ML20083R771
| ML20083R771 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 03/30/1984 |
| From: | Ullrich W PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC |
| To: | Murley T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| References | |
| NUDOCS 8404240302 | |
| Download: ML20083R771 (16) | |
Text
.
PHILADELPHIA ELECTRIC COMPANY 23ol MARKET STREET F).O. BOX 8699 PHILADELPHI A. PA.19101 1215)841 4000 March 30, 1984 Docket Nos. 50-277 50-278 Dr. Thomas E. Murley, Administrator Office of Inspection and Enforcement Region I U. S. Nuclear Regulatory Commission 631 Park Avenue King of Prussia, PA 19406
SUBJECT:
Preliminary Report of the Fracture Mechanics Asialyses for Weld Acceptability on the Weld Imperfections Identified by Philadelphia Electric Company in Licensee Event Report 2-83-24/1T.
REFERENCES:
1.
Licensee Event Report Narrative Description to Dr. T. E. Murley, NRC, from R. S. Fleischmann, PECo., dated October 24, 1983.
2.
Licensee Event Report 2-83-24/lT Attachment to Dr. T. E. Murley, NRC, from M. J. Cooney, PECo., dated November 7, 1983.
3.
Letter, dated February 17, 1984, from T. T.
Martin, NRC, to S. L. Daltroff, PECo.,
(Inspection No. 50-277/84-05; 50-278/84-05).
Dear Dr. Murley On October 24, 1983, Philadelphia Electric Company Engineering and Research Department notified the Electric Production Department that certain Class I piping radiographs were improperly read by Eastern Testing and Inspection, Inc.
(ETI) during previous outage modification work.
These discrepancies were identified during an audit of ETI by Philadelphia Electric Company's Construction Division Quality Control Section.
This notification by the Construction Division identified seven welds with radiographs that had been improperly 8404240302 040330
/
PDR ADOCK 05000277 S
PDR f
1(I
o e
Dr. Thomas E. Murley March 30, 1984 Page 2 examined.
The identification and location of these seven welds are as follows:
Wold Peach Bottom ID No. of Welds Unit No.
Location Number 3
2 CRD Scram 655-2-12 Discharge Volume 655-2-39 655-2-40 3
3 Reactor Water 686-3-7 Cleanup System 686-3-8 Between Inner 686-3-11 Isolation Valve and Drywell Penetration 1
2 Reactor Water 686-2-15A Cleanup System Between Inner Isolation Valve and Drywell Penetration Upon identification of these seven radiographic interpretation deficiencies, the ETI radiographs were re-interpreted and unacceptable indications were confirmed by a Philadelphia Electric Company NDE Level III.
The weld deficiencies were characterized as having indications of incomplete fusion, porosity, or crater cracks, or a combination thereof.
Defects were estimated to be 0.010" to 0.060" in depth and 0.5" to 3.0" long.
Philadelphia Electric Company subsequently requested Stone and Webster Engineering Corporation to evaluate the effect of these indications on fitness of the weldments for continued service.
Stone and Webster Engineering Corporation's initial evaluation determined that it would take more than 1000 cycles to propagate each of the seven aforementioned defects to twice their current depth.
These defects, even at twice their current depth, would be acceptable for continued operation.
Therefore, these seven reported radiographic indications were determined as acceptable for continued operation for the life of the plant.
Dr. Thomas E. Murley March 30, 1984 Page 3 Immediately following the discovery of these seven discrepancies, Philadelphia Electric Company reviewed the 131 welds that had been radiographed by ETI in safety-related systems at the Peach Bottom facility.
Philadelphia Electric Company's review identified sixteen additional welds with previously unidentified defects.
The identification and location of these sixteen additional welds are as follows:
Location Number 8
3 CRD Scram 655-3-3 Discharge Volume 655-3-4 655-3-5 655-3-8 655-3-16 655-3-24 655-3-27 655-3-30 5
2 Core Spray 389-2-1 System 389-2-4 389-2-7 389-2-11 389-2-15 1
3 Core Spray 389-3-7 System 2
2 Feedwater System 381-2-8 381-2-14 Following identification of these sixteen additional radiographic interpretation deficiencies, Philadelphia Electric Company requested Stone and Webster Engineering Corporation to perform fracture mechanics analyses for weld acceptability determination for all twenty-three welds with imperfections.
Licensee Event Report Narrative Description dated October 24, 1983, (Dr. T. E. Murley, NRC, from R. S. Fleischmann, PECo.) and Licensee Event Report 2-83-24/lT attachment dated November 7, 1983, (Dr. T. E. Murley, NRC f rom M. J. Cooney, PECo.) notified the Nuclear Regulatory Commission of these findings, and a commitment was made to submit to the NRC a summary report of the fracture mechanics evaluation upon
l Dr. Thomas E. Murley March 30, 1984 Page 4 completion of Stone and Webster Engineering Corporation's analyses.
On January 26, 1984, the NRC requested that PECo provide, as quickly as possible, an update as to the status of the fracture mechanics analyses.
In addition, the NRC requested that Philadelphia Electric Company perform microdensitometer readings to confirm the size and depth of the weld imperfections.
In response to the NRC's request to perform microdensitometer readings, Philadelphia Electric Company and the NRC selected five welds for confirmatory examination.
l Microdensitometer readings were performed on February l
14, 1984.
Unfortunately, the readings were performed with a machine that had not been calibrated.
Philadelphia Electric Company is in the process of re-performing the readings with a calibrated densitometer.
These readings are expected to confirm the findings of February 14, 1984, and are expected to be complete and reported within ten (10) days.
A review of the location of each of the 23 welds with I
imperfections revealed that all the welds were located in the l
following four systems of the Peach Bottom Atomic Power Station:
l I
- Reactor Water Cleanup System
- Core Spray System
- Feedwater System Conference calls were made to the Regional Office of the NRC reporting on the preliminary fracture mechanics evaluation performed on the ' worst case' weld imperfection in each of the four systems.
These findings were reported to the NRC, Region I, on February 1, 2, 10 and 17, 1984 for the Scram Discharge Volume, Reactor Water Clean-Up, Core Spray and Feedwater systems, respectively.
The following is a summary of the description of the l
welds that were determined by analysis to be the ' worst case' imperfections within each of the four systems:
1
Dr. Thomas E. Murley March 30, 1984 Page 5 System Weld No.
Material Scram Discharge 655-2-39 6 inch schedule 80 pipe Volume ASTM /A-106/ Grade B Carbon Steel Reactor Water 686-3-8 6 inch schedule 80 pipe.
Cleanup ASTM /A-312/ Type 316L Stainless Steel Core Spray No particular 10 inch and 12 inch weld was schedule 80 pipe ASTM /
determined A-316/ Type 304
' worst case' Stainless Steel Feedwater No particular 10 inch schedule weld was 100 pipe ASTM /A-106/
determined Grade B Carbon Steel
' worst case' Additional details of the worst case weld, the methods utilized during the evaluation, and the findings of these evaluations are described in the attached ' Radiographic Indications and Fitness of Affected PBAPS, Units 2 and 3, Systems for Continued Service' (Preliminary Evaluation).
The findings of the report are summarized on Table 2,
' Minimum Number of Loading Cycles for Propagation of Postulated Cracks to an Acceptable Limit'.
In accordance with Table 2, the weld defects that can operate safely with the least number of loading cycles (240) until the propagation of the crack would approach the acceptable limit of the analysis are both in the Feedwater system of Peach Bottom Atomic Power Station Unit 2.
Please be advised that the preliminary fracture analyses performed on each of these two welds were based on code allowable stresces, when in fact, the actual stresseu on these welds are considerably less than code allowable.
Stone and Webster Engineering Corporation, the consultant performing the fracture mechanics analyses on each of the twenty-three (23) welds, has been provided with the actual stresses on both Feedwater System welds.
It is expected that the actual stress fracture mechanics analyses on these two welds will reveal that these Feedwater System welds can experience the calculated number
O Dr. Thomas E. Murley March 30, 1984 Page 6 of loading cyclec projected during the design life of Peach Bottom Atomic Power Station and still be acceptable for continued operation.
In addition, as the calculated number of loading cycles for the worst weld defect within the other three systems, scram discharge volume, core spray and reactor water cleanup are in excess of the design life of Peach Bottom Atomic Power Station, Philadelphia Electric Company, based on the attached preliminary evaluation, has concluded that the weld defects within these systems are also acceptable for the design life of the plant.
A final determination will be made by Philadelphia Electric Company upon receipt and review of Stone and Webster Engineering Corporation's final report on the fracture analyses.
Based on conversations with our consultant, Philadelphia Electric Company expects to receive a final report on the fracture analyses of all twenty-three (23) welds by Stone and Webstet Engineering Corporation by May 1, 1984.
Submission of this final summary report to the NRC will take place following our review of that document.
Should you require additional information, please do not hesitate to contact us.
Very truly yours, W. T. Ullrich Superintendent Nuclear Generation Division Attachment cc:
A. R. Blough, Site Inspector NRC Document Control Desk i
)
ATTACHMENT PHILADELPHIA ELECTRIC COMPANY RADIOGRAPHIC INDICATIONS AND FITNESS OF AFFECTED PBAPS, UNITS 2 AND 3, SYSTEMS FOR CONTINUED SERVICE (Preliminary Evaluation) for 1
PEACH BOTTOM ATOMIC FOWER STATION UNITS 2 AND 3 DOCKET NUMBERS 50-277 and 50-278 Submitted To THE UNITED STATES NUCLEAR REGULATORY COMMISSION March 1984 h
I
~
e.
NITACllMF.NT i
RADIOGRAPIIIC INDICATIONS AND FITNESS OF AFFECTED PBAPS, UNITS 2 AND 3, SYSTEMS FOR CONTINUED SERVICE (Proliminary Evaluation)
1.0 INTRODUCTION
A number of radiographic indications in the scram discharge volume, reactor water cleanup, core spray, and feedwater systems were reported by Philadelphia Electric Company (PECo).
A preliminary fitness for service analysis has been performed by Stone & Webster Engineering Corporation (SWEC) based on fatigue crack propagation methodology consistent with ASHE Section XI (Raf. 1) rules.
The results of this preliminary analyuta are briefly summarized holow.
Each of the following four subsections incLudco stress analysis, cyclic crack growth analysis, and acceptance evaluation.
The effect of temperature, environment. and residual stresses is taken into account as appropriate.
Both constant (self-similar growth) and variable (per Ref. 2) flaw depth to length ratios are used.
2.0 SCRAM DISCHARGE VOLUME (SDV) 2.1 Strese Analysis Unit 2 contains three welds with a total of six indications.
Two of the welds are girth butt welds joining 90-dog L.R.
elbows to straight pipa, and the third is a girth butt weld joining straight pipe members.
Unit 3 contains 8 welds with a total of 17 indications.
Five of the welds are girth butt welde joining 90-des L.R.
elbows to straight pipe, two join 45-dog albows to straight pipe, and one is a butt veld joining an 8 in. x 12 in..weldolet to a tank.
The following cyclic stress contributions are evaluated:
1.
Stresses due to the free end displacement caused by thermal expansion (based upon Code, Ref. 3, allowable stresses) 2.
Internal pressure stress 3.
Transient through wall thermal stress due to fluid temperature change (the linear part (aT,)).
These stress contributions are conservatively considered to act concurrently.
The stress components normal to the face of the indication are avsluated.
Thes.e stresses are separated into membrann and bending componentn.
In the case of elbow / pipe gLrth butt welda, the maximum potmissibic moment load is determined from the Code relationship (Ref. 3.
Para. 104.8.3, Eq. 13a) 1 1
i-------- -------- ------ - - - - - - - - - - - - - - - - - -
l ATTACllMENT iM g 8^
Y The maximum membrane plus handing strans to than calculated from the relationship given by Rodabaugh, Iskander, and Moore (Ref. 4) where the maximum stress at the end of an elbow in S, = 1.48 where R, t, and r are the bond radium, nominal thickness, and mean radius of the elbow.
M/Z is the pennissible nominal stress 8 /1.
This vnlue of 5, is separated into membrane and bending as follows:3 S /t and S
~S
-S
/1 S,
=
3 b
m A
The pressure stress (Ref 3,
Para. 102.3.2, d) is taken as entirely membrane and normal to the girth weld.
The AT, contribution is determined through a time-dependent heat transfer analysis.
This stress is compressive at the inner surface and tensile at the outer surface.
The free end displacement stresses'and the pressure stresses are adjueced by the ratio of the nominal thickness over the minimum thickness (tmin) where t,g, = 0.875t, ~A e is the nominal thickness and A is the corrosion allowance.
n The stresses for the worst case (weld No. 655-2-39) are shown in Table 1.
2.2 Cyclic Crack Crowth Analysia The fatigue crack growth analysis addressed the worst case, i.e.,
- 655-2-39 for which a 60-mil-deep indication was reported.
In the calculation, a 0.125-in.-deep crack was postulated to allow for any un-certainties in the indication sizing.
Fatigue crack growth este data of Ne' 1 developed for ferritic materials in high-temperature water.avironment were used in the crack propagation analysia.
Using the above conservatively estimated stresses, the crack depth of the propagating postulated crack was evaluated as a function of a number of the load eyeles for constant and variable crack depth to length ratios.
2
l NITAciml:NT h
t i
2.3 Acceptance Evaluation l
l j
The acceptance criterion of IWB-3612(a) of Ref le i.e.,
[
I 4 la !60 1
was used as suggested in paragraph IWB-3620.
t More K is the maximum stress intensity factor and X is the crack arrestkoughnessgivenbyFigureA4200-1inAppendixAofNf.1.
4 i
J j
Since the indications were reported in the weld metsi, fracture toughness of the weld metal, STA5.1 E7018, was used in the analysis.
j The evaluation has determined that the above criterion is met even when l
the number of loading cycles is greater than 1000 (see Table 2).
J 3.0 CORE SPMAY (CS) i 3.1 Stress Analysis t
j The Unit 2 C8 contains five welds with eight indications.
Four of these l
welds are at pipe / elbow girth butt welds and the fif th le a girth bute weld l
adjacent to the RVF nomale safe and (on the pipa side). The stresses for
[
this latter veld are shown in Table 1.
These strasses are determined based on Code permissible moments and in addition are determined from the loadu l
taken from the PEco stress analysis.
The Unit 3 Cs contains one weld with two indications.
The weld is a I
pipe / elbow girth butt wold.
The indications are such that for the pur-pose of this preliminary evaluation, the Unit 2 result suffices to cover.
i Unit 3.
l The CS system does not experience any operating transients (cyclic r
stress) due to C8 operation.
The CS system esperiences RPV pressure cycling and free and espansion stresses caused by RPV espansion/ con-l traction while the C8 pipe remains assentially at ambient temperature.
3.2 Cyclic Creek Growth Analysis F
The depth of the reported indications varied between 5 and 25 mila.
An initial postulated crack depth of 1/16 in. wae' assumed in this analysis.
l The.al..istic were,erfor.ed for stresses b d o. the stte.e soiyets submitted by PRCo and stresses based on the Code allowable values.
Crack growth rate data f rom Ref. 5 and high-temperature properties of 308 weld stainless steel are used in this analysis.
4 2
4
)
i 3
i l
ATTACllMI:NT l
3.3 Acceptance Evaluation j
The acceptance criterion of IWB-3640 (Ref. 1) which covero nuatenttic piping applica to the core upray pipe welds. According to this critorion.
the allowable end-of-evaluation period flaw depth to thickness ratio in I
i t
l this case is 0.75.
It has been determined that under the most conservativa i
assumptions, i.e. using streason based on Code pemissible moments. the number of cycles satisfying the above criterion exceeds 12,000 (con Tabla 2).
l 4.0 REACTOR WATER CLEANUP (RWCU) f 4.1 Stress Analysis i
THe RWCU system of Unit 2 contains one weld with two indications. The weld is a pipe / penetration sirth butt weld.
L i
The RWCU system of Unit 3 contains three welds with five indicational two l
j of those welds are pipe / elbow girth butt welds and the stresses are determined in the same manner as for the SDV systca.
The third weld is a girth butt weld joining two 45-deg albows.
These elbows are out of plano I
and thus the longitudinal stresses are considered entirely membrann.
The i
maximum stress to determined through the series solution of Rodabaugh and Georga (Raf. 6) and is found to be S
4 2.5 H/Z in the direction normal to the weld.
Tne computed cyclic stresses are alwwn in Table 1 4.2 Cyclic Crack Crowth Analysis The analysis was performed for two worst c'avest 1.
The deepeat postulated crack of 0.125 in. (weld 646-2-15A) and 2.
Thehipeetstremelevel(weld 646-3-4).
The latter wee found to ha a limiting case.
Crack growth rate data used for the RWCU system were the same as for the core spray. Resulta presented below refer to weld 646-3-4.
L 4.3 cceptanca Evaluation Using the acceptanca criterion described tu Section 3.3, it was found l
i that at least 40.000 cycles are required for a postulated crack to grow l
to the nazimum allowable depth of 0.25 in.,
i.e.,
to 75 parcoat of the l
thichname (Table 2).
l i
1 4
i
ATTACllMI:NT 5.0 FEEDWATER STARTUP BYPASS (FW) 5.1 Streme Analysis The Unic 2 bypass contains two volds with three indicatione.
One weld to at a girth butt veld joining n wolded end valve (900 lb) to the largo end of a 10 in. x 8 in. Schedule 100 reducer.
The expansion strasses are based on the maximuta Code permissible moment at the small end of the reducer, adjusted upward by 12 parcent based on PEco stress analyois results to account for the chtage in moment arm. Thu longitudinal presoure stresu contains a bending component caused by the greater thickness of the valva.
The other veld is at a pipe / elbow girth butt weld.
The acrosses are based on the permissible Code moments and are computed in the manner described under the SDV system.
The streuses for both of these welds are shown in Table 1.
5.2 Cyclic Crack Growth Pontuluted cracks with an initial depth of 1/16 in. (reported slepth wan O.010 to 0.015 in.) were assumed to grow due to the cyclic loading into the weld metal, SFA5.1 E7018.
The crack growth rate for ferritic materials in water reactor environunent wan used in this analysis.
5.3 Acceptance Evaluation The acceptance criterion given in Section 2.3 required evaluation of Kg" from Tigure A4200-1 in Appendix A of Ref. 1.
For the weld metal, SFA $.1 E7018, the temperature difference T-RTu,,,, was was fodW( from estimated and th9 crack arrest f ractura toughness.
K,,
Figure A4200-1.
Based on the crack propagation resulto, it was shown that the puutulated cracks will remain with the allowabia limits for at least 240 cyclan based on the assumption of self-similar growth and 550 cycles based on the assumption of variable flaw depth to langth ratio (Table 2).
.=_ _
'l
~.
j f
ATTACllHENT 1
1 6.0 CONCLUSI0018 The preliminary results are summarised in Table 2, together with tho l
estimated number of loading cycles over a 40-year parlod.
Based on these results, the following conclusions can be drawns
^
i 1.
In the scram discharge volume, core spray, and reactor water cleanup systems, the number of loading cycles required to propagate postulated cracks to an acceptable limit considerably l
j exceeds the estimated total number of cyclic events over the plant lifetime.
i i
2.
Based on cyclic stress data presented in Section 5.1, post-
)
utsted cracks in the feedwater startup bypass line era not expected to reach the allowable Itait within at least 1H yeara of plant operation.
A more realistic numensment based on variable crack depth to length ratio shows that postulated eraaks will remain within the allowable limit over the plant 1
lifetime.
I J
1
)
i i
)
)
i i
)
+
1 j
I s
i I
a
- ~_ _
'/
NrrAcilHI:NT l
REFERENCES 1.
Rules for Inservice Inspection of Nuclaar Power Plant Compouants. ASHE i
Boiler and Fressure Yeonel Code, 1983 and Winter 1983 Addenda.
2.
Newman, J. C. and Hsju, I.
S., Engineering,Frneture Mechanics, v. 15 No. 1-2, p. 182,'1991.
Power Pi ing Amerier.n National Standard ANSI 831.1-1973 with Addenda 3.
F to and including Suasser 1975.
4.
Rodshaugh I.' C.
- Iskander,
- 8. K.; and Moore S. E.8 End Effects on l
Elbows Subjected to Homent Loadings,' 9ANL/Sub-2913/7, 1978.
l 5.
Basford, W. W., Journal of Pressure Vessel Technology, v. 101, p. 73, Fehuary 1979
?
6.
R.sdebaugh, E. C. and George,I*. H., Trans. AF.NE, v. 79, 1957.
4
?
t i
7
ATTACEMEhT TABLE 1 AFFLISD STRESSES FDR THE FRELIMINARY FATICLE ANALYSIS Pressure Expansion Thermal Sua of Wald Stress Strees Stress Cvelic Stress System hr G.(ksi)
%(ksi)
((ksi)
Qksi) g(ksi)
%(ksi) s'.(ksi) 6.(ksi) i m
686-2-15A 5.56 0
4.58 0
0 0
10.14 0
EECW 686-3-7 5.56 0
7.80 10.90 0
0 13.36 10.90 BMC5 63-4 5.56 8
9.39 11.87 0
0 14.95 11.87 i
EMCU 686-3-11 5.56 9
10.49 0
0 0
16.03 0
Save 655-2-12 Save 655-2-39 5.91 0
13.76 15.40 0
0 19.61 15.40 Sur*
655-2-40 Set 655-3-3 SDF 655-3-4 Set 655-3-5 SBT 655-3-4 sur 655-3-16 SDF 655-3-24 EDT 655-3-27 SDF 655-3-30 CS 309-2-1 5.26 0
8.19 0
0 0
13.46 0
CS*
309-2-1 5.26 0
19.51 0
0 0
24.77 0
CS 309-2-4 CS 309-2-7 CS 309-2-11 CS 389-2-15 l
CS 309-3-7 Fue 381-2-8 7.71 0
18.62 20.05 0
0 26.33 20.C5 796 381-2-14 7.71 8.67 17.13 0
0 C
24.84 8.67
- P y=== ion stresses are bas,ed ou code allouables l ".,
^
8
s l...
ATTACllMENT l
TABLE 2 l
I MINIMUM NUMBER OF 1.0ADING CYCL.ES FOR PROPAGATION OF POSTULATED k
CRACKS TO AN ACCEPTABLE LIMIT Calc. Number of Cycles Estimated Estimated g
to Acceptable Limit Number of Number of Const a Variable a Cycles Cyclon Ovar Unit System 1
1 per Year 40 Yearo 2 and 3 Scram discharge volume 71.000
>1.000 15 600 2 and 3 Core spray
>12,000 50,000 13 520 2 and 3 Reactor water cleanup
>40.000
>> 40,000 28 c1,100 2
Feedwater startup bypass 240 550 13 520
(~18 yrs)
(~42 yrs) 1 i"
.i I
l A:
I g
9 I
e j
.......,f.
. _j t