ML20080J390

Final Deficiency Rept DER 83-9 Re Low Pressure Safety Injection Pump a Suction Strainer Containing Foreign Matl & Torn Pump B Strainer.Initially Reported on 830217.Trains Flushed.Condition Not Reportable Per 10CFR50.55(e)

ML20080J390
Person / Time
Site:	Palo Verde
Issue date:	09/06/1983
From:	Van Brunt E ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR
To:	Sternberg D NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
References
10CFR-050.55E, 10CFR-50.55E, ANPP-27729-BSK, DER-83-9, NUDOCS 8309260343
Download: ML20080J390 (16)

Text

r

~. ,. , ,

b~d Arizona Public Service Company P O. BOX 21666

• PHOENIX, ARIZONA 85036 September 6, 1983 ANPP-27729-BSK/RQT U. S. Nuclear Regulatory Commission Region V .s Creekside Oaks Office Park 1450 Maria Lane - Suite 210 Walnut Creek, CA 94596-5368 Attention: Mr. D. M. Sternberg, Chief Reactor Projects Branch 1 1

Subject:

Final Report - DER 6(-9 A 50.55(e) Reportable Condition Relating To Unit 1 LPSI Pump "A" Suction Strainer Contains Foreign Material, And Pump "B" Strainer Is Torn File: 83-019-026; D.4.33.2 Re fere nce: (A) Telephone Conversation between T. Young and G.

Duckworth on February 17, 1983 (B) ANPP-23238, dated March 19, 1983 (Interim Report)

(C) ANPP-27333, dated July 18, 1983 (Time Extension)

Dear Sir:

Attached is our final written report of the deficiency referenced above, which has been determined to be Not Reportable under the requirements of 10CFR 50. 55(e) .

Very truly yours \

\

cL- c( . cut E. E. Van Brunt, Jr.

ne p APS Vice President, Nuclear Projects Management ANPP Project Director EEVB/RQT:sn . -

Attachment t' g

-o ^3

. : 3'.

cc: See Attached Page 2 7',

^ At3

'd:

l' }

C 21 8309260343 830906 PDR ADOCK 05000528 S PDR \\

f5'd

F-L., _ ,.

~

- Mr. D. M. Sternberg Der:83-9 -

Page 2:

cc: Richard DeYoung, Director

. 0ffice of Inspection and Enforcement U. S. Nuclear Regulatory Commission i Washington, D. C. 20555 T. G. Woods, Jr.

G. C. Andognini J. A. Roedel

, D. B. Fasnacht A. C. Rogers B. S. Kaplan W. E. Ide J. Vorees J. R. Bynum-

, D. D. Green P. P. Klute A. C. Gehr

' W. J. Stubblefield W. G. JBingham -

R. L. Patterson R. W. Welcher R. M. Grant D. R. Hawkinson L. E. Vorderbrueggen G. A. Fiorelli S. R. Frost J. Self D. Canady Records Center

- Institute of Nuclear Power Operations i 1100 Circle 75 Parkway - Suite 1500 Atlanta, Georgia 30339 t

i T

4

--y .- e,-q g - ,.- , , , - , , - > - w.,-., . r-,,-r+4w--,r= -+t- ----*--, r= r a- y e -r --- -

te- * --- --

Q .,

.- ~

FINAL REPORT ~- DER 83-9

. DEFICIENCY EVALUATION 50.55(e)

ARIZONA PUBLIC' SERVICE COMPANY (APS) c PVNGS UNIT 1 i

I. CONDITION DESCRIPTION -

4 Guidelines for preoperational cleaning of fluid systems are provided in/ANSli45.2.1,-Section 7.- In accordance with these guidelines, strainers were installed on the suction side of the Low Pressure Safety Injection (LPSI) ~ pumps to circumvent pump damage during flushing and preoperational phases of; this . system.

A Startup Quality Assurance Stop Work Notice was initiated af ter initial system operation of the LPSI system. Unusual pressure drop readings across the pump suction strainers led Startup to inspect the strainers. Findings of this inspection indicated that the

. cleanliness level of the system.did- not meet requirements. This

-Deficiency Evaluation Report was.then initiated to provide an

-F engineering . evaluation of the following cleanliness conditions:

1. Purge dam material... weld slag, metal chips, seeds,.etc., were foundnin the Train A pump suction strainer..

2. The train B pump suction strainer was torn,. indicating that-it-may have passed a heavy object.

II. ANALYSIS OF SAFETY IMPLICATIONS The following'are the evaluations of the two noted conditions:

El. Attachment A provides a complete review of purge dam usage / fluid' system cleanliness at this project. This report indicates that:

(1) purge dam material may not be removed during ' flushing if the -

. line is not flushed 'at 4 f ps or greater; (2) the miscellaneous debris found in the strainer. was either trapped during the initial' flush / operations or'was introduced into the system after the flush; and (3) the purge ' dam like material taken from the l LPSI pump suction strainer was tested and fou'nd to dissolve in

- water when the temperature reached 130*F.

2. Attachment B provides the referenced Stop Work Notice disposition,_ documenting that.the Train B pump impeller'showed no evidence of damage. Further investigation indicates that the strainer broke due to workmanship defects in the strainer itself and - not f rom impact by. debris.

i -

{' i 5

r t

1

r Based upon the above discussion, there is no evidence to indicate that, if lef t uncorrected, the LPSI system would fail to perform its safety _ related function during the expected life of the plant. This condition is consequently evaluated as not safety significant and therefore not reportable under the requirements of 10CFR50.55(e).

I III. CORRECTIVE ACTION

1. . Completion of the referenced Stop Work Notice requirements have verified that both LPSI trains have been sufficiently flushed to meet cleanliness requirements.-

2. Attachment A provides recommendations and a plan to preclude recurrence of the subject condition in Units 2 and 3.

l

l.

• o o ATTACHMENT A to DER 83-9

. Bechtel Power Corporation Engineers-Constructors Palo Verde Nuclear Generating Station P.O. Box 49

~

Palo Verde, Arizona 85343

[ B/ANPP-!F-007 March 14, 1983 Arizona Nuclear Power Project

)

P.O. Box 21666 - Fail Station 1742 Phoenix, Arizona 85036 Attention: Mr. G. c. Andognini Vice President, Electric Operations

Subject:

Purge Dam Material File: D.3.03.1

Enclosure:

Report en "Purce Dam Usage At NES" dated February 28, 1983

Dear Mr. Andognini:

The enclosed report on " Purge Dam Usage at FVIES" documents the historical background, the testing ccrpleted, the practices / experiences frcan other nuclear plants and the associated conclusions. As a result of this investi-gation we recomend the following -

1) Inspect low velocity flushed stainless steel lines in Unit 1 starting with the lines flushed at the lowest velocity and working up in line velocity until cleanliness is confimed. -

2) Ccrplete the Units 2 and 3 stainless steel piping installation using Dissolvo h1D 35/60 and have Startup follow a flushing / inspection program based on the Unit 1 experience. ,

On the basis of your March 10, 1983 telephcne concurrence, we are proceeding to inplement the above reccranendations and will keep you apprised of the results.

l i .

~

l Very truly yours, BEClf"EL POWER CCRPORATION l

V J. D. Houchen

! Assistant Project Manager JDH:ba l'

l' cc: J. M. Allen G. E. Pankonin

I p W. G. Bingham W. J. Stubblefield ll*

v '

J. R. Dynum E. E. Van Brunt D. B. Fasnacht W. H. Wilson f -

D. R. Hawkinson J. L. Zerucha J. E. Kirby .

9

O

~

. O February 28, 1983

( '\

PURGE DAM USAGE AT PVICS I) Historical Background Purge dars have been used for the inert gas field welding associated with stainless steel pipe since construction started at PV!GS in 1976 Frcm a search of procurenent records, the project has purchased only "Dissolvo" water soluble purge dam material manufactured by Gilbreth International Corporation. Initially three types of purge dars were purchased for use on the project as follcws:

h1D-35 (light weight cauge paper .003")

(for pipe under 4" I.D. )

b1D-60 (heavy weight gauge paper .006")

(for pipe ext W 4" I.D. )

WID/CPV (Type KID-35 coated with water soluble PVA film where)

(greater gas retention is required )

NPC IE Information 1:otice ::o. 81-07 " Potential Prchlcra With Uater Soluble Purce Dam Materials Used During Inert Gas Welding" was issued on March 16, 1981. This notice addressed teno-Sol purge dam material manufactured by Chris-Craft Industries which had a PVC (polyvinyl alcohol) film. The notice advised that (1) the solubility of the purge dam material in water rapidly approaches zero if the raterial is heated and held at temra-tures in excess of 300*F (2) the purge dam material hardens and becores difficult to break if subjected to terperatures in the range of 300*F to 400*F and becmes brittle if heated to 450*F and (3) purge dam raterial heated above its threshold tcrperature is not soluble in carenly used laboratory solvents. As a result of this ITRC IE Inforraticn f otice, Bochtel discontinued the use of Q,issolvo h1D/CPV which has a polyvinyl alechol film in April, 1981.

'Ihe status of the piping installation in each unit as of April,1981 along with the current status is shown in the following table.

Unit 1 Unit 2 Unit 3 Size Percent Pipe Percent Pipe Percent Pipe Pipe Installed Installed Installed 4/81 12/82 4/81 12/82 4/81 12/82 Large 98 100 73 98 20 , 76 Small 92 100 29 91 2 28 0 '

(1)

O

.. O

~

4

(

Refer to attachments 1 and 2 for graphics.

It is assumed that the percentage of stainless steel pipe installed is directly related to the percentage of total pipe installed. 'Ihe approxi-mate anuunt of large stainless steel pipe per unit is 43,000 LF and small stainless steel pipe per unit is 54,000 LF for a total of 97,000 LF per unit.

In addition to discontinuing the use of Dissolvo WID/CPV, purce dam test-ing was initiated which led to the following revisions in 1.7/P CCI Uo.

101.0 " Welding Control" on May 7, 1982.

1) No wadding of purge dam material will be allowed.

2) Purge dam material shall be placed a mininra of six (6) inches fran the center line of the weld.

3) Water soluble purge dam material will not be used for repair of systems which have been successfully flushed without the ex-pressed permission of Startup.

In January 1983, the LPSI system was refilled and put into cperatien after it had been velocity flushed and initially coerated for startup purposes.

Various pieces of miscellaneous debris (ie weld slag, metal chips, seeds, etc) and what appeared to be purge dam material were subsequently found in the LPSI pu::p suction strainer.

Unit 1 and 2 con +ht spray headers were initially filled with water, soaked and drained several times since velocity flushing was not possible.

A subsequent inspection of the spray headers in January, 1983, indicated that purge dam material was still present.

As a result of these recent findings, the Project needs to outline a course of action to address purge dam re:uval when it is not feasible to velocity flush.

II) Testing Su m -

l The following is a sum of the purge dam testing that has been .acecrn-plished in chronological order.

1) M&OS Report for Palo Verde Project dated January 1982;

Subject:

" Purge Dam Flushing".

Tested Dissolvo materials used at the jcbsite (including ITD/CPV) heated to various terrperatures frcrn 194'F to 500*F using' construction welding procedures in a water flushing test loop.

Test danonstrated that purge dams (including insoluble material) will be flushed frcrn the piping system with anbient terperature water at velocities of 3.6 to 12.5 fps.

(2)

o O

~

2) 1305 Preliminary Report for Palo Verde Project dated February,1983;

Subject:

" Paper-Like Material Fran Flushed Piping".

Tested material taken frcra the LPSI pump suction strainer and Dissolvo WLD/CPV sa ple taken frcun the jobsite warehouse quarantine area.

Test demonstrated that both materials did not dissolve in 70'F water but did disintegrate when water temperature was increased to 130'Fi.

3) M&QS ICM dated February 15, 1983;

Subject:

"Effect of Tenperature on Purge Dam Material Solubility".

Tested solubility of Dissolvo h1D-35, WLD-60, WID/CPV received frcxn jobsite and hTD/CPV received frcan Gilbreth after the pacer was heated to te peratures ranging frcra a-bient to 500*F and soaked in ambient terperature water.

Test demonstrated that hTD-35 and WID-60 dissolved after being heated to 300*F but at 400*F turned dark brown and did not dissolve.

LD/CPV received frczn jcbsite did not dissolve. WID/CPV received frc:n Gilbreth dissolved at tecperatures starting at a.bient. At 500*F all sa ples burned up.

4)  !%QS ICE dated February 18, 1983;

Subject:

" Purge Dam Paper".

Bridgeport Testing Iaboratory tested Dissolvo hb/CPV received frcra jobsite, WID/CPV received frcxn Gilbreth and raterial taken frc:n LPSI pu p suction strainer.

Test indicated that h1D/CPV received frcra jobsite and WID/CPV re-ceived frcun Gilbreth are similar but not the same and material taken frczn the LPSI pump surtion strainer is not the Dissolvo purge dam raterial used at the jobsite.

III. Information from Other Plants -

Utility contacts were made using the APS " Notepad" ccrmunications system.

• Ten plants including Perry, Millstone, Enrico Fermi, Ginna, Peach Bottcn, Limerick, San Onofre, Grand Gulf, McGuire and Perkins fed back information on the use of water soluble purge dars.

'Ihe following is a suntaary of the questions asked and responses -

1) Was water soluble purge dam material used in initial plant construc-tion? If so, please specify manufacturer and type if known'.

8 plants used Dissolvo raterials 2 plants had no rect./d of material used s 10 total (3)

'~

O O

L

2) Has water soluble purge dam material been found in operating systems after initial hydro test, flushing and startup operations? - If so, please indicate extent of problems.

9 plants reported no history of problems 1 plant reported no experience to date 10 total

3) Is water soluble purge dam material being used by the cperating plant raintenance organization?

, 7 plants using Dissolvo materials .

2 plants plan to use Dissolvo materials 1 plant considering the use of Dissolvo materials 10 total .

Refer to attachment 3 for specifics on the utility responses.

i IV) SD2ARY N;D CONCwSIOt:S

'"he' usage cf purge dars has received censiderable attention since NBC 2 Inforration Notice 81-07 was issued in March,1931. Iaboratory tests and the Unit 1 flushing program have verified that Dissolvo WID/CPV (with polyvinyl alcohol film) and Dissolvo WID 35/60 that was overheated by being located too close to a weld will not dissolve under normal flushing conditions but can be cleaned out of the affected lines by flushing at an adeaunte velocity. Se Mt.CS Purge Dam Flushing Report dated January, 1982 demonstrated that purge dams (including insoluble raterial) will be flushed frcan piping systems with ambient ter:perature water at a velocity of 3.6 fps. ~

An example of lines where a velocity flush was not used is the contain:ent spray headers which were filled, soaked and drained. Consequently a consider-able amount of peye dam material was found in these lines. Se containnent j spray headers will be hydrolased and class "B" cleanliness verified.

Be miscellaneous debris (ie weld slag, netal chips, seeds, etc) and what ,

appeared to be purge dam material found in the LPSI pump suction strainer was either trapped during the initial flush / operations or was intmW into the system after the initial flush.

An operating problem could be encountered in safety related systems if a large quantity of purge dam material collected that would interfere with equigent operation. (ie block punp suction strainers, block spray nozzles, etc.) so rertoval of bulk purge dam material is required. We conclusion is that low l velocitv flushed stainless steel lines (uo to 4 fos) (ie containment headers, l containment sump lines, etc) in Unit 1 should be insoected starting with the e

lines flushed at the lowest velocity and workina up in line velocity until I

cleanliness is confirmed.

Refer to attachment 4 which outlines the prcpesed Unit 1 Purge Dam Game Plan.

9 i

, (4) l- - _ _ _ _ _ _

O The Unit 2 flushing / inspection program should be developed based on the Unit 1 experience.

Since most of the Unit 3 pipe has been installed since 4/81, it is expected that considerably less undissolvable purge dam material will be found. However the Unit 3 flushing / inspection program should still be had on the Unit I and 2 experience.

Cancerning the continued use of Dissolvo 35/60 in the remaining piping instal-lation of Units 2 and 3 the alternatives to using water soluble purge da s have the following disadvantages -

1) Rmoveable purge dam - if left in a system has more serious irplica -

tions than dissolvable or undissolvable purge dam material due to the bulk and durable construction of the dam.

21 Valve cperaticn by Construction to effect a dam has potential seat /

operatcr damage implications.

3) Pipe caps can be used where there is an open pipe end available, however there wulld be a significant increase in the quantity /cest of purging gas regaired.

D/en if the above alternatives were utilized, water soluble purge dars utuld still need to be used to make scne welds where the purge gas quantity required .

would be prohibitive.

The survey rade of the operating plant maintenance practices (attachment 3) indicates that they use, or are consideri:x3 the use of, Dissolvo WID 35/60 purge dam material. -

Frcn an cverall Proiect standocint it is recarended that Construction cerlete the Units 2 and 3 stainless steel piping 2nstallat1on usina Dissolvo MD 35/60 anc that Startun follow a flushing /insoection procram based on Unit 1 exper1cnce. ~

JDH:ba (5)

! a

.?.$rf:o/%" p PALO VERDE NUCLEAR GENERATING STATION

/ LARGE PIPE SCHEDOLE UNITS 2 & 3 STATUS DECDBER 1982 g -

INSTALI(D UN!T k0.2 INSTALLE D UNIT 20.3 TOTAL IN!! MD TO DATE TN!! M0 TODATE DESCRIPfl0W flJIDO LF) - ACT $CH ACT SCH ACT $CN i ACT SCM CCNTAINHENT 15.l* 0.0 0.0 14.7 15.1 0.8 0.6 8.8 8.0 NAIN STEAM SS 3.1 0.0 0.0 3.0 3.1 0.2 0.2 f.4 1.5 AUX!L I ARY 38.9 0.0 0.0 38.9 38.9 0.9 1.0 32.2 25.7 CONTROL 5.7' O.0 0.0 5.9 5.7 0.2 0.2 5.7 3.8 TURBINE 66.5 0.0 0.1 65.3 66.5 f.2 2.0 50.8 42.0 rutt 3.98 0.1 0.0 3.9 3.9 0.4 0.1 3.2 0.1 -

R CwASTE 8.1 0.0 0.0 7.7 8.C 0.5 0.5 5.5 2.4

• DIE!!L 2.1 ' O.0 0.0 2.1 2.1 0.0 0.1 2.0 0.3 i '

R LowANCE O SUST0TA 143.4 , 0.1 0.1 141.5 143.3 4.2 4.7 109.6 83.2 . _ . _ . . _

T.W.I. 1 0.N.$. 1.1 ' O.1 0.1 0.7 f.0 0.4 0.0 0.4 c.0

,. TOTA 144.5 l 0.2 l C.2 l 142.2 l 144.3l 4.63 4.7 l110.0 l e3.e .

I 15C .

l

.J* ,.

e***

i l

i . ,

/

soym3 t.F

~

/

~ ~

1.

g -. L j .

7 l i

/

j

- I i

1. i

/

i J.~

1. k$ r s '

d i

/

5 E

4 N /

- l

I /

u  ! /

- i s /

LS(4. k

_.-_ _.. _ 4*l <

lJ ITS ..;

C ,

1II I III III II I IIi m3 t c . ,

A1"T a f*that"WP 1

- . _ . - __m. _ . , _ . , . _ _ , _ , . , . . . _ - . _ _ , - . - , . - _ _ _ . _ _ _ . , . - . . . . . _ . _ , . - _ , _ -. ._ . _ . - -- -

PALO VERDE NUCLEAR GENERATI STATION F

SMALL PIPE SCHEDULE '

UNITS 2 & 3 STATUS DEEMBER 1982

' .\ .t .

INSTALLED UNIT NO.2 INSTALLED UNIT NO.3 THis MD 10 DATE THi$ M0 70 DATE TOTAL D[sCRIFTION (1.000 L Fl ACT SCH ACT $CH ACT SCH ACT SCH CONTAIN$ENT I .5 0.4 0.3 15.8 17.0 0.0 0.6 6.6 7.0 MAIN STIAM SS 3.3 0.0 0.1 2.9 3.3 0.0 0.1 0.0 0.7 AUXIL IMY 42.0 0.8 0.6 39.7 41.5 2.6 1.7 14.2 19.5 CcNTROL 7.8 0.0 0.0 7.5 7.8 0.3 0.4 S.7 3.5 TURBINE 48.1 1. 3 0.6 47.1 46.7 1.2 1.7 10.4 23.6

, FutL 4.1 0.1 0.0 4.1 4.1 0.4 0.0 1.2 0.0 RA0 WASTE 17.4 0.2 0.3 16.2 16.3 0.7 0.9 2.3 2.6 DIESEL 3.4 0.0 0.0 3.4 3.4 0.2 0.1 1.6 0.1 ALLCWAN2 0.6

$USToiA 145.0 2.8 ,1.9 135.7 140.1 6.2 5.5 42.0 57.0 -

T.M.I & 0.N.S. 6.7 0.1 0.4 1.6 1.1 0.8 0.0 1.2 0.0 TOTE j 151.7 2.9 2.3 13!.3 141.2 7.0 5.5 43.7 57.0 183 g

. I i

3g 1 ...e _..=*

~

l l _

p' ,"

( f /

1

./ /

_ l /

l 3

.. acy r >..

1sc s <r  :

' ~ A 9  :

l 8 Tl /

\

= -

/

l I

~

t  :

S 3

_ l l * /

g i__ i . A

_ _ apseu _m7 9 /_ __ _ _.

s _./.' .

t i

- 1 p' .

I i

/

l <

3 ,

_l_ ,_d E Nt - I t- ' A **

i l l l l l l l l l !

s lin i ism l un l ime I let l usz l im3 g im on ATTACILE.r 2

- - - -- ,.m ,,,,-,,,,,,-,--,---.,,m

..  : O O Informaticm from Other Plants -

t'., A) Utility Question 1 Question 2 Has Question 3

• B) Plant _Rerarks Has Purge Dam Purge Dam been Is Purge Dam C) Individual Paporting used in found after Material being construction Hydro / Flush /SU used by Operaticns operating Plant 9 Mainterance

1) A) Central Area Power Initially used No experience Plans to use Has developed Coordination Group Pbn& sol. Now to date. Dissolvo program to B) Perry usina Dissolvo WID-35 ensure that C) Iarry Liechrran WLD-35 tbno-Sol is renoved frtra affected systes.

e l 2) A) !brtheast !b record of !b history of Using Dissolvo Utilities material used. prcblems. WID-60

! B) Millstone

't C) Skip Gilbert 1

3) A) Detroit Edison Used Dissolvo ib history of Plans to use B) *.rico Fermi i:LD-60. prcble s. Dissol"0 C) J. R. bblien WID-60

4) A) Rochester Gas & !b record of tb history of Using Dissolvo Electric material used. problems.

B) Ginna l C) Charlie Anderson /

Terry Schuler -

l 5) A) Philadelphia Used Dissolvo. tb history of Using Dissolvo Electric Co. problems.

B) Peach Bottcra .

Limerick l C) Bob Brown

6) A) SCE Used Dissolvo No history of Considering B) San Onofre WID-60, problems. use of C) tiel Ellis/ Mike Dissolvo Haroin

7) A) Mississippi Power Used Dissolvo Ib history of Using Dissolvo

& Light WID-35 & problems B) Grand Gulf WID-60 C) Bell

8) A) Duke Used Dissolvo No history of Using Dissolvo B) Pt:Guire problens, except Perkins tape residue has C) Holland been found after d Preco testinc.

e l, ATIA01MDTP 3

(-  ::

9 February 23, 1983 UNIT 1 PUR2 DAM GAME: PIAN

1) Assume worst case -

A) ht Dissolvo WID/CPV, or overheated WID 35/60, or sorce unknan material was used as purge dam material in the Unit 1 stainless steel piping installation of "O" systems.

B) ~ ht ' the purge dam material will not dissolve and has to be flushed out.

2) Assume that "Q" syste::s flushed at a velocity of 4 fps plus are '

adequately cleaned - (based on lab test results) . * ~~~

A) Review flushing records to determine which lines were flushed in excess of 4 fps. All syste:s in this category can be accepted as being clean of purge dam material. If any purge dam material is subsequently found in these systers it can be attributed to material being trapped during the flush or :aterial being added to the system after it has been flushed.

1) "Q* Systens or parts of "Q" systcrs that were flushed below a velocity of ' 4 fps would be indentified on a set of P&ID's for evaluation based on the following:

A) Lines upstream of strainers, filters, deuneralizers, heat exchangers or any other devise that will filter out purge dam -

material will be accepted if there is indication available for monitoring differential pressure / flow. ,

. t l B) Lines that will be operated during IFr will be accepted.

l C) Lines that will be included in the safety injection precps between FFT and Fuel Ioad will be accepted.

! D) Lines that do not meet the above criteria (A, B, or C) will be

inspected for evidence of purge dam material starting with the l lines that have the lowest flush velocity with credit being l taken for any previot3 inspections. If purge dam material is found, the inspection and corrective action will be expanded until the project is ensured that bulk purge dam material is reroved frcrn low velocity flushed lines.

Reference P40S Repcrt for Palo Verde Project dated January, 1982.

Subject:

" Purge Dam Flushing".

M:ba ATTACHMENT 4

e - ATTACHMENT B to DER 83-9 p W EM2EEin ph IPUNYLJtD 12ntTIy3 DIE @@53]D%EST CotaPANY CORREsPoNDEWCE PVNGS-JEK-M83-22 o s- January 19, 1983 to Cil Pankonin sie a 625'.

rnom John Kirby - A g

sie a 6 2.'. 9 j ,

Eno i sven ct Response To Stepwork Notice S-83-005 rile s': ,

1. Ins;tet1on to determine the extent of damage was conducted on January 18 19F 3. with the following results:

a) All of the strainer SIB-TO1 was accounted for. De size of the section cf straf ner which was broken loose indicates that no foreign material of greater mass entered the pump impeller. D e lack of deformation of both

- the st rainer-itself and the piece broken loose would indicate the" failure -.

was ne: due to impact.

b) ne levelier of the putrp was inspected (and photographed) in place, with no de w e indicated. A person physically felt the leading and trailing edges of the Impeller and noted no indication of damage.

c) It appears that the scratches were not raised by any material which could l

have gone through the parted strainer. The scratches were too long to be l nssecinted with that type of situation. It is more likely that they were caused during construction when the pump was apart.

I 2. Cleanlinesn of both LPSI trains will be verified by the following actions:

a) Th- suction side of both LPSI pumpa will be verified clean by operating cach pur.p in the long term recirculation mode and periodically inspecting the puttp :uction strainbrs. %e D/P across the strainer will be moniter- -

ed and inspected at a 10 PSID incrasse. Maintain a minimum 10 PSIG NPSH.

b) LPSI pump "B" will be operated with flow through the SDCMX lines. His '

flow path was used to flush the B Containment Spray (CS) piping after initial flush.

c) LPSI purcp "B" will be run in the recirculation mode to the Rb'T. Dis flow path was also used while flushing ",B" CS piping.

di Ar5 vi;; pe r fe r:: an enzoing inspection of Safety Injection piping dt.rinc valec dinasne:chlies that may be required prior to fuel load, ei The re nainine LP51 pipinc trill Be verified clean during LPSI full flow -

t ent en t!.c Reactor vessel with head and internals removed follewing HPT.

r *

"! f-

.? R2cponss to See Notica 5-83-005 continued

- Page 2

3. Potential causes/ opportunities for entry of foreien mateeist into the LFst System were investigated by a survey of all work activities that have taken place since the system flush. STR's. SIV0's SWP's. CHWO's MTS report. '

DCP report, NCR reports and SIM's were reviewed for possible inclusion of purge damn material into the system and none were found. There were several veld items that were less than 1 " pipe size or that particular section of the system still required. flushing. There have been numerous valves dis --

smbled that would have allowed entry of foreign material, however all work has been performed to Grade 'B' cleanliness criteria. There has been an ongoing inspection of piping by the PSE's when the system is breeched and this inspection vill continue.

4. Evaluation of systen cleanliness shall be a continuing process based upon the above tentioned considerations.

• ~ ~ ~

cc: G. C. Andernini C. F. Connell J. D. Houchen -

R. M. Taylor J. 1.. Zerucha l

PSD l

l l

l l

.