ML19327A590
| ML19327A590 | |
| Person / Time | |
|---|---|
| Site: | Three Mile Island  |
| Issue date: | 07/31/1980 |
| From: | Weiss E SHELDON, HARMON & WEISS, UNION OF CONCERNED SCIENTISTS |
| To: | Atomic Safety and Licensing Board Panel |
| Shared Package | |
| ML19327A583 | List: |
| References | |
| ISSUANCES-SP, NUDOCS 8008060428 | |
| Download: ML19327A590 (15) | |
Text
_ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
, .,,L
=
UNITED STATES OF AMERICA [ .' .
NUCLEAR REGULATORY COMMISSION v~
~. . , . - - v'sy;.
%l *-
7 BEFORE THE ATOMIC SAFETY AND LICENSING BOA" '
C -
- ~cQ p &..
-4 .
, . - -e-r f.,
u- . j,
- ', D' -
- l:;) ,
In the Matter of ) 'c;j l ..
) s- \p METROPOLITAN EDISON COMPANY, et al., ) Docket No. 50-289
) (Restart)
(Three Mile Island Nuclear Station, )
Unit No. 1) )
)
UNION OF CONCERNED SCIENTISTS' REVIEW OF CONTESTIONS The Board directed all intervenors to review their contentions with a view toward determining whether certain should be voluntarily f
withdrawn. UCS's review has indicated that our resources are insuf-ficient to adequately pursue all of the contentions admitted by the Board. With the Board's permission, UCS will therefore withdraw Contentions #6, 8 and 12. We stress that considerations of finan-cial and manpower resources dictate tne withdrawal of these conten-tions; it is our continuing _ review that they pose serious safety questions and we urge the Board to adopt them as Board issues.
Contention #6 is as follows:
- 6. Reactor coolant system relief and safety valves form part of the reactor coolant system pressure boundary. Appropriate qualification testing has not been done to verify the capability of these valves to function during normal, transient and accident condi- _ . .
tions. In the absence of such testing and verification, compliance with GDC 1, 14, 15 and 30 cannot be found and public health and safety is endangered.
8008060[k
' =
It is established that the failure of a power-operated relief valve to close was a najor contributor to the TMI accident. The s taff has noted that"[t]his and other operating experience raise a significant question about the performance qualification of two types of valves in the primary coolant boundary; relief and safety valves."1/ The importance of proper operation of these valves is explained at pages A A-7 o f NUREG-0 5 78. (Attached as " Appendix A") It is noted therein that proper operation of relief and safety valves is
" vital for conformance" with the basic design criteria for nuclear power plants. (Id. at A-6 ) The report goes in to explain that these valves are currently qualified (at most) for flow of saturated s team, despite tne fact that some transients and accidents, as well as " alternative core-cooling methods" can result in solid-water or s team-water flow through the valves.
Therefore, the s taf f calls for qualification of the valves "under expected operating conditions, which would include solid-water and two-phase flow conditions." (Id)
Up to this point, we have no disagreement with the staff description of the problem and its significance. However, the staff proposes to permit TMI-l to go back into operation without even determining the criteria or plan for performance testing of the valves, much less review of test results. Task II. D of the Action Plan.
1/ NUREG-0 5 78, "TMI-2 Lessons Learned Task Force Status Report and Short-Term Recommendations ," July 1, 19 79, p. A-7. (hereinafter "NUREG-0 5 78 " )
4 .
e i
The licensee has joined in an EPRI/NSAC generic program
- for valve testing. (SER. p. C8-11) Thusfar the industry group has submitted a test program, which NRC has not yet reviewed.
(Id. See also , - Action Plan , May, 1980, 11. D-1). The staff states that "[p3 reliminary discussions with EPRI also indicate
- that meeting the clarified requirements of NUREG-0578 is feasible."
(SER, p. C8-ll, emphasis added. ) To say that meeting the require-ments is " feasible" is f ar from concluding that they are or even will be met. Moreover, Task 11. D of the Action Plan indicates that the staff is currently unable to show compliance with General Design criteria 1, 4, 15 and 30 and will not be in a position to establish such compliance until some unspecified time after the 4
test program is completed - now called for by July 1, 1981.2/
On the basis of these facts, UCS believes that there is no rational basis for finding reasonable assurance that the plant can resume operation safely. The improper operation of relief -
I
and safety valves directly threatens safety: "The failure of I one or more of~these valves to close results in a cirect viola-tion of the reactor coolant system pressure boundary integrity."
' - (NUREG-0 5 78, p. A-6) There is a history of such valve failures daat includes the TMI accident and other instances. At present, the valves are unqualified for expected plant conditions. Indeed, it is not even known "whether these past instances of improper operation resulted from inadequate qualification of the valve or from a basic unreliability of the valve design." (Id. at A-7) 2/ Affidavit of Robert D. Pollard, June 26, 1980 p. 2. A copy is Ittached and marked " Appendix B."
l
~
Current NRC regulations are not met under these circumstances, including GDC 1, 4, 15 and 30. We urge the Board to adopt this contention as its own.
UCS Contention 8 is as follows:
- 8. 10 CFR 50.46 requires analysis of ECCS performance "for a number of postulated loss-of-coolant accidents of different sizes, locations, and otner properties sufficient ot provide assurance that the entire spectrum of postulated loss-of-coolant accidents is covered." For the spectrum of LOCA's, specific parameters are not to be exceeded. At TMI, certain of these were exceeded.
For example, the peak cladding temperature exceeded 2200 fahrenheit ( 50.46 (b)(1)) , and more than 1% of the cladding reacted with water or steam to produce hyrdrogen ( 50.46 (b)( 3 )) . The measures proposed by the staff address primarily the very specific case of a stuck-open power operated relief valve. How-ever, any other small LOCA coula leac to the same consequences. Additional analyses to snow that there is adequate protection for the entire spec-trum of small break locations have not been per-formed. Therefore, there is no basis for finding compliance with 10 CFR 50.46 and GDC 35. None of the corrective actions to date have fully addressed ..
the demonstrated inadequacy of protection against small-LOCA's.
In response to a Licensee interrogatory, UCS reviewed documents submitted by the licensee presumably in an effort 4
to meet short-term item 1.d of the Commission's Order of August 9, 1979, to wit: " Complete analyses of potential small breaks and develop and implement operating instructions to, define operator action." Our response to that interrogatory, dated April 28, 1980, which critiques the material submitted is attached and marked " Appendix C."
We will not here repeat all of the points contained therein.
In summary, the analyses suffer generally from the same deficien-cies as pre-TMI small break analyses. In fact, many of them are pr e-TMI . Assumptions with regard to the availability of equip-ment to function after an accident do no t distinguish between 4
safety and non-safety grade equipment. Thus, many of the analyses rely on the operation of systems and components not classified as safety-grade (and therefore not designed, procured and qualified to the safety-grade standards) to mitigate the accidents and transients considered. This is a violation of a basic precept of regulatory policy that, in evaluating the adegacy of protec-tion-against LOCA's, the applicant is not permittec to rely on the availucility of non-safety grade equipment.
In addition, the analyses improperly rely in many instances on operator action to initiate or control protective functions, despite NRC rules requiring such functions to be automatically initiated. A number of examples are discussed in the attached UCS answer to the licensee's interrogatory.
UCS believes that conformance with the Board Order and reasonable assurance of safe operation require, at a minimum,
4 small-break analyses under the following conditons:
- 1. Equipment which does not meet NRC regulations for s tructures , systems, and components.
important to safety should be assumed to fail or to function depending upon which assumption yields the most adverse results I 1
on core cooling capability.
- 2. No operation action is credited. Or, in the alternative (which we do not believe is acceptable), at least the analyses should consider the effects of operator errors of commission and omission, such as failure to initiate transfer of LPI suction from the borated water storage tank to the contain-ment sump or too early a transfer, when adequate prositive suction head is unavalla-ble.
- 3. Demons trate that the leak rate from all systems outside containment and the racia-tion shielding for those systems is acceptable assuming fuel failing in excess of 10 CFR 50.46.
We urge the Board to adopt this crucial contention and to
. retain independent experts to review the record and present testimony.
UCS Contention No. 12 is as follows: _.
L._
t
- 12. The accident demonstrated that the severity of the environment in which equipment important to safety must operate was under-estimated and that equipment previously deemed to be environmentally qualified failed. One example was the pressurizer level ins trument.
The environmental qualification of safety-related equipment at TMI is deficient in three
?
respects : 1) the parameters of the relevant accident environment have not been identified
- 2) the length of time the equipment must operate in the environment has been underesti-mated and 3) the methods used to qualify the equipment are not adequate to give reasonable assurances that the equipment will remain operable. TMI-l should not be permitted to resume operation until all safety-related equipment has been demonstrated to be quali-fied to operate as required by GDC 4. The
- criteria for determining qualification should be those set forth in Regulatory Guide 1.89 or equivalent.
This contention was limited by the Board to equipment witnin the containment and auxiliary buildings.
NRC regulations require that " structures, systems anc compo- ,,
nets important to safety" must be qualified to operate under L
expected plant conditions, including of course, the post-accident environment (GDC 4). It is not contestea that, during the TMI accident, equipment previously deemed to be qualified
'during the original NRC review of TMI-2 failed. One pertinent example was the pressurizer level instruments. This demon-strates a clear need to thoroughly review the qualification program for such equipment.
UCS has discovered certain facts through discovery which show that the status of equipment qualification is far more muddled and problematic for TMI-l than for TMI-2. A copy of the pertinent portion of a deposition which UCS took of the licensee and its contractors is attached. It shows that the review of environmental qualification for TMI-l was performed in accordance with a 12 year old standard which fails to approach the specificity and strictness of current require-l
~
ments and that Ode licensee had no present intention of re-assessing the qualification of safety-related equipment prior to restart of Unit 1.
The importance of this entire issue is highlignted by the Commission's recent Memorandum and Order In the Matter of Petition for Emercencv and Remedial Action, CLI-8 0 - 21, May 27, 1980. A copy is attached. The Commission found that older standards for qualifying safety equipment are inadequate and i
l required all licensees to demonstrate compliance with a stricter f
standard by June 30, 1982. However, this does not close the matter. In addition, the Commission strongly reprimanded licen-sees for failure to address this issue fully and promptly and
'a *
- noted that the staff's review to date has disclosed the shock-ing fact-that .
[ a]lmos t none of the equipment as yet examined meets all aspects of the DOR guide-lines which include the areas wnich any qua-lification judgment must address. Deviations from the guidelines include such things as an inadequate test sequence where not all of the service conditions were addressed, incomplete documentation of tests performed, no considera-tion given to aging and the fact tnat the com-ponent installed in the plant is not identical to the component tested. . . CLI-80-21, S1.op.
at 10.
4 Therefore, the staff was directed to continue a "high priority" review of licensees' environmental qualification documentation, providing bi-monthly progress reports and ordering the shutdown of plants to replace equipment when necessary. (ld. at 12) The Commission has recognized that a committment to be in compliance with strict environmental qualification guidelines by 1982 is not by itself necessarily enough to ensure safety in the interim. The Commission speci-fically stated "These deadlines, however, do not excuse a license from the obligation to replace inacequate equipment promp tly. " (Id) Licensees have not been given carte blanche to operate with unqualified equipment for two years.
The status of the TMI qualification program is unclear.
Our latest response from the staff is that they "will conduct a reevaluation of the adequacy of equipment qualification for TMI-1 in addressina the licensees response to IE Bulletin 79-01B.3/ Resolution of this issue is necessary for a finding I
1 1
3/ NRC Staff Response to Union of Concerned Scientists First and l Second set of Interrogatories, April 10, 1980, Question 108.
l l
c e
- * +
that the~ plant can be safely operated. UCS does not have the resources to analyze.the data that must form the basis for.its resolution. Again,'we urge the Board to acopt this contention and to retain independent expertise.
Respectfully submitted,
/ .
-,i '-r '
u a
-/. ~
Ellyn R. Weis's' HARMON & WEISS 1725 I Street, N.W.
Suite 506 Washington, D.C. 20006 DATED: July 31, 1980 e
a .
4 l
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION {
l BEFORE THE ATOMIC SAFETY AND LICENSING BOARD o :
)
In the Matter of )
)
METROPOLITAN EDISON ) Docket No. 50-289 COMPANY, et al., ) (Restart)
)
(Three Mile Island )
Nuclear Station, Unit )
No. 1) )'
)
CERTIFICATE OF SERVICE I hereby certify that copies of the " Union of Concerned Scientists' Review of Contentions," have been hand-delivered and mailed postage pre-paid this 31st day of July, 1980, to the following parties:
- Secretary of the Commission Dr. Linda W. Little ATTN: Chief, Dccketing & Service 5000 Hermitage Drive Section Raleigh, North Carolina 2761T U.S. Nuclear Regulatory Commission Washington, D.C. 20555
- George F. Trowbridge, Esq.
Shaw, Pittman, Potts &
- Ivan W. Smith, Esquire Trowbridge Atomic Safety & Licensing Board 1800 "M" Street, N.W.
Panel Washington, D.C. 20006 U.S. Nuclear Regulatory Commission Washington, D.C. 20555
- James Tourtellotte, Esq.
Office of the Executive Dr. Walter H. Jordan Legal Director 881 W. Outer Drive U.S. Nuclear Regulatory Oak Ridge, Tennessee 37830 Commission Washington, D.C. 20555 i
- hand-delivered
s .
l , ,,.
h[ .
1 NRR Lessons Learned Task Force l Short-Term Recommendations 5
- TITLE: Performance Testing for BWR and PWR Relief and Safety
,' Valves (Section 2.1.2)
- 1. INTRODUCTION i
3 General Design Criteria 14,15, and 30 of Appendix A to 10 CFR $0 recuire that the reactor coolant pressure boundary be designed, fabricated, and erected to ;
- the highest quality standards and be tested to ensure an extremely low proba- l trility of abnormal leakage, rapidly propagating failure, and gross rupture. I These criteria also require that the design conditions of the reactor coolant ;
boundary not be exceeded during any condition of normal operation, including !
anticipated operational occurrences.
Proper operation of reactor coolant system relief and safety valves is vital !
for conformance to these design criteria. The inability of a sufficient number 1 of these valves to open could lead to a violation of the integrity of the ;
reactor coolant system pressure boundary. The failure of one or more of these
~
valves to close results in a direct violation of the reactor coolant system l
, pressure boundary integrity.
When the reactor coolant system relief and safety valves open, the flow through these valves is normally saturated steam. Some reactor coolant system transients and accidents as well as alternate cors-cooling methods can result
- in solid-water or two phase steam-water flow through these valves. Present qualification requirements for these valves include only flow under saturated steam conditions.
^
The purpase of this recommendation is to require qualification of relief and safety valves under expected operating conditions, which would include solid-water and two phase flow conditions.
- 2. DISCUSSION The reactor coolant system relief and safety valves are connected to the pressurizer steam space on PWRs and to the main steam line on BWRs.
1 On PWRs, transients and accidents that result in increasing reactor coolant l
- system temperatures can cause an. expansion of the coolant volume in the l reactor coolant system so that the pres.surizer fills with water. As the system pressure increases, two phase and solid-water flow can occur through the reactor coolant system relief and safety valves.
On BWRs, transients or accidents requiring operation of the high pressure ~
coolant injection system or operation of the reactor core isolation cooling system can result in two phase or solid-water flow through the relief and safety valves if the reactor vessel level instrumentation malfunctions.
l In addition, on both PWRs and BWRs, certain alternative core cooling methods l require coolant injection with ECCS systems and coolant discharge through A-6
5 y .
I relief and safety valves. These cooling methods may result in two phase or h solid-water flow through the relief and safety valves.
g C
i Solid-water or two phase flow through the relief and safety valves can greatly E increase the dynamic forces on valve internals, piping, and supports over
$ those that would be expected from saturated steam flow conditions. Present j ASME qualification requirements for safety valves include only flow under saturated steam conditions. Because the safety analyses have not given credit pL for the pressure-relief capability of the power-operated relief valves, the ASME Code also does not address qualification requirements for these valves.
To date, there have been a number of instances of improper operation of relief I and safety valves. These' examples include valves opening below set pressure, valves opening above set pressure or failure to open, and valves failing to
[s reseat when open. The failure of the power-operated relief valve to reseat ,
was a significant contributor to the TMI-2 sequence of events.
- It is not clear whether these past instances of improper operation resulted 4
$ from inadequate qualification of the valve or from a basic unreliability of the valve design.
f ,
j Appropriate qualification testing of the relief and safety valves can verify the capability of these valves to function under the required conditions, a
thereby minimizing the possibility of multiple common-mode failure of these l
e valves due to challenges from conditions for which the valves are not qualified'.
Qualification testing will also provide some of the information necessary for ,
l assessing the basic reliability of the valve design since failures or successes
},
a of qualified valves will be a partial indication of valve reliability. .
f Current valve test facilities may have to be modified or expanded to test
[ - valves under various flow regimes since two phase slug flow and solid-water flow regimes will require higher mass flow rates and can result in greater
' dynamic forces. The time period for completion of this qualification testing s
has been chosen to allow for modification or expansion'of the test facilities.
The extended time for completion of this qualification testing is considered appropriate since this testing is considered to be confirmatory of valve performance capability.
I It should be noted that this requirement for qualification testing does not
! include testing under ATWS conditions at this time. Analyses of ATWS events j have shown that the pressurizer relief and safety valves could discharge two-phase and'subcooled water at pressures in the range of 2800 psi to 4800 psi 6 and at temperatures in the range of 650*F to 700*F. It is possible that the I f t,nal resolution of ATWS in- PWRs (expected in calendar year 1980) would permit
! some plants to reach a peak pressure of 3800 psi subject to showing that the
! integrity of the primary' coolant systems is maintained. It may be prudent, therefore, that test facility modifications include the capability of testing during ATWS conditions since it is likely that adequacy of any ATWS solution would depend on the verification of acceptable valve behavior.
1 A-7 b ._
.g:mmmemo vu-;m.o,,,um,_,.,_ _
t t
t k- 3. POSITION .
Pressurized water reactor and boiling water reac valves accidents.
under expected operating conditions for design s
operating conditions through the use of analyses of accidents The and operational occurrences referenced in Regulator forces on the safety and relief valves are maximized. Reactor the highest predicted by conventional safety analysis of associated control circuitry piping and supports as well as the valves themselves.
w.
O A-8
.