ML19332C402

Submits Supplemental Info to 890721 Response to Generic Ltr 89-08 Re Erosion/Corrosion Program at Plant,Per Request. Existing Program Ensures That Erosion/Corrosion Does Not Unacceptably Degrade High Energy Carbon Steel Sys

ML19332C402
Person / Time
Site:	Waterford
Issue date:	11/17/1989
From:	Burski R LOUISIANA POWER & LIGHT CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
GL-89-08, GL-89-8, W3P89-1592, NUDOCS 8911280078
Download: ML19332C402 (8)

Text

_ _.

r,

, ;aa

• e

-i L

{

> a IPL bdisl ANA / sir sAnoNNeSTnter . e.O. sex 60340 POWER & LIGHT NEW ORLEANS, LOUISIANA 70160 * (504)5954 100 UtbkNfr$S 'N  !

W3P89-1592

, A4.05 QA

' November 17, 1989

, f U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555

' Subjects NRC Generic Letter 89-08 -

Erosion and Corrosion Induced Pipe Wall Thinning l Supplemental Information Gentlement-The NRC issued Generic Letter 89-08 under 10 CFR 50.54'f) to obtain affirmations regarding eros!.on and corrosion monitoring programs. Upon reviewing LP&L's original rubmittal (re W3P89-3081, dated July 21, 1989),

members of the NRC Staff requested additional information about the erosion / corrosion program at Waterford 3. This letter identifica recommendations in NUREG 1344, Appendix A and compares them to Waterford's erosion / corrosion program.

The original erosion / corrosion program at Waterford 3 had been developed before the NRC issuad NUREG 1344. During Refuel 2. LP&L performed aumerous inspections and evaluations. Since these inspections LP&L revised the program to incorporate the experience gathered during Refuel 2. When the NRC issued Appendix A of NUREG 1344, LP&L evaluated its p-ogram against the NUMARC guidance. As shown in Attachment A, LP&L felt that some aspects of its program exceed those recommended by NUMARC. In other aspects, LP&L found the methodology in the LP&L program to be equivalent to the NUMARC guidance and chose not to make any changes. Therefore, as illustrated in Attachment A, LP&L did not strictly adopt all of the methods recommended by NLHARC. However, the program in place at Waterford ensures erosion / corrosion does not unacceptably degrade high energy carbon steel systems.

" L 8911780078 891117 PDR ADOCK 05000392 P PNU

g ,

w+ -

_ ,l:lR? : .

)M*ll ..

c 1 % >. ..

t li 1- W3P89-1592- t Page 2 l

. November.:17, 1989:

y Please contact Steven E. Farkas-of my-staff at (504) 464-3383- for any

. [

additional information you require, t

-t x .

(Verytrulyyours, f i

q' ,

, p R.F. Burski-L Manager- . '

i

[:

Nuclear Safety'& Regulatory Affairs j i

a I L _

, -RFB/SEF/pi i

+

l b

Attachment I p- c c =- ' Messrs. . R'

. D. Martin,: NRC Region IV

• F.J. Hebdon,-NRC-NRR, j D.L. Wigginton, NRC-NRR .i E.L.-Blake, t

. W.M. Stevenson  ;

NRC Resident Inspectors Office -[

Ms. J. Higdon  ;

3

.?

f f".

o .

l y

t L

I I

~?

Y '

.r k

P u-E _.

-l

2 Cerrent Implement % ion MethodM 3--

PAGE LINE- . NUREG-1344 App. A R :quirs: q Determine population oi piping products The current Waterford program licts _overL473 piping 2-1 21 listed in Table I which are contained in products. The ' or*ginal list combined two independent ' _ d the portions of piping systems' listed ~- studies,'one by LP&L and a.second'by an engineering Table 2. consultant. The studies sought to find piping products susceptible to erosion / corrosion at Waterford based on industry experience, system operating conditions-and configurations.. All components are in the i secondary plant 3 e.g., main steam, feedwater,' steam generator blow down. No stainless steel products (e.g., primary plant products) are inspected because they are _not -

cusceptible to erosion / corrosion. 'LP&L's method for selecting components is equivalen?. to the NUREG guidance'.

The piping subsystems may be grouped into The erosion / corrosion component list has enough 2-1 23

'information so LP&L.can readily sort ptoir.g'; oducts examination categories that have similar characteristics. by either system, type, or size. This Afo< nation allows Lron to implement the inte't of the FUREG.

2-1 25 Based upon EPR1 methodology or engineering As stated above, plant design and industry experience judgment or analysis, determir.e which form the bases for the Waterford component examination piping products and locations are most schedule.

susceptible to erosion / corrosion by considering the effect of the parameters listed in Table 3.

2-1 28 List these in descending order of LP&L selects components for inspection either at severity (likelihood of erosion / corrosion randon, or if the components meet a remaining life to exist). Unusual operation conditions criteria. Candidates are examined unconditionally (e.g., extended recire line flow or if the program predicts the component's end-of-life others) which are different from normal occurring before three fuel cycles passed. Candidates -

operation conditions shotild be. selected at random may or may not be inspected depending considered in generating the list. on LP&L outage resources. This implementation is equivalent to the'NUREG guidance.

2-1 32 Choose the 10 most susceptible, at a The program leads to selecting more than 10 product /

minimum, of piping product / locations locations. Approximately 200 ccmponents were from the list...This will conservatively selected for examination during the second refueling outage. For the third refueling, the program selected over 180 components for

_ examination. Components _are prioritized for selection' based on previous data and industry experience. LP&L's program greatly exceeds the recommendation stated in.lthe NUREG.

G

+ = ~ v w w + 2m.* w w - e -

+.c +--M

s

-p

~

PAGE LINE 'NUREG--1344' App. A Re'q drmt ' Current Implement = tion Method:' g 2-1 34. In addition,'S piping' product / locations ~ The program directs randomly scmpling a population of' I" should be selected 'at random. those components without an initial baseline inspection. During the third refueling outage, LP&L selected approximately.90 components at random.. This program feature exc'eeds the NUREG guidance.

2-2 8 An EPRI model for selecting the most The engineers who prepared the studies that' selected;

'the original components for the program manually "%h susceptible products / locations considers- W (1) piping material alloy content, (2) screened candidate components based cm the engineers' water chemistry variables of operating _ experiences, industry experience, and criteria' temperature, pH, oxygen concentration published by EPRI and INPO. That technique and the water treatment used, and (3) implements the intent of the'NUREG guidance. LP&L the hydrodynamic variables of pipe does not' employ the EPRI model'for~ selecting inspection diameter, component geometry and flow candidates; however,LLP&L's selection technique rate.

manually implements the equivalent of the EPRI model.

2-3 1 For each piping product found that has LP&L-projects wall thickness at the end of the aext' a wall thickness which is below code cycle by the following steps. Rather than assume minimum requirements or expected within initial wall thickness was.at the nominal dimension, the next refueling cycle or expected LP&L assumed the initial wall. thickness was at nominal operating cycle (+ 10% margin of that plus 10%. The rate of wear comes from the slope between-time) to be below code minimum. that point and the actual dimension measured in an requirements and which is known to be inspection. To . rate, thus established, is used to caused by erosion / corrosion, the project the remaining wall thickness at.the end inspection shall be expanded to additional of the number of effective full-power days susceptible components in the examination (calculated by Waterford reactor engineers) for the category based on engineering judgment. following cycle. For shallow defects, i.e., greater than 75% remaining wall, the relatively steep slope gives LP&L a predicted wall thickness below what the Appendix A method predicts.:The LP&L program specifically describes the criteria used to expand the-examination population upon discovering an unacceptable minimum wall thickness problem.

2-3 7 The additional samples shall be taken The program' implements this guidance.

from similar or like components in the examination category (i.e., sister train or similar arrangement) or components in proximity to the area of concern.

8 e

i, w- .- -.+,e e u.c.. e a . , .e.-,.c ,b. . . . . . -,

i

~

NUREG-1344 App. A R.quir = ' Current Implementrtion Method 2-PAGE LINE 2-3 9 The inspection of additional samples by The program implements this guidance.

this criteria is also required for piping product / component determined to be unacceptable in any additional test lot.

2-4 11 UT or RT may be utilized for the initial LP&L requires that NDE personnel certifications comply round of inspections with personnel with a procedure based on SNT-TC-1A. The program meets qualified in accordance with SNT-TC-1A. the NUREG guidance.

2-4 12 NDE procedures should be reviewed by The program implements this guidance, qualified Level Ill.

2-4 14 Utilize current industry experience The program engineers review industry experience gained in performance of inspections. obtained from " Nuclear Network," and INPO SOERs.

This program is the responsibility of the same group that is responsible for the 10 year IS1 program.

The IS1 exposure keeps them current on state of.the art inspection techniques. The ISI personnel involvement exceeds the requirements stated in NUREG-1344.

Establish benchmark or zero wear point If initial examination data were not available, LP&L 2-4 15 at Tnom, 100% wear at Tcode min. assumes that the initial wall thickness was nominal thickness plus 10%. The minimum thickness allowed comes from code requirements or engineering analysis.. This predicting technique is more conservative than the recommendation; therefore, the program exceeds the guidance in the NUREG.

2-4 18 The area of interest is laid out in a All components >6" in diameter are gridded before grid and thickness readings are recorded examining. When a component fails to meet the for the points where the grid lines acceptance criteria, test personnel refine the grid.

cross...The first step would be to The procedure requires measuring discrete intersections collect data at the grid line to facilitate finite element calculations that might intersection points. The second step become necessary. By using relatively small grid would be to scan the' entire area of spacing, LP&L's program is equivalent to the NUREG interest and record the location and guidance.

thickness for any point that is more than 20% below the average...

S m . w

n.

NUREG-1344 App'. A Rsquiraa: Current Implement' tion Method:-

PAGE LINE 2-4 29 Partial grids--primarily for elbows. Waterf ordicompletely grids all components-inJthe See appendix B. program greater than 6 inches in diameter.' If supports or other interferences exist,.LP&L collects data,from as close as'possible to the grid location. -By. gridding a larger area than required, LP&L exceeds the i recottmendations in the EUREG.

2-4 37 Quick scan--this type of scan may be ' This option is not implemented for components used as'a preliminary inspection or greater than 6 inches in diameter. Large pipes scoping work where a large amount'of that can be accessed internally will be' visually pipe is being evaluated for further examined.from the inside and may be scanned via NDE examination. procedures. The internal exam efficiently-localizes vulnerable surfaces. LLP&L, thus, implements the equivalent of the NUREG guidance.

2-5 12 Automatic scanning. This option is not implemented.

2-5 15 Marking--components should be Implemented by the program.

appropriately marked-(i.e., with high temperature paints, low stress stamps, etc.) for reference for future inspections.

If RT methods are utilized, refer to The RT method is not utilized; therefore, the program 2-5 17 EPRI Document 1570-2 and related does not refer to EPRI 1570-2 and related references.

references.

2-5 20 The NDE results should be used to Implemented by the program.

calculate the'approxinate erosion /

corrosion rate and the number of cycles remaining before the component reaches minimum wall thickness.

I i

. ~ .

, . _ . % fr- - . , . . , . . . , . ,

-~ - -~ , __.

LINE NUREG-1344' App. A Requiraat. ~ Current Implementrtion Methodi Q PAGE 2-5 '22 'If the calculations indicate that an See'the entry for Page 2-3,-Line 1 and~ 4 area will reach code minimum' allowable Page 2-6; Line'5.

wall ^ thickness within 1 refueling cycle:

or expected operating cycle (+ 10% margin of_that time), the questionable component must be repaired or replaced-unless the results of an engineering analysis show that there is an acceptable safety.margir_

for continued operation beyond that point.

' Erosion / Corrosion Rate Calculation The LP&L program uses the Appendix A expression 2-5 27 to calculate the wear rates.

?.-6 2 Each utility should choose its own LP&L implements this NUREG guidance.

method for-engineering evaluation within Code requirements.

The current program includes over 470 components. LF&L 2-6 5 Each utility should perform future inspections based on the results of examined approximately 200 components during Waterford's the initial inspections. Timing of second refueling outage. Of these. LF6L scheduled the inspections would be based on . approximately 90 for re-examination during the third predicted wear rates utilizing initial refueling outage. In the same inspection period inspection data and related acceptance LP&L plans'to examine approximately 90 more guidelines. components for the first time. Upon concluding the third refueling outage, over one-half of the program population will have been. examined.

LPSL keys the need to examine again on the components remaining life (RL, a function of the calculated wear rate). The shorter the RL, of course, the fewer the number,of refueling outages (RFs) that pass before the next inspection. For example, when an RL53 cycles, LP&L re-examines it at-the next RF. This method is equivalent to the NUREG guidance.

4 O

. . ~ . - - , -

.. _ - - - - - ~_ _ - ,

, u~ ..

,g

.lg .

i NUREG-1344' App. A RequirJni:

Implementation Method.. Il1 PAGE LINE.

14 All other units should perform .The program selects a variety of components to inspect 1 during each refueling outage based en (1) previous -

inspections at their most susceptible- resuh.s, (2) industry experience, (3) requests from-

~

locations as a minimum at the next

' refueling outage or within 18 operating . plant personnel, and,(4) random selection. ' Refueling months, whichever is sooner after' outages occur roughly every eighteen months.

release of the EPRI model.

3-1 20 General inspection results for each. LP&L'will add this feature'to the erosion / corrosion nuclear unit will be provided by program.

" Nuclear Network" entries by the utility. ,

Program follow-up will be provided INPO visited Waterford in' September, 1989. At that 3-1 21 time INPO. concluded that LP&L satisfactorily. implemented-through ongoing INFO plant evaluations beginning in . Tune, 1987. an erosion / corrosion inspection program.

1 4

W 4

4 m._ ..

, . , , .. . ~ , _ . , ,,,y~. .

,