the beam with an Ames dial
arranged to measure elongations
in a 2*/;-in. gage length.

“F*
ROCKWELL
HARONESS, Results of Beam Tests
The results, of the beam tests
are plotted in Figs. 7-10, and a
typical tested specimen is shown in
Fig. 6. Tables 2 and 3 contain
analyses of the test data. In the
plots of Figs. 7-10, the vertical
HARONESS | © scale in each case is the apparent
TENSILE stress, M/S, where M is the ap-
STRENGTH
60 60 100 plied moment and S is the section
60 60 40
MARONESS, ROCKWELL F HARONESS, ROCKWELL“F” modulus related to the tension
TENSILE ULTIMATE AND YIELO STRENGTH, KS flange.
NOTES THE ROSS SECTION 'S SYMMETRICAL ABOUT THE CENTER \ NE 4NO OR AWN TO SCALE In Figs. 7-9, apparent stress is
ALL PROPERTY DETERMINATIONS WERE MADE AT LOCATIONS ALONG THE PLATE CENTER LINES plotted vs. centerline deflection,
THE EXTENT OF REDUCED-STRENGTH ZONE °, S SHOWN FOR THE REGIONS AROUND THE WELD Figs. 7 and 8 giving the results for
ON THE TENSION FLANGE
the unspliced beams and Fig. 9
Fig. 4—Mechanical-property surveys on welded-box-beam the results for the spliced beams.
cross section. (Specimen A-3, 6061-16, as-welded) In Fig. 10, apparent stress is plotted
vs. average strain across the tension
flange for specimens A-3 and B-3.
For specimen B-3, the strains were
measured at two locations in the
tension flange; in one case the
gage length included the butt weld,
and in the other case it did not.
Both curves are shown in Fig. 10.
Also plotted in Fig. 10 (as a
dotted line) is a constructed 10-in.
gage length stress-strain curve for
the tension flange of specimen
B-3. This curve was constructed
by adding the elongations measured
on the 2*/;-in. gage length centered
on the transverse weld to the elonga-
Vi a oe ERO RR RR NE NS NEN tions for the remaining 7°/; in.,
7 and then dividing these total elonga-
606! - \
ra SPECIMEN A-4 tions by 10 in. to convert to strain.
/ REHEAT TREATED AND It was assumed that the remaining
i AGED AFTER WELDING “Sa.
“Sn.
“Sn,
“en,
en
a 7 °/s in. of the 10-in. gage length
x would have developed the strains
1 that were measured on the 2*/s-in.
—— a & WS
100 80 .~— «60 60 80 IC gage length that did not include the
HARDNESS, ROCKWELL "F" HARONESS, ROCKWELL “F" butt weld. Since there were no
transverse welds in specimen A-3,
NOTES:
THE CROSS SECTION iS SYMMETRICAL THE CENTER-LINE ANDO DRAWN TO SCALE the curve in Fig. 10 and the yield
THE HARONESS READINGS WERE MADE ALONG THE PLATE CENTER-~LINES strength determined from it are
THE EXTENT OF REDUCED STRENGTH ZONE, by, WAS TAKEN AS ZERO independent of gage length.
Fig. 5—Hardness survey on welded-box-beam cross section In Figs. 7-9, the calculated curves

420-s | OCTOBER 1960

 were determined assuming elastic
action and considering deflections
due to both shear and bending.
In the calculations, the modulus
values used were those given in
Reference 4 for the respective
alloys. These values are recorded
in Table 1. In Fig. 10, the straight-
line portions represent the modulus
for 6061-T6—10,000,000 psi.

Comparisons of Results of Beam

Tests with Calculated Strengths
In Reference 5, the yield strength
of awelded member is given as:
he - hye = Ari/A(fyo
t/Alfy— fy)
aa fy) (1)
where
fy yield strength of welded sec-
tion.
fy. = minimum yield strength of
unaffected base metal.
minimum yield strength in
heat-affected zone.
minimum yield _ strength
across butt welds (corre-
sponding to 0.2% offset on a
10-in. gage length).
reduced-strength area of cross
section due to longitudinal
welds.
reduced-strength area of cross
section due to transverse or
localized welds. (Portions of
the cross section within the
reduced-strength zones for
both a longitudinal and a
transverse weld should be in-
cluded in A,; but not in A,,. REPRESENT DAT
In the same reference, the ulti- PEPRESENT DATA
mate strength of a welded member Fig. 7—Apparent stress vs. deflection for beams containing
is given as: no transverse welds. (Specimens A-1 and A-2)
po f A,/A(f
where
S. tensile strength of welded
cross section.
ha minimum tensile strength of
unaffected base metal. STRESS AT O
minimum tensile strength in / PERMANENT SE
heat-affected zone.
net area of cross section.
= total reduced-strength area of
cross section (where A, ,
plus A,,). M/S,
KSI
Also in Reference 5, the above
formulas are stated to be applicable STRESS AT 0.002 IN/IN
to tension flanges of beams if the OFFSET STRAIN: 312 KS
ratios A,/A, A,/A and A,./A PERMANENT DEFLECTION=:O./9IN
are based on the net area of the + + —
tension flange plus the outermost ~fyh 2 19.4 KSI
one-third of the net web area TENSION
FIBER,
between the tension flange and the APPARENT
STRESS
ON
EXTREME
neutral axis. ie Ae SPECIMEN A-4
The yield strengths and ultimate SPECIMEN A~5 606I-T6 REHEAT- TREATED
606I-T6, AS- WELDED AND AGED AFTER WELDING
strengths of the welded box beams
were calculated by use of formulas 4 6 8 Le) 2 4 6
1 and 2, giving the results in Table
3.* The values for A,./A and CE NT ER LINE DEFLECTION, IN.
A,./A were determined using the SOLID POINTS REPRESENT DATA DETERMINED WITH INITIAL TEST
values of 6, shown in Figs. 2-5. SETUP, WHICH RESTRAINED MOVEMENT OF REACTIONS
In calculating A,, for the trans-
OPEN POINTS REPRESENT DATA DETERMINED WITH REVISED SETUP,
WHICH PERMITTED FREE MOVEMENT OF REACTIONS
*The different base-metal properties in the Fig. 8—Apparent stress vs. deflection for beams containing
webs and the flange were considered in the calcu-
lations by using a weighted average no transverse welds. (Specimens A-3 and A-4)

WELDING RESEARCH SUPPLEMENT | 421-s

 —— | —_+j | ; . -
| STRESS AT 0.'8 IN
: ‘PERMANENT SET =
, 42.8 KS!
—— —SS a
* 37.6 KS! STRESS AT 0.18 IN. | fy*fyn= fyo =42.0 KSI
PERMANENT SET =
37.8 KSi STRESS AT 0.002 IN./IN
OFFSET STRAIN ON IO IN
—* a ; GAGE LENGTH « 27.0 KS!"
STRESS AT 0.18 IN. PERMANENT DEFLECTION
STRESS
EXTREME
ON PERMANENT SET = 20.17 In.
25.5 KS! ty = 25.9 KS!
fy = 19.9 KSI
TENSION —fyy 215.5 KS!
FIBER,
APPARENT |
T py 4 ——_____+—______+—_
SPECIMEN B-! SPECIMEN B-2 SPECIMEN 8-3 SPECIMEN B-4
5154-H34 5456-H32! 606!-T6, AS WELDED 606!-T6, REHEAT-TREATED
= | i| ANDO AGED AFTER WELDING
4 7 6 fe) 2 2 4
CENTER LINE DEFLECTION, IN.
NOTES: SOLID POINTS REPRESENT DATA DETERMINED WITH INITIAL TEST SETUP, WHICH RESTRAINED MOVEMENT OF REACTIONS
OPEN POINTS REPRESENT DATA DETERMINED WITH REVISED SETUP, WHICH PERMITTED FREE MOVEMENT OF REACTIONS
Fig. 3—Apparent stress vs. deflection for beams containing transverse welds near mid length

versely welded beams, only the

2%9N GAGE LENGTHWELDNOT
THE TRANSVERSE INC transverse
. weld
. in the . flange was
nal a an considered since those in the webs
were not at the same cross section.
Since it was impractical to establish
us CeowsTRucTeED cuRve FoR \oIN by tests the 10-in. gage-length
GAGE LENGTH INCLUDING THE
TRANSVERSE WELC yield strength across the butt welds
T T ] P 2% im Gace LENGTH incLuDING in the tension flanges, the values of
OO2 OFFSET YIELO STRENGTH: Si Ons THE TRANSVERSE =. a used in calculating the weighted
+ + + + a Ee SG : average yield strengths in Table
TENSION
FIBER 3 were determined from the mini-
STRE
APPARENT } 4 a . Se eee CARs der, abe mum yield strength in the heat-
oe Se ee 6 Cars affected zone by multiplying by the
ratio 1.36. (In six tests of trans-
002003 004 005 006 0 OOF 002 003 004 005 006 verse butt welds in plate specimens
AVERAGE STRAIN ACROSS CENTER SECTION OF TENSION FLANGE, IN AN of these alloys made in connection
Fig. 10—Apparent stress vs. strain for tension flanges of the as-welded 6061-T6 beams. with another investigation, this
(Specimens A-3 & B-3) ratio varied from 1.32 to 1.42

Table 2—Summary of Beam-test Results

Section
Total modulus
center with
Moment deflec- respect
Alloy Maximum arm at Maximum tion at to tension
Spec. and Filler load, maximum moment, maximum flange, Condition of beam
number temper metal Bases load, in. in.-Ib load, in. S,, in? after testing
Specimens with longitudinal welds only
A-1 5154-H34 5154 58 , 450¢ 25.08 733,000 12.99 14.52 Nofracture
A-2 5456-H321 5456 76,0004 25.41 965 ,000 9.49 14.80 No fracture
A-3 6061-T6« 4043 61,900: 24.00 743,000 5.10 13.94 Fracture in tension flange
A-4 6061-T6° 4043 62,1007 24.39 756 ,000 1.72 14.08 Fracture in tension flange
Specimens with longitudinal and transverse welds
B-1 5154-H34 5154 57 , 300° 24.00 . 14.50 Fracture in tension flange
B-2 5456-H321 5456 67 ,7004 24.86 ’ 14.91 Fracture in tension flange
B-3 6061-T6* 4043 50 , 500: 24.00 ‘ 14.11 Fracture in tension flange
B-4 6061-T6° 4043 52,0004 24.25 , 13.91 Fracture in tension flange
@ As-welded.
” Reheat treated and aged after welding.
¢ Maximum load determined with the original test setup.
4 Maximum load determined with test setup using reactions on rollers.

422-s | OCTOBER 1960

Table 3—Comparison of Beam-test Results with Calculated Strengths
Ratio of Ratio of
Calculated apparent Calculated apparent
weighted stress at weighted stress at
average yield to Apparent average maximum
Apparent yield calculated stress at ultimate load to
stress strength weighted maximum strength weighted
Alloy at yield of tension average load, of tension average
Spec. and ——A,,/A’°—— A,,/A” (M/S), area, yield (M/S), area, ultimate
number temper Webs Flange flange psi f,, psi strength psi , psi strength
Specimens with longitudinal welds only
A-l 5154-H34 0.71 0.38 0 29, 000° 26 , 8007 1.08 50,500 , 900 1,3]. ¢
A-2 5456-H321 0.34 0.22 0 40 ,500° 36 , 000: 1.3 65,200 , 900 249.4
A-3 6061-T6¢ 0.45 0.42 0 31,200/ 31,900 0.98 53,300 , 500 . 38"
A-4 6061-T64 0 0 0 44 600° 42,000 1.06 53,800 5,700 . 189
Specimens with longitudinal and transverse welds
5154-H34 0.71 0.38 0.62 25 , 500° 19,900 47,400 , 700
5456-H321 0.34 0.22 0.78 37 ,800° 29,300 56 ,500 , 200
6061-T6° 0.45 0.42 0.58 27 ,000/ 25,900 43,000 , 900
6061-T64 0 0 0 42 ,800° 42,000 45,300 , 700
® Ratio of reduced-strength area due to longitudinal welds divided by total area.
» Ratio of reduced-strength area due to transverse welds divided by total area
© As-welded.
4 Reheat treated and aged after welding.
€ Determined from Figs. 7-9 at a permanent center deflection of 0.18 in
/ Determined from Fig. 10 at a permanent strain of 0.002 in./in
? Maximum load determined with test setup using reactions on rollers
h Maximum load determined with the original test setup.
Test discontinued before fracture.
Values of fy: determined from the average curves in Figs and 3 were used in eq 1

with an average value of 1.36.) f,. might result in permanent deflec- heat treated after welding, deflec-
Also in Table 3, the calculated tions of as much as '/7 of the span tions at fracture were roughly
weighted average yield strengths length. In the final column of 1/1; of the span (longitudinal welds
are compared with measured yield Table 3 is given the ratio of the only) and '/ of the span (trans-
strengths for the beams. For speci- apparent stress at the maximum verse welds). Corresponding deflec-
mens A-3 and B-3, the yield strength test load of the beam to the calcu- tions for the 6061-T6 beams heat
was determined as the apparent lated weighted average ultimate treated after welding were only
stress (M/S) at which a permanent strength. These ratios are repre- 1/, to '/. as great.
strain of 0.002 in. /in. was developed sentative of the values of plastic 3. For all beams, except the
in the extreme tension fibers (see shape factor which one would expect transversely welded 6061-T6 beam
Fig. 10). The permanent center for the relatively husky box section heat treated after welding, ratios of
deflections corresponding to these tested, especially when one con- calculated stresses (M/S) at maxi-
yield strengths were determined siders that specimens A-3, B-1 mum load to weighted average
from Figs. 8 and 9 as being 0.19 and B-3 were tested in a setup in strength values for the tension
and 0.17 in. for specimens A-3 which the measured ultimate load flange determined in accordance
and B-3, respectively. For the was probably 5 to 10% high. with the design methods proposed
remaining specimens, the yield in Reference 5 were greater than
strength was considered as_ the Summary and Conclusions 1.1. Ratios of this magnitude
apparent stress at which a perma- Welded-aluminum _ box beams, would be expected for a cross sec-
nent center deflection of 0.18 in. both with and without transverse tion of this shape.
was developed. This represents the splices, were tested and the results 4. At loads corresponding to the
average for specimens A-3 and compared with design recommenda- weighted average yield strength of
B-3 and is '/ 4. of the span. tions made in Reference 5. The the tension flange, calculated in
The measured yield strength and beams were made from alloys 5456- accordance with the design proce-
calculated weighted average yield H321, 5154-H34 and _ 6061-T6. dures in Reference 5, values of
strength values shown in Table Some of the 6061-T6 beams were permanent set were less than '/ i
3 are also plotted on the curves in heat treated after welding. The of the span, indicating the proposed
Figs. 7-9, demonstrating that when results of these tests can be sum- design method to be reasonably
the apparent stress in the tension marized as follows: conservative.
flange reached the weighted average 1. Aluminum-magnesium alloy
yield strength, only a modest 5456, 5154) beams with longitu- References
amountof permanent set—less than dinal welds, but no transverse welds, 1. Holt, M., and Matthiesen, R. B., “Static
Tests of Welded Aluminum Alloy Beams rue
'/9 Of the span length—had de- deflected up to the limit of the test WELDING JOURNAL, 34 (7), Research Suppl., 313-s
veloped in the beams. Also shown setup (deflections equal to about 320-8 (1955)
2. Anon., Tentative Methods of Tension
on the curves of Figs. 7-9 are the '/, of the span) without fracture. Testing of Metallic Materials, ASTM Designation
values of f/f, (base metal yield Transversely welded beams of the E8-57T, pp. 105-108
strength) and f/,, (minimum yield 3. Anon., Specifications for Aluminum Alloy
same alloys reached deflections on Sheet and Plate (tentative), ASTM Specification
strength in heat-affected zone) for the order of '/. of the span before B209-59T, pp. 195, 197 and 198
the individual beams, demonstrat- fracture, or about 5 times the 4. Anon., Alcoa Aluminum Handbook, Alumi-
num Company of America, pp. 23-24 (1959)
ing that /,, is an ultraconservative elastic deflection corresponding to 5. Hill, H. N., Clark, J. W., and Brungraber,
value of yield strength for a welded the ultimate load. R. J., “Design of Welded Aluminum Structures,”’
Proc. ASCE, Jnl. Structural Div., 86 (ST6), 105
beam, while stressing a beam to 2. For the 6061-T6 beams not 109, 117 (June 1960)

WELDING RESEARCH SUPPLEMENT | 423-s

Comparative Properties of Aluminum-Alloy Weldments

The effects of welding on various aluminum

alloys are compared on the basis of static mechanical

and fatigue properties

BY |. L. STERN, H. V. CORDIANO AND V. A. DiGIGLIO

ABSTRACT. The static mechanical and

fatigue properties of butt welds in '/,- Table 1—Base Plates and Filler Materials Evaluated
in. thick aluminum alloys have been Base plate Nominal Type of '/:.-in. diam Nominal
investigated and compared. In addi- (alloy and temper) composition” bare electrode composition?
tion, information was obtained as to
the effects of welding position on the 6061-T6 (Solution heat- 1.0 Mg, 0.6 Si, 4043 5.0 Si
over-all results. Alloys included were treated and then artifi- 0.25Cu, 0.25Cr
6061-T6, 5154-H34, 5086-H32, 5456- cially aged)
H321 and 5454-H34. 5154-H34 (Strain-hardened 3.5 Mg, 0.25 Cr 5154 3.5 Mg, 0.25 Cr
It was shown that minimum yield and stabilized, half-hard
strengths of welded butt joints in temper)
various aluminum-magnesium alloys 5086-H32 (Strain-hardened 4.0 Mg, 0.45 Mn, 5356 5.0 Mg, 0.10 Mn,
approach the values for the correspond- and stabilized, one-fourth 0.10 Cr 0.10 Cr, 0.10 Ti
ing annealed base plate. Reductions hard temper)
were also observed for the tensile 5456-H321 (Strain-hardened 5.25 Mg, 0.8 Mn, 5556 5.25 Mg, 0.8 Mn,
strength. The effect of welding on the and stabilized, one-fourth 0.10 Cr 0.10 Cr, 0.10 Ti
6061-T6 alloy resulted in reduced hard temper)
tensile properties. 5454-H34 (Strain-hardened 2.75 Mg, 0.8 Mn, 2.75 Mg, 0.8 Mn,
and stabilized, half-hard 0.10 Cr 0.10 Cr, 0.10 Ti
Fatigue strengths (large scale) at
100,000 cycles for flat welded joints in temper)
various aluminum-magnesium alloys @ Percent of alloying elements. Aluminum and normal impurities constitute remainder
were found to be about the same and
approximated 14,500 psi. The corre-
sponding fatigue strength for the are replacing the 6061-T6 alloy for Description and Apparatus
6061-T6 weld assemblies was approxi- many applications. With the exception of the 6061-T6
mately 11,000 psi. This represented a While data relative to the proper- alloy, all alloys were of the alumi-
significant reduction from correspond- ties of welds in both types of alloys num-magnesium type in which the
ing base-plate properties. The change have been published, there are sig- tensile properties of the alloy are
from flat to vertical position welding nificant gaps in the available infor-
resulted in a decrease of fatigue controlled by strain hardening. In
mation, so that, in many cases, per- the case of the 6061-T6 alloy, the
strength. formance under the conditions of
The fatigue strength (small-scale, properties of the plate are developed
Krouse-plate-type) at 1,000,000 cycles field application cannot be ac- by heat treatment. All plates and
ranged from approximately 9000 to curately predicted. In addition, bare electrodes were in conformity
13,500 psi for the various alloys welded the occurrence of failures in alumi- with the chemical compositions of
in the flat and vertical positions. num weldments, which would be the applicable specifications, listed
predictable on the basis of an analy- in Table 2.
Introduction sis of information from diverse All welding was performed with a
The heat-treatable 6061-T6 alumi- sources, indicates a lack of apprecia- standard semiautomatic, inert-gas-
num alloy has been extensively used tion of joint properties on the part shielded metal-arc welding set, em-
for welded-structural applications. of some designers and fabricators. ploying a “push’’-type wire-feed
However, aluminum - magnesium In view of the above, an investi- unit and a 400-amp d-c, variable-
type alloys are finding increasingly gation was initiated to determine the voltage motor-generator power
wider application to weldments and effects of welding on various alumi- source. Helium was used as the
num alloys proposed for structural shielding gas. The two fatigue-
I. L. STERN is Head, Welding Development applications and to determine the
Section, H. V. CORDIANO is Acting Head, testing machines used for this work
Mechanics Branch and V. A. DiGIGLIO is properties of the resulting weld are described as follows:
Supervisory Welding Engineer, Naval Material assemblies.
Laboratory, New York Naval Shipyard, Brook- The specific objective of the Large-scale Fatigue-testing Machine
lyn, N. Y.
Paper presented at AWS 4ist Annual Meeting phase of the investigation reported The machine which is described
held in Los Angeles, Calif., Apr. 25-29, 1960. herein was to acquire information in detail in Reference 1 is of the
The over-all investigation has been sponsored relative to the properties of the vibratory type and may be classified
by the Navy Department, Bureau of Ships. 1/.-in. thick butt-welded assemblies as a constant-load repeated-bending
The opinions or assertions contained in this listed in Table 1, which would be machine. The _ repeated-bending
paper are the private ones of the authors and are useful as guides to the designer and moment which is generated by
not to be construed as official or reflecting the
views of the Naval Service at large. fabricator. rotating eccentric disks is impressed

424-s | OCTOBER 1960

Table 2—Chemical Composition Requirements
Aluminum
alloy and Material ————-Elements, %*% Others, %
temper specification Me Mn Si Each Total
6061-T6 Fed. Spec., QQ-A 0.80-1.20 0.15 0.40-0.80 0.05 0.15
(plate) 327b (Mar. 7, 1958)
5154-H34 MIL-A-17357B 3.1-3.9 0.10 0.05
(plate) (May 6, 1955) (Si + Fe)
5086-H32 MIL-A-19070 (ships) 3.5-4.5 20-0.70 0.40 aad 0.05
(plate) (Oct. 7, 1955)
5456-H321 MIL-A-19842 (ships) 50-1.00 40 ae ; 05
(plate) (Mar. 8, 1957) (Si + Fe)
5454-H34 MIL-A-21598 (ships) 50-1.00 40 05
(plate) (Nov. 18, 1958) (Si + Fe)
4043 MIL-E-16053G ne 05 oul 0.0008 05
(electrode) (Apr. 28, 1959)
5154 MIL-E-16053G | 10 45 ia : .2 0.0008
(electrode) (Apr. 28, 1959) (Si + Fe)
5 356 MIL-E-16053G 05-0.20 50 0.0008
(electrode) (Apr. 28, 1959) (Si + Fe)
5556 MIL-E-16053G 05-0.2 50-1.00 40 0.0008
(electrode) (Apr. 28, 1959) (Si + Fe)
5554 MIL-E-16053G 2.4-3.0 05-0.2 50-1.00 , 40 0.0008
(electrode) (Apr. 28, 1959) (Si + Fe)
@ Composition in percent maximum unless shown as a range Remainder is aluminum

Fig. 1—Setup of specimen on large-scale fatigue-testing machine Fig. 2—Small-scale (Krouse-plate-type) fatigue-testing
machine and test setup

on the specimen assembly con-

sisting of a welded-aluminum speci
men fastened by bolts to a heavy-
steel base plate. The specimen
assembly is so supported that the
impressed bending moment is uni-
form along the span. The cross-
sectional dimensions and spacing of
the aluminum specimen and steel
base plate cause the neutral surface
of the composite section to fall in
the steel base plate. Under these
conditions the outer fibers of the —_AlIS! 304 GORROSION-RESISTING STEEL BACKING STRAP.
'/-in. thick aluminum plate are CLAMPED IN POSITION, REMOVED AFTER WELDING ST SIDE
approximately 2'/, in. away from
the neutral surface thereby pro- JOINT CLEANING PROCEDURE—Joint surfaces freed of oil with Sovasol No 2 and cleaned with
a corrosion resisting steel wire brush
viding for nearly axial loading in the WELDING EQUIPMENT AND 2>OWER SOURCE—Standard semi-automatic, inert-gas-shielded
specimen. The machine applies metal-arc welding set employing a ‘‘push"’ type wire feed unit and a 400 ampere d.c variable
completely reversed stresses to the voltage motor generator
SHIELDING GAS AND FLOW RATE Helium, 100 CFH
specimen since it is not set up with ELECTRODE DIAMETER— ein
any intentional initial stresses. The CURRENT AND POLARITY—D.C Reverse.
frequency of stress reversal is de- WELDING POSITION FLAT VERTICAL
pendent on the stiffness of the Amperes 170-195 135-160
specimen assembly. The machine Volts 9-33 25-29
NUMBER OF LAYERS
maintains this frequency constant lst Side c?
until failure occurs, at which time 2nd Side* i1
the safety cut-off switch is auto- TOTAL 4 3
matically actuated. Figure 1 shows * The root of the first side chipped to sound metal prior to welding the second side
a specimen setup in the machine WELDING PROCEDURE—Preheat—None
Interpass Temperature—150° F Max. (air cooled)
for test. Postweld Treatment—Air Cooled
Technique—String beading with layer thickness not in excess of
Small-scale Fatigue-testing Machine Direction of Welding—Reversed between layers
Distortion—Maximum deviation from flat surface did not exceed
This machine is identified as a
Krouse-plate fatigue-testing ma- Fig. 3—Joint design and welding procedure

WELDING RESEARCH SUPPLEMENT | 425-s

 chine. It subjects a_ cantilever-
LOCATION OF LARGE SCALE FATIGUE SPECIMENS beam-type specimen to repeated
—— FLAT OR VERTICAL PosITION — bending to a predetermined alter-
wes - - nating deflection. The deflection
} — for any desired stress is determined
Dired iON r FImAL ROLL {
by static loading of the specimen
with dead weights at the time it is
mounted in the machine for fatigue
testing. The machine applies a
completely reversed stress at a
frequency of 1750 cpm. The width
of the specimen is tapered in order to
provide a constant stress along the
LOCATION OF LARGE SCALE TENSILE SPECIMENS length. Figure 2 shows a specimen
FLAT _+}- VERTICAL set up in the machine for test.
[position ee _ e
: Procedure
q { Details of joint design and pro-
Mf cedure are shown in Fig. 3. As
yy 3 indicated therein, all assemblies were
} ! R '/,in. thick. The general procedure
Boia! 3 | employed in fabricating assemblies
a leet a “1 and selecting specimens is indicated
ole vail— oo in Fig. 4. Assemblies for large-
ae scale specimens were 36 xX 38 in.;
LOCATION OF SMALL SCALE FATIGUE AND TENSILE SPECIMENS assemblies for small-scale specimens
fe Farpat erecusens were 18 X 19 in. Test specimens
res8panty at tle 3) PLATE TENSILE SPECIMENS were oriented so that the direction
Be re —_— of final roll of the plate was per-
\~| iy I pendicular to the longitudinal axis
Lid
dna
|=at of the specimen (parallel to the
=x / direction of welding). The maxi-
J =>
R R mum deviation from a flat surface
oe for all test specimens did not exceed
pie
ppv avalon
wal
watiee
= eeEE
ee
rt —-| oy or
*4]4)—j)— aD '/,in. Welds were made with each
gence » a R— HELO IN RESERVE plate and electrode listed in Table 1
SPECIMENS 4 SPECIMENS
yrenSu.§ pa ears a in the flat-welding position. In
VERTICAL PosiTow HARONESS SPECIMENS
, vertical w
Fig. 4—Layout of test assemblies and specimens escrager Beewe erya age oy
5154-H34, 5086-H32 and 5456-H321
alloys.
After fabrication, each assembly
was examined visually and radio-
ORIGINAL DESIGN graphically to determine weld
quality and soundness and then
\+—€ oF wer NG" Beles dvifiod CoOi) sectioned. Initially, the large-scale
/\ ‘ - specimen of the original design
> | mo | shown in Fig. 5 was used for fatigue
o and tensile tests. However, in the
—_—@ -}+—_@—_+_@+- course of tests, some failures outside
the gage section were encountered
e—_o—o o ) ® | due to the moderate stress concen-
® a) tration existing near the fillet and
Tion
1FINAL
or
ROLL
Ommec at the bolt holes. In view of the
above, the modified design shown in
a 38" Fig. 5 was used for subsequent tests
BREAK ALL MODIFIED DESIGN whenever it was necessary to mini-
ee mize failures in these areas. The
4 aarineg “pitacin following tests were conducted with
<N the weld reinforcement intact:
9 ) Large-scale Fatigue Tests
2 These tests were conducted on the
specimens indicated below in ac-
> cordance with detailed procedures
o>
6 O
outlined in Reference 1.
? (a) Eight specimens of each of the
OF
DIRECTION
ROLL
FINAL
alloys (10 of the 5154-H34) welded
in the flat position. (b) Eight speci-
mens of each of the alloys welded in
the vertical position. (c) Eight
Fig. 5—Sketch of large-scale specimens for fatigue and tensile tests base-metal specimens of each of the

426-s
| OCTOBER 1960
alloys (9 for the 6061-T6 alloy). stress-strain data. The per cent and the yield strength (0.2% offset
The operating speed of the elongation in 2 in. was not deter- in 2-in. gage), tensile strength and
machine was held constant through- mined because the nonuniform plas- per cent elongation in 2 in. were
out the test at a value within the tically deformed fractured surfaces determined. Stress-strain curves
range of 460 to 475 rpm established could not be fitted together. were recorded by autographic
with each specimen assembly. All means.
specimens were tested until a fatigue Small-scale Tensile Tests
Three plate and four welded Macroetch Examination
crack at least 2 in. long developed
adjacent to the weld or in the plate. specimens were machined for each Three specimens of each alloy
Since it was desired to determine alloy as shown in Fig. 6. Tensile were saw cut from representative
the fatigue strength for a life of tests were conducted in accordance weld areas. Each specimen was
100,000 cycles, and since no infor- with the procedure of Reference 2, polished and etched with a 2% solu-
mation was available to indicate
what this stress level should be, an
estimate of desired fatigue stress was Table 3—Deviations from Radiographic Requirements
made from related available infor-
mation and four specimens were Deviate from
Plate alloy Bare Welding Specimen Specimen x-ray standard
tested at this level. Depending on andtemper’__ electrode position type nos. MIL-STD-437 (Fig. 7)
whether the median life obtained
6061-T6 4043 Fiat Large fatigue 7 and 8 Excessive porosity
was greater or less than 100,000 6061-T6 4043 Vertical Large fatigue 1,2and3 Excessive porosity,
cycles, a second stress level was 6061-T6 4043 Vertical Large tensile 1,2and3 Excessive porosity”
estimated which would bound the
fatigue strength for 100,000 cycles, ! Results were based on overhead position weld standard of MIL-STD-437 (see Fig. 7). No
and four additional specimens were standard for vertical position welds was provided
NOTE All welds, other than the 6061-T6 assemblies listed above were in conformity with
tested. The data were plotted on MIL-STD-437A (see Fig. 7)
semilogarithmic paper, and the fa-
tigue strength at 100,000 cycles was
determined by interpolation along
a straight line fitted to the data by ne
the “method of least squares.”
The stress levels are assumed to be
sufficiently close to permit assump- BREAK ALL
tion of a straight-line variation over SHARP CORNERS ROLL
FINAL
OFTION
DIREC
the small interval on semilogarith-
mic paper.
Small-scale (Krouse-plate-type)
Fatigue Tests
Krouse-type specimens were ma-
chined as shown in Fig. 6. Em- (ALL SCALE mn TENSILE
hk 114 SPECIMEN
1 8D
———
ploying the standard Krouse setup
and operating procedure, four speci- BREAK ALL +
SHARP CORNERS
mens of each alloy were tested to
failure at each of two stress levels
below and above the fatigue
strength) at 1,000,000 cycles with a
completely reversed stress cycle at a
frequency of 1750cpm. Thesestress
levels were estimated by the same
procedure used for the large-scale
specimens and the fatigue strength
for 1,000,000 cycles was determined
by interpolation along the straight
line established by the ‘method
of least squares.”” The 1,000,000
cycle level was considered as a
reasonable compromise between a
desire to obtain data relative to
long-term fatigue life and the ac-
quisition of data within the time
permitted.

Large-scale Tensile Tests

Two welded specimens of each
alloy were tested. Yield strength
and ultimate tensile strength were
determined in accordance with the
procedure of Reference 2. Tensile
strains were measured with an SR-4
clip-on gage. The yield strength _—Bien ak . setae
e
(0.2% offset in 2-in. gage length) bi Lay wR
was determined from the plotted Fig. 7—X-ray standard for bare aluminum-alloy electrodes

WELDING RESEARCH SUPPLEMENT 427-s

Table 4—Summary of Fatigue and Tensile-test Results
Fatigue strength
(stress ratio: —1)
at 1,000,000
cycles
(small-
Average tensile properties at 100,000 scale
Yield cycles Krouse-
Plate strength, Reduction of base (large- plate-
alloy (0.2% Tensile plate properties, % scale type
and Electrode Welding offset), strength, Elongation Yield Tensile fatigue fatigue
temper wire position psi psi in2in., % strength strength Elongation tests) tests)
6061-T6 4043 Flat 20 ,800 33,800 48 25 60 11,200 10,600
Vertical 18,800 29, 500 53 34 64 9,100 9,300
Plate 40,100 44,900 as = 18 , 300°
5154-H34 5154 Flat 19,700 38, 900 12 14,500 11,300
Vertical 18,800 39,200 41 11 37 11,100 11.200
Plate 32 ,000 44,000 a es 23,500
5086-H32 5356 Flat 23,600 44,900 14,200 12,800
Vertical 24,400 44,000 17 35 10,000
Plate 29,500 46 , 400 a 22,300
5456-H321 5556 Flat 27,900 50,700 ee
MNReeRID
me 40 14,400 13,500”
Vertical 27,300 46 ,600 18 58 11,300 10,000
Plate 33,300 55, 300 22 ,000
5454-H34 5554 Flat 20,100 40,000 44 10 15,000 11,100
Plate 36 ,000 46 ,500 Nm
eh
KF
WONMWWwWwWwOWDDoanon
NOWRWUNWOOWOWWSO 22,400
@ A better estimate for this value is 20,000 psi since failure occurred near shoulder where there is a moderate stress concentration.
> This is a minimum value since 7 out of 8 specimens failed at clamped end.

tion of sodium hydroxide. They in conformity with the radiographic straight line drawn through the
were then examined at magnifica- standards of Reference 3 (see Fig. data used for interpolation. Typi-
tions of 7 and 15 diam for under- 7) except for the deviations noted in cal failures for large- and small-scale
cutting and internal discontinuities Table 3. Since Fig. 7 does not fatigue specimens are shown in
such as cracks, excessive porosity contain a standard for vertical- Figs. 9 and 10, respectively.
and lack of fusion. position welds, specimens welded in A summary of the tensile-test
this position were judged by the results is given in Table 4. Figure
Hardness Survey 11 illustrates the type of fractures
overhead-position standard.
A Rockwell “‘F’”’ hardness survey encountered. Figure 12 is a graphi-
Fatigue strengths based on a life
was taken across the weld and heat- cal presentation of the fatigue and
affected zone of each of the macro- of 100,000 and 1,000,000 cycles
for the large- and small-scale speci- tensile-test results. The tensile
etched specimens. properties of the base-plate ma-
mens, respectively are given in
Table 4 along with tensile-test terials conformed to the require-
Results
ments of the applicable specifica-
Results of the visual and radio- results. The fatigue strengths were
tions.
graphic examinations indicated that established by interpolation as de- Results of the macroetch examina-
each specimen was satisfactory in scribed in the procedure. Figure 8 tion were satisfactory and _indi-
weld appearance and profile and was shows a typical plot of data and the cated freedom from internal dis-
continuities except in those weld
areas noted in Table 3 for some of
the 6061-T6 alloy assemblies which
contained excessive porosity. The
results of the Rockwell “F”’ hard-
ness survey are shown in Figs. 13
through 16.

Analysis
In considering the various results,
it should be noted that the data
referred to herein are based on the
specific thickness, joint design and
welding procedure described and on
properties of joints with weld re-
inforcement intact. Welding of
te a | these alloys with other processes,
DASH LINES [SHOW in heavier plate thickness, or con-
95% CONFIOENCE
Enve ore ditions of greater restraint, could
result in more deleterious effects
te) than those indicated herein.
CYCLES TO FAILURE 4
Fig. 8—Typical plot showing straight line used for determining fatigue strength The averages of the large-scale
of large-scale specimens at 100,000 cycles by interpolation and small-scale tensile-test results

428-s | OCTOBER 1960

Fig. 9—Typical failures for large-scale fatigue specimens. (Reduced by approximately one-half upon reproduction)

exhibited close correlation. Ac- figure, the yield strength (0.2%

cordingly, these averages were com- offset in 2-in. gage) across the
bined in obtaining the values shown welded joint had been exceeded be-
in Table 4 and Fig. 12 for average fore the occurrence of the failure
yield and tensile strength. through the bolt holes. Therefore,
The 6061-T6 alloy was the only although failure occurred away from
heat-treatable alloy included in the the weld, the welded area was
investigation. Tensile properties of stressed beyond the elastic range
this alloy in the annealed condition and should be considered as a
are usually specified as maxima. structural failure. The _ yield
Typical yield and tensile-strength strengths listed in Fig. 17 are similar
data for the annealed condition of to those noted in Table 4 for corre-
the 6061-T6 alloy have not been sponding welded joints, wherein fail-
included in Fig. 12 since it was felt ure occurred within the 2-in. gage
that they should not serve as a section across the weld.
basis for estimating the ultimate As indicated in Table 4 and Fig.
deleterious effects which could be 12, the fatigue strengths at 100,000
introduced by the heat associated cycles for the various aluminum-
with welding. In view of the magnesium large-scale specimens
above, minimum yield and tensile welded in the flat position ranged
strengths for welded assemblies of from 14,200 to 15,000 psi. The Fig. 10—Typical failure for small-scale
this alloy could not be estimated. results are considered approximate (Krouse-plate-type) fatigue-test speci-
Table 4 and Fig. 12 indicate the because of the scatter normally mens. (Reduced by one-half upon re-
relationships between yield found in the results of fatigue tests production)
strengths (0.2% offset in 2-in. gage and because of the relatively small
of the base plates and welded as- number of specimens tested at each
semblies. As indicated by stress level. For this reason, the dif-
Reference 4, as well as the hardness ferences noted in the approximate
data of Figs. 13 through 16, the fatigue strengths of the various
width of the area of maximum heat aluminum-magnesium alloys are not
effect is relatively small ('/, in. considered significant. The fatigue
from each side of the center of the strength at 100,000 cycles of ap-
weld). Consequently, the yield proximately 11,200 psi for the
strengths indicated would be de- 6061-T6 alloy flat welds is con-
creased if measured over a 1-in. sidered to be significantly lower
gage length. As indicated by the than the values given above for
literature, as well as Fig. 12, the aluminum-magnesium alloys.
yield strengths of welded joints in The corresponding fatigue
aluminum-magnesium alloys ap- strength of the 6061-T6 plate was
proach the yield strengths of the found to be approximately 18,300
annealed condition. Asa corollary, psi. A better estimate of the
it should be noted that the extent fatigue strength of this plate is
of the deleterious effects on the 20,000 psi since failure occurred at
half-hard temper alloys was greater the end of the reduced section near
than that on the one-fourth hard the fillet where there is a moderate
temper alloys. stress concentration. The base
Table 4 and Fig. 12 also indicate metal values for the aluminum-
that, while the effects of welding magnesium alloys ranged from 22,-
decreased tensile strength, the rela- 000 to 23,500 psi. These values are
tive degree of effect was not as great considerably greater than the corre-
as that observed in the case of the sponding properties for welded as-
yield strength. semblies.
Figure 17 illustrates a failure Table 4 and Fig. 12 indicate that Bond -Weld Type Fracture
through the bolt holes in a 5086- the fatigue strengths (large scale)
H32 alloy large-scale tensile speci- at 100,000 cycles for the 5154-H34,
men of the original design which was 5086-H32, 5456-H321 and 6061-T6 Fig. 11—Type of fractures in large- and
mentioned in the procedure. As alloys were reduced by 3400, 4200, small-scale tensile specimens. (Re-
indicated by the results noted in this 3100 and 2100 psi, respectively, duced by one-half upon reproduction)

WELDING RESEARCH SUPPLEMENT 429-s

 P- BASE PLATE MAXIMUM
56,00 F-WELDED FLAT AVERAGE
V- WELDED VERTICAL MINIMUM
TENSILE PROPERTIES
52,000 lend
DOTTED LINES INDICATE TYPICAL PROPERTIES OF BASE
PLATEIN ANNEALED CONDITION
48,000 Furveud PROPERTIES(STRESS CYCLECOMPLETELY

44,000

40,000

36,000

32,000
PS!
STRESS~—
28,000

24,000

20,000

16,000

12,000

8,000

4,000

.¢] PFV dé FV PF
PROPERTIES
PLATE e06i-T6
WiRE ——~ * 4043 5154 5356 5556 5554
% MINIMUM VALUES
Fig. 12—Bar graph of fatigue and tensile-test results

when the welding position was

changed from flat to vertical. The
PLATE-606!-T6 ELECTRODE -4043 reductions are attributable to a
WELDS _ combination of the following factors
i ; which are inherent in vertical weld-
| ing with any aluminum alloy:
a) The resulting deleterious
notch effects associated with
increased height and _ ir-
regularity of the reinforcing
bead.
(b) The relatively slower welding
speed which results in a higher
total heat input.
(c) The slightly greater extent
*F* +— |
WARONESS-RocxweLt
of distortion encountered in
vertical welding.
= fl
-
In view of the above, it is ap-
parent that a similar decrease can
be anticipated for vertical welds in
‘20 18 10 05 0 05 © 18 20 20 18 10 08 O OS 10 1S 20 the 5454-H34 alloy.
OF WELD - INCHES
DISTANCE FROM CENTER Fatigue results obtained on the
Fig. 13—Hardness surveys small - scale (Krouse - plate - type)

430-s | OCTOBER 1960

specimens, and shown in Table 4 PLATE-5I154-H34 ELECTRODE-5I54
and Fig. 12, show a greater range FLAT POSITION WELDS ____ VERTICAL POSITION WELDS _
+ ‘mais leis sales | T
than the results obtained with the
large-scale specimens. A value for
the 5456-H321 alloy welded in the
flat position has not been indicated
because failure of seven out of eight
of these specimens occurred in the 4, | }
ASSEMBLY SIZE- 36" 387 a ASSEMGLY SIZE 5x
plate at the clamped enlarged end. t+ ' |
¢ TYPICAL HA | oF ANNEALED
| ||PLATE
However, the fatigue strength of
these specimens at 1,000,000 cycles
is at least 13,500 psi, thereby making
the range for the small-scale speci-
mens still greater. It should be "F*
HARONESS
ROcKweLi
-
noted that the fatigue strength of 4 4 +
the small-scale specimen of this ASSEMBLY SIZE —1¢ "x9 | MBLY SIZE- 16"x 8
alloy is consistent with the high D EAT 4 ty |WwELO
tensile strength of the welded plate. YZ | = | WY)Wi
The clamp-end failures noted above 20 15 10 OS 0 OS 10 18 20 20 15 10 OS O OS |
were due to fretting corrosion which
DISTANCE FROM CENTER OF WELD—-INCHES
may occur when two surfaces of Fig. 14—Hardness surveys
any material are placed in contact
with each other and subjected to
the slightest relative movement,
even though the movement is micro-
scopic. When a similar specimen PLATE - 5086-H32 ELECTRODE -5356 PLATE ~ 5456~-H 32! ELECTRODE- 5556
FLAT
T 7 WELDS
7 POSITION : FLAT; POSITION WELDS
was tested in plastic-lined steel
jaws in lieu of bare steel, clamp-end
failure was not encountered.

Discussion &_ j/_ASBEMBLY SIZE - 36x 38",

Consideration of the over-all re- 3 a st a ae | = ame
sults indicates the importance of the %9
(— TYPICAL HARDNESS OF ANNEALED PLATE
weldability factor in aluminum de- «9
sign and fabrication. Design stand- aa
ards which are based on base-metal zaos
properties alone are inadequate. «<
r —— 4 =
The effects of welding on base-plate
properties will vary between alloys 60}++——_|_ ASBEMALY
SIZE -18'x19"| _+ WELD | ~| ASSEMBtysizé-18"x
19" |
(both heat-treated and __ strain- 4 [ HEAT AFFECTED
i! ZONE —~>}
hardened) and tempers. In general,
the error which will be introduced
by such practice could be greatest 20 15 ©
0 OS LO 15 20 0S 20 15 10 OS
for those materials whose mechani- DISTANCE FROM CENTER OF WELD - INCHES
cal properties have been enhanced Fig. 15—Hardness surveys
the most by heat treatment or strain
hardening.
It is apparent that proper design
and fabrication of aluminum weld- PLATE -5454-H34 ELECTRODE- 5554
ments of the type under considera- FLAT POSITION WELDS
tion must recognize that the weld | T T
and adjacent area could have com-
pletely different properties than
those of the base metal. One
approach is to consider the base
metal adjacent to the weld on the ASSEMBLY |SIZE T 36"x 36")
basis of the properties of the weld =e te 2s
and heat-affected zone; another | q TYRICAL HARDNESS DF ANNEALED PUATE |
would be to locate welded joints in 90} \ ——+- + —4+—__4 + —_—— |
noncritical areas. In the case of
small structures of heat-treatable 60 ae + = | TT Tt
alloys, the possibility of heat- =e"
treating the entire structure might -HARONESS
RockwEL"F*
ast Genllae (eal costes dell eal celles cilia | oS 1
be considered. + ——_—_—_}—__ —_—_+—_+ + + $+———+—__——_+—_4
The degree to which these princi- ASSEMBLY SIZ] ~18fx 19) }— WELD
ples have been taken into account is +4 HEAT - AFFECTED ZONE
indicated in Table 5. As indicated
therein, except for the 5086 alloy, 20 15 LO OS O OS 10 15 20
the maximum-allowable design-
stress values are less than '/; of the DISTANCE FROM CENTER OF WELD- INCHES
yield strength of the strain-hardened Fig. 16—Hardness surveys

WELDING RESEARCH SUPPLEMENT | 43l1-s

Table 5—Iindustrial Allowable Design Stresses for Welded Joints in Various Aluminum Alloys
Minimum plate tensile properties Maximum allowable design-stress values for welded joints (max
Plate Yield temp, +150° F)?
alloy strength Tensile _ ASCE Paper 970
and (0.2% offset), strength, Elongation ASME Boiler and API STD 12G vol. 82, no. ST. 3
temper psi psi in21n., % Pressure Code (Ref. 6)” (Ref. 7) (Ref. 8)
6061-T6 35,000 42,000 10 or9 Values given not applicable 6,950 8,000°: “
6061-0 22 ,000 (max) 18 to welded construction
5154-H34 29,000 39,000 10 7,350 (case no. 1174-2)
5154-0 11,000 18 eae
5086-H32 28 ,000 12 8,700 (case no. 1222)
5086-0 14,000 14 ace
5456-H321 33,000 12 10,400 (case no. 1248)
5456-0 19,000 16 we
5454-H34 29,000 10
5454-0 12,000 eS
Bw
EK
wwrt
SSRRSS
8333388 18
@ References 4 and 5 indicate that design stress values for welded joints should be based on the annealed plate properties and an appropriate factor
of safety should be applied.
> In the design of fusion-welded joints, the allowable stress values for the annealed condition shall be used.
Section J, Reference 8 states that ‘‘..., the minimum tensile strength in the heat-affected zone has been assumed to be 24 kips per sq in. for de-
sign purposes. The corresponding minimum yield strength for design purposes has been selected as 15 kips per sq in.”
¢ Section K-1, Reference 8 states that—'‘Fatigue tests indicate that welded members designed in accordance with the requirements of these speci
fications, and constructed so as to be free from severe re-entrant corners and other unusual stress raisers, will safely withstand at least 10,000 repeti
tions of maximum live load without fatigue failure regardiess of the ratio of minimum to maximum load.”

material and all are appreciably

below the yield strength of the
annealed material. It is through
such considerations of the effects
of welding on base-metal properties,
that the fabrication of serviceable
aluminum weldments can be as-
sured.
Conclusions
It is concluded that the minimum
yield strengths (0.2% offset in 2-in.
gage) of welded joints in the various
aluminum-magnesium alloys under
investigation will approach the
values for the corresponding an-
nealed base plate (see Fig. 12).
Reductions were also observed for Yield strength
the tensile strengths. The greatest Plate Welding” Specimen 0.2% Tensile® Location of
material position § ;, no. offset), psi strength, psi failure
reduction and closest approach to 23,150 43,600 Through bolt holes
5086-H32 Flat 1
annealed properties occurred in the (Electrode 5356) 2 23,700 43,400 Through bolt holes
case of the alloys with the greater @ Based on cross-sectional area of the gage section.
degree of strain hardening (AI
5154-H34 and Al 5454-H34). Fig. 17—Failure through bolt holes of large-scale
tensile specimen (not modified)
It is further concluded that the
fatigue strength at 100,000 cycles,
based on the large-scale specimen
is similar for welded joints in all vertical positions. The correspond- Material Laboratory, New York
of the aluminum-magnesium alloys ing value for the 6061-T6 welded al- Naval Shipyard, for their permission
investigated and is approximately loy is approximately 10,000 psi. for publication.
14,500 psi for welds in the flat
position with reinforcement intact. Acknowledgment References
The authors wish to express their 1. Cordiano, H. V., “A Unique Machine for
The corresponding value for the Large Scale Fatigue Testing,” ASTM Special
6061-T6 weld assemblies is approxi- appreciation to T. J. Griffin of the Tech. Pub. No. 216 (June 1957
mately 11,000 psi. This represents Bureau of Ships, Navy Department, 2. Federal Test Method Standard No. 151
July 17, 1956)
a significant reduction from corre- for his active interest and advice, 3. “Military Standard, X-Ray Standard for
sponding base-plate properties. The as well as to their colleagues in Bare Aluminum Alloy Electrode Welds,”
MIL-STD-437A (Dec. 9, 1958
change from flat to vertical-position the Material Laboratory, especially 4. Cook, L. A., Channon, S. L., and Hard,
welding also results in a decrease in K. J. Pon, welding engineer, and P. A. R., “Properties of Welds in Al-Mg-Mn Alloys
Abramov, mechanical engineer, 5083 and 5086,” THe WeLpinc JouRNAL, 35 (2),
fatigue strength. The extent of 112-127 (1955
these reductions is reflected by who assisted in many phases of the 5. “Welding Alcoa Aluminum,” Aluminum
Fig. 12. program and E. A. Imbembo, Co. of America, Pittsburgh 19, Pa. (1958)
6. ASME Boiler and Pressure Vessel Code and
The fatigue strengths at 1,000,000 metallurgist, and M. J. Berg, me- Case Interpretation Nos. 1174-2, 1222 and 1248
cycles for the small-scale (Krouse- chanical engineer, who were avail- Unfired Pressure Vessels).
“American Petroleum Institute Specifica-
plate-type) fatigue specimens fall in able for consultation. tion for Welded Aluminum-Alloy Storage Tanks
the range from 10,000 to 13,500 psi Appreciation is also extended to Tentative),”” API Std. 12G (March 1957).
the Bureau of Ships, Navy De- 8. ASCE Specifications for Structures of
for the various aluminum-magne- Aluminum Alloy 6061-T6, Proceedings, ASCE,
sium alloys welded in the flat and partment, and to the Director of the Paper 970, Vol. 82, No. ST 3 (May 1956).

432-s | OCTOBER 1960

All-Position Welding of HY-80 Steel

With the Gas-Shielded Process

Approval for all-position welding of HY-80 steel is currently being sought through

qualification in Navy tests including the explosion-bulge test

SY c. R. SIBLEY

Introduction been fabricated by this process for and does not fall out of a vertical
explosion testing. The results of or overhead joint. This type of
The present widespread use of HY- the explosion tests were also ac- transfer in carbon-dioxide shielding
80 steel in the atomic-submarine ceptable and gave added evidence of is known as “dip transfer.”” It is
construction program stems from the performance of the weld metal capable of depositing high-quality
two outstanding properties of this deposited by the gas-shielded proc- weld metal in all positions on mild
quenched and tempered steel. First ess. Military specification MIL- steel of various thicknesses. '
HY-80 steel exhibits moderate E-19822 (Ships) dated March 1, Pure carbon-dioxide shielding
strength combined with excellent 1957, applies to the bare-steel weld- cannot be used to weld HY-80 steel
toughness and ductility at all ex- ing wire type MIL-B88 which is with the MIL-B88 welding wire be-
pected service temperatures. Sec- used in the gas-shielded welding of cause the tensile and impact prop-
ond, this steel can be readily HY-80 steel. Currently, the gas- erties are reduced below those re-
welded with existing processes to shielded metal-arc process is being quired in MIL-E-19822 (Ships). A
form weldments which are tough used in shipyards where submarines little carbon dioxide in argon pro-
enough to withstand repeated bal- are constructed and repaired. In duces a desirable dip-transfer type
listic loading without failure. addition, some subcontractors are arc; therefore, it was necessary to
In the submarine program, HY-80 welding with this process at loca- determine if the amount of carbon
steel plates between *, and 3 in. tions remote from the shipyard. dioxide necessary to develop a good
thick are joined by groove welded The use of the gas-shielded proc- transfer for all position welding
butt and tee joints. The weld ess is currently restricted to the would also produce weld metal of
metal used to fabricate these joints flat position of welding because, adequate strength and toughness
must meet the requirements es- with an open arc in argon, the weld necessary to weld HY-80 steel.
tablished by the Navy. The estab- metal is too fluid to control easily
lished military specifications call for in joints not in the flat position. Test Method and Results
weld metal that can be deposited As the use of HY-80 steel grew, so
without defects and has tensile did the need for developing a Shielding Gas
properties similar to those of HY-80 method to weid in all positions with The two gases, argon and carbon
plate. In addition, the weld metal the gas-shielded process. This dioxide, were mixed in a ““Y”’ con-
must exhibit excellent impact paper is concerned with the de- nection upstream of the welding
strength. The minimum value re- velopment of a method for all- gun. Proportioning each gas from
quired is 20 ft-lb at —60° F ina position welding of HY-80 steel a cylinder into the ““Y”’ connection
Charpy V-notch specimen. The with MIL-B88 welding wire. permitted a full range of mixtures
weld-metal properties are taken in from pure argon to pure carbon
the as-welded condition since this History dioxide. A total flow rate of 25
is the status of HY-80 weldments in In the welding of mild steel in cfh was used with each shielding
service. the vertical and overhead positions gas composition.
The gas-shielded metal-arc weld- with the gas-shielded process, it is Initially, arcs were run in argon-
ing process was one of the first necessary to maintain a short arc rich mixtures to determine the
processes to deposit weld metal that length. With argon shielding, a amount of carbon dioxide required
qualified under the requirements short arc is not stable, causing to give an arc that is stable and
for HY-80 established by the Navy. spatter and porosity to develop. easily handled in any welding posi-
This welding process is the only With 100% carbon-dioxide shield- tion by the operator. As the per-
semiautomatic process approved at ing, a short arc can be maintained centage of carbon dioxide in argon
this time. In addition to mechani- without excessive spatter. The increased to 20%, an arc similar to
cal testing of weld metal, panels have weld metal in this instance is still the normal dip-transfer arc was
too fluid for out-of-position welding. obtained. Further additions of car-
If a proper power supply is used bon dioxide caused little change in
C. R. SIBLEY is associated with Air Reduction with carbon-dioxide shielding, how-
Co., Union, N. J the arc action or spatter level found
Paper presented at AWS 4ist Annual Meeting ever, an arc is established in which with the 20% addition.
held in Los Angeles, Calif., April 25-29, 1960 metal is not overheated by the arc Vertical-up deposits were made

WELDING RESEARCH SUPPLEMENT | 433-s

with 0.035-in. diam MIL-B88 weld- plus 80% argon. Charpy V-notch ing, are summarized in Table 1
ing wire in shields of pure car- specimens were machined from each From the results of the impact
bon dioxide, equal proportions of deposit to check the values at —60‘ tests, it was concluded that the 20%
carbon dioxide and argon and a F. The results of the impact carbon-dioxide mixture should be
mixture of 20% carbon dioxide testing, and the conditions of weld- evaluated further to establish its
use for all-position HY-80 welding.
To determine the tensile properties
of weld metal deposited by this 20%
Table 1—Properties of 0.035-in. MIL-B88 Vertical-up Weld Deposits carbon-dioxide mixture, all-weld-
Impact Properties for Three Gases— metal vertical-up deposits were
Impact strength, ft-lb Charpy V-No'ch made and 0.505 tensile bars were
Testing 50% CO., 20% CO., machined from the weld metal.
temperature ° F Pure CO, 50% argon 80% argon Typical tensile-test results of the
— 60 12, 16, 13 25, 23, 18 29, 29, 27 weld metal are included in Table 1.
— 80 No samples No samples 23, 23, 22 These tensile and Charpy-impact
taken taken results of vertical deposits meet
—100 No samples No samples 20, 17 the established requirements of
taken taken MIL-B88 weld metal deposited in
Typical Tensile Properties— the flat position with argon-oxygen
Variable 20°; CO, 80°; argon shielding gas shielding. Based on this qualifying
Bar 1 Bar 2 performance, the shielding gas mix-
ture of 80% argon plus 20% carbon
Tensile yield strength, psi 98,750 98 ,500 dioxide was established for all-
Ultimate tensile strength, psi 110,000 108 ,650
Elongation in 2 in., % 19 20 position HY-80 welding. Trial de-
Reduction of area, % 57.7 58.5 posits were made in the over-
Fracture type Cup-cone Cup-cone head and horizontal positions with
Hardness, Rc 21 21 the 0.035-in. welding wire for
arc performance and_ technique
Welding Conditions— studies. No testing of these de-
Argon + 20% CO, impact deposit Argon + 20% CO, tensile deposit posits was made except X-ray in-
spection which revealed no faults.
Explosion-bulge Plates’
Arrangements were made through
the Bureau of Ships to test ex-
plosion-bulge plates at the Naval
Research Laboratories which were
welded in the vertical-up position
by thisnew method. Two designs of
explosion panels were fabricated, a
Arc groove butt joint in 1l-in. HY-80
amp plate, and an assembly of two fab-
No. of passes
150 Arc voltage ricated tee sections attached to
170 Arc current l-in. HY-80 plate by groove and
170 Welding speed.........variable, 2-6 ipm fillet welds. Photographs of these
165
Avg. welding speed, 2 ipm, all passes

Joint design—Web
Double bevel
50 deg included angie
'/s-in. root opening
sin. root land
Welding conditions
Root passes Capping passes
land 2 3and4
Joint design Welding conditions Amperes 160-180 140-150
Root passes Capping passes Volts 20 19.5
land 2 3and4 Wire feed, ipm 396 330
Double bevel Amperes 150-160 Travel speed, 3 4
60-deg included angie Voits R 19.5 ipm
i/--in. root opening Wire feed, ipm 330 Fig. 2—Fabricated tee section for
'/ie-in. root land Welding speed, ipm 4
explosion-bulge testing 1- and °/;-in.
Fig. 1—Fabricated butt weld for explosion-bulge testing 1-in. HY-80 plate HY-80 plate

434-s | OCTOBER 1960

 two types of explosion panels which
measured 24 x 24 in. are shown in Table 2—Chemical Analyses of HY-80 Plate and MIL-B88 Welding Wire
Figs. 1 and 2. Three welded butt
joints and two welded tee joints Analysis by weight percent
were included in this series. Each 0.035-in.
Specified HY-80 1-in. plate 5/,-in. plate MIL-B88 weld-
joint was welded in four passes em- plate comp., up to US steel heat unknown ing wire ht.
ploying a weave technique. Aver- 1'/, in. No. 69L099 origin X10161
age values of current, voltage and Element
travel speed are included in the
Carbon .22 max. 0.14 .12 05
figures. The chemical analysis of Manganese .10/0.40 0.25 .23 42
the plates, welding wire and de- Phosphorus .040 max. 0.012 009
posited weld metal for this series of Sulfur .045 max. 0.020 015
explosion panels are listed in Table Silicon .12/0.30 0.25 52
2. The */;-in. HY-80 plate formed Nickel .00/3.25 2.15 43
the web of the tee connection be- Chromium .85/1.90 1.35 08
tween the two 1-in. plates. Vanadium ae. a cae 15
Explosion testing was done at Molybdenum 0.15/0.65 0.27 0. 46
©oooroeo°cor
0° F, with borderline performance of Chemical analyses of 0.035-in. MIL-B88 weld metal from ht. X10161
the butt plates reported by the
Analysis by weight percent
Navy. The fabricated tee sections Pure CO, Argon + 20%
performed as well as those made in Element shielding CO, shielding
the flat position; all panels failed Managanese 0.90 0.99
at the toe of the fillet weld. An Silicon 0.36 0.40
investigation was initiated at this Vanadium 0.12 0.15
point to determine the cause for Molybdenum 0.35 0.37
the failures which occurred in the
butt welds. Since the tensile and
impact results were satisfactory,
some other factor must have caused
these bulge plate failures. The
most probable defect to occur in
the explosion panels was thought to
be lack of fusion at the locations of
starts within the joint or between Table 3—Properties of 0.062-in. MIL-B88 Vertical-up Weld Deposits
the weld deposit and the plate.
Such defects were not apparent from (Shielding Gas: Argon + 20% CO.)
X-ray examination; so, it was Typical impact properties
decided to side-bend test some four- Testing Impact
pass vertical-up welded butt joints temperature, strength
in HY-80 plate. These welds were a ft-Ib
made with the same procedure used —60 38, 36, 2
on the explosion-bulge plates. Stops —80 28, 22, 2
and starts were intentionally made Typical tensile properties
so that the position of stops and
starts could be noted and side Variable Bar 1 Bar 2
bend samples could be cut at these Tensile yield strength, psi 108 , 750 105 ,600
points. The side bend test proved Ultimate tensile strength, psi 112,000 111,800
adequate in revealing fusion defects Elongation in 2 in., % 18 19
that were not easily detected in Reduction of area, % 50 57
normal X-ray inspection to 2% Fracture type Cup-cone Cup-cone
Hardness, Rc 24 24
sensitivity. To bend 180 deg in a
side-bend sample without failure or Welding conditions
flaw requires complete interbead Impact deposit Tensile deposit
and plate fusion. Figure 3 illus-
trates the side-bend performance in 45°—
joint areas which contain weld
starts and those containing no > |
starts. Since several weld starts
were made in the explosion bulge
plates, it is reasonable to assume
that lack of fusion existed at these
points. In further side-bend test-
ing, it was found that the first pass
on the backside of a double-vee
butt joint also could be poorly
fused in areas not attending weld
No. of passes
starts. These side-bend test re- 170 Arc voltage
sults clearly indicated the cause of 180 i CE ccna kadaxcctsboaax eee 175 amp
borderline explosion performance r 170 Welding speed... variable, 2-6 ipm
and the need for a method of weld- 4 180
ing which would not cause lack of Avg. welding speed, 2 ipm, all passes
fusion.

WELDING RESEARCH SUPPLEMENT | 435-s

 Welding-wire Size for 0.035, 0.045 and 0.062 MIL-B88 to those shown earlier for the 0.035-
Up to this point, 0.035-in. diam welding wire at 185 amp are 6.3, in. wire. Side-bend data were
MIL-B88 welding wire had been 6.0 and 5.0 lb per hour, respectively. obtained as the next step in com-
used exclusively for the out-of- The same shield-gas mixture of pleting the evaluation of the 0.062-
position welding on HY-80 steel. 80% argon plus 20% carbon dioxide in. welding wire. Side bend samples
Due to the small arc that forms was used with the larger wires to of double-vee vertical-up weld de-
with this size wire, the utmost test their operation against the posits made in four and six
diligence must be exercised by the 0.035-in. wire. Since the operating passes were completely satisfactory
operator to ensure complete cover- characteristics of these larger wires in performance. These different
age of the joint area by the arc. were not widely different from those pass techniques were evaluated
To make the arc area larger, and of the smaller wire, deposits were to ensure a range of energy input
thus make a joint easier to weld, made to obtain mechanical test data for this type of welding. A photo-
welding wires of 0.045 and 0.062 on the weld metal. The tensile and graph of the six-pass lower energy
. in. diam were evaluated. The melt- impact test results of the weld input side-bend samples made with
ing rate of these larger wires was metal from 0.062-in. MIL-B88 weld- the 0.062-in. wire is shown in
only slightly below that of the ing wire are shown in Table 3. Fig. 4. No failures have been
0.035 wire since the same current The typical operating conditions found in any of the side-bend
was used. Values of melting rate and resultant properties are similar specimens tested from welds made
by two welding operators. This
absence of fusion defects should
produce an improvement in the
explosion-bulge plate performance.
To make the capping weld passes
somewhat easier to deposit, a 0.045-
in. diam welding wire may be used
in place of the 0.062-in. wire for
finishing a weld. No difference in
surface appearance is caused by this
capping technique and its use may
not be desired by some operators.
Regardless of how the weld is
finished, the 0.062-in. wire should
be used in the root passes to ensure
proper fusion. Bend-test results
are not altered by this smaller
wire capping technique, as shown
in Fig. 5 where both four- and six-
pass deposits are illustrated. Drop-
weight* test data were obtained for
vertical-up butt welds made with
the 0.062-in. welding wire in four
and six passes. A four-pass de-
posit was made entirely with 0.062-
in. welding wire and one was made
with 0.062-in. wire in the root and
0.045-in. wire for capping. Both
these deposits were high-energy-in-
Samples without starts. Satisfactory. Samples with starts. Failed—Note lack of
Note extreme ductility in root fusion in root of bottom sample and
in upper portion of top sample.
Welding conditions: 1l-in. HY-80 plate, double-vee butt joint, 60-deg included angle, '/s-in. root
opening, 4 pass weld. Pass 1, 160/170a, 20.5 v, 445ipm W.F. Pass 2, 170/180a, 21 v,
492ipm W.F. Pass 3, 140/150a, 19.5 v, 339 ipm W.F. Pass 4, 150a, 20 v, 357 ipm W.F. Table 4—Drop-weight Test Data
Fig. 3—Side-bend samples; vertical-up double-vee butt joint; Vertical-up weld deposits; MIL-B88 weld-
0.035-in. MIL-B88 wire; shield gas, argon + 20% CO, ing wire; shielding gas—argon + 20°;
CO.; double-vee butt joint in 1-in. plate
Testing variables
Notch location Weld metal
Six-pass deposit: Specimen size - 2 Pm
Arc current 180 amp 125/, in.
Arc voltage 19 v Hammer height 13 ft
Avg weld speed 4ipm Cooling medium CO.—liquid
(all passes) nitrogen
Avg energy input 52,000
joules/in. Nil-ductility temp:
4-pass 0.062-in. wire
100,000 joules/in. —80° F
Fig. 4—Side-bend samples; 4-pass 0.062- and 0.045-
vertical-up double-vee butt in. wire, 80,000-
joint; 0.062-in. MIL-B88 wire; 100,000 joules/in. —60° F
shield gas; argon + 20% 6-pass 0.062- and 0.045-
CO, in. wire, 50,000-
80,000 joules/in. —60° F

436-s | OCTOBER 1960

 hard-surfacing deposit was placed erator-qualification test has been
on the weld at the center of its prepared which includes all-weld-
length and a sharp notch ground metal tests for tensile and impact
into the hard deposit. This is properties plus joint tests for side-
called a crack-starter plate and bend properties. Each require-
its purpose is a screening test to ment of the qualification must be
determine if the presence of the met before the operator is put on
crack will cause brittle failure in the production.
weld metal, heat-affected zone, fu- The limited amount of work done
sion zone or plate. In this test, on developing techniques for over-
the four-pass vertical-up deposits head and horizontal welding used
were satisfactory; that is, the crack 0.035-in. welding wire, and tests
did not run in any direction. revealed lack of fusion in most de-
Left The performance of the regular posits. It is expected that the
4-pass: Passes 1 and 2, 0.062-in. wire. Passes bulge plates did not meet the pre- 0.062-in. wire will improve the
3 and 4, 0.045-in. wire. 175 amp; 20 v:
2 ipm (av of all passes); 104,000 joules/in. scribed 10% plate thickness re- fusion in these positions as it did
Right duction at the crown of the bulge. in the vertical. Results of this
6-pass: Passes 1-4 0.062-in. wire. Passes § Failure in the heat-affected zone type of welding are not available
and 6, 0.045-in. wire. 175 amp; 2lv: passes occurred on the second round of presently but the same operating
1-4, 2.8 ipm, 79,000 joules/in.; passes 5-6,
3.lipm; 72,000 joules/in. firing, but three or four rounds are conditions of current and voltage
required to obtain a value of 10% as used in vertical-up welding will
Fig. 5—Side-bend samples; vertical-up
double-vee butt joint; MIL-B88 wire; reduction in plate thickness. This be employed. These tests of over-
shield gas, argon + 20% CO failure in the heat-affected zone is head and horizontal welding are
blamed on the adverse effect upon expected to be done in 1!/,-, 1'/2- and
this zone of the high-energy input. 2-in. thick plate.
put welds. The nil-ductility-tran-
sition (NDT) temperature was — 80 Trial deposits of six-pass welds
made with the 0.062-in. MIL-B88 Summary
F for the 0.062-in. wire deposit, and
welding wire were much lower in Even though it has been shown
-60° F for the deposit capped
with 0.045-in. wire. An additional energy input, as shown in the side- that good-quality weld metal can
bend samples of Fig. 4. More be deposited in the vertical position
drop-weight test was conducted on a
explosion-bulge plates will be made with this new welding technique, the
six-pass deposit in which the first
employing lower-energy input tech- failures in the plate heat-affected
four passes were made with 0.062-
niques which should alter the struc- zone during ballistic testing have
in. wire and capping was done
with 0.045-in. wire. The nil-duc- ture of the heat-affected zone in the withheld process approval for weld-
HY-80 plate making it tough instead ing HY-80 steel. By modifying the
tility-transition temperature for this
of brittle. welding technique to lower the
weld metal was 60° F. The
data of these tests are recorded in energy input it is expected that
Table 4. The higher NDT tem- Discussion satisfactory ballistic testing will be
perature values for the welds capped There is no approval, at this writ- obtained and thus permit the use
with 0.045-in. wire in both the ing, for all-position welding of of this welding method on HY-80
four- and six-pass sequence left the HY-80 steel with the dip-transfer steel fabrication.
four-pass deposit made with 0.062- gas-shielded process. Approval is The dip-transfer welding tech-
in. wire as the first choice for explo- limited to flat-position butt and nique, using a shield-gas composi-
sion-bulge plate testing. tee joints and argon plus oxygen tion of argon plus 20% carbon
A series of l-in. HY-80 bulge shielding. Expansion of this ap- dioxide, can be used for quality
plates was welded vertically-up proval to cover all-position welding welding of other types of steel in
using the 0.062-in. MIL-B88 weld- is currently being sought through all positions and thicknesses
ing wire and a four-pass technique. qualification in Navy tests includ-
Radiographic inspection of the com- ing the explosion-bulge test. Sat- References
pleted welds prior to firing was isfactory performance in all areas of 1. Tuthill, R. W., “Dip-Transfer Carbon Di
oxide Welding,”’ WELDING JOURNAL, 37 (10), p
satisfactory. Two types of explo- testing is required for approval. 976 (1959
sion-bulge plates were prepared for The welding operators who are to 2. Hartbower, ( E., and Pellini, W. S
“Explosion Bulge Tests of the Deformation of
testing at 0° F. One was a regular use this process for all-position weld- Weldments,” Jhid., 30 (6 Research Suppl.,
bulge test in which the welded plate ing of HY-80 steel must qualify in 307-s (1951
is forced to bend into a bulge at the laboratory sample welding before 3. Pellini, W. S., “Notch Ductility of Weld
Metal,” Jbid., 33 Research Suppl., 217-s7
plate center. In the other test, a starting production welds. An op- 1956

WELDING RESEARCH SUPPLEMENT 437-s

Weld Strength and Dimensional Stability

of Cold-Worked Stainless Steel

Program is aimed at obtaining data on the

properties of butt-welded joints in 301 stainless steel

in various tempers and gages

BY L. STEMANN AND E. E. WEISMANTEL

SUMMARY. As strength and weight been avoided for welded structures. or contribute undesirable weight.
requirements of welded structures be- It is generally considered that the Thus, it is important to know the
came increasingly critical, the use of heat-affected zone adjacent to the actual yielding and failure character-
annealed-strength values for welded weld, being basically an annealed istics of fusion welds.
joints in cold-rolled stainless steel was band, is a limiting boundary on the The annealing effect of welding
questioned. Therefore, a program was
conceived to obtain data on the proper- strength attainable from the struc- on cold-worked materials is gener-
ties of butt-welded joints in type 301 ture, if it were to be fabricated ally understood. However, actual
stainless steel in various tempers and utilizing fusion welding. data on the extent of the annealed
gages. Tensile tests on fusion butt Proper weld-joint reinforcement zone and on the effect on mechanical
welds showed that the yield strength or preferred orientation of the properties of fabricated parts are not
(0.2% total deformation) and ultimate joint with respect to the major- generally available. Because data
strength were considerably higher than stress direction can, in many cases, had been obtained which indicated
the annealed properties generally used eliminate the problems associated that weldments of cold-worked
Tests were included in the program with the reduced base-metal stainless steels could withstand
to evaluate the changes in dimensions
and mechanical properties which took strength adjacent to fusion welds. much higher stress levels than if the
place in base metal of various tempers Typical techniques of this type are annealed properties were used in
upon exposure to elevated tempera- shown in Fig. 1. In this manner, design, the following testing pro-
tures. The data obtained show that joint strength can be boosted to gram was developed. This testing
significant dimensional changes do values approaching that of the program considered the effect of
occur in the '/,-hard and full-hard primary base-metal structure. It fusion welding on the properties of
tempers upon exposure to temperatures is obvious, however, that these thin-gage stainless steel in various
above 800-900° F. A marked de- design tricks may become unwieldy tempers ranging from ',,-hard to
crease in room-temperature mechanical full-hard conditions. Work was
strength of full-hard 301 stainless steel performed on various gages of
occurs after exposure above 1050° F.
materials generally ranging from
Introduction LOAD 0.020 to 0.125 in. thick. The
The properties of cold-worked aus- FUSION WELD OBLIQUE TO properties of the weldment were
tenitic stainless steel are attractive DIRECTION OF LOAD evaluated by measuring the 0.2%
offset yield strength for a 2-in.
for many potential uses in highly
stressed sheet-metal structures. gage length across the weld line
The exceptionally high strength and also by determining the ulti-
attainable, the good ductility as- mate strength and the total elonga-
sociated with these high-strength tion for the given gage length. It
levels, the excellent corrosion resist- ++ 4 must be recognized that the meas-
——— a ured 0.2% offset yield strength is
ance, the lack of notch sensitivity
and good toughness at low and FUSION WELD REINFORCEMENT BY not the true yield strength of the
INCREASED BASE METAL THICKNESS
high temperatures all make these annealed zone but rather a measure
of total deformation as the remaining
materials attractive to the design
engineer. However, because the structure would see it.
yield strength of annealed austenitic When 0.2% deformation has oc-
stainless steels is very low, much of curred across a butt weld in a
the potential of these alloys has total gage length which includes
hard base metal, reinforced weld
FUSION WELD REINFORCEMENT BY nugget and annealed heat-affected
L. STEMANN and E. E. WEISMANTEL are SPOT WELDED DOUBLERS zone, the actual deformation has
associated with the Space Atomics Div. of the obviously taken place principally
Budd Co., Philadelphia 32, Pa. Fig. 1—Weld-joint design techniques to
produce welds approaching base-metal in the annealed zone. Such yield-
Paper presented at AWS National Fall Meeting
held in Pittsburgh, Pa., Sept. 26-29, 1960 strength in cold-worked materials ing need not be considered an

438-s | OCTOBER 1960

 008‘
000°

00S ‘€/
000‘

‘ST

000 °68

02T
€ZI 008
00T

96
000

002°

1 1
006

000
TOT
006°

121
008

201
°S6

002T
405
4[jy
000 ¢ 000'8I
006
89I

? 1

SUPPLEMENT
006 OZT 009° 91
000 é2I 002
SSI
002

891 000

.0S0T
000 vet 000

40}
000802

IST 000

44

4
[

7 1
006 ‘202 602
221 000°
006
L0¢

061 008
14 T 40) 4 002
£02 008 221 000°
‘uBueijs 000621

000
PIPIA yesjjo 000

681
9ST

000°
00S 102 00S
O€T 000

ISI
00S
Z2I

‘yyBuess
peaiscses-sy

yBueiys

eae
}eES}jo }asSjjo einjyessdwe}

isd
Yisd
%

*“ulzZ

alisue}
eayewiin

RESEARCH
%2°0

eunsodx
“ulfo9

“urz ut %2°0 ‘isd

uolyuones

*“ul2 ul %2*0 ‘isa 7 ul ‘u}Buels %2'0 ‘isd

‘isd

r
isd isd % }OSj}jO

ul
yu
‘uyBuesjs
*‘uyBueijs
ayIsue} uones ‘yu}Bueiys eyisue} uones ‘yyBueis
PIPIA

ayewnin uojy PISIA

0z0°0—-
uoly

—psey

uolz PI®IA
ejisue} uones
———_——_———-pey */:—0v0'0 —_ - psey?
jiny

eyewiyn
———————-psey tiny
9€¢0°:0
aanjesedwial Wiooy ye [9893S SSOjUIE}S TOE @dAY Jo Sansedosg jed}UeYIOW OY} UO eANjesodwis| sinsodxy jo S}99jj3—7Z 9/91

WELDING
009°09 00S'TS

00 000‘ 8: oc 008
008

O0L
‘8b

006 '6/ 008 000

9

000
000

000°/8
008009'S9

000 'vET

00S
€8
000°09

00S" 008‘ 98 009 bri

000° 00068 006 ‘EbI

008
98
°29

006 101
008 18 000°

000'00T 000°68 000

000
8tT 009

801
008

98
18

000°006 008‘
000 82 000
000

POT
006
EE€T
000
88
S8
006‘ PET
006 89 000°06 006 veT
000 £8 000° 66 008° 22

16 ‘£2
00S ‘08 000° 26 002
°92T 000
96
000° OT
000‘ 000°

2 000

06 000

00081
26
6 000°

‘trl
2
000 +6

tl 002
22 Zi 008

009
nnsc*

00006
000°201

006
VUS

008 68

000'
000°

000
000'88
000 '99 008 000 69 0008 006 ‘88
'6EI 000
009°29 000 000°82 00088 00‘ v6
bbl
002‘ tbl 000 81 00S
000'ErI 2vI 000 v8T 000‘
000°€9

O2T 00S
9IT
000° 000°€S 00S 64 000°

00S
SIT
000

021
tt 008
br
000°

E21
009° 008 dt 62
000
t9 009‘

004
b2T000221

00S
oT

000°19

92000°00008
000
82

221000°
008 €8
O'e!

006°19
000°SS

00018 000°S8 €Z Ov
000

000
29
tS

Buejs 000°
“ulzZ o'er

00618 000°28 6t
£2 000 v8 920°0 H*/1
I2I 000

‘u}yBuels
yyBuels
b2I 000‘

jejyew-eseg—
j|euIPN}PBuo})
yesyjo % y@SjjO oM % yesjjo ul sedwes

isd
isd
%

isd
eyIsue}

eyisue}
ayewnin
uones

‘sseu “ulz ul 20 ‘sd ‘sseuyoiut

eyewnin
%2'0 ‘isd

sensedoid
ulz ui
%

ulz ul %2'0 ‘isd

(U0}Ie4IP
isd

jedX!ueYyIopW—T
yES}jO

‘uy
uoly ut

uones ‘yuBuses uoines *‘yyBuels ajisue} pueH uoines ‘yyBuess

‘yuyBuesds ‘isd %2°0

papjem-yjNg
uojy PI@IA uoja PI®IA ayewnin uoly PISIA
ajisue} ‘ylBuesys

Seljiedoig
SSOjUIeIS
ayewnin PIPIA
—(juIOl Pjam 0} @S4BASURI}) Seledold Pjem-VNgG - —— —_—— —— ——

9/Ge]
1923S
PGA]
TOE
jo
 438-s | OCTOBER 1960

1000
IN_
PSI

mass a | a om | | wee oe SS om a
100+ 2"GAGE
LENGTH
PS!
1000
IN 040 060 080 .100_ ./20
ULTIMATE
TENSILE
STRENGTH 0 .020 .040 .060 -080 100 ~=«.120 OFFSET
0.2%
STRENGTH
YIELD .020
THICKNESS IN INCHES
THICKNESS IN INCHES
. 2—Room-temperature yield and ultimate strength of butt welds in various thicknesses of Type 301 stainless steel

undesirable condition since satis- tion, some residual stresses are

factory additional ductility re- present which have little effect on
mains in the over-all weld joint. the basic metal strength but which
aa
For the particular structural ap- affect the dimensional stability of the
° plication under consideration, serv- base metal after exposure for a
°°
ice conditions included large varia- short time at elevated temperatures.
z RS a tions in stress level and exposure Additional tests were run on this
==x 200 400 600 800
TEMPERATURE °F temperature. Thus, in addition to program to evaluate the magnitude
zw2° the need for weld-joint information, of dimensional changes that would
4 063 data were required to show the occur for both the transverse and
a° + HARD effects of exposure at elevated longitudinal dimensions with consid-
a
.N° temperatures. The actual limiting eration of the thickness and the
= temperatures for the use of cold- temper conditions of the material
worked Type 301 stainless steel used.
==
°z were not known. It was important
ot See oe a eee to determine the temperature at
«= 200 400 600 Test Program
” TEMPERATURE °F which relaxation occurred, the rela-
[o)a 020 tive amount of dimensional change Since this program was basically
w that occurred and the effect of an evaluation of butt-welded
> FULL HARD
— exposure temperature on the prop- strength of thin-sheet materials,
”_—s erties of the material (at both all butt welds were made without
°Cs exposure temperature and room edge preparation. All welds tested
temperature). Tests were run to were produced by the gas tungsten-
Nad evaluate what this effect might be arc method. Amperage, voltage
°o
with exposure to elevated tempera- and filler-metal addition were ad-
tures. justed to give welds with complete
1 ! | A cold-worked material such as penetration and with uniform weld
200 400 600 800
TEMPERATURE°F Type 301 stainless steel has resid- build-up of 20°% base-metal thick-
ual deformation stresses. These ness. X-ray inspection was used to
Fig. 3—Elevated-temperature yield stresses are directly related to the determine weld quality. Areas
strength of Type 301 butt-welded strain hardening required to obtain which contained porosity and other
stainless steel specific strength levels. In addi- defects, even though they were

ULTIMATE

|
—=H
2 DESIGN MIN |
-——BUTT WELDS IN >H
2 3H DESIGN MIN.
—tH DESIGN MIN. -BUTT WELDS IN 3H

1000
PSI
ANN. DESIGN MIN. BUTT WELDS IN 7H—~
— ANN. DESIGN MIN.
ZL
ULTIMATE
TENSILE
STRENGTH C= 2 a 4 i
400 600 800 1000 STRENGTH
YIELD
0.2
AT
%OFFSET ~ 200 400 600 800 1000
TEST TEMPERATURE °F. TEST TEMPERATURE °F
Fig. 4—Comparison of mechanical properties of butt welds to base-metal design minimums in Type 301 stainless steel

440-s | OCTOBER 1960

 LONGITUDINAL DIRECTION TRANSVERSE DIRECTION
ANN.

6CHANGE
INLENGTH
ININ. —.008 -—
6IN.
CHANGE
IN
LENGTH |
ROOM — EE SE —— |
Temp 700 90 1000 1100 120 ROOM 700 800 900 1000. 1100
EXPOSURE TEMPERATURE °F TEMP EXPOSURE TEMPERATURE °F.
Fig. 5—Dimensional changes in Type 301 stainless steel as a result of exposure to elevated temperatures

not large enough to be cause for are base-metal properties for com- perature data with design minimum
rejection in production welds, were parison. In all cases, both yield curves for Type 301 stainless steel
not used for testing. Standard 2- and ultimate strength has been is shown in Fig. 4. It is of primary
in. gage-length sheet-metal speci- decreased. The measured strength, importance to note that, while
mens were machined from the test however, remains considerably butt-weld strength is lower than the
panels. Tensile testing was at higher than that for annealed mate- respective base-metal design prop-
room temperature, 400 and 700° F. rial. It should be noted that, in erties, a considerable advantage
Tensile tests met the requirements these tests, the yield points have was maintained over annealed prop-
of ASTM E8-46. been determined for a 2-in. gage as erties at all test temperatures.
To evaluate the effect of stress- mentioned. As gage length de- The dimensional changes that
relieving temperatures on the di- creases, the measured offset yield occur in Type 301 stainless upon
mensional stability of cold-worked strength and elongation of butt exposure to elevated temperatures
materials, tests were performed on welds will approach that of the are plotted for longitudinal and
unwelded base-metal strips each of annealed material. transverse directions in Fig. 5.
which was exposed to elevated The effects of metal thickness on These data show that the dimen-
temperatures for 1 hr. Test speci- mechanical strength of butt welds sional change associated with the
mens were machined very accurately are plotted in Fig. 2. On '/,-hard cold-worked austenitic stainless steel
and measured to the nearest one and full-hard material, the 0.2% is generally a shrinkage or contrac-
ten-thousandth of an inch in length. offset yield is lower for heavier tion, the amount of which depends
These specimens, approximately 6 gages__ tested. The ultimate upon the degree of cold work,
in. long, were placed on a flat plate strength, however, does not change temperature of operation and ori-
in a furnace preheated to the desired appreciably with gage. The data entation of the sheet with respect
temperature and held 1 hr at the on '/.-hard materials are more to the rolling direction. In general,
temperatures of 700, 1050 and scattered and no general trends are it can be said that the amount of
1200° F. After heating, the speci- apparent. No appreciable differ- contraction is less in the transverse
mens were remeasured. Standard ences were measured on any gages direction than in the longitudinal
tensile specimens were machined of this temper when tested at direction if temperatures and tem-
from these pieces and tensile testing similar temperatures. pers are the same.
performed to determine the changes The yield strength of the butt Within the accuracy of measure-
in mechanical properties which oc- welds as a function of the test ments for this program, the an-
curred. temperature is illustrated in Fig. nealed and -hard material did
3. It can be seen that the effects of not undergo significant changes in
Results gage thickness, noted above for dimensions below 1050° F. Expo-
All data obtained on butt welds ; and full-hard material, were sure of these tempers at 1200° F
in cold-worked 301 stainless steel maintained at all test temperatures. did cause very slight longitudinal
are shown in Table 1. Also included Comparison of these elevated-tem- growth and transverse shrinkage.

— FH ULTIMATE
7-FH YIELD
~
$4 ULTIMATE
P,/$H YIELD

| ULTIMATE
_ 44
== | H YIELD
—.008 PS!
1000
STRENGTH
YIELD
-
CHANGE
6LENGTH
IN
it | | 4| | 0.2%
&ULTIMATE
OFFSET
TENSILE
ROOM
TEMP 700 800 900 1000 100 1200 900 1000 1100 1200
EXPOSURE TEMPERATURE °F TEMP EXPOSURE TEMPERATURE °F
Fig. 6—Effect of exposure temperature on dimensional changes and room-temperature mechanical properties of 301 stainless steel

WELDING RESEARCH SUPPLEMENT 44l-s

 The '/.-hard and, particularly, change in length is not directly must be examined carefully.
the full-hard tempers of the Type related to a reduction in the The foregoing test program has
301 stainless steel showed more strength developed by cold work. indicated that fusion-welded cold-
pronounced shrinkage both longitu- The '/.-hard and full-hard material worked stainless steel can be used
dinally and transversely after expo- apparently aged at 700° F, showing and design can be based upon
sure at 1050 and 1200° F. The an increase in strength even though strength levels higher than that
decrease in length of ' .-hard stain- appreciable dimensional contraction which would be indicated for the
less is 0.006 in. /ft longitudinally after occurred. It is expected that the annealed material. In _ addition,
exposure to 700° F and 0.012 in. ft contraction observed is related to testing has shown that significant
after 1200° F. For the full-hard combined effects of precipitation, changes in dimensions can occur
specimens, the decrease in the tempering of martensite formed after exposure to elevated tempera-
same direction was 0.012 in. ft after during cold working and relief of tures and that the effect of tempera-
700° Fand 0.018 in. ft after 1200° F. residual stresses. It should be ture on the properties of the cold-
The changes in dimensions for the noted that there is a marked change worked material are great at tem-
annealed and '/,-hard specimens are in room-temperature tensile and peratures above 1000-1050° F. The
negligible regardiess of the grain yield properties of full-hard stain- data obtained from this study
direction. less steel after exposure above might vary somewhat from heat to
All data obtained on the me- 1050° F. This factor, combined heat of the material depending
chanical properties at room tempera- with the dimensional changes, in- upon processing history and com-
ture after exposure to various ele- dicates that the use of cold-worked position. It is suggested that addi-
vated temperatures are compiled in 301 stainless steel in critical stress tional testing would be required to
Table 2. These data show, as applications, even for the very short obtain a more statistically sound
illustrated in Fig. 6, that the exposure in this temperature range, evaluation of these effects.

PB 161195 Recent Developments in tations of using explosive energ

Titanium Brazing (DMIC Memo in metal-working operation.
RESEARCH NEWS 45). March 1960. 8 pages. 50 PB 161202 Review of Problems
cents. Developments of improved in Using Flat-rolled Materials in
brazing filler metals are discussed Air- and Space-weapon Systems
and evaluated. (DMIC Memo 52). April 1960.
Defense Metals Reports 22 pages. 75 cents. Problems in
PB 161198 Brazing for High-
Seven technical memorandums temperature Service (DMIC Memo producing and using sheet materials
prepared by the Defense Metals In- 48). March 1960. 15 pages. 50 intended for service at high tempera-
formation Center, Battelle Memorial cents. The extent of advancement tures are reviewed.
Institute, on the characteristics and in the field of brazing for service PB 161203 Notes on the Diffusion
working of titanium, molybdenum, temperatures in excess of 600° F is Bonding of Metals (DMIC Memo
beryllium, tungsten, steel and var- summarized. 53). April 1960. 9 pages. 50
ious alloys, have been released to PB 161199 The Determination cents. Diffusion bonding is dis-
industry through the Office of of Oxygen, Nitrogen, Hydrogen and cussed as a joining technique ob-
Technical Services, Business and Carbon in Molybdenum, Tungsten, taining coalescence of clean, closely
Columbium and Tantalum (DMIC fitting parts by applying pressure
Defense Services Administration,
U. S. Department of Commerce. Memo49). March1960. 20 pages. with or without heating’ the
The Defense Metals Information 50 cents. The determinations of assembly.
center was established at Battelle the interstitials in various refractory
metals are discussed with considera- Welding Bibliography
at the request of the Assistant
Secretary of Defense for Research tion as to the interstitial element Selective bibliographies, listing the
and Engineering to provide informa- sought. latest Government research reports
tion on beryllium, titanium, refrac- PB 161200 Diffusion Rates and and other technical documents avail-
tory metals, high-strength alloys Solubilities of Interstitials in Refrac- able to science and industry, have
for high-temperature service, corro- tory Metals (DMIC Memo 50). been published by the Office of Tech-
sion and oxidation resistant coatings April 1960. 12 pages. 50 cents. nical Services. A welding bibliog-
and thermal protection systems. This memo reports on the diffusion raphy, supplement to Catalogs of
In addition to issuing formal reports, rates and solubilities of interstitials Technical Reports (the former name
the Center prepared these memoran- in refractory metals. of the Selective Bibliographies), is
dums covering information gathered PB 161201 Bibliography on Ex- available from OTS, U. S. Depart-
in more limited areas. plosive Metal Working (DMIC Memo ment of Commerce, Washington 25,
The following may be ordered from 51). April 1960. 17 pages. 50 D. C., and is identified as: SB-402
OTS, U. S. Department of Com- cents. This memo lists works in- Welding (supplement to CTR’s
merce, Washington 25, D. C. vestigating the advantages and limi- 324-331). Price 10 cents.

442-s | OCTOBER 1960

Studies of Methods for Sealing Ends

of Reactor Fuel Rods for PWR

Various types of pressure bonding, resistance welding

and fusion welding are evaluated to find a useful method

for making the end seals

BY : - VAGI AND OD. C. MARTIN

ABSTRACT. As part of the WAPD* a metallurgical bond absolutely free lateral upsets would expose the fuel
program for developing fuel rods for of voids or porosity was required for to the cooling water. Therefore, it
PWR, + methods for sealing the ends of heat transfer. In addition, to per- was desirable to produce a mini-
Zircaloy 2-clad, uranium-12 w/o mo-
lybdenum-cored fuel rods made by mit machining the fuel rod without mum of lateral fuel upset during
coextrusion were investigated. At- exposing the core, a minimum welding. The first tests showed
tempts were made to join Zircaloy 2 amount of core upset was desirable. that the lateral fuel upset could be
end plugs to the rod by means of pres- Exposed core material would lead to controlled by varying the extension
sure bonding, resistance welding and rod failure and contamination of the of the rod out of the clamping dies.
fusion welding. water-cooling system. A very small amount of lateral upset
Exploratory tests indicated that was produced with extensions of
resistance upset welding was promising. Resistance-upset-welding in. or less. Lateral fuel upsets
Tension-test fractures occurred in the Tests produced with s- and -in. ex-
Zircaloy 2 end caps away from the
weld joint when materials were vac- Exploratory resistance - upset- tensions are compared in Fig. 1.
uum treated prior to upset welding. welding tests were made in an open For the remaining tests, the fuel-rod
Percussion welding and flash welding atmosphere and in an open atmos- extension was in. or less.
were eliminated from extensive con- phere with supplementary argon To obtain additional information
sideration because molten core alloy shielding. A 30-kva flash-welding on the conditions required for weld-
was extruded out along the cladding- machine was used for all resistance- ing, a number of end seals were made
to-end-cap bond line when these proc- for tension testing to determine the
esses were used. Attempts to fusion upset-welding tests. Rods for mak-
ing simulated fuel-rod end seals were strength of the bonds. These end
weld an overlay on the ends of the fuel
rods resulted in end seals that had in. diam and 2 in. long. The seals were resistance upset welded
uranium at the surface and also con- welding tests initially were on in an open atmosphere, and fuel-rod
tained gas porosity. square-butt joints in as-extruded extension was in. Welding-cur-
rod. However, when gas porosity rent settings were chosen arbitrarily,
Introduction was discovered in welds in as-ex- using a setting that produced good
As part of the WAPD program of truded rod, additional tests were welds by visual inspection. Weld-
PWR _ fuel-element development, made using materials degassed in a ing current was 14,000 amp ac at
methods for sealing the ends of vacuum at 925° C for 25 hr. Also, 65% phase shift. For identification
s-in. diam fuel rods were investi- various joint designs were tried, but purposes, this group of welds was
gated. The reference fuel element there was no noticeable effect on designated Series A. Results of
was a rod of ' ,-in. diam uranium- either strength or gas_ porosity. tension tests on six samples (A-1 to
12 w/o molybdenum core, clad by Tension tests and metallographic A-6) are included in Table 1. Ex-
coextrusion with 0.030-in. thick Zir- examinations were made on many of amination of the fracture surfaces
caloy 2. To find a useful method for the samples. In addition, a large showed that the welds contained gas
making the end seals, various types number of welds were made for porosity. The fracture surfaces,
of pressure bonding, resistance weld- development of nondestructive in- Fig. 2, also showed that gond bonds
ing and fusion welding were eval- spection methods. were made between the Zircaloy 2
uated. A reference process readily cladding and end caps. Metallo-
adaptable to high-production rates Resistance-upset-welding Tests graphic examinations of another
was desirable. The work reported Using As-extruded Materials specimen (A-8) from this group also
here was carried out in 1954.{ The first test in a series of resist- disclosed the porosity shown in
In the initial phases of this study, ance-upset-welding tests was made Fig. 3.
to relate welding conditions to The photomicrographs in Fig. 4
J. J. VAGI is a Welding Engineer of the Metals
Joining Div. and D. C. MARTIN is associated lateral core upset. After the ends show the bond lines that were
with the Battelle Memorial Institute, Columbus, of the fuel rods are sealed, it is typical of welds in this group. In-
Ohio
probable that excess weld metal will clusions are evident along the clad
* Westinghouse Atomic Power Division be machined to the original diameter end-cap bond line, but not at the
+t Pressurized Water Reactor
t Publication delayed for declassification of the reference alloy rod. Large fuel end-cap bond line. In addi-

WELDING RESEARCH SUPPLEMENT 443-s

 seers | VVIVOeEMN 2904

1 HF-30HNO,;-30 lactic acid etch

Fig. 3—Gas porosity in a resistance
(a) (b) upset weld (A-8) made using
1 HF-30HNO,-30 lactic acid etch. (a) Fuel-rod extension, */1 (b) Fuel-rod extension, as-extruded materials. About X 2
in.
Fig. 1—Comparison of core upset produced from various fuel-rod extensions. X 2!/, tion, the black areas in the fuel ad-
jacent to the bond line were thought
at first to be voids caused by shrink-
age. Current evidence indicates
that the dark areas may be voids
caused by a constituent’s being
dragged out of the metal because of
overpolishing during the metallo-
graphic preparation.
Welding tests were continued in
order to determine whether gas
porosity could be eliminated by a
small change in both fuel-rod exten-
sion and welding current. A large
number of end seals, Series B, were
made for tension tests and for non-
Fig. 2—Gas porosity at tension-test fracture surfaces of resistance upset welds destructive tests. The weld bonds
made in an open atmosphere using as-extruded materials. X 2'/2 in tension-tested samples of Series

Table 1—Tension-test Results on Resistance-butt-welded End Seals Made Using As-extruded Materials
—Welding conditions’— Breaking
Fuel-rod Phase strength, Location of Gas
Designation extension, in. shift, % psi fracture porosity Remarks
Welded in air
Series A
A- 27 ,600 Along weld plane Extensive Upset machined to */;¢ in.
21,600 Along weld plane Extensive Upset machined to °/;, in.
22,600 Along weld plane Extensive Upset machined to °/;, in.
42,900 Along weld plane Extensive Upset machined to °/,, in.
36 ,000 Along weld plane Extensive Upset machined to °/,, in.
29,400 Along weld plane Extensive Upset machined to °/,, in.
4 >>>>P>
2 w
54,500 Along weld plane Extensive As upset
55,000 Along weld plane Extensive As upset
70,000 Along weld plane Extensive As upset
73,000 Along weld plane Extensive As upset
65,000 Along weld plane Extensive As upset
36 ,800 Along weld plane Extensive As upset
51,000 Along weld plane Extensive As upset
59,500 Along weld plane Extensive As upset
86 ,000 Along weld plane None Weld machined to '/, in. diam
73,900 Along weld plane None Weld machined to '/, in. diam
Doakhwnr
eOOYNAHSWNH®
RE 47 ,300 Along weld plane Extensive Upset machined to °/,, in. diam
AAA
AA
kbs 71 45 ,000 Along weld plane Extensive As upset
71 47,200 Along weld plane Extensive As upset
Welded using supplementary argon shield
Series C
C-1 Along weld plane Extensive Upset machined to °/;. in
C-2 Along weld plane Extensive Upset machined to °/,, in
C-3 Along weld plane Extensive Upset machined to °/;.in
C-4 Along weld plane Extensive Upset machined to °/,;.in
* Secondary current = 14,000 amp.

444-s | OCTOBER 1960

 Zircaloy 2 Zircaloy 2
end cap end cap

Bond line
Bond line
Zircaloy 2
cladding

1 HF-30HNO,—30 lactic acid etch. Fuel-to-end

end-cap bond line. x 75. (Reduced by '/2 cap bond line. xX 500. (Reduced by upon
upon reproduction) reproduction)
Fig. 4—Weld bond lines obtained in as-extruded materials by resistance upset welding in an open atmosphere

B had higher tensile strength, Table believed that a degassing treatment mercury. The combinations of
1, but gas porosity was not changed prior to welding might assist in re- material welded, identified as Series
significantly. Tension-test fracture ducing gas porosity in the weld. D, are given in Table 2.
surfaces of some of the Series B Exploratory welding tests were Tension tests on the seven samples
samples are shown in Fig. 5. made using various combinations of showed that high-strength bonds
Two samples (B-9 and B-10) had degassed materials. The degassing were obtained with the vacuum-
much higher tensile strength than treatment was performed in a good treated materials. Sample D-6
other specimens from the same vacuum at 925° C for 25 hr with a failed along the weld plane; the re-
group. These two samples had the starting pressure of 1 « 10~° mm of maining samples failed outside the
cladding machined off, mainly to
determine whether the end cap was
bonded to the fuel. The tension-
test results indicated good bonding.
Tension tests also were made on
samples, Series C, welded in an open
atmosphere with supplementary
argon shielding supplied through a
hollow tube around the weld joint.
Table 1 also includes results from
tension tests on Series C samples.
The breaking strengths of these
samples appeared about the same as
in the other series. Also, gas
porosity was about the same.
In addition to the square butt- Fig. 5—Porosity at tension-test fracture surfaces of resistance upset welds
joint design used for most of the made in an open atmosphere using as-extruded materials. About X 2
tests, several other joint designs
were studied. These included:
Fuel rod Zircaloy 2 end cap
1. Recessed fuel matching end
cap.
Beveled fuel rod flat end cap.
Beveled end cap flat fuel rod. Both materials
Beveled end cap matching vacuum treated
fuel rod.
Flat end cap—fiat fuel rod;
seal bead fusion welded around on nr =ilanenel Sanita
joint.
The results of tension tests and
examinations of fracture surfaces
did not indicate any improvement End cap only
vacuum treated
in strength or reduction in gas
porosity.
Resistance-upset-welding Tests
Using Degassed Materials
Resistance-upset-welded end seals
made using as-extruded materials Fuel rod only
vacuum treated
showed gas porosity in nearly all of
the welds. The surface of the gas
pockets appeared bright and shiny,
indicating that a nonoxidizing gas
was causing the porosity. It was Fig. 6—Tension-test results on various combinations of degassed materials. X 2/

WELDING RESEARCH SUPPLEMENT | 445-s

 Bend Tension Bend
fracture fracture fracture Table 2—Materials Combination for
Series D Welds
Treatment
Sample End cap Fuel rod
D-1 Degassed Degassed
D-2 Degassed Degassed
D-3 Degassed Degassed
D-4 Degassed As extruded
D-5 Degassed As extruded
D-6 As extruded Degassed
D-7 As extruded Degassed

D-6
Fig. 7—Fracture surfaces of resistance upset welds made
using degassed materials. X 2'/, weld in the Zircaloy 2 end cap.
The tension-tested samples are
shown in Fig. 6. Samples D-3 and
D-4 were subjected to bending to
produce failure along the weld plane
for examination of the weld bond for
gas porosity. Only a small amount
of porosity was found in Sample
D-6. None was found in Samples
D-3 and D-4 (Fig. 7). The results of
the tension tests are given in
Table 3.
To examine weld bonds in the
vacuum-treated materials further, a
longitudinal cut was made through
the axial center of sample D-2. No
evidence of gas porosity in the weld
As-weilded. X2 Machined and etched. X 2 was found when the polished-and-
etched surfaces were examined. In
addition, several cracks were ob-
served along the core-clad bond
line at the juncture of the original
fuel and the upset fuel.
Hydrogen Analyses
Observations of metallographic
samples and fracture surfaces of
resistance-upset welds in as-extruded
materials indicated that gas porosity
was causing poor bonds. Gas poc-
kets in individual samples appeared
to form in a group along the fuel-
clad bond line in the core alloy. It
— was found that degassed samples
were relatively free of gas porosity.
“ne boy The hydrogen contents of degassed
“s ore
ae i‘ and as-extruded materials are given
in Table 4.
The hydrogen analyses indicated
1 HF-30HNO,-30 lactic acid etch. As welded 1 HF-30HNO,-20 lactic acid etch. Welded and that the degassing treatment was
xX 30 machined. X 30
beneficial in hydrogen removal. It
Fig. 8—Flash-welded end seals showing fuel alloy trapped appears that, in the fuel rod, hydro-
along the cladding-to-end-cap interfaces gen is concentrated in the cladding.
On a volume basis, for the material
involved in bonding, there is less
hydrogen in the as-extruded core
than in the degassed end-plug mate-
rial.

Flash-welding Tests
Attempts were made to seal the
ends of PWR fuel rods by flash
welding using the 30-kva flash-
welding machine. Visual inspec-
1 HF-30HNO,-30 lactic acid etch. (a) Flat-disk end cap. (b) Recessed end cap
tion indicated that good bonds
Fig. 9—Pressure welds made using a 200-kva spot-welding machine. X 2 were made. However, examination

446-s | OCTOBER 1960

 Bond line

End cap End cap

“ee
a #
1 HF-30HNO;- 30 lactic acid etch Pressure weld A 30 HNO;-30lactic acid etch Pressure weld B
Fig. 10—Fuel-to-end-cap bond lines in end seals made using a
200-kva spot-welding machine X 500 (Reduced by upon reproduction)

Table 3—Tension-test Results on Resistance-butt-welded End Seals Made

Using Vacuum-treated Materials
Breaking
strength,
Designation psi Location of fracture Remarks
Both materials vacuum treated
D-1 68 ,900 Zircaloy 2 end cap Not examined for porosity
D-2 67 ,500 Zircaloy 2 endcap Metallographic section showed no
gas porosity
D-3 68, 100 Zircaloy 2 end cap Bend-fractured surfaces in fuel rod
along bond line; showed no 1 HF-30HNO
porosity weldA
End cap only vacuum treated
D-4 71,800 Zircaloy 2 end cap Bend-fractured surfaces in fuel rod
along bond line showed no
porosity
D-5 70,000 Zircaloy 2 end cap Not examined for porosity
Reference alloy only vacuum treated
D-6 74,500 Along weld plane Very small amount of porosity was
visible at X30 magnification in
fuel rod
D-7 79,900 Zircaloy 2 end cap Not examined for porosity
Fuel-rod extension, * phase shift supplementary helium shielding; and machined to
in. diam
1 HF-30HNO;-30 lactic
weld B
Fig. 11—Cladding-to-end-cap bond lines in
at high magnifications showed that and '/, in. thick, were attached to end seals made using a 200-kva spot-
large amounts of fuel were trapped «-in. diam, -in. long fuel rods. welding machine. X 75. (Reduced by
along the cladding-to-end-cap inter- In addition, several end caps, upon reproduction)
faces, Fig. 8. It was believed that in. diam, had a sin. deep recess to
it was impossible to eliminate the fit on the end of the fuel rod. Ex-
trapped fuel alloy at the bond line, amples of the welds are shown in
and no further studies of the process Fig. 9. Impact tests indicated that Table 4—Results of Hydrogen Analyses
were made. good pressure-welded bonds were on Degassed and As-extruded Materials
made only when the fuel rod was de-
formed considerably. Figures 10 Hydroger
Pressure-welding Tests content, ppm
and 11 show the bonds produced in As
Attempts were made to attach the samples given in Fig. 9. Be- Material extruded Degassed
Zircaloy 2 and caps to the ends of cause of the excessive deformation in. diam Zire
the PWR fuel rods using a 200-kva required to get good bonds, this aloy 2 rod (end
spot-welding machine. The process method was not investigated exten- plug)
used was similar to resistance-upset sively. -in. diam alloy
welding, the major difference being rod with a */;.-in
that pressure is applied continuously Percussion-welding Tests hole drilled axi
to the weld when the spot-welding The chief advantage of percussion ally to remove a
machine is used. The flash-welding welding leading to its consideration portion of the
machine used was cam driven and, for sealing ends of fuel rods is that fuel
U-12 w/o Mo fuel
when the maximum rise of the up- only a thin surface layer on each alloy only
setting cam was reached, all pres- part becomes molten. In the sys- sin. diam fuel
sure on the weld was released. tem used for percussion-welding rod
Zircaloy 2 end caps, */;, in. diam tests, capacitors precharged to 150 v

WELDING RESEARCH SUPPLEMENT | 447-s

 sors | VU IT YVUODCN 1700U

5/ye-in. diam
«in. diam Zircaloy 2
Zircaloy 2 '/w-in. by 90-deg
fuel rod fuel rod

Fig. 12—Percussion-welding-test results.

to 840° C and subjected to a pres-

sure of 7500 psi for 10 min. When
the welded bar was removed from
the molybdenum sleeve, the diam-
eter at the weld was measured and
was found to have increased less
than 0.002 in. The weld appeared
very good to the eye, as can be seen
from the photograph in Fig. 14.
When subjected to tension, how-
ever, the specimen broke at a low
load (2360 lb = 31,000 psi). The
fracture surfaces showed that little,
Fig. 13—Fusion-welded end seals. X 3
if any, metallurgical bonding had
occurred between the Zircaloy 2
end cap and the uranium-12 w/o
molybdenum core. However,
bonding between the Zircaloy 2
cladding and the end cap caused
visible necking at the point of frac-
ture.

Zircaloy 2 end cap Weld Fuel rod Discussion

Fig. 14—Pressure-bonded end seal. Area around weld was The results of exploratory tests
swabbed with ammonium bifluoride to remove discoloration indicated that resistance-upset weld-
ing was the most promising method
for end capping the PWR reference
were connected to an open circuit, using the gas tungsten-arc welding fuel element. High-strength bonds
one terminal of which (fuel rod) was process. Tests were made on fuel connecting the end cap to the fuel
fixed. The Zircaloy 2 end cap was rods having the fuel recessed to a rod were resistance upset welded,
held in a collet terminating the pis- depth of */,; in. and attempts were but gas porosity and shrinkage
ton rod of a force cylinder. Upon made to melt the Zircaloy 2 edge cavities remain a problem. Gas
application of air pressure to the top and then to overlay the end of the porosity was reduced to a low level
of the cylinder, the end cap was fuel using '/»-in. diam Zircaloy 2 and a significant increase in tensile
moved toward the fuel rod. The filler wire. During the tests, it ap- strength was obtained when mate-
capacitors were discharged when peared that considerable alloying rials were vacuum treated to remove
the ionization potential across the occurred between the fuel and the hydrogen prior to welding.
air gap was reached. The arc be- Zircaloy 2 overlay. Examination Because of the expulsion of molten
tween the parts produced a melting of fusion-welded end seals indicated metal, two resistance-welding proc-
of the abutting surfaces immediately that the outer surfaces of fusion esses were eliminated from exten-
prior to contact between the rapidly welds were contaminated with ura- sive consideration early in this in-
approaching workpieces. Figure 12 nium. In addition, considerable gas vestigation. These processes were
shows joint design and results of porosity was found in the seals, as flash welding and percussion weld-
percussion-welding tests. shown by sections from two samples ing. Because of the rapid alloying
Figure 12 shows that melting in Fig. 13. Since it appeared im- of molten Zircaloy 2 filler metal and
occurred over small portions of the possible to make overlays without cladding with the uranium-12 w/o
abutting surfaces even when a heat uranium contamination, no further molybdenum core alloy, overlaying
concentrator was used. Percus- work on the method of sealing was by tungsten-arc welding also was
sion-welded end seals probably could carried out. eliminated as a feasible method for
be made with equipment of higher making the end seals.
capacity. However, indications are Pressure-bonding Tests Pressure bonding the end seals in
that some fuel would be trapped at One attempt was made to pressure a vacuum or by upsetting, using a
the cladding-to-end-cap bond line bond a Zircaloy 2 end cap to a fuel spot-welding machine, also ap-
due to the expulsion of molten metal rod. The bars were faced off on a peared to show some _ promise.
when the two parts are brought in lathe so as to have clean, flat sur- However, pressure bonding in a
contact. Consequently, no _ at- faces. These surfaces were butted vacuum system might restrict the
tempt was made to do further work together inside a molybdenum ability to produce large quantities
on a machine of larger capacity. sleeve, and the assembly was in- for production of the fuel rods. Be-
serted into the pressure bonder. cause of the erratic results obtained
Fusion-welding Tests The bonder was closed and evacu- when attempts were made to spot
Fusion-welding tests were made in ated to about 0.1-micron pressure. weld end seals, the method was not
an open atmosphere and in argon The specimen was induction heated investigated extensively.

448-s | OCTOBER 1960

 Volatilization Phenomena

in High-Temperature Brazing Filler Alloys

Development program results in oxidation-resistant

brazing filler alloys containing liquidus temperature depressants
which can be volatilized during brazing

BY WILLIAM LEHRER AND HARRY SCHWARTZBART

ABSTRACT. Experimental brazing filler binary Ni-In alloy can be considered tion resistance to the alloys, the
alloys have been developed containing practical alloys for high-temperature removal of indium (the temperature
temperature depressants which have usage. depressant) would necessarily result
been volatilized during the brazing of A direct effect of metallic vapors on
stainless steel leaving joints of high wetting and flow of the filler alloy was in an alloy of higher melting tem-
remelt temperature. The mechanisms observed and subsequently discussed perature. Carrying this premise
by which remelt temperature is in- with respect to their influence on further, the removal of any tem-
creased have been studied for the surface-tension relationships. It has perature depressant after flow would
range of alloys investigated which con- been shown that in the presence of result in a more suitable high-tem-
tained nickel, chromium, germanium, metal vapors, discoloration of the base perature filler alloy from the stand-
iron, lithium and phosphorus. The metal, wetting, flow and skull forma- point of high-temperature strength.
two main mechanisms are (a) dissolu- tion of the filler alloy are functions of The desirability of an alloy that
tion of the base metal in the filler the rate and quantity of metal vapor may be applied at relatively low
metal and diffusion of constituents of removal.
the filler metal into the base metal and brazing temperature, and subse-
(6) volatilization of the filler metal con- quently have a much greater remelt
Introduction temperature, was obvious and be-
stituents.
It has been shown theoretically and The objective of this program came the objective of this research
experimentally that, in order for re- was to develop oxidation-resistant program.
melt temperature to increase, there brazing filler alloys containing liq-
must be solid solubility of the diffusing uidus temperature depressants Fundamental Considerations
or volatile element in the filler alloy. which can be volatilized during The compositional changes that
Although compositional changes may
result from either mechanism, a remelt vacuum brazing. This program had may be brought about in the filler
temperature change does not neces- its inception as a result of the alloy alloy during a brazing cycle may
sarily occur. The degree to which a development work conducted at occur by various mechanisms. Fur-
compositional change affects remelt Armour Research Foundation under thermore, these mechanisms may
temperature is proportional to the Contract No. AF(600)-33406, Task operate in different modes and, also,
solid solubility. No. 73022, ‘Development of Oxida- result in a remelt temperature de-
In this program, vacuum pumping tion and Liquid Sodium Resistant crease as well as the desired in-
was the most efficient vapor-removal Brazing Alloys.” In this study, crease. The major mechanisms
method among several discussed and indium and lithium were added to may be generally described as (a
investigated. Furthermore, under
vacuum the _ surface-area-to-volume nickel-, iron- and chromium-base metallurgical change due to solution
ratio of the filler alloy becomes less alloys as brazing temperature de- of the base metal and the selective
rate controlling due to boiling. Boil- pressants and as aids to flow when diffusion of filler-alloy constituents
ing, however, causes porosity in the used in a hydrogen atmosphere on into the base metal and (6) compo-
joint. Inconel-stainless brazements. A\l- sitional change due to volatilization
Of the alloys investigated, a 61% loy compositions were shown that of an alloying element. A secondary
Ni - 39% In alloy exhibited the great- flowed between 1750 and 1900° F. mechanism, which has been ob-
est remelt temperature rise due to Lithium - bearing Ni - Cr - Ge alloys served in other brazing research at
volatilization alone, whereas a 94% also withstood 500 hr of atmos- the Foundation, results in the re-
Ni—6% P responded only to dissolu-
tion and diffusion to exhibit the largest pheric oxidation at 1650° F. The moval of one of the filler-alloy
rise in remelt temperature due to this presence of appreciable indium, constituents to a slag by reactions
mechanism. Ni-Cr-In-Ge alloys ex- however, resulted in a loss of oxida- with base-metal oxides or the atmos-
hibited a substantial rise in remelt tion resistance. It was recognized phere. This mechanism was ig-
temperature due to the operation of that the desirable flow and tempera- nored on this program, since it is
both mechanisms. This alloy and the ture depressant properties of in- secondary and was not observed to
dium could still be utilized provided occur. Although volatilization was
WILLIAM LEHRER is Senior Product Engi that a means was established for the major mechanism investigated
neer, Raytheon Semiconductor Div., Newton, the elimination of the element after in this work, an understanding of
Mass. Formerly, Research Metallurgist, Armour
Research Foundation, Chicago, Ill., and HARRY flow had occurred. Oxidation re- the diffusion and solution mech-
SCHWARTZBART is Supervisor, Welding sistance, after indium removal, anism was requisite, since in all
Research, Armour Research Foundation, Chicago, should then be that of the Ni-Cr-Ge practical brazing operations both
iil.
Paper presented at AWS National Fall Meeting alloys. may be expected to operate simul-
held in Detroit, Mich., Sept. 28-Oct. 1, 1959 As a corollary to restoring oxida- taneously.

WELDING RESEARCH SUPPLEMENT 449-s

 result in a substantial rise in the
solidus temperature. Note the cri-
terion of solid solubility——the greater
the solubility, the less diffusion is
required to accomplish the solidus
rise. One factor, that of diffusion
| rates, has not been discussed. De-
scribed above is the equilibrium end
| } | result only—the time depends en-
4 tirely on the diffusion coefficients.
| | Initial saturation occurs very rapidly
1] (Cr,-Cr2) in a matter of seconds.
| .
x %oY—~ y z Y This has been shown to be true by
Tea Y—> Bredzs and Schwartzbart who
A.—-Eutectic system, no solid solubility .- Eutectic system, appreciable solid
solubility proved the saturation of liquid
Fig. 1—Hypothetical binary equilibrium diagrams copper filler with an _ iron-base
metal occurred in less than 2 sec
at 2000° F.? On the other hand,
Change of Composition Through the horizontal tie line AB is at- it has been shown‘ that diffusion
Solution and Diffusion tained. The filler-alloy composi- into the base metal is time consum-
The silver brazing of iron-carbon tion, Cr:, is now enriched in Y ing; therefore, in practice, complete
alloys is the only practical system and is in equilibrium with Y, crystallization would not occur at
in which no changes of composition composition unchanged. A _ cross brazing temperature.t
occur during brazing in either the section of the hypothetical test
filler alloy or the base material. sample would show dissolution of Change of Composition Through
The lack of mutual solubility in any the base metal. The solidus tem- Volatilization
proportion* in the liquid or solid perature, however, is unchanged Although the room-temperature
state makes this so. Other than even though filler-alloy composi- vapor pressure of commonly used
this one case, no brazing filler tion has changed from Cy, to Cr. metals is so small that loss by vapor-
alloy-base metal combination exists The system with solid solubility, ization is an nth order effect, the
in which a change in composition however, presents a less simple same is not true at elevated tem-
through dissolving or solutioning picture. Again, an alloy Z-Y of peratures. Dezincing of brasses,
is not to be expected to some degree. composition Cy, is molten at T, in for example, at elevated tempera-
The dissolving (in washing) of contact with base metal Y; but, tures is a well-known phenomenon.
base metal is even more _ pro- from Fig. 1B, the system is not at The process is considered one of
nounced in high-temperature braz- equilibrium. Furthermore, even at fractional distillation with the high-
ing where the high-temperature composition Cy,, after enrichment of vapor-pressure zinc evaporating out
filler alloys are essentially of similar the filler alloy with Y, equilibrium of the copper solvent. An impor-
component type as the heat-resist- is not attained; the tie line AB tant point to recognize is that this
ant base metals; e.g., nickel-base indicates this. In this case, the is a surface phenomenon and is de-
filler alloys are used primarily in composition of the base metal at pendent upon the diffusion rate of
the brazing of nickel-cobalt- and the interface must also change. the solute in the solvent. An
iron-base heat-resistant alloys—all This is accomplished through a example of the effect of diffusion on
mutually quite soluble in liquid diffusion of Z into Y resulting in a loss through volatilization can again
nickel. The result of this solubility composition of the base metal at be illustrated through a commonly
is a drive towards equilibrium in the interface of Cy. Thus, dif- observed phenomenon. Although
the polynary liquid-solid system. fering from the first case, transfer zinc is volatilized from a_ brass
The various effects of diffusion on of both components occurs, Z into surface at elevated temperatures,
the remelt temperature are shown Y and Y into Z both acting to the alloy color does not turn the
by a series of hypothetical binary enrich the filler alloy in Y. characteristic copper color because
equilibrium diagrams in Fig. 1. Equilibrium in the hypothetical of the rapid diffusion of zinc to the
Figure 1A is a diagram representing system above, however, is not yet surface replenishing the zinc con-
a eutectic system with no solid complete. Although Cr; is in equi- tent continually. On the other
solubility, whereas Fig. 1B illustrates librium with the base metal at the hand, when silver-copper alloys lose
appreciable solid solubility. These interface at composition Cm, a copper at the surface at elevated
diagrams represent the composition potential energy gradient still exists temperatures through oxidation and
of a molten filler alloy, Cr, at between the solid interface and the flaking, the color does change be-
temperature 7’, in contact with a bulk of the base metal at composi- cause of the inability of the
base metal of composition, Ca). tion Cy;. This energy gradient copper to diffuse rapidly through
The base metal, for the sake of results in a depletion of Z from the the silver and replenish the surface.
simplicity, is one of the components interface through solid diffusion of Volatilization of constituents in
of the filler alloy. Z away from the interface. The filler alloys will be dependent,
Considering first the eutectic sys- source of Z is the liquid filler therefore, not only on the relative
tem with no solid solubility, at alloy which, as depletion of Z vapor pressures of the constituents
T, an alloy of composition Cr; occurs, commences to crystallize but also on the rate of diffusion to
is not in equilibrium with the base at the interface. This process con- the surface. Obviously, a large
metal Cy;. As Y is soluble in the tinues until the filler alloy has surface-to-volume ratio of the filler
liquid, it dissolves until an equi- entirely solidified at composition alloy in the joint would reduce
librium condition represented by Cy;. It is obvious that, given the diffusion effect on the end result.
enough time at temperature T),
* Solubility of Ag in Fe is ~0.0004%; in the average composition of the filler + Nore: This statement excepts lengthy
Fe,,, ~ 0.0002%. alloy may change sufficiently to anneals resulting in diffusion bonding.

450-s | OCTOBER 1960

 The vapor pressures of the ele-
ments considered in this program
have been plotted in Fig. 2 as a
function of temperature. The data
for these curves were obtained
from Smithells.‘t The temperature
range chosen is that normally
used in high-temperature _ braz-
ing applications. One ele-
ment, phosphorus, is not shown in
the curves due to its enormous
vapor pressure in the temperature
range plotted (vp = 760 mm Hg
at 536° F). It is appropriate at
this point to define ‘“‘vapor pressure’”’:
“the pressure exerted by a vapor
when a state of equilibrium has
been reached between a liquid, solid
or solution, respectively, and its
vapor is called the vapor pressure
ofa liquid, solid or solution.’’> The
vapor pressures plotted in Fig. 2, VAPOR PRESSURE, MM Hg
however, describe the equilibrium Fig. 2—Vapor pressure-temperature curves of pertinent elements
pressure of these metals in the liquid
or solid state (the vapor pressure
curve is continuous) as pure metals,
but describe nothing regarding their be a function also of the volume of tainer is infinite in practical brazing
solutions. For dilute solutions, Ra- the system—.i.e., the larger the con- operations, the time is not. The
oult’s Law may be applied to de- tainer, the greater number of atoms quantity of atoms leaving the solid
termine the reduction of the vapor in the gaseous state required to surface will depend upon the vapor
pressure of the various constituents attain 0.9u. pressure of the solid-gas equilibrium.
of the alloy system.’ For present In a sense, an infinite volume It is readily seen that a substance
purposes, however, the vapor pres- may be created by removal of the of high vapor pressure tends to
sures of the pure elements have gaseous atoms from the system as supply many times more atoms in a
been used due to the complexity they leave the solid surface. given time to the surroundings than
of the filler alloys and their devia- If the silicon atoms are carried a lower-vapor-pressure substance
tion from ideality. Furthermore, away from the solid surface, equilib- in the effort to attain equilibrium,
since the vapor pressures of the rium is disturbed and more atoms even to the extent that the higher-
filler-alloy constituents differ by leave in an effort to restore the vapor-pressure material will boil.
orders of magnitude at the brazing equilibrium. The gaseous atoms Incidentally, the boiling process,
temperature, the evaporation proc- may be removed by entrainment in by its agitation, reduces the depend-
ess during brazing may be con- a moving gas, by vacuum pumping, ence of volatilization on diffusion
sidered as a pure distillation, with- by condensation, by reaction with rates in the liquid, and subsequently
out the occurrence of azeotropes. another material placed in the fur- reduces the importance of the sur-
By this is meant that the element nace or by combinations of these. face-to-volume ratio. However, in
with the highest vapor pressure Thus, if the system in which evapo- terms of the practical objectives of
will volatilize completely, after ration is being carried out provides this program, it might be anticipated
which the element with the next for the continuous removal of the that volatilization without boiling
highest vapor pressure will vola- gaseous atoms, the solid will com- would yield sound, dense joints,
tilize. pletely volatilize, given sufficient whereas volatilization with boiling
We will now consider the implica- time. would yield porous joints.
tion of the word “equilibrium” in Regarding the method of removal An example of the relationship
the definition of vapor pressure. of the gaseous atoms and the result- between vapor pressure and quan-
It can be seen from Fig. 2 that an ing rate, both the vacuum pumping tity of metal volatilized may be
element such as silicon will vola- rate and the inert gaseous flow rate taken from Fig. 2. It can be seen
tilize at a temperature of 2100° F are so extremely variable that that, at equilibrium between mol-
so as to exert a pressure of 0.9u. either may be the primary carrier. ten 68% Ni 32% In and their
That is, the solid silicon atoms will In the static situation, the presence vapors (using the vapor pressures
leave the surface of the bulk silicon of a condenser (as for example, a of the pure elements), the vapor
at this temperature at the same rate water-cooled end of a closed retort contains 7000 atoms of In to 1 atom
as gaseous atoms return to the sur- will create a gradient of concen- of nickel. This represents 13,500
face when the pressure of the tration of the gaseous atoms to g of In to 1 g of nickel.
silicon atoms is 0.94. As the proc- cause flow away from the solid The vaporizing of a volatile sub-
ess continues, the solid is con- the rate in this case being a function stance from an alloy, even if suc-
sidered to be in equilibrium with its of the mean free path of the gas- cessful, may or may not, however,
vapor. Since pressure and volume eous atoms in gaseous diffusion. aid in raising the useful tempera-
are related, it is obvious that the We may conclude that the quantity ture limit (solidus) of a filler alloy.
quantity of solid vaporizing will of solid vaporized is a function of Whether this is accomplished de-
t In computing the curve for nickel, the con the volume, and the volume is es- pends upon the relationship be-
stant, D, in the equation log p A/T sentially infinite in a_ practical tween the compositional change
By + C log T + DT, in the Smithells Handbook brazing operation. and the solidus, as was discussed
was found to be 1.31 x 10 instead of the
0.131 as listed Although the volume of the con- previously for compositional changes

WELDING RESEARCH SUPPLEMENT 45l-s

through diffusion. These consider- compositions selected were those cooled to act both as a quench prior
ations are the equilibrium re- that showed good wetting and flow to removal of a specimen from the
lationships in the solid state. Again, characteristics and oxidation resist- furnace and as a condenser for
reference is made to Fig. 1. The ance without indium in the pre- metallic vapors. The brazing pro-
analysis in the case of volatilization vious program cited. Binary alloys cedures were as follows:
is not nearly as complex, but the A, B, J, K and L were selected 1. Under vacuum at <0.5u—15
conclusion is the same. In Fig. primarily to simplify understanding min at temperature.
1A, it is readily seen that 100% of and subsequent explanation of any 2. Under helium at 5 psig—15
the volatile constituent in alloy observed phenomena. Alloys B, min at temperature.
X-Y must be removed before the D, E, J and L, although interesting 3. Under helium at 5 psig—15
solidus temperature increases. In from a fundamental standpoint min at temperature, followed
1B at temperature 7’, the amount particularly Fe-Sb and Fe-Ga) by evacuation to predeter-
of Z (the volatile constituent) that were eliminated from study for mined absolute pressure.
must be removed before the solidus reasons shown in Table 1. Alloy . Under helium at 5 psig for
rises must bring the alloy composi- N deserves separate mention as this total time of cycle 3 (25 to
tion into the solid solution range. alloy was formulated in an attempt 35 min at temperature).
Solid solubility, therefore, is a to strike the eutectic composition,
criterion in the case of volatiliza- if any, of the quaternary Ni-Cr- To determine the effect of the
tion as well as in the case of diffu- Ge-In system. The composition dissolution of the base metal and
sion-—the greater the solubility lim- was obtained by slowly melting diffusion of the filler alloy in the
its, the less volatilization required. alloy H under an argon atmosphere absence of volatilization, other T-
and analyzing chemically the first specimens were encapsulated in
Experimental Program drops of incipient melting alloy. very small Vycor tubes under a
Based upon the fundamental con- After the composition was deter- helium atmosphere and brazed at
siderations and the primary ob- mined, a large melt was made in a 2120, 2020 and 1920° F for the
jectives of the program, experi- manner similar to that described same length of time as allotted to
ments were carried out to examine for the other alloys. those specimens brazed in the re-
the following factors: Approximate solidus temperatures tort. It was felt that the small
of the alloys were determined by finite volume would rapidly saturate
1. The effect of interdiffusion of with vapors, and compositional
filler and base metal on the encapsulating small angular pieces
of alloy in Vycor glass, evacuating, changes would be primarily through
remelt temperature in the ab- diffusion and dissolution.
sence of volatilization. backfilling with argon and observing
the temperature at which the alloy To measure the _ volatilization
2. The effect of volume of braz- effects of the filler alloys en-
ing container on remelt tem- commenced melting. This was in-
dicated by the rounding of the sharp tirely separated from other primary
perature. factors, i.e., dissolution and dif-
3. The relative effects of dynamic corners. This temperature was
taken as a rough approximation of fusion of the base metal, a small
vacuum and of infinite-volume amount of each alloy was weighed
static helium atmospheres on the solidus temperature (+10° F).
All alloys were crushed to powder to the nearest milligram, placed ina
remelt temperature. dried alumina crucible and melted
. The effects of metallic vapors and preplaced along one side of the
capillary formed by a 3-in. T-speci- under 0.1u dynamic vacuum. The
on wetting and flow. time chosen at temperature was that
men of 304 stainless steel. Braze-
Eleven alloys were arc melted ments were made at 2120, 2020, comparable to a _ brazing cycle.
under inert gas, in water-cooled 1920 and 1820° F. These braze- After cooling under vacuum, the
copper crucibles and with a non- samples were reweighed and _ per-
ments were made in a 2-in. cent compositional change was cal-
consumable tungsten electrode; each
alloy contained a volatile constit- diam Inconel retort sealed and culated. After the vacuum treat-
uent (Ga, In, Li, Sb or P). The connected in such a manner as ment and reweighing, the incipient-
compositions of these alloys are to allow a vacuum or a positive melting temperature of the alloys
shown in Table 1. With the excep- helium pressure to be applied. was redetermined in the manner
tion of the binary alloys, the Both ends of the retort were water- previously described.

Table 1—Composition of Alloys Containing Vclatile Elements

Alloy ——————___—_—_—_—_—_——_Composition (w/o), % Solidus
no. Ni Cr Fe In Li Sb Si temp., ° F Remarks
61 at isan 39 athe ree 1675 + 10
at ae 48 ~ te 52 1825 + 10 Eliminated—formed brittle joints
64.6 17 vyBs 9. 1885 + 10
oOoOw>33.3 he 16. Eliminated—nonhomogeneous and er-
ratic in mp
m 38. a ud er Soa 7 pee ee Eliminated—lithium separated as 2nd
liquid phase
38. 28.6 eee ; ax avr ve ; oa os M52
28. _ as ' a ps i sai .-. 1800+ 10
30. 30.8 “~ _ oe pata : ore
33. iid aad “io are ; ea -.. 1885+10
89 Ete tek , ae gi int ret pap ws Eliminated in favor of alloy M
1900 + 10
64 ‘
raAer-roOn Eliminated—was not homogeneous after
successive arc melts
94 ay han nas eas ae 7“ at .-. 1600+10
z= 35 ian tee ja ive sabi --- 1750+ 10

452-s | OCTOBER 1960

 The method of determining the such low deviations, the remelt discussed in four separate sections.
remelt temperature of brazements apparatus is seen to be accurate These are effect of dissolution
is based upon the mechanical fail- and reproducible. Therefore, the and diffusion on the remelt tempera-
ure of the joint under load at the scatter evidenced by the determi- ture, (b) the effect of volatilization on
onset of melting. nations of remelt temperature of the remelt temperature, (c the com-
A schematic drawing of the the experimental alloys is an indi- bined effects on the remelt tempera-
apparatus and sampling procedure cation of compositional hetero- ture and the effect of brazing
is presented in Fig. 3. geneity along the length of the cycle on wetting and flow.
Holes for the lifting wire were T-brazement.
punched into the vertical member Effect of Dissolution and Diffusion
of the T-specimen samples cut from on the Remeit Temperature
Results and Discussion
the brazed 3-in. T-specimen (Fig. The results of brazing stainless-
3B). A chromel - alumel thermo- The results will be presented and steel T-specimens at three tempera-
couple was spot welded to the
horizontal member of the sample.
A small stainless-steel fixture, affixed
with a wire from which to hang a
100-g weight, was slipped over the
horizontal member of the sample.
The thermocouple leads were inserted
into a Dewar flask mercury cold
junction maintained at constant
temperature and connected to a
rapid recorder-potentiometer. With
the lifting wire in a lowered posi-
tion, the sample and appurtenances
were attached (see Fig. 3C). The
automatic recorder is turned on
and the sample lifted into the hot
zone of a furnace maintained at
2400-—2500° F.
The load of 100 g was determined
to be the minimum load to insure
against binding when incipient melt-
ing occurs. A minimum load was
sought so that early parting of the
two members would not occur from JS
creep failure or fracture of the filler Fig.3 -Determination of meiting temperature of filler alloy in brazements
alloy while wholly solid. The sam-
ple size was minimized and the
furnace maintained at temperature
so as to have a high heating rate Table 2—Remelt Temperature of Filler Alloys After Brazing in Confined
Static Helium Atmosphere (304 stainless steel-304 stainless steel)
again to minimize errors through
possible creep failure. Solidus Brazing Remelt Temp. change due
When the joint failed, the hori- Alloy temp., ° F temp., ° F temp., ° F to brazing, ° F
zontal member dropped from the AE 1675 + 10 2120 1801 + 67
furnace due to the dead-weight load. 2020 No joint formed
This was immediately reflected in 1920 No joint formed
a drop in temperature of the hot CE t 2120 1966 + 118 significant change
2020 > 2290" 405
junction of the thermocouple and
1920 No joint formed
recorded on the potentiometer chart. 2120 1968 + 55 4.22 4 66
After failure, the thermocouples 2020 1904 + 18 No significant change
were cut from the samples, the in- 1920 1893 + 14 No significant change
sulators moved back and the ther- 2120 2270° +470
mocouples rewelded to the next 2020 No joint formed
group of samples. 1920 No joint formed
Prior to testing experimental al- 2120 +172 + 88
2020 No significant change
loy brazements with the above ap-
1920 No significant change
paratus and procedure, the remelt 2120 +181 + 51
temperature was determined for 2020 No significant change
copper and silver as a check on the 1920 +60 + 33
accuracy of the test. T-specimens 2120 +70=+ 27
of 304 stainless steel were brazed 2020 +19 15
with these pure metals at 2000 and 1920 No joint formed
1800° F, respectively, under vac- 2120 2009 + 88 +409 + 88
uum. The remelt temperature of 2020 No joint formed
copper was found to be 1951 1920 No joint formed
2120 2236 + 27 +486 + 37
8° F (mp 1981° F), and that of 2020 1994 + 68 +254 + 78
silver was found to be 1765 2° ¥ 1920 1870 + 83 +120 + 93
mp 1761° F). Since the checks 1820 1854 + 75 No significant change
of the accuracy of the melting
points of silver and copper yielded a One sample,

WELDING RESEARCH SUPPLEMENT 453-s

 tures in the Vycor tubes are listed at the lower temperatures. possible to raise the remelt tempera-
in Table 2. It has been postulated Significant effects of dissolution ture of alloy A more than alloy M if
that, in the confined volume, the and diffusion are noted for the binary one considers only the solubilities
filler alloy and base metal would be Ni-P alloy (M) and the low melting of indium and phosphorus in nickel.
in equilibrium with their vapor and Ni-Cr-Ge-In alloy (N). Alloys H (Nickel dissolves a maximum of
compositional changes due to vola- and I, closely approximating the 25% indium, and no or little
tilization could not occur. The composition of alloy N, show some phosphorus.) The fact that the
data obtained in these experiments rise in solidus temperature through actual behavior is the exact reverse
were limited by the poor wetting diffusion and dissolution; and ac- of this must reflect the different
and flow characteristics of the filler tually, if the limits of the error be diffusion rates of indium and phos-
alloys under these conditions. Dis- applied, they appear to be ap- phorus in stainless steel. When
cussion of this, however, will be proaching the same solidus temper- alloy A melts in contact with the
found in a later section. ature after diffusion at 2120° F base stainless steel, some dissolution
An exception to the premise that and, by inference, the same com- immediately occurs, the filler alloy
the filler alloy would attain equilib- position. Alloy N, however, ap- now contains some stainless steel
rium when encapsulated was alloy pears to be in an entirely different (mostly iron), and the solid solu-
G. The vapor from this alloy, the quaternary valley that probably bility of indium in the filler alloy
only one containing lithium, in- has. definite solid solubility to has been reduced. Furthermore,
variably etched the Vycor capsule. account for the gradual rise in the indium, being insoluble in iron, does
Even though this observation is solidus temperature with diffusion not diffuse into the base metal,
certain evidence of the volatilization temperature as postulated earlier. and the remelt temperature is not
of the lithium in the alloy, the ob- The relative solubilities of in- raised appreciably. In contrast to
jective of encapsulation—to inhibit dium§ and phosphorus in stainless this behavior, when alloy M (Ni-P)
volatilization—-was not accom- steel are manifested by the behavior melts in contact with the base stain-
plished. With the continual reac- of alloys A (Ni-In) and M (Ni-P). less steel, dissolution again occurs
tion of lithium vapors with the Although some solidus tempera- with pickup of stainless steel.
Vycor, the lithium melt-vapor equi- ture rise was recorded for A, Phosphorus now has solid solubility
librium was never attained. the rise is smaller than that of alloy in the filler alloy because of the
Table 2 indicates a greater in- M. Based upon the previous fun- presence of iron from the base
crease in remelt temperature with damental discussion, it should be metal; furthermore, phosphorus dif-
increasing brazing temperature. fuses out of the filler into the base
This is illustrated by the behavior of metal, in contrast to indium’s in-
alloys F, H, I, K and N. Even § No equilibrium diagram has been determined ability to do this. The net result is
for Fe-In although the insolubility of indium in
though all these alloys with the iron may be inferred. The Liquid-Metals Hand- an appreciable rise in remelt tem-
exception of K are Ni-Cr-Ge-In book’ states that “only superficial attack was perature, all of which fits the funda-
observed on an 18-8 stainless steel crucible used mental mechanisms previously dis-
alloys, this does not imply unique to hold indium at 1200° F for 21 days in contact
behavior on the part of this system; with the atmosphere. In similar tests, Monel cussed and the known phase dia-
the other alloys may have shown the and nickel crucibles were perforated in one day.” grams.
Iron has been suggested as best for possible use
same behavior, had they flowed with liquid indium Alloy K, (Fe-P), showed little

Table 3—Incipient Melting Temperature Changes and Calculated Composition of

Experimental Filler Alloys After Vacuum (0.1) Melting
First Second Loss, %
Vacuum volatile volatile lst 2nd -Solidus temp.,
melt element, element, Sample weight, gm vol. vol. Final Before, After,
temp., ° F % % Before After A element element composition’ +10° +10°
1920 39 In 61 Ni 0.6077 0.5275 0.0802 36. 26.2 In 1675 1700
2020 saa 0.6580 0.4451 0.2129 83. 9.8 In 1675 2275
2120 0.6058 0.4123 0.1935 81. 10.3 In 1675 2275
1920 9.2 In 9.2
Si 0.6598 0.6042 0.0556 91. 0.8 In-10.1Si 1885 1885
2020 0.6490 0.5932 0.0558 93. 0.6 In-10.1 Si 1885 2025
2120 0.6045 0.5434 0.0611 100. 0 In- 9.2Si 1885 2035
1920 4.7In 28.6Ge 0.6197 0.5831 0.0366 100 0 In-29.1Ge 1885 1885
2020 0.6044 0.5669 0.0375 100 0 In-28.8 Ge 1885 1885
2120 0.6028 0.5670 0.0358 100 0 In-29.1 Ge 1885 1885
1920 4.7 Li 47.6 Ge 0.6216 0.5776 0.0440 100 0 Li-48.6 Ge 1800 1775
2020 0.6422 0.5661 0.0761 100 0 Li-46.1Ge 1800 1775
2120 0.6099 0.5473 0.0626 100 ay 0 Li-46.9 Ge 1800 1925
1920 231n —«-15.4Ge 0.6141 0.4703 0.1438 100 ERONWNW
ON 0 In-19.5 Ge 1890 2010
2020 0.6354 0.4885 0.1469 100 0 In-20.0 Ge 1890 2025
DDDUD2120 0.6300 0.4747 0.1553 100 ~ 0 In-18.3 Ge 1890 2035
1920 16.61n 16.6Ge 0.6033 0.5023 0.1010 100 0 In-19.7Ge 1885 2010
2020 0.6073 0.4932 0.1141 100 0 In-17.8 Ge 1885 2025
DDD
Mwre 2120 0.6023 0.4914 0.1109 100 worewo0 In-18.1Ge
ow 1885 2010
MN 1920 10.5P 89.5Fe 0.5980 0.5918 0.0062 9.9 9.2P 1900 1910
2020 0.6144 0.6065 0.0079 12.2 9.3P 1900 1915
2120 0.6246 0.6080 0.0166 25.4 8.0P 1900 1935
1920 6P 94 Ni 0.6247 0.6196 0.0051 13.6 $.2 P 1600 1625
2020 0.6243 0.6168 0.0075 20.0 4.8 P 1600 1625
ZEZ=EAAATITIITIOS
2120
zDuymwmwrwD
wren
wr 0.5797 0.5642 0.0155 42.5 VEE
cooooof 3.4 P
oNnrasensenswoooocos 1600 1650
' First letter of code identifies alloy composition.
+’ Other elements in same proportion as original alloy to make balance.

454-s | OCTOBER 1960

tendency to a temperature rise due
to dissolution and diffusion. Since
the contribution of the rate of diffu-
sion of phosphorus through stainless
steel is the same for both alloys K
and M, (Ni-P), the difference in
behavior between the two alloys
must be due to the difference in
phase diagrams, specifically to the
cut through the filler-alloy composi-
tion and the base stainless steel.

Effect of Volatilization on the

Remelt Temperature
Table 3 lists the results of melt-
ing weighed filler alloys in alumina
crucibles, holding at 1920, 2020
and 2120° F under 0.14 vacuum for }
the same length oftime as in brazing
(15 min), followed by reweighing. seer: oe,
In the absence of the base metal, dis- 30 4c 5
w/O INDIUM
solution and diffusion were elimi- Fig. 4—The Ni-In equilibrium diagram with composition and
nated as factors in any compositional incipient melting point of experimental alloys
changes. The loss in weight of the
sample before vacuum melting and
after was considered to be due to the
loss of the most volatile element. ports the criterion suggested earlier Both alloys have lost all their in-
In some cases, the loss in weight for rise in melting temperature of a dium content and are left with
exceeded the amount of the most filler alloy; i.e., the constitutional substantially the same Ge content.
volatile element, and it was assumed system of the alloy, whether binary Similarly, their solidus tempera-
that the additional loss was due to or polynary, must have solid solu- tures are the same. This may be
the next most volatile element. bility for the volatile element prefer- directly attributable to the same
On the basis of these assumptions, ably terminating close to the eutectic Ge content, this element acting as
the final composition of the alloys composition, if brazing is to be done the temperature depressant. As
was calculated. To complement at the lowest possible temperature. the loss of indium resulted in a
the discussion of these results, one Even with solid solubility, if the small temperature rise, we may as-
simple equilibrium diagram has limit of solubility is very much sume that these quaternary alloys
been reproduced in Fig. 4, the Ni- different from the eutectic, too great have solid solubility for indium.
In binary system. Superimposed a loss of the volatile constituent Alloy F, containing 4.7% indium,
upon the diagram are the calculated may be necessary before a signifi- exhibited no rise upon total loss of
composition and incipient melting cant rise in the solidus temperature the volatile element. This indicates
temperature of alloys A, A-R1, is realized. that in this composition the indium
A-R2 and A-R3. From the dia- The case of solid solubility illus- is present as an insoluble phase.
gram, as postulated earlier, it can be trated in Fig. 1B has now been ac- Alloy C, which initially con-
seen that the percentage of indium tually demonstrated. An extreme tained 9.2% Si and 9.2% In, is an
in nickel must be reduced to at case of the opposite condition shown interesting illustration of the re-
least 25°% before an increase in the in Fig. 1A is demonstrated with the lationship between solubility and
incipient melting temperature can results shown by alloy M (Ni-6P) rise in remelt temperature. After
be realized. That the postulated in Table 3. Ni-P alloys fit the treatment at 1920° F, only 0.8 In
criterion is justified is demonstrated hypothetical case of Fig. 1A in remained in the alloy, yet no rise
by the experimental results in that no appreciable solid solubility in remelt temperature occurred,
Table 3 for alloys A-R1, 2 and 3. exists. Although 42.5% of the indicating that the limit of solid
After a vacuum treatment at 1920 phosphorus was volatilized out of solubility had not been reached.
F (alloy A-R2), the Ni-39%In alloy MR-3, leaving a Ni-3.4P After treatment at 2020° F, 0.6%
alloy lost, by volatilization, 36.7% alloy, little significant increase in In remained and the remelt tem-
of its indium; this left an alloy of the solidus temperature was real- perature rose 140°. Thus, the solid
73.8Ni 26.2In. The incipient ized. Indeed, nearly 100% of the solubility line for In in this alloy
melting temperature observed was phosphorus in the Ni-6P alloy is steep. Furthermore, comparing
1700 + 10° F. It may be seen from would have to be volatilized before C-R1 and C-R3, since the remelt
the diagram that the expected equi- a temperature rise would occur. temperature rise was equal in these
librium incipient-melting tempera- An intermediate case is that of alloys, at different Si contents (10.1
ture is 1667° F. On the other alloy K (Fe + 10.5 P). In this and 9.2%), 9.2% Si is above the
hand, alloys A-R1l and A-R3, vac- system some solubility does exist, solubility limit in this alloy.
uum treated at 2020 and 2120° F, but the solubility limit is low We have shown, to this point,
lost 83 and 81.9% of their indium, 2.8% P). It can be seen from data illustrating the changes in
respectively, to result in alloys of Table 3 that this limit was not remelt temperature of filler alloys
Ni-9.8 In and Ni-10.3 In. The reached even after the 2120° F due to dissolution and diffusion
incipient - melting temperature of treatment and, again as predicted, alone and due to volatilization alone.
these alloys was found to be 2275 little significant rise was realized. We will now discuss their combined
t 10° F. The proximity of these Alloys H and I, although of effects in brazed joints, in dynamic
temperatures to the hypothetical different initial composition, behave vacuum and in static helium, in
solidus line in Fig. 4 strongly sup- under vacuum in the same manner. systems of infinite volume.

WELDING RESEARCH SUPPLEMENT 455-s

 Table 4—Remelt Temperature of Filler Alloys After Furnace Brazing at 2120° F (304 stainless steel to 304 stainless steel)
Remelt temp., ° F Remelt temp.
Solidus <0.5,y, He + 0.5n, Helium, difference due
temp., vacuum vacuum, 15 min, Temp. change, ° F— to vacuum,
"Fy a b c a b c a minus c
1675 + 10 > 2354 2312+ 5 1829 + 59 >670 637 + 15 154 + 69 >+516
1885 + 10 2342 > 2349 2266 + 30 464 >451 381 + 40 +83
1885 + 10 2289 + 7 > 2359 2196 + 13 404 + 17 >474 311 + 23 +93 + 40
1800 + 10 2368 + 6 >2278 >2399 568 + 16 >478 >599 >—31+ 16
1890 + 10 2272 + 27 2269 + 24 2229 + 2 382 + 37 379 + 34 339 + 12 No signif. effect
1885 + 10 2339 + 21 > 2325 2241 + 26 454+ 31 > 440 356 + 36 +96 + 67
1900 + 10 2087 + 59 2083 + 10 2110 + 13 187 + 69 183 + 20 210 + 23 No signif. effect
1600 + 10 > 2307 > 2385 2148 + 7 >707 + 10 >785 + 10 548 + 17 >+159 + 27
1750 + 10
=-ZTO70>
2 2300 + 16 2336 + 6 2278 + 13 550 + 26 586 + 16 528 + 23 No signif. effect
» One reading only.
» Vacuum of 0.1 «w, under helium for 25 min.

Table 5—Remelt Temperature of Filler Alloys After Furnace Brazing at 2020° F (304 stainless steel to 304 stainless steel)
Solidu Ss —__——_——-Remelit temp., ° F Remelt temp. diff.
temp He + 0.54 H e+ 0.lu Temp. change, ° F— due to vacuum, ° F
Alloy F vacuum, a Helium,* b vacuum, c Helium,’d a b c d a minus b cminusd
1675 + 10 > 2190 2333 + 30 2362 + 25 2106 + 170 >515 + 10 658 + 40 687 + 35 431 + ? +256 + 215
1885 + 10 2221 53 2124 + 44 2 194 + 33 1912 + 19 336 + 63 239 + 55 309 + 43 No signif. +282 + 72
effect
1885 10 2222 + 46 2279 + 2231 t 44 2181 + 45 337 + 56 394 + 43 346 + 54 No signif. No signif.
effect effect
1800 10 2370 + 3 2337 +7 > 2239 2332 +7 570 + 13 537 +17 > 439 + 10 No signif. ?
effect
1890 + 2216 + 34 1935 5 + 2102 + 36 + 44 45+ 33 + 36 +272 +77 +217 + 82
1885 + 2231 + 5 2190 + 2328 + 16 + 66 + 50 t 10 No signif ?
effect
1900 + 1990 + 11 2010 + 2035 + 39 90 + 21 +19 + 43 No signif. No signif
effect
M 1600 + 2166 + 8° 1880 + » 2244 566 + 95 + 91 +286 + 186
N 1750 + 0.lu vac 2172 + 54 0. lu vac.: a-d:
2335 4 585 + 17 +163 + 81
15 min at brazing temperature
+’ Time at brazing temperature equal to total time of He + 0.1lyu vac. cycle.

Table 6—Remelt Temperature of Filler Alloys After Furnace Brazing at 1920° F (304 stainless steel to 304 stainless steel)
Remelt temp., ° F
Solidus Helium Remelt temp. difference due
temp., O.1u +0.1u Temp. change, | to vacuum, ° F
Alloy "Ff vacuum, a vacuum, b Helium,’ c a b Cc a minusc b minus c
1675 + 10 > 2500" > 2437 2224+ 117 >825 + 10 >762 + 10 >+276 + 137 >+232 + 137
1885 + 10 2122 + 38 2113 + 46 1913+ 21 237 + 48 228 + 56 28 +209 + 79 +200 + 87
1885 + 10 2282 + 78 2352 + 39 2138+ 83 397 + 88 467 + 49 No signif. +214 + 142
effect
1800 + 10 >2491° 22349 No joint formed >691 434
1890 + 10 2360 + 40 2391+ 4 > 2345 470+ 50 501 + 14
1885 + 10 2282 + 46 2073 + 76 > 2054 397 + 56 188 + 86 ? ?
1900 + 10 2082 + 12 1934 + 35 2041 + 7 182 + 22 34+ 44 +41 + 39 —107 + 61
1600 + 10 2348 + 24 2277 + 16 No joint formed 748 + 34 677 + 26
1750 + 10 2299 + 10 2246 + 17 2290 + 19 449 + 20 496 + 27 91 + 49 No signif.
effect
2 One sample only.
» Time at brazing temperature equal to total time of He + 0.1u vac. cycle.

Table 7—Remelt Temperature of Filler Alloys After Furnace Brazing at 1820° F (304 stainless steel to 304 stainless steel)
Remelt temp., ° F
Helium
Solidus 0.1u +0 lu Remelt temp. difference
temp., vacuum, vacuum, Helium,’ Temp. change, ° F due to vacuum, ° F
il a b c a b Cc a minusc b minus c
1675 + 10 1737° No joint 1845 + 46 62 + 10 170 + 56 —108 + 66
formed
1800 + No joint 2256 + 43 2273 + 39 456 + 53 473 + 49 No signif.
formed effect
1600 2250 + 20 2397° No joint 650 + 30 797 + 10
formed
N 1750 + 2133 + 21 2162 + 70 1870 + 22 383 + 31 412 + 80 +263 + 63 +292 + 112
* One sample only.
» Time at brazing temperature equal to total time of He + 0.1s vac . cycle.

456-s | OCTOBER 1960

 Results of Brazing in an Infinite Volume
Tables 4 through 7 list remelt
temperatures of the experimental
filler alloys after brazing in vacuum,
in helium followed by vacuum, and
in 5 psig static helium only. As
pointed out before, the brazing sys-
tem used may be referred to as A- WIRED T-SPE
PREPARED F
having infinite volume in that equi- POWDER BEFORE
librium was never achieved in
any of the process variations in-
vestigated. Vapor - metal atoms
were continuously withdrawn, either
by vacuum pumping or by conden-
sation at the cold ends of the retort
in the case of the static helium.
Each table presents the remelt
temperatures of specimens brazed at
one temperature, 2120, 2020, 1920
and 1820° F in Tables 4, 5, 6 and 7,
respectively. Through the course Fig. 5—Evaiuation of wetting and flow properties
of the experimental program some
slight variations were introduced to
the brazing cycle. Originally, spec- remelt temperature of these alloys in its dependence on volatilization
imens were brazed at a maximum was principally by the dissolution- in raising remelt temperature. Re-
pressure of 0.54 Hg. With refine- diffusion mechanism, and that the ferring now to Tables 4 through 6
ment of techniques, this maximum difference in behavior of the Fe-P and to the final columns, which
pressure limit for vacuum brazing and the Ni-P binary alloys is due to show how much higher the remelt
was fixed at 0.lu Hg. A second the solubility-diffusion characteris- temperature is with the vacuum
variation was the time at tempera- tics discussed in detail previously. processes than with the helium
ture under helium. Specimens The rise in remelt temperature processes, alloy A consistently ex-
brazed at 2120° F were main- will now be discussed in terms of the hibited a difference. For this alloy
tained under vacuum at tempera- operating mechanisms and the effects it may be concluded that a vacuum
ture for 15 min. Comparison speci- of time, temperature and vacuum is more efficient in removing the
mens under helium were held also versus helium at 5 psig pressure. metal vapors than is the helium
for 15 min. The process involving Effect of Time on Remelt Tempera- system. For the remaining alloys,
helium for 15 min plus the time for ture. Although time is a factor in in accordance with their lesser
pump-down to 0.lu Hg required influencing the change in composi- response to volatilization shown in
25 to 35 min total. Accordingly, tion, it should be clearly re-empha- Table 3, the data are not sufficiently
experiments were conducted in static sized that even large compositional reproducible to show any distinct
helium with 25 to 35 min at tem- changes are not necessarily reflected advantage of vacuum brazing.
perature for comparison purposes. in the remelt temperature. Effect of Brazing Temperature on
Reference to Tables 4 through 7 Tables 4 through 7 indicate gen- Remelt Temperature. Theoretically,
shows that remelt temperature was erally that a difference in time be- one would expect an increase in
raised by all the brazing processes tween 15 and 30 min does not remelt temperature with an increase
investigated. affect remelt temperature. Al- in brazing temperature. Both the
The greatest increases in remelt though the data are not conclusive vapor pressure and the diffusion
temperature were shown by alloys in this regard, the general observa- coefficient increase exponentially
A, G, M and N. Alloy M is par- tion can be made by comparing with temperature, and this should
ticularly outstanding in that it has columns a and b of Table 4, columns result in more rapid change in
the lowest initial solidus tempera- b and d of Table 5, columns a and filler-alloy composition by either
ture, 1600° F, and joints were b of Table 6 and columns a and b mechanism, dissolution-diffusion or
obtained with vacuum brazing at of Table 7. The justification for volatilization. As long as equilib-
the lowest brazing temperature uti- drawing conclusions regarding the rium is never reached, as indeed it
lized, 1820° F. effect of time by comparing data was not, then a faster rate of change
The alloy exhibiting the smallest from brazements made in vacuum in composition should result in a
increase in remelt temperature is to brazements made in _ helium higher remelt temperature in a
alloy K, Fe —-10.5% P. This alloy plus vacuum is the knowledge, to be given time. The reproducibility of
also did not respond to the treat- discussed later, that volatilization the data, however, is inadequate to
ments investigating the individual occurs in helium. prove this hypothesis.
effects of dissolution-diffusion or Effect of Vacuum Versus Helium at
5 Psig Pressure on Remelt Tempera- Brazing Characteristics
volatilization, as shown in Tables
2 and 3. The comparison between ture. The relative effects of brazing So far, only the remelt tempera-
the performance of alloy K, exhib- in a vacuum or in helium at 5 psig tures of the filler alloys have been
slightly greater than atmospheric discussed. Inthissection, there will
iting the smallest rise in remelt
pressure to avoid contamination in be described another very important
temperature, and alloy M, among the case of leaks) on remelt tempera- facet of the filler alloy—its ability
those exhibiting the greatest rise, ture would be expected to be to flow and make a sound joint.
is interesting since they are both greatest for those alloys which Filler alloys were purposely pre-
binary phosphorus alloys. This owe their increase in remelt tem- placed along the capillary rather
bears out the observations from perature primarily to volatilization. than at one end of the T-specimen
Tables 2 and 3—that the rise in From Table 3, alloy A is outstanding in an effort to eliminate variations

WELDING RESEARCH SUPPLEMENT | 457-s

 in composition from sample to The dull to black surface appear- tom of the ““T”’ in contact with the
sample cut from the same T-speci- ance of the stainless-steel base metal retort boat was invariably bright,
men. This procedure tended to found after brazing naturally raised as was side B of the T-specimen (see
obscure the flow properties of the questions as to the quality of the Fig. 5) if the alloy did not flow.
filler alloy in the sense of their vacuum and the helium gas used. Although the degree of discolora-
usual definition, i.e., length of flow Accordingly, as an atmosphere tion of these specimens can, with
along the ““T.” It was recognized check, a blank T-specimen was run certainty, be assumed to be related
early in the study, however, that after every brazing cycle under the to the degree of volatilization, no
the atmosphere cycle had an effect same conditions as the _braz- quantitative relationship could be
on the wetting and flow properties; ing cycle. All blank specimens ascertained. Itisa safe conclusion,
and, because of this, many T-speci- emerged from the furnace retort as however, that in practical brazing
mens did not yield a full number bright and shiny as they were prior processes, discoloration of parts
of samples for remelt temperature to heat treatment. The atmos- may not result from moisture in the
investigations. This is evidenced pheres used, therefore, could no atmosphere as is usually presumed
by an occasional asterisk in many longer be suspect. In the presence but, rather, is due to volatilization
of the preceding tables of data of the filler alloy, however, the color of constituents in the filler alloy.
indicating that the data were from of the base metal varied from Table 9 has been prepared to aid
one sample only. shiny through green to black. Ap- in the analysis of the results re-
Throughout this investigation, parently, the discoloration or surface corded in Table 8. Although Table
careful notes were made concerning contamination originates in the 8 may be independently used to
the appearance of all T-specimens filler alloy and could be a result of arrive at the conclusions discussed
after brazing. Table 8 represents a the volatilization process itself. below, Table 9 has been topically
full compilation of all the observa- Furthermore, in many cases, the separated into the two major phe-
tions made. The wetting and flow contaminated layer was quite nomenological variants—namely,
observations were based on an thick and flaky and could be scraped variations in brazing characteristics
empirical scheme illustrated in Fig. off, leaving a shiny stainless sub- from (a) helium in retort to helium
5. Other observations as percent surface. Another significant ob- confined and (6) vacuum in retort
skull, color and erosion are self- servation was the location of the to helium in retort. Under these
explanatory. surface contaminant(s). The bot- conditions, wettability, flow, skull

Table 8—Brazing Characteristics of Experimental Filler Alloys

a ae EEE
Alloy no. & He + Helium Helium + Helium + Helium-15
composition 0.lu vacuum 0.54 vac 15 min Helium-encap 0.54 vac. 0.1u vac. min Helium®*
A Wet Wet (0°) Wet Some wetting Wet Wet Wet Wet
61Ni- 391n 109% flow 100% flow 100% flow 50% flow No flow 100% flow 100% flow 103% flow
No skull No skull No skull No skull No skull No skull No skull No skull
Shiny Dull Dull Green Green Green Green-black
Gassy fillet Gassy fillet
Cc Some wetting Wet (0°) Wet Wet Partial wetting Wet Wet Wet
64.6Ni-17Cr - No flow 109% flow 100% flow 100% flow No flow 100% flow 30% flow No flow
9.2In -9.2Si No skull No skull 5% skull No skull 5% skull No skull No skull No skull
Shiny Shiny Green-black Shiny Green Dull Green-black Green
F Wet Wet Wet Some wetting Wet Wet Wet Wet
38. 1Ni - 28.6Cr - 100% flow 100% flow 100% flow 100% flow 100% flow 100% flow No flow No flow
4.7 in-28.6Ge No skull No skull No skull 10% skull No skull No skull No skull No skull
Green Green Green Dull-green Dull-green Dull Green Green-black
Erosion
G Wet Wet Wet Wet Wet Wet Wet Wet
28.6Ni-19.1Cr - 100% flow 100% flow 109% flow 100% flow 100% flow 100% flow 100% flow 100% flow
4.7Li-47.6Ge 5% skull No skull No skull 35% skull 30% skull 20% skull 20% skull 10% skull
Dull Dull Shiny Dull Dull Dull Dull
Erosion Slight erosion Heavy erosion Slight erosion Slight erosion Slight erosion Slight erosion
H Wet Wet (0°) Wet Wet Wet Wet Wet Wet
30.8Ni — 30.8Cr - 100% fiow 100% flow 100% flow 100% flow 90% flow 95% flow 100% flow 109% flow
23.0In -15.4Ge 10% skull No skull 10% skull 20% skull 50% skull 50% skull 20% skull 20% skull
Green Dull Green Dull Green-black Dull Green Green
Slight erosion Gassy fillet Gassy fillet
| Wet Wet Some wetting Wet Wet Wet Wet Wet
33. 3Ni — 33.3Cr- 100% flow 100% flow No flow 80% flow 75% flow 90% flow 80% flow 100% flow
16.61n - 16.6Ge 20% skull No skull 10% skull 50% skull 20% skull No skull 40% skull 30% skull
Shiny Green Green Dull Green Dull Green-black Green-black
Gassy fillet
K Wet Wet Wet Wet Wet Wet Wet Wet
89.5Fe-10.5P 100% flow 100% flow 100% flow 100% flow 100% flow 100% flow 103% flow 100% flow
No skull 20% skull No skull 65% skull 20% skull 5% skull 20% skull No skull
Dull Shiny Green Dull Dull-green Shiny Green Dull green
Erosion
Wet Wet Wet Some wetting Wet Wet Wet Wet
100% flow 80% flow 100% flow 40% flow 100% flow 100% fiow No flow Negl. flow
No skull No skull No skull No skull No skull No skull No skull No skull
Dull Green Black Green Green Dull Green Green
Erosion Erosion Erosion Erosion
N Wet (0°) Wet Wet Wet Wet (0°) Wet
35Ni — 24Cr - 109% flow 100% flow 65% flow 100% flow 100% flow 95% flow
261n -15Ge 20% skull 20% skull
Green Green Green Dull-bright Green Green

* Brazing time comparable to helium + 0.14 vacuum cycle time.

458s | OCTOBER 1960

remaining and color changes have capusle; and at Zim° &F, there difference in rate of removal of
been recorded as increasing, decreas- was no difference 100% flow Ooc- metal vapors, or that the tempera-
ing or remaining the same. curred in both cases. ture is above the liquidus in both
As an example of the use of Table The data in Tables 8 and 9 can cases. Any difference in skull for-
9, consider alloy G. The behavior be utilized to draw conclusions mation in the two processes neces-
while brazing in helium in the concerning the relative efficiencies sarily indicates a difference in rate
infinite volume retort and behavior of removing volatilized metal vapors of removal of metal vapors. The
while brazing in confined helium, when brazing in vacuum or in data of Tables 8 and 9 indicate a
i.e., in the small sealed capsules, helium at atmospheric pressure. difference in skull formation between
may be compared. At 2120° F, This may be done by considering the two processes, from which it
color change decreased a_ small amount of skull formation and color may be concluded that vacuum
amount, wetting and flow were the change. It was pointed out pre- brazing was more efficient in re-
same, and amount of skull increased viously that, in complex or binary moving metal vapors.
appreciably. At 2020° F, wetting alloys with limited solubility, a Now, let us apply the same analy-
decreased a small amount, flow change in composition does not sis to color change. A process
decreased appreciably, color change necessarily cause a change in solidus which is very efficient in removing
remained the same and skull again temperature. The liquidus tem- metal vapors allows little time for
increased appreciably. At 1920° F, perature is, however, directly in- reaction with the specimen surface;
wetting decreased appreciably, and fluenced even by slight composi- thus, the color change should be
flow, skull and color change re- tional variations. Depending on slight or, at any rate, less than that
mained the same. In other words, the composition of the eutectic in for the less efficient process. The
considering the effect of tempera- the system, the liquidus may in- data of Tables 8 and 9 show that
ture on the relative flow of alloy G crease or decrease due to compo- the specimen surfaces were darker
in the helium infinite-volume retort sitional changes. Thus, a compari- after helium brazing than after
versus helium confined, at 1920° F son of skull formation in the two vacuum brazing, once again con-
there was no difference between brazing processes indicates the effi- firming more efficient removal of
the two processes (from Table 8 ciency of removal of metal vapors. metal vapors during vacuum braz-
it is seen that no flow occurred in If both processes yield the same ing.
either case); 2020 F, flow was amount of skull formation, this Generalizations regarding wetting
better in the retort than in the neans that either there is no and flow are difficult to draw when

on 304 Stainless Steel T-specimens Under Various Brazing Conditions

F — eens BIG? Fo
Helium + Helium +
Helium-encap 0.1 vacuun 0.lw vac Helium Heliun encap lu vacuum 0.lu vac. Heliun.® Helium-encap?
De-wet Wet Wet (0°) Wet De-wet some wetting Some wetting Some wetting
No flow 95% flow 100% flow 80% flow No flow No flow No flow 30% flow
No skull No skull No skull No skull No skull No skull No skull No skul
Black Green-black Green Black Green Green-black

Wet (0°) Wet Wet Wet Some wetting

100% flow 60% flow 5% flow Neg. flow No flow
No skull 40% skull No skull No skull No skul
Shiny Dull Green-black Black
Wet Wet Wet Wet Wet
15% flow 100% flow 100% flow No flow 90% flow
10% skull 5% skull 10% skull 5% skull 5% skul
Dull Shiny Dull Green Dull

Some wetting Some wetting Some wetting Some wetting De-wet Wet (0°) Wet (0°)
No fiow 5% flow 5% flow No flow No flow 10% flow 100% flow
90% skull 90% skull 10% skull 90% skull 0% skull 50% skull 10% skull
Dull Dull Dull Dull Dull Dull-green
Erosion Erosion
Wet Wet Wet Wet Wet
100% flow 50% flow 100% flow 100% flow 100% flow
20% skull 60% skull 70% skull 70% skull 80% skull
Dull Dull Dull-green Green Dull

Wet Wet Wet Wet Wet

100% flow 75% flow 50% flow 90% flow 100% flow
30% skull 70% skull 70% skull 5% skull 40% skull
Dull Dull Green-black Green Dull

Wet Wet Wet Some wetting

50% flow 100% flow No flow 100% flow No melting
10% skuli 40% skull 40% skull 30% skull
Green-black Green-black Green-black Green Dull-green

De-wetted Wet Wetted De-wet De-wet Wet De-wet De-wet

No flow 100% flow 100% flow No flow No flow 100% flow No flow No flow
No skull No skull No skull No skull No skull No skull No skull No skull
Black Black Dull Black Black Dull-green Green Green-black
Erosion
Wet Wet (0°) Wet Wet Wet Wet Wet Wet Wet
100% flow 100% flow 90% fiow 95% flow 100% flow 80% flow 66% flow 50% flow 100% flow
20% skull 20% skull
Dull-bright Green Green Green Dull-bright Dull-green Green Green Dull-bright

WELDING RESEARCH SUPPLEMENT | 459-s

Table 9—Analysis* of Wetting, Flow, Degree of Surface Contamination and Skull Formation Change Due to Brazing Cycle
2120° F . 2020° F 1920° F 1820° F ~
Alloy Decrease Same Incr. Decrease Same _ Incr. Decrease Same Incr. Decrease Same incr.
From helium in retort to helium confined
< S C* WF Ss
oi WF Th Ss
2nm xe oC?" |ie wS
WF Ss
WF _ WF
WFS Ww
WF S
+nmoO S ace
Z2ex~-IrQONOPS
Os0200NSsw Ww ,* SV%SEE9OSNN7,
ST7ecee
Ww
= ro} 3 vacuum to helium in retort
WFS Tbe n oO WSs
WFS**C No comparative n ”4 Ww
WFSC conditions F WSs
WFC Fre WS
WFSC — Ww
asa C S Ww
WFS Cc $**Ct+ WF
WFS Cc WF Sc
IA
ZzSex~-
NODS + +F + Cc wWrtFt+ C¢
® Key: W = wetting, F = flow, C = color, S = skull, * = small change, ** = very small change.

comparing vacuum to helium. Spe- the encapsulated specimens is finite ever, its flow is retarded at 2120° F
cifically, alloys H, I and N, all of rather than infinite as in the retort, and eliminated entirely below this
similar chemistry (containing in- the concentration of metal vapors is temperature. In these cases of
dium), appear to have better flow higher. Furthermore, the speci- poor flow when encapsulated, the
under helium than in vacuum at men surfaces are completely envel- surface of the stainless steel was
lower temperatures. Another in- oped by the metal vapors. In generally green to black. It has
dium alloy, A, flowed better under this connection, it is interesting already been established that this
helium than in vacuum at 1820° F. that the color of the encapsulated is not oxidation but contamination
The indium alloys generally flowed specimens was uniform over all by the metallic-vapor phase. The
well at the higher temperatures in surfaces, whereas, in the retort, presence of germanium and chro-
the vacuum and the helium infinite- color changes only occurred on mium in the Ni-In alloy results in
volume processes. From Table 3, specimen surfaces directly exposed quite a different effect. These
recall that the indium alloys showed to filler alloy. alloys tend to improve their wetting
the greatest rise in remelt tempera- The effect of metallic vapors on and flow properties or at least
tures due to volatilization. In view interfacial tension has been investi- remain the same when encapsulated.
of the greater efficiency of metal gated. Generally, a foreign metallic Alloy N not only has improved flow
vapor removal by vacuum than by vapor-solid interfacial tension is but results in a brighter T-specimen
helium, the observation that in- greatly reduced from that of an when encapsulated.
dium alloys flow better in helium inert-gas-solid interfacial tension. Some of the high-indium alloys
than in vacuum at lower tempera- In an equilibrium wetting situation, exhibited definite signs of boiling,
tures is not surprising, regardless the y.—-. interfacial tension is referred to in Table 8 as gassy
of the exact mechanism by which usually the largest vector. Reduc- fillets. Alloy A, containing 39% In,
indium promotes flow. tion of this vector by the substitu- which showed the _ greatest in-
Alloy G, which contains lithium tion of metallic vapor for inert crease in remelt temperature due to
as the volatile element, behaved vapor could, theoretically, result in volatilization, illustrates the effects
similarly to the indium alloys in the transformation of a wetting of temperature and degree of
that flow was superior in helium situation to a nonwetting. The vacuum on porosity in the fillets.
to vacuum at the lower tempera- wetting of the encapsulated speci- Porosity was observed at 2120
tures. Apparently, lithium behaves mens by the filler alloy, there- and 2020° F, which corresponds
similarly to indium in its funda- fore, should be less than in the to the temperatures of greatest
mental role of affecting flow. retort. Examination of Table 9 indium loss (from Table 3, over
Alloy M, Ni-P, in contrast to the reveals this to be true in some cases. 80% loss of indium). No porosity
indium alloys, flowed better under In others, there is no difference. was observed at 1920° F where,
vacuum than in helium at the lower Of the phosphorus alloys, K and M, from Table 3, a loss of indium of
temperatures. This alloy, at these the Ni-P flow was generally un- 37% was noted. This bears out the
temperatures, lost less than a quarter affected (this alloy flows poorly dependency of volatilization losses
of its phosphorus through volatili- under infinite helium) whereas the on diffusion rate of solute atoms
zation (Table 3). Its behavior Fe-P flow was affected adversely through the molten-filler alloy. The
cannot be rationalized as easily as under helium encapsulation. agitation of boiling makes the
that of the indium or lithium alloys. An outstanding case of the effect process less dependent on diffusion.
Let us now discuss the wetting of vapors on the surface tension From Table 8, it is seen that
and flow properties during brazing forces is exhibited with the Ni-In porosity was obtained only at
in the confining capsules as com- alloy, A. This alloy generally vacuums of 0.1u, and not at 0.5u.
pared to infinite-volume helium flows well in the infinite-volume This dependence of boiling on
brazing. Because the volume in brazements. Encapsulated, how- pressure is expected.

460-s | OCTOBER 1960

 Of all the alloys investigated, the liquid filler alloy to the surface, f. When metal vapors are present,
alloy G most consistently showed and the _ surface-to-volume ratio discoloration, wetting, flow and
maximum erosion of the base metal; are rate controlling. Boiling under skull formation are functions of the
this was, as expected, a function of vacuum minimizes the dependence rate and quantity of metal vapor
temperature. From Table 2, it is on diffusion rate, but causes porosity removal.
seen that this alloy also showed one in the joint. g. With indium-bearing alloys,
of the highest increases in remelt 4. In regard to volatilization better flow properties are obtained
temperature in the absence of alone, indium-bearing nickel-base at lower temperatures in helium
volatilization, although it has been alloys showed the most promise. brazing than in vacuum brazing.
stressed before that compositional Of these alloys investigated, ranging This is due to the rapid loss of
changes are not necessarily reflected from a binary containing 61% indium under vacuum, with the
in remelt temperature increases. Ni-—39% In (solidus temperature, attendant loss of its contribution to
1675° F) to Ni-Cr-In-Ge quater- alloy flow. The same is true of the
Summary and Conclusions naries (solidus temperatures from lithium-bearing alloy.
1. Experimental brazing filler 1750 to 1890° F), the binary ex- h. In addition to the 61% Ni-
alloys have been developed con- hibited the greatest remelt tempera- 39% In alloy as a usable filler alloy
taining temperature depressants ture rise, as well as the highest resulting from this program, the
which have been volatilized during remelt temperature. This alloy 35% Ni-24% Cr-—26% In-15%
the brazing of stainless steel, leaving could be considered a practical al- Ge quaternary (alloy N) is also
joints of high remelt temperature. loy for high-temperature usage, and attractive. It has a low solidus
The mechanisms by which remelt presumably chromium could be temperature, 1750° F, an apparently
temperature is increased have been added to impart oxidation resist- low liquidus temperature as indi-
studied, for the range of alloys ance without affecting the above cated by the small skull formation,
investigated, containing Ni, In, considerations. even at a brazing temperature of
Cr, Ge, Fe, Li and P. The two 5. In regard to the rise in remelt 1820° F, and flowed and wet well
main mechanisms are (a) dissolution temperature due to dissolution of under all conditions. Its remelt
of the base metal in the filler metal the base metal and diffusion into temperature rise was consistently
and diffusion of constituents of the the base metal, in the absence of high with either vacuum or helium
filler metal into the base metal, and volatilization the alloys yielding the brazing, with the exception that at
(6) volatilization of constituents of greatest rise in remelt temperature a temperature of 1820° F vacuum
the filler metal. These have been were a 94% Ni-6% P binary and brazing yielded higher remelt tem-
studied individually, as well as in Ni-In-Cr-Ge quaternaries. An peratures than helium brazing. In
combination, during brazing in 89.5% Fe-—10.5% P binary alloy regard to the fact that the oxidation
various types of cycles. exhibited the least rise in remelt resistance of the joint obtained
2. It has been shown theoretically temperature. The difference in using the 61% Ni-— 39% In binary
and experimentally that in order behavior between the Ni-P and might be impaired by the 10% In
for remelt temperature to increase Fe-P binaries was explained on the remaining after brazing, additional
either by the dissolution-diffusion basis of phase diagrams and dif- time or more efficient removal of
mechanism or by volatilization, fusion rates. indium vapors during brazing would
there must be solid solubility of the 6. In regard to the practical solve this problem.
diffusing or volatilizing element in brazing experiments which were
the filler alloy. Secondly, although carried out in vacuum, in helium Acknowledgments
compositional changes may result followed by evacuation, and in
from the operation of either mech- helium at 5 psig, the following This work was carried out under
anism, a change in remelt tempera- conclusions were reached: Contract No. AF33(616)-5654,
ture (solidus temperature) does not a. The remelt temperature of all sponsored by the Wright Air De-
necessarily occur. The degree to the alloys investigated increased velopment Center with Lt. George
which a compositional change affects appreciably during all of the braz- Haley and Major E. M. Kennedy
remelt temperature by either mech- ing cycles studied, except 89.5° acting as monitors. ‘The authors
anism is proportional to the solid Fe—10.5% P, which responded are grateful to the Air Force for
solubility. little. permission to publish these findings,
3. In regard to the mechanisms b. Because of the greater effi- and to Officers Haley and Kennedy
by which volatilization can be con- ciency of vacuum brazing versus for their advice and assistance.
trolled during practical brazing, helium brazing in removing metal
four methods are available for vapors, those alloys especially re- Bibliography
removing volatilized metal vapors: sponsive to remelt temperature rise 1. Canonico, D. A and Schwartzbart, H
vacuum pumping, entrainment in a through volatilization (particularly Development of Oxidation and Liquid Sodium
Resistant Brazing Alloys WAD¢ Technical
flowing gas (protective or reducing), the 61% Ni- 39% In alloy) showed Report No. 57-648, September 19
condensation in a cold portion of greater rise in remelt temperature 2. Bredzs, N., and Schwartzbart, H., “‘Metal-
lurgy of Bonding in Brazed Joints, Part I
the retort or reaction with another under vacuum than in helium braz- THE WELDING JOURNAL, 37 (11 Research
material placed in the furnace which ing. Suppl., 493-s to 498-s (1958
will not itself adversely affect the 3. Bredzs, N., and Schwartzbart, H., “‘Metal-
c. A difference between 15 and lurgy of Bonding in Brazed Joints, Part II,”
brazing process. In the present 30 min at brazing temperature did Ibid., 38 (8), Research Suppl., 305-s to 314-s
program, metal vapors were re- not affect remelt temperature. 1959).
4. Smithells, C. J Metals Reference Book
moved by vacuum pumping or by d. The volatile elements have a Interscience Publishers, Inc., Ne York, p. 613
condensation at the cold ends of a direct effect on the wetting and flow 1955
5. Lange, N. A., Handbook of Chemis Fifth
retort containing helium at 5 psig properties of the filler alloys through Ed., Handbook Publishers, Inc Sandusky,
pressure. It has been shown by their influence on surface tension Ohio, p. 1661 (1944
several means that vacuum pumping relationships. 6. Dushman, S Scientific F oundations of
Vacuum Technique, John Wiley & Sons, Inc.,
was the more efficient. In the e. The presence of metal vapors New York, pp. 764-778 (1949
absence of boiling, the diffusion rate in the retort affects the color of 7. Liquid-Metals Handbook, NAVEXOS]P-73:
rev.), AEC, Dept. of Navy, Washington, D. C
of the volatilizing element through the base metal. p. 172 (June 1953

WELDING RESEARCH SUPPLEMENT 461-s

 Proposed Procedure for Testing

Shear Strength of Brazed Joints

Paper by F. M. Miller and R. L. Peaslee was published

in the April 1958 issue of THE WELDING JOURNAL
Research Suppl., 144-sto 150-s.

Discussion
BY W LEHRER AND H. SCHWARTZBART

The authors are to be congratulated standard test or tests: (1) to provide well as to the design of brazed
for their good _ intentions in a common basis for comparing honeycomb-sandwich panel-wing
“grabbing the bull by the horns.” brazing alloys and procedures among sections. We think not. How-
They have made a start in an area different organizations, and (2) to ever, be that as it may, we agree
in which all brazing research people provide data to be used by design wholeheartedly that some stand-
have felt a need, and which has re- engineers in designing brazed struc- ardization of testing procedures is
ceived less attention than it deserves tures. Any one of several arbitrary highly desirable.
from those others of us who have tests may fill the first need. The This brings us to the main reason
been rendered somewhat immobile final choice will be based on simplic- for this discussion. We propose an
by our realization of the difficulty of ity, economy and reproducibility, extension of the Miller-Peaslee test
the problem. In this regard, it and it is in the area of reproduci- which we feel will greatly increase its
doesn’t much matter whether the bility that more evaluation needs to value. Miller and Peaslee propose
test proposed is accepted as the be done on the proposed test as well that the test be carried out at a
standard test or not; if it generates as on other available tests. In re- given single overlap so that failure
further activity toward some stand- gard to the need for data to be used will always occur by shear through
ardization of testing procedures it in the design of brazed structures, it the brazed joint. We feel that this
has accomplished a worth while ob- may still be questioned whether a procedure would not adequately
jective. single test can provide meaningful describe the characteristics of the
As the authors point out, there data of universal application to all brazed joint. It would be much
are two reasons for developing a brazed structures. In other words, more descriptive to present the
can the same test provide meaning- characteristics of a joint in a given
W. LEHRER and H. SCHWARTZBART are ful data to be used to design a sleeve base metal, brazed with a given
associated with Armour Research Foundation
Chicago, Til. type of joint loaded in tension, as filler metal with a given brazing

120

T A-AVERAGE SHEAR STRESS IN FILLER METAL 10 A-AVERAGE SHEAR STRESS IN FILLER METAL
B -AVERAGE TENSILE STRESS IN BASE METAL B - AVERAGE TENSILE STRESS N BASE METAL
100

90}

T FAILURE | FAILURE 8 Y TENSION FAILURE BY SHEAR

BY SHEAR | THROUGH BASE METAL THROUGH FILL ER METAL
T THROUGH
BASE
METAL
T FAILURE 1073
Psi
x
SHEAR
BY
THROUGH
METAL
FILLER

a= = ; a *
5 6 7 8 9 1.0 5 6 7
VERLAP, NCHES OVERLAP, INCHES
Fig. Dl—Strength of brazed joints. Copp er Fig. D2—Strength of brazed joints. Mild-steel
base metal, BAgla filler metal base metal: BAgla filler metal

462-s | OCTOBER 1960

 A -AVERAGE SHEAR STRESS IN FILLER METAL
B -AVERAGE TENSILE STRESS IN BASE METAL

FAILURE ALWAYS BY SHEAR

THROUGH THE FILLER METAL

x x
Fig. D4a—Copper brazed with BAgla
|

i ee
*
Fig. D3—Strength of brazed joints. 4130 steel base
| metal BAgia filler metal

0 aE = qxctAlemeeetl EE a |
8) ! 2 3 eee
4 5 6
IVERLAP, INCHES Fig. DS—Mild steel brazed with BAgla
procedure, as curves of average
shear stress in the brazed joint and
average tensile stress in the bas2
material as functions of overlap dis-
tance. Describing the properties of
the brazed joint by the strength at a
single value of overlap distance is
considered comparable to describing
the fatigue properties of a metal by
citing only one point on the S-n
curve. This is not done; the entire
S-n curve is plotted to describe
fatigue properties.
To illustrate these ideas, a series
of copper, mild steel and 4130 steel
specimens were torch brazed with
BAgla filler metal and evaluated
using the Miller-Peaslee test. The
sheet thickness was 0.125 in. and base metal and the tensile stresses overlaps would still result in rela-
overlap distance was varied from plotted are average. tively low-strength joints with fail-
0.031 to 1 in. Figures D2 and D5 present the ure in shear of the filler metal be-
The results are presented graphi- data for mild steel brazed with cause of the stress gradient within
cally in Figs. D1 through D3, and BAgla. In this case, failure changes the filler metal. Considering the
Figs. D4 through D6 are photo- from shear through the filler metal wide range in bulk mechanical prop-
graphs of the fractured specimens. to tension through the base metal at erties, of base and filler metals, that
Consider first the behavior of an overlap of about 0.7 in. are encountered in brazing, it is felt
copper brazed with BAgla (Figs. As Figs. D3 and D6 show, 4130 that the sort of representation util-
D1 and D4). At very low overlap steel brazed with BAgla fails by ized in Figs. D1 to D3 is more in-
distances, less than 0.05 in., failure shear through the filler metal at all formative than reporting the strength
occurred by shear through the filler overlap distances up to 1 in. at a single value in answering
metal. At overlap distances be- It is apparent that joint strength the questions of the average shear
tween 0.05 and 0.275 in., failure and the location of failure are func- and tensile stresses at failure of a
occurred by shear through the cop- tions of the bulk strength and joint with a given overlap, and the
per, and above 0.275 in. by tension ductility of the filler metal, and the overlap distance required to yield a
through the copper. Although we bulk strength and ductility of the joint of 100% joint efficiency.
use the words “shear” and ‘“‘ten- base metal. When brazing copper At first appraisal, it may appear
sion” it can’t be emphasized too with BAgla, extremely small over- that more specimens are required to
strongly that the shear is not pure laps would be required if the base determine the curves of strength vs.
shear and the tension is not simple, metal were not to be loaded to ' overlap than to determine the
uniaxial tension. The stress and its yield point, as is recommended in strength at one overlap. In
strain states are nonuniform through the paper. On the other hand, in practice, however, the preparation
the filler metal, and the shear brazing a ductile base metal with a of the curves may well require fewer
stresses plotted are average. Simi- filler metal of comparable strength specimens. If the standard test is
larly, bending stresses exist in the but practically no ductility, large to consist of a determination of the

WELDING RESEARCH SUPPLEMENT 463-s

 men is required at each overlap dis-
tance; the averaging process con-
sists of drawing the curve.
Furthermore, Miller and Peaslee
indicate that the proper overlap
should be of a distance so that the
maximum stress applied to fracture
the brazed joint should be equal to
or less than '/, the yield strength of
the base metal. With a new brazing
filler alloy, numerous specimens
may very well be required to deter-
mine this optimum overlap prior to
obtaining the desired average shear
stress.
It appears, therefore, that the plot
Fig. D6é—4130 steel brazed with BAgla of the strength vs. overlap curves is
not only more informative but in the
last analysis, involves less time and
strength at a single overlap dis- to allow for experimental deviations. effort than the determination of a
tance, several specimens will have to In determining the curves of single average shear stress of a
be tested and an average computed strength vs. overlap, only one speci- filler alloy.

Authors’ Closure

The authors wish to thank Messrs. capacity of that particular joint de- In the case of number one, the
Lehrer and Schwartzbart for their sign. shear value is well above the base-
comments and recognition of the The authors would still like to use metal cohesive strength. The num-
need for some sort of standardiza- the test to determine an “‘apparent ber two value is about equal to
tion or uniform approach for deter- unit shear strength” figure, even the base-metal cohesive strength.
mining the strength properties of though certain researchers have The number three value is much
brazed joints. Their clear explana- argued the actual need or validity of higher than either of the other two,
tion of the complex factors of this such a figure. These arguments and probably equal to the base-
problem and their comments on the are theoretically sound, but the fact metal cohesive strength.
correct and incorrect use of any such still remains that millions have been These figures prove that, when
test data will no doubt be of con- spent for conducting expensive the overlap is shortened and the
siderable help to researchers working strength tests on brazed joints. average stress on the base metal is
on problems connected with brazed- These test results have always been reduced sufficiently, a more realistic
joint mechanical properties. expressed as “unit shear strength” “apparent unit shear strength”
Their proposed extension for the which has resulted in much mis- figure can be obtained. This is be-
use of the original test is quite leading information. This has led cause the very short overlap has
unique and definitely would reveal design engineers to worry about and reduced the localized stresses in the
additional information. The test- question the low strength values re- joint area to the point where the
ing of several bars with overlaps ported in certain tests when actually average shear stress is close to the
from very short to very long would the brazed-joint bond strength was maximum shear stress. The “ap-
be good for revealing the load- very high. parent unit shear strength” then,
carrying capacities of a particular The test procedures advanced by would be more realistic than if it had
brazed .joint. However there are Messrs. Lehrer and Schwartzbart been taken from the last specimen
other design factors besides overlap, have shown very clearly that it is which had the longest overlap.
that also would affect the load- possible to get very strong bond On the other hand the suggestion
carrying capacity of a brazed struc- strengths in the joint area even by the discussers that joint strength
ture. For example, the effect of though the base metal and filler factors other than “apparent unit
the overlap on a shaft-insert or a metal cohesive strengths may be strength” be considered deserves
tongue-and-groove type of joint relatively low. For example let us consideration. It is true that
would probably not be the same as study the reported average shear mechanical factors as affected by
on a single lap joint of the proposed stress in the first specimen in Figs. joint design will have an effect on
design. D1, D2 and D3. (These speci- the resultant load-carrying capacity
It: is not clear as to how the dis- mens have the minimum overlap.)
of a brazed structure. The exten-
cussers plan to use the data shown 1. BAg-1 brazed copper, approx.
on the graphs in Figs. D1 to D3. sion as proposed by Messrs. Lehrer
average shear stress 48,000 and Schwartzbart may be a relatively
There may be several things that psi.
could be determined by study. simple way to determine some of
2. BAg-1 brazed mild steel,
Perhaps, for example, the areas these factors. It is hoped that other
approx. average shear stress
where the lines on the graphs 42,000 psi. researchers will do as well in the
straighten out could be used to find 3. BAg-1 brazed 4130 steel, analysis and use of the proposed
the minimum overlap necessary to approx. average shear stress shear test procedure for brazed
produce the maximum load-carrying 95,000 psi. joints.

464-s OCTOBER 1960

Above, left: Automatic weldes IS Casy to handle and operate Above right: Close -up of one type of automatic unit available
for welding tubes to tube sheets Note uniformity of sample welds above

Mechanized welding process makes the permanently tight

heat exchanger tube joints needed in modern power plants

The design and operation of modern produce consistently high quality joints. filler-metal addition, most of the tube
steam powel plants require that tubes and tube sheet combinations employed
Inert-gas tungsten-are (TIG) weld-
in condensers and feed-water heaters in condensers and heat exchangers.
ing has the advantages of simpli ity,
be installed permanently tight to pre-
good control, and protective shielding Joints, whether a few hundred or sev-
vent contamination of the boiler water.
without the use of flux. Tough, depend- eral thousand in number, are uniform
Welding the tubes to the tube sheets able equipment for automatic welding in geometry and quality. Typical are
will insure this. In most cases, how- with this process such as that shown the results obtained with tubes of Ar-
ever, welding cannot be done manu- in the illustrations above —is widely senical Admiralty-439 on tube sheets of
ally for a varietv of reasons A mech- available. These units make it possible Naval Brass-450 as show mn in the 20X
anized process must be employed to to weld, without the complication of micrograph reproduced at the left,
below.
Technical Assistance: The Metallurgi-
cal Department of Anaconda American
Brass Co. will gladly help you in the
selection of suitable alloys and in weld-
ing procedures and the design of joints.
This service is available without obli-
gation. See your Anac onda representa-
tiveor write: Anaconda American Brass
Company, Waterbury 20 Conn. In
Canada: Anaconda American Brass
Ltd.. New Toronto. Ont

ANACONDA
TUBES and PLATES for
in Oe CONDENSERS and HEAT EXCHANGERS
Mic rograph 20X of typic al weld section. Tube is of Arsenical Admiraltv-439 and Ana onda America n Brass ( ompany
tube sheet is of Naval Brass-450.
For details, circle No. 36 on Reader Information Card
To weld large diameter pipelines, and for many other
high tensile welds, use new EASYARC 10Mo—tops among
E-7010-Al electrodes. Its 1/2% moly matches pipe
composition. It’s best on vertical-down passes, from
12 to 6 o’clock, for root, capping and facing beads. And
don’t forget Airco’s EASYARC 10, the E-6010 electrode
— for welding plain steel pipe, or for root beads on
high tensile pipe. Available in 1/8”, 5/32”, 3/16’.

Now— better, faster pipeline weldin

with Airco electrodes...

new EASYARC 10Mo and EASYARC 10

If you're welding pipelines of the high strength steels — use Airco’s 119 electrode types: stainless, mild steel, low alloy,
Airco electrodes. They offer unbeatable welding character- low hydrogen, metal powder, aluminum, bronze, cast iron
istics, whether pipe diameters are 3 or 30”, walls thin or and hardfacing.
thick. They're designed for the job, and are all Airco- Phone your Authorized Airco Distributor. He’s in your
engineered. Classified Telephone Directory under “Welding Equipment
hey are also ideal for many other industrial applications. and Supplies.”
Whatever your stick welding application—choose from

On the west coast—

Air Reduction Pacific Company
AiR REDUCTION SALES COMPANY internationally—
Airco Company International
A division of Air Reduction Company, incorporated In Cuba—
Cuban Air Products Corporation
150 East 42nd Street, New York 17, N. Y. in Canada—
Air Reduction Canada Limited
More than 700 Authorized Airco Distributors Coast to Coast All divisions or subsidiaries
of Air Reduction Company, ine.
For details, circle No. 37 on Reader information Card