GLOVEBOX PRESSURE RELIEF AMD CHECK VALVE*

Kenneth L. Blaedel

Engineering Sciences Division

Mechanical Engineering Department
University of California
Lawrence Livermore National Laboratory
Livermore, California 91550

ABSTRACT

This device is a combined pressure relief valve and check valve providing
overpressure protection and preventing back flow into an inert atmosphere
enclosure. The pressure relief is embodied by a submerged vent line in a
mercury reservoir, the relief pressure being a function of the submerged
depth. The pressure relief can be vented into an exhaust system and the
relieving pressure is only slightly influenced by the varying pressure in the
exhaust system. The check valve is embodied by a ball which floats on the
mercury column and contacts a seat whenever vacuum exists within the glovebox
enclosure. Alternatively, the check valve is embodied by a vertical column of
mercury, the maximum back pressure being a function of the height of the
column of mercury.

UCID—20695

DEB6 014940

*Work performed under the auspices of the U.S. Department of Energy by

the Lawrence Livermore National Laboratory under Contract W-7l05-Eng-18.

OJSTRIBUTION OF THIS DOCUMENT 1S UNUNITED

^
 INTRODUCTION

LLNL employs a number of inert atmosphere enclosures ("gloveboxes") with

attached glove ports and which house toxic and radioactive materials. The
gloves represent a hazard in that they could burst if the glove box were
accidentally over presssurized thereby releasing toxic material to the
environment. As a precautionary measure theses enclosures are fitted with
pressure relief devices which, because of the toxic material contained in the
box, must vent into an exhaust system and not into the laboratory. The
pressure relief setting should be set to limit the difference between the
glovebox pressure and barometric pressure in the laboratory, since it is this
pressure difference which causes the gloves to rupture.

A problem has arisen with the current apparatus because the pressure
varies in the exhaust system by more than the specification on the setting of
the pressure relief device. The need is for a passive pressure relief device
which is sensitive to room barometric pressure, but not downstream pressure in
the exhaust line.

A second problem has arisen with the current apparatus when the glovebox
is under vacuum and has caused the mercury to draw about 30 Inches up the tube
from the glovebox to the bubbler. The mercury in the tube has come from the
reservoir and hence the immersion of the tube into the reservoir has decreased
considerably. Under this condition, if the reservoir is seismically
disturbed, then the tube of mercury can gulp a bubble of air which can carry
some mercury back up into the glovebox. The need is for a passive check valve
which does not require a great volume of mercury to actuate.

BACKGROUND INFORMATION

Engineering Safety Note ENS 82-910 (attachment #H) describes a prototype

relief and check valve device, but which exhibits the two problems described
above. This safety note, however, establishes the basic design goals
including, the minimum tube sizes for required flow, the pressure at which the
gloves burst, and the pressure at which the relief device should actuate.

A memorandum from Kelly to Landingham dated 2 July 1981 (attachment #3)

discusses the pressure drop across various in line mercury vapor filters.

DESCRIPTION OF THE APPARATUS

The device described below meets the requirements of Safety Note ENS 82-
910 and addresses the two problems described above. The valve assembly
comprises the following items as labeled on the attached drawings,
AAA8H-100334 and AAA84-111177 (attachment (71):

Item tk, the reservoir, structurally connects all the component parts and
contains the meeury.

Item #B, the mercury, prevents the gas from the enclosure from entering
the laboratory and provides the pressure relief for the enclosure.

-2-
 Item ffC, the annulus to the exhaust, is the passageway from the valve
assembly to the exhaust system which conducts escaping gas from the
enclosure to the exhaust,

Iten ID, the check valve, conducts escaping gas when the pressure in the
enclosure is greater than the setting of the relief and blocKS flov; when
the pressure in the enclosure is less than the pressure in the exhaust
system.

Item #E, the duct and filter to the laboratory, allow the barometric
pressure in the laboratory to act on the surface of the mercury in the
reservoir.

FUHCTIOK OF APPARATUS

This apparatus uses the barometric pressure in the laboratory as a

reference for the relief pressure and remains unaffected by the pressure
fluctuations of the exhaust system.

The valve assembly must function during three states:

1. The glovebox under vacuum.

Gloveboxes are evacuated as normal procedure when establishing an

inert atmosphere inside. The valve assembly must therefore act as a
check valve, that is, not allow flow into the enclosure under this
circumstance. This is accomplished by the check valve. The
embodiment here uses a hollow steel ball which floats on the column
of mercury. When a vacuum is drawn, the mercury rises up the column
until the floating ball contacts the seat thereby closing off any
further rise of mercury up the column.

2. The glovebox at normal operating pressure.

While the glovebox is at a pressure slightly below the barometric

pressure in the laboratory, the height of the mercury in the tube to
the glovebox is above the level In the reservoir in communication
with the laboratory. The height of the mercury in the annulus is
also above that in the reservoir. Thus, the glovebox is sealed off
w
from bc h the exhaust system and the laboratory.

3. The glovebox venting at the relief pressure.

Should the pressure rise in the glovebox, the level of the mercury in
the tube to the glovebox will decrease until it reaches the bottom of
the column tube at whicri time the gas from the glovebox will escape
from the column and will rise up through the annulus leading to the
exhaust system. Because the gas is much less dense than the mercury,
it will not tend to migrate downward and escape to the part of the
reservoir in communication with the laboratory. The setting of the
relief pressure is determined by the depth which the column tube is
immersed below the level of the reservoir in communication with the
laboratory. Because this depth is not significantly influenced by
the pressure of the exhaust system, the pressure at which the
glovebox will vent is not significantly influenced by the pressure of
the exhaust system.

-3-
 TESTltiC THE APPfcBWOS

The attached report by Jess Squires (attachment #2) describes the testing
Of the relief device. With a H.37 inch diameter reservoir, a change of 1 inch
Of water in the exhaust system suction caused a change of only 0.1 inch of
water in the relief setting. There are a number of ways to minimize this
further if the !0 fold decrease in sensitivity proves inadequate.

The attached memorandum by Ed Schmitt (attachment 05) describes the

testing of the check valve.

Attachments

1, Drawings: AAA84-11033^ and AAA81-111177-

2, Report: "Testing of Mercury Bubbler, Mk II".

3, Memorandum: "Glove Box Upgrade Mercury Bubbler Filter", 7/2/81).

H. Safety Note: "Vacuum Glovebox Installation, Building 24!", ENS-82-910.

5. Memorandum: "Check Valve Development for riercury Bubbler, Mk II", 1986

-H-
ATTACHMENT 1. DIMENSIONING ANO TOLERAHCIMr, K t
ANSI V14.5M.ln2.
2. SURFACE TfXTUIE M l A N t l MA.I-1V7I.

*. RU_ RliEKVCHZ. WITH ME-g£.Ug.V TO J. LEVEL C£HeE£fOd»o£

£ > fcFTEE. TTJC^E^Z. ~ \ U _ 1 i O B T ^ M F . C ! i £ T T\T» CJ=

_E.VE!_ i K i a t K T C e . S C R E W T £ S A " 7 - VTTME S L S B F K E
OF T H S . W H . ' a i j S . f . EITOW F I L L » O l E i , S TO i T E M - 7 '
1
TO vesawkaMim ?v*. vrb TOiniQM.

« • 7EPTHFDe.eruE.P PEEi^LIEE.1
iUHq aim Ma TO
.CO 0 'zm
.ID IJiA 040
.zo S.CS .c*?a
.!£> l.(ol -H7
,3S OS^ -V7Z.
.40 10.16, .I9&
-50 12.7 .Hfc
.40 L5.Z4 . 2 ^ !

^® .70 I7.7& -3-44 1

MUT.HE.Y,. U 4 - 7 Q U M C
SCE,. S C H C A g i
P*L'JGi, P I P E . I/Z-14VAMPT.
475Q-ifa1'*l, Wk&H&a. . UlqlCKl . 1./4
rUJG.TAIUTiF-'F ' j m O M 3/4 £££.^304
NUT. UklwDM. 3 / 4
LE*E.1- N P t W C f c ^ a i W 7AS-Q2
53lO-Z2Zl>^ O - R ^ M t ^ , i - a / S H 2 * 3/16. W S U M M 4 mtixt-afc b
ez-ic^o^o VIEWPORT a X S ^ E - T
•iCWTOET W K J C C W
iEWTOETC*P
CHE.CVC VAX-Vg.
WEJLE'MgMT

DiSCI»tiON/aAIHiM

own P F souzxTa&4 CIMMICAIION CW£V,\. BA-VCa 1 A \

B'UB'Srl-E.'W.. CHE£.*~ VKLVE.

IimWVCTIOH Hi
U W I I H C I MVttHOM

•c^SM-12
AAAert-\V333_rOC
| (UCH1NKM |<tg||HlllNO 01
 ATTACHMENT 1 . OlMINWOWtNC ANO tOMANClHG r e t

. smrACt rexnjtc m ANN tui-\n%.

. flu. Olh" A l t »M i-otH^S

A U . MftTiut r»Arars. re ae
o-.o'o c».c*ci»yte Fir
fc ALl JOIX, JrtwC . . F"-.e?
0'4 FuSipu WCLR A i RCOb
T ALL WCLQ6 TO 3C VAC T i G ^ r L M K
T
Rl\rC HOT 3 S t t C C o l*|0"*mf

1 T ^ C , . ^ = O . O . S » » » C«K .0
&i-»57&i
1

z. mso-mj'i r-«»OP.Mu-.o-.< J ; . « K '

LADHH CR. CA 0 C « . Li «•> W K . D . C . J o * ^ T
\
?.
e t

SUCtr . , 1 , ™ <
5 H e : r O B I TUK
j « CM*
3c^ tr.Ej
c <•

i 5.
1 <-.« V, N . » SO. « JO* C « « * I ,
^ = .~r. „ » , « „"w = , = » <e S 3
? TVO.-G ..*• » » . . « . » » « ^ t
Ti ™ « , , . . « » , c , « n
'
• M M S

IIOO ' • H A « in «l IMC X)

'"•
OWN 1. MuruttY \i/Bm
cue-.' OLOO i-ii
CM * * f (.7VWX£>0&\ * 4 * 1
- - { p •», 5 0 ,- - • Ml M X U H , , ,„
trvojiS Zt~A I"** (K33.£? 'i 'i,
( ««(.-£
...iu... 1 wt.2-«^'."
1—
1 ., ~;^;r .,
;
*™ ;AAAc-i-l" 77-QC
" 7
3
t M
•l'-l»"!'°
__ """•'• °"'"~~
 ATTACHMENT # 2

Jess Squires
August, 1985

Testing of the Mercury Bubbler, Hk II

The purpose of this experiment Is to establish the influence of the

vacuum exhaust system pressure on the relief pressure of the mercury bubbler.

In actual use, the vacuum will be supplied by a hood, or "green line" at

a negative gauge pressure of about 5 in. H2O. For the test, however, a line
with more vacuum was used and regulated with a Hoke valve. The glovebox
overpressure was simulated with a throttled compressed air line. The testing
apparatus is shown schematically In Figure 2.1.

The testing procedure consisted of presetting the vacuum pressure, then

cracking the compressed air valve until the relief pressure was reached. The
relief pressure is immediately followed by a slight pressure drop, indicating
an escaping bubble.

The data are tabulated in Table 2,1 and shown graphically in Figure
2.2. Ideally, the curve should be a line with zero slope. The observed slope
in the data is due to the change in the level of the Mercury exposed to
barometric pressure. Specifically, as Mercury is drawn up the annular line to
the exhaust system with changing suction of the exhaust, the height of Mercury
in the reservoir subject to barometric pressure slightly decreases.
Therefore, the immersion height of the relief tube from the glovebox also
decreased thereby decreasing the relief pressure.

The slope in Figure 2.2 is 0.1, hence for the mercury bubbler shown in
drawing AAS.SU-110333, a change in exhaust suction of 1 inch of water causes a
change in relief pressure of 0.1 inch of water. There are several ways to
further minimize this effect. For example, one simple method is to increase
the ratio of the reservoir area to the area of the annular tube to the exhaust
system. For example, a reservoir of 13 inches diameter would decrease the
sensitivity to .01, that is a change in exhaust system pressure of 1 inch of
water would change the relief pressure .01 inch of water.
 • /•A/,--: ^ jo

•'"-V- \ ^rt-\):

UtiEf ^tif\ £lo\ir?--•;*. pfr^-isj/?-:

r
1 4?
VACUUAV fit-

..A> fls^r —
i

lA^n /

K -V.J7

f/O^ £-1
V4 McUOiifytimr MEkuRE 4> FlArt W' fi
MH
1
0 sicarjb M M fi'M

'^/^^f^^ifi^^^^^^^^
 h^MpMl.J/J.J

r ...^-
1 .:,-r..Ji
.•ujjjy ./„ -*UU.
{.L.'ixMO"-'}
- l / ? - ^ -/-^ '•£-£''

r„,,v
;.Jj„>
s,.u*
'

o" )
'.•

J
•'•-'

-<..> Hy r-i 7 y.i> ~> 'J ir-

i •»
? • /?>• / / / -•i •iih

?o / ^ /*-( >c -h 71*

.~Y 20~> 2^-' i-^ "n-y 'z-.'-'T

-> n ; ;-r
? 2. ^ ^>/
?^ 2^: 2W 5.0 1:7 2/7
2J2 7.0 2 z-% 1,7 2/?
i-j* - j : l-(. 2-1^ ^
~-<r ^"v lob z-r 2,<? 22£J
L 10% 231' -? '• 3/4 ZZO
C-
- • o • lOi, i?^ £.^ 120 7.1'

6-; 2o/ 20i /•<? ^^ : 221

9-6 2<2> 2J- '•6 122 22 5
5-V 20°/ 7o? AV 111- 'ilV
5-2 Ut 161 /•* 2J.o l-i&

6*o 10°i 20=7 /•o lii -1-5

fc? 210 2/J a is 2.-3 22 (.
fc£ 2/0 IiO o£. i> 2ii

yA HI - i " o-W 'ilv 21?

2/2 2/J o-Z. Zln 22?
f-2
o 11^ *25*

J^^2.i-^t iOit
 ATTACHMENT # 3
interdepartmental letterhead
Mai'S'a'OiL 369

July 2, 1984
2218L

TO: K. L. Landingham

FROM: B. B. Kelly

SUBJECT: Glove Box Upgrade Mercury Bubbler Filter

Introduction

Over the past year much effort has been dfvoted to upgrading the safety
standards of glove boxes in buildings 222, 231, and 241. Recrtly, upon
completion of an upgraded system in Building 241, it became apparent that
further research was needed in th° area of filters for the mercury
bubblers. The following diagrams and test results were obtained to
complete the box upgrade

Test Procedure

The purpose of t h e test vas to determine what the pressure drop (b P)

across the filter would be at various flow rates. Besides the obvious
filtering requirements this would also help determine the maximum s a f e
venting rate for the box. Four tests were conducted, one without a filter
in the line and t h r e e more with the following filtersi

1. USA $8760 Hersorb Chemical Cartridge

2. In-line filter assy. MSA #78001

LLNL #4180-33332

3. Gas filter, David Trip Assoc, Inc.

LLNL - 8

The test of the filters vas accomplished by flowing air and measuring t h e
pressure at a point before and after the filter. (To test filter HI a SST
housing was fabricated on-site, see photo.)

The flov rate was measured chrough a flow meter and the pressure
difference vas noted for the various flov rates, see graph. For filter f/3
the pressure drop was so great that a less sensitive pressure gauge had to
be used.

Universityo','California
|[[E Lawrence Livermore
\~d National Laboratory
R. L. Landingham ~2- July 2, 1984

Summary

The MSA 'mersorb* filter will be adequate for removing mercury vapor from
the box atmosphere and ksep particles from the box contaminating the
mercury bubbler. The maximum flov rate for the argon backfill should be
3ess than 230 1/min, As a word of caution, we found that i t was necessary
to treat the glove box mercury filter and bubbler exhaust vent to dri
train as a whole vnit for setting the level in the mercury bubbler.

ASBA has not set a standard for mercury filters, but the USA filter
does meet'the industry standards for mercury vapor removal. They are made
vith an indicator to tell when it's time to replace. This filter is
designed to stop mercury vapors only—it will not stop mercury liquid.

An in-line filter housing was designed (see drawing) to fit onto the
mercury bubbler. As an added attraction other HSA filters will fit into
this housing. We found it useful with the .3 micron dust filter in
protecting the leak detector from exposure to beryllium dust in one of the
boxes.

B. Earl Kelly ~f

BBKijg

Distribution
K. i . Bj'^edcJ, L-34X-
J . C. Coker, 1-144
R. tf. Goluba, L-144
F. E. Green, L-362
E. J . Bammerel, L-369
C. t . Boenig, L-369
R. L. Landingham, L-369
P. W. Linnes, L-369
S. C. Jxird. 1,-327
a. V. Mahler, L-144
T. C. Seitz, 1,-358
E. B. Schmitt, L-369
E. W. Southvick, L~369
?. W. Wilson, L-358
 o D
HSA #8760 A
SST Cas Filter Kersorb Jn-Jiue
o
Test Stand David Trip, Inc. Filter paper Filter
Liters/iain. Resistance Inches of Water 4130-33332
20 .50 .80
30 1.10
40 1.00 1.50
50 1.40
60 2.20
70 6.94 (.25 psi) 2.00 2.60
80 .50 2.50 3.00
90 15.28 (.55 psi) 2.95 3.50
100 3.35 3.95
110 3.95
120 27.80 (1.00 psi)
130 2.00
140
150
160
170 1.50
180
I 11
0

PLUMBING DIAGRAM

X Valve

u
r—a
Regulator

Flow meter
Brooks #R-BH-25-4

®
A? Gauge, 0-8 inches
or 0-3 psi gauge

Filter assy.
 "T"

180

170 -0
•130--
<•>
150 -

HO • 90 •-

130 ©

50 1/minj ^
120 4 1- +•
.25 .50 ,?5 1.00
110
P,S,I._

100 £} —

90 -0-
80 j j ) „ ^0- -A

70 -EJ- A-

60 ^&-

50 -B-
40 -0- --&- ^-4es*-~stafK3 resistance^
1
1
30 ~A—k- ^©-^S^rgai-ft ^TI1P» j l ^ C t ._;

20 S ^ S A ^ B r e Q - W W * - ' •-••! -= ; - •{--'--; -,'

—1- -J__i ._J„
)
]0 U-A- S> I n - l i n e papet-Hlter /4180?3332-fj

1 ••?•
0
1.0 1.5 2.0 2.5 3.0 3.5 4.0
Inches of H,0
 Wares-

l
F ^ > tveLD To Be \XAcuuff rr&tsr,
" FMier 7-<? FUSS mcvu*? r&ir
LeAm /?*)W MeT ~™> excsvo
i x to-* /ir»?o: tre^/ssz

r/iB MO.

£^¥

/O TVBE l£ CO X .Cf9ttfXS.0litP3£. J.%%

j
rvee / o-P. x .O49tfxs.oj,fe (. tiTM-Azm 47/O- I4ZCH
T5S-,
rues g .o. x .o&jwxs.ci,-
a
<t7/0-l-fZM
i : 3
TUBS ^•O.D.X.O'>9t^XS.Si. ' ' L Jg/ , 3

C
7Z/B£ iJr aa. X.ea-9tVJ!Z.C6- SZ iSh <-7/o-t<*za£

B3-/'5S79 /SOpy XX&-06

S3-//5579 BO0f 774 B- OS

S3-//S579 eooy 7?iB-a$
S3-//557? BOPf 7XB-03
B3-/ISS79 BODY rxlB'OZ
MATERIA! / DESCRIPTION

m.fi%zrs#sGi)&'"VE ClASSlFICAflON MAJOR UNIT £M&A$f&7&y & S & '

No. P.EQ'0
at. /Ktft&vatM
APPROVED &tfM£&r
™0£ElP.flU JiJSg SUBASST. / * 7 2 5 ^ » C - £ ^ ' / - 77&*'°

OF THE UNIVERSITY OF CAt'FOtNIA

PER ASSV.
SHOWN m£3-MF576 OETAH &&£>y W£'U3*?£M7~
tAWRENCE LIVERMOAE LAtC'ATORY,
lEPKODUCTION PROHIBITED WITHOUT
kP&8?-//557f-00 3)
10 Uwenc*\,lveimoiB Nnilortol laboratory
MECHANICAL ENGINEEH1N0 fiEPARTMEKT
UNIVMJITY OF CALIFORNIA
PERMISSION Or THE MtCHANlCAl
ENGIMIERIHG DEPARTMENT. ^z>/?-i2. | imn-Z-CT.^-.
 A/ores-
J. Ii§/ytACMlAJ£0 SUFF f &TCCK.

f j > H/ELO TO BS ISACOllJ*7 T'6//~

F>E*? *7&£.-7t. - &>'333 (l-£.M£. SFS<z)
L£AK */FT£ /Jor TO £XC££0

TAB tVO.
B>y¥v

/a 7V0£ tg O.&. X.e#<?*//< £.06?.a3t. &ggi V7M4Z&

7-jJBS' / P.O. x.o+?n'JtS.c6t.t?j6. ^gS g >)7a-i<tzcx-
r z ^ g g ^ a g . •t-.g^y.^.yg.as^ggz. £ggs ASTMJZit
7VB£g o.o. x .c«9Hrxe,a!>?.ojt. c-5g-; ASTIj-AW)
7U8£ /f eg?. X.O<9tfrS.Q&?.a3d. cffigj; 4S!W4zi<! fJlQ. /4-SOe
S3-/I5S7B COt/EJZ. 7X0-Q6
63-//SS73 tCOl/SfZ 7VI0-OS'
<COIS£*g. 7X&-0&
63-//SS7S <ZOf&?? 7»e-as
S3-//SSTS czofmfz. 7-AP-oz.
MUTHIM. / DESCRIPTION REQ'O S P E C NO.

wwoctTsrsooeF <Sa

NO. REO/D
aH,if<t;6aiOWH/
APPROVED ^ c T y & y ^
122. OFfrarusE
HlIT
rHft e o t U M t N r 15 THE W O M t T Y
WUORWTT
M&jee&^y
tfW?A*S7^y
7-&*)&
&2+/

iOCUMtNr 15 THE W O M t T Y
SHOWN QHS3-//55?4- DETAIL €&&&& &/£•£.&/*?&*/7~
OF THE (JNIVUStTY O f CAltrCHNIA
DRAWIWO TO.
UWKENCI MVE1M0RE 1/UOUTOtY,
3% U v n e n c B U v e r m o r i Nntlonit laboratory it»«O0UCTION MOHUITED WITHOUT
MECHANICAL ENOINEEHWa DEfAHTMEHr • ' H M I S 3 I Q N OF THE MtCHAHICi.1
W I W H l T V Of CALITMNIA ACCT. HO. -527/^- /Z. SHEET (..Of t
IMQlNEEIINO DEPARTMENT.
 4. IHTTAU. / nsr vx CJIS Ci-ies.

SZ- 'C9i"iB

&unt nor.

.i'SSMSls

•' l E G E N D

-IX]- VAJ.VE

-7 /<C LOi^~^
* & - . _2J «.MPO^"-'5 &UJE

0 % VACUUM S1GE

U BE6JLATO!;
ol

4*IVC £9.
ni.T«
 LAWRENCE LIVERMORE LABORATORY • UNIVERSITY OF CALIFORNIA FILE NO. FACE

ENGINEERING NOTE ENS R?-910 1

MECHANICAL ENGINEERING SAFETY MOTE Stan Lord
VACUUM GLOVEROX INSTALLATION - BLDG. 241 November ??, 1°8?

ATTACHMENT # 4
MECHANICAL ENGINEERING SAFFTY MOTE

VACUUM GLOVEROX INSTALLATION - RLDG. W

Prepared h v ; <?<pt&?&:
Stanley C£/Lor?
Responsible Engineer
Engineering Design Section
Engineering Sciences D i v i s i o n

Checked by: *£/~c<z^e \J(, C^~A^n^&^<z__.

Diane i . Chambers, Engineer
Engineering Design Section
Engineering Sciences D i v i s i o n

Reviewed by
.lanj^s C. Swingle, Leader 7/
Engineering Design Section
Engineering Sciences D i v i s i o n

Approved by: V .
Jens Hahler," Acting"Jivision Loader
Engineering Sciences Division

Distribution:

Rlake, A.
Chambers, D.
Hammer?!, E.
•lohnson, J .
Kotowski, H.
l./indinqham, R.
Lirtnes, P.
L o r d , S.
Southwick, E.
Swingle. J .
v-inSarit. . V . , E.
Viprne R./Mahler, J.
M. E. F i l e s

U 3 1 3 (ftEV.B/m
 LAURENCE UVERU0R6 LABOfcATORV • UNIVERSITY Of- CALIFORNIA F T t l HO.

ENGINEERING NOTE NAME

2
MFCHANTCAI. ENGINEERING ^AFFTV NOTE Start l.nrd
VACUUM GLOVEROX INSTALLATION - RLOfi. ?.V

DESCRIPTION

This i s a general enqineerinq note t o o u t l i n e an updated method f o r

/ n
i n s t a l l inq gloveboxes i n Rldg. ? • With qloveboxes such as
AAA67-1055°1 and others used i n the b u i l d i n g the atmosphere i s
controlled. Work i s moved i n t o and removed from the work chamber
through an a i r lock ( e n t r y p o r O . A vacuum t i g h t door i s i n s t a l l ? ; ' i n
t h e opening between t h e two chambers. Evacuation and b a c k f i l l i n g of
e i t h e r the a i r lock or work chamber are independently c o n t r o l ! e r f . The
f o l l o w i n g three systems are r e q u i r e d f o r a glovebox of t h i s t v p e .
I n s t a l l a t i o n of these systems i s o u t l i n e d on schematic AAA#?-11?79O-0A

1. Vacuum system
?.. Argon system
3. Venting system

With these three systems the experimenter can o b t a i n and maintain an

i n e r t atmosphere i n which to c a r r y out the work.

Authorized personnel i n B u i l d i n g ?41 w i l l use the gloveboxes.

HA7ARDS

Operation of t h i s type of a glovebox has the f o l l o w i n g p o t e n t i a l hazards:

1. Rupturing of gloves by excessive pressure.

?. Mercury contamination by t h e r e l i e f device.
3, lackstreaminq of pump o i l i n t o the chambers,
i. Room contamination hy the a i r lock atmosphere.
c
i. Window breakaae.
fi. Earthquakes.

U-398 (MV.8/71)
I.AfRENCE LIVHPHORE LABORATORY • UNIVERSITY OF CALIFORNIA rue

ENGINEERING NOTE iPIS *?• am

NAME
I -L
MECHANICAL ENGINEERING SAFFTY NOTE Stan Lord
VACUUM GlQHF.qnx TNSTALLftTTQM - HI.OG. W " * " Piovemher 7?. toa?

Glnve Rupture P r o t e c t i o n

I n a s e r i e s of gloves tests conducted by Hazards Control i n June 19B2 a

minimum burst pressure of 0.379 Dsi was observeH. This was f o r a 15 m i l
b u t y l NORTON #8B1532 g l o v e , the kind used i n the b u i l d i n g . See Hazards
Control memorandum dated June ? 1 , 1f>8?. i n t h e appendix. Rased on t h e
t e s t data i t was decided to set the mercury r e l i e f device at 0.196 psi
f o r glove pressure p r o t e c t i o n , approximately 50% or" the minimum burst
pressure. At t h i s pressure the end o f the r e l i e f l i n e is submerged
about 0.10 inches which provides an approximate glove pressure
p r o t e c t i o n S.F, of ?. This i s h a l f the S.F. r e q u i r e d f o r pressure
vessels but i t is considered adequate since an i n t e r n a l pressure
c o n d i t i o n w i l l cause the gloves t o protrude which i s e a s i l y observed by
the o p e r a t o r .

Mercury Contamination

The mercury pressure r e l i e f device i s a source o f mercury contamination

to the glovebox, a i r l o c k , and <-oom. When e i t h e r the glovebox or a i r
lock i s evacuated the mercury w i l l suck back toward the chambers. To
prevent t h e mercury from e n t e r i n g the chambers t h e connecting l i n e must
extend at least 33 inches ahove the surface o f the mercury. Such a
column i f f i l l e d w i t h mercury i s iQ% more than a standard atmosphere.
Also there must he s u f f i c i e n t mercury i n the r e s e r v o i r to f i l l the l i n e
and prevent sucking a i r . A mercury l i n e f i l t e r i s i n s t a l l e d at l e a s t 30
inches ahove the surface o f the mercury. Nothing else is i n s t a l l e d i n
the l i n e .

An nyer pressure c o n d i t i o n in e i t h e r the work chamber or a i r lock i s

r e l i e v e d through i t s mercury r e l i e f device. The qas discharge picks up
n ^ c i r v contamination as i t bubbles through the pool of merciirv and must
not lie discharged d i r e c t ! v i n t o the room. Instead i t is vented through
a l i n o and passes through a HfPA f i l t e r before being discharged outside
ahove the roof of the b u i l d i n g s .
 . t "WHENCE L1VERM0RE LABORATORY • UNIVERSITY OF CALIFORNIA

ENGINEERING NOTE FHS 37-910

MECHANICAL ENGINEERING SAFETY NOTE Stan Lord
VACUUM GLOVF.ROX INSTALLATION - BLDG. ?41 November ?g, 103?

1
Pump Q i Backstreaining

Onlv one pump fa mechanical pump such as a Sargent-Weichi i s used f o r

evacuation. Should i t f a i l t o f u n c t i o n o i l can backstream and
contaminate the chambers. This is prevented by the Leybold-Heraeus
Secuvac valve i n the l i n e . When there is a f a i l u r e the valve c l o s e s .

Room Contamination from the A i r Lock

The atmosphere i n the a i r lock is a source of lovs-level contamination .

This occurs when i t s o u t s i d e door i s opened and the pressure i n s i d e the
a i r lock i s above ambient. To prevent t h i s the pressure i n s i d e the a i r
l o c k i s reduced to less than Jne atmosphere by opening the vent v a l v e ,
valve 2 0 , f i r s t . Hazards Control recommends a minimum a i r v e l o c i t y of
FO f e e t per minute through the door opening. See t h e i r memorandum dated
December 20, 1982 i n the appendix. The outside a i r lock door and vent
. valve are i n t e r l o c k e d so t h a t the door can only be opened when the vent
valve i s open.

Window Breakage

The glovebox window i s a l u c i t e window f i " t h i c k ) made o f a s e l f

e x t i n g u i s h i n g m a t e r i a l (Elexson p l e x i g l a s s , SE3K Rubber gaskets are
r e q u i r e d on both s i d e s .

Earthquakes

The glovebox s h a l l be secured t o the f l o o r w i t h a p p r o p r i a t e f a s t e n e r s .

Both mercury r e l i e f devices are to be fastened t o the frame above t h e
floor.

Ll-398 (AH.8/J1)
 L4WRENCE UVERMORE LABORATORY . UNIVERSITY OF CALIFORNIA FILE HO. P*GE

ENGINEERING NOTE 5
MECHANICAL FMGINFERTNG SAFETY MOTE Stan Lord
VACUUM GLOVEROX INSTALLATION - BLDG. 24! November ??, 193?

RELIEF DEVICE PRESSURE SETTING

There are several models of mercurv r e l i e f devices used i n the

building. These models have r e s e r v o i r s of various diameters. Only t h e
Q
s i x - i n c h diameter model shown on drawing AAA8?-10 148 i s to be used on
1 1
the work chamber with gloves. A l l models i n c l u d i n g AAA8?- 09 4R may bt
used on the a i r l o c k . There must be s u f f i c i e n t mercury i n the r e s e r v o i r
to f i l l the connecting l i n e to a minimum height of 33 inches above the
surface of the mercury. This column of mercury serves as a check valve
and prevents t h e suck back of mercury when the chambers are evacuated.

The proper depth o f f i l l of mercury i n the r e s e r v o i r i s obtained by the

r e l i e f pressure s e t t i n g . These s e t t i n g s are shown i n Table 1 . Relief
pressures are t o be set so t h a t the pressure corresponds to t h a t of a
Wallace Tiernan absolute pressure gage, or e q u a l . For a 33 inch column
of mercury the r e l i e f pressure decreases as the r e s e r v o i r diameter
increases.

INSTALLATIONS GUIDELINES

1. Glovebox

Each glovebox being i n s t a l l e d s h a l l be a p p r o p r i a t e l y anchored t o the

f l o o r per t h e LLNL earthquake tiedown p o l i c y . For t h i s note the
glovebox i s to be the vacuum type and s h a l l have a l u c i t e window w i t h
a minimum thickness of one i n c h . I t s h a l l be ?. s e l f extinguishing
material (Elexson p l e x i g l a s , SE31. Hookup o f the qlovebox i s shown
5
on uwg. AAAfl -n?7nn.

U m (SEV. B/71)
 LAWRENCE LIVERM0R6 LABORATORY . UNIVERSITY OF CALIFORNIA FILE 4 0 . PAGE

S U J E
ENGINEERING NOTE ENS 8 2 - 910
N«ME
6
* " MECHANICAL ENGINEERING SAFETY NOTE
Stan Lore
VACUUM GLOVEBOX INSTALLATION - BLDG. 241 * " November 2 2 , 19R2
TABLE 1,
; ^
(p-£ty TT^ A - cfirAk- ( o l - ^ A (K- £) ^•07 50. J

7
ci
'* .4
I
J
V•

t f t 5 Q.O<*
t U^ Re^iwow- A .
1
A- 0.75
M = 3^> —v-
-i=0.06
Jr\~ 0."5&l ?** (0.775 w c ^ K ^ "^ """
V U
R E u i e P PRES3UR& SH-XTINIG^ JH
OlA- 4.25" tiiAT.4-.SO" OlfN = s.oo" 1 T>IP •. - (j. 0 0 "
<

Jr\ (p-»J M«&b - J l f r ^ Jkli* Ug) A ( ^ P AG^JUr-'M-H^

3\ D.3S& 0.72& o.=>\& 0.647 0.2^6 o,4^o 0.17c O.SS&
^ 1
O.Bt ) 0.752. o.^z& j o . t t a 0.264- j O.S?7 t 0.\fc\ ' o.t^

3*> 0.^1 1 0.775 0. Z.1Z 0.5A4- : o . \ S 7 O.i&V

34 c.v\z 0.711 0 . ^ 4 & ; 0.7 \ 0 O.ZSO 0.51\ 0.\^ o.sqa
B% 0.404 o.&zz 0.3&T 0.12>D o.z&<\ 0.&S7 o.\q& 0.404
S£ 0.4\S> !
0.&4<^ 01*6*1 0.7^»\ o.z<\7 0.&04 i 0.204 0.4\&
| c 8t
a? o.4a7 ' ^ o.=7f 0-772- O.BOB 0.62.1 D.£\0 0.42-7
l a 0.43^ O.&l^ 0.5&1 o.7qa 0.3,»2> O.G^S o.z.vs 0.4S&

sq 0.450 \0A\& 0.4O0 O.S\4 O.S2-1 'o.teS 0.2.Z\ 0.4S0

AO 0.A-6Z 0-^40 0A\O o.&as J 0. Z2>o '• 0.6710.22-7 0.4&\

 L4«RENCE UVERM0R6 LABORATORY • UNIVERSITY OF CADFORNiA f I L E HO. TAGC

HJFCT
ENGINEERING NOTE ENS ?,?- l°9
7
""- MECHANICAL ENGINEERING SAFETY NOTE Stan Lord
VACUUM GLOVEflOX INSTALLATION - BLDG. ">A\ ""November ?p, ins?

?. Vacuum Pumo

Only one vacuum pump is to be installed on each ojovebox. With the

pump both the air lock and work chamber are to be evacuated. A
Sargent-Welch, Model 139? or 140?, mechanical purnn, or equal, is
recomnended for this service. A Levbold-Heraeus Secuvac, model KF-?0
or KF-3?, valve shall he installed on the pump i n l e t port to prevent
backstreaming in the event of a power f a i l u r e . For pump stoppaqes
due to other causes, i . e . , belt f a i l u r e , a CPA Fail Safe Detector,
Q
model RfiO, is i n s t a l l e d . This device w i l l close the Secuvac valve.
'See Appendix for equipment brochures).

3. Mercury Relief Devices

Two mercury r e l i e f devices are installed on each glovebox. One is

installed on the work chamber for glove pressure protection and the
other on the a i r lock. For glove pressure protection, only r e l i e f
device AAA82-10°148 is to be used. This model and the other models,
such as AAA 65-115700, may be used on the air lock. Each device is
to be secured to the frame above the floor and the r e l i e f lines from
the work chamber and air lock shall extend at least 33 inches above
the surface of the mercury. A mercury f i l t e r sha71 he installed in
each r e l i e f line at least 30 inches above the surface of the
mercury. There a^e to be no other obstructions in the l i n e s . Vent
lines are to he connected to the building negative air exhaust
system. Exhaust from the work chamber and a i r lock shall pass
through a HEPA f i l t e r before entering the trunk line of the negative
air system.
 4WRENCE LIVERMORE LABORATORY -UNIVERSITY OF CALIFORNIA FILE HO. PAGE
•

("itTs"
'

ENGINEERING NOTE EN< P^-910 8

MECHAfJVCAL ENGINEERING SAFETY NOTE Stan Lord
VACUUM RLOVEROX INSTALLATION - BLDG. ?4J November ??, 1"92

a. Lines

The main lines f4710-741381 in the vacuum system shall he a minimum

1
diameter of .1Z5 inches. Lateral lines to the glove ports etc. may
T
be smaller. he vent lines shall also he 1.125 inch minimum except
f o r the lateral to the individual components. These lines
/4710-14137! may be 0.875 inches 0,0. A }/? inch (0.625 0.0.)
1
diameter line f47 0-14l36l is recommended f o r the argon line . This
size line VJI'11 prevent the need for excessive pressures in order to
obtain a reasonable f i l l time.

*. Valves

Use Worcester type ball valves (4560-41(5121 i n vacuum and larger

lines of the vent system. Hoke tvpe valves may be used for other
requirements.

fi. Gages

Use compound Bourdon tube gages (6685-200351 and T/c gages

Q
(P9fiO-5fi5lOl where shown on drawing AAA8?-H?7 0. Use a Rourdon tube
tvpe pressure gage i n the argon system.

7. Regulator

Install a pressure regulator in the argon supply line.

8. Air-lock Door Latch

Install an interlock latch on the outside door of the air lock. A

small diameter steel cable shall be used to connect the latch to the
handle of valve 20, drawing AAA82-112790. Adjust the cahlp so that
the latch w i l l only be open when valve 20 is ooen. This insures *
negative pressure in.the a i r lock when the outside door is opened.

Ll-M» ( K T . I / m
 L.1WRENCE UVERMORE LABORATORY - UNIVERSITY OF CALIFORNIA Hie no.

ENGINEERING NOTE ENS ftg-910

MECHANICAL ENGINEER TNG SAFETY NOTE Stan L o r d

VAOJUM GLOVEHOX INSTALLATION - BLDG. ?41 November ??, 1982

TESTING

A f t e r the i n s t a l l a t i o n i s completed a l l the f e a t u r e s of the glovebox are

to he checked by an authorized i n d i v i d u a l . This i n c l u d e s a vacuum check
of the glovebox and a 0.3 p s i g pressure t e s t .

LABELING

A f t e r t e s t i n g a l a b e l s u f f i c i e n t l y l a r g e t o be e a s i l y seen s h a l l be
attached t o the glovebox i n a conspicuous p l a c e . The f o l l o w i n g
i n f o r m a t i o n should be on the l a b e l :

Assemble AAA fi7-itW!

I n s t a l l a t i o n AAA82-112790
Safety Note ENS SP-WS
Date o f Test
Date of Retest
Tested hy:

U-338 (BEV.6/7I)
 LAWREMCE LIVERHORE LABOR*TOR* • UNIVERSITY OF CALIFORNIA F l i t NO.

ENGINEERING NOTE
MECHANICAL ENGINEERING SAFETY NOTE
10
«*MEStan Lord
VACUUM GLOVEROX INSTALLATION - RLDG. 211 D»TE November ??, 1°8?

A°PEN0IX

0 Landingham Memo 3 paqes

0 Patterson Memo 3 pages
0 Leybold-Heraeus Data 2 paqes
0 CPA Fail Safe Data i page
a Graph of l u r s t Test i paqe
s
0 Hazards Control Memo 4 pages
0 Dwg. AAA 8?-"2790 1 paye
0 Hazards Control Memo J/page
/
/

./

LL-MJ (K».e/71)
 /•:• >. . ::. • :>• ' -'J

369

£•1 2-8022

April 12, 1982

0962H

TOi Distribution

FROM: Richard L, Landingharo

SUBJECT: IMPPOVE RELIABILITY AND SAFETY OF VACUUM GLOVEBOXES

A vide variety of vacuum gloveboxes have been designed, built, and

installed in Building 241 and other locations at LLNL. The original
designs met the need of the programs funding that worlc. As the glove-
boxes are used for other programs, the plumbing and electrical setups
are modified to meet the needs of the newer programs. Host of the
gloveboxes were not originally setup to handle toxic or radioactive
materials. The glovebox line in Ttoom IBS? of Building 2£1 is one
exception.

1 believe a basic design must be adopted that all these gloveboxes

should be upgraded to and any modifications then documented in OSP's
and/or on the glovebox itself. I propose this basic design should
include all the features shown in the attached diagram. This diagram
includes the following features:

1. Negative exhaust system on the entry port which is connected

to a filter box.

2. A connection latch between this exhaust system and.entry port

door to prevent opening this door without negative exhaust
being drawn through the port to the filter box.

3. Only one vacuum pump is connected to the box system to

evacuate the glovebox, gloves, and/or entry port. If the
pump fails, the glovebox, gloves, and port bleed back to
atmospheric" pressure equally with less chance of inploding
the gloves or entry port. Pump exhaust is vented through the
filter. A normally closed vacuum valve is located between
the pump and the glovebox system which will close if
electrical power to the pump is lost.

hm A mercury (Hg) bubbler trap is installed on the glovebox.

Only a line filter is installed between the glovebox and Hg
bubbler to remove Hg vapors; no values or mechanical relief
valves are allowed. These bubblers mast be secured to the
glovebox, filled with Hg to bubble between 0.2 and 0-3 psi,
and exhaust through the filter. No Hg bubbler is required on
 Distribution -2- April 12, l'Jb2
D962R
the entry port so long BG the b a c k - f i l l gas conies from the
glovcbox. If B gas l i n e i s attached t o the e n t r y port
d i r e c t l y from a high pressure (>. 3 p e i ) s o u r c e , a Hg
bubbler roust be i n s t a l l e d Dn tbe entry port s i m i l a r t o that
i n s t a l l e d on the glovebox.

5. Pressure/vacuum gauges should be i n s t a l l e d on tbe glovebox,

glove evacuation l i n e , and the entry port for rapid determi­
nation of the pressure i n these a r e a s .
i
6. The l u c i t e window (>_ 1" t h i c k ) should be made o f s e l f
extinguishing material (Elexson p l e x i g l a s s , SE3), and have
rubber gaskets on both s i d e s .

7. Final checkout of these gloveboxes w i l l include a vacuum and

pressure (_< 0 . 3 p s i ) check.

Since most of the gloveboxes do not meet the above s p e c i f i c a t i o n s ,

I propose immediate upgrade on those now being used programmatically
and continue upgrade on maintenance accounts u n t i l a l l the i n s t a l l e d
gloveboxes are up to these minimum standards. Any m o d i f i c a t i o n o f
these standards or handling of t o x i c m a t e r i a l s within t h e s e gloveboxes
should be reviewed by our F a c i l i t y Managers and appropriate Section
Leader(s).

I would appreciate any comments to improve on t h i s b a s i c d e s i g n

and on any programmatic impacts generated by t h i s upgrade. This
upgrade could be considered as one portion of tbe general upgrade
proposed by Matthias J . Kotowski on December 23, 1980 Csee attached
memorandum).

Richard L. Landingham
Section Leader
Ceramics, Corrosion, & Thermochemistry

RlX:rm
Attachments
Distribution
Bender, C. L-326
Hammerel, G. L-369
Kelly, E. L-351
Kotowski, H. L-384
Lepper, J. L-33B
Mara, G. L-217
Meisenheiroer, R L-370
Bobbins, J. L-326
Short, D. L-217
Southwick, E. L-369
 K»Ut«. Eikhvifc

Lint Ft It

A m in orn'

Grieve- Soy
4/v/rx.
 J4
IrHartii'parlmantaltclt'jrhsad
tAti'StaikmL- 384

E*t: 2525?

December 23, 1980

TO: H. W. Patterson
FROM: Industrial Safety
SUBJECT; Proposal for Glove Box Upgrade
So far this year there have been four glove box incidents in Team 1 areas
that have come to our attention: The two incidents in B/332, one in 8/2*1,
and one in B/231. Reports are available for the first three incidents.
The last incident (B/231) did not result in a breach of the glove box
containment and does not qualify for an incident report because of the
fairly benign nature of the work carried out in the box.

While we are aware of the four cited incidents, there have been, perhaps,
many more near-misses and minor incidents this year that never were brought
to our attention.
Review of these known incidents indicates that the following are factors in
the causal process:

• Operator error caused by lack of understanding o' glove box

systems and their limitations.

• Insufficient considertion Df human factors in the design of glove

box systems that leads to operator error.

• Unforgiving design of glove box systems; e.g. lack of interlocks,

pressure relief, etc.

It's significant to note that these incidents invariably seem to stem from
general safety type events, such as mechanical failure, pressure and
operator error, even through the major problem is contamination by
radioactive materials. We therefore believe that it is appropriate for us
to raise this issue and to propose that the following steps be taken to
improve glove box safety.
rt&commsndations:
1. That an interdisciplinary Laboratory committee be formed to establish
standards for glove box design and use. We would envision that the
committee would have representatives from Mechanical Engineering,
Hazards Control, and Users.

University of California
{•j_£3 Lawrence Livermore
ts?d National Laboratory
 15
-2-

2. That mandatory standard designs be adopted f o r nil nc-w glove boxes and
c o n t r o l systems. Separate designs would be needed t o cover a i r boxes,
i n e r t f l u s h i n g boxes, i n e r t s t a t i c boxes and vacuum boxes.

3, That mandatory minimum requirements be e s t a b l i s h e d f o r e x i s t i n g glove

boxes.

h. That a l l operations r e q u i r i n g t h e use o f standard glove boxes (2 and 3

above) be i d e n t i f i e d by type or c a t e g o r y . ( T h i s would enable
exemption o f low r i s k work from unnecessarily s t r i c t requirements.J '

5. That standards be established f o r what work i s p e r m i s s i b l e i n glove •

boxes i n terms of compressed gas use, flammable l i q u i d s , exothermic •
r e a c t i o n s , heating equipment, e t c .

6. That minimum t r a i n i n g requirements be e s t a b l i s h e d t o teach a l l

personnel the features and l i m i t a t i o n s o f standard g l o v e boxes as
o u t l i n e d i n item 5 above.

Discussion:

1. Standard Design:
The only Laboratory guidance f o r design o f glove boxes i s contained i n
the Mechanical Engineering Safety Manual and H e a l t h and Safety Manual
Supplement 12.03. This guidance i s q u i t e general i n n a t u r e and the
designer has a great deal o f l a t i t u d e when he designs a glove box.
Consequently, there are many d i f f e r e n t designs f o r glove boxes
intended f o r s i m i l a r s e r v i c e . T h i s c o n d i t i o n i s improved only
s l i g h t l y by the f a c t t h a t some components are more o r l e s s standard
and by t h e p r a c t i c e of copying from successful designs already i n
s e r v i c e . I t might be argued t h a t the c u r r e n t system i s t h e best
possible since the box designers can t a i l o r each box e x a c t l y t o the
needs o f the operation, p r o v i d i n g t h e best p o s s i b l e p r o t e c t i o n f o r the
hazards presented by t h a t o p e r a t i o n . While t h i s may have been the
i n t e n t , past experience i n d i c a t e s t h a t t h i s i s n o t n e c e s s a r i l y t r u e .
We see t h e f o l l o w i n g problems which would be lessened by standard
glove box design:

• Some designers have much more experience, t a l e n t o r i n t e r e s t i n

glove box design than o t h e r s . Consequently, the q u a l i t y of glove
box design v a r i e s , and there are same t h a t are not as well
designed as i s d e s i r a b l e ,

• Hazards Control design review i s very d i f f i c u l t . Since each

glove box i s an e n t i r e l y new system, a thorough system safety
a n a l y s i s would be a p p r o p r i a t e . This i s u s u a l l y not possible
because of time c o n s t r a i n t s and lack of d e t a i l e d drawings and
engineering documentation. F u r t h e r , the review f r e q u e n t l y takes
place a f t e r much of the plumbing i s i n place a l r e a d y . This
e f f e c t i v e l y l i m i t s Hazards Control input t o minimun requirements;
human factors considerations and p r e f e r e n t i a l recommendations can
r a r e l y be accommodated.
 76

u Operators and maintenance personnel must learn the peculiarities

and safety procedures for each glove box. When on individual has
to deal with many different boxes this may be difficult;
operators may become confused about which glove box has what
capabilities and may make errors. >
i
2: Standard Operating Parameters
Operating parameters are generally discussed in glove box DSPs. The
review process for such OSPs is intended to assure that all proposed
operations can be carried out safely in the glove box. Vhile this
usually works well, there is little technical guidance for the
experimenter or reviewer. Some examples of problem areas are;•
« How can gas be used in a glove box without undue, risk of over- •
pressure? What are the pressure, flow, and total volume limits?
• Under what circumstances can flammable liquids and gases be
safely used in glove boxes? What flushing rates are required?
Is it wise to permit them in static boxes under any
circumstances? Are oxygen monitors needed?

a H D W much heat can be introduced into a glove box by a furnace?

Is there a potential steam explosion problem if a water-cooled
furnace is used in a glove box?

• What limits should be established for chemical reactions in glove

boxes?
3. Operator Training
While the vast majority of personnel using glove boxes undoubtedly are
well trained in contamination control techniques, additional training
is needed to assure that all glove boxes users understand what steps
must be taken to prevent accidents that could result in containment
failure. This would not take the place of detailed training for a
particular procedure or specific glove box; rather, it would cover the
standard design features and operational parameters and explain their
Purpose. To assume training of all personnel we would envision that
this training be mandatory for all personnel who work in or on glove
boxes.

Matthias J. Kotowski "

Safety Engineer
Industrial Safety Group
MJ<:tdb
 w
SECUVAC® Valves 17

SECUVAC valves are vacuum safely valves which are used to isolate
the vacuurti system from the mechanical pump in case of powerfailure.
They have the following features:

O Close immediately on p o w e r failure

a Co r.st t p a f i until intake litis >> evacuated
• l e a k rate lass than 1 0 - ' Torr Itr/see
• Can be rrtouiited in any positron

.Specially designed for connection to mechanical pumps, SECUVAC

". .valves are oleetro/nagnstieally opened, but close duelo-tho pressure!" ; "j
';.; difference between'.atmdspheric'pressure and "icuuiti.' OneSECUVAC:*•£.*
~\va'lve performs the" functions of two valves/i.e. a rapiJI'f'cTosing vac-"''
•^uum valve for isolating, the vacuum chamber <and/or-vapot. puro_^}^-.'- 1

= Vffom'ihe'bKlting'purrJp'tirid arc'aif inlet-valve for veptiflrf the "tecking""' -.'•

Flpump.frecau.seisolation and venting follow at.a-'predeteimined intervaV7;-^~f>-;_r~
1-
""o? time (brought about by means of devices-built into the valve), t h V " > - .
:__imounVof air drawn inio-the vacuum chamber is less the>45c>0_SJ'ofc_=r. •"•-=—•
_
!"J Jtfpei.eniVoJ the vatoe plaie.'sor1ace ar'ea. o'jt opening.fh&.vatVRtneC -i'.KF!9.'W'' SECU'VAevllv« KF 3i"
:

'•• 'pressure iise" in the fore-vacuum'4ine of standard pump^sets comprising-*: • . - ! " " ' . •'
J
:_- of difhjsTorr and-backirrg pumps,-does not exceed the critical'backing-^-*;:-^
* pressure- of the diffusion pump;By parallel connection of the d r i v e ^ '££,
v__mow of the backing pump and the_solenojd-.coii_of.1hj*^SECUVACTr-?. - f '^-i.. -
f,
"valv'ff"_'the vacuum chamber and the" vapor pump ar4 protected from""" *'J"' ~"-'""-
ai\ inrush of air on power failure and ihB backing pump is vented. The '
venling time, is so fixed that sucking back of oil from the pump Is
avoided.
The SECUVAC is a right angle valve. Conductance and comer dimen­
sion across the valve correspond to those for pipe bends of the seme
7
nominal diameter. The leak late is less than 1 0 ~ Torr Itr/see. The values
from KF 10 W KF 32 are of steel and KF SO, 3" and 4" are of light
alio/.

Applications

As vacuum-securing isolation watvc% between backing pump and

vacuum chamber or vapor pump - for protecting the vacuum chamber
from inrush of air when itie backing pump stops due to a power failure
FiD. 1.2 SECUVAC M t , . 3 "
or twitch-off.

rilnll In Vftti G*mn>r

 w 18

_JS=»~7~
l***

_tt. T£-i-r O
ft ••7-" - \ Y ••••{._•_•:!
- i- -„— ?;•:
AV-- --.--.' 4 - • • - • - - - ? ~ 4 1 • • -

Qr>iSt ^'jr
Fi'O. * 3 D r i y 4 f n g Of tfie.SECUVAP vaWft*3'*jpd V - . . . ' ; •".•.-• > '**•
* :fjrfi-»4^Sl*ndiidjr9WtfngxOn^ .

Technical data

Size KF10 KF20 KF32 KFSO 3" 4"

1=/,, 2 2 27. 474 67,

2V. 37.. 3V, S7„ 7'/ 2
fl'/i
3'/,, 37, 37. 4V. 6"/„ 77.
6% 67, 67. 7V, 10V« 127,,
1.7 2.4 3.5 4.B 8 15
1
AmptKia* « 1 1 5 V, 60 Hz )
0.17/0.11 0.17/0.11 0.17/0.11 0.17/0.11 0.2(70.11 0.26/0.11

SECUVAC vilv« . . . . C*t. No. 273 01 -1 273 02-1 27303-1 273 05-1 272 77-M 272 BD-M
o
J,«44llMjA.imMw?>MKnf.'l4ctifTi»pvuir<»qiil»<lfoiconri.<<ingl>i>rilv,.
') J » V. 10 Hi. ITKXM. nwfliW* « i nqunt.

LEYBOLD-HERAEUS, INC.
 CPA FAIL SAFE — MODEL # 9860
MECHANICAL PUMP FAILURE DETECTOR FOR
BELT DRIVEN PUMPS

-
i-
Price $135.00
• Protect your vacuum systems from mechanical pump failure

• The mechanical pump failure detector has a moving magnetic field that is detected by a transducer, mounted on the belt
guard ol the mechanical pump. The output of the transducer is electronically monitored by a missing pulse detector. If
the puMeyceases to rotate, the circuitry actuates a relay which is used to interlock the vacuum system.

• For thin tlm evaporators, sputtering systems and other diffusion pumped pumping systems. It the mechanical pump stops
lor any reason - belt breakage, vane freeze up or motor failure, the valve at the foreline of the diffusion pump should be
closed immediately, to stop air and mechanical pump oil or oil vapor from being sucked bach into the hot boiler of the
Oilfusion pump At the same time, the diffusion pump heater should be turned off. and the hi-vac valve closet to prevent
Contamination of the wo/king chamber.

• All ol this can easily be done by connecting the mechanical pump failure detector to a mullipole power relay which can
close both valves, and turn off the diffusion pump.

• for other pumping systems which use s belt driven mechanical pump, the mechanical pump failure detector can be con­
nected d«ectly to a sonic alarm, visual alarm or interlocks that require two <2) amps or less to operate. For interlocks
that reqiwe more than one (J) relay contact or currents in excess of two (2) amps, an auxiliary relay can be easily con-
necifd.

SPECIFICATIONS
1 D?tec*.sttie motion ol the mechanical pump pulley.

2 Provides one form C (single pole douBle throw) 2 amp relay contact for interlock purposes.

3 - Operates on 11S V AC line. Requires no special or auxiliary power supplies-

1 • Ppwrr (equipments 1 ] 5 V AC • .015 amps

5 • Nominal we o'- the transducer bo« 2" deep x ZV." wide n AK" Ions

7 2 5 K I F E H RCJ. • S U N N Y V A L E , C A , 3 4 0 0 6 . (<3QS) 7 3 3 - 9 8 3 3
•t/77
 — i/i
^-1 (V

v 11 J L
z. § l-S zzm
3
S3 o

s
cj S-
£-A-

St X\\\\\1
y' / y / y * /
00 f>

It I////// ' / / / / / / ,
c
S
.«V

m
 21
/?;.•••?.-,-., .•;••.•• .!.:il-.!t( :!!•,:•!

L-< 2-5165
June 21, ]982

TO: Eugene Haimierel

FROM: Safety Science Group, Hazards Control Department
SUBJECT: Bursting Pressure of Various Gloved Box Gloves
M. H. Chew and P. E. Barry determined the bursting pressure and approximate
volume of a 30-mrn neoprene and butyl gloved box glove in 1975, attachment 1.
To confirm their data and expand the data base, a series of gloved box gloves
have teen burst (15-, 25- and 30-mrn}. , .
Each glove was attached to the properly sized gloved port, 5" or 8" and secured
with two 0-rings. The mounted glove was inserted into the port with the out­
side edge of the port pressed against a rubber coated aluminum plate and held
in place by four or si.", bolts. This arrangement corresponds to the configura­
tion used to secure gloves on LLNL vacuum boxes. A 1/8" x 17 NPT inlet in the
plate equipped with a needle valve and compressed air line was used to inflate
and btirst the test gloves. Pressure within the inflated gloves was measured
by a calibrated magnehelic gauge. The results of our tests are listed on the
attached table.
As pressure was applied to the test glove it became rigid and the pressure rose
significantly. After the glove began to deform the pressure dropped quickly
(< 30 sec) and leveled off at the burst pressure. As can be seen from the
times reported to burst, there is in some cases a significant variation. Since
our ialet flow was set at a constant rate, this fluctuation in time must be
related to the variation in elasticity of the gloves we tested.
Stan Lord observed all of our tests and participated in the data collection and
experimental observations. If you require any additional information or tests
don't hesitate to give us a call.

inson, Leader
Safety Science Group

Charles Harder
Safety Science Group
cr: J. Becker
R. Griffith
 22

Glove Bursting Data

Initial Burst Time to

Glove Type Pressure psi Pressure psi Burst (Min)
1 Cat. #5N153Z Glove #1 0.90 (25. OJ* 0.72 (20.0)* 7:'.S
Cuff Dia. 5"
15 Mil Glove #2 1.22 (34.0) 0.68 (19.0) 3:30
Neoprene Norton
2 Cat. #8BSF-1532A Glove §1 0.67 (18.5) 0.53 (14.8) 9:01
Cuff Dia. 8"
15 Mil Glove #2 0.65 (18,0) 0.57 (15.8) 9:14
Butyl Norton
3 Cat. #831532 Glove #1 0.43 (12.0) 0.40 (11.0) 7:47
Cuff Dia. 8"
15 Mil Glove €2 0.43 (12.0) 0.38 (10.5) 7:30
Butyl Norton
4 Cat. #8B3032 Glove #1 0.83 (23.0 0.68 (18.8) 7:49
Cuff Dia. 8"
30 Mil Glove #2 0.86 (24.0) 0.66 (18.2) 7:04
Butyl Norton
5 Cat. #8N3030 Glove £1 2.16 (60.0) 1.76 (49.0) 3:39
Cuff Dia. 8"
30 Mil Glove #2 2.27 (63.0) 1.84 (51.0) 4:18
Buta-Sol Charco
6 Cat. #7BT-2532 Glove #1 0.68 (19.0) 0.46 (12.8) 8:43
Cuff Dia. 7"
25 Mil Glove #2 0.64 (17.8) 0.49 (13.5) 19:30
Butyl Norton
•inches of water
 23

ENVIRONMENTAL PROTECTION-

Gltivcd lloic Expansion and Pressure Test Both gloves were quickly filled with air to simulate
In a icccot experiment in Building 241, an inert a sudden expansion. A moiinmctcr was used to
hard vacuum gloved box was used to perform measure the resultant pressure On the glove. Both
experiments with unstable hydrides. Theoretically, if gloves leached an equilibrium pressure within a few
the hydride* were to disassociate, a large volume of seconds. However, the free volume obtained by the
gas under piessure would expand into the gloved box. butyl glove was approximately twice that of the
A 2-psi pressure relief valve was deemed necessary for neoprene glove before bursting.
safe relief of the entire system. The relief valve would The results of the tests, arc given below:
be adequate if all the gloved ports covered with the
gloves inside the box were properly secured. However,
during the times when the gloves were in use, the weak
Bursting Approximate
link in the system would be the bursting pressure of Glove pressure volume
the gloves.
A quick experiment was set up to determine the Neoprene 1.1 pa 0.5 m 3

bursting pressure and the approximate free volume of Butyl 02 m 3

the Iwo standard types of gloves normally used at LIX.

Clove A is fabricated from 30-mil neoprene, and
Glove B is fabricated from 30-mil butyl rubber with
a durasot coating. Both are manufactured by Charleston In conclusion, the bursting pressure of the gloves
Rubber Co. should be considered in designing a pressure relief valve
Each glove was attached to a standard Sin. gloved for the system.
port and was secured with an aeroseal clamp. An The following photographs show a sequence of the
aluminum plate was attached to the gloved port with glove expansion. The black glove is the neoprene glove
the proper hose-barb connection. (fig. 44).

i Z~—,.L

Fig. 44(A). Nroprcne {Jove expansion test.

 24

. :t

i. 'i
Ji_iii

Fig. 44(B). Butyl rubber glove expansion (est.

*
 26
Interdepartir.vi nnl Ivttrxi'ufid

f>
' 25267, 25170

December 20, 1982

TO; Stan Lord L-341 -

FROM: Jerry Schweickert, Industrial Hygiene Groups^)
RosswUson, Health Physics Group (/
SUBJECT: Airflow into Glove Box Openings

The Health and Safety Manual Supplement 12.03, "Work Enclosures for Toxic
and Radioactive Materials" addresses airflow requirements for glove box
openings. The inward airflow through any opening to control hazardous
materials is 150 linear feet per minute (1/pm) when feasible. However, a
lower value may be accepted in some situations.
For the case of vacuum boxes which are not designed to move large volumes
of air, a rate of 150 lfpm is difficult to achieve. This criteria is valid
for the box proper where loose te«ic materials will be used and an
accidental breach could cause external contamination. In pass through
ports, an airflow less than 150 lfpm is acceptable if hazardous materials
are suitably contained or in a nondispersive form. At no point in the
opening should the flow drop below 50 lfpm, which is the minimum needed to
overcome room drafts or turbulence from passersby.
After modification, e?ch glove box should be evaluated by the Hazards
Control Safety Team to ensure its adequacy for the intended use.

JS:RW:jre
cc: J. Becker L-327
P. Linnes L-369
Bldg. 241 File
Date File

University of California
23 Lawrence Livermore
National Laboratory
 ATTACHMENT # 5
Interdepartmental letterhead

Ma!Statk>nL369

February 28, 1986

EDS-86-19

TO: K. L, Blaedel

FROM: E. H. Schmitt £]*£.

SUBJECT: Check Valve Development For The Kereury Bubbler, MK11

A number of accidental spills of mercury have occurred while a glove box

is under vacuum when the bubbl .-r has been jarred allowing a burst of air to
push the mercury into the vacuum box not only contaminating the box contents
with mercury but air as well. A solution to this problem Is to place a ball
and seat check valve between the glove box and the bubbler to act as a
positive vacuum seal and hence limit the amount of mercury required while the
glovebox is under vacuum. Conventional check valves available through vendors
are not compatible with many of the glove box contents. The possibility of
corrosion or a reaction between the glove box contents and the check valve
could lead to a catastrophic failure.

The cheek valve design shown in the attached drawing (AAA8U-111179)

provides a high conductance flow channel during the pressure relief state by
pushing a hollow stainless steel ball (item 6) located in a cage (item t)
against a flow through seat (item 1). Alternatively, the check valve provides
a seal against flow during the state where the glove box is at vacuum by a
hollow ball floating on a column of mercury thus being forced against a Sil
Guard seat (item 2 ) , A aight glass Is provided in the design to allcw an
outside observer to view the location of the stainless ball.

EHS:bjh

University ot Caltnria
LAWRENCE UVERMORE LABORATORY
 NOTES
UNLESS OTHERWISE SPECIFIED:

1. DIMENSIONING AND TOIERANCING PER

ANSI YM.5M-I982.
2. SURFACE TEXTURE PER ANSI B46.1-I97*.
J. F*R.£iS FiT ITEM 4 OW ITEM!

OfZlliUT TO A\A5IO POtZTS

1(3.00)

SOC HD> C A P 5C<g | Q - 3 ^ . X "L'A 1 6 CTgCS

'OT^ING I.T37 I P X.IP3 W E3UMR N ' Z - I ^ z l PARICtCtt
BOLL, MOLLOWj & /.Z5Q1-CCZ X0W, <Z€1£S WO
<U
Bt-iine'f SIGHT GLASS
e^-Miisj

a FMU/llNl STK NO
O'PPEfe
LOw£(Z.
CAP
CAP
DESCRIPTION/MATERIAL SPEC MQ Itutt
DWH |. MfffclW/ CIASSIF1CAJI0N
^M BUBSLei? • E f l C C . CHECK
Hn/ygiLTs^soog^j^^l THIS DOtlKE'T IS
THE K l f E l K OF CHECK vWLVE
THE
;,.*»,».V, *± I I H I V E l S l t t OF C U I F O I I I A
Atf iTT- DAtt
LAHEICE LIVE "0FE
LIIOM10«Y
l

ttFROBUUlOl PK0HUITE& 1H0WM OH A •.It DRAWING NO

IAWIEMCE UVERMORE

u. irs/-r A tvtos T&*Z.O/S U NATIOMAl UHOJIATORY

MECHANICAl ENGINEERING DEM
uuiuimiT of conroiMiA
UILESS PEAM1TED BY THE
HECHAI1CAL EKSI•EEAIHS '
DEPAATAEAT j
AAA64-IIII79 -0A
ISHEtl / oTT