ML20030C174
| ML20030C174 | |
| Person / Time | |
|---|---|
| Site: | Grand Gulf  |
| Issue date: | 02/20/1981 |
| From: | Mcgaughy J MISSISSIPPI POWER & LIGHT CO., TENNESSEE VALLEY AUTHORITY |
| To: | James O'Reilly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| References | |
| REF-PT21-81 10CFR-050.55E, 10CFR-50.55E, AECM-81-80, NUDOCS 8108250463 | |
| Download: ML20030C174 (8) | |
Text
oipic.IAL Copy.
))
MISSISSIPPI POWER & LIGHT COMPANY Helping Build Mississippi P. O.
B O X 16 4 0. J AC K S O N, MIS SIS SIP PI 3 9 2 05
'#Ysf,DTN*AUd'#
19- 055- 03/> v February.,I 1
in 79-doY-d 3 4 -
O,.
h h '[ k 4
q Q~
ys Office of Inspection & Enforcement g
I)
U. S. Nuclear Regulatory Commission Region II
!?
'3 N
"~
D th jf Z
101 Marietta Street, N.W.
3 Suite 3100 3
cj g6
- ~
Z
~
.k)0, ] g T/g Atlanta, Georgia 30303
\\g Attention:
Mr. J. P. O'Reilly, Director C/
N
- s
Dear Mr. O'Reilly:
SUBJECT:
Grand Gulf Nuclear Station Units 1 and 2 Docket Nos. 50-416/417 File 0260/15525/15526 PRD-79/04, Final Report, Inadequate Crimping in PGCC Cable Pin Connectors AECM-81/80
References:
- 1) AECM-79/26, 3/27/79
- 2) AECM-79/54, 5/29/79
- 3) AECM-80/126, 11/19/79
- 4) AECM-80/18, 1/31/80
- 5) AECM-80/239, 9/30/80 On February 23, 1979, Mississippi Power & Light Company noti-fled Mr. W. B. Kenna of your office of a Potentially Reportable Deficiency (PRD) at the Grand Gulf Nuclear Station (GGNS) construc-tion site. The deficiency concerns inadequate crimping on certain Power Generation Control Complex (PGCC) interconnecting cable pin l
connections supplied by the General Electric Company. Our final l
report is attached.
l l
A draft response was discussed with your Mr. Ton. Conlon on l
February 5 and 10, 1981, at which time it was agreed to delay this response until February 20, 1981.
Yours truly,
[p / J. P. McGaughy, Jr.
EWC:mt Attachment cc:
See page 2 l
h825046381022O 1-l I
S ADOCK 05000416 PDR Member Middle South Utilities System
%5 m
Mis"I'/%IPPI POWER O LI2HT COMPANY Mr. J. P. O'Reilly AECM-81/80 NRC Page 2 cc:
Mr. N. L. Stampley Mr. R. B. McGehee Mr. T. B. Conner Mr. Victor Stello, Director Division of Inspection & Enforcement U. S. Nuclear Regulatory Commission Washington, D.C. 20555 Mr. G. B. Taylor South Miss. Electric Power Association P. O. Box 1589 Hattiesburg, MS 39401 i
l
[
~
Attachm:nt to AECM-81/80 Page 1 of 5 FINAL REPORT FOR PRD-79/04 I.
Description of the Deficiency Certain Power Generation Control Complex (PGCC) interconnecting cables supplied by the General Electric Company have inadequate crimps on the connector pins. This deficiency was noted in January 1979 from inspections made on similar equipment for the Susque-hanna Nuclear Station. Pennsylvania Power & Light Company notified USNRC Region I of the deficiency on January 9, 1979.
Subsequent investigations conducted by Bechtel Power Corporation revealed that the potential for similar conditions existed at the Grand Gulf Nuclear Station (GGNS). MP&L then reported this condition to USNRC Region II as a Potentially Reportable Defi-ciency (PRD) on February 23, 1979.
Also, MP&L subsequently expanded the scope of PRD-79/04 to in-clude the potential problems concerning damaged wire insulation, pin anomalies, suspect lug crimps on Termination Modules (T-Mods),
and panel connectors.
The deficiency was applicable to delivered PGCC equipment for Grand Gulf Units 1 and 2.
The T-Mod lug crimping problem was also applicable to delivered equipment for Unit 2 and for T-Mods manufactured for Unit 2 which were subsequently installed in Unit 1.
II.
Analysis of Safety Implications Mississippi Power & Light considers this deficiency reportable f
under 10CFR50.55(e) in that extensive tests, inspection and repairs were required to substantiate the adequacy of the con-nectors, connector pins and cables.
It should be ackncwledged that the General Electric Company disagrees with our evaluation with.egard to 10CFR50.55(e). However, both companies concur j
that this deviation does not constitute a substantial safety hazard as defined in 10CFR Part 21.
l I
l l
Attachment to AECM-81/80 Page 2 of 5 The successful completion of qualification testing of typical PGCC cables /connecto'hs, inc]uding low force crimps, to the full requirements of IEEE 323-1974 and 344-1975 confirms the continuing functional integrity of crimped connections under worst case conditions of Main Control Room environments.
III.
Corrective Actions Taken MP&L has implemented the following program through GE and the constructor for insuring that the connectors for the intercon-necting cables in the PGCC have adequately crimped pin connet-tions:
A.
A complete review was made of the Manufacturing and Quality Assurance processes, procedures, tooling and records asso-ciated with pin crimping at General Electric factory in San Jose and any deficiencies were corrected. In-process de-structive testing on pin crimp samples was instituted as a Quality Assurance procedure.
B.
For those cables manufactured prior to the establishment the new controls the following is applicable.
1.
General Electrie implemented an extensive inspection program for. Grand Gulf Unit I cables / connectors and panel connectors with repair and reinspection as required.
T-Mod lug crimping inspection was performed on Unit 2 on site as well as Unit 1.
Unit 2 cables / connectors and panels were reinspected at San Jose prior to shipment to site. The summary of this inspection / reinspection
(
program is provided in Figure 1.
I 2.
It was determined that tests per 10CFR50 Appendix B were required to verify the technical decisions declared previously concerning the adequacy of PGCC cables / con-nectors to meet intended design functions.
In July 1980, General Electric successfully completed the extensive l
testing required to qualify PGCC cables / connectors and panel connectors to the full requirements of IEEE 323-1974 "IEEE Standard for Qualifying Class IE Equipment for Nuclear Power Generating Stations" and IEEE 344-1975 l
Attachmint to AECM-81/80 Pcge 3 of 5
" Criteria for Seismic Qualification of Class IE Elec-trical Equipment for Nuclear Power Generating Stations."
In order to confirm the functional adequacy of the typical PGCC crimped connection, twenty-four (24) of the connec-tor test samples used in the Qualification Test Program were assembled to include a range of crimp strengths from 2 to 65 pounds - there were over six hundred (600) active contact-to-wire crimped connections in the test lot.
It should be noted here that, in preparing test sample crimps, the proper operation of calibrated crimping tooling yielded crimps normally above the twenty (20) pound range for 20 AWG wire in 16 AWG contact and above the 40 range for 16 AWG wire in 16 AWG contact.
Only by upsetting the crimping process and tooling was it possible to create abnormally low strength crimps.
The lot of twenty-four (24) test samples - sixteen (16) containing 16 AWG wire (approximately 300 pins) and eight (8) containing 20 AWG wire (approximately 300 pins) was subjected to accelerated thermal aging equivalent to forty (40) year life.
Eleven (11) of the test sam-ples were then subjected to seismic and then to envi-
- nmental testing. Finally, all twenty-four (24) test samples were destructively tested.
The destructive crimp pull testing results showed that about 7% of the crimps pull tested were low force crimps.
The qualification test results have demonstrated clearly that crimps of strengths in the range of 1.5 pounds to 65 pounds on 16 AWG and 20 AWG wire in 16 AWG copper pins meet full qualification requirements for the design life of the power plant.
It is the General Electric conclusion that wire / pin configurations of 16/12 in cables and 16/16, 16/12, 20/16 and 20/12 in panel con-necuors are also qualified on a similarity comparison basis.
Attachmsnt to AECM-81/80 Page 4 of 5 The dies of crimping tools are such that for the appli-cable wire / pin combinations the desired mechanical pro-perties will result in the crimped joint. The crimp joint strength is defined by the pin barrel dimples and the retained wire within the pin barrel.
The low force crimp configuration for 16/16 and 20/16, using copper pins, was tested.
No low force crimps were included for 16/12 and 20/12. However, the crimping process and tooling for 20/12 samples was calibrated prior to assembly of 20/12 test samples by destructivelj testing to assure crimp strength well above the eight (8) pound lower limit.
Since the required crimp joint strength was achieved for all configurations tested, it follows that any low force crimp in the 16/12 and 20/12 configuration would have the same properties and behavior on a similarity comparison basis as those of the 16/16 and 20/16 low force crimp configurations which showed positive results in the testing program.
Similarity in the basis for comparison is explained and justified as follows:
IEEE 323-1974 allows for a process of analysis which is a logical reasoning that leads from stated premises to the conclusion concerning specific capabilities.
Further type tests can be made on sample equipment which is hardware that is the same as that used in production.
The hardware used in the test is identical to the pro-duction hardware - only the process of crimp depth is different.
In analyzing the 20/12 wire / pin combination as compared to the 20/16 configurations, only the barrel diameter changes. However, the barrel merely provides the means of containing the wire.
It is the dimple in
- -m
Attachm nt to AECM-81/@0 Page 5 of 5 the barrel which provides mechanical force against the wire.
The 20/12 tested configuration shows that the barrel contains the wire and allows the dimples to provide the necessary force against the wire as was demonstrated by the destructive testing. Since the Qualification Testing clearly showed negligible effects of aging, vibration and environment on crimp-joint strength and the pre-qualification destructive tests showed the 20/12 configuration to have adequate crimp strength, the 20/12 configuration is qualified.
Since the hardware is the same, and the tests for the 16/16 and 20/16 combinations were undercrimped, the above comparative analysis meets the intent of IEEE 323-1974 as an analysis / type test.
With regard to using " brass" pins (Bendix type) in lieu of the tested copper pins for low force consideration, it is noted that both copper and brass may crack in the dimple of the crimp.
In the case of the copper, 13 pins which passed the qualification test, had cracks in the dimple area of the crimp.
Since brass and copper have a close relationship with respect to elongation (roSO% for copper and.63% for brass),
we can expect brass, in the case of the 20/12 combination to have a similar discontinuity in the dimple of the crimp.
i The low force crimp condition for combinations between 20 and 12 AWG for pin and connector, including both " copper" and " brass" pins are qualified for use in nuclear power plants based on the successful completion of test conducted.
1 4
Test results demonstrate the qualification of PGCC cables /
connectors, including low force crimps, to the full re-quirements of IEEE 323-1974 and 344-1975. The results 1
also confirm the adequacy of the 8 and 15 pound pull force values for testing crimped 20 AWG and 16 AWG wire connections in the field.
CABLE / CONNECTOR INSPECTIONS GRAND GULF 1 AND 2 TOTAL POPULATION MIL-STD-1050 0.65 ACTUAL 0.65 AQL (1)
PINS / LUGS FDI/FDDR DESCRI'PTION AQL sat 1PLE SIZE SAMPLE SIZE ACCEPTANCE REJECTED REMARKS CONN. _
PINS (PINS)
(PINS)
NU[tBER_(PINS 1 WAGN UNIT 1 CABLES CAT I SITE 1680 (2) 17 6
2670 25 3
(3)
SAN JOSE 4350 37 9
INSPECTED If TOTAL 1416 20711 315 CAT II SITE 154 (2) 2 0
SAN JOSE 400 6
2 TOTAL 897 7176' 200 554 7
2 ACCEPTED CAT III SITE 300 (2) 4 3
765 9
3 SAN JOSE TOTAL 502 75041 200 1065
'13 6
ACCEPTED 475 (2) 6 22 CAT IV
' lTE a
SAN JOSE 190 3
13 TOTAL 1486 28234 315 665 8
35 INSPECTED 1(
UNIT 2 CABLES 4301 63625 1250 6224 44 17 ACCEPTED UNIT 2 PANELS 3925 50976 500 3146 29 9
ACCEPTED e
I 500 4208 26 20 ACCEPTED WAHX UNIT 1 PANELS 4050 52600 S3 8784 200 8784 65 59 INSPECTED 1 NfT WAKW S
WBFA 1 01 1342 125 1342 15 0
INSPECTED 1 2T i
g g
fT JB1-768 FLAT HEAD PINS 21 34 42680 500 21340 130 35 T
FAILURE RAT ACCEPTED.
JB1-768 792 38016 500 38016 200 98 INSPECTED 1 AM Et "J"
p (1) Accepted numbers obtained from attached MIL-STD-105D Extrapolation Chart - Figure 2.
(2) Site data obtained by Site electricians in special study with 100% GE observation.
(3) 100% inspection of all connectors having less than 48 pins. On 48-pin connectors, outer two rows were pull tested.
If no failures, the inner rows were exempted.
(4) CAT I - 20 AWG conductor in 16 AWG pin; CAT II - 16 AWG conductor in 12 AWG Bendix pin; CAT III-16 AWG conductor in 16 AWG pin; CAT IV-16 AWG conductor in 12 AWG pin.
FIGURE 1 i
..