ML20090K318
| ML20090K318 | |
| Person / Time | |
|---|---|
| Site: | Midland |
| Issue date: | 05/16/1984 |
| From: | Levin H TERA CORP. |
| To: | Lauren Gibson CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| References | |
| NUDOCS 8405240148 | |
| Download: ML20090K318 (6) | |
Text
,
3 b
L J P ltiCl_P_AL ST,A_F L p
a
//'gp JF]RP 1
DjRA 3:
A RA i
D ASP May 16,1984 PAO SCS h///l.
3GA 1
^#
.Fils N
ENF i
Mr. Louis Gibson Manager, Engineering Midland Project Consumers Power Company 1945 West Pornoll Rood Jackson, MI 49201 s
Subject:
TERA Comments on Bechtel Response to Confirmed items C-144
Dear Lou:
TERA has reviewed Bechtel's April 27,1984 letter (No. 149292) which contains on outline of their proposed response to Confirmed item C-144. Our comments, which define our odditional information needs, are ottoched.
Should you require any clarification of our discussion, please contact Mr. Doug Witt at our Berkeley office. Information pertaining to the disposition of this item may either be discussed at our next review meeting in mid-June or vio correspondence.
~
Sincerely,
= >/
i.cw l
Howard A. Levin Project Monoger Midland IDCVP l
Attachment cc:
J. Cook, CPC J. Keppler, NRC, Reg ill D. Eisenhut, NRC, NRR R. Erhardt, CPC r
l D. Budzik, CPC D. Quamme, CPC (site) f R. Whitaker, CPC (site)
D. Hood, NRC, NRR J. Taylor, NRC, l&E T. Ankrum, NRC, I&E J. Milhoon, NRC, I&E E. Poser, Bechtel R. Burg, Bechtel J. Agor, B&W J. Korr, S&W (site)
IDCVP Service List 0
HAL/djb MAY 181634 8405240148 840516 DR ADOCK 05000329 PDR TERA CORPORATION
. 7101 WISCONSIN AVENUE BETHESDA MAIMAND 20814 301 654 8960
\\
SERVICE LIST FOR MIDLAFO INDEPEtOENT DESIGN g
Ato CONSTRUCTION VERIFICATION PROGRAM cc:
Harold R. Denton, Director Ms. Borboro Stamiris Office of Nuclear Reactor Regulat. ion 5795 N. River i
U.S. Nuclear Regulatory Commission Freeland, Michigan 48623 Washington, D.C. 20555 Mr. Wendell Marshall James G. Keppler, Regional Administrator Route 10 U.S. Nuclear Regulatory Commission, Midland, Michigan 48440 Region I!!
799 Roosevelt Road Mr. Steve Godler Glen Ellyn, Ill,inois 60137 2120 Corter Avenue St. Paul, Minnesoto 55108 U.S. Nuclear Regulatory Commission Resident inspectors Office Ms. Billie Pirner Garde Route 7 Director, Citizens Clinic Midland, Michigan 48640 for Accountable Government Government Accountability Project Mr. J. W. Cook Institute for Policy Studies Vice President 1901 Que Street, N.W.
4 Consumers Power Company Washington, D.C. 20009 1945 West Pornoll Rood Jockson, Michigan 49201 Charles Bechhoefer, Esq.
At rnic S fety Licensing Boord Michael I. Miller, Esq.
U.S. Nuclear Regulatory Comm,ission Isham, Lincoln & Beale Washington, D.C. 20555 Three First National Plazo,
[5 Chic o, ll r is 60602 6125 N. Verde Trail James E. Brunner, Esq.
Boca Roton, Florido 33433 Consumers Power Company 212 West Michigan Avenue Jerry Harbour, Esq.
Jackson, Michigan 49201 Atomic Safety and Licer. sing Board U.S. Nuclear Regulatory Commission s.
S"'.'
W shingion, D.C. 20555 57 n set Niw Midland, Michigan 48640 Mr. Ron Collen Michigan Public Service Commission f
Cherry & Flym 6545 Mercontile Way
[
Suite 3700 P.O. Box 30221 l
Three First National Plaza Lansing, Michigan 48909 Chicago, Illinois 60602 Mr. Paul Rau Ms. Lynne Bernobei Midiond Daily News Government Accountability Project 124 Mcdonald Street i_
1901 Q Street, NW Midland, Michigan 48640 Washington, D.C. 20009 t
l i
l l
~
\\
i TERA COMMENTS ON DRAFT OUTLINE FOR BECHTEL RESPONSE TO OCR ITEM C-144 General Comments The draft outline includes specific items which collectively cover all of the concerns raised in the OCR, and generic information regarding aspects of the opproach to piping analysis which in general result in conservative design. The latter should be useful as background to place the OCR issues in perspective; however, our review of your response to the OCR will place heavy emphasis on information which is specific to the Midland project. Such informction con be of two types:
1.
Evidence that the specific technical concerns are not significant in magnitude for the piping and pipe supports in the Midland plant (i.e., the design of Midland is such that inclusion of the factors noted in the OCR would produce insignificant changes in the design of piping and supports).
2.
Information which supports the position that margins due to design measures specific to the Midland plant (i.e.,
margins in addition to those inherent, in " standard" indus-try methods) offset the effects of the OCR concerns.
It is recognized that rigorous mothematical proof of the above is not practical and might necessitate unwarranted analysis. In lieu of this, o careful examino-tion of a sample of Midland's piping may provide the Justification we require to disposition the OCR.
A key requirement for establishing the validity of conciusions reached from such on approach is the validity of the sample.
Statistical proof is not necesscry, but the representativeness of the sample for the specific concern must be demonstrated. It is also expected that the stresses, foods, displacements, etc., will be offected to some extent by any alteration in onalytical method.
The issue is whether such differences are significant.
Guidance for ossessing significcnce is provided in Standard Review Plan Section l
3.7.
(
In general, the draft outline does not include sufficient detail to allow us to determine if the Bechtel response to OCR-C-144 will provide the type of information discussed above and enable us to disposition the OCR. Specific comments are provided below to facilitate the preparation on your resi onse.
i Specific Comments
!!A Revision 0 of Regulatory Guide 1.92 was implemented in the NRC licensing review of plants that filed operating license applications offer April I, 1975. Not all plants complied with all the regulatory positions in the Guide; however, the position regarding closely spaced modes has been odopted by a portion of the industry.
During the May 3,1984 OCR meeting, FSAR Appendix 3D was discussed as documentation of the studies performed by Bechtel addressing the Regulo-tory Guide 1.92 issue. We had previously reviewed Appendix 3D, and more information is required to disposition this aspect of the OCR.
The detailed basis for sample selection needs to be presented.
For example, the results show that pipe support and nozzle foods increase by more than 10%. This increase was then' determined to be acceptable, because design margin was available in the existing design.
How was existing design margin considered in selecting the analysis problems; that is, was the sample biased so that it contained problems where the margin was low? For the nozzle load assessment, we need to know not only the applied foods, but also the alloweble foods and how these were determined.
This type of information needs to be odded to that presented in Appendix 3D.
A second issue is the concern that the margins which were available at the time of the study may be diminished as construction proceeds. TERA would need to see a discussion of the measures which will be taken to prevent a decrease in design margin as construction change regests are evoluoted in the future. Any revision to o piping analysis has the potential for increasing support or nozzle loods. If these increase sufficiently, the 2
~
(
orgument used in the FSAR for neglecting the effects of closely-spaced modes is no leger valid, because the Justification is dependent on the margin between existing copocity and foods calculated.
t IIB lt was at one time on accepted industry proctice to cut off dynamic analysis at 33 Hz. Various approaches have more recently been employed to account for the higher frequency participation. The opproach used on Midland is more conservative than using a 33 Hz cutoff to the dynamic onalysis; however, it will always neglect some portion of the response. The question is whether that portion neglected is ever of practical significance.
We concur with point I of the Bechtel outline, and also recognize the shortcomings of results of calculated response at high frequencies as mentioned in point 2. Residual response con, however, be approximated in on acceptable manner using computer programs that calculate the effec-tive mass not included in the dynamic response at lower frequencies, and then use equivalent static methods for that mass. This method has been applied to certain plants including, for example, the SMA Seismic Margins Evoluotion of Midland. The Bechtel program ME101 appears to have this some capability based on our review of the users' manual. This would provide a basis for comparing such results with the present Bechtel f
method. As in the item IIA discussion, sampling technique is critical to the validity of the conclusions.
I IIC Industry practice has changed in the lost ten years regarding support stiffness. It may not be practical to reonalyze piping to confirm that the effects of support stiffness are sufficiently small to neglect.
As on alternative, industry design practice for pipe supports frequently included methods which essentially bounded the stiffness in a manner which l
mitigated the impact of this issue. One opprcoch to responding to this concern is to dernonstrate that measures were token to ensure that engineering consideration of support stiffness was taken or will be taken during as-built confirmation.
3
g The offsetting effect of thermal and seismic loads only applies to higher temperature piping. Adequate consideration needs to be given to low temperature piping where seismic support loads may dominate.
i lilC The Seismic Margins Evoluotion piovided by SMA for the seismic margin earthquake shows that many of the supports sampled were very conservo-tively designed. Credit for such conservatism con be taken to offset the OCR concerns, provided it con be demonstrated that it was a result of a controlled design process. There needs to be assurance that the conservo-tive methodology was universolly applied and that it will continue in the design revision process, if the conservatism is used to offset uncertainties arising from the piping analysis methods addressed in the OCR. Sompling con be used to demonstrate post conservatism inherent in the design process; however, without evidence of specific controls which led to the conservatism, the sample would have to be large and statistically random.
l l
l l
l i
4
.- -