ML20096F397
| ML20096F397 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 09/04/1984 |
| From: | Lorion J CENTER FOR NUCLEAR RESPONSIBILITY |
| To: | |
| Shared Package | |
| ML20096F384 | List: |
| References | |
| 84-496-03-LA, 84-496-3-LA, CLA-1, LA, OLA-1, NUDOCS 8409070430 | |
| Download: ML20096F397 (10) | |
Text
~
fl AFFIDAVIT OF JOETTE LORION I, Joette Lorion, being duly sworn, say as follows:
- 1. I am Research Director of the_ Center for Nuclear Responsibility, a non-profit, nuclear information and' resource center. I have been writing and researching nuclear safety issues since 1978, and have acted as a consultant to Dan'Rather, 20- 20, CBC, The Wall Street Journal, New York Times, Government Nuclear Oversight Committees, and others.
I have personal knowledge of the matters stated herein, and I believe them to be true and correct. I incorporate, by reference, all the statements' by Dr. Gordon Edwards in his of- affidavit of August 30, 1984, anc include them with my own is support of Intervenors' contentions (b) and (d).*
- 2. Intervenors' contention (b) states:
Whether the entirely new computer model used by the utility, for calculating reflood portions of the accidents, meets the Commission's ECCS Acceptance Criteria: specifically, whether a 2.2% reduction in reflood rate is misleading because for a small decrease in reflood rate, there results a large increase in fuel temperature. Reflood rates are critical if below 1 or 2 inches per second.
In support of contention (b), it is clear that the Westinghouse Emergency Core Cooling System Evaluation Model utilizing the new "BART-A1: COMPUTER CODE FOR BEST ESTIMATE ANALYSIS OF RFLOOD TRANSIENTS" (BART) >
is'a contrived computer model, which consists of an uncertain l l
8409070430'840904 PDR ADOCM 05000250 l G PDR ,
(2) computer code loosely coupled with other models prepared for other fuel core designs, and does not equate a computer code specifically designed for this technology and does not constitute compliance with 10 C.F.R. 50.46.
DISCUSSION In a loss of coolant accident (LOCA) , it is in the critical time period from about 15 seconds after the rupture to 30 seconds that control of the accident must be gained by the emergency core cooling system's (ECCS) operation,and the fuel temperature excursions halted. It is of utmost importance that adequate cooling water flow upward through the core as quickly as possible in this period to prevent overheating and core melt. If the vertical flooding rate,once emergency cooling reaches the core bottom, is below some critical value (presently believed to be in the vicinity of 0.7 inches per second), then the accident will proceed out of control. .
In order to ascertain that a LOCA will be controlled by the ECCS in an accident, reactor manufacturers, licensees, and the NRC resorted to computer calculations. In 1967, when the ECCS was designed, it was believed-that if there were no inhibition to reflooding, the water level would rise in the core at between 8 and 10 inches per second. As studies continued, it was found that adverse circumstances could drastically reduce reflood rates. At the present time, reflooding rates expected-for an accident in the operating reactor are between 0.9 and 1.5 inches per second. Thus, the lowering of the reflood rates at Turkey Point to 1.17, means that there are, at best, relatively small safety margins between effective cooling and loss of control. And, that this reduction is not sufficiently conservative to guarantee safety margins.
- 9. ..
(3) .
In 1975, the American Physical Society's Study Group on Light Water Reactor Safety,-stresse the importance of conservatism on reficod rates when they stated, We_ thoroughly support the AEC's recommendation concerning PWRs that, 'the calculated reflood rate should have a substantial margin over the rate that is just sufficient to turn the plant around.' (A.P.S. Report 5.31, 1975)
WHY BART CODE DOES NOT MEET REOUIREMENTS OF 10 C.F.R. 50.46 The computer code and analysis performed by Westinghouse to support the lowering of the reflood rate-at Turkey Point and subsequent reduction in safety margin is an uncertain, modified, and patched up code that does not meet the requirements of 10 C.F.R. 50.46. In fact, there is already a new improved version according to licensee's Affidavit of Marvin J. Parvin, page 3, which states, "The BART grid rewet model, which is now undergoing NRC review, is an improved version of the BART code and accounts for increased heat transfer due to the spacer grids."
NRC acceptance, although conditional, of 'this unfinished and uncertain computer code in order to grant FPL a license amendment that would allow them to reduce safety margins ( reflood rates ) at Turkey Point, shows that they clearly have not met the responsibility to establish with finality that the public health and safety will be protected by adoption of this code.
In fact, the NRC seems t6 be falling into a pattern 6f conduct that the A,EC warned against in a report in 1971 which stated, 1 As reactor designs and their operating characteristics changed, the analyses methods were ' patched up', rather than redeveloped, with the net result that over all existing methods are inefficient, inflexible, and do not adequately represent the physical phenomena intended. (see ' Water Reactor Safety Program Argumen -
tation Plan,' USAEC, Nov. 1971, Exhibit 1026, ECCS Hearing Record. )
E .; ..
~
l (4)
One should also consider a warning by Dr. Alvin Weincerg, Director of Oakridge National Laboratory to then AEC Chairman James Schlessin in Feb. 1972,-which stated with respect to codes in reactor safety analysis, that he had"a very basic-distrust of very elaborate calculations of complex situations, especially where the calculations have not been checked out by full scale experiments. "
It seems tha~t the'AEC and Dr.
Weinberg were referring to the type of model that has been accepted as a basis for lowering of the reflood rate at Turkey Point.
It-is clear that this rushed,(FPL needed the code to start-up Turkey Point
- 3 with the new fuel core design, see letter Uhrig, FPL, to Eisenhut NRC, July 6, 1983, regarding Pressurized Thermal Shock.).
uncertain code was adopted more for the sake of expediency than its technical accuracy.
In fact the letter of acceptance ~for the BART code from Cecil O' Thomas ,NRC, to Mr. Rahe of Westinghouse,
& SER,Dec. 21, 1983, points to numerous uncertainties in the code and gives a conditional acceptance to the " extent specified under the limitations delineated in ^
i the report and associated NRC evaluation." !
Some of the uncertainti a contained in the BART SER are:
a)That to thethe smallfuel mixed break core.LOCA analysis did not give much weight 3
b)BART does swelling not as model have a gap by required heat transfer K.
Appendix model or cladding i c)BART was accepted without a grid spacer model because it was still being reviewed by the NRC staff.
d)Only basisone single 1 test for parameter was performed in the BART topical as a assessment.
e)The assumption of constant pressure made in BART may preclude consideration of the oscillating antigravity reflood phenomena, i f)The Flecht Seasta data comparison were from a series of tests conducted to a 15 x 15on fuel rods in a 17 x 17 assembl y and extrapolated assembly.
V. ..
( 5) .
Other uncertainties are pointed out on page 14 of the SER, which states that "many more experiments were used to develcp the old empirical carry-over rate correlations," and on page 17 which states that," additional confirmatory validation of the BART/WREFLOOD model for reflood rates less than inch /sec is required."
Other' areas which BART neglects entirely are:
a)BART does not address or compute the probability that steam generator tube failure and steam binding could stall the reflood.
b)BART does not compute the possibility or consequences of gross pressure vessel rupture.
c)BART does not take into account the aging to the system and components at Turkey Point.
d)BART has not conducted actual experiments on a mixed transitional fuel core, and instead adopts a purely hypothetical percentage for thermal hydraulic resistance.
e)BART does not analyze a small pipe break accident and the oscillating anti-gravity reflood phenomena that could stall reflooding.
In short, it seems that the BART-Al computer code approved by the NRC Staff on December 21, 1983, with the subsequent amendment being granted on December 23, 1983, was more of a rushed-up, patched-up methodology, loosely coupled with computer codes models prepared for other fuel core designs,&does not meet the requirements of 10 C.F.R. 50.46 that, The reactor core and associated coolant, control, anc protection systems be designed with appropriate margin to assure that specified acceptable fuel design limits are not exceeded during any condition
, of normal operation, including the effects of anticipated operational occurrences.
Since it appears that assumptions supporting the BART model were based on personal judgment rather than technical expertise or laboratory experiment, it is necessary that the Board carefully examine the accuracy of the underlying scientific data before approval of design changes that will lower a safety marging in this reactor. It is clear that an approach which tailors completion of a scientific study to time requirements, obviously does not foster technical accuracy or meet the requirements of 10 C.F.R. 50.46, nor is it the proper methodology to use in a field where lives and property are at risk.
m _
V. ,. -
(6)
- 3. This is in support of Intervenors' Contention (d), which states:
The proposed decrease in departure in the nucleate boiling ratio (DNBR) would significantly and adversely affect the margin of safety for the operation of the reactors. The.
restriction of the DNBR safety limit is intended to prevent overheating.of the fuel and possible cladding perforation, which would result in the release of fission products from the fuel. If the minimum allowable DNBR is reduced from 1.3 to 1.17 as proposed, this would authorize operation of the fuel much closer to the upper boundary of the nucleate boiling regime. Thus, the safety margin will be significantly reduced. Operation above the boundary of the nucleate boiling regime could result in excessive cladding temperature because of the departure .from the nucleate boiling ratio (DNB) and the resultant sharp reduction in heat transfer coefficient.
Thus, the proposed amendment will both significantly reduce the safety margin and significantly increase the probability of serious consequences from an accident.
In the Affidavit of Dr. Gordon Edwards, incorporated by reference, herein, he has given numerous reasons why the decrease in departure in the nucleate boiling ratio (DNBR) wou'ld adversely affect the margin of safety of the Turkey Point reactors. I would, however, offer a few points of embellishment.
In response to the claim in Affidavit by Edward A. Dzenis, August 8, 1984, page 9, that the change in minimum DNBR for the different correla-tions in no way implies a reduction in the safety margin of the nuclear reactor, I wish to quote recent remarks made by Robert B. Pollard of the Union of Concerned Scientists (with whom the Center has consulted on contentions (b) and (d)) nade in a Southdade Newsleader article, dated August 1, 1984, " Fuel Core Design Changes at Turkey Point Debated,"
Exhibit A:
Running the fuel at a hotter temperature then you should increases likelihood of a meltdown...It's quite clear that safety depends on the temperature of the fuel-not getting the fuel too hot or the !
cladding melts. It's clear they have cut down on safety.
l
c .
s (7)
In the same article, Demetrios_Basdekas, a nuclear safety engineer i
with the NRC states, It certainly cuts down the safety margin, reducing the amount of fuel.from which you get the power-squeezing more from less fuel density, that presents a problem...If you want to be realistic, cut down (5. percent) on power.
It is clear that the above quotes of experts in their fields, along with those of Dr. Edwards, prove that there is a genuine issue of material fact to be heard regarding the decrease in DNBR at Turkey Point, in that these changes do not meet the requirements of 10 C.F.R. Part 50 and Appendix A.
DISCUSSION 10 C.F.R. Part 50, Appendix A requires that the " reactor core, associated coolant, control, and protection of systems shall be designed with appropriate margin to assure that specified acceptable fuel design limits are not exceeded during any condition of normal operation, including the. effects of anticipated operational occurrences."
Thus, if all FPL calculations are correct, Turkey Point should be getting exc'ellent cooling and there should be no failure. Yet, in reality, this is not the case. On August 31, 1983, FPL was granted an amendment that allowed them to increase the radioactive iodine in the primary system by e t.ctor of 4. The-document states that "the licensee was informed by their fuel vendor Westinghouse on August 12, 1983, that the problem of fuel failure has been occuring elsewhere throughout the country."
Yet, nowhere in analysis performed to support the decrease in DNBR, does FPL take into account the fact that they are working with a fuel core that is already experiencing. fuel failure, which could cause
.them to bein violation of 10 C.F.R. Part 50, as a result of the
y-,
(8) decrease in DNBR in the Turkey Point Units. This is a serious omission in light of the. fact that 'there have been two releases of radioactive iodine gas from the plant on October 3, 1983, and Jan. 5, 1984.
UNCERTAINTIES IN DN3R CALCULATIONS af The DNBR calculations f ail to take into account the The fact that Turkey Point is already experiencing fuel failure.
calculations are based on an ideal situation.
b) The DNBR analysis was performed with the assumprion of a homogenous fuel core, rather than a mixed core, and may not reflect actual hydraulic' resistances between the LOPAR and OFA fuel. Clearly this does not meet the requirements and an analysis of the transitional mixed core should be performed.
c).The small LOCA analysis evaluation did not analyze the hydraulic resistance of the mixed core and the effect on peak cladding temperature.
d) No data for the 15x15 Optimized Fuel Assembly Critical Heat Flux is available.
e) The analysis was performed with the WRB-1 calculation that had been approved by Westinghouse for the 17x17 OFA and applied to the 15x15 OFA fuel. An analysis should be performed for the fuel core design in question.
It is clear that the fact that FPL is pushing the DNBR closer to the bulk boiling region means that the safety margins have been reduced and the Turkey Point reactors are in greater danger of experiencing fuel failures that could lead to a serious accident.
It is obvious that the NRC Staff has once again granted a license amendment that reduced a safety margin on the basis of uncertain,
-incomplete , and borrowed data that does not equate a sufficient analysis for the Turkey Point reactors, nor does it constitute compliance with 10 C.F.R. Part 50, Appendix A.
It is known that fuel failure and radioactive emissions are occurring at Turkey Point possibly because the fuel is being run hotter and for longer periods of time to achieve economy of operation.
I, ., .
l
.n (9) .
It-~is certain that FPL could have stayed within existing safety margins at Turkey Point if they had been willing to follow the European
. experience and derate the Turkey Point Units.
By permitting FPL utilize an unproven technology at Turkey Point and lower the DNBR based on " unreal" analysis, so that they can alleviate the pressure vessel embritt.lement problem without power penalties, the Coinmission permits FPL to experiment in the field, rather than the laboratory, and increases the chances of massive fuel failure and a serious nuclear accident at Turkey Point.
Again, I point to the words of Demetrious Basdekas in the Southdade Newsleader article that discusses the fuel core design change, I wish that we could be more prudent with the public's safety than we are. I wish we were doing more.
This Licensing Board can do more by carefully analyzing the data upon which the lowering of the DNBR at Turkey Point is based.
Further deponeth sayeth not. .
- i
?C7N Chi.w Joette Lorion State of Florida )
.) ss County of Dade )
Subscribed and sworn before me this 4th day of September, 1984. -
My commission expires:
NOTARY PUBUC STATE OF FLO11DA BO?CED THilU CEN!t:( 1* gum 4CE U'O. y ,
MY CoMMIS$loN DilRIS JULY 1619S6 /' 4 .
7 NOTA PU$LIC ()
L __
.s h&M A *
~
Bauth Babe Neum reaber _
Our 73rd Year of Service, No.110 WedfEday, Aur:st1,19M 5S.achens t.rm -
D Fud core design changes at Turkey Point debated Turkey Point By LII.Ll AN MART 1N Contenued from 4A organimauan which studies
~~
. saend in a four.p.rt oms rh. ... to.i a..iyo Choos rsca. at is. pra',a"*re,ked d with the asow it wouid aise ree plant reactor unit 3 in D.C.mbe 1983 cna gC,,fory ltyf,*, tug ybu,t HOMESTEAD
, ag,haify,yag
, (( ar m,>t
, eg
- The in unit 4 ha April 1984 w.r. pr.Ys nt!Y. taame cuenchanted witn the quences." Gilliland sa g,
""f,';l P't',""w,,,r,',J o m.o.ur.s iO .=ur. .mbrmi.m.at woula a
- 7,la,, teaer,*'=f=u;,ed m , b, ibeToesiMs%'r,ead operibon en 1972, were to last unut scrty next century But notOCCur* my supervsaor as o what g in " control Mop time" -
the reactor vessels were could or could not say." he Coolro3 rvD 4*e dropped to sald af the UC3. stop the becoming unacceptably bnttle wtuch 'h, nuclear rescuen -
and would make operating the nata the change in design, have ever reacted it sunsafe However, these> changes in you get less neutron bom- incide# could cause a minor plant a risk after 19a3 - level of embrtrtlement) It the pressure vessel can cause bardment, tnst at the same bst to tturn intorea acter g the ma)or about 20 years before the
- sil not reach it in the other problems for the plants. Lime, the houer you run the #cMdn't be shut down fast expected life span - unless Islet me of those reactors." Because less uranium fuel somethmg was done, accord. Bruns sand. would be placed in the core ~
core, the more dangerous tn enough, said Joette Lonon, eng 11 project 6ons by Nuclear N uclear safety engmeers if the dummies are used ~ case of an emergency, he director of the Center for R _ rul; tory disagree sad Nuclear Restpnsibility, an Commmsson the plant should reduce tNitCt reports. There tr u,o way to ehmt- -Running the fuel at a organizauce concemd wtut Powef. hatter temperature than you the safe operataan af nuclear But a spokesman for nye tne problem. but steps "It certataly cuts the safety Turt ry Poant sasd thei can be taken to cut down the margm reducing the volume shmid 'sacreases the likels. er plants in south sminttlemert is not a pro, radiatmn hitung the venet hood of a meltdown," said da.
of the fuel from which you get in a request for a hearmg W ra ss pie plant. m alls. sand Dematnos the power - squeetag more Poua.rtL
"%s showed the NRC et was Basdekas, nuclear safety from less fuel density. That gg s quite clear that safety and leave to intervene on tM not true," seed Tony Bruns, engmeer with the NRC in depends on the temperature amendment request for fuel presents a problem," said spokesman for Flonda Power W ashingtoo. D.C. Basdekas. of the fuel - not getung the enee design chanses an 1M snd Light Co.*s nuclear power One method which could "if you want to be realisue fuel too hot or the clacomg lanan's attorney, M artz plant. slow down this process ts La 4 cover surroundmg uranium H< wider, listed the totiowing as cut down (5 percenti en lueil melts. It's clear they
. There are theertes which redesign the fuel core. In this power," be said. ** eneof theconsnderauons.
say that the metal in the have cut down on safety," he "T* se increased fuel core case " dummies" (inert metal But utthues don't want a saA $ temferatt.res generaDy would vessets could lose its ductihty fuel rodst or parually reduce their power because it Bruns said he doesn't feel ; eurved safety margins and f reststance to fractures. but burnedeut fuel rods - are means losmg money. that poses any problem and I specifically would result In he does not feel it would have placed around the outer shells " Megawatt means make-a-erw been reachad at the of the fuel core to reduce the buck," said Basdekas they doWL expect it to unnaceeptable s*eumg or pl~ nt radiation hattmg the malls of overheat. *>owing of fuel rods. During an "I wish inat we could be The eempany did, however, the vesset. But it will not stop more prudent "It does not add to the accident, fuel rod s*enhng with the danger, it just maugates the mak7(%wges 6n the fuel core the radiation, only decrease due to higher temperatures desi gn, the amount, sa6d Basdea as.
pubtsc's safety than we are. l. neL1ron bombardment. It displaces Cooling water and mish *e *ere doing more,. ,
"We Lnew Pt would be a "It wtu not reverse the Basdetas sald. extends the life of the plant," i tmpedes Insertlee of contro!
problem down the road." damage " he said. "Is it But FPL does not expect samt Bruns, adding that no* rods by the physical phenom-Bruns said the plant can reach its hcens-sufflesent? In my opinion, the core to overheat running ed infettme of 40 years - trem enon of increased size. This The ces fuel design no " at full power said Bruns. gg 2to2012.
I coul81 result tB a significant changes made at the plant The new fuel core design "When you rearrange the Joe Gilhland, spokesman l #
Increase to the possibility reactor unit 3 in December changes at Turkey Pomt fuel core. the fuel doesn't for the NRC in Acanta, Ga.,
, and/or increase of an acci-1983 End in un6t 41a Aprnt 19s4 include using the anert fuel burfi as t.ot." Bruns saHL f deet" wirs prevenuve measqres to rods Bruns said with their The change in design will strees with Basdekas about Bruns said that control ensure troorttuement would new design embrittlement us!! the new design not eluninat- drop time is not a problem, if mcrease the heat in certain ing the prob 6em. He does, not occur. The new design not occur. parts of the reactor core, said however, disagree with the n was the NRC would not etrnges keep the radeoactive "It totally ehmmates the Bob Pollard. nuclear safety have given them clearance danger.
neutrona from hittmg the problem as far as Turkey engmeer with the Union of sor the design changes.
wr.Ils. he said. Pomt es concerned." Bruns concerned Scient sts tUC5s tr. ,
"We feel tt is under control "I <lon't thmk me would said. W ashmaton, D C, a non-pactit See TUR K EY POINT,3A We de't feel n poses a
~ , @m," Bruns said.
Which would Pollard con-sider preferable, une the new destgn and run at less power er continue with fuD power?
"The preferable is netther,"
\ Peuare salat T
"That is when the NRC doesn't do its . There's ne excuse to cut the safety margin when R doesn't etop j the problem, only shows it down," he saleL In the Oct it tasue of the New York Times U.S. Rep Edw ard J. M arkey w as trusted as saying that there to a cozmens between the NRC and the utntues wtuch makes i them locapable of watching over the nuclear industry. The l
i NRC has he "a la y Fratber 'tha:t a wat Ntarkey sate Giltuand argun w:tn thw cftartes Every plant hu 'two or three 11 censed inspectors working there fu1141me. Tte
~
document room of the contalas correspoe h between the NRC and utscos which shews that "we're ready nitplet.ing at sur own expense," he said "Last year we had L5e tarrest amount of civil penalues in our htstory,"
Gumand sait coresee ww .re seene maciee-Nos to peree