COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services

 AWS C5.5-80

Recommended Practices
for Gas Tungsten
Arc Welding

Prepared by
AWS Committee on-Arc Welding
and Arc Cutting
Under the Directionof
AWS Technical Activities Committee
Approved by
AWS Board of Directors, September 11,1979

AMERICAN WELDINGSOCIETY
2501 N.W.7th Street, Miami, FL 33125

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C5.5 8 0 W 0 7 8 4 2 b 5 0 0 0 2 5 8 7 2 .

Library of Congress Number: 80-65185

International Standard Book Number: 0-87171-193-1
American Welding Society,2501 N.W. 7th Street,Miami, FL 33125
01980 byAmerican Welding Society.
All rights reserved.
Note: By publicationof this standard, the American Welding Society does not insure anyone utilizing the standard against
liability arising from the use of such standard. A publication of a standard by the American Welding Society does notcarry
with it any right to make, use, or sell any patented items. Each prospectiveuser should make an independent investigation.
This standard is subject to revision at any time by the responsible technical committee. It must be reviewed everyfive years
and if not revised, it mustbe either reapproved or withdrawn. Comments (recommendations,additions, or deletions) and
any pertinent data which may be of use in improving this standard are requested and should be addressed to AWS
headquarters. Such comments will receivecareful considkration by the responsible technical committee and you will be
informed of the committee’s resp.onse. Guests are invited to attend all meetings ofAWS committees to express their
comments verbally. Procedures for appeal of an adverse decision concerning your comments are provided in the Rules of
Operation for AWS Technical Committees. A copy of these Rules can be obtain@ from the American Welding Society, 2501
N.W. 7th Street, Miami, FL 33125.
Printed in the United States of America

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 .
.
..
.
..
.
..
.
..
.
..
.
.".-~
AWS C 5 1 5 80.- 0 7 8"4 2~6 5~0 0_0 2~5 8_8 4
_ _ _ _ _ _ _

Contents
Personnel .................................................................................... V

Foreword .................................................................................... vii

I . ScopeandDefnitions ....................................................................... 1
1.1 Scope ............................................................................... 1
1.2 DefinitionsandHistory ................................................................. 1
1.3 Development of Process Variations ........................................................ 1
1.4 GrowthinAdaptability ................................................................. 2
1.5 Process Limitations .................................................................... 2
2 . Recommended Practices for GasTungstenArc Welding ............................................ 2
2.1 Materials andJointDesign .............................................................. 2
2.2ShieldingGas ......................................................................... 6
2.3 Arc Initiation Methods .................................................................. 6
2.4 Welding Current: Types and Application ................................................... 8
2.5 Arc Voltage .......................................................................... 10
2.6 WeldingSpeed ........................................................................ 11
2.7 TungstenElectrode .................................................................... 14
2.8 Filler Metal .......................................................................... 16
2.9 Fixturing ............................................................................ 17
2.10 Welding Schedule and Procedure ......................................................... 18
2.11 Welding Equipment Setup............................................................... 18
3 . Welder and Welding Engineer ................................................................ 22
3.1 Welder Training andQualification ........................................................ 22
3.2 Areas of Responsibility ................................................................. 22
4 . QllalityCoiltrol ............................................................................ 23
4.1 Inspection and Test Methods ............................................................. 23
4.2 Specifications ........................................................................ 23
5 . Typical Equipmentfor Process Applications ...................................................... 23
5.1 Manual Gas Tungsten Arc Welding ....................................................... 23
5.2 Semiautomatie Gas TungstenArc Welding .................................................. 25
5.3 Automatic Gas Tungsten Arc Welding..................................................... 26
6 . Safepractices .............................................................................. 37
6.1 Introduction .......................................................................... 37
6.2 Safe Handling of Shielding Gas Cylinders and Regulators ...................................... 37
6.3 Cylinderuse ......................................................................... 37
6.4 Gases ............................................................................... 37
6.5 MetalFumes ......................................................................... 37
6.6 RadiantEnergy ....................................................................... 37
6.7 Noise andHearing Protection ............................................................ 38
6.8 Safe Handling of Welding Equipment ...................................................... 38

iii

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C 5 - 5 B O W 0 7 B q 2 6 5 0002587 b

Personnel

AWS Committeeon Arc Welding and Arc Cutting

R . í?Hemzacek, Chairman International Harvester Company

L. C . Northard, Ist Vice Chairman Tennessee Valley Authority
J. R. Hannahs, 2nd Vice Chairman Midwest Testing Laboratories
H. W Raths, Secretary American WeldingSociety
a! L. Ballis Columbia Gas Distribution Company
L. R . Colarossi Pittsburgh-Des Moines Steel Company
S. M. El-Soudani Aircraft Engine Group, General Electric Company
N A . Freyfag The Budd Company
R. E . Garner Boeing Company
J. E. Hinkel Lincoln Electric Company
J. A . Hogan Hypertherm, Inc.
D. J. Kotecki Teledyne McKay Company
T. E . Junk Westinghouse Electric Corporation
R. K.Lee Consultant
R. A . Manley Naval Ship Engineering Center
R. D. Mann Airco Welding Products
E . R. Pierre Edward R. Pierre Enterprises
L. J. Privoznik Westinghouse Electric Corporation
P. W Ramsey A. O. Smith Corporation
G. R. Rothschild Airco, Inc.
H. S. Sayre Consultant
W K. Scattergood Suntech, Inc.
B. L. Shultz General American Transportation Corporation
M. D. Stepath Arcair Company
E. P. Vilkas Astro-Arc Company
N S. Wamack Combustion Engineering, Inc.
G. K. WilleckP Miller Electric Manufacturing Company
F. J. Winsor Foster Wheeler Corporation
*AdvisoryMember

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C5.5 8 0 W 0 7 8 4 2 6 5 0 0 0 2 5 7 0 2 W

PERSONNEL

Subcommittee onGas lhngsten Arc Welding

E . P. Vilkas, Chairman Astro-Arc Company

H. W Raths, Secretary American WeldingSociety
N. Chappel Consultant
H . R. Conway Huntington Alloys, Inc.
J. C . Downey E G&G Idaho, Inca
I . S. Goodman Westinghouse Electric Corporation
G. K . Hicken Sandia Labs
R . G. Hirsch Norfolk NavalShipyard
J, Lmzafame Daniel International
D. C.Leach Harrison Radiator Division, General Motors Corporation
R. A. Manley Naval Ship Engineering Center
J. O. Nelson Grumman Aerospace Corporation
L . C . Northard Tennessee Valley Authority
G . RRothschild
. Airco, Inc.
F. A. Shaikh Carolina Power andLight Company
T W Shearer, Jr. Fisher Body Division, General Motors Corporation
G. R.Stoeckinger Roy E. Hanson, Jr, , Manufacturing Company
J. C . Wormeli Quality Systems, Inc.

vi

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Foreword

Gas tungstenarc welding was introducedas a practical fabricating process approximately thirty-five years
ago. In the past several years,rapid advances have been made in the development oftechniques for automatic
applications, andgas tungsten arc welding is now accepted as the only practical method in some metal joining
applications.
Sufficient data have been recently gathered and organized to yield an authoritative source of sound
technical practices for gas tungsten arc welding. Accordingly, the AWS Committee on Arc Welding andArc
Cutting and the Subcommittee on Gas Tungsten Arc Welding have prepared these recommended practices.
These recommended practices are based on a survey of gas tungsten arc welding as used in the metal
fabricating industry.
The description of gas tungsten arc welding and itsfeatures is presented here as clearly andconcisely as
possible. The Committee hasdeveloped these guidelines in the hope thatthey would lead to further
development of the gas tungsten arc welding process and, thus, to higher quality andperformance standards.
Comments on this publication willbe most welcome. Theyshould be addressed to:
Secretary, AWS Arc Welding andArc Cutting Committee
American Welding Society
2501 N.W. 7th Street
b Miami, FL 33125

vii

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C 5 - 5 8 0 078qZb5 0002592 b =

Recommended Practices for

Gas Tungsten Arc Welding

1. Scope and Definitions (1.6 mm) diameter tungsten electrode can be used.
The slow response of the dc generator to changing
1.1 Scope. GTAW has become indispensable as a tool for conditions was first improvedby attaching aunitthat
industry in the relatively fewyears since its inception. This superimposed high frequency ionization on the wetding
has occurred through improvement of equipment and by current. This unit permitted the welding of aluminum.
development of new GTAW processcontrolsand Then a high frequency stabilized ac welding machine
techniques. The information prepared in this document was used, and results on the aluminum alloys being
describes the established standards and enumerates the welded werefar superior to those of the existing de power
recommended practices. sources. Although there are other differences between
1.2DefinitionsandHistory. Welding in an inert gas DCRP and DCSP, the phenomenon of cathodic cleaning
atmosphere was first considered in the late 1920’s. In associated with electrodepositive welding (DCRP) is most
1930, a patent was issued to Hobart and Devers covering important in welding aluminum alloys. The mechanism
the use of an electric arc within an inert gas atmosphere. responsible for removingsurface oxides from aluminum is
Various others experimented with argon and helium as generally agreed to be best explained by the theory of
shielding gases, but cost considerations were against atomic sputtering. Alternating current, with current flow
adoption of the process. in one direction and then in the other, allows the advan-
The first commercia1 development ofgas tungsten arc tages of cathodic cleaning without the disadvantages
process equipment occurred in the aircraft industry. Lead- associated with DCRP welding. Power sources were
ing the development were Russell Meredith and V. H. then developed specifically for GTAW (often called TIG
Pavlecka. In 1941, these menand their associates de- welding).
veloped the first practical electrode holders (torches) pro- Welding studies have revealed that the type of power
vided withmetal nozzles; helium gas was fed through the supply used for GTAW has a profound effect on the arc
electrode holder to protect the electrode, weld pool, and characteristics and properties of the weIds. However,until
adjacent heated areas of the workpiece from atmospheric recently, little was done to study the effect of power supply
contamination. They found that the gas shield must pro- waveform on the characteristics of GTA welds, because
vide full protection, since even a small amount of en- suitable equipment was not available.
trained air can contaminate the weld. The process was 1.3 Development ofProcess Variations. GTAW is
called “Heli-Arc’’ welding, and a patent was issued to adaptable to both manual and automatic operation and can
Meredith in February 1942. be used to produce continuous welds, intermittent welds,
The first gas tungsten arc (GTA) welding was done on and spot welds. Because the electrode is essentially non-
magnesium alloys withdc generators of the rotating type, consumable, a weld can be made by fusion of the base
utilizing dc reverse polarity (electrode positive). Today, dc metal without the addition of filler metaI. A filler metal
reverse polarity is almost never used, since its limitations may also be used, depending upon the requirements that
are well established. For example, an electrode about 3/16 have beenestablished for the particular joint. The hot wire
in. (4.8 mm) in diameter must be used when welding with method of filler metal additionwas introduced in 1966as a
lOOA DCRP, to avoid overheating. If the circuit polarities tool for increasing deposition rates of GTAW. Filler metal
are switched to electrode negative (DCSP), then a 1/16in. addition is desirable in manyapplications. Automatic

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS c5.5 B O rn 0 7 ~ ~ 1 2O O~ O5Z ~ W B m

2 / RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

tungsten electrode-to-workpositioning devices have been Weld penetration has been controlled for many years by
developed to allowwelding of contoured parts with a various designs of backup tooling. Other methods of con-
uniform heat input and more efficient filler metal addi- trolling weld penetration have been developed recently,
tions. The completely automatic GTAW process has the when backup tooling became either impossible or imprac-
following basic variables controlled from the start to the tical to apply to the butt joint. One basic approach was to
finish: improve the fundamental GTAW process controls, thereby
(1) Welding current with up- and downslopes assuring constant weld penetration. The first adaptive
(2) Arc voltage with up- and downslopes, which is controls were developed in order to compensate for such
equivalent to arc length control variables as weld joint gapping, thickness, and mismatch
(3) Welding speed and to control weld penetration or weld geometry.
(4)Inert gas flow In the GTAW spot welding process, tungsten electrode
(5) Filler wire feed rate and position tip life is the greatest limiting factor. Nevertheless,multi-
Since the GTAW process was most practical for welding ple electrode GTAW spot welding equipment is presently
materials such as titanium, many weldments were pro- used successfully by several automotivemanufacturers.
duced within enclosures or so-called dry boxes. This pro- Many new dissimilar metal joining application prob-
cess .variation requires vacuum pumping equipment to lems have been solved by development of proper filler
evacuate the ambient atmosphere from the dry box so it metals.
can be filled with inert gas.
Another GTAW process operationwas arc spot welding
with or without filler metal addition. As arc initiation
methods were improved, newGTAW process operations
were developed, employingseveral tungsten electrodes 2. Recommended Practices for Gas
positioned around the part to be welded andinitiated either Tungsten Arc Welding
sequentially or simultaneously.
1.4 Growth in Adaptability.GTAW involves a number of 2.1 Materials and Joint Design. The GTAW process is
basic functions. The arcis usually considered the heart of perhaps the most flexible of all hsion welding processes
GTAW. However, processing and positioning of the parts in the variety of metals welded. Variations in wkld joint
to be welded and measurement of the welded product are, design are limited only by the characteristics of the par-
in many applications, as important as the handling of the ticular metal and joint efficiencyrequirements. In addition
welding arc as a tool. to the wide choice of joint designs and metals that may be
The growth of GTAW process adaptability must be welded with thisprocess, the filler metal (if used) may be
attributed to innovations in electrode holder designs, de- added to the weld pool by manualor automatic means, or
velopments in dc and ac power sources, the evolution of as a preformed shape.
automatic arc starting systems, automatic positioning Most of the commercially availableweldingpower
and process sequence controls, and a concentrated effort sources offer precision control over the welding variables;
to establish gas tungsten arc welding as the process for and along with the excellent visibility of the exposed gas
fabrication. shielded arc, the addition of filler metal maybe precisely
controlled to suit the metallurgical requirements of the
1.5 Process Limitations.GTAW requires continuous and particular weldment.
efficient weld metal shielding. Backup shielding is also This section includes guidelines and accepted practices
required in manyapplications. This basic requirement concerning materials and joint designs. Specific informa-
tends to limit the process to indoor types of applications; tion concerning base metals and their respective joint
however, with proper shielding techniques, field welding designs is discussed in depth in the AWS handbooks.
is also readily accomplished. 2.1.1 Composition and Quality. Most of the metals
The duty cycle of the GTAW equipment generally ex- weldable by other fusion processes may be welded bythe
ceeds the intended service requirements. However, many GTAW process. Successful welding requires the base
problems have been experienced through neglect check- in metal to be of a chemical composition suitable for weld-
ing the welding duty cycle and matching it with the duty ing. Thechemical composition of the base metal should be
cycle of the rectifier, water cooling efficiency, tungsten known prior to welding so that variables such as filler
electrode holder efficiency, and the efficiency of tooling metals, metallurgical properties, preheating or postheat-
with respect to heat dissipation. ing, and other important considerations may be included
The fact that the gas tungsten arc has the shape of a cone in the application of the welding process. In some cases,
rather than a cylinder required development of various the chemical composition of the base metal may be un-
techniques in manual application and arc positioning con- known and various weldability tests are conducted. These
trols in automatic applications. It should be noted that the may be destructive or nondestructive tests and are gener-
widely used arc voltage control method requires a consid- ally conducted on small samples. These tests easily deter-
erable amount of practical understanding, since arc volt- mine whetherthe material is compatible with GTAW,
age is not a linear function of arc length. Base metals that can be traced to a particular chemical

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices / 3

composition are usually referenced in the AWS hand- Many variationsare derived from these five basic joints.
books, whichdeal with specific welding details. Neverthe- Most are derived from the butt joint. These basic joints
less, certain chemical compositions that appear to be represent the minimum amount of preparation prior to
within the limitations for the major elements may display welding. They also provide the most economical welding,
undesirable characteristics during and after welding. since minimum filler metal, minimum preparation, and
These characterristics are generally found to bereIated to minimum setup time are required.
unreported trace elements within the metal. Lack of weld Weld requirements for complete joint penetration, high
penetration when using fixed welding conditions, exces- joint strengths, or other special requirements usually re-
sive porosity, andmicrocracking are a few of the undesira- quire some modifications to the square butt joint as the
ble characteristics that have been identified and traced to thickness increases over sheet gages. A few of these vari-
various trace element effects. ations are shown below.
Although the possibility of trace element problems
exists, the chemical certifications generally provide suffi-
cient information for determining the relative weldability
of a particular metal. The final test of quality metal is the
M Single-V
Da Double-V
l2ElSingle-U
actual welding; and, in certain instances, the addition of a
filler metal having a slightly different but compatible
chemistry may help to eliminate or minimize a metal While there are no fixed rules governing the use of a
problem. particular joint design for any one metal, certain designs
2.1.2 Joint Design. Due to the variety of hase metals were developed for variousreasons. Areas for which each
and their individual characteristics (such as surface ten- joint type (single-V, double-V, single-U) is best suited are
sion, flowability, melting temperature, etc.), jointgeome- presented in the following paragraphs.
tries or designs that provide for optimum welding condi- 2.1.2.1 Single-V Butt Joint. This joint configura-
tions should be used. Factors affecting joint designs in- tion is the most widely used of all joint designs when
clude metal thickness, weld penetration requirements, and complete joint penetration is required from one sideonly.
joint efficiency requirements, along with the metals There are three variables of this joint design: root open-
characteristics previously mentioned. ing, thickness of root face, and angle of bevel. All vari-
The first consideration in joint design is provision for ables must be considered prior to joint preparation.
proper accessibility. The jointopening must be adequate to The amount of root opening and thickness of the root
permit the arc, proper gas shielding, and filler metal to face depend upon whether the GTAW process is to be
reach the bottom of the joint. If manipulation is required, manual or automatic, whether filler metal is to be added
the opening must be sufficient to allow forproper manipu- during the root pass, andwhether a backing is to be
lation. In determiningcorrectjointdesign, the characteris- employed. The amount of angle depends upon the thick-
tics of the weld metal must be considered. For example, ness of the metal and the clearance needed for arc move-
the high nickel alloys are very sluggish when molten and, ment to assure adequate fusion on both sides of the joint.
therefore, the weld metal does not spread. The weld metal These variables are generally determined by welding test
must be placed at the proper location in the joint. There- samples which have been prepared using a variety of
fore, high nickel alloy joints must be more open than those combinations. Whilespecific informationon joint designs
for carbon and alloy steel in order to provide space for may be found in the AWS handbooks or in the metals
manipulation. suppliers' literature, the following design is presented as a
The five basic joints (butt, lap, T; edge, and corner) guide.
shown in the figures below may be used for virtually all
metals.

Most Metals
T - U8 to 1/2in. (3.2to 12.7 mm)
A -60" to 90"
B - 3/32 in. (2.4 mm) or less
Corner Edge C - 1/16 in. (1.6 mm) or less

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS c 5 4 ao m O X N Z H O O O Z ~I ~m

4 / RECOMMENDED FORGASTUNGSTEN ARCWELDING

PRACTICES

2.1.2.2 Double-V Butt Joint. This joint is primarily 2.1.3 Joint Preparation and Tolerances. After the
used when thicker sections that require complete joint particular joint design has been established, the most
penetration are welded. It provides the economic advan- important item for consideration is the method of joint
tage of a reduced amount of welding. Its use is usually preparation. There are many ways to removemetal to
confined to sections over 1/2 in. (12.7 mm) where weld prepare a given joint angle, land thickness, or geometrical
distortion may be a problem. With careful weld pass configuration. However, many GTAW problems or sup-
sequencing, the double-V joint permits welding on both posed problems are a direct result of using improper
sides to hold weld distortion to a minimum. methods for joint preparation. Chief among these is the
The following design is presented as a guide. improper use of grinding wheels to prepare joints. Soft
materials such as aluminum become impregnated with
microsized abrasive particles which, unless subsequently
removed, will resultin excessive porosity. Grinding

"Qf3
1 wheels should be cleaned and reserved exclusively for
material being welded. The ideal joint preparation is ob-
the

tained through the use of a cutting tool such as a.lathe for

T round or cylindricaljoints or a milling cutter for longitudi-
nal preparations. Care must be exercised in the choice of
cutting fluid (if any) to be used, and cleaning after cutting
or turning should be with safety-approved solvents that are
B free of residues.
The required exactness of joint preparation also de-
pends upon whether the GTAW is to be done manually or
Most Metals by automatic means. Manual welding conditions can tol-
erate greater irregularities in joint fit-up than automatic
T - 1/2 in. (12.7 mm) and over welding. The particular tolerance for a given application
A -60" to 90" can be determined only by actual testing, and this toler-
B - 3/32 in. (2.4 mm) or less ance should be specified for future work.
C - 1/16in. (1.6 mm) or less
Beveling is usually not required for butt joints 1/8 in.
(3.2 mm) or less in thickness in carbon, low alloy and stain-
less steels, and aluminum. For high nickel alloys, beveling
2.1.2.3 Single-U Butt Joint. This information con- is usually not required for'material 3/32 in. (2.4 mm) or
cerning the single-U groove generally applies to all con- less in thickness. For greater thickness materials, the joint
should be beveled to form a V-, U-, or J-groove. Other-
figurations obtained by preparation methods involving
wise, erratic penetration will result, causing crevices and
operations or tools for other than straight line metal re-
voids that are potential areas of accelerated corrosion.
moval. These types of joints areused on a variety of metals
Notches resulting from erratic penetration can also act
and thickness ranges. The type of metal to be weldedoften
as mechanical stress raisers and cause early mechanical
dictates a specially prepared joint, even thoughjoint prep-
failure in the weld joint.
aration costs are higher. Certain metals in the refractory
group, such as the titanium alloys, tend to possess unusual Normally, a double-U or double-V joint design is prefer-
fluidity during welding, and a specially prepared joint is red for material over 112 in. (12.7 mm) 'thick. The added
cost of preparation-ìsjustified by the decreased amount of
the best answer for obtaining a satisfactory root pass. Butt
welding metal and lower welding time needed to complete
welds in aluminum piping are often made with a joint
the joint.Also, less residual stress will be developed than
as shown in illustration (A) for control of root pass pene-
with the single groove design.
tration when welding in all positions without internal
As shown in Fig. 1, V-groove joints are normally bev-
backing.
eled to provide approximately a 60 degree groovefor
This joint design isusually applied after other less
carbon, low alloy, and stainless steels, and an 80 degree
expensive avenues have been explored. Since the specific
groove forhigh nickel alloys. A 90 degree groove is com-
design for this type of joint varies considerably, two exam-
monly used when welding aluminum with ac GTAW. The
ples are shown.
U-groove joints are generally beveled to 7 to 9 degree side
wallsfor carbon, lowalloy,and stainless steels. A 15

m degree side wall is recommended for high nickel alloys,

and a 20 to 30 degree side wall bevel maybe required for
U-groove joints in aluminum. Single bevels for T-joints
between dissimilar thicknesses of ferrous metals should
Used where metal Used where locating
fluidity creates surfaces are required have an angle of approximately45 degrees. An angle of as
problems in obtaining for accurate alignment much as 60degrees may be necessary for aluminum alloys.
root pass 2.1.4 Production Conditions. Cleanliness is one of

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices I 5

60"
V-Groove

L1116 in.

J-Groove
U-Groove
4 7"
,-3/16-5116 in. R 112 in.
7"

L 3 1 3 2 in. f 118 in.

Double-
Double-
U-Groove
r\ Eo
7"

r
Metric Equivalents
In. mm
1.6 1/16
3/32 2.4
118 3.2
L 3 1 3 2 in. 3/16 4.8
3/16-,5/16 in:R 5/16 8.0
112 12.7

Fig. 1"Qpical joint designs for ferrous metals

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 the most important and also one of the most often over- lower arc voltage at any givencurrent value andarc length,
looked requirements for successful joining of metals. a smoother and quieter arc, and easier arc initiation. Since
Many substances often used during normal manufacturing it is a heavy gas (atomic weight 40), lower flow rates are
processes can cause welding difficulty if not thoroughly required to provide good shielding.
removed.Examples include grease, oil, paint, cutting Helium is used in welding heavy sections where the
fluids, marking crayons and inks, processing chemicals, higher arc voltage characteristicis advantageous or greater
and machine lubricants. Because it is frequently impracti- penetration is desired, or both.
cal to avoid the use of these materials during processing An excellent reference for the selection of shielding
and fabrication of the alloys, it is mandatory that the metal gases is Table 23.2 of Section 2 of the AWS Welding
be thoroughly cleaned prior to any welding operation. A Handbook, 6th Edition. A similar table (Table 1) has been
minimum area of cleaning should extend112 in. (12.7 mm) included here.
from the joint on each side. The cleaned area should 2.2.2 Shielding Efficiency. The purpose of shielding
include the edges of the workpiece and the interiors of gas is to protect the weld area and the tungsten electrode
hollow or tubular shapes. from contamination by the atmosphere. Nozzle size, flow
The cleaning method depends upon the composition of rates, and the type of gas utilized are critical factors in the
the substance to be removed and should not cause any process. See paragraph 2.7.2.3 (Atmospheric Contamina-
problems with the metals being joined. Shop dirt and tion) and Tables6 and 7.
materials having an oil or grease base can be removed by 2.2.3 Methods of Shielding.The shielding gas is gen-
vapor degreasing or swabbing with non-toxic solvents. erally delivered around the tungsten electrode through a
Paint and other materials not soluble in degreasing sol- concentric nozzle. Nozzle size must be adequate to provide
vents may require the use of methylene chloride, alkaline coverage of the weld area. Automatic welding setups may
cleaners, or special proprietary compounds. include supplemental shielding by the use of leading or
Oxides should also be removed. For some metals, wire trailing shields that increase coverage area.
brushing is sufficient. Howeyer, wire brushing may not be 2.2.4 Economics. Argon is obtained from the lique-
sufficient for metals that have refractory oxides, such as faction of air and isone of the three primary productsof air
aluminum, stainless steel, and high nickel alloys. Accept- separation plants. The other two primary products are
able methods for removing the oxides include grinding, oxygen and nitrogen. Although argon is less than one
abrasive blasting, machining, or pickling (except forAl). percent of the yield, it is readily available.
Aluminum welds arb often made by decreasing the weld Helium is obtained from wells controlled by the Bureau
joint opening when ac GTA welding is employed. The list of Mines, witha small amount produced by private indus-
of good practices shown below should be followed forall try throughrecovery from natural gas. Limited availability
production conditions. places a premium on this gas.
Metal Production Conditions Prior to Welding Shielding gas systems using a total of less than 10 O00
(1) Metal shall be free of foreign substances, oxides, mois- cubic feet (283 O00 liters) per month normally employ
ture, marking crayons, etc. individual cylinders. Distribution manifolds fed bycylin-
(2) Special cleaning of metal shall be accomplished when der banks or trailers are used for higher consumptionrates.
required by specifications. High consumption rate users of argon should consider a
(3) Weld joint edges shall be clean and uniform. Edges bulk liquid system. Since normal losses due to boiling of
shall be checked forlaminations or other discontinuitiesin the liquid average one percent of the volume in 24 hours,
accordance with the applicable specification. withdrawal rates, hours per day on line, and vesselcapac-
(4)Weld joint fit-up shall be in accordance with the ap- ity must be carefully evaluated.
proved procedure.
2.3 Arc Initiation Methods
(5) Metal identification shall be maintained for accurate
documentation of welded components.
2.3.1 Touch Start. The tungsten electrode tip is
momentarily placed in contact with the workpiece and
Adhering to the above steps will, in most instances,
quickly withdrawn.Inert gasflow during this procedure is
assure that the metal is ready for welding and that adequate
assumed. This starting method is not acceptable for criti-
quality control measures have been observed.
cal applications, since small tungsten particles may be-
2.2 Shielding Gas come embedded in the workpiece.
2.2.1 Quality, ljpes, andMixtures. Theinert 2.3.2 Carbon Start. The tungstentip is positioned
monatomic gases argon and helium are used as shielding close to the work and the resulting gap is momentarily
for GTAW. Mixtures of argon andhelium are useful when bridged with a carbon rodor block. Afterthe arc is formed,
some balance between the characteristics of both is de- the carbon is withdrawn or the arcis moved overthe joint
sired, Additions of hydrogenand nitrogen havebeen to be welded. This method is undesirable when critical
utilized in special applications. Weldinggrade argon and weld applications are involved, since carbon particles may
helium are supplied with purity of 99.995 percent or be entrapped in the work. Also, the required application of
greater. Dewpoint is -70" F (- 57" C)or lower. the carbon rod is frequently impractical.
. Argon is the most commonly used gas as it provides 2.3.3 High Frequency Start. This method takes ad-

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices I 7

Table 1
Advantages of shielding gases
Welding
hielding type Metal gas Advantages
Manual
welding
Argon
Better arc
starting,
cleaning
action,
and weld quality; lower gas
consumption.

speeds
welding
Argon-helium
Aluminum
High possible..
and
Machine
Argon-helium
Better
weld quality, lower gas
flow than required with straight
helium.
Magnesium Welding
Helium (DCSP) Deeper penetration and higher weld speeds than can be obtained with
argon-helium.
spot
Welding
Argon
Generally
preferred for longer
electrode life. Better weld nugget contour.
Ease of starting, lower gas flows than helium.
Manual Carbon
control,
especially
Welding
pool
Better
Argon
steel for position welding.
Machine
Welding
Helium
Higher
speeds
obtained than with argon,
Manual
Welding Argon Permits controlled
penetration on thin
gage
material
(up
to 14 gage).
Argon
Excellent
control of penetration on light
gage
materials.
Argon-helium Higher heat input, higher welding speeds possible on heavier gages.
Stainless
Machine
Argon-hydrogen Prevents undercutting,
produces
desirable weld contour
at low current
steel Welding to(up 35% H2) levels, requires lower gas flows.
Argon-hydrogen-
helium An excellent selectionfor high
speed
tube
mill
operation.
Helium
Provides
highest
heat
input
and
deepest
penetration.
ArgonEase of obtainingpoolcontrol,penetration,andbeadcontouron thin
gage metal.
Copper,
nickel,
and
Cu-Ni alloys Argon-helium
Higher heat
input
offset
to high
heat conductivity of heavier gages.
HeliumHighestheatinputfor weIding speedonheavymetalsections.
Argon Low gas flow rateminimizesturbulence and aircontamination of weld;
Titanium improved heataffected zone.
HeliumBetterpenetration for manualwelding of thick sections(inertgasbacking
required to shield backof weld against contamination).
Silicon-
cracking
Reduces
Argon
bronze this “hot
metal.
ofshort”
Aluminum-
ArgonBronze Less penetration
metal.of base

vantage of the ionization characteristic of gases. The ap- tion systems. FCC regulation Part 18 establishes the
plication of an alternating voltage in the kilohertz fre- maximum high frequency radiation permissible.
quency range causes the gas between the tungsten tip and 2.3.4 Pilot Arc. A small current is maintained between
the work to ionize. This establishes a conductive path, electrode and torch nozzle to provide a conductive path
permitting the weld current to start flowing. for the main weld current. This method is most often used
Since high frequency tends to erode the tungsten elec- in GTA spot welding equipment.
trode tip, automatic circuit arrangements are usually pro- 2.3.5 Hot Thngsten Arc. The tungsten is resistively
vided which shut off the high frequency as soon as thearc heated to a cherry red. At this temperature, the gas in the
has been established. Special design precautions are re- immediate vicinity of the tungsten electrode tip is easily
quired to prevent the high frequency from radiating too ionized by the application of a reasonable open circuit
much energy and causing interference with communica- voltage from the weldingpowersupply. This starting

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C 5 0 5 8 0 0784265 00025q7 7 W

8 / RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

method is also used in GTA spot welding equipment. 2 , Reverse polarity (electrode
positive)
The temperature control of the tungsten and the preheat-
ing effect of the hot electrode are disadvantagesof this arc
starting method.
2.3.6 Impulse Start. The momentary application of a
high voltage (usually from a bank of capacitors) between
the tungsten electrode and the work creates an ionized path
.
Straight polarity (electrode negative)
through which the weld current starts flowing. Poor reactor design

2.4 Welding Current: Qpes and Applications. Three v)

$ I Reverse polarity (electrode

positive)
basic categories of available weldingcurrent are alternat-
ing current, direct current, and programmed or manipu-
lated current. Programmed current is a category combin-
ing some of the features of ac and dc.
2.4.1 Alternating Current. Alternating current flows Straight polarity (electrode negative)
Good reactor design
first in one direction and then inthe other. Whenchanging
from positive to negative flow, the current must pass
through zero, which extinguishes the arc. The arc must Fig. 3 -Clipped wave form
then be reignited or it will remain extinguished. In the
GTAW process with some metals, the electrons flow Another method to stabilize the arc without balancing
from the tungsten to the work more readily than in the the current wave output is to utilize a superimposed high
reverse direction. This difference in resistance to electron frequency.
flow with direction of current is greater with some metals The advantages of balanced current flow are:
thanwith others. Aluminum, magnesium, and copper (1) Better work surface cleaning action
with oxide films are particular examples. Whenaluminum (2) Smoother, more stable arc
is welded using alternating current, the electron flow is (3) Less electrical upset in welding power source result-
from the plate to the tungsten during the reverse polarity ing from unbalance and occasional fully rectified load
(electrode positive) halfwave, and from the tungsten to the current
plate during the straight polarity (electrode negative) half The disadvantages of balanced current wave are:
wave. Hot tungsten is a better emitter of electrons than the (1) A larger diameter electrode is required.
aluminum workpiece; thus,the straight polarity half wave (2) A higher open circuit voltage is required for ba-
has a higher magnitude of current, which results in an lanced wave welding machines.
unbalance in current. This unbalance, spoken of as partial (3) Cost is usually considerably higher than conven-
rectification of the arc, is shownin Fig. 2. tional power sources.
Balanced wave current is most desirable for mechanized
welding andhigh speed welding applications.
v)
3 Reverse polarity (electrode
positive) 2.4.2 Direct Current. Power supplies commercially
n I available fordc welding are motor generators, transformer
rectifiers in combinationwith saturable reactors or magne-
tic amplifiers, and, more recently, transformers with pre-
= 01
cisely controlled silicon controlled rectifiers (SCRs). The
Straight polarity (electrode negative)
more sophisticated power supplies are most frequently
usedin conjunction with semiautomatic and automatic
Fig, 2 -Unbalanced current wave tungsten arc welding processes.
All power supplies incorporate some form of current-
to-voltage characteristic commonly referred to as “con-
stant current characteristic” whereby the welding current
At times, the electrode positive half cyclefails to ignite is essentially unaffected by variations in the arc length
and complete rectification takesplace. This is sometimes (voltage). Two types of direct current GTAW to be consid-
referred to as “clipping.” This clipping effect is shown in eredare: direct current straight polarity (DCSP), and
Fig. 3. direct current reverse polarity (DCRP). Each type has its
Several methods ofwave balance have been used to own characteristics and areas of application.
equalize the negative and positiveportions of the current Straight polarity (electrode negative) generates greater
wave. Some of these are: heat at the positive (work) end of the arc and less at the
(1) Series capacitors negative (electrode) end. Without the severe heating of the
(2) Resistor-rectifiercombinations electrode, smaller diameter electrodes can be used; i.e., a
(3) Batteries 1/16 in. (1.6 mm) diameter electrode is capable of carry-
(4) Wave balance ratio ing 125 A.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices I 9

Fig. 4 -Direct current-straight polarity (DCSP)

DCSP is normally used for welding practically all met- grain size, and out-of-position weldingcapabilities.
als. The molten weldpool is narrower andthe penetration Two such programmed current techniques are square
deeper than with DCRP and ac welds. Results are nar- wave ac pulsation and pulsating dc.
rower heat-affected zones, less distortion, and faster weld- Square wave ac pulsation varies from ac balanced wave
ing rates. welding inthat the generated wave form is from a precision
DCRP is rarely used but does have a particular advan- dc current source which is switched in alternate pulses of
tage of surface cleaning on metals whose oxides cause reverse and straight polarity direct current. The time the
problems to the welding operation. This cleaning action is current is flowing in one direction is adjustable from 11120
beneficial for the welding of aluminum and magnesium. to 1/6 second. This allows a programmed current of time
This same action occurs in the reverse polarity half cycle and magnitude that will provide some surface cleaning
of ac welding. action while on DCRP, as well as provide penetration
Withreversepolarity, the electrode is positive,and patterns and depths approaching DCSP and good arc
negative electrons strike the tungsten, causing electrode stability and controllability. Studies have also shown that
overheating. Since the electron flow heats the eIectrode weld pool agitation can be introduced with programmed
and not the plate, the weld pool is shallow and wide. alternating pulses that aid inproducing consistently high
This excessive electrode heating limits the application of quality welds(see Fig. 6).
DCRP; i.e., a 114 in. (6.4 mm) diameter electrode has an The pulsating dc envelope pattern is composed of two
allowable current-carrying capacity of approximately 125 different current levels selected to best suit the intended
A. application. Time for each current level is selected to yield
2.4.3 Programmed Current. Welding powersupplies a program that produces consistent weld metal quality. The
have also been developed capable of providing various frequency of the ripple is 60 hertz. The ripple ratio is
types of “modulated” current with or without actual polar- related to the ac reactance of the power supply and the
ity reversal. These “current pulsation techniques” have external dc inductor usually used in a single-phase full
been developed forthe purpose of controlling or improv- wave rectifier system and is a factor of system design (see
ing such items as weld root and bead contour,penetration, Fig. 7).

Welding

1/4 in. (6.4 mm)

diam. electrode

E’ectrons

-
Fig. 5- Direct current reverse polarity@CRI’)

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 10 I RECOMMENDED FORGASTUNGSTEN ARC WELDING
PRACTICES

welds. Although helium gas helps, it is difficult to manu-

ally weld aluminum with DCSP, as very short (approx.
0.060 in. C1.5 mm]) arc lengths are required. Therefore,
automatic methods of controlling arc length or arc voItage
were developed.
The voltmeter on the power supply indicates the total of
all voltage drops in the cables, connections,tools, tungsten
holder, electrode, and arc. The voltage most nearly repre-
senting the conditions of the arc is that measured between
the tungsten electrode holder and thework. Electrode tip
geometry is also an important factor affecting dc arc
voltage, as indicated by Fig. 8 .
Figure 8 illustrates that at the same tip-to-work distance
the arc voltage is higher with a sharper cone tip on the
electrode.
2.5.2 Methods of ControllingArcVoltage. Two
2.5 Arc Voltage methods of arc voltage control are employed in automatic
2.5.1 FactorsAffectingArcVoltage. The voltage GTAW.
drop between the tip of the tungsten electrode and the work The first method is to move the tungsten electrode at a
is influenced primarily by the type of welding current and preset distance to the work independent of the arc voltage
the shielding gas used. The arcvoltage is proportional to and current. These systems vary from simple visual ad-
the arc gap length and gas composition. Shielding gases, justment to the use of elaborate magnetic and eddy current
such as helium and argon-helium, produce greater voltage devices.During welding with theconstant position
drops than argon for the same arc gap. The difference method, a strip chartrecording of arc voltage will indicate
between an argon and a heliumvoltage drop is approxi- changes in arc length (penetration). The conformity of
mately 4 volts. Therefore, helium gas is used for deeper electrode tip used from weld to weld, as well as electrode
penetration, particularly when DCSP welding aluminum tip deterioration, can be observed on a voltage recording
alloys. During manual ac welding, the operator is not meter. This observation may indicate the difference bet-
concerned with the arc voltage. However, he controls the ween defective andtrouble-free welds.
arc lengthvery carefully in order to make acceptable The second method is to automatically control move-

too x

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices I 11

I 1 I I I I I I I I I I I I
Argon shielding gas

15”
-
Arc

1O0 150 200

Direct current straight polarity (A)

Fig. 8 -Voltage and current for two different eIectrode tapers,

as measured at power supply and directly across arc

ment of the tungsten electrode at a preset arc voltage. The 2.6 Welding Speed
voltage is held constant by a reversible gear motor, which 2.6.1 PrinciplesGoverningSpeedSelection. Arc
raises or lowers the electrode to change the arclength to a penetration is usually inversely proportional to welding
preset voltage level.The basic schematic of an automatic speed. In automatic in-place butt welding of tube and
GTAW voltage control system is given inFig. 9. In such a pipe, for example, proper welding speed programming
system, the measured arc voltage consists of two distinct provides constant complete penetration welds in butt
voltage drops: the I-R drop due to the arc itself, and the joints.Programming of welding currentrather than
drop in the tungsten electrode. The voltage drop in the welding speed can also produce acceptable tube welds.
tungsten electrode begins to rise immediately after arc However, the constant heat densiiy required to produce
initiation, due to heat caused by the welding current. As constant arc penetration is easierto develop and to
the voltage drop increases, the AVC (arc voltage control) reproduce with a variable welding speed system.
will shorten the arc length to maintain the sum at a con- The welding speed pattern is affected by several factors.
stant value. The characteristic “diving” of the tungsten As illustrated in Table 2, the thermal conductivity of the
electrode often encountered in GTAW with AVC heads is tube metal is a major factor controlling speed.
partially due to resistance changes that occur during the 2.6.2 Consistency of Heat Input.Experimentallyop-
warm-up period of a weld. The record of the tungsten tip timizing the welding speed pattern minimizes distortions
geometry is also important for transferability ofGTAW caused by thermal expansions and contractionsduring the
parameters from machine to machine in a plant or between welding. In the welding of highly conductive metalssuch
plants for a given job. as aluminum, fast welding speeds are preferred to keep
~ ~ ~~

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C 5 0 5 8007 8 4 2 6 5 .O002603 7

12 I RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

Amplifier “-c Motor

circuit T+ circuit

Error voltage

“AVC h e a d y

Signal J
circuit

t
Welding
current

I Work I. I SUPPlY

Fig. 9-Automatic GTAW voltage control system

Table 2
Typical welding speed program settings
with automatic in-place tube welding system

Tube diameter
of Wall Constant Final
butt joint thickness speed speed Total
~- start
delay, weld time,
Tube alloy in. mm in. mm S idmin. mm/s in./min. mm/s S
- - “ ~~- ”

Stainless steel
(AM-350) 0.028 3.2
0.71 7.85 0.4 12.4 3.7 8.7
Titanium
1/2 (6A14V) 12.7 0.037 3.7
0.94 1.2 1.5 4.2 1.8 20.2
Copper-nickel
(90-10) 0.050 2.1 1.27 0.7 0.87 7.1 3,O 33.7
Aluminum
0.049 4.7
1.24 1.0 1.9 8.7 20.7 15.4

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C5.5 B O 0784265 0002604 9

Recommended Practices / 13

ahead of the conducted heat. Welding

Alloys prone to thermal cracking cannot be welded af r head
high speeds, since the associated steep thermal gradients
Bevel gear
would contribute to crack formation. Fairlylowweld electrode holder
speeds are applied to circumvent this cracking problem.
Lowweld speeds are often used in combinationwith
Gas
preheating the base metal to further reduce the possibility
nozzle
of thermal shock.
The size of the molten weldpool is directly influenced
by the weld speed. Only a small weld pool cqn be carried L 1-1/2in. titanium tube
when welding in positions other than the flat position 1/8 in. thick wall
(downhand). Careful selection of the weld speed is there-
fore required. The problem of in-place welding and Fig. 10- Cross section of a typical
controlling the weld pool in heavy-wall tube butt joints overlapping melt-thrubutt joint weld
has beensolved for the ferrous and Iow conductivity
metals, such as titanium, with the development of welding
current pulsation controls. A programmed weld speed
effectively produces a series of overlapping melt-thru
spot welds, each of which solidifies before the next spot
is made.
This pulsed current technique was further improved by multiple-passwelding heavy wallscause distortions which
the development of the step pulse or incremental welding can be reduced by selecting a suitable combination of
arc travel control (see Fig. 10). Figure 10 illustrates a welding and torch cross-seam oscillationspeed. Adjusting
cross section of a typical steppulse melt-thru weld the dwell time at the end of each strokeprovides the proper
joining 1-1/2 in. (38 mm) diameter,1/8 in. (3.2 mm) thick “wetting” action of filler metal to wall joint and eliminates
wall titanium tubing. the occurrence of “cold shuts.” Also, proper selection of
Figure 11illustrates the step pulse welding control prog- dwell time in the 2G (vertical) positionprovides control of
ram which is indispensable in fully controlled penetration weld bead sag during multiple-pass weldingof pipes.
(root) passes. Metal as thick as 1/4 in. (6.4 mm), which is In welding long seams, the welding arc should be pre-
greater than the practical GTAW melt-thru welding thick- ciseIy positioned overthe joint to bewelded. The follow-
ness range, has been successfully welded. ing seam tracking techniques have been developedto suit
Thethermalexpansionsandcontractionsfrom specific applications.

~~~ ~ ~

Table 3
Types of tungsten electrodes
AWS
classi-
of Type tungsten Available
fication (avg. alloy) finish+ Color code Remarks
Pure
EWP Green Cleaned Provides
arc good stability for ac welding. Reasonably
andgroundgoodresistance to contamination. Lowest current-
carrying capacity. Least expensive. Maintains a clean
balled end.
Zirconia,
EWZr Brown Cleaned Preferred when tungsten contamination of weld is in-
0.15 to 0.40% and ground tolerable. Excellent for ac weldingto
due favorable
retention of balled end, high resistance to
contamination, and goodarc starting.
EWTh-1 Thoria, Yellow Cleaned
0.8 to 1.2% and ground Easier arc starting.Highercurrent capacity. Greater
EWTh-2
Thoria, Red Cleaned arc stability. High resistance to weId pool
1.7 to 2.2% ground
and contamination. Difficult to maintain balled end on ac.
EWTh-3 Thoria, Blue Designed primarily for ac welding to improve
to 0.35 0.55% balled end.
*Clean finish designates electrodes are chemically cleaned and etched. Ground
that finish designates electrodes
with a centerless ground finish to provide
maximum smoothness.
Centerless ground tungsten electrodes are used where minimum resistance
loss at the collet-electrode contact point
is desired.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 3 Pulse, high currenttime

nn
t- -
Pulse, low current time
a
Y

c,
I I-( I\
Pulse, high current
E!
3
V

:
I\
-
1 I .I
I
I I
I I
I
I

I
c- Final current

Initial
current- I
I I Weld time I
I I 4 Final
I I I I
slope t
rc
start
-c._I Time (sec.)
Start
downslope

Programmer startsto
increase motor (travel)
speed: distance between
pulses increases.*
Ba,
P
ô z
c,

SE
"-
t I I I
I
Motor
delay
time
- I
I
Constant
speed
time
1I'
I
Upslope
speed
time-=
I
I
~un-out
time
L
I
I
Time (sec.)
*Penetration is reduced

Fig. 11-Step pulse programmer

(1) Stylus. A stylus with a suitably shaped tip is (e) Variations in the location of a reference type
mounted ahead of the welding torch, and the tip rides shim mounted parallel to the joint to be welded as ob-
inside the joint tobe welded. Variations are translated into served bya mechanical sensing element
electrical control signals, causing a servomechanism to ( f ) Variations in the location of the joint to be
adjust the welding torch so it accurately positions over the welded as observed through a closed-circuit television
joint. camera
(2) The required servocontrol signal can also be ob-
tained byusing joint sensing techniques based on: 2.7 WngstenElectrode
(a) Variations inelectrical capacitance caused bya 2.7.1 Type. Tungsten is employed as an essentially
variation of distance of the weld joint in relation to the nonconsumableelectrode for the GTAW process. lbngsten
sensor element has melting point of6160" F (3392" C ) and a boiling
(b) Variations in magneticreluctance point of 10 700" F (5906" C). It is virtually impossible to
(c) Variations in the frequency of a timed circuit vaporize a tungsten electrode during welding, provided
formed between the sensing element andthe weld joint the electrode is used within the current-carrying capacity
(d) Variations in the location of a reference line range for its specific type (see Table 4) and diameter, with
drawn parallel to the joint to be welded as observed by an sufficient inert shielding gas. Tungstenretains its hardness
optical sensing element even at red heat.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices I 15

Table 4
Typical current ranges for tungsten electrodes
a

frequency
DCSP,
DCRP,
High
unbalanced High frequency
balanced wave,
A A wave, ac, A ac, A
EWP EWP
EWTh-1,
EWTh- 1, EWTH-1,
EWP
EWTh-1,
EWTh-3
EWP
EWTh-3
Electrode d i m . , EWTh-2, EWTh-2, EWTh-2, EWTh-2,
in. mm EWTh-3 EWTh-3 EWZr EWZr
0.010 0.25 up to 15 b up to 15 up to 15 b up to 15 up to 15 b
0.020 0.5 5-20 b 5-15 5-20 b 10-20 5-20 10-20
0.040 0.1 15-80 b 10-60 15-80 10-80 20-30 20-60 20-60
1/16 1.6 70-150 10-20 50-100 70- 150 50- 150 30-80 60- 120 30- 120
3/32 2.4 150-250 15-30 100-160 140-235 100-235 60-130 100-180 60- 180
1/8 3.2 250-400 25-40 150-210 225-325 150-325 100-180 160-250 100-250
5/32 4.0 400-500 40-55 200-275 300-400 200-400 160-240 200-320 160-320
3/16 4.8 500-750 55-80 250-350 405-500 250-500 190-300 290-390 190-390
114 6.4 750-1000 80-125 325-450 500-630 325-630 250-400 340-525 250-525
a. All valuesare based on the use
of argon as the shielding gas. Other current values may
~~
be used, depending upon the shielding
gas, type of equipment,
and application.
b. These particular combinations are not commonly used.

Tungsten electrodesare commercially available in cates metal contamination.

diameters from 0.010 to 1/4 in. (0.25 to 6.4 mm) and in 2.7.2.3 Atmospheric Contamination. The most
lengths to 24in. (610 mm). Electrodes may be pure common causes of atmospheric contaminationare insuffi-
tungsten or tungsten alloyed with zirconia or thoria. Table cient gas flow, excessive eIectrode extension, or insuffi-
3 lists the various types of tungsten electrodes commonly cient postweld purge time. The gas flow rates will vary
used along with their AWS classifications, color codes, widely, depending upon the shielding gas used, nozzle
availabIe finishes, and normal applications. size, and specific weldingconditions. Manual welding
2.7.2FactorsAffecting Electrode Life. Although requires flow rates of at least 15 cubic feet per hour (7.05
the tungsten electrode is considered nonconsumable, liters per minute) with argon. Helium requires slightly
electrode life is determined by operating conditions. The higher rates. The effectiveness of the gas shielding, even
life of any tungsten electrode is shortened by excessive with adequate flow, can be greatly diminished with exces-
welding current, metal contamination, and atmospheric sive electrode extension. A rule-of-thumb for electrode
contamination. extension is:the electrode should extend beyond the nozzle
2.7.2.1 Excessive Current. Good practice requires a maximum of one times the diameter of the nozzle;
use of an electrode diameter consistent with the operating i.e., when using anumber 6 nozzle (3/8 in. [9.5 mm]), the
current being used. Table4 lists various types of tungsten tungsten electrode should extend not more than 318 in.
electrodes with typical current ranges for each. Currents (9.5 mm) beyond the nozzle. Also, the effectiveness of a
greater than the recommended range for a respective elec- gas shield can be increased with the use of a gas lens
trode size and type will cause melting and vaporization of attachment. A too short shieIdingpostflow may also cause
the electrode tip. The molten weld metal may be contami- atmospheric contamination. The postflow should be suffi-
nated and possibly alloyed with tungsten as a result. A ciently long to allow the tungsten to cool below itsoxidiz-
given size electrode has greatest current-carrying capacity ing temperature.
with DCSP; less with ac; and still less with DCRP. Water can be another source of electrode contamina-
2.7.2.2 Mefal Contamination, The most common tion. Water contamination may result from leaks in the
type of electrode contamination is metal contamination. water-cooled torch caused by loose connections, pinched
Themetal may comefrom the filler metal, the molten weld O-rings, defective seaIs,etc. Also, water contamination as
pool,orthe base metal being welded.Particularly in a result of condensate forming in thewelding torch head
manual welding, the welder may dip the electrode into the can occur with no leaks in thetorch. Condensate can occur
weld pool or touch the electrode with the filler metal being when cold water coolant is circulated through the torch
added. The contaminated surface of the electrode must be during warm, humid days. This problem can be corrected
removed. If a specific electrode geometry is used, the by using warm cooling water.
electrode must also be reprepared so that clean tungsten is Tungsten electrode contamination, by either atmos-
available for welding. When contamination of the elec- phere or water, is apparent by the color change on the
trode occurs, the welding arc becomes more diffuse. This electrode. The color may range from a bright blue or
change in arc action is obvious to the operator and indi- purple to black. If the tungsten has a blue-black color, the

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 16 / RECOMMENDED
PRACTICES
FOR GASTUNGSTEN ARC~WELDING

oxidized portion of the electrode must be removed andthe

cause of contamination determined and corrected before
welding is continued.
2.7.3 Electrode Preparation andTip Geometry The
tungsten electrode may be used with no end preparation,
with the end beveled to a specific included angle, or with
the electrode end “balled.” Various electrode tip geomet-
ries affect the weld bead shape and size.’ For example,
with DCSP at 90 A welding on 1/4 in. Alloy 600 plates,
changing the end bevelor included angle from 30 degrees
to 120 degrees (going from a pencil point to a blunt tip)
will increase the width ofthe weld beadand decreaseweld
penetration. However, as the welding current magnitude is
increased, the opposite is true. For example, when weld-
ing 1500 A285 alloy steel at 300 A DCSP, changing the
electrode tip included angle from 30 degrees to 120 de-
grees will decrease weld bead width by a factor of 2 and
increase weld penetration by 45 percent. The degree of
taper or beveling of the electrode tip also influences the
tungsten’s erosion rate. Tapers overa length of 2 to 3 times
the diameter will minimize electrode erosion.
. When electrode geometry is achieved by grinding
methods, the grinding should be with a special fine grit
hard abrasive wheel, used only for preparing tungsten
electrodes. The electrode may be contaminated with parti-
cles from wheels used for grinding other materials. The
electrode should be held at right angles to the abrasive
wheel face during grinding so that the grinding direction is
90 degrees to its length (see Fig. 12).
Another method of fungsten preparation is to set the
welding power source amperage specified in Table 4 for
the specific electrode type and size. An arc is then initiated Filler metal AWS Specification
~~

with a high frequency arc-starting mechanism on a water-

cooled copper block. As soon as the arc is established, the Carbon steel A5.18
electrode end will show a bright orange color. As the Copper A5 .’7
current is increased, the color will change to a brilliant Chromium and chromium-nickel A59
, white and the electrode will melt, forming a ball on the Nickel A5.14
end of the electrode.. At this point, thë current is shut off Aluminum A5.10
. and a uniform diameter ball will remain on the end of
Titanium A5.16
the electrode. Surfacing AS. 13
Regardless of the type of the electrode geometry
selected, the most important factor is that a consistent The filler metals are usually alloyed to resist porosity
electrode geometry be used once a welding procedure is and fissuring in the weld deposit due to high arc currents,
established. Since changes inelectrode geometry canhave highpooltemperatures,andpoolagitation.Filler
a significant influence on weld beadshape and size, elec- metals for weldingcarbon and low alloy steels are alloyed
trode tip preparation is a welding variable that should be with manganese and silicon. Tifanium is most often used
included in the welding procedure. to control porosity for weldingthe stainless steels andhigh
2.8 Filler Metal. nickel alloys.Manganese, columbium, molybdenum, or a
2.8.1 Composition. The filler metal compositions combination of these can be used to control cracking,
used for GTAW are in general the same as those used for 2.8.2 Quality and Identification. The wire surface
gas metal arc welding (GMAW). These filler metals nor- must be free of all drawing lubricants and oxides to ensure
mally have compositions similar to the base metals being high quality welds with the correct composition filler
welded. However, forsome considerations, such as corro- metal. Nearly all manufacturerssupply filler metal that has
been properly cleaned, packaged, and identified. How-
ever, once the filIer metal has been received bythe user, its
1. The Effect of Electrode Geometry in Gas Tungsten-Arc Welding, proper storage and care become an in-house function.
Weldirg Joftrrral,November 1965. - Opened packages, dirty gloves, andgreasy or dirty weld-
.~
~--
2. Investigation, Welding Jounrul, April 1975.
- ~ .~ ~
. ~ing
. equipment are all poor shop practices which may
-- - .~

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices / 17

contaminate the filler metal and result in defective welds. Other examples of preforms are T-rings or Y-rings,
Good shop practice is to keep the filler metal in a clean, which are roll-formed and then machined to close toler-
dry cabinet which is well identified so that filler metals ances for components that require precision welding, such
will not become mixed. Filler metal should always be as certain aerospace or nuclear applications. Machined
returned to the proper storage cabinet when not in use. preforms are normally used in conjunction with the au-
In instances where special cleaning, packaging, or tomatic welding systems, where precise control of the
identification of filler metals is desirable, a specification welding variablescan bemaintained.
may be prepared by the user for purchasing the needed
filler metal. 2.9 Fixturing
2.8.3 Feeding Techniques. The three methods of in- 2.9.1. Principles Governing Fixture Design
troducing filler metal into a weld joint are: (1) handfeed- 2.9.1.1 General. The decision to use fituring for
ing by the welder; (2) automatic feeding by a wire feeder; the fabrication of a weldment is generally governed by
and (3) preplacing the filler metal, either as a consumable economics. Close tolerance requirements for exceptionally
insert or as a filler metal overlay. high quality work may also dictate fixturing. A decision
The torch should be held at between 75 and 90 degrees not to build fixturing usually means that the part to be
to the work forall manual welding. A slight inclination in welded can be hand fitted and tack welded together and
the forelíand positionis necessary for good visibility. An will be entirely self-supporting during the welding opera-
acute angle can cause aspiration of air into the shielding tion tnd that the resulting distortion can be tolerated or
gas and thus contaminate the weld. corrected by straightening operations.
The filler metal should be carefully added at theleading Some functions of a weld fixture are:
edge of the molten weld pool to avoid contact with the (1) To locate parts relative to the assembly
electrode. The hot end of the filler metal should always be (2) To maintain alignment during welding without ex-
kept in the protective atmosphere of the shielding gas. cessive restraint (should notpromote weld cracking)
Excessive agitation of the molten pool should be avoided; (3) To control distortion in the weldment
it should be kept as quiet as possible to prevent vaporizing The size of the weldment in itself may require that support-
the deoxidizing elements. In manual welding, the filler ing fixturing be made, or a large number of small as-
metal is introduced much the same as the filler metal is semblies might be produced more economically with
addedin oxyacetylene welding. In automatic welding, the locating and clamping fixturing.
wire is fed mechanically into the leading edge of the weld The fixturing decision is even more complex when the
pool. Preplaced filler metal, such as a consumable insert, welding process appIication is considered. Selection of the
is often used, particularly in pipe welding applications. most economic process to meet the quality requirement
Once the insert is tacked into place, the welderneed might require movement ofan extremely large weldment,
concentrate only on fusing the filler metal to the work- with very highfixturing costs. Selection of another weld-
piece. ing process (more expensive in itself) mightpermit weld-
Regardless of the method of filler metal addition, it ing progression by the weld station alone, with the large
should be remembered that these filler metals contain part stationary. Thus, once it is determined that fixturing is
elements specifically added to resist hot cracking and required, economics of fixture constructionwill determine
porosity. Maximumbenefitfromthoseelementsis whether to move the work or the torch. The GTAW process
achieved whenthe completed weld consists of at least 50 is readily adapted to allthe foregoing considerations.
percent filler metal. 2.9.1.2 Effect of Heat Sink.Achieving an accepta-
2.8.4 Filler Meta1 Size and Shape. Filler metal is ble GTA weldgenerally means that a stable, reproducible
available either as stright lengths, normally 18 or 36 in. thermal pattern was produced. The heat received into the
(460 or 900 mm) long, or on spools or reels. Spool size molten weldpool goes initially to melting base metal and
will vary from a 2 lb (0.9 kg) spool for the very small preheating ahead of the weld. Nearly all of the heat is
diameter filler metals to 12 and 25 lb (5.4 and 11.3 kg) initially dissipated through the base metat, and thermal
~ S 50 to 60 lb (23 to 27 kg) reels. The diameter
S ~ O O and gradients are established, depending upon the geometry,
range is from 0.020 to 1/4 in. (0.5 to 6.4mm). The sizes the mass involved, and base metal diffusivity. If metallur-
most often used are 1/16, 3/32, and 1/8 in.(1.6, 2.4, and gical or quality considerations dictate higher thermal gra-
3.2 mm) nominal diameters. dients and chill rates than the weld assemblyconfiguration
Filler metal shapes other than round wire are usually in willfurnish,thensupplementalchillingcanbeac-
the form of preformed rings or rectangular wires. The complished with the addition of chill bars. The influence
preformed filler metal insert is of the same composition as of these bars depends upon their mass, location, and dif-
other forms andis fused into the weld joint during the first fusiuity. Also, the condition of the metallic interface be-
weld pass. The advantages of preforms are uniformity of tween the chills andthe base metal is of major importance.
results and precise control over the chemical composition Oxides and foreign material can reduce heat transfer by
of the weld deposit. Pipe welding is an example of a making intimate contact impossible. The chills should be
preformed shape providing for uniformity of the internal flexible and ductile and held in place by uniform unit
weld condition. pressure of sufficient magnitude.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS CS.5 B O W 07842bS 0002b07 B W

18 / RECOMMENDED FORGASTUNGSTEN ARCWELDING

PRACTICES

High chill configurations can be difficult to weld, since practices, including all joinf weldingprocedures, involved
they demand more of the arc heat for preheating. When in the production of a weldment.” Usual informationto be
welding speed islowered to increase heat input, heat considered for a welding procedure specificationincludes:
utilization becomes even less efficient. Erratic high chill Base metal andfiller metal
systems produce variable nuggetand heat-affected zones, Amperage
since the heat available for melting is notstable. Low chill Voltage
systems are sometimes used to eliminate these problems. Welding speed (travelspeed)
Whatever chill approach is taken, it mustbe compatible Wire feed speed and number of passes
with the welding conditions thatare developed. Shielding gas type and flowrate
2.9.2 Clamping. Pressures required for thermal con- Backup gas type and flowrate
siderations can be a factor in weld consistency. Pressures Tungsten size, type, and electrode tip preparation
required to align and hold parts before, during, and after Arc length
welding have largely been determined by trial and error Joint design
and are not usually a problem to a tool designer. Well- Fixture type (if used)
formed parts generally require only nominal pressures for
(b) Welding schedule: “A chronological listing of
alignment and containment during the expansions and
welding sequence and related activities whichmustbe
contractions of welding. Weld tooling can be classified as
followed in their respective order for the purpose of re-
tool supported or part assembly supported. Toolsthat
producing the desired weld.” The welding procedure
support as well as clamp are normally heavy andcapable
will reference the welding schedule for thespecific
of producing more than enough closing force, while as-
information.
semblies with self-contained tooling are held with lower
pressures. These concepts have come to be loosely known (c) Certain factors must be declared prior to the es-
as “hard” and “soft” tooling. On large structures, soft tablishment of the welding schedule or procedure. These
tooling has economic advantages but usually requires a factors directly relate to the level of effort that will be
low chill utilization of the clamping surfaces to maintain placed on schedule and procedure establishment, and will
weld bead consistency. Preweld dimensional control of include but are not limitedto the following:
parts is usually stringent with the soft tooling approach. Quality of weldrequired (level of quality to which the
The GTAW process yields excellent results with either weld must be tested and accepted)
chill concept. Welding variables will usuallybe different Specificationrequirements (personnel qualifications,
between hard and soft tooling to account for heat sink procedure qualifications)
differences. Production schedule
Clamping or closing a tool can be a tedious manual Tooling ability or capacity
operation (such as turning down a number of C-clamps), The welding schedule and procedure may be established
or simply engaging a lever or pushing a button on automa- after determining the above factors. The procedure and
ticsystems. Generally, pneumaticormechanical- schedule may be quite involved, depending upon the in-
pneumatic systems are quick closing, while completely fluencing factors, or may be minimal in the case of a
mechanical systems take longer. Pneumatic systems usu-
manually produced weld.
ally are directly expanding, with hoses between the tool 2.10.2 MonitoringandRecording. Instrumentation
and the chills. Pressures up to 100 psi (69 kPa) are com- for monitoring and recording welding activities has be-
mon in the hoses. Very uniform mating pressures can be come increasingly important with the mechanization of
obtained with these systems. Mechanical-pneumaticsys- the weld system. The cost of commercially available
tems may consist of a series of levers, one arm to a work equipment varies from a few hundreddollars for a single-
contact shoe and the other to a portion of an expandable channel monitor up to several thousand dollars for a
hose. These systems can use various lever arm ratios for multiple-channel recorder. Increased demands for quality
mechanical advantage and apply high clamping or align- control measures require incorporating these recording
ment forces. instruments into weld systems in order to compare weld
2.10 Welding Schedule and Procedure (See Tables 6 & information with known standards.
7.) Recorders are used to provide accurate data regarding
2.10.1 Establishing Welding Activities.The majority shielding gas flow, backup gas flow, weld travel speed,
of industrial welding activities are governed by predeter- shielding gas moisture content, weld temperatures, and
mined welding schedules and procedures. Establishing amperage and voltage values.
these procedures and schedules may be quite involved, The need fora recording instrument is dependent upon
depending upon the application and specification re- the weldment specification or the degree of necessary
quirements. The following definitionsarelisted to quality control measures. Also, recorders used with auto-
distinguish between a welding schedule and a welding mated weld programs provide definite and useful quality
procedure. measurement.
(a) Welding procedure: “The detailed methods and 2.ll Welding Equipment Setup. GTAW equipment is

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Table 6
Typical welding procedure for manual
of carbon steel
gas tungsten arc welding
Material thickness, in. (mm) 1/16 - 118 (1.6 - 3.2) 118 - 114 (3.2 - 6.4) 114 - 112 (6.4 - 12.7)
Joint design Straight butt Single-V-groove Double-V-groove
Current, A 50-100 70-120 90-150
Polarity DCSP DCSP DCSP
Arc voltage,V 12 12 12
Travel speed As required As required As required
Electrode type EWTh-2 EWTh-2 EWTh-2
Electrode size, in. (mm) 3/32 (2.4) 3/32 (2.4) 118 (3.2)
Filler metal type E7OS-3 MOS-3 E7OS-3
Filler metal size, in. (mm) 1/16or 3/32 (1.6or 2.4) 3/32 or 118 (2.4 or 3.2) 3/32or 118 (2.4 or 3.2)

Shielding gas Argon Argon Argon

Shielding gas flow rate, cfh(literlmin) 20 (9.4) 20 (9.4) 25 (11.8)
Purging gas Argon Argon Argon
Purging gas flow rate, cfh (literlmin) 5-7 (2.4 - 3.3) 5-7 (2.4- 3.3) 5-7 (2.4- 3.3)
Nozzle size 318 (9.5) 318 (9.5) 112 (12.7)
Nozzle-to-work distance, in. (mm) 1/2(12.7)max 112 (12.7)max 1/2 (12.7)max
Preheat, min 60" F (16" C) 60" F (16" C) 60" F (16" C)
Interpass temp., max 500" F (260" C) 500" F (260" C) 500" F (260" C)
Postweld heat treatment None None None
Welding position F, H,V, OH F, H, V, OH F, H,V, OH

Table 7
Typical welding procedure for manual
gas tungsten arc welding of stainless steel
Material thickness, in. (mm) 1/16- 1/8 (1.6- 3.2) 118 - 114 (3.2 - 6.4) 114 - 112 (6.4 - 12.7)
Joint design Straight butt Single-V-groove Double-V-groove
Current, A 50-90 70-120 100-150
Polarity DCSP DCSP DCSP
Arc voltage, V 12 12 12
Travel speed As required As required As required
Electrode type EWTh-2 EWTh-2 EWTh-2
Electrode size, in. (mm) 3/32 (2.4) 3/32 (2.4) 3/32 (2.4)
Filler metal type ER-308 ER-308 ER-308
Filler metal size, in. (mm) 1/16 or 3/32 (1.6or 2.4) 3/32 or 118 (2.4or 3.2) 3/32 or 118 (2.4 or 3.2)

Shielding gas Argon Argon Argon

Shielding gasflow rate, cfh (litedmin) 20 (9.4) 20 (9.4) 25 (11.8)
Purging gas Argon Argon Argon
Purging gas flow rate, cfh(literlmin) 5-7 (2.4- 3.3) 5-7 (2.4- 3.3) 5-7 (2.4- 3.3)
Nozzle size, in. (mm) 318 (9.5) 318 (9.5) 112 (12.7)
Nozzle-to-work distance, in. (mm) 112 (12.7)max 112 (12.7)max 112 (12.7)max
Preheat, min 60" F (16" C) 60" F (16" C) WF(160C)
Interpass temp., max 500" F (260" C) 500" F (260" C) 500"F (260" C)
Poshveld heat treatment None None None
Welding position F, H, V, OH F, H,V, OH F, H, V, OH

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 20 / RECOMMENDED FORGASTUNGSTEN ARCWELDING
PRACTICES

designed to perform numerous functions controlling weld variable follow relatively rapid variations in the input over
quality, described and analyzed in the discussion of pro- a fairly wide range, as well as to conform withthe input in
cess variables. The number of variables controlled by the the presence of load disturbances. Such a system must
operatorduring welding depends upon the welding contain a closed loop. An automatic arc voltage control
equipment setup.The basicvariables include welding system is an example. This type of control system is
current, arc voltage, travel speed, wire feed, and inert gas classified as a servomechanism and is defined as “an
coverage of the molten pool. error-sensitive, follow-up, amplifying system permitting a
Refinements of manual GTAW include foot controls for wide range of input command remotely located from the
welding current, travelspeed controls, wire feedman- element being controlled.” This type of servomechanism
ipulators, and on-off switching. is employed by the new feedback controls generally de-
Automatic GTAW is extensively used. The degree of scribed as adaptive controls.
mechanization varies from simply mounting the torch in a There is need for a means of compensation for uncon-
bracket that moves over the workpiece to an automatic trollable input variables even witha properly functioning
operation for the total weldingcycle. Thedegree of GTA weld qualitymonitoring system. For example: In the
mechanization is usually determined by the number of welding of a large booster, mismatch, gapping, tack
identical welds and the speed and quality desired. The welds, and heat sink variation caused a variation in grc
aerospace and the nuclear industries use automatic GTAW penetration, even though the process was controlled with
extensively, notnecessarily because of the large quantities very close accuracy. The problem demonstrated a need to
of production parts involved, butbecause the weld quality automatically close the loop (or feed back information)
required can often only be achieved withthe control inher- between the work and the welding equipment; i.e., find a
ent in automatic welding. way to feed back signals that could change weld setting
A workable weld processoperation has to be carefully automatically. Potential solutions required automatic
analyzed and available welding equipment studied to de- compensation (1) to increase the welding current when a
termine the degree of automation possible for a welding local increase of heat sink was taking heat away from the
application. weld area, thereby reducing penetration; (2) to reduce the
2.11.1 Electrical Design Variables.Feedback control weld current when penetration increased (due to the vary-
has been used to establish the basis for analysis of the ing gap width betweenjoint faces); and (3) to increase the
required degree of automation in the electrical design of welding current when a mass of tack weld was ap-
welding equipment. Feedback control can be established proached, and thenreduce current when leavingthe vicin-
in open-loop or closed-loop systems if a value for com- ity of the tack weld.
parison with a reference or desired valueis present for the Several designs of adaptive controls are in use. Each
variable to be controlled. adaptive control uses a sensor that continuously derives
In an open-loop control system, the static correspon- information concerning the dynamic penetration of the
dence between the controlled variable (systemoutput) and weld in progress. A second element constitutes a simple
the reference quantity (system input) is a function of the computer, which accepts information from the sensor and
calibration of the control mechanism. In a closed-loop increases or decreases the welding current a proper
system, the amount of corrective response is determined amount to assure uniform preset weld penetration. A block
by the controlled variable as well as the input command. diagram of an adaptive GTAW penetration control system
Therefore, the static correspondence between the output is illustrated in Fig. 13, Additional information on adaptive
and input is more even in a closed-loop system than in an control systems is published in “Automation of the Gas
open-loop system. lbngsten Arc Welding Process” by E. P. Vilkas, Welding
Control devices may be further subdivided on the basis Journal, 45 ( S ) , pp. 410-416 (1966).
of functional area and the nature of the input. 2.11.2 Mechanical Design Variables. Measurability
Some GTAW process variable controls are designed for has been used as the recommended basis for mechanical
the specific purpose of maintaining a correspondence design. Mechanical measurability is vital to automatic
between the controlled variable andaninput which GTAW. Measurability of the mechanical welding machine
changes infrequently. A sysfem of inert gas flow control variables can be broadly classified as static and dynamic.
may be considered an example of an open-loop control Positioning of the parts to be welded is almost as important
system. The output is caused to change infrequently in as the mechanism of a wire feeder, a wire guide, and the
accordance with the input (state of flow rate and pressure arc length control dynamicresponse in many applications.
control devices). The flow of inert gas is in no way affected The setup for automatic GTAW may be simple or com-
by the welding arc and does notdiffer from that called for plex, as dictated by the configuration of the component
by the inert gas control. Suchan open-loop control system parts and the alloys being fused. In tube and pipe welding
is properly termed a “regulator,” since its purpose is (without the addition of filler metal) the setup is relatively
merely to adjust the output to a desired constant value. simple and allowsthe entire welding operation to becom-
- Other controls are designed tomake the controlled pletedwithout constant observationand adjustment of

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Recommended Practices I 21

a Uncontrollable
input variable
I
Measurable
output
Penetration variables Gas tungsten Controllable input variables
*
sensors arc welding
t t t t
Corrections Electrode
Arc Current Travel Seam
* head sourcefeed speed track

Sequence control
A
Penetration
Adaptive
measurement t
control
' system
A A conditions setup Machine
Desired Actual
penefration conditions
criteria

Fig. 13-Block diagram of adaptive GTXW penetration control system

mechanism or controls by an operator. GTAW of hyd- tions. Reproducibility of tungsten electrode position with
raulic, pressurant, fuel, and instrumentation lines in the respect to theweld joint is assured by measurability of all
aerospace and nuclear industries is performed automati- mechanical variables.
cally, The automatic weld sequence with correct weld The basic elements in the tube or pipe welding system
settings eliminates human error in tube welding applica- are illustrated in the block diagram in Fig. 14.

i""-""" -J L """_ "I

controls Mechanical controls ElectricaI

Fig. 14-Diagram of basic elementsin tube or pipe welding

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 22 1 RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

The mechanical variables are controlled by the follow- Changing from manual to mechanized welding usually
ing means: requires a sound knowledge of welding, a sophisticated
(1) Tube or pipe outside diameter tolerances andalign- machine design, and a high initial investment. Neverthe-
mentoftwo tube ends. Spring type, replaceable tube less, GTAW is readily adapted for automation through the
clamp inserts accomodate tube OD tolerances and accu- availability of a wide variety of mechanical and electrical
rately and positively position andalign tube ends without controls,
deformation. Also, in tube or pipe GTAW, a unique nozzle design
(2) Mechanical guidance of the ring type electrode hol- creates a miniature chamber.The nozzle surroundsthe weld
der. A split type precision housing of high heat and wear joint with inert gas without creating any gas pressures
resistant material insulates and contains the electrode hol- during the complete weld cycle. This allows the welding
der. The split type electrode holder accurately maintains of such difficult-to-weld reactive materials as titanium
the electrode-to-joint spacing alignment andisdriven and others. The gas purges solelythrough a designed
around the joint by a miniature high torque dc gear motor, clearance between the nozzle and the outside of the tube.
with tachometer, mounted within the welding head Many GTA welding process variables, such as mate-
handle. rial, inert gas, and electrode quality, are regulated by
Another GTAW application with a high degree of material quality specifications and are relatively constant.
measurability is arc spot welding of automotive compo- However, material preparation, including cleaning pro-
nents .3 cedure or process controls, eliminates many welddefects
The more complex assemblies may be welded automat- if provided with a closed loop feedback feature.
ically or semiautomatically but require elaborate tooling,
fixtures, and the techniques of a skilled machine operator
who can make applicable machine adjustments.
Semiautomatic pipe welding with filler metal addition
is not especially complex. Pipes maybe welded in place if
clearances allow the weld head to travel around the pipe
circumference, Otherwise, a manual weld withthe aid of 3. Welder and Welding Engineer
mirrors and unrealistic positions is required. Setups re- 3.1 Welder Tkaining and Qualification. The type and
quiring such manual welding are more likely to con- quality of work which the individualis expectedto perform
tain defects that can be avoided with semiautomatic pipe dictates the extent to which a welder or welding operator
welding. should be trained. The welder responsible for joining
Automatic GTAW may beeasily adapted for high pro- nuclear pressure vessels and pipes requires more training
duction of either simple or complex small assemblies. than one responsible for flat position fillet or lap welds
These types of assemblies have included electronic com-. only.
ponents, hydraulic assemblies, power plant parts, and The training program is by one or a combination of two
instrumentation closures, among others. The welding general methods: (1) within the plant itself, and (2) by
setup has provided for automaticloading and unloading of outside schools. The training program should be geared to
parts, the necessary purging with inert gas, and the weld- the company’s needs andbe a mixture of basic theory and
ing operations. The various operations may also be per: practice. Minimum requbements for the training of wel-
formed in an atmospheric chamber, to accommodate the ders are listed in AWS E3.1-75.
inert gas requirements either for welding or in the filling Once the individualhas been allowed sufficientpractice
and sealing of inert gas within the component assembly time (which varies depending upon the particular assign-
being welded. ment) andhas attended a training program, a qualification
2.11.3 Environmental Design Variables.Gas shield- test is usually given.
ing has been used as the basis for analysis and design of Welder qualification tests are primarily given to the
equipment setup with respect to environmental variables. individual to determine that person’s ability to produce
Manual GTA welding of many complicated assemblies is sound welds that meet the requirements of the particular
best performed in so-called dry boxesorchambers. code or specification involved. There are several types of
High quality inert gas atmosphere in a chamber is qualification tests, and many companies have their own
achieved by means of vacuum pumps evacuating the at- test which the weldermust pass. A qualification test
mosphere and inert gases back-filling the chamber. The should be designed to represent the type ofwork the
size of the chambers used depends upon the requirements. welder willbe required to do, even though noone test can
For example, in nuclear applications, room size chambers apply to all types of work.
have been used. Mechanized welding is usually employed
for small assemblies, such as nuclear fuel rods, which 3.2 Areas of Responsibility. Responsibilitiesof the wel-
must be welded ina chamber and can be rotated. der, the welding engineerltechnologist, and the welding
technician vary, depending upon the company and itspro-
3. Vilkas, E. P., Automation of the Gas Tungsten Arc Spot Welding duct. Generally, these responsibilities are as shown in the
Process. Welding Journal, January 1966. following chart.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Typical Equipmentfor Process Applications 1 23

Welding
engineer1
Welding
Welding
operator,
Responsibilities technologist technician welder
manual
Prepare welding
specifications *
Purchase equipment
(prepare specifications) *
Prepare welding procedures * *
Recommend NDE activities
and assist when necessary * *
Provide up-dating training
sessions on new processes *
Provide feedback
(improvements, suggestions) * *
Follow-up shop work * *
Provide plant assistance
and troubleshooting * *
Establish work priorities for
T
new weld development
Qualification and testing of
weld personnel *
Oversee preventive
maintenance of equipment I

Produce welds per

approved procedure *

4. Quality Control producer must establish a quality levelthat applies to the

specific product. The producer's specificationmust define
4.1 Inspection and Test Methods. GTA welds are in- the test procedure and the acceptance standards to assure
spected by the same methods used for other fusion proces- uniformity in testing and evaluating the product;
ses. The as-welded surface ofGTA welds is generally
adequate for inspection without conditioning.
Visual inspection is one of the most valuableinspection
methods and should include all factors affecting the qual- 5. mical Equipment for
ity of the weld such as edge preparation, cleaning, and Process Applications
alignment, as well as the visible condition of the com- 5.1 Manual Gas nngsten Arc Welding
pleted weld surface. Common weld defects disclosed by 5.1.1 General. Manual GTAW is most applicable
visual inspection are incorrect weld size, inadequatejoint where complex shapes preclude the use of automatic
penetration, undercutting, surface porosity, and cracking. methods. Manual electrode holder manipulation is gener-
The welder should detect these conditionsand take correc- ally used for irregularly shaped parts that require short
tive action to preventrejection in the final stages of product welds or forwelding in difficult-to-reach areas. The
fabrication. GTAW equipment required for manual welding is illus-
Other inspection methods used, depending upon the trated in Fig.15.
requiredproductquality level, areliquidpenetrant,
Arc striking may be accomplished as discussed in
magnetic particle, ultrasonic, eddy current, and radiog-
Section 2.3.
raphic. Section I, 6th edition, of the AWS Welding Hand- Once the arc is started, the electrode holder (torch) is
book and AWS Welding Irzspectioninclude information on held with the electrode positioned at an angle of about 75
these methods of inspection. degrees to the weld pool surface. To start welding, the arc
4.2 Specifications. The purpose of any inspection is to is usually moved in a small circle until a suitable size pool
ensure a predetermined quality level in the finished pro- of molten metal is obtained. When adequate fusion is
duct. Since the qualityrequirements may vary widely with achieved at any one point, a weld is made by gradually
service conditions, design factors, and economics, each moving the electrode along the parts to be welded in order
~~

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C 5 - 5 B O W 0 7 8 4 2 6 5 0002bL5 3 W 1

24 I RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

Note: Sometimes a water circulator is used.

Electrode holderA

Welding
machine

yas
I
I,
I
Fig. 15- Schematic diagramof gas tungsten arc equipment

to progressively melt the adjoiningedges. Solidificationof moves in the opposite direction, but the filler rod is at all
the molten metal follows progression ofthe arc along the times near the arc and feeding into the weld pool within the
joint and completes the welding cycle. shielding gas blanket.
Welding is stopped either by gradually withdrawing the All thestandardtypejoints,suchassquareand
electrode from the workpiece or by tapering off the cur- V-groove, T, and lapjoints, may be welded by this
rent. The former system is used often when dc power is process.
employed. The latter is more common with ac welding, The GTAW electrode holder must have sufficient weld-
since the high frequency oscillators normally used with ing current capacity to prevent overheating. Collets must
alternating current usually require on-off switching for be available to accommodate the correct sizes of tungsten
control of the high frequency spark. Foot controls for electrodes. Each diameter and type of tungsten electrode
current and on-off switching are used for high quality ac has a recommended welding current range (see Table 4).
and dc welding. The nozzles of an electrode holder are made from various
The metal thickness and joint design, together with heat resistant materials in different diameters, shapes, and
metallurgical properties, determine w,hether filler metal lengths. Length and shape are selected on the basis of joint
need be added to the joints. Filler metal, if added, is accessibility and the required clearance between the noz-
applied by manually feeding the filler rod into the pool of zle and the work. The nozzle should be large enough to
molten metal in the arc region, in muchthe same manner provide complete inert gas coverage of the molten weld
as in oxyacetylene welding. One of the most frequently metal.
used techniques forfeeding filler rod is illustrated in 5.2.2 Power Sourceand Welding Current. The
Fig. 16. power source is chosen to provide the type of welding
The filler rod is usually held at an angle of about 15 current needed for the metal under consideration. The
degrees to the surface of the work and slowlyfed into the welding current may be ac, DCSF’, or DCRP. The welding
weld pool. The filler rod must not be removed from the current magnitude is adjusted to provide the desired
protection of the inert gas shield during welding. Another amount of heat. Alternating current is usually preferred
method is to press the filler rod in line with the weld and for the manual GTA welding of magnesium, aluminum,
melt it along with the joint edges. This method is often and their alloys because of the cleaning action (oxide
used forthe root pass in multiple-pass weldingof V-groove removal) that takes place in the electrode positive half-
joints, A method used frequently in weld surfacing and in cycle ofthe alternating current. Power sources have been
making large welds isto feed filler metal continuouslyinto designed to produce a balanced wave alternating current
the weld pool by oscillating the filler rod and arc from side (half-cycles of equal magnitude) for the welding of
to side. The filler rod moves in one direction while the arc aluminum and its alloys. The balanced wave increases the

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Typical Equipmentfor Process Applications 1 25

Direction of
weldina

A. Develop the pool

back torch
B. Move D. Remove rod

C. Add filler metal E. Move torch to leading

edge of pool

cleaning action. Ac or dc power sources normally used for intermittently by the arc welding gun trigger. Fillermetal
welding with coveredelectrodes may be used for manual feed is controlled byadjusting feed speed, feed time inter-
welding. Special dc or ac power sources that are designed val, and dwell time between feed intervals.
Sinceeach feed
specifically for use in GTA welding aie available with interval can be used with any of four dwell times, these
automatic means for controlling the flow of gas and water variables can be adjusted to deliver the right amount of
and the start and stop of welding. (Forfurther detail, refer filler metal at the right time.
to Welding Handbook, í'th ed., Vol. 2, Chapter 1, Arc Semiautomatic GTA spot welding equipment is an as-
Welding PowerSources.) sembly of apistol-like electrodeholder, a water-cooledgas
5.2 Semiautomatic Gas 'hngsten Arc Mklding. Semi- nozzle, a tungsten electrode caficentricallypositioned
automatic gas tungsten arc welding is defined by AWS
as welding with equipment which controls only the filer
metal feed. The advance of the welding is manually
controlled.
Semiautomatic electrode holders for gas tungsten arc
welding wereintroduced about 1952but were neverwidely
used. A semiautomatic electrode holder is an assembly of
a handheld, water-cooled electrode holder with an attach-
ment thatbrings the filler wire into the arc area. The filler
wire is fed to the arc through the flexible conduit by a
motor-driven wire feeder. The filler wire fed ahead of the
arc helps guide the electrode holder and establish travel
speed. The filler metal is then melted by the arc and
deposited in the joint.
Figure 17 shows a filler metal delivery system used with
semiautomaticcold wire feed GTAW With thissystem the
welder does not haveto feed filler metal into the weld pool.
The system delivers filler metal to the welding arc in
adjustable increments to suit the metal welded and the
welding configuration. The filler wire is fed at a preset
constant feed rate that can be operated continuously or

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C5.5
~" -
80 W 07b-IZb5-~0 0 0 2 b 3 1 7
~~ ~~ ~~ ~
~~

26 / RECOMMENDED FORGASTUNGSTEN ARC WELDING

PRACTICES

with respect to the nozzle, a miniature filler wire feeder Torch

Circular Hold-down
with wire guide arrangement, and a trigger switch for backing bar-,
sequence start. The filler wire addition to the arc iselec-
tronically controlled in conjunction with the spot weld
programmer controls.There is a slight time delay while the
molten spot is established (filler metal added), then the
filler wire is retracted a precise distance as the spot weld
current starts the downslope time. The retraction is neces-
sary to avoid freezing the filler wire with the spot weld
metal.
The nozzle configuration is varied to fit the contour of
the weldment. Arc penetration is controlledby adjustment
of arc length andthe amount and time of current flow. Adjustable
Multiple pulses of current are preferred to one long roller support
( 1 o f 2)
sustained pulse in some applications. Variations in shear
strength, nugget shape, and penetration of the spot weld Floor line
can be minimized with precision control of all variables
(including tungsten electrode tipsand pressure of the
nozzle on the weldments). For additional informationrefer
to A. Lang and T. Rutkay, "Gas Tungsten Arc Spot Weld-
-
Fig. 18 Setup for automatic weldingof
ing 2219 Aluminum on S-1C Booster Program," Welding
longitudinal buttjoints
Journal, 45 (6), 1966.
5.3 Automatic Gaslhngsten Arc Welding. The amount
of automation or mechanizationapplied to GTAW depends The clamping mechanism is composed of two rows of
upon the quantity of identical welds, the accessibility, hold-down fingers that firmly hold the workpiece during
quality control requirements, degree of perfectionre- welding. Each rowmay be independently actuated by
quired in the weldment, and availablefunding. GTAW can means of manual or foot-operated valves that control
be controlled with various devicesthat accept information mechanical, magnetic, hydraulic, or pneumatic pressure
about the desired valuesof the process variables on systems. The intensity of clamping force may be control-
punched tape or cards. In some devices, the input data are led by a pressure regulating device.
stored on a memory drum. Less sophisticatedcontrols that The backing bars are removable to permit use of vari-
makeuse of cam-actuatedpressureormechanical ously contoured groove openings for different thicknesses
switches are also available. and types of metal. The bars may also contain relief
The aerospace industry uses automatic GTAW exten- grooves that permit flow of shielding gas to the underside
sively, not necessarily because of large quantifies of pro- of the weld (as in Fig. 18) and may have provision for
duction parts, but because the quality required for aero- thermostatically controlled heating or cooling.
space designs often can be achieved only withthe control Pedestal boom manipulators can be used for rapid posi-
inherent in automatic welding. A typical example is the tioning and welding in any direction to produce internal or
fabrication of large diameter rocket motor cases where external longitudinalor circumferential(girth) welds. This
longitudinal and girth welds aremade by automatic equipment is suitable for use with powerrolls or
GTAW. headstock-tailstock welding positionerson tanks, vessels,
Longitüdinal seamers, often called stake welding fix- pipes, and similar weldments. Rotating positioners, to
tures, are designed to locate and clamp both sides of a which fixtures or other work-holding tooling devices are
straight butt joint, as shown schematicallyin Fig. 18. The mounted, are readily adapted to use with pedestal boom
butt joint may be between two plates, or it may be the manupulators for small production runs or special fabrica-
longitudinal seam of a flat, conical, circular, or rectangu- tions.
lar workpiece. A longitudinal seamer consists of a base Power supplies,. fiiler metal supplies, portable coolant
that contains a weld backing bar to which the work is tanks, automatic welding heads, and remote controls can
clamped by hold-down fingers and an upper beam struc- be mounted on a pedestal'boom manipulator. A typical
ture (not shown in Fig. 18) that supports a traveling car- pedestal boom manipulator that includes some of these
riage and track carrying the welding head, wire-feed features is shown in Fig. 19. Pedestal boom manupulators
mechanism, and clamping fingers. The upper beam struc- can be moved manually on integral wheeled bases to the
ture is cantilevered, secured to the base, and hinged at one work location, or they can be powered to travel at welding
end for support, with the opposite end remaining open to speeds on rails.
permit loading and unloadingof the workpieces. Apivoted Skate welding machines are lightweight motorized
locking latch is provided at the loading end to restrain the travel carriages on which seamtracking devices, a filler
cantilevered upper beam when the clamping force is metal feed system, and the automatic welding head are
applied. mounted. The skate welding machine shown in Fig. 20

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Typical Equipment for Process ApplicationsI 27

In skate welding large assemblies, a mobile service

carriage that contains the power supply,the gas supply, all
the control elements, and the operator is synchronized to
travel with the moving skate carriage during welding.
Such equipment (shown in Fig.21) is satisfactory for long^
duty cycles on large assemblies.
Rotating positioners are probably the most adaptable
type of fixtures for manual or automatic welding, The
positioner worktable is a circular or square plate that has
been machined to a flaf surfaceand is fitted withT-slots for
bolting downvarious jigsor work-holding tools. The
worktable can be rotated for 360 degree circumferential
welds in any plane from flat to 10 degrees past vertical.
Variable rotational speeds and provision forsmooth start-
ing and stopping are mandatory for automatic welding
operations. Pedestal boom manipulators (see Fig. 19) are
used to position the automatic welding head over the joint,
and the positioner worktable may be started manually or
be sequenced to start automatically
The automatic GTAW process may be programmed by
-
Fig. I9 Use of a pedestal boom manipulator N/C (numerical control). Control information is received
from a 1 in. (25 mm) wide, eight channel tape that is
for making a circumferential girth weld
punched in a binary coded decimal (BCD) format. The
information is transmitted in the form of electrical com-
enables precision welding -of massive structures that can- mands to servomechanismson each machine axis, as well
not be handled by conventional rotating positioners and as those provided forthe various welding functions. Feed-
longitudinal seamers. Straight or curved track sections are back devices are used to relay actual conditions to the
attached directly to the workpiece by bolting or by suction machine control unit, which corrects for any differences
cups, permitfing movement of the welding carriage in a between command and actual 'values. Such a system is
continuous path along the weld seam. The skate carriage capable of providing relative motion between the torch and
can be operated in horizontal and vertical directions for workpiece in as many as eight axes, delivering up to
öut-of-positionweIds. twenty controllable welding functions. Lineal axis move-

~. .

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 28 I RECOMMENDED FORGASTUNGSTEN ARCWELDING
PRACTICES

ment is programmable toan accuracy of 0.0002 in. One type of portable welding head for small diameter
(0.005 mm), and rotary motion to an accuracy of 0.001 1/4 to 4 in. (6.4to 102 mm) O.D.pipe is shown in Fig. 22.
degree. The welding functions can be programmed by This welding head is used in one application forcomplete
t-codes in increments as fine as 0.4 A of welding current, joint penetration weldswith a uniform inner bead, in
0.03 volts of arc voltage, and 0.1 in./min (0.04 m d s ) of stainless steel pipelinesof 0.109 in. (2.8 mm) and 0.154in.
wire feed speed. Depending upon the type of tape reader (3.9 mm)wall thickness. The welding programmer in
and means of executionand electrical storage, the welding this instance is up to 100ft (30.5 m) away from thepoint of
variables can be changed anywherein a welding sequence welding, and the welding variables are monitored by a
as frequently as 0.15 seconds. They may not be changed multi-channel recorder for the quality assurance program
simultaneously, however. This in-process change capabil- requirements.
ity is quite essential for such situations
as weld starts, weld The ability to reproduce quality welds witha minimum
overlaps, tapered sections, changingjoint mass conditions, of personnel is the primary advantage of this type of pipe
welding around radial corners, backstepping, variations in welding equipment. Others are that only small clearances
fillet size, and combined weld types. For more informa- are required around the stainless or carbon steel pipe and
tion, see “Advancement of the Numerically Controlled that wall thicknesses up to 0.154 in. (3.9 mm) can be
Gas TungstenArc Welding Process,”G. R. Stoeckinger, et welded in one fusion pass without wire addition (see Fig.
al., Welding Journal,49 (3), pp. 183-193, (March 1970). 23).
Some specific applications where automatic GTAW is Automafic Pressure Vessel Welding
being used in industry are presented below. Various types of equipment may be adapted for cir-
Automatic E4be and Pipe Welding cumferential (girth) welds. The major limiting factors for
Development of better equipment has increased the automating the welding operation are the ingenuity of the
usage of automatic GTA pipe welding. Completely prog- personnel assigned to the task and the funding available.
rammed pipe welding systems are available to weld pipes One automatic welding operation utilizing equipment
in the fixed position up to 40 in. (1.02 m) in diameter, from several manufacturers is shown in Fig. 24. The
using wire feed additions, arc oscillation, and arc voltage welding head is mounted on a long side beam carriage that
control. Power supplies using current pulsation and vari- can be easily adjusted vertically for different vessel
able speed control have helped adaptation of the GTAW diameters. The welding head uses arc voltage control, hot
process to out-of-position welding. wire feed, and arc oscillation to weld formed heads to

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Typical Equipmentfor Process Applications I 29

pressure vessel shells. Hot wire feed permits high deposi-

tion rates comparable to other weId processes, and the
welds easily meet the stringent radiographic soundness
requirements. The drive rolls that support and rotate the
vessels are sequenced to start with arc initiation. The
welding operator is required to monitor the welding arc
and make the minor adjustments needed during the weld

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AUS C5.5 80 07@142La5 UB%12b2L 9

39 / RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

cycle. This type of complete system may reduce the weld- fluids within the tube bundle.
ing costs by as much as fifty percent. In one method of GTAW, the tubes are recessed below
Automatic NuclearWaste Container Welding the surface of the tube sheet approximately the same
Remote welding applicationsof automatic GTAW in the distance as the wall thickness of the tube, and the tungsten
nuclear field are growing. One exampleis the use of this electrode melts the corner or edge of the tube sheet as it
process to remotely weld the top cap on containers con- progresses around the tube periphery. Filler metal may be
taining remotely loaded radioisotopes. These high quality present either as a tube sheet overlay or wire addition. This
welds are simply madeby the GTAW process, USihg creates a weld throat or minimum leak path equal to the
equipment incorporating commercially available current tube wall thickness.
pulsation and weld programming features. As shown in Figure 26 shows the operator centering the machine
Fig. 25, the mechanical equipment for holding and center- over a tube hole by using a mechanical centering device.
ing during welding is designed to meet specific needs. In After centering, the device is removed andthe weld torch
this particular case, the GTAW torch was also shop built to is lowered to the workpiece, and the automatic cycle,
allow tungsten replacement by remote manipulator con- which includes programmed current and speed controls, is
trol. started. On tube heat exchangers of this size, the welding
This demonstrated flexibility of the process and the sequence is important to minimize tube sheet distortion.
associated equipment is partially responsible for the con- Certain codes may require a weld throat dimension
tinuous growth and acceptanceof its use in nearlyall types greater than the wall thickness. In this situation, filler wire
of industry. additions are made automatically after the first fusion pass
Automatic Tube-To-TubeSheet Welding is completed.
Various designs of heat exchangersare in use through- Automated Pipe Welding
out industry. Fastening the tubes to the tube sheets in these Automatedpipeweldingequipmentrepresents a
heat exchangers is often accomplished by mechanical ex- technological improvementover conventionalmanual pro-
pansion of the tube into a small ring groove machined into cesses. It was developed to satisfy the increasing demands
the tube sheet holes, Automatic GTAW of the tube ends to for an improved method of achieving consistently high
the tube sheet minimizes leakage when using corrosive integrity welds forthe critical requirements of nuclear pipe

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Typical EquipmerztforProcess Applications I 31

fabrication (see Fig. 27). and ease of operation of the welding heads. Direct reading
All-position installation and removal of the welding control settings, easily accessible manual adjustments,
head is accomplished in a few minutes, and the head is and a clear view ofthe weld zone contribute to minimum
precisely guided around the pipe by rigid or segmented operator training time and allow compensationfor uneven
belt guide rings. The independently adjustable dwell time wall thickness,offset joints, uneven lands, and weld bead
at oscillator excursion limits and the precise relationshipof sag.
the filler wire to the electrode tip permit extremely rapid
filler wire addition. Control of penetration (root) passes Typical Equipment and Applications
for all types of joint designs and high quality fill passes is Figures 28 through 35 were chosen to illustrate automa-
ensured by the accuracy and repeatability of the controls tic GTAW equipment and types of product application.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 32 I RECOMMENDED FORGASTUNGSTEN ARCWELDING
PRACTICES

Typical welding system integrating powersupply, automatic weld programmer, torch holder,
and parts for semiautomatic welding,mediumvolume production.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Typical Equipmentfor Process ApplicationsI 33

2 x 2 in. (51 x 51 mm) semiconductor case edgewelded for a hermetic seal.

The torch automatically tracks the shape.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 34 I RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

One programmer with power source systemenables one operator to alternately load 0.6 in. (7 mm) O.D. capsule into fixture every
6 seconds. Welding labor perpart, only 3 seconds; welding amperage switched electronicallyfrom onetorch to the other. Fixture has water-
cooled clamps, common drive for alignment, automatic torch refraction, and complete weld program.
Fige30 -WOshtic~ma ~ t ~ m dGTA
ic dm stdion

Simple fixture provides method of locating parts and torch precisely for each weld.
Fig. 31 -Ckcuderen~alwelding system

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 Typical Equipmentfor Process ApplicationsI 35

Typicalwelding system integrating power supplx automatic weld programmer, torch holder, and parts for semiautomatic welding,
medium volume production.
Fig. 3 2 - S e d a n t ~ m a ~GIA
c welding m a c h e

Small thin partswelded on welding fixtures as well as automatic machinewelding makes GTAW an economical production process.
This is anassortmenf of small thinparts, production welded. Note 25-cent piece in the center for relativesize.
Fig. 33-Qpimlprduc~onpieces

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 36 / RECOMMENDED
PRACTICES
FORGASTUNGSTEN ARCWELDING

of welded assembly in
Two 4 x 8 in. (102 x 20 mm) crystal can relays edge weldedCRS header lip to cupro nickel case. Note size
comparison to dime at toD of Dicture.

Coil ends being butt welded together with no increase in strip thickness.
Fig. 35-GTA welding m a c h e

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C505 8 0 078L1265 0002628 I

Safe Practices I 37

6. Safe Practices ing arc acts on the oxygen in the surrounding atmosphere
to produce ozone. The amount of ozone produced will
6.1 Introduction. The general subject of safety and safe depend upon the intensity and the wave length of the
practices in welding, cutting, and allied processes is cov- ultraviolet energy, the humidity, the amount of screening
ered in ANSI 249.1: “Safety in Welding and Cutting” affordedby the welding fume, and other factors. The
(published by the American Welding Society), and ANSI ozone concentration will generally be increased with an
249.2: “Fire Prevention in Use of Welding and Cutting increase in weIding current, by the use of argon as the
Processes.” Welding personnel should be familiar with the shielding gas, and when aluminum is welded. Test results,
safe practices discussed in these documents. based upon present sampling methods, indicate the aver-
In addition, there are other potendal hazard areas in arc age concentration of ozone generated in the GTAW pro-
welding and cutting (besides fumes, gases, and radiant cess does not constitute a hazard under conditions of good
energy), such as noise and the handling of cylinders and ventilation and welding practice. (See ANSI 249.1 for
regulators, which warrant consideration. Those areas that welding conditions requiring Ventilation and particularly
may be associated with the GTAW process are briefly when weldingis accomplished in confined spaces.)
discussed in this section. 6.4.2 Nitrogen Dioxide. Some test results show that
high concentrations of nitrogen dioxide are found only
6.2 SafeHandling of ShieldingGasCylinders and within six inches of the arc. Natural ventilation reduces
Regulators, Compressed gas cylinders should be handled these concentrations quickly to safe levels in the welder’s
carefully. Knocks, falls, or rough handling may damage breathing zone, so long as the welder keepshis head out of
cylinders, valves, or safety devices and cause leakage or the plume of fumes (and thus out of the plume of weld
accident. Valve protecting caps, when supplied, should be gases). Nitrogen dioxide is not thought to be a hazard in
kept in place (handtight) except whencylinders are in use GTAW.
or connected for use. For further information, see CGA
Pamphlet P-1, “Safe Handling of Compressed Gases in 6.5 Metal Fumes. The welding fumes generated by the
Containers.’”j GTAW process can be controlled by natural ventilation,
general ventilation, local exhaust ventilation, or by re-
6.3 CylinderUse. The following should be observed spiratory protective equipment as describedin ANSI 249.1.
when setting up and using cylinders of shielding gas. The method of ventilation required to keep the level
(1) Before connecting a regulator to the cylinder valve, of toxic substances in the welder’s breathing zone within
the valve should be opened moment&ily and closed im- acceptable concentrations is directly dependent upon a
mediately (“crackingyy), to clear the valve of dust or dirt number of factors, among which are the material being
that otherwise might enter the regulator. The person open- welded, the size of the work area, and the degree of
ing the valve should stand to one side of the cylinder confinement or obstruction to normal movement where
outlet, never in front of it. Then, the cylinder should be the welding is being done. Each operation should be
secured. evaluated on an individualbasis inorder to determine what
(2) After the regulator is attached, the adjusting screw will be required. Acceptable levels of toxic substances
should be released (by turning it counterclockwise).Then associated with weldingand designated as time-weighted
the cylinder valve should be opened slowly to prevent a too average threshold limit values (TLVs) and ceiling values,
rapid surge of high pressure gas into the regulator. have been established by the American Conference of
(3) When the weId area is left untended, the source of GovernmentalHygienists and bythe Occupational Safety
the gas supply should be shut off. and Health Administration. Compliance with these ac-
6.4 Gases. The major toxic gases associated with welding ceptable levels can be checked bya sampling of the atmos-
areozone, nitrogen dioxide,andcarbon monoxide. phere under the welder’s helmet or inthe immediate vicin-
Phosgene gas could be present as a result of thermal or ity of the helper’s breathing zone. The maximum allowa-
ultraviolet decomposition of chlorinated hydrocarbon ble particulate matter (welding fume) which may be pre-
cleaningagents,suchastrichlorethyleneandper- sent (as measured over an eight hour period for a forty hour
chlorethylene, located in the vicinity of welding opera- weekly exposure) within the welder’s breathing zone is
tions. Degreasing or other cleaning operations involv- listed in Table S. Sampling should be in accordance with
ing chlorinated hydrocarbons should so belocated that AWS F1.1, Method for Sampling Airborne Particulates
vapors from these operations cannot be reached by Generated by Weldingand Allied Processes?
radiation from the welding arc. 6.6 RadiantEnergy. Any personnel within the im-
6.4.1 Ozone. The ultravioletlight emitted bythe weld- mediate vicinity of the welding arc-should have adequate
protection from the radiation produced by the welding arc.
4. ANSI 249.1 is available from the American Welding Society, 2501 Generally,the highest ultraviolet radiant energy intensities
N.W. 7th St., Miami, FL 33125 are produced when an argon shielding gas is used and
5. ANSI 249.2 is available fromthe American National Standards Insti- when aluminum is welded.
tute, 1430 Broadway, New York, NY10018
6. CGA P-1 is available from theCompressedGas Association, Inc.,
500
Fifth Avenue, New York,NY 10036 7. See also AWS filmes and Gases in the Welding Environment.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services
 AWS C S - 5 8 0 W 0784265 0 0 0 2 6 2 9 3 W !

38 I RECOMMENDED FORGASTUNGSTEN ARCWELDING

PRACTICES

Table 8
Particulate matter with possibly significant
fume concentrations in the welder breathing zone
matter
Particulate
weldedbeing
Material
Aluminum and aluminum allovs AI. Mg, Mn. Cr
alloys
Magnesium Mg, Al, Zn
Copper and Cu,
copper
alloys Be, Zn, Pb
allovs
nickel
and
Nickel Cr,Cu,Ni, Fe
Ti allovs
titaniumand Titanium ~~ ~~ ~~ ~ ~~ ~~ ~

Austenitic stainless steels Cr, Ni, Fe

Fe,Carbon steels* Cu, Mn
*Also Cd and Zn for plated materials.

The filter glass shades recommended for GTAW, as protected against exposure to noise generated in welding
presented in ANSI Z 49.1 as a guide, are: and cutting processes, in accordance with paragraph
Shades 1910.95, entitled “Occupational Noise Exposure” of the
When welding ferrous material 12 Occupational Safety and Health Standards, Occupational
When welding non-ferrous material 12 Safety and Health Administration, U.S. Department of
Flash goggles 2 Labor?
Leather or’woolclothing that is dark in color (to reduce
reflection which could cause ultraviolet burns to the face
6.8 Safe Handlingof Welding Equipment
and neck underneath the helmet) is recommended for 6.8.1 No welding should be done until all electrical
GTAW. The greater intensity of ultraviolet radiation will connections,power supply, weldingleads,welding
cause rapid disintegration of cotton clothing. machines, and work clamps are secure, and the welding
power source frame is well grounded.
6.7 Noise and Hearing Protection. Personnel shall be 6.8.2 The work clamp must be securely attached to the
work before any welding is carried on.
6.8.3 Welding cables and hoses should be kept free of
8. The choice ofa filter shademay be made on the basis of visual acuity
grease and oil. The line supply disconnect switch should
and may therefore vary widely from one individualto another, particu- be placed in the “OFF” position whenthe welding power
larly under different current densities, materials, and welding processes. source is left unattended.
However, the degree of protection from radiant energy afforded by the 6.8.4 The positions of all power source controls for
filter plate or lens, when chosenallowtovisual acuity, will still remain
in initial start or resetting the disconnect switch should be
excess of the needs of eye filter protection. Filter plate shades asaslow
shade &have proven suitably radiation-absorbefit for protection from the checked.
arc welding processes. 6.8.5 The insulation on the electrode lead cable, gun,
9. See also AWSArc Welding and Cutting Noise. and electrode holder should be in good condition.

COPYRIGHT American Welding Society, Inc.

Licensed by Information Handling Services