ML18005A709
| ML18005A709 | |
| Person / Time | |
|---|---|
| Site: | Harris  |
| Issue date: | 11/18/1988 |
| From: | Loflin L CAROLINA POWER & LIGHT CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NLS-88-247, NUDOCS 8811290523 | |
| Download: ML18005A709 (44) | |
Text
~,"AQCELZRATED DISI IBUTION DEMONSTRATION SYSTEM REGULATORY INFORMATXON DISTRIBUTION SYSTEM (RIDS)
ACCESSION NBR:8811290523 DOC.DATE: 88/11/18 NOTARIZED: NO FACIL:50-400 Shearon Harris Nuclear Power Plant, Unit 1, Carolina AUTH.NAME AUTHOR AFFILIATION LOFLIN,L.I.
Carolina Power
& Light Co.
RECIP.NAME RECIPIENT AFFILIATION Document Control Branch (Document Control Desk)
SUBJECT:
Submits summary of 880923 presentation made to NRC re wall thickness of RCS piping at plant.
DXSTRIBUTION CODE:
A001D COPIES RECEIVED:LTR ENCL SIZE:
TITLE: OR Submittal:
General Distribution NOTES:Application for permit, renewal filed.
DOCKET.5 05000400 05000400)~
8
(, RECIPIENT
)'
- COPIES, ID CODE/NAME '"'TTR ENCL PD2-1 LA 1
.0 BUCKLEY,B 1
1 RECXPIENT
COPIES
'"'D CODE/NAME LTTR ENCL PD2-1 PD 2
2 A
INTERNAL: ACRS NRR/DEST/ADS 7E NRR/DEST/ESB 8D NRR/DEST/RSB 8E NRR/PMAS/ILRB12 OGC/HDS1 RES/DSIR/EIB 6
6 1
1 1
1 1
1 1
1 1
0 1
1 ARM/DAF/LFMB NRR/DEST/CEB 8H NRR/DEST/MTB 9H NRR/DOEA/TSB 11 EG FILE 01 1
0 1
1 1
1 1
1 1
1 EXTERNAL: LPDR NSIC 1
1 1
1 NRC PDR NOZE 'IO ALL "RIDS'i RECXPZERIS PLEASE HELP US TO REDUCE HASTE!
COWZACT 'IHE DOCUMENT CONZROL DESK ROOM Pl-37 (EXT. 20079) m EIZhGMTE YOUR NAME FKM DISTRIK7ZZGN LISTS H)R DO(XIMEMXS YOU DGNiT NEED!
TOTAL NUMBER OF COPIES REQUIRED:
LTTR 25 ENCL 22 I
D 8j
'A)
D D
8
1 0
<
~ II tt ~. 'I!' ll Ap. I ~C p Y" P 1 4$ C 1 Document Control Desk NLS-88-207 / page 2 In conjunction with the root cause assessment, the need for corrective actions has been examined. The procedural error which prevented earlier identification of the below general design minimum wall thickiiess is precluded from happening in current ISI procedures by identification of the acceptance criteria within the ISI procedure. The review and approval process for these procedures includes two technical reviews and two safety reviews which check for the correct acceptance criteria. At the conclusion of the September 23, 1988 meeting, the NRC Staff requested follow-up activities by CPRL in four areas. Each item is listed below along with the closing action. 1. Submit a letter documenting the information discussed in the CPRL presentation.
Response
This letter provides the requested documentation.
2.
Address CPRL's actions regarding corrective actions based on the root cause assessment.
Response
This letter provides the requested information.
3.
Provide the revised Class 1 Stress Report for NRC review.
Response
The revised Class 1 Stress Report has been completed by Westinghouse and is available for NRC review. CPRL willcoordinate any activities related to this report at the NRC's convenience.
Make the detailed stress calculations performed for this issue available for NRC audit.
Response
As subsequently discussed with the NRC Staff, the subject calculations are available at the Westinghouse offices in Monroeville, Pennsylvania for inspection and audit by the NRC. CPRL willcoordinate any activities related to these calculations at the NRC's convenience.
On the basis of the information provided in this letter, CPRL considers this issue closed.
Except for consideration of corrective actions as outlined above, no further action is required. If you have any questions regarding this information, please contact Mr. 3. D. Kloosterman at (919) 836-8055.
Yours very truly, L.. Lof in LIL/JDK/lah (50753DK)
Attachments Manager Nuclear Licensing Section CC:
Mr. W. H. Bradford Mr. B. C. Buckley Mr. M. L. Ernst
Attachment l to NLS-88-207 List of Attendees Shearon Harris Meeting September 23, 1988 Name Bart Buckley Leonard Loflin Al Watson Scotty Hinnant Lee Williams Max Thompson 3im Kloosterman Ed 3ohnson Dick Tome Mark Gray Evertt Rodabaugh L. B. Marsh Mark Hartzman Alan Herdt Robert Hermann Dick Becker D. M. Verrelli 3im Coley P. T. Kuo G. C. Lainas Or anization NRC/NRR/P D2-I CPRL Licensing Harris Project V.P.
Harris Plant General Manager Nuclear Engineering Nuclear Staff Support CPRL Licensing Westinghouse - PSD Westinghouse - PCE Westinghouse - PSD Consultant for NRC NRC/DEST/MEB NRC/DEST/MEB NRC/Region II NRC/DEST/MTEB NRC/DLPQ/PEB NRC/Region II NRC/Region II NRC/NRR/EMEB NRC/A/D RII
ATTACHMENT2 TO MLS-33-207 WALLMEASUREMENT PROGRAM DESCRIPTION AND RESULTS
SUMMARY
WALLMEASUREMENT PROGRAM DESCRIPTION o
CONTROLLED BY PLANT PROGRAM PROCEDURE PLP-606 COORDINATED BY NUCLEAR ENGINEERING DEPARTMENT.
o EXAMINEDALL REACTOR LOOP PIPING PRIMARY WELDS USING FOLLOWING STANDARD ULTRASONIC INSTRUMENTATION:
USK-7 USK-75 USL-37 SN:
2727-0180 SN:
31059-881 SN:
211650 o
UTILIZEDINDEPENDENT CONTRACTOR TO TAKE.MEASUREMENTS BRW/MQS.
o DATARECORDED AND TRANSMITTEDTO WESTINGHOUSE FOR FINAL APPROVAL.
o WESTINGHOUSE REVISED CERTIFIED STRESS REPORT.
g Eh.ego< vfssEL IX-SN sfEAM 4Eg.
g,.C. I'9MB
~
~
l2EAcqc g VEssfl-IX -SN 5-RC-2.- F'e-2 5-RC I-YA'-I 5-K-2.- Py-I 5-gc-2-Fe-4 g-gc-2-5-yW-I 5'-P~- Z-I=e-S 5-Rc Py-7 g-gc-2-Iw'-5 5-Rc I=W-8 5-AC 4-A/-C R.Each<
COOLa)f 0'i O'I g~
FIELD f sHoF WELv LocATlogs
RESULTS
SUMMARY
o OF THE 36 PRIMARY LOOP WELDS, IO WERE FOUND WITH LOCALIZEDAREAS BELOW GENERAL DESIGN MINIMUMWALL CONFINED TO SMALLAREA OF WIDTH ALONG PIPE (0.5" TO l.5")
MOST BELOW WALLAREAS <io ALONG PIPE CIRCUMFERENCE INFORMATIONTAKENON PIPE WALLBASE METALVERIFIED BOUNDING OF ISSUE TO WELD AREAS
SHEARON HARRIS NUCLEAR POWER PLANT REACTOR COOLANT PIPING WALL THICKNESS FIELD ACTUALS LOOP 1
WELD GENERAL DESIGN MIN.
PER NB 3600 LOCALIZED DESIGN MIN.
PER NB 3200 LOWEST NOTED WALL THICKNESS DIFFERENCE FW 1
2.27 2.04 2.48
+ 0.440 I
FW - 2 2.18 2.36
+0.180 FW
- 3 2.42 2.18 2.35
+ 0,170 FW
- 4 2.15 2.03 2.21
%0.1 80 FW
- 5 2.15
- 2. 03 2.88
~0.850 FW 6
2.42 2.18 2.36
+0,180 FW 7
2.42 2.18 2.42
+ 0.240 FW
- 8 2.42 2.18 3.1 0
+ 0.920 1
- VW -1 2.27 2.04 2.34
+ 0.300 5
- VW -
1 2.1 5 2.03 2.22
%0.1 90 3
- VW - 5 2.42 2.1 8 2.48
+0.300 4
- VW -
6 2.42 2.18 2.52'0,340 NOTE: ALL DIMENSIONS ARE IN INCHES
SHEARON HARRIS NUCLEAR POWER PLANT REACTOR COOLANT PIPING WALL THICKNESS FIELD ACTUALS LOOP 2
GENERAL DESIGN MIN.
PER NB 3600 LOCALIZED DESIGN MIN.
PER NB 3200 LOWEST NOTED WALL THICKNESS DIFFER EN CE FW 1
- 2. 27 2.04 2.44
+ 0.400 FW
- 2 2.42 2.18 2.36
+0.180 FW
- 3 2.42 2.18 2.36
+0.180 FW
- 4 2.15 2.03 2.06
+ 0,030 FW 5
2.15 2.03
+0.61 0 FW 6
2.42 2.18 2.37
+ 0.1 90 FW 7
2.42 2.18 2.40
+ 0.220 FW
- 8 2.42 2.18 2.56
+ 0.380 1
- VW -1 2.27 2.04 2.24
+ 0.200 5
- VW -
1 2.1 5 2.03 2.1 6
+ 0.130 3
- VW - 5 2.42 2.18 2.48
+0,300 4
- VW -
6 2.42 2.18
- 2. 48' 0.300 NOTE: ALL DIMENSIONS ARE IN INCHES
SHEAR ON HARRIS NUCLEAR POWER PLANT REACTOR COOLANT PIPING WALL THICKNESS I. IELD ACTUALS LOOP WELD GENERAL DESIGN MIN.
PER NB 3600 LOCALIZED DESIGN MIN.
PER NB 3200 LOWEST NOTED WALL THICKNESS D IFFER EN C E FW 1
2.27 2.04 2.36
+ 0.320 FW
- 2 2.42 2.1 8 2.44
+ 0.260 FW
- 3 2.42 2.18 2.34
+ 0.1 60 I
- 4 2.15 2.03 2.21
+0.180 FW
- 5 2.15
- 2. 03 2.90
+0.870 FW 6
2.42 2.18 2.31
+0.1 30 FW 7
2.42 2.18 2.40
+ 0.220 FW
- 8 2.42 2.18 2.52
+ 0.340 1
- VW -1 2.27 2.23
+ 0.1 90 5
- VW -
1 2.15 2.03 2.22
+0.180 3
- VW - 5 2.42 2.37
+0,190 4
- VW -
6 2.42 2.18
- 2. 44.
+ 0.260 NOTE: ALL DIMENSIONS ARE IN INCHES
ATTACHMENT3 TO NLS-88-207
SUMMARY
OF METHODOLOGY FOR EVALUATION
SEGPT-ERJ-118 PIPE STRESS CRITERIA FOR PIPE MALL THICKNESS DEVIATIONS PREPARED BY:
CHECKED BY:
SE&PT-ERJ-118
SUBJECT:
PIPE STRESS CRITERIA FOR'PIPE WALL THICKNESS DEVIATIONS PURPOSE:
This document provides pipe stress criteria for pipe wall thickness deviations in ASME Class 1 piping.
The deviations considered are local regions in which the pipe thickness is below the minimum thickness required in Section NB 3640.
These deviations, which may be found by ultrasonic NDE testing of the piping system, may be caused by; a) excessive grinding of welds on inside or outside diameter, or b) excessive radial weld shrinkage, or c) shop fabrication errors in the piping components.
Deviations in the, body of elbows are not considered.
APPROACH:
The ASME Code permits use of the more rigorous analysis of Section NB 3200 in place of the simplified analysis of NB 3600.
Z.
Criteria for Prima Membrane Stress:
The. following criteria may be used in place of NB 3641.1 based on NB 3217 and NB 3213.10.
Case A:
When regions of "thin wall" satisfy the following requirements:
a) total length of "thinned" region along pipe axis LT < 1.0 Rt b) distance along pipe axis between adJacent regions LA > 2.5 Rt c)
R
=
average pipe radius d) t
=
minimum wall thickness in thinned region e) region may be completely around the circumference.
f)
SZ
> 1.1 S, where SIL and S 'are defined in Item 2 below.
m'or Shearon Harris Loop Piping:
Pipe Hot leg X>>legl Cold-leg R(in. )
15.74
- 16. 81
- 14. 93 1.0 Rt 3.97 4.10 t 3.86 2.5 Rt 9.92 t
10.25 9.66
SE&PT-ERJ-118
~ Include weld from pipe to SG inlet.
2.
- Then, SIL 1.5 S
Local primary membrane stress intensity due to:
internal design pressure, axial force from deadweight, and axial force from other design mechanical loads as defined in piping D-Spec.
Code allowable stress. at design temperature.
For Shearon Harris, Design Mechanical Loads are:
Deadweight Case B:
1.
- When, SIL 1.1 S
in the thinned region, where SIL is defined as in Case R, Item 2 above.
2.
- Then, regions of thinned wall need not satisfy criteria in Item 1
of Case A above.
In this case, the extent of the thinned region should be limited by good engineering Judgment (e.2.
L> < 4.5 ~Rt, A.e.,
LA + 2 L> from Case A).
In Case A or B, the allowable stress, S
, may be increased if a reduction in design temperature can be gustified.
m'n Case A or B, the calculated stress intensity, SI may be decreased if a reduction in design pressure can be gustified.
Calculation of local membrane stress due to internal ressure:
S Radial
= - 0.5 PD Axial
=
(
I
)
PD Do DI 2-2' Hoop
=
I D
D P
2t PD
Design pressure DI
=
Maximum inside diameter t
=
Minimum wall thickness in thinned region D
=
DI + 2t 0
SE&PT-ERJ-118 For Shearon Harris Loop Piping (based on weld undercut dimensions, Ref. 1):
Pipe Hot leg X leg
+
Cold leg D
(in. )
29.21 31.21 27.71
~ Includes weld from pipe to SG INLET PD
=
2,485 psi S
=
17 550 psi TD
=
650'F for TP 304N SS Calculation of local membrane stress due to deadwei ht:
SX 0.7854 (D 2 DI2 FX
=
Axial force (+ tension)
D (zv) 0.0982 (D 4 - D 4) 0.7854
- D 2) o I
o I
V
=
Shear force MX
=
Torsional moment Calculation of SI SIL
=
S - S S
=
1/2 [S
+
+ S ] +
Hoop 3
Radial II. Criteria for total primary stress, secondary membrane plus bending stress and peak stress.
The following criteria may be used to satisfy the Code requirements for Level A, B, C and D service in sections NB 3653,
- 3654, 3655 and 3656.
SEGPT-ERJ-118 A.
Pressure Stress Ratio Replacing o (nom) with Do
+ I D
2 D 2 2t nom Do -DI New to old stress is, fp fp
=
2t nom D (nom) 0 D
'2 DI2 Do -D 2
I D 'I from Part I above 0
'For Shearon Harris Loop (excluding body of elbows)
Pipe (in.)
Do(
) f
+ Do - I 2
D 2 Do
+ DI 2
2 2tnom Do (nom)
Hot leg X leg
~
Cold leg 2.45 2.60 2 32 33 9
36.2 0.1445 0.1436 0.1444 Includes weld from pipe to SG INLET.
B.
Moment Stress Ratio Replacing Do nom 2(.0491)>(Do(
) - DI(
)
)
with Do 2(.0491)
(Do - DI )
4
, ratio of new to old stress is, fm D
- D o(nom)
I(nom) fm
"-(
Do(nom)
D D
DI 0
SE&PT-ERJ-118 For Shearon Harris Loop:
Pipe Hot leg X leg
~
Cold leg o
I D0 18094.
21926.
15406.
D 4 o (nom)
I (nom)
)~fm =
Do(nom)
Includes weld from pipe to SG inlet.
The pressure stress ratio, fp, and moment stress ratio, fm, is applied to the left, hand side of Code Equations 9,
10, 11, 12 and 13 to obtain the calculated stresses in the thinned region.
Revised values of Equations 10 and 14 must be used in the calculation of the cumulative usage factor.
C.
Stress Indices The Code stress indices for flush butt welds are applicable provided that:
a) the extent of the thinned region satisfied item I, Case A,
- above, b) the radial weld shrinkage satisfies Item IIIbelow, c) visual inspection indicates that pipe surface at weld is sufficiently flat to permit inservice ultrasonic examination.
III.'riteria for Radial Meld Shrinkage Radial weld shrinkage is not anticipated in the thick wall (> 1.0 inch) primary loop piping.
The criteria below assure that the discontinuity produced by radial weld shrinkage does not require any increase in the stress indices of NB3680.
The decrease in pipe outside radius is limited to 0.25 times the nominal wall thickness, t The minimum pipe circumferenceg C
f is therefore given by the Pollowing values based on an initial diameter nam'qual to the nominal pipe diameter, Do:
C 3.1416 (Do.. - 0.5 + t ~
))
P (nom)
'nom)
SE&PT-ERJ-118 For Shearon Harris Loop Piping:
Pipe Outside Circumference Nominal Minimum Design Design.
C (in.)
8nimm Allowed Hot Leg 106.50 106.22 102.65 X-Leg+
113.73 113.44 109.64 Cold Leg 100. 97 100.75
- 97. 33
~ Include weld from pipe to SG INLET.
SE&PT-ERJ-118 REFERENCES 1.
"Calculated Minimum Wall Thickness for Straight Pipe Under Internal Stress - CQL", File CQL-140/7, 9/14/78.
2.
ASME III, 1977 Edition through Winter 1979 Addenda.
3.
"Strength of Materials - Part II", S. Timoshezko, D. Van Nostrand Co.,
3rd Edition,
- 1956,
- p. 205.
4.
"Piping Design Specification, ANS Safety Class 1", 955239, Revision 3, 2/19/86, S. 0. CQL-137.
5.
Reactor Coolant Pipe Shop Fabrication - E-Spec
- G678843, Revision 1,
3/6/72, S. 0.
CQL-140.
6.
Reactor Coolant Cast Fittings - E-Spec
- G678865, Revision 2, 6/16/72, S. 0.
140.
7.
Reactor Coolant Cast Fittings - CQL - E-Spec
- 679144, Revision 4, 9/3/74, S. 0.
CQL<<140.
8.
Reactor Coolant Piping Shop Fabrication - CQL - E-Spec
- 679187, Revision 1, 2/19/75, S. 0.
CQL-140.
9.
"Table of Pipe and Elbow Properties",
S. Tate, 3/12/79, CQL-145/14A.
10.
"Stress Indices for Girth Welded Joints, Including Radial Weld Shrinkage, Mismatch and Tapered-Wall Transitions",
E.
Rodabaugh, S.
- Moore, NUREG/CR 0371, 9/78-11.
Reactor Coolant Seamless Forged Pipe - E-Spec.
G-678866, Rev. 2, 10/24/73, S. 0.
140.
SHEARON HARRIS NUCLEAR POWER PLANT REACTOR COOLANT PIPING WALL THICKNESS MINIMUM WALL THICKNESS CRITERIA AND CHECKLIST CODE PARA-GRAPH NB 3213.1 0 LENGTH FOR STRESS
)1.1Sm DIST.
NEXT DISCONT NB 5640 GENERAL MINIMUM WALL THICKNESS NB 3213.10 MIN.
WALL TO MEET 1.5Sm MIN.
WALL TO MEE 1.1Sm 1 973 SUMMER ADDENDUM NB S650 MIN.
WALL MIN.
WALL TO MEET TO MEE FQTN 9 EQTN 1 2 MIN.
WALL TO MEET EQTN 1 3 NOMINAL WALL THICK.
MIN.
WALL FOR USAGE FACTOR 1 979 WINTER ADDENDUM NB 3650 PIPE O.D.
(AVG.)
MIN.
RADIUS CIRC.
o.sW~ 2.5WRt tnom 0.8 t Cp COLO LEG LIMITS 1.93M c
(%2. 75) 9.66%c
(~1 3. 75) 2.15 2.21 1.42 1.93 2.03 NOT GOVERN.
ING Nor GOVERN
~
ING 2.32 1.86 1 4.93 97.33 HOT LEG LIMITS 1.985 %t
( 2.84) 9.92M c
(~14.18) 2.27 2.33 1.49 2.04 NOT Nor GOVERN. GOVERN.
ING ING 1.99 2.45 1.96 1 5.74 1 02.65 X-OVER LEG LIMITS 2.05M t
(~3.03}
1 0.25K
(~1 5.1 4) 2.42 2.48 1.60 2.18 NOT NOT GOVERN-GOVERN-
.ING ING 2.03 2.60 2.08 1 6.81 1 09.64 NOTE:
ALL DIMENSIONS ARE IN INCHES
ATTACHMENT0 TO NLS-88-207 ROOT CAUSE ANALYSIS AND
SUMMARY
ROOT CAUSE ASSESSMENT OF AREAS WITH LESS THAN DESIGN MINIMUMWALL THICKNESS IN REACTOR COOLANT SYSTEM PIPING WELDS SEPTEMBER 28, 1988 MEM/NRCRoot/1/OS4
Table of Contents Section 1.0 PURPOSE AND SCOPE 2.0 PERSONNEL INVOLVED 3 '
SUMMARY
OF CAUSE 4.0 INFORMATION COLLECTION 4.1 Data Collection 4.2 Personnel Interviews 4.3 Physical Measurements 5.0 EVALUATION PHASE 5.1 Physical Root Cause 5.2 Programmatic Root Cause 5.3 Bounding the Minimum Well Issue in Main Steam and Feedwater Piping 6+0 CONCLUSIONS 6.1 Physical Root Cause 6.2 Programmatic Root Cause 6.3 Process Control 6.4 Problem Scope
~Pa e
4 4
4 5
5 7
9 9
9 10 11 12 12 14 15 15 MEM/NRCRoot/2/OS4
e A.
TABLES
~Aecdicee l.
2 ~
3 ~
4 ~
Physical Problem Analysis Programmatic Problem Analysis Personnel Interviewed Piping Weld Joint Data (Design vs Actual) Loops 1, 2, and-3 MEM/NRCRoot/3/OS4
1.0 PURPOSE AND SCOPE Localized areas of less than general design wall thickness were discovered in large piping and fitting weld joints of the Reactor Coolant System.
The purpose of this study was to establish the primary root cause (physical and programmatic) for the observed wall thickness and to evaluate the scope of the issue in respect to other plant piping systems.
2 '
PERSONNEL INVOLVED A task team of personnel experienced in plant construction and problem analysis techniques was assembled.
Personnel selected were independent of the current operations, site engineering, and maintenance organizations.
This team met to consider,
- evaluate, and study this issue on a daily basis from August 29,
- 1988, through September 19, 1988.
At least 75 man days of effort were expended by the team members during this period.
Members Or anization C. H. Griffin G. T. Lew D. A. Morrison C. L. McKenzie M. F. Thompson H. L. Williams HE6E Center, Metallurgy SHNPP, Special Projects
- SHNPP, ONS
- SHNPP, Ops.
QA/QC NSSS NED 3'
SUMMARY
OF CAUSE The physical cause of the minimum wall issue includes a number of primary contributing factors'he factors include:
Physical dimensions of piping and fittings upon receipt of the material.
')
Physical fabr:"ation procedures such as. installation, joint fit-up, welding, and Preservice Inspection (PSI) surface preparation resulted in the removal of sufficient" material to reduce wall thickness below minimum values in isolated areas.
3)
A program based on the use of judgement to check for wall thickness when doubt existed was in effect.
A program to test for PSI suitability failed to detect below minimum wall thickness because an incorrect thickness was employed for comparison purposes in the data forms utilized by PSI personnel on special pipe.
The problem is confined to large diameter special pipe which was prepared for PSI ultrasonic inspection.
This potentially includes selected Reactor Coolant System, Main Steam
- System, and Feedwater System piping.
MEM/NRCRoot/4/OS4
3.0
SUMMARY
OF CAUSE (continued)
The bounding of the minimum wall issue will be discussed in Section 6.4 of this report.
Refer to Tables 1 and 2 for a detailed problem analysis outline.
4' INFORMATION COLLECTION 4.1 Data Collection Various plant records were reviewed in order to bound and research the problem.
These record types included:
a.
Field Change Requests
[FCR/PW] (includes permanent waivers and design changes) b.
Non-Conformance Reports (NCR) c.
Deficiency and Disposition Reports (DDR) d.
Correspondence Files e.
Piping Installation Packages f.
Piping Spool Piece Data Packages g.
PSI Data h.
Fabrication Drawings i.
Process Control Instructions and Procedures j ~
Preservice (PSI) and Inservice (ISI) Inspection Procedures k
Installation and Fabrication Speci:fi'cations
\\
A review of records was conducted in order to ensure that documents pertinent to the minimum wall issue were not missed.
The fact that a given document file did not contain pertinent information was useful in the evaluation.
The following paragraphs outline the results of the. data search and only the important results are stated.
The jurisdiction of work control was an-important consideration and certain key dates were crucial.
The Reactor Coolant System (RCS) was transferred to the work control jurisdiction of the Start-Up Unit on February 22,
- 1985, by Release for Test Boundary (RFT 2005.001)
~
Prior to this point in time work was controlled by construction procedures; after this time work control procedures were defined by the Start-Up Manual.
Most of the RCS piping work pertaining to PSI preparation was accomplished prior to system turnover and was subject to Construction work control procedures.
MEM/NRCRoot/5/OS4
4.1 Data Collection (continued) 4.1.A.
Some 633 FCR/PW documents pertaining to the Min. Wall Evaluations'(as well as 90 DCNs) were xeviewed.
4.1.B.
99 NCRs relating to piping minimum wall violations were located.
Only a few of these related to the RCS and PSI issue All pertinent DDR documents were filed in the. RC loop installation and data packages.
General correspondence files were searched for documents of interest.
There were a number of documents providing background information but no references that pertain to specific RCS or PSI wall thickness issues 4.1.E.
Piping Installation Packages for the RCS loops were reviewed in detail.
These packages contain specific records and as-built data for each joint.
The work performed by construction on each RCS piping joint is documented in these packages.
References to DDRs,
- NCRs, WDRs, WTNWs etc. are included.
4.1.F.
Piping Spool Piece Data Package.
This document is similar in nature to item E above but contains data documenting the work pexformed by the vendor (Southwest Fabricatoxs and Cameron Iron) of the material.
4.1.G.
PSI Data consists of the data observed and recoxded during the PSI/ISI program.
4.1.H Fabrication Drawings contain the specific as-designed and as-built data.
EMDRAC Drawing 1364-2680 series is of interest to this issue.
Process Control Instructions and Procedures include those documents controlling work on the RCS.
This includes the Stax't-Up Manual and Construction Welding and Fabrication procedures.
4.1.J.
Inservice Inspection Procedures control the ISI/PSI work process.
They include ISI-110, ISI-201, and ISI-501.
4'.K.
The various specifications of interest include Westinghouse Shop Order 140 documents for the Reactor Coolant System.
MEM/NRCRoot/6/OS4
4.2 Personnel Interviews This phase of the investigation was based on interviews of key personnel.
The intent of the interviews was to examine activities that were relevant to the minimum wall 'condition.
While the main focus was on the RCS, the investigation also examined the processes and activities relative to other piping systems for the purpose of bounding the problem.
Personnel were selected for interviews based on their involvement with the welding process (from fit-up to final QC signoff), the PSI program and the work control procedures that controlled these activities.
Each of these areas were discussed relative to the potential physical and programmatic root causes that contributed to the minimum wall problem.
A summary of some of the key interviews are included below listed by functional title.
These summaries provide the pertinent facts that were used to develop the root cause determination.
Complete documentation of each of the interviews can be found in the backup documentation files.
In addition, a list of interviews listed by name and functional title is provided as Table 3.
The interviews that are not summarized in this report do not provide anyadditional information of significance but they do collaborate the facts already stated.
4.2.A.
CP&L PSI/ISI Su ervisin Pro ect En ineer Discussions with the PSI/ISI Supervisor indicated that he had no recollection of any specific actions similar to the weld preparation being performed on the Reactor Coolant Loop.
He stated that all activities were performed with paper work and that any activity that would remove metal would have required a Work Traveler Non-Welding (WTNW) ~
The only activity that was performed without paperwork was light buffing to remove surface contaminants
~
There was no document allowing this buffing on Class 1 piping.
Craft support for PSI work was provided by Welding Engineering.
The nominal and minimum wall values were taken from the M-30 Specification or the Line Lists 4.2.B.
NES Site Su ervisor and Pro ect En ineer for PSI The NES Site Supervisor reported to the CPBL PSI/ISI Supervisor and acted as the coordinator between CPKL and NES.
The responsibilities of his position included Non Destructive Examinations including ultrasonic thickness measurements.
He indicated that minimum wall spot checks were made during the PSI walkdowns to ISI-110.
They also looked for weld crown and surface configuration problems.
If any problems were identified they were worked by construction.
He said that he thought some light buffing was performed on the RCS.
The most probable cause of the minimum wall condition was thought to be confusion as to what minimum wall for special pipe was Values of nominal and minimum MEM/NRCRoot/7/OS4
4.2.B.
NES Site Su ervisor and Pro ect-En ineer for PSI continued) wall were recorded by clerks (contractors) who reported to the PSI/ISI Specialist.
Discussions were held concerning minimum wall but he did not remember the source of the values on the ISI-110 data sheets.
The procedure required minimum wall checks, and if the data sheets were checked "Sat." it meant the thickness was checked.
4.2.C..
CPSL Level III NDE A/
C Su ervisor Discussions with the CPKL NDE QA/QC Supervisor indicated that WTNWs were utilized on the RCS and that they should be filed with the spool piece packages.
He remembers that welds were thicker than what was stated by Welding Engineers because exposure times for radiography would always have to be greater than what calculations indicated.
His recollection was that we had an extra quarter inch of weld metal.
An NCR was written by the ISI group saying the NDE group had accepted some welds that were not acceptable by the Code.
The NCR was closed when NDE proved to the ISI group that the indications were nonrelevant.
4.2.D.
General Foreman PSI/ISI Pre aration Grou The PSI/ISI General Foreman had four crews working for approximately a year on PSI preparation.
They utilized Weld Data Reports (WDRs) and WTNWs for work control.
Crews were trained, supervised and closely watched by NES.
Numerous Penetrant Test indications were worked on.
The groups performed their own information PTs and made repairs where necessary.
Final visual and PT was performed by QC.
D-Meter checks were utilized when minimum wall problems were suspected.
Minimum wall was a major consideration and evaluated immediately.
4.2.E.
Pro ect A En ineer The Project.
QA Engineer believes a lot of wall thickness checks (VT) were performed but it appears they were for information only, since documentation can not be found.
He believes grinding, including PSI cleaning of welds, contributed to the cause and used up our. margin.
The NES profile sheets did not identify the areas below design minimum wall because they used the wrong minimum wall criteria (12 1/2Z).
Documents used to control work were WTNWs and NCRs.
Also, he remembers a letter in the files that allowed light buffing with a flapper wheel (this letter did not apply to Category 1 piping).
MEM/NRCRoot/8/OS4
During the fit-up process, dimensional checks were made with calipers on the weld area end preps with measurements of the land area (depth of the counterbore) of primary interest.
Measurements were checked against the Westinghouse fabrication drawings.
No problems were detected.
There was no real concern with minimum wall since, per the Welding Engineer, there was acceptable minimum wall with the counterbore and the counterbore was to be -filled in.
QC would not have used the line list for minimum wall values.
They would go to the drawing, the Welding Engineer, or Westinghouse with the final call being the Welding Engineer.
The final visual of welds utilized a six inch scale to check for suspected low areas.
4.3 Ph sical Measurements Some surplus RCS piping remains in storage.
The surplus includes fittings from Units 2, 3, and 4 as well as piping from Units 2 and 3.
Specific "as built" documents were not available for this material.
Therefore, specific dimensions could not be field verified.
Dimensions taken and observations made supported the general Unit 1 observations.
For example, a vendor weld was measured at.020" over design minimum wall thickness in a localized area.
Extensive measurements of RC loop piping joints were made by an independent contractor.
The detailed results of these measurements are found in the documentation files for this study.
A summary of the results is compared to existing joint data in Table 4.
5 '
EVALUATION PHASE The task force worked as a team to evaluate and determine the root causes of the problem.
The Kepne'r-Tregoe problem analysis technique was used along with brainstorming and data correlation to establish the root causes.
Information obtained from data reviews and interviews was used in the various steps of the evaluation.
The evaluation was performed'n two steps.
The first step was to determine the physical root cause of having piping less than the original design minimum wall thickness.
The second step was to determine the programmatic root cause for not identifying the existence of these areas during the fabrication process.
The root causes are explained belo~.'.1 Ph sical Root Cause The fact that there are a number of small localized areas in and adjacent to some of the welds in the Reactor Coolant loop that have wall thicknesses less than the design minimum is attributed to the following two items.
MEM/NRCRoot/9/OS4
5.1 Ph sical Root Cause (continued)
The pipe and fittings for the Reactor Coolant loop are large diameter and heavy walled and, because they are constructed of stainless steel, their cost is considerable.
Therefore, when designing. this Loop it is reasonable that efforts were made to not overdesign the system.
Accordingly, since stainless steel was used and the chemistry of the water being contained is controlled, the amount of wall thickness specified above the design minimum was comparatively small (0.060" for design minimum of 2.15", 2,27" and 2.42").
The design agent did,not require an allowance for erosion or corrosion on stainless steel piping in the RCS.
In a few cases fabricated spool pieces were accepted with areas of wall thickness closer to the design minimum than specified.
The most extreme case being at vendor weld six of loop 3 where, in one area, the wall thickness was measured to be 0.005" above the design minimum.
Therefore, when the RCS loop pipe arrived on site, it had areas of wall thickness that were very close to the general design limits.
2.
An inherent part of the installation of piping systems is grinding on the surfaces of the two pieces being joined by the welding process.
Grinding is performed to obtain proper fit-up, to remove indications discovered during welding and to prepare the completed surfaces for nondestructive testing.
In addition, the completed surfaces (vendor welds and field welds) for which ASME XI applies must be prepared for PSI/ISI.
This consists of light grinding, sanding, and buffing.
Each of these operations tend to remove material from the weld and its adjacent areas.
- Hence, the root cause for wall'hicknesses less than the design minimum, is that. some areas of the RCS loop pipe were very close to design minimum wall when received on site and the inherent grinding of the fabrication process took some of areas below the design minimum wall.
5.2 Pro rammatic Root Cause The team evaluated the program in effect at the time the RCS piping was received, installed and tested to determine what program elements should have caught and identified the subject localized areas in and adjacent to some of the welds with Less than design minimum wall thickness.
ASME,Section III does not require minimum wall checks as a "hold point."
The Code does require that. care be taken to avoid violating minimum wall MEM/NRCRoot/10/OS4 5.2 Pro rammatic Root Cause (continued) specifications.
Consistent with the Code, the program required that care be exercised in conducting grinding activities so as not to violate minimum wall thickness.
The fact that the actual wall thickness in some areas was close-to the general design Limits was not realized and special. wall thickness measurements were not deemed necessary.
The program, developed relied on inspection personnel and others to call for a minimum wall verification check if there was cause to suspect that minimum wall thickness was of concern (such as when grinding was performed). The program also required an inspection of wall thickness, of welds and adjacent base material when the PSI walkdown was performed to determine acceptability for performance of PSI.
Upon review of inspection sheet information which was entered prior to these walkdowns, it was observed that incorrect minimum wall thickness specifications were listed.
When used by the PSI inspectors, the incorrect thickness lead them not to identify areas that were of concern with respect to minimum design wall thickness.
The incorrect values were apparently based on using a formula for standard SA376 schedule pipe where 12-L/2X of the nominal pipe wall thickness is subtracted from the nominal pipe wall thickness in order to obtain the fabrication minimum wall thickness.
- However, the RCS piping was special pipe where the use of the fabrication tolerance formula did not apply since the design documents specified the min. wall thickness.
The programmatic root cause was that a specific hold point was not established for QC measurement of wall thickness and that the incorrect minimum wall values (schedule pipe versus special pipe) were listed on the PSI walkdown inspection sheets.
These incorrect values led the PSI inspectors to accept wall thickness measurements that were less than the specified design. minimum thickness.
Problem analysis worksheets-'re shown in Tables 1 and 2 and Table 4 displays data used for correlation.
A review of PSI inspection sheets was performed on other special piping to determine if other problems exist outside the Reactor Coolant system.
Similar minimum wall thickness determination errors were noted in several of the PSI inspection sheets for other systems using special piping.
The observed wall thickness values were compared to new. design minimum values determined by NED and no cases were found to suggest that a wall thickness problem exists in any other special 'piping not discussed in this report.
5.3 Boundin the Minimum Wall Issue in Main Steam and Feedwater
~Pi in In an effort to bound the minimum wall condition, PSI records were reviewed for approximately 70 Mainsteam (MS) and 61 Feedwater (FW) (Code Class
- 2) weld joints.
With the exception of a few of these weld joints, MEM/NRCRoot/LL/OS4 5.3 Boundin the Minimum Wall Issue<<in Main Steam and Feedwater Pi in continued) the PSI walkdown weld inspection sheets reflected the correct minimum specified wall thicknesses for the piping located inside the containment building.
The utilization of the correct minimum wall thickness values for comparison resulted in several minimum wall thickness violations being observed and documented by PSI personnel.
These violations were later evaluated for acceptance/rejection by NED.
For-those few cases where the incorrect minimum wall thickness value was used, the documented PSI weld profile dimensions did not indicate that the'oints exhibited any minimum wall vi,olations.
Stress analysis was performed on these joints by NED.to ensure that wall thicknesses at'east as low as the value utilized by PSI would be acceptable.
For those MS and FW piping welds (Code Class
- 2) located outside the containment within the break exclusion zone the PSI personnel had, in most cases, used (incorrectly) the minimum wall thickness values from the line list as nominal values'n many cases, a
fabrication tolerance value of 12 1/2X was subtracted from the nominal values to derive minimum thickness criteria.
Although the observed PSI profiles for these joints did not document any missed areas of minimum wall thickness violations, NED stress analysts vere requested to evelusne this piping to determine if the ~desi n
minimum wall thickness for these pipe welds was greater than the value derived by subtracting a
12 1/2X'fabrication tolerance from the minimum value specified in the CPRL line list.
The calculations performed by NED indicated that the design minimum wall thicknesses on these lines was such that usage of the 12 1/2X fabrication tolerance off the line list values would not result in the use of values less than the design minimum wall thickness for the pipe.
This effort excluded the MS elbows which were thoroughly reevaluated for minimum wall issues in late 1983 (see DDR 84-500).
It is, therefore, concluded that for these lines the use of the 12 1/2X fabrication tolerance would not have caused
- the, PSI personnel to miss documenting a design minimum wall thickness violation..
6.0 CONCLUSION
S The results of this investigation revealed that. there are some weld joints in the Reactor Coolant Loop Piping with localized areas that exhibit thicknesses less than that required by the original, design specifications for the pipe.
6.1 Ph sical Root Cause The root cause of this condition is to be attributable to several key factors relating to the design, fabrication, acceptance, and PSI of the system, as detailed below.'
MEM/NRCRoot/12/OS4,-
6.1 Ph sical Root Cause (continued)
Ao Stainless steel pipe is not normally purchased as standard size and schedules for the diameters and thicknesses required for this piping.
Therefore, the Reactor Coolant Loop was purchased as Special Pipe (i.e., Non-Schedule) by Westinghouse from Cameron Iron.
Southwest Fabricators performed the required shop machining and fabrication.
Westinghouse specification requirements provided for minimal surplus material thickness in the counterbore weld prep regions.
The difference between the minimum fabrication thickness, specified by Westinghouse to Southwest Fabricators, and the design minimum thickness was 0.060 inch (approximately 1/16 inch).
This margin was, in some
- cases, reduced based on normal fabrication practices in which areas of the pipe were manufactured with less thickness than other areas.
In. no case was a wall thickness less the than design minimum accepted by Westinghouse.
The supplying of this pipe. with little surplus thickness is due in part to the high cost of stainless steel material.
B ~
During site installation, the material thicknesses specified in the SHNPP Piping Line List and on site process control sheets (i.e.,
Weld Data Reports) for the Reactor Coolant Loop piping were the specified minimum wall thicknesses.
Plant engineering and craft personnel were not aware of the tight tolerances to which this pipe had been fabricated and were therefore not aware of the fact that the counterbored ends exhibited thicknesses which in many cases were near the design minimum wall values for the pipe.
C ~
Prior to completion of welding, the root I.D. surface of each weld joint was blended with the surrounding metal by grinding and polishing.
Upon completion of the welding for each joint, the weld crown over the pipe 0.
D. was also ground flush with the surrounding base metal to better facilitate later PSI/ISI by ultrasonic (UT) techniques.
There is no definite evidence to indicate that, after final QC and. ANI acceptance of the completed weld joints, craft personnel, acting under the request of PSI Technical Support personnel, performed additional polishing of any of the weld joints.
In summary, it appears that the initial condition of the pipe was such that very little extra thickness was available to compensate for metal removal during the fabrication/installation of the piping.
This thickness was used up during fit-up, weld-out, NDE surface preparation and PSI surface preparation of the weld joints The result is some localized areas exhibiting wall thickness below design minimums.
MEM/NRCRoot/13/OS4 6.2 Pro rammatic Root Cause The apparent reasons that this condition was-not detected during the construction and/or PSI baseline inspections are as follows:
A.
As required by Section III of the ASME Code, plant procedures called. for care to be exercised in grinding in order to avoid the unnecessary removal of material.
Site procedures did not require routine thickness checks of the pipe as an inspection hold point after installation.
Thickness measurements were requested during installation only if a wall thickness question existed.
Because plant engineering personnel were not-aware of the tight tolerances to which the pipe was manufactured, no special program was implemented for this system to provide, for closer monitoring of the wall thickness during installation and PSI surface preparation.
B ~
During the PSI baseline examinations for this RCS pipe, Technical Support ISI personnel had applied incorrect design minimum wall thickness criteria on the data sheets required by procedure ISI-110.
Apparently, ISI personnel subtracted 12-1/2X off the nominal fabrication tolerance dimensions for the ASME SA376 pipe to derive the minimum wall thickness'here were instances during the baseline inspection where the PSI UT inspectors measured and recorded thickness readings which were actually less than the values required by design.
However, these values were not recognized as wall thickness problems since the incorrect minimum wall thickness value was being used for. comparison and evaluation.
If the correct minimum wall thickness values had been utilized for the evaluation of the baseline UT reports, it is evident that. these minimum wall th'ickness problems would have been detected.
This is demonstrated by a number of wall thickness discrepancies
- detected, reported, and evaluated on'ther piping.
A review of site Nonconformance Reports and Field Change. Requests revealed numerous instances where PSI baseline inspections had documented potential minimum wall problems and processed them for. engineering evaluation.
It is concluded that no data'r information available to plant engineering, QA/QC, or Technical Support ISI personnel previously identified these minimum wall thickness problems on the RCS pipe.
The questions, which are now known to exist, were not documented or submitted for evaluation.
MEM/NRCRoot/14/OS4-Cl
6.3 Process Control The results of the investigation indicate that the plant did have process control of grinding activities during the construction phase.
Plant personnel (i.e.,
QA/QC, engineering, and craft) were apparently also aware of the potential for minimum wall questions on pipe..
These conclusions are based on a review of requirements stated in plant procedures and. in the plant's ASME QA Program Manual which were in eff'ect during construction.
They are also supported by the numerous Nonconformance Reports and Field Change Requests documenting the evaluation of several hundred potential minimum wall thickness questions detected during the construction and PSI time frame..
The results of the investigation indicate that this condition is limited to the Reactor Coolant Loop piping.
This is the only stainless steel "special" ordered pipe at the plant which also required that the weld crowns be ground flush to facilitate PSI/ISI ultrasonic examinations.
Considerations that bound the issue include:
A. 'iping welds which were prepared for PSI/ISI ultrasonic examination.
Other welds have weld crowns which are not ground flush with the pipe surface; these welds would not approach the design minimum wall thickness.
B.
Special wall (non-schedule) pipe.
This is a bound since there was a greater margin of material between pipe wall nominal values and the design minimum wall thickness for schedule piping.
Specifically, for rolled plate (seam welded) piping, the fabrication minimum wall thickness is derived by subtracting
.010 inch from the specified nominal thickness.
For seamless
- pipe, the-fabrication mini'mum wall thickness is derived by subtracting 124X off the nominal thickness'he design minimum wall thickness should be even less than this calculated fabrication minimum value.
However, for the special pipe at the HNP, the design'inimum wall thickness is only approximately 7X less than the nominal value.
A sampling of PSI inspection sheets showed that errors were not being made in determining the minimum wall thickness requirement for schedule piping.
C.
Stainless Steel Class I RCS Loop Pipe.
This is a bound since the design margins for the MS and FW systems are greater.
A review of the PSI inspection. sheets on other than stainless steel special pipe (MS and FW systems, Class
- 2) showed that, where the inspection sheet had an error in the minimum wall thickness, the value used met calculated minimum wall thickness criteria.
MEM/NRCRoot/1S/OS4.
PttTSICAL CAUSE TABLE I PROBLEM ANALYSISWORKSHEET State Deviation Areas of RC Loop Piping have design minimum wall thickness probibms.
Specily the Problem
, What
'dentity.
Thickness probluz)s:)~ !.,
Stainless Steel..:. '.L ~
!'Weld areas Class I system hsttgrgaction Ill.r Butt Melde Par Vt)!!)'ai'<W!lrrr~ "-
Special (Non-Standard)"'pipe IS NOT Standard pipe Cracks/Linear indications
Socket Molds,)r;)r>g(q<!~$~r,'~ulgp yt<
~":-,)~vj~>> );evlr.r )t.'; I.') )!)irwb)rp qrar)irv;:.
Distinctions ol IS compared with IS NOT
~
~ t)'r',
'v '!r~: )'rgg v3fi~
C"at'fyvt':::..'4"'M$53Fjk5MJRMP Qw ~ )r) lpf4y j')ztt rra))gv)Qllvlo Changes in distinctions (list dates) ruel~irl; aa
< v)r rrl)
~
~ Xr)
+ ~
~ '
~ ~,')
~ ~
Where Iocation Weld zone areas Large piping)16" Containment Building ISI piping Circumfrential defects Over counter bore Pipe Elbows Longitudinal Small pipe When timing Pirst discovered during PSI After veld crowns)s vere ground e
f Discovered during installation.
Discovered receipt inspection.
Discovered vendor inspections.
Discovered veld preparation..
\\t
~
r ~
'! )
r ~I
'" all'! C3)P',:tr r
') W r.
il'li i
'! ~ ~
i Is3r(tjlff! 4) 'retd P!' tt f~l" r.)r L.'.d'.r!.:.f'4)T'> 'I >r.)r )
~
Extent magnrfude Deviation vorst case 2.42 to 2'1 (. 11 inch) 14 out of 36 vclds Through pipe
) 12tsI of nominal vali thickness
)t
~
~
Develop Possible Causes from experience. changes, distinctions Test for Probable Cause against speciiications (list assumptions from destructive test)
Original Hatcrial Specified vrong Fabriciated wrong Inadequate Installation Fit-Up trelding Grinding Post Installation
- PSI Grinding vt ~
~
does not explalm )
Mhy not identified before PSI Why not identified before PSI Acceptable inspections also 14 of 36 welds Acceptable inspections explains only if:
marg n
Dcvigtions Deviations margin accepted not giving accepted'ot 'giv
~
~
I Determine Most Probable Cause Well thickness margin was small (less than 12tsX) at the beginning and was used in construction and PSI surface preparation.
Verify True Cause steps
- 1) Comparison of Fab Drawing measurements to current measurements
~
~
~
~ re
~
2)
Ph sical Ins ection of excess i in E /BL25 002 PROGRAMMATIC CAUSE TABLE II PROBLEM ANALYSISWORKSHEET V
State Deviation Progress did not identify existing design minissuss wall thickness-problcns.
Specify the Problem What identity '.
Min. vali probleiss...'
'i. i.'.
RCS piping Meld areas Special pipe (Non-',standard),t';>$ ->
Butt welds
~ ~ s ~ us'isahvrseit!ix xi"t" Iltrs '4 o IS NOT Crack/t.inear indication
~ a<<
'-;<>I:4,;,ct"'4'"jssi's"',4c,'4oRN)5..":
~ s!~<<e! s~'
<!crier tis!~!WHJ~~/re r!r~
Distinctions of IS compared with IS NOT a'Lpfj)
< "I:r>'
~1'."1'"
I II J!!!!!:A I.
I/!yt'ajtttyr!r. j F;"jt": Xfl.'.(,"'QpA$
'!s'i!ar ia is!i.i'i+svatsr'\\,
Changes m dlstrnctrons (ttst dates)
F~P"
~r <<~Itrif
~
'!i
i s Ks4
's!!s! Iii'0 ',>
Where location tteld xone areas Large piping ) 16" Containsent Building ISI piping Circunfrential Defects Pipe Flbovs Longitudinal Snail pipe When timing First discovered during PSI After weld crowns were ground Discovered Discovered Discovered Discovered cion.
during installacion receipt inspection vendor inspection before veld prepara-
+':f:!4'.
.ri)I 1. ': t rs-.'4 j.'Ia Mat
~': '6: ii"jf x
Extent magnilude Deviation worse case 2.42 to 2.31 Through pipe
(.11 inch) 14 out of 36 welds
) 12/2 of Nossinal wall thickness Develop Possible Causes from experience, CI>anges, distinctions No specific code requiressent to lseasure nin. vali v/o due cause to suspect.
Inspectors using )udgeaenc co deterssine if rsin. wall requires checking PSI used vrong criteria for cain. wall (12tsZ of Nossinal) during walkdovns..
Ignorance of tsccual wall 'thickness.
No progran to test for wall thickness.
Test for Probable Cause against spea%cations (list assumptions from deslructive test) dOeaneteXplain:
PSI walkdown PSI walkdovn inspection PSI walkdovn inspection
\\
PSI walkdovn inspection explains only if:
~
I Inadequate visual inspcccion by QC.
Misinterpretation of requireuent for Min. vali (12.52 in lieu of design specified Grinding progran did not account for lsspact of cumulative rcparis.
PSI walkdovn inspection i ~
~ ~
~
s pplies to PSI'valkdovn '
Determine Most Probable Cause PSI used wrong criteria for nin. vali (12taI of Noainal) during vslkdovn..
Verify True Cause steps Sanple check PSI sheets for stand" pipe.
Check other systen with special pipe PSI inspection check.
~
" -"!-".ys
'~! E /BL25 002
Table 3
Personnel Interviewed TOM= BROMBACK SCOTT LARSON CP&L Pxoject Engineer over PSI/ISI Group NES Site Supervisor
& Project Engineer for
- PSI, ED BETZ SAM WHITLOCK BILL DOWNS DAVID SHOCKLEY STEVE FRESHWATER MIKE PUGH EMORY UPCHURCH ROLAND PARSONS RAY HANFORD RANDY SELLERS WARD MERCER STAN PRUITT 4
'Z CP&L Level III NDE QA/QC Supervisor DANIELS General Foreman over PSI Prep.
Group NES Contractor CP&L QA/QC Technician CP&L Project Mechanical
& Piping Engineer CP&L Project Q.A. Engineer CP&L Welding Engineer CP&L Project Manager CP&L Resident Engineer Metallurgy/Welding DANIELS Piping Foreman CP&L Q.
C. Inspector CP&L PSI/ISI Specialist MEM/NRCRoot/18/OS4 Table 4
Loop 1
Loop 2
Loop 3
Fab Xla De alga Mfa Fab Letaal PSI Lctaal Labatlt Feb Ila De alga Xla lab Letaal PSI Letaal Labailt lab Ila Deeiga ila lab hctaal PSI hetaal Lsballt F1-i R.SS 2.27 Z.S80 2.48 2.9$
2.27 2.975 2.44 2
~ 99 2.27 2.S86 2.3d F1-2
- 2. 48
- 2. 42 2.98 2.48 2.48 2.63 2.4 M 2.Sd 2.48 2.42 2.54 2.6 M 8.44 FY-3 2.48 2.42 2.59
.$ 6M 2.96 8,48 2.42 2.55 R.Sd 2.48 2.48 R.49 e.e w 2.95 ~
F1-4 2.21 2.16 e
e 2.25 8.81 2.21 2.16 2,25 2.0d 2.81 8.15 2.25 2.2i F1-5 2.21 2.15 N/h 8.0 M
2.88 2.81 2.16 N/A 2.9 M 8.84 2.21 2.15 N/L 2.04E 8.00 P1-d 2.48 2.42 2.600 R.S5 R.Sd 2.48 2.42 8.52S 8
4 8.$ 7 8.48 2.48 8.4S2 8.5 E 2.91 P1-7 k b 2.48 2.42 2.512 2.4 M
8.42 8.48 2.42 2.50d 2.48 8.48 2.401 e
2. 5 E
8. 4,0 o
FV-8 2.48 2.48 S. I
$.10 2.48 8.48 S.i E 2.68 8.48 8.42 9;12E 2.62 i-V1-i R.ss 2.27 R.Sd4 2.46 2.$ 4 R.SS 8.27 2.885 8.95 M 2.24 2.SS 8.27 2.800 8.$ 7 8.29 3-71-5 a
. e 8.48 2.42 4-71-d
- 2. 48 2.42 5-V1-1 2.21 2.15 8.170 2.15 M 2.616 2.4 M
2.549 2.55 M 2.48 2.62 2.22 8.48 2.42 2.688 8
~ 55M 8 48 2.48 8.42 8.517 2.65 I 2.48 2.42 2.48 2
48 2
1
- 2. id 8.21 2.16 2 Ri 2.16 Z.S8M
.10$
2.6O6 2.68 E 2.97 e
b 2.486 8. 8 E
2.200 2.$
M 2.22 Measured E
Estimated from seal'e on profile Below'esign Min Fall Below Fab Mln Wall