ML20198E431

Affidavit of G Edwards Re Intervenor Contention (D).Dnbr of 1.17 for Optimized Fuel Assembly Fuel in Transitional Mixed Core Unwarranted Unless Detailed Studies Done.Adoption of Lower DNBR Value Would Allow Operation at Hotter Temp

ML20198E431
Person / Time
Site:	Turkey Point
Issue date:	11/06/1985
From:	Geoffrey Edwards CENTER FOR NUCLEAR RESPONSIBILITY
To:
Shared Package
ML20198E427	List: ML20198E424 ML20198E431 ML20198E435 ML20198E444
References
OLA, NUDOCS 8511140024
Download: ML20198E431 (6)

Text

- . . . - - _ , - .

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION In the Matter of )

FLDRIDA POWER & LIGHT COMPANY )

(Turkey Point Plant, Units 3 & 4) ) l l

AFFIDAVIT OF GORDON EDWARDS

, REGARDING CONTENTION (d)

I, Gordon Edwards, being duly sworn, state as f ollows:

1. I am a professor of mathematics and science at Vanier College in Montreti, a professional consultant on matters related to nuclear power reactors, and the President of the Canadian Coalition for Nuclear Responsibility. A summary of my professional qualifications and experience is attached to a previous af fidavit which was filed by the Intervenor, Ms. Lorion.

2. The purpose of this Affidavit is to address the

" Licensee 's Motion f or Summary Disposition of Intervenor's Contention (d)" and the " Licensee *s Statement of Material Facts as to Which There Is No Genuine Issue to be Heard With Respect to Intervenor's Contention (d)" dated September 20, 1985, as well as the "NRC Staff Response to Licensee's Second Motion for Summary Disposition of Conta.ition (d)" dated October 15 1985. The arguments advanced in these documents in relation to Contention l

l l (d) are not only inconclusive but invalid. I do not concur in the I

l . conclusions which are reached in those documents and in the supporting affidavits.

3. As set forth by the Licensing Board, the three genuine questions as to material f acts which remain f or litigation are:

(i ) Whether the DNBR of 1.17 which the amendments impose on the OFA fuel in Units 3 and 4 comepnsates f or the three

uncertainties outlined by the Staff in its December 23, l 1983 SER on the amendments, at 4.

l I

8511140024 851107 PDR ADOCK 05000250 G PDR i

(ii) Whether, if the DNBR of 1.17 does not compensate for those uncertainties, the SRP's 95/95 standard, or a comparable one, is somehow satisfi ed.

j' (iii) Whether, if that standard is not being satisfied, the reduction in the margin of safety has been significant.

Order at 64. The Licensee has indicated that, in its opinion, the answer to the first question is "no" but that the answer to the other two questions is "yes". The NRC Staff Response has supported the Licensee in all three of these answers. The f ollowing discussion identifies several important shortcomings in the arguments adduced by the Licensee and the NRC Staff.

4. The Licensee readily admits that the DNBR of 1.17 which the amendments impose on the DFA fuel in Units 3 and 4 does not compensate for the three uncertainties outlined by the Staff in its December 23, 1983 SER -- uncertainties related to rod bow, increased hydraulic resistance due to the transitional mixed core, and applicability of the WRB-1 correlation to the 15x15 DFA fuel design. Indeed, both the WRB-1 correlation and the Licensee's own plant-specific safety analysis have been 4

predicated on the assumption of a homogenous core rather than a l

mixed core. On the face of it, theref ore, the applicability of the DNBR of 1.17 to the DFA fuel in the transitional mixed core is highly questionable, based as it is on a hypothetical i extrapolation of results obtained f rom a consideration of homogenous cores only.

5. Rather than perf orming the appropriate studies, based on the assumption of a mixed core, with all the complications and dif ficulties that that assumption entails, the Licensee and the

NRC Staff have accepted the results of studies based on the more convenient (but incorrect) assumption of a homogeneous core, and have attempted to compensate for the obvious inadequacies of such an approach by assigning numerical penalties to those results.

In the absence of any detailed study of the transitional mixed core itself, however, such numerical penalties are little more than " educated guesses" which cannot be regarded as having the same status of reliability as the WRB-1 correlations which have been painstakingly derived for homogenous cores. For this reason, it would seem prudent not to countenance the use of a DNBR of 1.17 for the DFA fuel in Units 3 and 4 until the transitional period has passed and the core is truly homogenous, as assumed in all the relevant studies which were done in developing the WRB-1 correlation as well as in the plant-specific safety analysis.

6. As the Licensing Board has pointed out, The behaviour of a prudent driver depends on how clearly he can see the edge of the road. If the edge is shrouded in fog, the prudent driver will steer clear of where he thinks the edge might be, and the more in doubt he is about where the edge is, the more he is inclined to stay away from where he thinks it might be. If, however, the f og clears some, the driver, by moving closer to the edge of the now more visible road, does not increase his chances of going off the road.

l There are several points to bear in mind in applying such logic to the situation at hand. Firstly, the cauti ous driver has no particular incentive to drive closer to the edge of the road, j whereas there is an obvious financial incentive on the part of l FPL to operate Units 3 and 4 hotter rather than cooler. Secondly, the operator of a nuclear power plant is dependent on instruments l

and sensors rather than his eyesight to communicate the actual l

I l

l

condition of the fuel in the core of the reactor, and such instrumentation may be inadequate to convey an unambiguous message as to what is really going on in the core of the reactor.

Thirdly, the fog (in the case of Turkey Point Units 3 and 4) is related to the various uncertainties in the applicability of the analysis, some of which are statistical uncertainties (which have been studied in detail and are demonstrably subject to statistical laws) while others are related to omissions, oversights, incomplete analysis or inappropriate assumptions.

Fourthly, the degree of prudence exercised by a driver depends to some degree on the perceived hazard; if the road cuts straight across flat prairieland, the driver may be relatively unconcerned about the edge of the road, fog or no fog - but what if the road borders on a fairly deep ditch, and the shoulders of the road are soft?

7. As a mathematicien, I would not expect the behaviour of a transitional mixed core of LOPAR and OFA fuel to be simply a

" linear combination" of the characteristics of two hypothetical homogenous cores, one entirely composed of LOPAR fuel, and the other entirely comprised of OFA fuel. On the contrary, I would expect some non-linear interactions to intrude thmselves into the calculations. It is customary in all areas of science (pure and applied) to use linear approximations as a crude estimate of non-linear behaviour, but it is well-known that in the case of highly i non-linear behaviour, the resulting linear approximations can be quite far off the target. In particular, such linear approximations are not, in general, always conservative. As a 1

very simple illustration, both the NRC Staff and the Licensee utilize a rod bow penalty of 5.5 percent of DNBR, a transitional mixed core penalty of 3 percent, and an uncertainty of less than 2 percent for the application of the WRB-1 corrlation to the 15x15 OFA fuel design; they then add these to obtain a maximum total penalty of 10.5 percent. But this is only a crude linear estimate of the actual result of combining these three penalties, which (assuming independence) may be found by multiplying as follows: ( 1. 055) x ( 1. 03) x ( 1. 02) = 1.10838 : a total penalty of 10.838 percent, not 10.5 as claimed. If we go further and make the reasonable assumption that the three penalties are not independent of each other, but interact with each other, the resulting maximum total penalty could go considerably higher. For example, it is entirely likely the rod bow phenomenon might interact in a fairly complicated way with the already complicated non-uniform hydraulic resistance phenomenon. These comments are simply meant to reveal the crude nature of the approximations being used by the NRC Staff and the Licensee in their efforts to justify the use of a DNBR of 1.17 despite the uncertainties which are inherent in the transitional mixed core configuration. Much more detailed study would be needed to provide a realistic estimate of the appropriate DNBR to use for the OFA fuel in Units 3 and 4 because of the mixed core configuration. In the absence of such detailed study, it would be prudent to apply a more 1

conservative DNBR figure until the transition to a homogenous l l

core has been completed.

8. For the reasons outlined above, I believe that the adoption of a DNBR of 1.17 for the OFA fuel in the transitional mixed ccre is unwarranted unless detailed studies are done, proceeding from the assumption of a mixed core, and confirming that a DNBR of 1.17 is indeed appropriate under such circum-stances. It must be remembered that this amendment to the technical specifications is being requested to support FPL's program for reduction of neutron bombardment and consequent embrittlement of the pressure vessel walls. In other words, FPL is undertaking a program of corrective action to overcome a serious safety problem which escaped analysis at the time the plant was originally licensed. Under such circumstances, it is my considered opinion that any unwarranted assumptions or incomplete analysis designed to justify FPL's wish to " drive closer to the edge of the road" should not be countenanced - not until the corrective action has been completed, at least.

9. Operationally, the adoption of a lower value for DNBR translates into permission to operate the reactor at a hotter temperature. Since the object of having a DNBR margin is to prevent overheating of the fuel and subsequent release of fission products into the primary coolant, it seems germane to observe that the releasable fraction of iodine-131 depends on the operating temperature of the fuel prior to the release.

The foregoing is true and correct to the best of my knowledge and belief.

E DTL Gordon Edwards 5 (N 0 A Ai A 3 F o A. c HG on F12 C T A E A 4. , TH o'S 5 BAf CP

• J F M S

• R- /9 15 gxh ws$ 5 5l

f. ,e f _