ML19220C544
| ML19220C544 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 04/10/1979 |
| From: | Hendrie J NRC COMMISSION (OCM) |
| To: | |
| References | |
| NUDOCS 7905110197 | |
| Download: ML19220C544 (42) | |
Text
..
$ls 0 'O _
'.~ r I
[
3'[ - 3 TESTIMONY OF CHAIRMAN HENDRIE SEFORE THE SUSCCMMITTEE ON NUCLEAR REGULATION OF THE SENATE CCMMITTEE ON ENVIRONMF.NT AND PUBLIC WORKS TUESDAY, APRIL 10, 1979 Mr. Chairman, as you and I agreed in arranging this hearing, we are here to present a preliminary account of what happened at Three Mile Island, our response to the accident, and some initial conclusions in order to put tMse matters on record at an early time.
I am very grateful for and a,'preciative of your understanding that we could not and ;Nald not divert any appreciable staff resources from the operations at Three Mile Island or from the support groups at Bethesda to prepare a more ex+.ensive record for this hearing.
There will, of course, be r. ore detailed investi-gations, and we will look forsard to future hearings in which a further accounting can be made.
First, by way of general comment, I want to emphasize my great personal concern, which I know is shared by my colleagues on the Commission and all members of our staff, that this accident at Three Mile Island Unit 2 has happened.
All of us are profoundly thankful, as I am sure you and all of our citizens are, that the radiation exposure levels to the public in the area of Three Mile Island have been low.
However, we do not take much ccmfort from the low rad.ction exposures experienced 'to cate since it is clear that the potential for substantially larger exposr of the public existed during the course of this accident.
I90511 0 6 ]
/ J. 7
_2_
The efforts of a very large number of people in the licensee's organiza-tion, the State of pennsylvania, the NRC, other Federal and State agencies, and industry groups are being applied to make sure that we can collectively maintain and improve the conditions that have prevailed thus far with regard to the offsite public and the workers at the site.
The rapid marshalling of very extensive resources frcm a wide variety of sources has been a heartening experience and I believe the record should include er, expression of our appreciation to the hundreds of individuals and organizations that l ave responded.
It is my view, and I am sure it is yours as well, that we cannot have en acceptable nuclear power program in this country if there. is any appre-ciable risk 'of events of the Three Mile Island kir.d occurring at nuclear pcwer plants.
The Nuclear Regulatory Commission must promptly carry out a searching review and evaluation of our own policies and procedures, in addition to our investigation of what has taken place at the Three Mile Island facility.
We must find out where our inspection and enforce-ment of safety-related operating requirements, cur design standards, and our reviews of possible transient and accident situations have scmehow been inadequate to prevent the Three Mile Island accident.
We already have put these elements of the staff that are not immediately involved in dealing with the situation at Three Mile Island tc work on this essential q}fb and major effort.
. The iiuclear Regulatory Commission's investigation of tne Three Mile Island accident is actively uncerway.
Of course, cur most immediate concern has been dominated by the operaticnal censiderations of limiting further releases and returning the plant to a safe and secure shutdown condition.
Consistent with our efforts in that direction and to the extent-could do so without interfering with the recovery operation, ou.* investigators have been at work gathering information.
The NRC investigation team is now at the Three Mile Island site in the field phase of its work, following several days of review of plant records and charts furnished to our Operations Center in Bethesda.
As you are aware, Mr. Chairman, the NRC deliberated very carefully concerning the possibility of serious generic concerns that might require shutdown of the other nuclear power plants designed by Babcock and Wilcox, the nuclear steam system supplier for the Three Mile Island plant.
The results of our review in this area led us to the conclusion that although shutdowns were not warranted at other B&W plants, it was important to require tnese other utilities to make changes in operating procedures and to carry out supplementary training of their cperators.
g6 Ju'
_4-These required actions were taken prcmptly and are "eing folicwed up by increased inspection activities.
First, an NRC Inspection and Enforcement Bulletin was transmitted on April 1, directing licensees operating B&W reactors to perform a series-of specific reviews and actions.
The licensees must respond by temorrow.
Our onsite inspectors will monitor compliance with this Bulletin.
The NRC Headquarters staff will review the responses prcmptly and act upon them accordingly.
This Bulletin was supplemented by another on April 5 that provided additional specific operating instructions based on our augn nted understanding of the events at Three Mile Island.
(Copies of
'hese Bulletins are attached for the record.)
Second, an inspector was assigned full-time at each operating plant having a B&W reactor by April 2.
In addition to assuring that the Bulletins are fully understood and folicwed, the full-time inspector --
with additional assistance frem the NRC Regional Offices -- will assure that some inspection activities will occur during eacn shift and that ea:h shif t will have a clear understanding of the guidance.
Regional and Headquarters staffs are staying in clese touch witn the onsite NRC inspector! to be sure that our instructions to the licensee: are being carried out.
Furthermore, the Ccmmission has sent a telegram to each of the licensees with B&W reactors to underssare the seriousness with which tne Commission views this situation. (A copy is attacned for the record.)
9 6 'a'J D
~,w-
~ ~ -
. Internally, an NRC Task Force on Generic Review of Feecwater Transients in B&W Reactors has been formed to examine the reactor and plant systems at these plants that provided protection against feedwater transients.
The Task Force report is expected to be completed before the end of this month. We will carefully review that information and take whatever further action may be appropriate.
Pending completion of our review of the responses to the April 1 Bulletin and in view of the actions which the NRC has taken and our current understanding that operator errors played an important role in the accident, we believe there is reasonable assurance that the plants can continue to operate without danger to the public health and safety.
Although Ccmmissioner Bradford agrees fully that the aforementioned specific proch 'ures are prudent and provide a considerably enhanced level of assurance, he would await the report of this Task Force before sharing fully in this conclusion.
Before descriting the situation at Three Mile Island, I should like to comment on President Carter's role.
Frem the outset we have kept the President fully informed concerning the accident.
He has been most helpful in providing us with nis personal views and suggestions and in making available to the NRC a communications network.
Moreover, his visit to the site on April I helped alleviate the growing anxiety of the people in the area.
9b
. In his recent eriergy address, the President expressed concern about the accident and directed thac an independent Presidential Commission be established to investigate the causes of the accident and to make recommendations on hcw the safety of nuclear pcwer plants can be improved.
The Nuclear Regulatory Ccmmission will, of course, cooperate fully with the Presidential Commission in every way.
Initial Events At about 4:00a.m. on March 28, 1979, an event occurred which lead to the accident at the Three Mile Island Unit No. 2 facility.
The sequence of events included a lots of feedwater flow, the sticking open of a primary relief valve, the premature turning off of the emergency core cooling system, and later the turning off of all reactor coolant pumps.
Shortly before 7:00a.m. a site emergency was declared by the utility as it became apparent that the reactor fuel had been damaged.
I would now like to have Mr. Darrell Eisenhut go through the detailed sequence of eventi, as we understand them new, in the flest hours of the accident. Then I will resume my statement.
96 332
_ At about 7:00 a.m. cn March 28th, the licensee notified the State of
"'nnsylvania. At 7:45 a.m., the licensee reached the NRC Region I office, and within h hour the NRC Incident Response Center in Bethesda was operating.
The White House was alerted at 9:15 a.m., and by 10:05 a.m. the first NRC Response Team was on the site, a group of five inspectors frcm our Region I office.
At 10:15 a.m. the Comission was briefed by the Director of the Office of Inspection and Enforcement.
NRC had 11 people and a mobile laboratory van frcm the Region I Office on site by evening; radiological assistance teams from Brookhaven Laboratory were monitoring the site vicinity frca mid-afterncen on.
An Aerial Monitoring Survey aircraft also was at the site by mid-afternoon.
NRC strength at the site increased to 29 on Thursday and 83 on Friday.
(A brief chronology of these response actions is attached for the record.)
From Wednesday on, there were continuing releases of radioactive gas evolving from the reactor cooling water which had been pumped frem the containment building to the auxiliary building.
Despite efforts to halt these releases, they increased on Friday morning.
This situation led to a decision by Governor Thornburgh -- en the Ccmission's recommendation --
to advise preschool children and pregnant women to leave tne 5-mile zone nearest the reactor.
The releases were greatly reduced on Friday afterncon when scme of the water in the auxiliary building was pumped to storage tanks emptied for that purpose.
NRC increased the level of its onsite team, and Harold Denton, the Director of Nuclear Reactor Regulation, was 96 333
. placed in charge.
Cctmunications and transportation support were supplied at White Hcuse direction.
Other Federal agencies provided additional assistance.
NRC received the full cooperation of Pennsylvania State government officials as well as other Federal agencies.
Ncwhere was this cooperation more apparent than in the vitally important area of radio..;ical monitoring.
For insnance, the Cepartment of Energy's Aerial Mcnitoring Survey airplane was on station over the site within a few hours.
Personnel frcm the Pennsylvania State Bureau of Radiological Health were making measurements frcm Thursday afternoon enward.
Radiological monitoring efforts were also carried out by HEW and EPA teams.
On Friday, it was recognized that the early core damage had formed a substantial quantity of hy 'rogen gas, some of which was thought to have collected in the reactor pressure vessel above the core.
One concern was that if the reactor pressure was decreased, the hydrogen bubble would grow and might interfere with the flow of ecoling water through the core.
Another was that oxygen generated by radiolysis of water might accumulate -- with the ;:tential for forming a flammable mixture.
The Ccmmission considered recommending additicnal evacuation measures beyond those previously rec mmended to Governor Thornburgh -- wnich I indicated in my press conference on Saturday afterncen -- but decided tha+
gh b
. increased protective action was not necessary.
I should note that at one time or another frcm Friday morning through Sunday, members of the senior staff and various Ccmmissioners propoced either stronger advisories or actual evacuation of people within a two or five mile circle around Three Mile Island.
However, the estimates frcm NRC staff at the site remained considerably more optimistic than those which could be based on the more limited information available to us in Washington or Bethesda.
Consequently, stronger recommendations were not made to Governor Thornburg.
The precise timing and content of the various preposed Ccmmission recommendations will certainly be made public as soon as all of the relevant tapes and notes can be reviewed and transcribed.
The information will then be promptly furnished to tne Congress.
By Sunday, April 1, it appeared that efforts to understand and manage the hydrogen bubble were meeting with scme success.
Over the follcwing few days, the bubble was reduced to negligible size by degassing with the pressurizer spray and the letdcwn ficw and by gas dissolving in the reactor coolant water.
Further analysis had also indicated that little, if any, oxygen could be evolved in the hydrogen-rich ccnditions of the reactor system. Thus, the period of immediate crisis passed and the reactor cooldcwn process could proceed without the hydregen bubble problem.
The entire recovery opera *. ion is being conducted in a way that minimizes the release of radioactivity frcm the site and that requires very careful 96 35:
3
~
-. consideration and concurrence by the on-site NRC team in every significant action taken.
With the reactor in a stable cooling status since early in the accident, we have been very cauticus about moving too rapidly in the recotery operations lest we encounter situations that have not been carefully thought out.
I am convinced that this is the correct approach, even though, frcm the public's point of view, it may seem ta prolong the emergency.
Let me now give you a brief status report en the current situation at Three Mile Island.
I was at the site again last Sunday and have been in close contact with Farold Centen and his team at the sito throughout the accident.
Conclusion The staff has identified six main factors that caused and ir. creased the severity of the accident.
These are as folicws:
1.
At the time af the initiating event, loss of feedwater, both of the auxiliary feedwater trains were valved out of service.
This was a violation of the plant Technical Specifications.
2.
The pressuri:er electromatic relief valve, which opened during the initial pressure surge, failed to close when the pressure decreased below the actuation level.
This failure was not recognized and the relief line closed for scme time.
g{6 e e m o
_ _ _ ~__. -. 3. Following rapid depressurization of the pressurizer, the pressurizer level indication may have lead to erroneous inferences of high level in the reactor coolant system. lne pressurizer level indication apparently led the operators to prematurely terminate high pressure injection flow, even though substantial voids existed in the reactor coolant system. 4. Secause the containment does not isolate on high pressure injection (HPI) initiation, the highly radicactive water frcm the relief valve discharge was pumped out of the containment by the autcmatic initiation of a transfer pump. This water entered the radioactive waste treatment system in the auxiliary building where scme of it overflowed to the floor. Outgassing from this water and discharge through the au.Miiary building ventilation system and filters was the principal source of the offsite release of radioactive neble gases. 5. Subsequently, the high pressure injection system was only intermittently operated to control primary ccolant inventory losses through the e'ectrcmatic relief valve, apparently based on pressuri er level ' ndi cation. Cue to the presence of steam and/or noncondensible voids elsewhere in the reactor coolant system, this led to a further reduction in primary coolant inventory. E. Tripping of reactor coolant pumps during the course of the transient, presumably to protect against pump damage due to pump vibration, led to fuel camage since voids in the reactor coolant system prevented natural circulation. 96 3Yl
- L. . As I remarked at the beginnirg, I am deeply concerned about the accident at Three Mile Island and its implications. Even thcugh, as far as we can determine at this time, no one was exposec to dangerous radiation levels, the accident was indeed serieur. The lives of the people in the neighboring communities were disrupted and many individuals experienced personal hardships. And, of course, the economic consequences of the accident are substantial. We cannot tolerate accidents of this kind and we ' take whatever steps are necessary to prevent them. '4e have a lot to learn from the Three Mile Island accident. I would like to emphasize here that I'm net just thinking of improved hardware or procedures, or other technical fixes. I'm thinking of the regulatory framework itself. I do not see the Three Mile Island accident and its implications as matters that can be addressed exclusively on a technical ~~' plane. For example, the total response of all parties involved reveals that a critical reexamination of the relationship among licensees, regulators and suppcrt groups during crisis management is needed. I do not mean to convey any impression of conflict amcng these groups arising out of the Three Mile Island accident. Indeed, cooperation has been very good. However, during these last 2 weeks we in NRC have been neavily involved 9h bJb
.-e--- e---- . in operational matters that we normally expect to be primarily the licensee's responsibility, subject to our review and approval. Our involvement in these matters has been appropriate under the circumstances, but we believe that we now ahould look more carefully at our preplanned role in crises. It is possible that legislative action may be needed in this area. We have been sorely taxed to assemble NRC personnel from our other Regional Offices and our licensing offices to provide onsite NRC staff supporc. We may find that our budget planning has inadequately considered the impact of short-term heavy demands such as the Three Mile Island accident. It is already clear to me that among the likely areas wnere improvements should be made are first, a substantial upgrading of reactor operator training with respect to plant transients and off-normal conditions. Second, we should put more emphasis in our staff reviews and safety regulations on safety measures to deal with plant transients criginating in the secondary or steam-producing side of nuclear ?cwer plants. Third, we must carefully reexamine the sensitivity of all plant designs to these transient situations and we must reexamine the automatic safety systems that deal with them. Finally, we must make certain that our licensing review and our inspection procedures are as capable as we can make them of p*u"enting the mistakes that were made at Three Mile Island. -70 9b bJ/ .m e. .es. m-w e}}