ML20086D328
| ML20086D328 | |
| Person / Time | |
|---|---|
| Site: | Vermont Yankee File:NorthStar Vermont Yankee icon.png |
| Issue date: | 06/01/1976 |
| From: | French J VERMONT YANKEE NUCLEAR POWER CORP. |
| To: | James O'Reilly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| Shared Package | |
| ML20086D325 | List: |
| References | |
| WVY-76-64, NUDOCS 8312010170 | |
| Download: ML20086D328 (4) | |
Text
{{#Wiki_filter:- O O VERMONT YAN KEE NUCLEAR POWER CORPORATION SEVENTY SEVEN GROVE STREET s RUTLAND. VERMONT 05701 M E PL,Y TO s ENGINEERING OFFICE June 1, 1976 TURNPIKE Mo AD 4 WESTBoRO, M A55 ACHUSETTS 0 t S81 TM.EPHONE 687 366-9011 WVY 7A u !D \\ -t United States Nuclear Regulatory Conunission .S Region I [? Office of Inspection and Enforcement / s 631 Park Avenue g M Y9) 3 King of Prussia, Pennsylvania 19406 i Attention: James P. O'Reilly, Director References (1) License No. DPR-28 (Docket No. 50-271) 9 b (2) Abnormal occurrence No. 75-17. - (A l - (3) GE Report NEDC-21,000, " Investigation of Cause of Cracking in Austenitic Stainless Steel Piping of Boiling Water Reactor Plants," 1975. (4) NRC Report NUREG-75/067, " Investigation and [ Evaluation of Cracking in Austenitic Stainless Steel i-Piping of Boiling Water Reactor Plants," 1975. lt f
Dear Sir:
This report is being supplied to you as a result of a verbal request from your office following the event at vermont Yankee which resulted in Abnormal Occurrence 75-17. Startup procedures were underway at vermont Yankee during August 1975 when a visual inspection of the reactor containment and its internals b revealed primary water dripping from some pipe lagging. Further examination I revealed that the source of the water was a Safety Class 1 pipe line - the l PSE instrumentation line - whereupon a cold shutdown of the plant was initiated. The PSE instrumentation line is a one inch schedule 80 304 stainless steel line.ihich originally comprised part of the Low Pressure Core I I Injection (LPCI) loop select logic. Each of the ten recirc system jet pump risers has one of these 1" sch. 80 PSE line penetrations. As a result of an ECCS upgrading implemented during the fall of 1974, the affected instrument lines were made non-functional by making them non-i flow lines which were dead ended outside primary containment. An ir. situ visual examination was made by verr.ont Yankee and an Abnormal Odcurrence (AO) report (Reference 2) was filed with the NRC l which attributed the failure to a fatigue mechanism. This conclusion was based on the following observations: I 8312010170 750909 ~^ 67/O-PDR ADOCK 05000271 S PDR i( . copy SENT REGIONN. i 1 L . - r - ~. - I
O O United States Nuclear Regulatory Commission June 1, 1976 Attn James P. O'Reilly, Director Page Two (1) Leaks in the 1" PSE line were located in two localized areas that appeared to be welding are strikes, (2) With the plant in a cold shutdown condition noticeable vibration was observed in the line at an area of improper support. (3) The pivot or cantilever point for the vibration appeared to be in the vicinity of the leaking. (4) Visual examination of the remaining nine (9) PSE lines revealed no vibration or improper supporting. The failed section of the pipe was removed and the two ends were plugged by Vermont Yankee personnel. The section of pipe containing the failed area was sent to the 6 General Electric Vallecitos Center for a failure analysis in an atter.pt to identify the failure mechanism. A subsequent analysis on the same section of pipe was also performed by Mr. Helmut Thielsch. Based on visual, dye penetrant, and metallographic examinations GE concluded: (1) The failure mode was intergranular, originating on the inside surface of the pipe in the weld heat affected zone (HAZ). (2) The most probable failure mechanism was stress corrosion cracking. l (3) The Type 304 stainless steel was not severely sensitized. The analysis by Mr. Thielsch confirmed that the failure was primarily due to intergranular stress corrosion cracking. His analysis yielded the following informations r (1) The failure of the pipe by cracking adjacent to the socket weld joint is due primarily to stress corrosion cracking. (2) The stress corrosion cracking progressed from the inside surface of the pipe towards the outside surface. (3) The stress corrosion cracking occurred only in the area which had been sensitized by welding. (4) In accordance with ASTM Specification A262 utilized to revaal sensitization by the oxalic acid etch test, the weld deposit would be considered severely sensitized. This heat-affected zone condition, however, is typical of stainless steel weld deposits involving Type 304 or Type 316 grades, as extensively installed in nuclear power plants and providing entirely satisfactory service. I i 4
o o United States Nuclear Regulatory Commission June 1, 1976 Attn James P. O'Reilly, Director Page Three (5) Contributing to this type of stress corrosion crackfr;g are residual welding stresses, a bending moment and fatigue. The bending moment in the pipe joints generally veries significantly around the pipe circumference. The stress corrosion cracking normally occurs in these piping systems at one location around the circumference cnly. Thus, this type of stress enhanced corrosion cracking has not, and is not expected to result in pipe rupture. (6) The cross section was also etched in KOH, utilized to reveal the presence of sigma phase. Sigma phase formation had not occurred. It is important to note the apparent discrepancies in the GE and Thielsch findings: (1) Thielsch concludes that there are areas of severely sensitized material - GE concludes there are no such areas. (2) Thielsch indicates that there is evidence of fatigue / bending moments, however, GE makes no mention of these indications. The failure analysis by Mr. Thielsch clearly indicates that there was a very complex mechanism present which eventually led to the failure of the pipe. His findings also substantiate the visual examination done by Vermont Yankee at the time of the failure, i.e., vibration, and GE's 4 findings of intergranular stress corrosion cracking. Actually, when one combines the Vermont Yankee and GE findings, one arrives at the conclusions of the Thielsch analysis-intergranular stress corrosion cracking with evidence of fatigue and bending stresses. At this time there is no way of assessing whether or not the failure would have occurred if the vibratory stresses discovered during the visual examination were not present. While fatigue failures of weld sensitized stainless steols in high cxygenated water are normally identified by their transgranular mode of crack propagation, they can indeed be intergranular (References 3 and 4). It is certainly reasonable to conclude that the stresses that resulted from the vibratory loading could have contributed significantly to this intergranular stress corrosion failure. Due to the complex nature of this failure and the phenomena of stress corrosion in general, no hard case based on materials behavior can be developed that would necessitate the removal of the remaining nine instrument lines, therefore, we have made no plans to removes them. e ,,,.,,gr-l,,. w., ,.ny -n,,..-, -.-w ,,,,.s-- s-, ,.n.
O O United States Nuclear Regulatory Commission June 1, 1976 Attn: James P. O'Reilly, Director Page Four We trust that this submittal is an adequate response to your request, however, should you desire additional information, please contact us. Very truly yours, VERMONT YANKEE NUCLEAR POWER CORPORAT40N j ll* % T , f.w _r_ J. L. French Manager of Operations WPM /kg i l l l 4 -aw. =}}