ML20212P566
| ML20212P566 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 03/05/1987 |
| From: | Eenigenburg E COMMONWEALTH EDISON CO. |
| To: | Bradley Davis NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| References | |
| 87-154, CAL-85-04, CAL-85-4, IEB-79-14, NUDOCS 8703160154 | |
| Download: ML20212P566 (5) | |
Text
'
Commonwealth Edison Dresden Nuclear Power Sttion R.R. #1
,f Morris, Illinois 60450 Telephone 815/942-2920 March 5, 1987 EJE LTR: 87-154 Mr. Burt Davis Regional Administrator Director of Inspection and Enforcement Region III U.S. Nuclear Regulatory Commission 799 Roosevelt Road Glen,Ellyn, Ill. 60137
References:
1) Confirmatory Action Letter (CAL) 85-04 dated April 5, 1985.
Dear Sir:
This letter is in reference to Confirmatory Action Letter (CAL) 85-04 regarding the required functional testing of twenty (20) safety-related mechanical and hydraulic snubbers during the 1987 Dresden Unit 2 refueling outage. Item 4.0 of this confirmatory Action Letter requires a verbal notification to Region III within 2 working days followed by a written report within 30 calendar days. Six (6) of these tested snubbers failed to meet the functional testing criteria. They are: Snubber #1527 Notification made to N. Chrisstimos by J. Brunner (PSA-10) February 6, 1987. Snubber #9 Notification made to J. Gavula by J. Welch (PSA-35) February 9, 1987. Snubber #1510 Notification made to J. Gavula by C. Smith l (AD 5500) February 17, 1987. Snubber #1301, Notification made to J. Cavula by J. Welch
- 1302, #1303 February 18, 1987.
(BP 2%) The written report for each snubber is attached. Sincerely, E. D. Een, ige urg Station M er Dresden Station RS:EDE:rg 8703160154 870305 ' I ( PDR ADOCK 05000237 PDR %W 9g eP g op
Snubber #9 Location - Recirculation Pt mp Bowl 2B-202 Descriptiota of Failure - During an ae:tivation test in the compression direction at 25% of rated load the anubber (PSA-35) locked up and failed to release, resulting in ar acceleration of 0.000G's. Supplemental tests were performed on Lt.ls snubber to 3nvestigate its failure. First, activation tests of 25%, 50%, 75% and 100% of rated load were conducted. Although some signs of iockup were noted at 25% of rated load, the snubber released as designed at all other loads. Secondly, multiple tests at and around 25% of rated load were performed. Signs of lockup were observed but a total lockup of the snubber was not repeated. l Analy'ais - Upon analysis it was determined that this low load activation does not invalidate the piping analysis methodologies used in the IE Bulletin 79-14 analysis because the snubber passed both the running drag and breakaway drag tests. Thus, the snubber would have acted normally under thermal loading and would have locked under dynamic loadings. This snubber can be considered operable at all load levels. This snubber is being replaced with a new PSA-35. In conversations with PSA, indications are that the cause of this test anomoly is that the retainer clip on the capstan spring is misaligned. 1 I l
y. Snubber #1510 ' Location'- Line 2-1502-24" Emergency Core Cooling System (ECCS) Suction. Header Description of Failure - The final running drag in both compression and tension were above the test criteria of' 2% of rated load (1,100 lbs) with peak up to 5% of rated load (2,750 lbs) for less than 500 mil?.iseconds. The average final running drag of this Anchor Darling 5500 anubber was 1986.2 lbs in tension'and 2249.7 lbs in compresulon. Peaks in the compression direction were as high as 3,500 lbs. Due to the fact'that-the previous AD5500 had only marginally passed the running drag test and that these type of failures had occurred elsewhere, all eleven (11) AD5500 snubbers were removed from Unit 2 and functionally tested. Of the eleven tested, six (6) failed the test in this manner. Analysis - 4 Anchor Darling was notified of the problem and reviewed the test data for all eleven AD5500 snubbers. Their analysis classified these failures as the previously identified problem with a ball screw assembly in which a ball skids within a ball _ return tube rather than rolling at the entrance and exit of the ball return tubes. As a corrective action, Anchor Darling disassembled the snubbers and ground the entrance and exit paths of the ball nut to allow for smoother operation. The ball was then reloaded to fill the ball nut and the snubbers reassembled. The snubbers were then retested. This functional testing showed that seven (7) of the anubbers were unable to pass the running drag test. Therefore, the decision has been made to replace all of this type snubber with PSA-35's. From a piping analysis standpoint, an analysis had previously been reported by Nutech in letter TWSH-86-011 dated March 13, 1987. The analysis evaluated the ECCS suction header for restricted thermal expansion of the piping resulting from such increased drag forces. Upon review of the Wyle test data, Nutech has determined that the previous snubber failure evaluations bound the current condition of six failed snubbers. As such, there is no operability concern associated with this issue for normal and accident conditions. i l'
e -Snubber' #1301. #1302. #1303 i. - Location - (1301, 1302)'Line'2-1303-12" isolation condenser condensate return line. (1303) Line 2-1302-14" isolation condenser ' steam supply line.' Description of Failure - .While performing'the' activation test, the-three isolation condenser hydraulic anubbers failed to lock uo within the required testing criteria of 1 to 10 in./ min. The results cr the snubber testing are as follows: Maximum Test Support Lockup Load Snubber I.D. Velocity Obtained 1301 2-1301-M-203 41.5 in./ min. 9406# -1302 2-1301-M-204 49.9 in./ min. 10526# 1303 M-1163D-79 > 49.9 in./ min. 593# ' Snubber 1303 with a lockup velocity greater than 49.9 in./ min. could not be activated on the test machine. -Analysis - The computer models of the two systems were reconstructed using the 79-14 piping analysis outputs. These two lines were then analyzed with the three " failed" snubbers removed from the models using standard Final Safety Analysis Review (FSAR) methodology. The analysis showed that both systems do not meet the FSAR criteria established during 79-14. License Event Report (LER) 87-005-0 on docket 50-237 has been initiated. The two piping systems were shown to meet operability criteria taking into account the test results of the three " failed" snubbers. The only snubber that was truly a failed snubber was #1303. The other two snubbers were determined to still be active f9r dynamic loading using a lower snubber stiffness. An analysis performed by Impell has determined that the two piping systems meet oper.tbility criteria. The methodology used 3s consistent with the NRC accepted I.E. Bulletin 79-14 Program. The snubber testing procedure used at the time of installation only required that the snubber lock-up and bleed visually when activated on a i test machine. No criteria was given to verify acceptability for these snubbers. Verification of a specific setting was not required when the hydraulic snubbers were originally installed. At that time, the test equipment used was not capable of verifying these settings. Plans had previously been made to replace these hydraulic snubbers with mechanical PSA-10 snubbers under modification M-12-2-83-47. These three snubbers are the last remaining safety-related hydraulic snubbers on Unit 2. I i- ,,-re,,,- -,-n--- --.-+-e,---,-- ~. -..... .,..n..,. --,.,,....----c---c
r-T Snubber #1527 -Location - Line 2-1504-10" DK. Low Pressure Coolant Injection (LPCI) system drywell spray line, outside drywell penetration X-150A. Description of Failure - While performing an initial-breakaway friction force test, the snubber failed to'auve in the compression direction. Snubber #1527 (PSA-10) was loaded to 15,000 lbs. in the compression direction ~(100% of rated load) with no movement noted. Upon removal of the snubber extension piece, it was discovered that the bottom snubber dust cover was dented indicating that the thrust bearing had been damaged. Internal examination revealed that the inner race of the thrust bearing was " corn cobbed". The outer races showed signs of fretting (in lairs approximately 1/16" apart). There was also an indication of a slight bend in the shaft. All these indications have implicated an overload as the cause of the failure. This is identical to a previous failure of this snubber during the last refuel outage. Analysis - A visual inspection was performed on the structural steel for the snubber, the local pipe supports, and nine-pipe welds surrounding the snubber location. No abnormalities or reportable indications were detected during the inspections. Furthermore, magnetic partical inspection of six pipe and drywell liner welds is being performed. Preliminary indications show all welds tested as being acceptable. From a piping operability standpoint, the analysis done following the previous failure of this snubber during the last refueling outage remains valid and is applicable to the current circumstances. A thermal expansion analysis was performed tc study the effect of having a rigid support instead of a snubber at the location. The analysis was performed both for normal operating and maximum operating thermal conditions. The piping stressos were found to be acceptable based on the operability criteria of twice the yield stress. The fatigue evaluation was performed assuming 40 cycles of 35 Ksi load applied in the direction of the locked snubber due to water hammer events during the life of the plant. The maximum pipe stress was 3605 Kai at the drywell penetration I-150A. The fatigue usage factor per ASME Section III, Division 1, Appendix I, Figure I-9-1 for 40 cycles at the above stress in the material was 0.003. This result is judged to have insignificant effect on the fatigue life. From this evaluation and the operability assessment, it has been determined that operation with this snubber locked up is acceptable. In an on-going effort to discover the root cause of this recurring snubber failure, a review is being performed of station operating and testing procedures, Technical Specifications, and system configurations for conditions that might cause water hammer. This review will be performed by Nutech and will include interviews with station personnel responsible for implementing these procedures. Also, monitoring equipment (strain gauge and linear variable displacement transducers) shall be installed on the snubber to detect the occurrence of future transients and to characterize future transients.}}