ML20080A562

Raises Several Issues Re Gpu Request for Commission Approval of Facility Restart.Issues Include Util Mgt Competency & Integrity

ML20080A562
Person / Time
Site:	Crane
Issue date:	12/06/1983
From:	Udall M HOUSE OF REP., INTERIOR & INSULAR AFFAIRS
To:	Palladino N NRC COMMISSION (OCM)
Shared Package
ML20080A548	List: ML20080A555 ML20080A562
References
NUDOCS 8402060120
Download: ML20080A562 (3)

Text

,

l'

~.

1.

asusTv senesen cos casss sTANtEY SCCWLt STAFF DRtCTCR Monat$ E. UDALL. AALI, CMAIRMAN AND CovN$tL Q'[

"g("""

ROY JONE$

^ " ' " " ' * ' * " ' " "

%ilL'= :ll=."' ~

==,5;.a* *^=-

COMMITTEE ON INTERIOR Ed 6 E:36,,.

AND INSULAR AFFAIRS E'IE"=~5n U.S. HOUSE OF REPRESENTATIVES

$"[$n$$'u"sn

' =,,,

maa's y,s* a,=y p -

g,haana 6m WASH.NGTON, D.C. 20515 0" " ""acO*-

"."."J'A 0."..". "'O.

m gtg,*4,,_

December 6, 1983 "n!?"'S".D C'.*EtO...

% l l".'=,',ll'.'M 0 0"$' C,."'

The Honorable Nunzio Palladino-Chairman United States Nuclear Regulatory Commission Washington, D.C.

20555 Dear M'r.

Chairman:

This regards several issues relatied t.o consideration of General Public Utilities' request that the Commission approve restart of Three Mile Island Unit 1.

The first issue relates to the Coinmission's current thinking on the need for holding and comple~ ting public hearings on management competence and integrity issues prior to any decision authorizing the restart of Unit 1.

On this subject, you said in your July 14, 1983 letter"to me that:

I would note that by memorandum dated May 27, 1983 the Comraission requested the NRC's Office of Investigations to conduct an investigation into the allegations of leak rate falsi. fica-t i ori.

The Commission does not intend ~to make a decision on TMI-l management. competency until the relevant portions of that investigation are complete. (emphasis added)

In light of NRC staff's recommendatibn yesterday that TMI-l,

be allowed to operate at 25% power prior to completion of the Office of Investigations reviews of competence'and integrity of GPU, should I infer that the Commission policy has changed from that.which was expressed to me in your July 14 letter?

On a related matter, I am aware of, and sympathetic with, sentiments expressed by Senator Specter and others to the effect tha't restart should not be allowed until the hearings are complete.

As Chairman of the Cornmittee having pritaa ry jurisdiction in the House over nuclear regulation, I do not 8402060120 840125 PDR ADOCK 05000289 U

PDR

--7--.

7,

,+wg w&w-w a

--"-e7 n-ie--

t>w wwes-- - -e

1 f,.

• e,

-2 The Honorable Nunzio Palladino December,6, 1983 believe it appropriate to express a personal view as to-when

. restart should occur.

It is, however,. appropriate and vital when questions of integrity and competence are involved (as is the case here) that the Commission take account of the views of the elected representatives of Pennsylvania.

Whatever the Commission decides in this matter should be accompanied by a full explanation of the manner in which the Commission took account of the integrity, competence and hardware issues which have been enumerated in the NRC staff studies and by intervenor groups.

Among such issues are the falsification of leak rate calculations, and the November 7,.

1983 federal grand jury indictment ~of=MetropolitanIEdistoT "Of..

particular concern is how the-Commission-weighed the r

implications of this indictment with respect to management

~

participation in, or awareness of, the leak rate falsification activities.

Finally, I am concerned about the Commision's potential use of the so-called "Sholly Provision" to grant an immediately effective approval of Unit 1 steam generator repairs prior to the completion of requested public hearings on the health and i

safety significance of those repairs.

As you know, I had qualms about the Sholly provision when it was proposed; I supported it with the understanding that it would be used carefully, and only for those license amendments which clearly pose no significant hazards consideration.

As the Commission is aware, both.the extent of the damage to ynit l's steam generators, and the means of repairing that damage is unprecedented.

Moreover, there is no serious question that the safe operation of steam generators is integral to the safe operation of a nuclear plant.

I cannot understand, therefore, how TMI's steam generators can be recommissioned without a public accounting of why their use poses no significant hazard to the health and safety of the public.

I am somewhat surprised to find that the staff has made a "no I

significant hazards" determination in the case of the TMI-l l

steam generators.

The facts available to me do not present the sort of circumstances I had in mind to trigger the Sholly provision when I supported that legislation (i.e. non-safety related, routine, license amendments so that the agency resources could be devoted to significant matters of public concern). I'm certain, members of the Subcommittee and the public would appreciate a full explanation by the Commission as to why the steam generator questions do not constitute "significant hazard" and thus can be considered under the Sholly provision if it is invoked.

(

l l

4 *.

j 0

. The Honorable Nunzio Palladino Decembe r 6, 1983 I appreciate your prompt attention to this matter, and look forward to the Commission's response.

As always, in addition to the collegir.1 view of the Commission on each of these questions, I welcome separate and dif fering views of individual commissioners.

Sincerely, e

h

,,%#)

'D Y

^

-MORRIS [UDALL Chairman e

9 9

Y

UCS 12/9/83 UNITED STATE OF AMERICA NUCLEAR REGULA'IVRY C0ffiISSION BEFORE THE COMMISSION In the Matter of

)

)

METROPOLITAN EDISON CDMPANY

)

Docket No. 50-289

)

(Restart)

(Three Mile Island Nuclear

)

Station, Unit No. 1)

)

UNION OF CONCERNED SCIENTISTS RESPONSE TO GPU LETTER OF DECEMBER 6, 1983, REGAPDING EMERGENCY FEEDWATER FLOW INSTRUMEN1ATION Intr.oduction OnDecember6,3983,counselforGPUtransmittedto'youadocumentwhichit characterized as "potentially relevant.and material to matters under adjudication in the plant design and procedures phase" of the m1-1 restart proceeding.

The doc ment is a letter frm H. D. Hukill, Director of WI-1, to the NRC, dated November 23, 1983.

The letter discloses that the emergency feedwater flow instr ments installed at 'IMI-l to ceply with t short-term '.,

-p - lessons learned" requirements do not meet NRC!s criteria.or.33PU's conmitment ir.

.~

the restart proceeding.

In essence, GPU requests an exmption frm those requirements, although its submission obscures this, as we will describe below.

There is no justification for granting such an exmption and UCS therefore urges that. you not permit the plant to operate without accurate feedwater flow instrmentation.

deMgg%

~

' Background

':he NRC's mI-2 Iassons Learned Task Force noted that "the need for an emergency feedwater systm of high reliability is a clear lesson learned from the MI-2 accident.1/

The Task Force recomnended, among other things, that safety-grade indication of emergency feedwater flow be provided for each steam generator and that <mch instrumentation be installed by January 1,1980. ! This requirement was later clarified to require, for m1-1 and other Babcock and Wilcox plants, two mergency feedwater flowrai.e indicators for each steam generator.

The implementation schedule was extended such that final design information was to be subnitted to NRC by January 1, 1981, and installation cmpleted by July 1, 1981.3/

During the restart hearing, the NRC Staff testified that: 1) each instrument should ha.re "an accuracy on the order of +10%," 2) the licensee had comnitted to installing two safety grade sonic flow devices on each of the two energency feedwater pipes, and 3) the licensee indicated that these new flow devices would have "an accuracy of better than +5%."Y fne Staff concluded, on that basis, that MI-l was in empliance with all the short-term requirements applicable tn the mergency feedwater flow indicators, subject only to the submittal of environmental qualification certification.5/

In a letter dated May 24, 1983, GPU advised the NRC Staff that two of the four installed emergency feedwater sonic flow devices would be replaced with lf NUREG-0578, "mI-2 Lessons Learned Task Force Status Report and Short-Term Recomnendations," July 197.9, p. 10.

2/

Id., pp. 11, A-32, and B-2.

_3/

NUREG-0737, Clarification of WI Action Plan Requirements," Novenber 1980, pp. II.E.1.2-4,

-5.

)-

4/

Staff Ex. 1,.NUREG-0680, " m I-l Restart," June 1980, p. C8-39.

5/

Id., p. C8-40.

See also NUREG-0680, Supp. No. 3, April 1981, pp. 38-39.

differential pressure transmitters.

On August 25, 1983, GPU notified the NRC Staff that further testing had shown the remaining scnic flow devices to be unsatisfactory and that, - by restart, all of the sonic flow devices would be replaced with differential pressure transmitters.

In a letter dated September 22, 1983 (copy attached), the NRC Staff informed GPU that the change to differential pressure transmitters was acceptable to meet the short-term lessons learned requirements pertaining to mergency feedwater flow indication.6/

Discussion In its November 23, 1983 letter, GPU discloses that its August 25, 1983, assurance that the differential pressure instrumcats for measuring mergency feedwater flow "are reliable and accurate and are designed to monitor the full range of systs flow requirements," is incorrect.

Tests have disclosed that with su rgency feedwater flow less than 100 gpn, oscillations cause the indicated flow to be inaccurate by more than 10%.

(GPU does not say hcw far outside ' the +10% accuracy requirement the flow indication lies.)

N,vertheless, GPU concludes that the emergency feedwater flow systm'as currently installed and tested at 2I-1 "is acceptable and meets the requirements of NUREG 0737 and our comnitment as reflected in the Partial Initial Decision of December 14, 1981."

6f We note the sharp contrast between the Staff's letter approving the differential pressure instrtunents and its earlier approval of the sonic flow devices in Nti?G-0680, p. C8-39, 40.

We earlier approval describes the sonic flow devices in great detail, which creates the impression that the Staff reviewed the design. We recent approval of the differential pressure transmitters provides neither a description of the design nor a basis for the Staff's approval. However, this contrast may not be significant because the earlier Staff testimony in NUREG-0680 is little more than a rewording of the material supplied by GPU. See Lic. Ex. 1, Report-in Response to NiC Staff Recomnended Requirments for Restart of Three Mile Island Nuclear Station Unit 1, p. 2.1-23.

Paragraph 1029 of the Licensing Board's December 14, 1981 Partial Initial Decision is as follows:-

The original EEW system design did not have any provision for indication in the control rom of emergency feedwater flow.

Safety-grade, redun-dant indication of EEW flow to each steam generator will be provided in the control rocm prior to restart.

Licensee has comnirted to perform a fa.1ctional test of the new ETW flow instrumentation prior to restart.

Based upon the Staff's review of the Licensee's design for providing safety-grade EEW flow indication in the control room and on the infor-motion that the flow transducers are qualified for operation in the assumed environment from a postulated main steam line break in the Intermediate Building, the Staff has concluded that Licensee is in compliance with the NURE. -0578 reconnendation, in its 2.1.7.b, for C

emergency feedwater flow indication to the steam generators.

The Staff will verify that the flow devices are installed and suitably qualified prior to restart.

(Citations craitted.]

Based on the information provided in GPU's November 23 letter, UCS concludes that the short-term lessons learned requirements for emergency feedwater flow indication have not been met at m I-1.

'Ihe basis for this conclusion follows.

GPU claims that its admittedly inaccurate emergency feedwater flow instrmentation does "not affect the functional capability of the EEW systs or the ability of the cperator to take proper action."

We agree that lack of reliable flow indication does not directly affect the functional capability of the mergency feedwater system, but it can (and during the mI-2 accident did) affect proper operator action.

During the mI-2 accident, the lack of EEW flow indication did not, in fact, affect the functional capability of the EEW system.

However, there was no EEW flow initially because two valves were closed, and the lack of EFW flow instrumentation delayed proper operator action for about 12 minutes.

GPU also claims that, if the EEW system is automatically initiated,

" accuracy of flow rate is not necessary at low flows." GPU apparently overlooks the fact that the EEW pumps are cooled by that flow.

If the EEW flow control

valves are closed, a failure of the pump recirculation flow paths could require pranpt operator action to prevent failure of the EEW pumps.

During manual control of EEW flow, GPU claims that the operator's attention would be focusal on the indication of steam generator level ard pressure, not on EEW flow indication.

Again GPU's mamory is short.

During the mI-2 accident, the operators focussad their attention on a few instruments and ignored others, to the detriment of core integrity.

Does GPU now propose to stand on its head the lesson learned that operators should not focus their attention on only a few instruments?

GPU's claim that the operators do not need to use EFW flow indication to control steam generator level also directly contradicts the lessons learned requirements. The requirement to install two EEW flow indicators for each steam generator in B&W plants like mI-l stemned fran the need "to provide the capability in the control room to ascertain the actual performance" of the emergency feedwater system.E This requirsment was relaxed, for Westinghouse and Combustion Engineering plants, to the extent that only one EFW flow indicator for each steam generator was required "for 'PWRs with U-tube steam generators because flow indication is of secondary importance in assuring steam generator coolirg capability tor steam generators of this design."

Because TMI-l uses a onca-through steam generator design, EEW flow is of primary impor ance.

c GPU's proposal also ignores the experience of the mI-2 accident in another respect.

Durirg the W I-2 accident, the operators ignored indications of extremely high temperatures in the core because they knew the instruments were not safety grade and thus potentially onreliable.

Now GPU proposes to operate y

NUREG-0737, p. II.E.1.2-4.

8/

Id.

B 8

TMI-l with. unreliable EFW flow indication and " assist the operators in understanding how the EEW flow devices are expected to perform under various EEW flow conditions,"

i.e., inaccurately.

Conclusion The'short-term lessons learned of NUREG-0737 specifically require for B&W

' reactors precisely what TMI-l does not have: emergency feedwater flow instrumentation Imeting strict, detailed performance criteria to ensure that operators can rely on them.

These requirements can not be met by installing faulty instruments and telling the operators' to ignore then.

'Ib even suggest such a response is disingenuous in the extreme.

Because GPU has failed in two attempts over a period of several years to design an accurate, reliable EFW flow indication system, it is reasonable to conclude that this further demonstrates the licensee's technical incompetence.

Since the industry-wide deadline for installing this instrumentation was July 1, 1983, we - assume that all other operating plants have met the pertinent requirements. 'Ihere is no reason to exempt 'INI-1.

'Iherefore, 'IMI-l should not be permitted to restart until it also complies wfth this "short-term" requirement.

Respectfully submittad, L

8 jfj7/

Ell IC Weiss

/f/

Gen al Counsel Robert D. Pollard Nuclear Safety Engineer Date3: Dc h r 9, 1983

+

d

1. 'j,.5s c:,

[o.,

[,' -

-.h UNITED STATES NUCLEAR REGULATORY COMMISSION E

WASHINGTON. D. C. 20555

's,

,o' Septerrbe r 22, 1983 D'acket No. 50-289 Mr. Henry O. Hukill Vice President GPU Nuclear Corporation P. O. Box 480 Middletown, Pennsylvania 17057

Dear Mr. HLkill:

By order dated August 9,1979, the Commission required, in part, that GPUN provide indication in the TMI-l Control Room of emergency feedwater system (EFW) flow to each steam generator prior to any subsequent restart.

To satisfy this requirement, GPUN committed to install two safety-grade sonic flow devices on the flow path to each steam generator to provide Control Room indication of EFW flow. This commitment was reviewed and found acceptable by the NRC staff as documented in the TMI-l Restart Safety Evaluation (NUREG-0680) pages Cl-5 and C8-39, 40 and NUxEG-0680, Supriement 3 pages 38-39.

Recently by letters dated May 24, 1983 and August 25, 1983, you advised us that, due to signal interference problems between the sonic flow devices and other unsuccessful test results, you have decided to remove the sonic flow devices and replace them with differential pressure transmitters. We have received the information provided in your submittals and, based upon that review, we conclude the changes you propose satisfy tr.e requirements of Part 2'to NUREG-0737 Action Item II.E.1.2 " Auxiliary Feedwater System Automatic Initiation and Flow Indication" and are therefore acceptable.

gincerely,

\\

'[

f s\\

k Jontr F. Stolz, Chief Op,4 rating Reactors Branch #4 ivision of Licensing cc:

See next page b

+

U

.1 UNION OF CONCERNED SCIENTISTS me c_,,er,a., m.... u.. s. no,. w. hi,,_. oc m3e. <2o2,2,s.seoo The ' State of the Nuclear Industry and the NRC:

A Critical View A presentation of the Union of Concerned Scientists to the Nuclear Regulatory Commission November 17, 1983 Eric E. Van Loon Executive Director Ellyn R. Weiss General Counsel Today marks an unusual occasion.

After more than a decade of UCS criticisms of the nuclear power program, many reports on ruclear safety, repeated appearances before Congress, years of litigation, and several debates in public forums, this is the first time that the Union of Concerned Scientists and the five NRC Commissioners have taken the opportunity to sit down together to di: cuss nuclear fundamentals.

UCS' apprecia tes this opportunity for a frank exchange of views.

Today, we wish to address briefly the condition of nuclear power in the United States, and then focus primarily on the role of the Commission and its regulatory responsibilities.

There is little doubt that nuclear power is in trouble in America today.

The basic facts are familiar to us all: the five year de facto moratorium on new orders, the cancellation of virtually esery plant ordered since 1974, the unpromising outlook for reactor sales for the rest of this decade.

And while lowered demand forecasts, soaring construction costs, and high interest ratee have doubtless contributed to this picture, something more, something particularly related to nuclear power is at work.

Since 1979, 26 coal plants have been ordered nationwide.

According to the National Coal Association, 21 new coal-fired units went into operation last year, providing 10,400 megawatts of power, compared to 3000 megawatts from new nuclear units.

These figures tell us,that utilities continue to reauire new capacity, but they aQ declining to choose nucl?ar energy to meet these needs.

The underlying reason for this direction, UCS believes, is _that nuclear power has lost the public's as well as the investors' confidence.

Two years ago this month, an NBC News / Associated Press poli signaiea a tenciamental shift in public attitudes:

26 Church 5, rect. Cambrid e. Stewohuar,i, o223ft. (617) 547-5552 Stein Office:

F

.i.,

4&,

- 4 x_.(

4

+J

-4 4*

A a.4 4m..aa.--,

. 2 t

A maj ority of the.public--56%--do not want more nuclear power plants built in the United States.This opinion.is the reverse of our findings four and a half years ago, when nearly two-thirds of the public favored new nuclear power facilities (Poll #72, 24

- November 1981).

Industry polls taken in the past six months confirm that majority opposition to nuclear power continues today.-

The American Nuclear Society's. Nuclear Report found in May.1983 that the 33% public support for nuclear power in a Cam 5 ridge Reports poll constituted "the lowest level of support nuclear power has ever had, including immediately after Three Mile Island, and represents a steady decline from overwhelming support in the early 1970's."

Why has the public become disenchanted with nuclear power?

UCS sees f

principal reasons.

First. nuclear power, unlike most other foms of q~ a_o_ur lectricity generation, entails a risk of catastrophic accidents.

Second, t_he public is justifiably concerneo aoout tne con 1 inn-n nr-e+Ma af an ramaly toxic and lona-lived radioactivEtastes which no one has yet isolated from the environment.

Third, industry, operations nave been marred by repeated instances of ineptitude that would be comic if they were not so potentially l

disastrous.

The _0fabio L,a n_ von Diueprint reversal, Zfiruner's ana s'ii di a n d ' s --

quality assurance problems, and GPU operators' cheating af ter the acciaent --

to name a ornmiamat faw -- are examples of repeated industry lanses which

' undermine the fervent assurances that in-the area of nuclear safety, nothing serious can no wrong.

The fourth factor affecting public confidence is eaually imnnetant and is more directly within your purview: the public, in our view, believes that it is no longer being protected by its covernment -- that there is no tough,'

effective cop on the nuclear beat.. A consistent oattern of Commission actions broadcasts the message, Intendt.d or not, that sa faty ic nnt this Commission's highest priorit'.

~

y

_ - While nuclear's economic problems are largely beyond your control, the pub,lic'c " fety -- and the nublic's faith -- are clearly in your bailiwick.

In order to restore a significant measure of confidence, the Commission must insist on safety first, institute measures to reinforce the cafety first" mandate within the NRC' staff, 'and act firmly toward an industry sorely in need i

l of discipline.

Only a firm and steady hand can restore the credibility necessary for you to regain public confidence and, in our view, the respect of the industry you are charged with regulating.

With this prescription on the table, let me return to a diagnosis of the problem.

In our view, the agency's shortcomings in its regulatory role have fallen rather consistently into three categories:

1 (1) The 1RC has been unable or unwilling to resolve fundamental reactor safety problems; (2) The Commission has repeatedly tolerated a slow, rel uctant or cbscuring response to known safety hazards; and,

~

(3) The Commission and its staff have been hostile to those raising p.safetyconcerns.

~

.c w--y y-m-,,y- -...

.,w.

y,,.,_,y

,w,%,,,-,,,.-,,,.,-..-,-v%,,%

,_,...,.-,,,,,.,y,,---*-T-*N'*-*v'eNe*"

y = t eN g* W 9ew-*-'""""9'-v*vvt9--7'"9

-"'8**"'**='r"

,4 c

I will present several examples of Commission actions which illustrate i

the first two points, and UCS General Counsel Ellyn Weiss will foll.ow with our views on the-handling of public participation at the NRC.

The first example we wish to highlight -- the agency's reaction to the Bingham Amendment, introduced by Congressman Jonathan Bingham after the Three Mile Island. accident -- casts doubt on how much the NRC knows about the level i

of safety at operating reactors. This amendmert to the 1980 NRC authorization ac t directed the egency to identify which operating nuclear plants meet current safety requirements, and which generic unresolved safety issues have technical solutions.

A maj ority of the then-NRC Comissioners (Ahearne, Hendrie and Kennedy) obj ected to this task, indicating that the resources required to collect the data showing which plants met Comission 1icensing requirements would strain the agency's manpower.

The_ Commission maiority i

conceded, in essence, that its staff was unable to determina which safety rejuTrements had been appliec in its licensing reviews.

Commissioners' Bradford and Gilinsky called this disclosure evidence of a " stunning disarray in the status of NRC knowledae of onerating olants..."

As 'you know, the Bingham amendment was never implemented.

Sadly, such delay and obfuscation has become a familiar pattern, and the Commission's enforcement.of safety standards has failed at crucial moments.

In recent months, we have seen two striking examples in the NRC's handling of the Indian Point emergency preparedness issue in June and the BWR pipe cracking controversy in July.

Of Indian Point, little need be. said.

A Commission majority selected the pla7t upwind from the nation's most densely populated area to announce, in effect, that it would not stand behind its emergency planning regulations.

And the BWR pipe-cracking sequence, with the Commision's quick reversal of an initial firm stance by the Staff undermined enforcement "in this and future cases. The pipe-cracking problem deserves a brief elaboration.

' Two decades ago, an Oak Ridge research report disc.losed the potential

. for cracking in stainless steel.

Stress-corrosion cracking was listed in 1975 as an unresolved safety problem slated to be resolved in 1976.

The discovery of widespread cracking in prierary and emergency piping systems in m3ny plants has brought immeciacy to the issue.

It is now apparent that current inspection techniques are grossly inadequate, and the cracks may be deeper and

- more numerous than inspections have indicated.

Our focus is directed, however, to the events of July 14th and 15th.

Harold Denton, not a frequent advocate of plant shutdowns, called on July 14th for closing five BWRs within 30 days for inspection.

The very next day, the Commission met with industry representatives and rescinded the order -- an i

4 action not likely to inspire public confidence in the agency's commitment to T safety.

Indeed, as Richard Vollmer of the Staff explained the recission to a reporter, " Industry gave them [the Commissioners] a good enough story that said [ safety)... was outweighed by the costs of downtime to the utilities."

Neither EPRI's round robin test results nor the August 9th Advisory Comittee on Reator Safeguards rebuke have moved tne agency to take the action that is closing the plants quickly to check each of them so clearly prudent thoroughly for potentially dangerous cracking.

l

.., ~..., - - - -. - - _ _ - -,. - - -, _,..,,

_4 Environmental Qualification: A Case Study in Delay A prime example of the NRC's failure to resolve a safety issue promptly is the " environmental indeed, at times, to obstruct safety progress qualification" of electrical equipment needed to mitigate an accident.

Nuclear plant safety systems must be able to function during an accident.

A regulat_ ion rannirino environmental cualification has been on the bo9s for many years.

In a 1978 ruling, the Commission acknowledged that g environmentai oualification is " fundamental to NRC regulation of nuclear power re_ac to rs. " On Octob:e 8.1982, more than four years af ter this statement, the g Statt i n fnmad the Commission that more than 80% of the safety equipment in operating -lants still had not been shown to be capable of f unctioning

$ properly w.... exposed to tha hmh enviennment caused by an accident. Equally disturbing is the fact that the NRC delayed this infomation coming before the public until last month when it was released in response to one of our Freedom of Information Act requests.

UCS first brought the environmental qualification issue before the Commission in November 1977 when we concluded that results from NRC-sponsored environmental qualification tests demonstrated the inadequacy of NRC's standards.

The day after we raised tnis issue, the NRC staff issued a statement claiming that UCS had misconstrued the safety significance of the test results and that none of the equipment which failed the tests was used in safety systems in any operating plant.

(Both these claims were later shown to be totally unfounded.)

Shortly thereafter, the NRC nonetheless asked nuclear operators to supply information concerning the environmental qualification of equipment in operating pl an ts.

In 1980. aftpr two and a half years of the utilitiat largely ignoring the NRC requests, the Commission gave the industry almost two h to demonstrate - that their safety untii June 30, 1982 more years equipment was qualifieo to operate in an accident environment.

The 5*,af f subsequently amended the 11onse of each operating plant to incorporate the 1982 deadline to ensure its enforceability.

(No licensee requested a hearing to oppose this license amendment.)

In its 1980 order, the Commission al so established a new and stricter set of qualification criteria, conceding the clear inadequacy of the prior standards.

A year later, in June 1981, a group of utilities asked the Commission to postpone the June 30, 1982 " deadline" another year on the sole ground that they could not meet it.

Without an extension, they claimed, they would have to shut down their plants.

Remarkably, the NRC staff responded to this request with a far more generous proposal - a two year extension --

and the Staff added exceptions that could, potentially, allow unlimited delay.

In j

arguing for this proposal, the Staff disclosed that its evaluation of about 20% of the equipment essential to safety in operating plants, led it to estimate that 15 to 40% of the equipment was " unqualified" and would need to be replaced.

The Staff argued. in effect, that the plants should be allowed to remain in operation until it determined which equipment was detective.

l When the June 30, 1982 deadline arrived, the Commission published an l

innediately ef fective ru_le waivino the " deadline" altogether, but provided no

opoortunity for public comment or hearings on thi sJ ti on,.

On October 8, 1982, the Staff provided the commission with a repart (SECY-82-409) which showed that its 15 to 40% estimate of unqualified. equipment had been

1) 44.6% of the equipment in operating plants has to be replaced, cptimistic:

physically modified, relocated, shielded, or tested further; and 2) 31.1% of the equipment lacked documentation to detennine whether it was environmentally qualified or needed to be replaced, modified, relocated, shielded, or tested.

Al tonether. 84.8% of _the electrical equipment relied upon to protect the pubfic in the event of an acciaent was not shown to be environmentally qiialified and only 6.6% of the equipment was fully Quali fieo. (ine remaining 7.5% of the equipment did not need to be qualified because it did not provide a safety function or was not, because of its location, exposed to the harsh accident environment.)

This status report, five years after the Comission began its efforts to showing get licensees to demonstrate the qualification of safety equipment, that only 6.6% of the safety-related electrical eauipment was a sorry statement.

However, the Commission apparently did not recoanize how sharply it contradicted the assurances given ny tne Staff in the sorino of 1982 before the Commission suspended the.1une 1982 deadline.

The imp 7ssion created by the Statt was tnat most equipment was QTiali fied cn that adequate justification for continued safe operation had been obtained for the equipment which was not shown to be cualified.

Unfortunately, the Staff neglected to tell the Commission that neither the Staff nor its contractor had independently verified the licensees' claims concerning equipment identified as environmentally qualified.

It was only afterwards that the Staff began reviewing those claims.

Apparently, it was this subsequent review which led the Staff to conclude that only 6.6% of the equipment was actually fully quali fied.

The most recent development in this tortuous saga is a June 30, 1983

[

D.C. Circuit Court of Appeals decision (in a suit filed by UCS) that the NRC had violated the Atomic Energy Act, the Administrative Prbcedure Act, and its l

own rules when it suspended the June 30, 1982 deadline withoui providing opportunity for public comment.

NRC lawyers argued to the Court that the NRC l

rule requiring an opportunity for public comment "in the manner stated in the

[ Federal Register] notice" had been met because offering no opportunity to comment was one way of regulating the manner of commenting.7he Comission's lawyers also argued that NRC's safety determination - that continued operation cualification program of the plants pending completion of the environmental was an will not present undue risk to the public health and safety insignificant " ancillary finding" or " explanatory background remark" which was not "not a part of the rule."

Of course, without such a finding the l

l Commission could not allow the plants to remain in operation.

Since the Appeals Court ruling, NRC resistance has taken still another 1

tack.

More than four months after the Court ordered the Commission to

" provide an opportunity for comment on the sufficiency of current documentation purporting to justify continued operation pending completion of environmental cualification of safety-related eauipment," no opportunity for public comment has been provided.

This is perhaps not surprising, however, since NRC attorneys told UCS after the Court ruling that even though UGS l

prevailed, NRC could exercise its discretion oy doing nothing.

' ' 4 The handling of the environmental qualification problem thus presents a glaring example of,NRC's disregard for its statutory mandate.

The Commission staff clainiv failed to protect the nublic -health and safety.

It repeatedly g yielded in the face of industry nonperfomance.

The_ aaency refused to meet the minimum legal requirements for public particii,ation.

It nas ignorea we express direction of a Federal Court.

And the Sta f f has kept important information about the safety of nuclear power plants from the public ana, at times, from the commissioners themsehe>.

The CRGR: A Safety Bottleneck?

More currently, the entire "backfitting" controversy -- and particularly the role of the Committee to Review Generic Requirements (CRGR) -- raises additional questions about the NR_C's priorities and perfomance.

Ac the Commission has acknowledged, establishmeinT of a DacKTitting rule involves "a number of broad policy questions", all of which, in our opinion, are extremely i

complex, and none of which have yet been adequately analyzed.

We appreciate the Commission's - effort to gain a better understanding of these issues by asking a number of thoughtful questions in its Advance Notice of Proposed Rulemaking.

We-will not now repeat our analyses of these issues, which we submitted in our comments of October 28.

P Although in our view an inordinate amount of time has been spent on resolving a backfitting problem that has neither been clearly defined nor proven to exist, UCS supports the concept of creating a more " rational" backfitting process which assures that requirements are well-considered and justified, and that licensees do not receive conflicting signals that could be detrimental to safety from different organizations and level s of Staff.

Ilowever, both the CRGR, and the backfitting proposals to date, aim less toward this goal and more toward establishing an irrational system in which licensees can avoid backfitting.

. Part of this irrationality derives from CRGR's use of cost-benefit l

analysis and probabilistic risk assessment (PRA).

While cost-benefit analysis has a certain superficial appeal (who is opposed to preventing wasteful expenditures?), this initial attractiveness cannot withstand close scrutiny.

The most fundamental problem is that benefits cannot be fairly quantified.

The " benefit" side of the ledger is the reduction of accident risk.

This reduction can only be measured by the use of a PRA.

However, as you know, the bottom-line results of any PRA are enormously imprecise for the most serious accidents; the uncertainty bands range to a factor of 100 or more.

This inherent lack of precision makes a quantitative balancing of costs against benefits essentially meaningless.

Cost-benefit analysis, like PRA, becomes a tool that can be manipulated to produce a preselected result.

Second, cost-benefit analyses done to date do not properly factor in the cost of all accidents.

Risk reduction is measured by the avoidance of public radiation dose, with an arbitrary monetary amount set for each man-rem.

The full economic costs of a serious accident, ranging in the billions of dollars, are not accurately incorporated.

Cgc+ WH t analysis, therefore, does not provide an objective basi _s.

for_ nuclear safety decisionmaking.

It is very likeiy to ^ e useo to forestall u

. safety improvements indefinitely. ~The Army Corps of Engineers' cost-benefit

7_

analyses, in which it juggled and rej uggl ed the figures to justify a predetermined conclusion, were legendary.

Protection of the public from nuclear accidents is too important to be relegated to the same treatment.

The legality of NRC's use of cost-benefit analvtis is also cuestionable.

The Atomic Energy Act requires the Commission to assure protection of the public health and safety, and any step necessary to fulfill that mandate must b e ta k e_n.

nRC case saw is remarkably consistent in its acceptance of the principle that costs are not a legitimate factor ie safety decisions.

Adoption of a cost-benefit approach represents a dramatic departure from the guiding principles NRC and AEC have always enunciated.

Nor can a process to assess the need for safety improvements be rational if it is not accessible to all sides in the debate. CRGR is almost completely insulated from public scrutiny, with limited access to relevant documents and only sketchy minutes of meetings.

There are no transcripts by which the public can ascertain how CRGR arrives at its decisions.

The best picture one can get from such a restricted view of CRGR is that of a black hole, into which issues fall and never emerge.

Many items forwarded to CRCR are sent back to the staff for more review and/or modifications which, although it is difficult to tell, appear to significantly weaken their effectiveness.

CRGR requires extensive, time-consu' ding c'ost-bene fi t analyses, and demands new analyses if it is not satisfiedywith the results of the first ones.

The resul t is a bottlenecks where -items sent back to the staff sometimes never resurface, with~

date in sight 'for the next CRGR review, much less a no deadline for final agency resolution.

Cl ea rly, CRGR's interest in cost-benefit analysis is' not balanced with an equally strong interest in resolving safety' problems..

+

Cases where C4G(appears to have caused long delays include:

• Regulatory Guide SC78-4, " Qualification and Acceptance Tests for Snubbers Used in Systems Important to Safety".

CRGR held 'two meetings on this topic, in March -- and again in May 1982.

CRGR asked the staff at Nuclear Regulatory Research (RES) to. analyze a number of aspects of the benefits of this Guide in light' ~of an-expensive estimate of implementation costs (supplied by utilities).

RES is still working on this item, with no date scheduled for another CRGR review.

• USI A-43 " Containment Emergency Sump Perfomance", where CRGR supported two Nuclear Reactor Regulation (NRR) proposals which would reduce requirements on future OL applicants, but disagreed with one proposal requiring utilities to perform plant-specific analyses.

At two meetings, in November and December 1982, CRGR complained that NRR's analysis of this requirement overstated the

~

benefits (which NRR claimed outweighed the costs) and understated the costs.

At the second meeting, CRGR told NRR to " review their risk reduction analysis in light of the, analysis perfomed by the Deputy Executive Director for Regional Operations and Gener_ic 'Reauirements (DEDROGR) staff with the objective of developino the most realistic assessment of averted radiological dose."

We understand that the package has gone out for comment, and NRR, in addition to considering the comments, is still doing cost-benefit work in anticipation of.another CRGR review.

s

-,-n,---

• The - Anticipated Transient Without Scrair Rule, where in November 1982 l

CRGR requested an extensive cost-benefit analysis, which RES performed before CRGR's second meeting on the rule in January -1983.

However, at the January maeting, CRGR decided that it was unable to determine the total safety benefits and costs of implementing all ' the modifications identified by the staff, and RES' agreed that - the ATWS rulemaking package would attempt to address them all.-

At ' the next meeting, af ter the Salem plant suffered two ATWS incidents, CRGR continued to look for ways in which utilities could avoid full implementation of the requirements, suggesting that, for example, a plant Ray reach 'an age where the benefits of full implementation may not justify the costs.

Four meetings later, in May 1983 CRGR was still requesting additional cost-benefit justification.

A casa where the mere prospect of the CRGR-required cost-benefit analysis seriously delayed progress is " Systems Interaction", an Unresolved

. Safety Issue which was later turned into Action Plan item II.C.3.

According to an NRC Staff member, in October 1981 approval was given by DST to a proposal for initiation of the methodology demonstration phase of the post-TMI 2

systems interaction program.

NRR's approval was needed for final selection of l

the pilot plants which would be used, but NRR took na action, and "the effort stalled at that point, apparently over concerns that developed in connection with cost-benefit estimates required for the expccted review by [CRGR] of any NRR approval action on this propcsal."

The emphasis on achieving cost-benefit advantages (to help in gaining acceptance-from CRGR), led to a proposal to combine the - systems interaction program with an already envisioned NREP/SEP i

combined review. program, to eliminate duplicative aspects of the three programs.

This delayed NRR approval of the October 1981 DST proposal again, as the new proposal was explored further.

The original proposal was eventually dropped altogether, and we understand that no approval has been given by NRR for implementation of any systems interaction methodology demonstration studies under any option.

The same Staff member concludes that

...the basic error involved was in RRAB, DST and NRR management (i) not taking a more aggressive posture with CRGR in presenting. the II.3.C related program proposal on its own merits, 'i.e. as a necessary program for timely resolution of; a USI, and (ii) not resisting ' the post-facto imposition of cost-benefit. criterion in a way - that... delayed excessively the progress of l

that necessary program." Thus, the NRC has made virtually no forward progress l

cn systems interaction, which has been an unresolved safety issue for years, and is now further away from its ultimate resolution.

Finally, we were utterly amazed at the Staff's triumphnt boast (in a Summary and Assessment of CRGR Activities To Date of February 28, 1983) that:

There has been a substantial reduction in the number of new 9eneric requirements imposed on reactor licensees since the inception of CRGR.

For Instance, the staff had projectes there would be 1900 new operating reactor licensing actions generated in FY-1982, whereas the actual number was less than 900'new actions. Most of this reductior, was in the number of multi-plant actions and can, I believe, he attributed to the existence of GR.

Thus, the RG had a major impact in the staff's ach evement of red ng the operating reac tor l

l-licensing backlog from about 5400 to 3600 actions during j-FY-1982.

l l

i

_.. _,..,. _ _. - ~,.... _. _ _ _ -.. _.,

_ _ _ _ _ _ _. _.. ~,,

9 Until then, we had been unaware tha CRGR' purpose was to clear the agency's rf,ggLa +a y decbat This by-the-num approach to sa f e ty'Tetp' 2 + 4 an. H tS, a

its attendant bias against any new reauirements, is hardly compatible with the mandate of the Atomic _ Energy Act.

^

The Commission, in its struggle to promulgate a backfitting rule, has acknowledgeo that many complex problems arise in detertnining what standards and methods should be used tc j usti fy backfits.

Yet the Commi ssion is ignoring these problems altogether in allowing Cy te operate as it does.

The Hartman Allegations: Unanswered Questions It has been over a week since indictments were handed down against the Metropolitan Edison Company for intentionally and systematically falsi fying leak rate calculations at TMI-2 for montns before the accident.

Because we are sensitive to the ex parte restrictions on discussing W merits of this catter which is before the Commission, we will not deal with the substance of the charges beyond what is now widely available publicly.

But we will raise several questions about the NRC's handling of this r..a tte r which you as Comissioners need to address because they go to the heart of your agency's effectiveness _ and credibility.

The first question is: Considering that this agency was infortned 'n tal 1979 of the facts underivino these indictments. how can it be that 41/2 yaes laTei~, NRC has not taken any action?

Indeed, these charges have not even been brought before the Licensing or Appeal Boards, nor has your own inve_synatinn_ %

been concluded.

!iew can it be that, in its only official communications to the TMI Restart l icen ti na Board on the matter. the 5taff u'iuscoten tne charges, u

failed utterly to disclose the evidence which it had in hand and, in fact, (aid i believe this is the most serious) actively misieo7he Boards and ThD public by downplaying the significance or tne cnarges?

Why_ i s there no discussion cf this issue in NRC's official accident 9 i nv e sti. gift _i n n e, either that Drepared by I & E or the so-called "Rogovin F i

Report," although the facts were bnwn to both Staff groups oL the t%e?

Why did NRC take no action whatever on the charges for almost a year",

from May 1979 until April 1980, until a television reporter discovered and Y publicized them?

How could the Staff have continued to enthusiastically endorse GPU's Canagement competence and integrity untii tne spring of 1983. een it had g

evjdence three years earlier that these charges were true?

Wha _t hacoened in the spring of 1983 to trigger the Staff's belated nf its endorsement and its riati ve te con fi rm the existence of with,d_t m l

" procedures for procedure compliar.ce" at THI?

Was it fear or oeing-embarrassed by the issuance or inoictmentse -

The answers offered to these ouestions thus far are weak and unsati sfying.

We are told that the Staff advised the Commissioners " orally" of the seriousness of the charges.

If the Commission was secretly told that

l.

s.

Hartman's charges were substantiated while at the same time the ASLB, which was adjudicating GPU's management competence on the record in public hearings, was led to believe that there were no serious questions about the utility's competence and integrity, that makes the adjudicatory process a charade.

Q aise constitutes an egregiout violation of the statutory prohibition aaainst secret, off-the-record contacts by the Staf f to the decision-makers.

It is possible that some -- even at NRC -- may not understand why the agency cares about the inteority of its licensees.

You have no. choice but to care.

Your inspectors and technical statt are-humanly capable of reviewing enly a tiny fraction of the work done on these plants during construction and operation. - A string of recent cases -- including South Texas, Zimmer, Midland and Diablo Canyon -- confirm that it is only too possible for a plant to be virtually completed with NRC approval all along, only to discover through a whistleblower or otherwise that your rules have been flouted.

In addition, hardly a week passes without an I & E Circular or Notice informing licensees of some operational problem or equipment failure, directing licensees to investigate, but requiring no response.

You trust the licensees to do what is necessary.

That is why you must be able to have total confidence in their integrity.

In the nuclear safety context, integrity cannot be separated from competence.

It is no less necessary to insist upon the highest standards from your cwn Staff.

.There are many honorable and hiahly competent people in this agjncy. They will give you what they oesieve you want.

If you wani.eifeciive andlorceful regulation, we suggest that you will need to take action in this case to hold accountable those who were responsible for this failure of enforcement.

RECOMMENDATIONS: A Question of Mindset From UCS's perspective, these examples illustrate an agency which has lo_st siaht 'of its mandate.

On the face of this recora, wnat can De done to bejin to restore confidence in the agency charged with being the public's protector, the firm cop on the nuclear beat?

Proposing a cure is difficult, since this is not a problem of rulas or statutory authority.

Instead, consistent with the central finding of the Kemeny Commission, it is a problem of mindset.

Unfortunately, the agency has a short institutional memory, and the strenoth of that insight has greatly faded in the past 4 years.

As a guiding principle, we cannot do better than to reiterate Kemeny's finding that " fundamental changes" are "necessary" in NRC's attitude, which must change from one of complacency to one Inat says nuclear Dower is oy 1ts very nature potentially aangerous, and, therefore, on~e~

Eu_s 161 O m ;" a = eF Jaa"*-dE O -^=dy

-"8'ki g +^ neamma+ maine accidente You as Commissioners may well feel and believe that changes have been major and that tough-minded vigilance is now the NRC nonn.

Whatever your intentions, however, your actions speak louder -- and convey an opposite gghgtn.

As one Commissioner has oublicly observed, senior staff which rinheri iW~titi months followina TMI to propose newer - stricter safety reouirements

- have switcned in the meantime to suggesting ways to roll back what are now

viewed as unnecessarilv nvamaalnus measures.

The Commission's decisions have set the norm-and ynor actions in recent years have not set a standard or vigilant public protection.

As critical to public confidence as the actual decisions you make is the manner in which the Staff carries out its extensive responsibilities.

The standards of integri ty and carefulness you demand of the Staff are as

- important as the standards you set for yourselves.

Arguably, they are even more important, because you depend heavily on the Staff to carry out your responsibilities.

The arcane nature of this technology, the relative brevity of the Commissioners' terms, the sheer vol ume of paperwerk that nuclear regulation entails -- all create a situation in which the Staff plays a central role in regulatory decisions.

You must depend on them for full and honest reporting to you, and unblemished integrity in dealing with the industry you regulate.

In addition to the cases already discussed -- the Staff treatment of the Ha_r tma n allegations and the handling of information about environmentar qualitication -- we could cite many exampies or statt actions which undermine ~

NRC's integrity as a regulatory agency.

The Staff civing an early draf t of the TMI Lessons Learned report to GPU and later deleted statements critical of the U+.ility; a similar early review;_or tne quality assurance stuay on viaolo p Canyon by the licensee. wi tn Ine no t i accepcing many of its proposea cnanges; the cattern of conduct described in cne noyt report en Ene Applegate allegations -- these are some of tne most pud 11CiZed DuL unquestionably not the only (and in bur view, not atypical) examples of, to put it in the vernacular, Staff "@" with the 1% it is supposed to keep at ann's length.

While this is not the proper forum to review specific instances of individual conduct, suffice it to say that Staff members heavily involved in these sorts of activities should not be rewarded with promotions, citations or financial bonuses, regardless of their other contributions to the regulatory process.

To do s'o sends a message that such conduct is accepted, even encouraaed. when it should be puni s he_d.

Your ability to influence the ~

befiavior of your large, long-entrenched Staff is not unlimited, and the l

signals you send down the ranks should never be ambiguous.

You will need to maintain an attitude of questioning skepticism as you proceed to issues on your regulatory agenda, some of which involve the central premises of nuclear regulation.

For your performanca

+n in en4' a a"h l i c_

confidence, you must avoid the old patterns of backdoor industry influence and I

resistance to outside advice. We will turn to a few of those issues now.

I Source Terms The source term issue is, unfortunately, painted by many as being far simpler tha;i it is.

One frequently hears talk of numerical reductions in the

" source term", as though the " source term" were one entity.

It is now quite clear that each accident sequence at each reactor will have its own unique I

source term that defines the quantity and chemical form of radioactive l

materials released to the environment as a result of a reactor accident.

1 I

l

12-The source term issue is described by many as having arisen from the TMI-2 accident.

While TMI kindled a new interest in these issues, source term has been debated and researched for more than ten years.

Industry spokesmen claim that the TMI accident " proved" that the release of radioactivity from a severe reactor accident had been drastically overestimated.

No such fact was proven at TMI.

What did emerge was a convincing demonstration under actual reactor accident conditions that iodine leaving the core under similar cqnditions is in m fnrm of a salt, namely cesium iodide, and nnt iodine vAa' o r.

This confirmed previous analytical work which was not credited in the WASH-1400 report due to conflicting experimental evidence.

The implication of this finding is that if the release of radioactive ma teri al s from the reactor core during an accident passes through a

" water-bounded" pathway, much of the radioactive iodine will be dissolved in water and not available for an airborne release to the environment if the containment fail s.

It is clear, however, that the TMI sequence is hardly characteristic of the spectrma of severe accident sequences which can occur in light water reactors.

The NRC's major research effort, principally conducted at Sandia, Oak Ridge, and Battelle Columbus Laboratories, has one main object: to model as realistically as possible the transport of radioactive material s from the primary system to the containment and subsequently to the environment if the containment fails.

Unfortunateiy, tnis research program includes only about~

20 accident seqtrences in five different reactors,. and it is not clear that these five reactors are representative of the entire LWR population.

Further, there are typically dozens of accident sequences that are " risk significant" at each reactor.

The uncertainties involved in accident phenomenology and the timing and mode of containment failure are considerable, and these uncertainties will greatly affect the results of state-of-the-art source term calculations such as those being perfonned in the NRC/Battelle program.

It is premature to be discussing the magnitude of p,ossible source term reductions and their implications for emergency planning and other regulatory concerns.

Rather, the Commission should evaluate its present source term research program, and reorient and expand it to permit useful comparisons of accident consequence calculations, employing the Reactor Safety Study source terms and the results from newer, s ta te-o f-the-a r t calculations of release-from-containment source terms.

We do not mean to imply here that reliable absolute risk comparisons could bc drawn; however, useful comparisons could be made on a conditional basis (i.e., given the release, what are the l

resul ts).

Only then will it be possible to determine the degree to which reactor accident risks have been overestimated, if at all, and the steps that need to be taken in the light of the new informatTdn~-

Safety Goals The Commission is mid-way through a two-year " evaluation period" in its cos. sideration of quantitative Safety Goals for commercial nuclear power pl a n +.s.

As the Commission should by now be aware, setting the Goals is much easier than convincingly demonstrating that any particular plant meets them.

This is true because the quantitative Sa fety Goal s and the art of probabilistic risk assessment (PRA) are inextricably intertwined.

Unless and

j g.

\\ O _

until PRA resul ts become significantly more reliable than they are now, quantitative Safety Goals are a diversion from more profitable (in terms of safety) pursuits.

We do not mean to say that the PRAs done to date have been worthless.

To the contrary, some significant insights into reactor risk have arisen from the~ NRC's PRA programs.

These insights fall into two categories:

(1) identification of generically applicable risk contributors, such as the interfacing LOCA for PWRs; and (2) plant-specific design and procedural flaws, such as the upper compartment drain plug situation at Sequoyah and the unreliable emergency feedwater system at Calvert Cliffs.

i We believe, however, that such insights could have been derived from a program which concentrated more on the evaluation of plant design using fault tree and event tree methods and other more traditional engineering analyses, as _ opposed to the quantitative analysis performed in a PRA.

In so doing, much of the controversy about assumptions, data base, and similar issues would be reduced and the costs of the undertaking.would be somewhat lower, allowing Gre reactors to be analyzed.

A Without reliable PRA results, quantitative Safety Goals have no meaning because it is impossible to assess compliance.

Worse, the results can be manipulated by a judicious choice of assimptions, - data base, and models to assure the compliance of any plant with the Safety Goals.

The Comission should not delude itself that the latter cannot be done; the nuclear community knows enough about PRA to make this sort of manipulation quite feasible.

We therefore urge the Commission to embark on an alternative program whereby qualitative evaluation methods are used to analyze many more reactors; indeed, all reactors could benefit from such an analysis.

In addition, qualitative Safety Goals could be developed; for example, such a Safety Goal

.could define goals for containment performance.

We believe that a comprehensive qualitative program will be far more useful to the Commission and contribute more significantly to the protection of the public health and safety than the present, PRA-based, quantitative Safety Goals program.

Even the NRC-sponsored Interim Reliability Evaluation Program (IREP) PRAs appeared to recognize the worth of qualitative analysis.

These reports state:

"The principal product obtained is the integrated engineering logic presented in the plant and system models and the insights into plant features contributing significantly to risk -- not the specific values computed for accident frequencies."

t The ' NRC Staff, in a more candid early version of the Safety Goal evaluation plan memorandum, admitted that without PRA, "... the establishment of numerical guidance would be an empty exercise."

We agree, and urge the Commission to consider the alternative we have suggested.

We also note that the Commission is sponsoring a study at Oak Ridge l'

National Laboratory of severe core damage precursors.

This program has been quite ' useful in putting actual operating experience into perspective by assessing the approximate extent to which historical events have contributed to the chance of a severe core damage accident.

However, this program has not examined the historical experience on containment failure.

We believe that

this would be a useful extension of the precursor orogram, especially in view of the claims being made in some quarters regarding the capabilities of various containment designs to mitigate severe accident risks.

Such an extension should exainine the degree to which quality assurance problems and variations in materials quality could impact containment reliability.

PUBLIC PARTICIPATION: Illusion and Reality To those of us who have represented intervenors in NRC proceedings --

and I have represented a broad range, including a state government, local and regional citizen groups, and national groups, most often UCS -- this agency presents c. puzzling paradox.

While its overridino oblication under the law is to protec t the public heal th and safety, the public is made to feel an unwelcome interloper in __NRC Droceedin,qs, caught between a Licensing Board that j

has torcefully been told to "eX Dedi t.e" licensing cases, a utility wrsich is y,

perhaps understandably eaaer to get rid of intervenors, and a statt which ot' ten simply seconds the utility's motion.

I think you would be astonished if you polled lawyers who have

""-'"^m wd

ny represented states and intervenors.

I "-

timer M. '; =cM3 Weed-ittorneys involveo ir. their n rsmC tat m

th,ev l ave wai n h te--- ;,,, i n s i uie y, e sc.t2t4-40 never seen a systen en intervenors "nnints of view.

And this, of course, is under the current rules.

New proposals to " reform" the licensing process would close the system even more.

Let me be more specific.

I will begin by citing one of the principal findings of NRC's own Special Inquiry Group:

nsofar as the licensing process is_ supposed, to. provide h

a publicly accessible rorum mr

>no ro sm u u on vi osi so i c i.r issues relevant to the constructinn and operation or a^

nuclear plant, it is a sham.

~

[

(Report to the Commissioners and the Public ["Rogovin Report"], p. 139)

As you know, the licensing process actually begins when an applicant for a construction permit submits the first version of its Preliminary Safety Analysis Report (PSAR) ar.d environmental information to the Staff for an I

initial " completeness" review.

The completeness review does not deal with issues of substance.

The substantive Staff review, and the documents that cmbody it, the Safety Evaluation Report (SER) and Draf t Environmental Statement (DES), are months from publication.

Even the Applicant's basic licensing documents are in preliminary form; it is typical for a dozen or sometimes many more amendments, containing the detailed information required to show compliance with NRC rules, to be issued in the months before and during the actual hearings.

However, the licensing " clock" starts when the public notice is given, even though NRC's fomal review is in progress and the case is not nearly ready for hearing.

This has two ex tremely un fortunate ef fec ts.

First, intervenors' attempts to gather the infomation necessary to file contentions and go to hearing are treated as " del ay. "

Ironically. intervenors must p f formulate their contentions before the Staff's own review is enmnlatod.

In acdition, the Licensing Boards, which are under orders from the Commission to move proceedings quickly, of ten view intervenors as impediment *,

to achieving that goal.

This dynamic has the counterproductive e'ffect of creating an incentive for applicants and Staff to withhold detailed safety information until relatively late in the process, after contentions have been fomulated and discovery completed.

At that point, an intersenor must move to amend its contentions, reopen discovery or modify the hearing schedule.

Such motions not only require intervenors to meet a special burden; they place the intervenor in the position of causing " delay", a position that is extremely unsympathetic to the Licensing Boards.

Even though the delay is not the intervenor's faul t, such niceties become lost in the current climate of "cxpediting" licensing.

By the time the Staff completes its SER review, it has as a practical matter informally " esolved" any differences it may nave had with the applicant.

From that point on, the NRC Staff acts as the applicant's advocate.

Its advocacy of the applicant's position places it from the very btginning in an adversary position towards any intervenor.

This fact explains what intervenors have learned by hard experience but which is always a shock initially:

No matter how technically credible an intervenor may be nor what legitimate issues it raises, the Staff makes virtually no attempt to meet with intervenors, to seriously consider whether their technical concerns have validity and what if any corrective action should be taken.

In ste.a4-th_g, Staff's immediate knee-jerk response is to find some justification for Cpposing the intervenor's positions on aii suu nan u ve -and proceFural 'ssuss, a stance which continues during the entire licensing process.

-- Throughout the process, it is the goal of the applicant, and generally al so the NRC Staff, to keep potentially troublesome information off the record.

While one can understand such an impul se on the part of the applicant's attorneys, it _i s not a proper position for the NRC, whoseg cbligation by law is to the public.

Rules are interDreted in the narrowest f pdTsible way to exclude relevant evidence aid disregard sef 'ous safety issues.

i I recently can.e across a staggering example of this syndrome, which little credibility among the oeople illustrates in a nutshell why NRC hat en who participate in its proceedings.

Following a Commission decision, the San Onofre Licensing Board ruled that 'intervenors may not present evidence showing that hospitals around the plant are inadequate to treat injured persons in an accident.

It was held that NRC's rules only require the applicant to list existing facilitie.,.

Therefore, "[b]oards are not to go behind the list of existing facilities to determine whether those facilities are adequate or inadequate...."

After a construction permit is granted, on the basis of preliminary dasign infomation, the level of Staff review drops dramatically.

The on-site inspector, occasionally augmented by other NRC personnel, is capable of auditing only a tiny fraction of the construction programs, and that audit l

consists largely nf reviewing paperwork.

Yet, as we have seen in the past few years, too many utilities have tolerated pervasive quality assurance di sco@d--T breakdowns during construction -- breakdowns which were never g

worlLers -an t cm tn linht My thenugh "the revelations or plant I

NRC_b_u t intervenors.

Zimer, Midland and South Texas are chilling illustrations of this patte7h.

l

.y, y -

The Eoards consistently refuse to consider safety issues brought to

~

their attention by the public during the construction period on the grounds that such issues should await the operating ifcense review.

The problem, of course, is that once the billions of dollars required to build the plant have b:en spent, unbiased consideration of many safety issues becomes impossible.

As an example, interver. ors in the Seabrook construction penni t proceeding raised the issue that, during an accident, the thousands of people who use the braches within a few miles of the plant could not be evacuated.

This is an issue that goes directly to the suitability of the site.

The Appeal Board in 1977 interpreted NRC's rules at establishing that an accident requiring k svacuation could never occur-therefore, the feasibility or infeasibility of cvacuation was 1,rrelevait to lic_ensing the plant.

N DT

@he TMI-2 accident happened.

Shortly thereaf ter, you

%Twoyear: later, t

promulgatea rules requiring emergency planning for the 10-mile zone around all plants, accompanied by the central finding that serious accidents can happen at even the safest plants.

The Seabrook plant was at that time not yet a fait accompi; I believE that construction of Unit I was approximately 25% compTbTe and Unit 2 was barely started.

Intervenors quickly petitioned the NRC to imediately consider whether adequate emergency protection could be provided to the public surrounding the Seabrook site.

The intervenors urged NRC to hold hearings on this issue before the plant was built, knowing full well that it would be too late afterwards,.

NRC refused to allow hearings, citing the lame __ fiction that the applDt was proceeding witn construction "at its own p2ril" and this issue coula De fully explorea auring tne operating license proceedino.

The intervenors, including most of the towns surrounding Seabrook and the. Commonwealth of Massachusetts, are now pursuing these issues in the cperating license hearing.

It is, however, far too late.

They are faced with a Hobson's choice because, of course, construction is not really at the utility's risk, it 'is fundamentally at the public's risk."

The choices are either to accept the direct risk to public safety or to accept the financial cost of abandoning the plant--a cost that would ultimately be paid in large part by the ratepayers.

In the recent past, the Commission has begun to actively interj ect itself into ongoing proceedings to narrow the issues beging considered or to remove them altogether from the hearings and, therefore, from the purview of en the record public participation.

This has happened several times in the TMI Restart case, in the Indian Point case, in Zimmer and in other cases.

While we do not dispute the Commission's right to maintain control over the proceedings, it should be noted that in each of these cases, the Commissioners intervened to remove issues from the hearings.

I am unaware of any instance where the Commission has intervened on its own motion in the other direction.

Two other recent developments add further reinforcement to the public's view that the Commission is hostile to its right to present arguments on the record which challenge those of the Staff and utilities.

One is the Catawba decision.

Duke Power Company, et al. (Catawba Nuclear Station, Units 1 and 2), CLI-83-19, duty 1,

Im.

Ine commission teck sua sponte review of an Appeal Board decision holding that it was unfair and confrary to the Atomic Energy Act for the Boards to require intervenors to meet the special standards

a for late-filed contentions when they could not have filed the contentions earlier; specifically, because of the unavailability of the emergency plan, it was impossible for intervenors to know whether it was acceptable and, if not, in what specific ways.

The Commission reversed the Appeal Board in a remarkable decision that frankly left many of us in disbelief.

Finally, I wnuld lika tn h,ino your attention to a doctrine devised by siege mentality NRC which epitomizes the mindset I have been discussing: tne that reflMTvely resists public participation.

It is the Commission's position,- ihoi, when it amends a license to add some'new requirement, tne uti.lity may claim a hearing to challenge the need for thF requirement, but the public is absolutely precluded from challenging its suff3c5cy tme the safety problem in nuestion.

inis i3 true even when i.ne puosic rirst orought thf safety problem to NRC's attention.

Ine utility and 5taff will negotiate a resolution privately, and the pijbric simply has no opportunity to present evidence on its sufficiency.

I regret that this practice has recently been

v. NRC.

While the upheld by the D.C. Circuit Court of Appeals in Bellotti -

case is still on-going and I personally believe that it was wrongly decided, this regressive policy should be changed regardless of whether it is legal or not.

UCS' views on current proposals to " reform" the licensing process, and our suggestions for meaningful change, should be well known to you since they have been presented in several Congressional hearings as well as formal comments on NRC's administrative proposals.

What I suspect that some of you do not fully appreciate is the effect of the atmosphere of mistrust in which these proposals are made and which has, unfortunately, shaped the debate on licensing changes.

Intervenors' worst fears about the direction of licensing

" reform" were confimed by the first set of NRC proposals, most of which resurfaced in DOE's bill.

They were a laundry list of the industry's complaints about public participation, which went solely in the direction of cutting off such participation, and which made not even a. single move toward correcting the gross imbalances in the process or encouraging the presentation of alternative views.

I commend you for establishing the Ad Hoc Committee to review these propo sal s, althought I must note that not one active representative of intervenors was invited to serve.

However, the well had been already poisoned, as members of the public perceived that the purpose of licensing " reform" was to exclude them rather than to address inequities and inefficiencies in the process.

The_ members of the nublic and reDresentatives of state and local governments who participate in NRC rulemakings and aajuaications are ent1Tkd_

to a fair ano accessible forum.

Iney oo not receive tnat.

wnoi. Iney too of ten get is a system which provides them with tne tecnnical formalities of it due process -- the grudging minimum required under the law -- but makes tKe 4 clear fenm the outset that they are seen as troub g ma nhetacles in 7

agency's path.

You pay iwo prices for this.

The Lirst prica is the NRC's lqv af credibility.

This applies not only to individuals who are opposed to all nuclear power, and as to whom you might therefore conclude that credibility is impossible to achieve, but also to Congre:.s, state and local go_vernments and other citizens' groups with pro-safetv and pro-consumer agendas whom you cannut aT roro to nave A5 automatic adversaries.

~

-18 N

The other price you pay is the loss of diversity in views on technical lhg and policy issues.

Many people outside the nuclear industry /NRC complex have valuable information and opinions.

I know from personal experience that such persons are frequently unwilling to participate in a process which they perceive as rigged.

I have not pulled any punches in this presentation and have at times used strong language because I believe it is necessary.

However, regardless of how UCS or I personally feel about the wisdom of future nuclear development, the fact is that there are over 80 plants licensed now with more en the way.

Tha_t reality dictates one mission we must share: to see that fairness, j those plants are constructed and operated safely.

Effective DuDiic participa u un and an aggncy which accords tne puoiic respe'ct ~and y

both in words and practice, are necessary to acnieve tnat goal.

---*-w--r-w--rr-

g v. w.

4o...

l.,',,.

Il Oli Of Comcenmen Scismrisis January 21, 1983 The Honorable Richard Thornburgh Governor Commonwealth of Pennsylvania Harrisburg, Pennsylvania 17120 4

Dear Governor Tho rnburgh:

On June 22, 1979, you wrote to the NRC, expressing your

" deeply felt responsibility for both the physical and psycho-logical security of the citizens" of Pennsylvania and advising the NRC c, f your " strong opposition 'to any plans to reactivata

[Three Mile Island] Unit 1 until,a number of very serious issues have been resolved."

Among these "very serbas issues" was whether the " Babcock & Wilcox reactor design flaws which may,

have contributed to the accident in Unit 2 have been fully discovered and corrected in Unit 1."

We regret to inform you that GPU and the NRC Staf f propose operation of TMI-1 without meeting, this most basic condition that you established.

While.a large number of " flaws" in the B&W design have been discovered, GPU. proposes to restart the reactor without correcting many of the most important defects.

the NRC Staff plans to allow the plant to To make matters worse, operate even when the equipment that has been installed or upg raded to correct known safety problems -is out of service.

5 Finally, GPU is urging the NRC Commissioners to allow TMI-l to restart despite the lack of evidence that reliable means exist to cool the-reactor.

We have attached a memorandum which-documents these issues in more detai,1 for you.

• UCS urges you to take two actions..

First, we believe you should notify the NRC that the conditions you set for restart of TMI-l have not been met.

The facts presented in the attached memorandum clearly support this conclusion.

A Governor's v4ews taktn very seriously by the NRC.

Had you not firmly stated as you did in your June 22, 1979 letter to the are youP position NRC, we believe that no hearings would have been. held b'efore restart of TMI-1.

Yo ur failure to comment on the restart,at this time would be interpreted as your approval to resume operation before the remaining safety problems are corrected.

1384 Massachusetts Avenue

• Cambridge. Massachusetts U2238

• Telephone (617) 547-5552 1725 i Street. N.W Sete 601

• Wasnington. O C 20006

• Telephone (202) 296 5600

i.

Second, we urge you to direct the Commonwealth's technical and legal offices to participate in the reopened hearing, now scheduled to start on March 1, 1983. - The active involvement of the Conmonwealth is essential to assure that the safety issues surrounding reactor core cooling will be properly explored and resolved.

For. three years UCS has provided the technical and legal expertise to counterbalance the combined e f fo r ts of GPU and the NRC to restart Unit 1 before its safety defects are fully corrected.

NRC's Appeal Board candidly concluded that the evidence as it stands today does not support a finding that TMI-l is adequately safe.

The NRC has set an accelerated timetable for the reopened hearing.

This hearing will address a number of important safety problems, and we expect GPU to use every available legal tactic to try to prevent a full airing of those issues.

Nor do we expect the NRC Staff to take an objective view.

The NRC Staff has consistently o ppo sed the conclusions reached by the Appeal Board and is now in the position of defending its past actions.

UCS simply does not have the resources to cope with the combined technical and legal forces of GPU and the NRC Staff under the accelerated ground rules established for this hearing.

. th e Commonwealth's active

. participation in the hearing is, therefore, essential for a fair review of the reactor cooling issues.

We recognize your wide-ranging responsibilities, and UCS would not bring these matters to your attention if we could rely on any other means to ensure that the NRC pe(forms the " careful and objective review" of the relevant safety issues that you urged it to conduct in your November 20, 1981 letter to Chairman Palladino.

Unless the NRC sees that the Commonwealth intends to take an

active, informed role in the reopened

hearing, we believe there is little hooe that the remaining ssfety problems at Unit I will be corrected.

In that case, restart of Three i

Mile Island Unit I would once again jeopardize the physical and psychological security of the citizens of Pennsylvania.

Sincerely,

(,L

+-6

-[%

Henry Kendall Chairman Board of Directors Robert D.

Pollard Nuclear Safety Engineer JL,

5i y.

El1yn R. Weiss General Counsel Enclosure w

~

11 011 0 Comcenmso SCIENTISTS M EMORANDUM To :

Governor Richard Thornburga Prom:

Union of Concerned Scientists Da te :

January 21, 1983 Re:

Safety Problems at TMI-l As the following facts establish, one of your basic that the conditions for restart of Three Mile Island Unit 1 design flaws which contributed to the THI-2 accident should be has clearly not

" fully discovered and corrected" in TMI-l been fulfilled.

1.

Important Design Flaws Remain Uncorrected Today, almost four years af ter the accident, many important design flaws remain uncorrected at TMI-1.

It should be emphasized that the design changes we refe'r to are those where there is no dispute that design changes are necessary.

Fo r not GPU, the NRC example, no party to the restart hearings Staff, the Commonwealth, or UCS -- challenged the necessity of coolant piping, upgrading the emerge,h point vents on the such modifications as installing hig reactor ncy feedwater system to meet the NRC cequirements applicable to safety

systems, and installing additional radiation shielding to protect equipment and personnel during and after a TMI-2 type accident.

The only issue here is the timetable to complete the necessary modifi Mtions.

Deadlines for correcting the serious safety problems disclosed by the accid 2nt were first set for January 1,

1901.

l l

However, those deadlines have.been extended over and over again.

As it stands today, the necessary design changes will not be implemented at THI-1 until the first refueling after restart.

We have no confidence that even this schedule wi-11 be met.

You may recall that when GPU anticipated restarting the plant in l

l l

1384 Massachusetts Avenuo. Cambodge Massachusetts 02238 + Tel (617) 547 5552

... -. - --.. 3 l

- +

late 1981 or early 1982, it had " committed" to complete the necessary modifications during the first refueling after restart

-- in other words, about a year after restart.

Now, more than a year later, GPU makes the same promise.

1his GPU commitment can not be relied upon.

Recently, GPU has begun to qualify its " commitments."

On j

.D cember 17, 1982, GPU Executive Vice President Philip Clark told the NRC that GPU would correct the most recently disclosed safety deficiencies in the emergency feedwater system during the first refueling after restart -

"if feasible."

In the context of that discussion, GPU's basic position is that it will make v

the changes necessary to protect the public when it is conven-ient and does not interfere with its first priority, operating TMI-1.

The NRC Staff, for its part, has been more than willir,g to extend the deadlines in response to GPU's pressure, even going so far as to solicit requests for extensions.

The NRC is also continuing its practice of classifying safety problems as generic in order to postpone decisions on dif ficult technical questions.

Although the President's Commission on the Accident at Three Mile Island (the Kemeny Commission) soundly condemned this practice, the NRC continues it even for the new generic safety issues that stem f rom the TMI-2 accident.

For example, in its March 1981 report to the Congress,

" Identification of New Unresolved Safety Issues Relating to Nuclear Power Plants," the NRC states :

l There is an increasing realization that one of the most important factors in the safety of nuclear reactors is the. reliability of the systems used for decay heat removal following the shut down of the reactor for any reason.

The principal means for r emov.ing decay heat in a pressurized water reactor [like TMI-1] under normal conditions immediately following reactor shutdown is through the steam generators using the auxiliary l

feedwater system.

[In TMI-1, this system is named

" emergency feedwater system."]

In addition to the WXiH-1400 study mentioned above, later reliability studies and related experience from the accident at ThLse Mile Isl,and Unit 2 (TMI-2) have reaffirmed that the loss of capability to remove heat through the steam generator is a significant contributor to the a core-melt event.

NUREG-0705, pp.

probability of A-13 A-2.

e

-3 In Ma rch 1981, NRC's " preliminary estimate" for completing its studies of this unresolved safety issue was April 1984.

In June 1982, the NRC Staff revised this to November 1984, with the schedule for implementing any new safety requirements in operating reactors still undetermined.

'1h us, relying solely on NRC's own statements, it is clear that the TMI-2 accident " reaffirmed" the need fo r reliable reactor core cooling systems to prevent a mel tdo wn.

The NRC also acknowledges that the "need fo r an emergency feedwater system of high reliability is a clear lesson learned from the TMI-2 accident."

(NUREG-0578, p.

10)

Nevertheless, the NRC s

Staff would allow TMI-1 to restart without installing the modifications which all parties agree are necessary to correct the known safety deficiencies in TMI-l's emergency feedwater system, the " principal means for removing decay heat" in TMI-1.

t In sum, GPU and the NRC Staff propose to operate TMI-1 without the modifications they both acknowledge are necessary to correct the safety deficiencies disclosed by the TMI-2 accident.

We recommend that you upd ate your 1979 statement calling fo r correction of TMI-l's design flaws by making clear your opposition to restart until at least those design modifications

-which all parties concede are necessary, plus those the Licensing Board finds necessary, have been fully implemented.

2.

Essential New Equipment Need Not Be Operable A great deal of existing equipment has been modified or new equipment has been installed to correct safety deficiencies disclosed by the accident.

Ho wev e r, GPU and the NRC Staff g

propose. to run TMI-1 even if that equipment is not operable.

Thus, while some of the design flaws revealed in the accident will theoretically have been corrected, Unit 1 would be free to operate in much the same condition as it was before the accident.

GPU has requested amendments to its TMI-1 operating license, endorsed by the NRC Staff, that would allow the plan t to operate without restriction if half of the new or upg raded safety equipment is nonfunctional.

Moreover, under this proposed ifcense amendment the plant could operate for a full

~

week after failure of the remaining safety equipment because, C'PU now asserts, the new or upgraded equipment is not ecsential for safety.

Fo r example, two devices have been added to inform the operators that the PORV is open.

The NRC Staff is ready to conclude that TMI-1 can operate safely even if one of those devices is always inoperable, and that it can' operate safely for

o

_4_

i a-week after failure of the other warning device.

The basis for i

this ' claim is that the operators can rely on the temperature indicator downstream of the PORV, even thoug h the operators ignored this indicator when the PORV stuck open during the THI-2

~

accident.

Similarly, two meters have been added to inform the operators that the reactur cooling system is approaching the boiling point.

Under GPU's proposal, one of those meters would

+

never have to be operable, and the other can be out of service for a week before the plant must shut down.

The basis for this proposal is GPU's assertion that the operators have other means to determine whether an unsafe condition is developing in tise 1

reactor.

Unfortunately, these means proved inadequate during the TMI-2 accident.

In effect, GPU and the NRC Staff now argue that this new equipment is not necessary for safety -- a position they did not take during the restart hearings.

We feel confident tha t the Licensing Board wo'uld not have approved restart had it known that the equipment necessary to correct the safety deficiencies disclosed by the accident would no t necessarily be operable while the plant is running.

UCS certainly would have challenged that position had it been taken, and we expect the Commonwealth would have as well.

In sum, the restart of TMI-1 without the full complement of new and upg raded equipment that has been installed to correct the design flaws disclosed by the TMI-2 accident violates the spirit and probably the letter of the Licensing Board's decision and should not be permitted.

In addition, the fact that GPU and the NRC Staff even made such a proposal is another indication that their first priority is operation of TMI-1 rather than protection of the public.

3. Adequacy Of Decay Heat Removal Rempins Unresolved Perhaps the most fundamental and dangerous flaw in TMI-1 revealed by the accident is the grave uncertainty of adequate decay heat removal.

Remarkably, the NRC Appeal Board's most t

recent,05 er (issued December 29, 1982) confirms what UCS,has d

argued Tor years: that the facts do not establish that viable and suitably reliable means of decay heat removal exist at TMI-1.

As you are probably aware, the Appeal Board ordered the hearing reopened because of this issue.

Nevertheless, on January 4,

1983, GPU asked the NRC Commissioners to allow TMI-1 to restart.

A copy of that letter and our response are attached for your attention.

v.

You will note that GPU's sole ground for this extraordinary plea is its claim that because similar safety problems may exist at other plants, it would not be " co nsi st ar nt" fo r. TMI-l to be treated differently.

In other words, if one or-more plants are ope ra ting with safety hazards as sericus as those at TMI-1, TMI-1 should also be allowed to operate.

Quite frankly, UCS can imagine no greater perversion of the 4

NRC's duty to protect the public than to let the allowable level of risk be dictated by the worst powerplant in'the country.

We also believe that it is especially i r res ponsi ble-for GPU to argue, in effect, that the people of central Pennsylvania are i

not er, titled to protection of their health and safety as long as the residents near.other nuclear plants face the same dangers.

In any case, your concerns as Governor of Pennsylvania must be, as they have been from the outset, with the sa fety' of your constituents.

Although the details of the deficiencies in the decay heat the issue can removal systems at TMI-l are technically complex, the evidence to be summarized in one simple, accurate sentence:

date cannot support a conclusion that Unit 1 can withstand the type of accident that occurred in Unit 2.

'Ih a t is why the Appeal Boa rd scheduled additional-hearing s to begin March 1.

The Appeal Board directed GPU and/or the NRC Staff to provide additional testimony in eleven areas.

In several of these

areas, the Licensing Board had ordered additional testimony during the original restart hearings.

Thus, this will be GPU's and the NRC Staff's third or four th attempt to prove that the f acts support their common position that TMI-l is safe enough to restart.

In addition to the very serious safety issues that necessitated the new hearings, the NRC Commissioners were informed by their Staff on December 17,

1982, that the TMI-1 the principal system used to emergency feedwater system is not remove decay, heat from the reactor after a shutdown capable of withstanding even a mild' earthquak,e.

Although GPU proposes to modify the system to withstand the largest historical ea r thquake for the plant

location, it has only committed to take this action during the first refueling after restart "if ' feasible," as we discussed above.

v Only GPU and the NRC Staff were allowed to participate in the Commission meeting at which the earthquake issue was discussed.

(The re fo re, UCS considers the meeting to have been an ex parte communication of the type prohibited by the AdmiEIstrative Procedure Act and NRC's Rules.)

However, all parties were permitted to comment on the matters discussed.

We j

were disappointed that the Commonwealth did not choose to comment.

UCS's' comments included the following points:

.-----.,--,-pg,-e,.,,

n- + - -, ~ _ - -

,-,n,w_,.--n,w-,,,,---,.,,,,m,,.-n.,

--.ew-,,,,v-, -,.-.,

,a-n _,w _ a _ _ -. - -, - aw..,

n. - - ~ ~ - -

t The inability of the TMI-l emergency f eedwa t.er (EFW) system to withstand even a mild earthquake an earthquake so mild that it can reasonably be expected to occur during the lifetime of TMI-l poses a substantial safety hazard to the public.

The finding that the TMI-l EFW system cannot survive-an earthquake is directly relevant to the issues in the restart hearing and to the Commission's pending decision regarding restart.

The need for emergency feedwater systems to be designed to withstand a much larger earthquake (designated as the Safe Shutdown Earthquake), and to incorporate pomps with diverse motive powe r (e.g.,

steam turbine and electric motor drives),

was recognized as early as 1972 by the Atomic Energy Commission.

GPU has been dilatory in providing all the information which, in February 1981, the NRC Staff requested be supplied within 120 days so that it could determine whether the license should be modified, suspended or revoked.

The NRC would violate its obligation to protect the health and safety of the public if it authorizes TMI-l to restart before the TMI-1 emergency feedwater system is modified to meet NRC's rules applicable to safety systems.

Conclusion i

TMI Unit I should not be allowed to restart until:

l 1)

All modifications which age necessary to correct the safety deficiencies disclosed by the accident have been made; i

l 2)

Its operating license requires that the full complement of safety equipment be operable prict to plant startup, and that the plant be shut down within a prescribed time after any soch equipmbit becomes inoperable; and i

3)

The evidence be fo re NRC's hearing board demonstrates I

that its decay heat removal systems are adequately reliable.

l t

e s

e se SHAw PITTMANe PoTTs & TROWBRIDGE e

a emerestassem or amortsseo=at Compoaateoms 6000 M ST R C C T. N. W.

WASH t N GTON. D.C.20036 83O29 823-sOOO osassaw.. ports. P C to.oasas a onmeta, o c Jomse se o setett, se a t==t*= J esau t*e *=

Estvant t pe rt esam.

• C.

Janets se owmota. P C.

Jaw a. testates cawio s.awete CC **stt a mano a attra a ss = t ammusmoor 4 tomet a snowsmooct. pe".

s=ctoo= J wrests. P C.

' E ' E CO**C "

tewoveev e aseemeot escoteeCa 6 as t ue es t e.eces o pot e s.

• C Joase a sacCustouces. e C e tanto Caes==orr.

• C J Pata*C= e.C=C v.

• C-eaoas saa aoes a eps seee gusaetta ** *E=Ottfoas conoc= a =Aseorsaw paut a. maata=

Jtreetv s. osa=Cota s=stus o sonrweCa, oc stoact e eneC=ats t. Ja.

• C manav es. otas&**Ects sea==asec = uteta=*=

twoamas.t s =seant. e c.

.sanoma c. rossoas e tenorsev eensatose e c.

J attwt= t ettvata, o c startert.vasso=

e g omet e. moor es, se, o c eancia a neent =setese tttcm sacn wenaw arta a u'itt.oc ocam o auuCa.

• C O as m.mecoewec, soose= a sanott a

's"us'*se ao tow mo at vows o. em so== a n=eset ta= ore oc some= c=GCt. o c so-aeos esmawtaw wone a

r et useo cs wCE w C=wec=wt.

• C C=amt t a e f reemen. o c ttf ut a =*C=osso=. Ja.

• C. s et eset= ee newtetta. P C.

C *.es t -seeawt aw

• Jones o came. Je moster 4 westaso*C menerene o meast. o c went= mon me emows. e c.

peeeup J seanvt, amoetw o E6tes moster me conoo wtwotum a==es t esce.neo J a t moast. e C Jamer s e maasum.

• C saneana J esono t ee staatte a mano aw a seter oc

• C mamoat a utst. e t so==st s corvut o onestet use s o scam;ama = mosteotte et moetet C a nset t e ta sme s e stsusma s aw'e ca omet w atst= se.pc n.Ceeamo t catta sose= r ot at,'

egowano m smarstansa=

weao troCE otsomam e eavsta

• C mostat e moneeses g%, g ecot?a a=t= stag antat asse t

• C=E t a rats om,as=to o a s ses wtestte. o c attvtse me. Lucas caseretLL notstria saanc t s t = a t a s 't e=

eent=== ct e sotto so. o c cow.o se avet=stre=

c tresse w=efitt st y wetso=

ttva = =oocasiaas taawes t emo==.<a eeana awot mou.c a. o c Cu ate m

seetetasecc maevtv cast t ta=Estt.osa se.*C mateas r snaverso oemt octassaan e cowar nec=a mo = a mo= te=**

c.m es t som o sosets. e C w Ctonea J etmanes s e c reeta e est e====

e=ot me=evoto = o c weette s...t C, o.at =w=.C e January 4, 1983 822-1026

~

Nunzio J.

Palladino, Chairman Victor Gilinsky, Commissioner John F. Ahearne, Commissioner Thomas M.

Roberts, Ccmmissioner James K. Asselstine, Commissioner U.S. Nuclear Regulatory Commission Washington, D.C.

20555 In the Matter of METROPOLITAN EDISON COMPANY (Three Mile Island Nuclear Station, Unit No. 1)

Docket No. 50-289 SP (Restart)

Dear Chairman Palladino and Commissioners:

In its Memorandum and Order of December 29, 1932 (ALAB-l 708), t6c Appeal Board ordered a limited reopening of the record in this proceeding to facilitate prompt resolution of matters related to decay heat removal under certain post-accident conditions. */

Licenscc is concerned that the Appc 21 Doard's actions could precipitate consideration. whether to again defer a Commission determination on lifting the suspension on THI-1 operation.

The Commission should not further defer its determination on immediate of fcctiveness because of the Appeal Boa rd 's *4cmoranitum and Order.

i

+/

It should lx: noted that t.he Appeal Doard's action does not its final decision on the subjects addressed in,the represent

_ _M[ ~

m v e

ae...

n..t

. J' O

s q.

SHAw. PITTMAN. PoTTs & TRoWORioGE

.... r-c....

..., c.

. c.

Chairman Palladino and Commissioners January 4, 1983 Page Two 1979, TMI-1 was shut down under an imme.liately effective In order to be continued until the conclusion of a oublic hearing, because of Commission concerns viewed as unique 'to this plant's operation.

No other plant was treated in a similar manner.

There can be no serious dispute that the questions presently removal are not being raised by the Appeal Board on decay heatare equally applicable to a numb unique to TMI-1 but in fact It is essential currently operating plants of similar design.

for bringing this unprecedented proceeding to completion that issues common to other operating plants, as they arise, be resolved for TMI-l consistent with their resolution for other Continued operation of other similarly con-operating plants.

figured plants would be irreconcilable with deferral of a decision the immediately ef fective suspension of TMI-1 on the basis to lift of the generic questions raised by the Appeal Board in its Memo-randum and Order.

Respectfully submitted, SHAW, PITTMAN, POTTS & TROWGRIDGE By:

, fs/ff/

MN/

/

Geofge F. Trowbridg Counsel for Licensee 1

GFT/tjc cc:

Service List Q, indicates hand delivery)

(Continued)

Memorandum and Order.

Rather, the Appeal Board by its reopening seeks supplemental information to clarify the record and to ex-plore further the position taken by both the NRC Staff and Licensee removal can be reliably accomplished either by that decay heat including boiler-condenser natural circulation, natural circulation, The using the cmergency feedwater system or by feed and biced.

Appeal Board han in fact stated that satisfactory demonstration as removal would resolve its to either of ther.c methods of decay heat conccrns.

,6

,i

~

Il Oli i Comcenmso SCisNTISTS January 11, 1983 Nunzio Palladino, Chairman John Ahearne, Commissioner James Asselstine, Commissioner Victor Gilinsky, Commissioner Thomas Roberts, Commissioner U.S.

Nuclear Regulatory Commission Washington, D.C.

20555 RE:

Three H i'l e Island Unit 1 Docket No. 50-289 (Restart)

Gentlemen:

1 UCS is astonished by the letter sent by Counsel for GPU to the Commissioners on January 4,

1983 GPU urges the Commission to make the Licensing Board's decision immediately effective, thereby lifting the NRC's 1979 orders requirJng THI-1 to rematn shutdown, despite the fact that the Licensing Board's decision has been invalidated in important respects by the Appeal Board.

The sole grounds offered in support of this extracrdinary plea is, in effect, that some other unidentified plants with safety deficiencies as serious as THI-1 may be operating.

This is, of course, an assertion which has yet to be proven.

However, the record is

clear, as the Appeal Board has

found, that the evidence 'does not support a conclusion that THI Unit 1

c. a n withstand an accident of the type that occurred in Unit 2.

Tigerinescapable conclusion of GPU's logic is that if one other plant could be found with equally serious safety problems.

THI-1 should be permitted to restart.

It is unthinkable that Nhc would seriously consider adopting a policy which would either explicitly or implicitly permit the tolerable level of risk to the public to be dictated by the 1:saa Mawactiusi lis Aounn Li' aire:alo+ M svenhnxit. t:2?:10 I,1.rt,tiosie (6111 ';4 / '>'i's/

D C ?Of KJ6

• T.f.?tthon ' (202).990 '41 Ks 1 ? ?8, I t fa W

• C.. iti$eMll

• W.iJustutori l

e

<, w m m w -

..*s

. e s'.

wors' plant or plants which nay be operating.

We can imagine no greater perversion of the NRC's unequivocal mandate to protect the health and safety of the pubite.

l Very truly yours, N

Robert D.

Pollard Nuclear Sa fety Engineer 5

Union of Concerned Scientists s

)

b

'b EllyM

. Weiss General Counsel for UCS Harmon & Weiss.

1725 I Street, N.W.

Suite 506 Washington, D.C.

20006 cc: THI-1 Service List k

e o

h I

,._.-,,-,m.,,,.-,,.n._.,

--