ML18058B321
| ML18058B321 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 01/07/1993 |
| From: | Slade G CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9301140206 | |
| Download: ML18058B321 (3) | |
Text
(
.:,.J i'
consumers Power
~
.. ~
POWERINli MICHlliAN'S PROliRE55 Palisades Nuclear Plant:
27780 Blue Star Memorial Highway, Coven, Ml 49043 January 7, 1993 Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 GB Slade General Manager DOCKET 50-255 - LICENSE DPR PALISADES PLANT - 10CFRS0.61 PRESSURIZED THERMAL SHOCK - ADDITIONAL INFORMATION On October 23, 1992, the NRC notified Consumers Power Company {CPCo) that the Oyster Creek surveillance weld was reported as having been fabricated with RACO 3 heat number W5214 weld wire and ARCOS BS flux and that the copper content of this weld was reported to be 0.28%.
This copper content is significantly higher than the 0.179% copper best-estimate value for welds with RACO 3 heat W5214 weld wire with nickel addition which was reported in, our June 5, 1992 revised Palisades Projected RTprs submittal.
In telephone conferences on November 24 and December 1, 1992, CPCo informed the NRC staff that because of {l) abnormally high copper content, {2) no nickel added to the weld, {3) use of a different flux, and {4) use of a different weld procedure, we concluded that we could not recommend using the Oyster Creek surveillance weld data as input for determining the best-estimate chemistry for RACO 3 heat W5214 nickel added reactor vessel welds.
Our position was based on the same rationale we used to exclude uncharacteristically low nickel or copper content data from our June 5, 1992 projected RTprs determination.
The staff asked that we place our conclusion on the docket. This letter fulfills that request.
We have determined that the data reported for the Oyster Creek surveillance weld is unlikely to be representative of data which should be used to determine the best-estimate copper content for welds made with RACO 3 heat W5214 weld wire with nickel addition and, for the following reasons, should not be included in data used to determine the best-estimate chemistry for the Palisades reactor vessel {RV) RACO 3 heat W5214 welds.
9301140206 930107
'?
PDR ADOCK 05000255 P
/
I
}
I.
~
~.
2
- 1.
The Oyster Creek surveillance weld was fabricated using a different flux than that used in the welds used to determine the best-estimate chemistry for RACO 3 heat W5214 nickel added welds.
As mentioned in our June 5, 1992 RTPTs submittal, ARCOS BS flux was replaced by Linde 1092 in order to produce more consistent quality in the weldments.
Although it has not been either qualitatively or quantitatively documented, this difference in type of flux could affect the weld copper content.
- 2.
The Oyster Creek surveillance weld was made using the.CE single arc with no nickel addition welding procedure SAA-33A (3).
The Palisades reactor vessel RACO 3 heat W5214 welds and the welds used to determine the best-estimate chemistry for the Palisades RACO 3 heat W5214 welds were made using CE welding Procedures SAA-33-H and SAA-33-J for single and tandem arc welds with nickel addition.
Use of a different procedure resulted in the Oyster Creek surveillance weld having a nickel content of less than
.10% and thus results in a weld which is not representative of welds made using CE welding procedures SAA-33-H and SAA-33-J.
- 3.
Items 1 and 2 above provide argument that the Oyster Creek surveillance weld does not physically qualify as a member of the data set used to derive the Palisades beltline weld best-estimate chemistry.
If it were assumed to be a member of the set, it would be a statistical outlier for the following reasons:
- a.
The purchase specification for the Oyster Creek surveillance weld specimen required the copper content of the weld to be less than 0.20%.
The reported copper content of 0.28% is so far out of specification that it suggests the fabrication of the specimen was not adequately controlled.
- b.
When the Oyster Creek surveillance weld data is compared to the tightly grouped RACO 3 heat W5214 copper data reported in our June 5, 1992 submittal, the 0.28% reported copper content exceeds the mean value of the other data by 5 standard deviations.
- c.
Our June 5, 1922 submittal contained a least squares chemistry factor of 217 degrees F which was derived from Charpy tests of power reactor irradiated specimens of heat W5214 nickel added welds.
It is well established that the CE weld procedure for the nickel added welds targeted 1% nickel.
CPCo in its June 5, 1992 projected RTPTs report and EPRI in its PTS report for H.B. Robinson 2 concluded that the nickel content for the RACO 3 heat W5214 nickel added welds should average 1.05% and 1.10% respectively. With nickel in the 1.05% to 1.10% range, copper would have to be in the 0.17% to 0.19% range in order to obtain the derived 217 degrees F chemistry factor.
The 0.28%
copper reported by Oyster Creek is clearly outside this range.
\\J 3
Therefore, Consumers Power Company's projected date for when the Palisades reactor vessel RACO 3 heat W5214 nickel added axial welds will exceed the 10CFRS0.61 RTPTs screening criterion remains the same as that reported in our June 5, 1992 submittal at year 2010.
If the Oyster Creek heat W5214 data were to be included~ that date would change to the year 2004.
However, for the reasons stated above, we have not included the Oyster Creek data in our projection.
~~~-
Gerald B Slade General Manager CC Administrator, Region III, USNRC NRC Resident Inspector-Palisades