ML20195J904
| ML20195J904 | |
| Person / Time | |
|---|---|
| Site: | Quad Cities  |
| Issue date: | 09/24/1998 |
| From: | Sparacino J DUKE ENGINEERING & SERVICES |
| To: | |
| Shared Package | |
| ML20195J807 | List: |
| References | |
| TR-VQ1500-02, TR-VQ1500-02-R00, TR-VQ1500-2, TR-VQ1500-2-R, NUDOCS 9906210150 | |
| Download: ML20195J904 (16) | |
Text
_
hh '
TECHN1 CAL DOCUMENT COVER SHEET p$"*3 $i"^
REVISION 0 DOCUMENT NUMBER: TR-VQ150042 RL V: 0 DOCUMENT TITLE: Clean ECCS Suction Stratner Head Loss Test Repart PROJECT NAME: Quad Cities Station ECCS Suction Strainer Replacement WID NUMBER: VQ1500 CLIENT: Comed, Quad Cities Station Units 1&2 (00010.05.0001.00.00000) l SOFTWARE USAGE I
- Pre-Use (Retain u:agefhun petor restsnons, (sn!! appucable)
Hardware Platform / Description of Functions, Features, Modules, Version I VerificationSoAware Name Operating System Libraries. Modeling Techniques l
l
- Review Seaware Capabilities, Review Opm Faw Nues, F.nsm Inmallation Tan Canplated, and Access Control Satisfied; per DPR-3.5.
DESIGN VERIFICATION METHOD: DESIGN REVI2W CRITERLA SOFTWAPIREVIEW CRITERIA fp,,I8 8 M'" m HA IIEM m Ha IIEW (V D DemipinputcomalySetaned a trsaawncap bilisi Revww.d O Alternate Calculation a tr." -- Adequate /Rossanable Q FOpen Emw Notices Revwwed 0 (^ - Noted for Verifush D B"Soawn Used comov s l
O Quali5 cation Testing tr a 3,,,.pnas.D ipMedwas a maSoa ,e Re.h.%L .s D (YDesign lupuis ' r- J in Desihn D Wiey Prograni Feehares Record.ed V O DMOutputfartheInputs l 0 [Interfaang C.. : Specified i
Preprer Verifier Apprmcr 1
. b. 11A.utz W .
Printed Name
/J088PhDsp DMd Md M Nm2 o* l //a4 /95) f/ahs 7-p#- #
9906210150 990615 PDP, Page
! d /h l ADOCK 05000254 ,
, P PDR
l O. D_ _E & 5 FORM DPR-3.9A
. PAGE 1 OF 1 REVISION 0 DESIGN VERIFICATION CHECKLIST
- 1. Were the inputs correctly selected and incorporated into design? N NA Y Are assumpnons necessary to perform the design activity adequately described and Y N reasonable? Where .necessary, are the assumptions identified for subsequent reverifications when the detailed design activities are ccmpleted?
^
- 3. Are the appropriate quality and quality assurance requirements specified? Y N
- 4. Are the applicabic codes, standards and regulatory requirements, including issue Y N A Q[A,)
and addenda properly identified, and are their requirements for design met?
- 5. Have applicab't. construction and operatmg experience been considered?
Y N
- 6. Have the design interface requirements been satisfied?
7.
Y N @
Was an appropriate design method used? Y N g 8.
9.
Is the output reasonable compared to inputs?
W N NA Are the specified parts, equipment and processes suitable for the required application?
Y N h
- 10. Are the specified matenals compatible with each other and the design environmental conditions to which the material will be exposed?
Y N
- 11. Have adequate maintenance features and requiremenu been specified? Y N @
- 12. Are accessibility and other design provisions adequate for performance of needed maintenance and repair?
Y N @
- 13. Has adequate accessibility been provided to perform the inservice inspectic:.
expected to be required during tLe plant life?
Y N h 14 Has the design properly considered radiation exposure to the public and plant Y N QjA) personnel?
- 15. Are the acceptance entens :.icorporated in the design documents rufficient to allow verification that design requirements have been satisfactorily accomplished?
h N NA
- 16. Have adequate presperational and subsequent periodic test requirements been appropriately specified?
Y N h
- 17. Are adequate handling, storage, cleaning and shipping requirements specified? Y N g 18.
19.
Are adequate identificanon requirements specified? Y N h Are requirements for record preparation review, approval, retention, etc.," Y N adequately spe:ified? Q93 l
Document No.tTR-1/Q 90 -0 A Rev.: O
Title:
Olean Eu.5 sudon Stru;ar bllossT;d-1 Verified By: [r _' _ Date:
'l
[f!i/
/
p y e. a. J W
\
.A DE&S V wer,-vgumar Technical Document TR-VQ1500-02
. Page 3 TABLE OF CONTENTS Section Title Page
- 1. Executive Summry 4
- 2. Introduction 5
- 3. Purpose 5
- 4. Test Apparatus 5
- 5. Methodology 6
- 6. Analysis of the Results 7 7 Summary and Test Conclusions 15 8 References 15 1
1 1
j l
, o D@$
(*wumen Technical Document TR-VQ1500-02
, Page 4 l
- 1. Executive Summary Duke Engineering and Services performed head loss testing of a clean Quad Cities ECCS suction strainer on April 30,1998 at the Fairbanks Morse Pump Co. test facility. The j head loss testing consisted of measuring and recording head loss data for a Quad Cities ECCS suction strainer in the test facility at flow rates ranging from approximately 2600 gpm to 12,000 gpm in increments of 1000 gpm. The results fram the head loss test performed indicates that the strainers will perform equal to or better than the head loss information supplied by the vendor, PCL Therefore, a Quad Cities ECCS analysis which is based on the supplied vendor information on the suction strainer will be conservativa in its 1 assessment ofECCS operability. l l
gyg V map =,u Technical Document TR-VQ1500-02
, Page 5
- 2. Introduction During a loss-of-coolant accident (LOCA) in a Boiling Water Reactor (BWR) nuclear power plant, pipe insulation may be L-.sporh from the contairanent to the suppression chamber, ne suppression chamber supplies water to the Emergency Core Cooling Systems (ECCS).
His insulation, combined with corrosion products and other debris, can migrate and collect on strainers installed on suction lines supplying the ECCS pumps. Relatively small amounts of fiber insulation combined with corrosion products have been sh w.n to result in a significant pressure drop across suction strainers, which could in tum result in ECCS flow degradation.
ne overall pressure drop attributed to the stramer is the sum of that which is caused by debris on the strainer and that due to the clean stramer. n's test was performed to provide information regarding the pressure drop across the clean strainer. This test report will document the results of the full-scale test that was performed to determine head losses caused by a clean Quad Cities Station, ECCS suction stramer, designed and built by Performance ContractingInc. (PCI).
- 3. Purpose DE&S performed the Quad Cites clean strainer head loss test procedure [Ref. 8.2] on April 28th of 1998, at the Fairbanks-Morris Pump Co. test facility, he objective of the test was to measure the head loss across a clean Quad Cities ECCS suction strainer at varied flow rates, including the design flow rate of 10,000 gpm, for companson to the head loss value theoretically calculated by PCI. He purpose of this report is to document the test data and head loss evaluation results.
nese test results will verify that the theoretically calculated clean strainer head losses previously submitted by the vendor for the Quad Cit;es ECCS strainers are in fact conservative. Le theoretical clean stramer hydraulic performance is provided in the PCI Head Loss Calculation document [Ref. 8.3].
- 4. Test Apparatus The Fairbanks Morse Pump Co. test facility has the capability of mounting and testing the svetion strainer per the criteria set forth in the Quad Cities suction strainer test plan [Ref.
8.1]. All criteria and prerequisites within the test plan and test procedure [Ref. 8. land 8.2]
were followed. He actual test apparatus setup is shown in Attachment A. He strainer was connected to a 20-inch diameter and 144 inches long suction pipe, which was connected to a 61 inch-pump spool piece with straightening vanes. The straightening vanes are used to prevent the pump impeller from imparting any rotation to the flow of water in the suction piping, he 6.1-inch pump spool piece has an instrument tep located 7 inches above the face of the flange that connects to the 144-inch long spool piece (See attachment A). ne suction piping and ECCS strainer were mounted vertically in to the pool of water and the pressure tap was used to measure the head loss across the strainer via a water manometer.
I i
)
'g g g V us, vurva Technical Document TR-VQ1500-02
, Page 6 A 17.5 x 12.75-inch venturilocated on the discharge piping was attached to a mercury manometer and was used to measure the flow rate.
A minor correction factor of 1.6% is required for the manometer used to measure the head loss. The correction factor is required due to the fmite dimension of the resen oir that contains the manometer measuring fluid as compared to the manometer tubing. His correction has been used in previous similar tests with the same instrumentation. The basis for this correction is documented in a letter by Fairbanks Morse [Ref. 8.4].
5.- Methodology he test procedure consisted of steps for measuring the head loss, at different flow rates, across the test setup which included the clean Quad Cities ECCS suction strainer. This information will be used to venfy the clean strainer head loss at the desis,n flow condition and to develop a clean stramer head loss curve as a function of flow rate. The tabular results of the test data are shown in Table 1, and the clean stramer head loss curve is shown in Figure 1.
Once the stramer and test apparatus were set up, the pump was started and the flow rate was set to approximately 10,000 gpm. The head loss of the test setup was then recorded at the flow rate of 10,000 gpm. The head loss of the test setup was recorded at each of the flow rates as it was adjusted up to 12,000 gpm, down to 3000 gpm, and back up to 12,000 gpm in intervals of j 1000 gpm. '
As steed earlier in this report, the measured head loss must be corrected due to the i mstrumentation by increasing the head loss by 1.6%. The results of this correction are shown in Table 1.
In order to detemune the actual head loss of the clean ECCS suction strainer, the head loss due to the suction piping and the velocity head of the fluid must be subtracted from the corrected test setup head loss data. Performing these iterations on each of the head loss values will produce the head loss due solely to the suctum strainer. The head lora due to the suction piping willbe calcule.ted as follows:
lu = f(UD)(v'/2g) Eqn. {l], (Ref. 8.5]
where:
lu = the head loss of the suction piping (ft. H2O).
f= piping friction factor L = length of the piping (ft.).
D = Ameer of the piping (ft.).
y = velocity (ft/sec.)
2 g = gravitational constant (32.2 ft /sec).
x
'6 DEES V cars,wrgui,= Technical Document TR-VQ1500-02 Page 7 He velocity head of the fluid in the suction piping will be calculated as follows:
hv = v2/2g Eqn. [2], [Ref. 8.5]
where:
hv = the velocity head (fl. ofH2O).
v = velocity (filsec).
. 2 g = gravitational constant (32.2 ft /sec).
He results of the calculated suction piping head loss and velocity head are documented in Table 1. nese values will be subtracted from the corrected measured test setup head, as shown in the following equation, to obtain the head loss solely due to the suction strainer. The results of these manipulations are shown in Table 1.
i hos = (hw x 1.016)- hu,- hv Eqn. [3]
where:
hts = the head loss of the strainer only (fl. H2O).
hm = the measured head loss across the test setup (fl. H2O).
hui = the head loss of the suction piping (fl. H2O).
- 6. Analysis of the Test Results !
The results of the Quad Cities clean strainer test are shown both in Table 1 and Figure 1. ;
Strainer head loss values were recorded at various flows ranging from approximately 2600 '
gpm to approximately 12,000 gpm. The plotted curve of strainer head loss test data verses the flow rate shows that the measured strainer performance matches the calculated strainer performance curve documented in the PCI report [Ref. 8.3]. He test curve has a shape similar to the calculated curve shown in the report. He test curve shows slightly lower head losses than the calculate? values. This is clearly seen when comparing the measured strainer head loss to the predicted strainer head loss at 10,000 gpm. He measured strainer head loss at approximately 10,000 gpm is approximately 1.85 ft ofH2 0. He predicted bounding strainer head loss at 10,000 gpm, as documented in the PCI report is 7.07 ft. ofH20. As can be seen, actual mercured head loss of the strainer is slightly less thar. .ne predicted head loss.
He curve shown in Figure I depicts the relationship between head loss and flow rate for a clean Quad Cities ECCS suction strainer. This relationship can be expressed mathematically with the following equation:
- g y g V ma,.gsma, Technical Document TR-VQ1500-02
, Page 8 4
Head Ims, hos (fi. wc) = (1.8887 x 10 ) Q'-(1.7545 x 10 4) Q + 0.1379 Eqn. [4]
where:
Q = Flow Rate (gpm)
Equation [4] was derived by performing a curve fit on the test data. To illustrate the accuracy of the curve fit, the calculated head loss results, using the test flow rates as input to the above equation, are shown in Table 2 along with the measured strainer head loss. Figure 2 shows the curves of both the measured head loss and calculated head loss vs. flow rate. As expected, Figure 2 shows that Equation [4] accurately predicts the test results. The above equation will accurately predict the head loss for a clean Quad Cities suction strainer for flows in the range of 2,600 to 12,000 gpm.
Traditional fluid flow theory would predict that the head loss for turbulent flow through a fitting would be proportional to the square of the fluid velocity, V, through the fitting, which is represented mathematically by the following expression:
2 hos = K(V /2g) Eqn. [5]
In Eqn. [3], K is referred to as the resistance coefficient. In reality, the flow through the strainer is very complex. De head loss through the strainer is dominated by turbulent flow through the strainer core tube but there is also a lesser contribution due to laminar flow through the perforated plate. nese multiple contributions are reflected by the fact that the head loss through the strainer is represented by a quadratic equation, Eqn. [4]. Furthe:more, the fact that the strainer head loss is dominated by turbulent flow through the core tube is reflected by the 2
fact that the Q term is much larger than the other two terms in Eqn. [4]. Because of the strong dependence on turbulent flow, the strainer head loss can be adequately approximated by Eqn.
[5], which gives rise to a strainer resistance coefficient, K.
In order to determine the resistance coefficient, K, for the strainer, the above quadratic equation is simplified so that the head loss is proportional to the square of the flow rate. As discussed above, the simplification is acceptable due to the relative insignificance of the second and third terms in Eqn. [4] as compared to the squared term. Herefore, the strainer head loss can be conservatively represented by the following equation:
4 2 Head Loss, hts' (ft. wc) = (2.3 x 10 ) Q Eqn. [6]
He calculated approximate strainer head loss is also shown in Table 2. Conservatism was 2
added to the stramer head loss calculation by changing the coefficient on the Q term from 4 4 1.8887 x 10 to 2.3 x 10 . He affect ofincreasing this factor, increases the calculated approximate head loss such that it adds conservatism at larger flow rates, as well as provides a better fit to the actr.al data at the lower flow rates. De calculated approximate head loss curve, along with the measured head loss curve, are shown in Figure 3. Dese curves indicate the discussed conservatism at higher flow rates as well as the better fit at the lower flow rates.
. 0 DEE5 a*apeguman Technical Document TR-VQ1500-02
, Page 9 l
l l As indicated in Table 2, the resistance coefBeient, K, for the strainer was determined to be 1.16. The resistance coefficient was determined using the following equation and the calculated approximate head loss values. The velocity of water, V, through the strainer was determined using a strainer core tube inner diameter of 19 inches.
K = 2(hu')(g)/V' where:
K = Resistance coefficient (dimensionless).
ha' = Strainer Head 1.oss (fi. wc).
g = gravitational constant (32.2 ft/sec').
V = Velocity through the strainer (ft/sec).
1
y lns, s f e
'20 Oo )
c 42335655893828841 31 45 oh)t . 3 "01 rL e sw o 42545081 19753235791 1 71 604494816 h o(f t
8
- 0 -
e g i ndh a f t
82.62851 1 2221 1 1 0000000001 5826 t gt p nea 5
'0 a a e ( 1 1 22 2
=5 P DQ 1
r S H t
e pT Li P 1
d V-iR d "T ey , )c t t r
e ).
aidwc 244423582261 622853244 l
uo a cl ef 603963075321 t
253520398951598930253 235703693 I n
n (nI 0
laVe H (v 1 1 21 1 1 1 0000000001 1 1 1 2 e pi a 2
C h iD P
go p dinh ep , ,
laps s )c 3404338551 7071 7328929 t i :
unL ow 582852975421 1 1 21 1 1 00000000001 245702592 co t 1 1 1 2 l
Cca atid( f 000000000000000000000 ue s SH s
o Lgnr dnitoo aitpcc 333333333442443333333 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 eiia 0000000000.0.0000000000 HPrF r
F 000000000000000000000 e
isn 1
ade r'l r 666666665555555666666 00000000000000C000000 bletSob nm + + + + + + + + + + + + ++ + + + + + + +
EEEEEEEEEEEEEEEEEEEEE a nyu 84048271 89949981 72840 TaeN leR 57975421 4721 2741 2 4.G.7 9 1 1 1 1 1 1 1 1 97646791 1 1 1 1 1 C
s i en gt c y) 08080 io ince t i 21 61 2 051 07222701 50 080 Ct c plios duie/ t 22222.16672267622766 0121 0 1 11 006001 1 1 1 21222 6 aS P u V (f 1 1 1 1 1 9876542456789 012 1 11 Q
dds s eer o )c t
cuLw es 007202895369626084208 2608331 85553448227662 r ad r
oe af t 6 3 2 3.6.9 e( 34543221 1 000001 1 22345 3 7 3 9.6 3.6 9 3 8 3 9.6 3.2 CM H ds es ro )
c 3253351 03863575227426 n
suL uw 6921 78863343326992094 521 adh ea t f
(
3 5 8 2 7 3 9 6 3.6 9 3 17 22345 345432211000001 2 9.t.5 21 Me H Sm Es de et E% r uRm a) 08080 29292 0500820280050 080 Du awp s
e o (g 09090 00200 8030282820308 9000996990009 292 090 8876432346788 002 g' MF l 1 11 1 1 11 1
gn-b 6o
?$
i H !
1
^
k g l 8 3 I l T i l li,
=g
! ~
l i gi :
l I
A !
$ l O
mg g g g g g g g Oe e e v M N s c Ek h4) ssol peeH Q$
0
s ~ .
tn 22 e01 -
m0 u0 ge )
c5 aP e K DQ o1 c(
ntn 000000000000000000000 l
a V ae 666666666666666666666 c -
t i sci 1 1 111 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 iRnT i
sf h ef e 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 c
e Ro T C t
n e 's ]
i c dt et6 h[
i f ea , .
f e
o t
aisqcmsn) 923295478958598745923 C
l u
c roL xoEw 082805732565652375082
. 37373841 8,531 3581 48373 e
c l
a p d n (f gt 223221 1 1 00000001 n 1 1 223 a Cpai s i
t s
s AeH U e
R d
n a
s s r L
o e n s]
a r h [4 t
i d
a 2 e t , .
l eH S ssqn )c 89698388308?803883896 ba de Tt a doEw eL t 525250041 1 6?
82628529753.?357925826 61 1 400525 t gf 1 2221 1 1 d l adn( 0000000001 1 1 22 l
e cu ai es a
C r
l C
aHU e
i n
a r
t S
n s o
a h l
e y ,
C nss l
s e
i t O r L o )c 42335655893828841 31 45 i
C ed w 42545081 8202851 1 1 71 60449481 9753235791 5826 6 d nat f .
122211 1 a
u ie(
a rH 000000000111 22 Q t St s e
T
)
m p
d g e
r( 08080 ue st 29292 8030282820308 09090 0500820280050 292 080 a
eRa 9000996990009 00200.8876432346788 090 002
$% a Mwo 1 1 1 1 1 1 1 4 l
@1 : F r*
. ~
s t n 23 e41 e mwg u
cs oiP a Do l
a v- O iac h
nr M c
e 1 T
0 0
, 0 2
f 1
s e
s mep,he.
s
' 0 a.p j
- o. 0 i )
e 0
mwwf(
I 0 w d t a 1 e .
H
/
r .
e n .
i a 0 0
M) t r
0 P S
n 8 G 2 a ee 2 (
t e
r uQ a g
id F et l
a u
- 0 0
0 R
w o
c 6 l u
F J
l a h, C s s
d yo n
a 0 l
nL d - 0 Oda e 0 r e) e r
u - 4 i nHc w s at s
_ r et f .
a t e ST(
M _
0 0
0 2 _
0 S 0 00 00 00 0 0 000 00000 0 _
0 5 0 5 05 0 5 0 _
E" 433221 1 0 0 D^* g3 d lJ mm*I g'
t n 24 e 01 -
m0 u0 ge c5 a DO o1 P 0 0
a V l
c R- 0 i
h nT 4 _
c 1 _
e _
T 0
0 -
0 2
/
s e
e 1 _
L .
d .
a e 0 's ] _
eh[t6 H *
- 0 t r das ems nq e
n ., 0 t aio E) i a
0 xL luordgc c pai nw _
t r
1 lapest .
S -
- CAHU( f d
M d
e m
m aa V 0 0
0 8
4 R
G
(
n a t e
3 es
- a e
sL pd ha e I
e
-V 0 0
0 R
m H 6 k
/
r e F s t
i n h s,
a r s t
S 0 lyo n L t
s 0 Oda e
m 0 r e e n H )c 4
i m iat s w tr et .
m p
ST( f p
A 0 d 0 e
d t
a 0 l
a c 2 a
C 0
0 0 0 0 0 Sh 0 0 0 0 0 s 0 0 0 0 0 E% 4 3 2 1 0 Dr* _v[ =0 .'4c( 5=z
r
, gMM Technical Document
' w wer,. gum = TR-VQ1500-02
, Page 15
- 7. Summary and Test Conclusions The analysis of the results from the head loss test performed at the Fairbanks Morse Pump Co.,
on the Quad Cities clean suction stramer, indicates that the strainers will perform equal to or better than the head loss information supplied by the vendor, PCI at 10,000 gpm. Therefore, a Quad Cities ECCS analysis which is based on the supplied vendor information will be conservative in its assessment of ECCS operability.
In order to describe the strainer head loss mathematically, a quadratic equation was derived from the test data. In order to det rmint the stramer resistance coefficient, K, the quadratic equation was simplified so that the head loss would be p oportional to the square of the flow rate. At lower flow rates, the calculated strainer head loss was slightly different than the stramer performance. This required that conservatism be added to the simplified strainer head loss equation by a slight increase in the squared flow rate coefficient . With this simplified equation, a resistance coefficient, K, was determined to be 1.16, between the Pow rates of 2,600 and 12,000 gpm.
- 8. References 8.1. DE&S Technical Document No. TPP-VQ1500-001, Rev. O, " Plan for Head Loss Testing of a Clean, Quad Cities Station ECCS Suction Strainer", dated 4/27/98.
8.2. DE&S Technical Document No. TPP-VQ1500-002, Rev. O, ". Head Loss Test Procedure for a Clean, Quad Cities Station, ECCS Suction Strainer", dated 4/27/98.
8.3. Performance Contracting, Inc. Report No. PCI-NPD-CE01, Rev. 2. "PCI Head Loss Calculations for Bare Sure-Flow
- Suction Strainers at Quad Cities 1, 2 and Dresdeu 2, 3 Nuclear Units", dated 5/19/97.
8.4.- Fairbanks Morse Pump Corporation letter, from Ken Hoskins, to David DeGrush of DE&S, Dated October 24,1997.
8.5. Crane Technical Paper No. 410, " Flow of Fluids 'Ihrough Valves, Fittings, and Pipe",
1988.
P g _ W M . W $ " W d D E [ M T "? M ii$ $ Mz,$ M
. TR -vasc eo z 9 Rw.o gg[- ,
Akhmd A a'
Pac,p lofI I.
m .-a Il I,
W8!! i e E r 9
- gr i ,
l @ i -
., 4 e !
- l '
l E l l ___ . :
6 :
l -
i s l
LLt.L
'I III III*
l, ! !!
e s et e
4 at a 18
< i l
. { I-
- 1 d
- i d> li Gi j
'l .S E< 11
. , i a f h 5Ig
- l fs
!h
~
l/ -
V.glr 1
N s e i , s
% :so -
+
c
- ca! b. l ;
a
- "5a
=
8 Ill i E
~
G l l 5
. . m E
j ils www I,
!I k
E Ci .!g' =; E li da .
w hg u I
=
C@.~. 5 .
g l' . EEE1 i
)/ w/ h kl N h k ' ' '
.o ..e
--- -- l, . s g
f _
. {\ . ' . En '
k4 c
a !.
-:%4 -l--M. -g- -
- - , r,i - - - -- - -
d, ,,
- * 'p ,,, _,I, _,
( .ui . .:
- .i .s l
. .o .ri *'
.u i .i
..n .. l g l
g d
3 rig at31 Ay 03ensv3a 3A Tim Iloessev .4 -
'"" ~
- OM ***%8"'t# eg OGue8" Ausf8% E 9Assepsm g emp enge g agpensum ge e ,et g 8't 9 8F'9 9