ML20106E485

Corrected Pages & Table of Contents to Findings of Fact & Conclusions of Law Re Dieckamp Mailgram Issue.Certificate of Svc Encl

ML20106E485
Person / Time
Site:	Three Mile Island
Issue date:	02/08/1985
From:	Bernabei L, Doroshow J GOVERNMENT ACCOUNTABILITY PROJECT
To:
Shared Package
ML20106E476	List: ML20106E474 ML20106E485
References
SP, NUDOCS 8502130324
Download: ML20106E485 (20)

Text

r7-277. Socond, Mosoley's cbservations of Diecksmp during his one strike the Board as peculiarly unreliable. First of all, the interview itself provided no opportunity to test Dieckamp's cred-ibility. The questions which Moseley asked Dieckamp were gener-ally conclusory, and asked for opinions not for facts. Tr.

29,898-900 (Moseley). Secondly, Moseley approached the interview, which focused on the mailgram, with an extremely narrow working definition of material false statement -- that is, one submitted to the NRC in some of fic*al manner, Moseley knew at the time he interviewed Dieckamp that he would not find the mailgram to be a material false statement. Tr. 29,893-897 (Moseley).

278. Third, the Board observed that Moseley appeared to find the entire issue of the Dieckamp Mailgiam beneath his atten-tion. Moseley appeared to be more interested in protecting Dieckamp than in determining whether information about the pres-sure spike and hydrogen burn had been improperly withheld from the Commi ssion.

279. Although he determined that the mailgram contained inaccurate statements, apparently Moseley never bothered to tell t he Commi s s i on. Tr. 29,846-847 ). Further, Moseley testified that he believed Dieckamp should have correted the mailgram. Tr. 29,946. However, he explained that it was a rather minor inaccuracy even though it had consumed great NRC

. investigative and hearing resources. Tr. 29,946 . When asked why he believed Dieckamp should have corrected the mailgram, Moseley explained simely that it would have saved himself a lot of grief.

Tr. 29,976 (Moseley). The Board finds that Moseley was not 101

~

8502130324 850211 PDR ADOCK 05000289 0 PDR

.___.._-,_~..__.._________:_,___..__._,__.__._... . , . . _ . _ . . .

sufficiently concerned in his investigation about the adequate flow of information to the NRC.

, 280. On the other hand, we find Gamble's criticisms of the i IE investigation and report to be incisive and determinative in l

our decision not to afford NUREG-0760 or its conclusions any weight. Gamble was an investigator f rom the NRC's Of fice of Inspector and Auditor assigned to the investigation to protect the interests of the Department of Justice, to ensure that any i

information be preserved which might be useful to any future i 1

criminal prosecution. Tr. 30,510 .

I 281. He made the following serious criticisms of the inves-i tigations

1) Moseley directed that the three major portions of the investigative report be drafted prior to any significant investigation;

2) Moseley attempted to restrict full and complete

, questioning of witnesses by imposing a protocol where interviewers could only ask questions on a pre-approved list; follow-up questions by other than the chief interviewer were permitted only at the end and after

! being approved by Moseley; and in some cases by entering into agreement with corporate counsel whereby the areas of question were restricted. Gamble ff Tr.

30,587 at 3-5; Tr. 30,548; 30,559; 30,561-564; 30,579-580; 30,660 (Gamble);

3) At least one original member of the Task Group, Ronald Haynes, appeared to have a conflict of interest.

Tr. 30,729 (Gamble);

4) Significant information which was already on the public record was never discussed during the course of the investigation and did not appear in the final report. Tr. 30,531 (Gamble);

5) Moseley did not employ investigative techniques which would have led to a better development of the factual record. Tr. 30,706-711 (Gambl e); TMIA Mallgram

, Exh. 2 2, Enclosu re 2.

4 i

, 102 4

y,,r r. y --,m,- -+-,w. - , , , . . , - . - - - - - - + - - - . , . , , - , - - - , ----n--.--w,- - , -

282. G:mble's criticisms lead us to find that the conclu-sions of NUREG-0760 are not supported by the facts. Moreover, Gamble's criticisms of NUREG-0780 provide further reason for this Board's refusal to credit the conclusions of that report. These include:

1) Significant facts were left out.of the final report or not fully developed, Tr. 30,532; (Gamble);

2) The report's characterization of Plumlee's testimony did not give it adequate weight, Tr. 30,714-719 (Gamble);

3) The Task Group's conclusions were not adequately supported by the facts, including conclusions regarding whether anyone in the Unit 2 control room properly interpreted the pressure spike, Tr. 30,804. (Gamble)

See TMIA Mailgram Exh. 24 at 1.

283. We give no weight to NUREG-0760, its conclusions or l Mos el ey's testimony on the issue before us.

I E. Lowe's Alleged Discovery of the Significance of the l Pressure Spike Late on March 29, 1979.

284. Licensee argues that Dieckamp was entitled to rely in sending his mailgram, on his understanding that William Lowe was the first to correctly interpret the pressure spike to indicate a hydrogen burn and core damage. Dieckamp ff Tr. 28,316. Lowe claims he was the first person to discover the significance of the pressure spike at 11:00 p.m. on March 29, 1979.

285. Lowe tes ti fied that on March 29 he-f ollowed Uni t 2 Superintendent Joseph Logan into the Control Room when operators had lost control of the pressurizer level. At that time Bensel~

showed him the containment building pressure strip chart trace showing 28 psig at 1:50 p.m. on March 28, 1979. Lowe says that he concluded immediately that the spike was caused by a hydrogen ignition and the hydrogen had been produced from a zirealloy-

.103

1 water reaction. Lowe testified that he asked for a second pres-sure reading and was pointed to the wide range trace at the bottom of the same chart. He also reviewed building. temperature i traces which confirmed the spike. Lowe, Thomas Crimmins and Moore then calculated the bubble size and the amount of zirconium cladding which would have to oxidize to produce that amount of hydrogen. Lowe contends that from these calculations they con-cluded the core was very seriously damaged. Lowe f f Tr. 6-7; 11-13.

286. The Board is not convinced that Lowe was the first person to discover the significance of the pressure spike. We have seen substantial evidence in this proceeding hearings that both Met-Ed operations personnel and GPUSC engineers reviewed and understood the significance of the spike on the first day of the accident.

287. Further, we believe ' that Lowe's testimony itself con-tains many contradictions. For example, although Lowe now con-tends he was the first to correctly analyze the spike, in a conversation in 1979 with a Special Inquiry Group investigator he explained in response to a direct question that in fact he did not know whether he was the first to recognize the signif nee of the spike. Tr. 28,154-155; 28,157-158 (Lowe); JME l-C (104).

288. Moreover, his prefiled written testimony is clearly misleading in seriously understating his understanding of the seriousness of the accident on March 28. Lowe states in his testimony that Thorpe informed him at about 4:20 p.m. on March 28 that' " core cooling is recovered". Tr. 28,160; Lowe ff Tr. 28,151 a t 3. Yet according to a memo Lowe dictated to the file on March 28, Thorpe in f act reported to Lowe at that time, that " Plant 104

t thinks core is recovered, but proof not yet es tabli shed." TMIA .

Mailgram Exh. 1.

289. The Board understands Thorpe's statement in this memo-randum to indicate that at some time prior to 4:20, Licensee's staff believed the core was uncovered. The language in this memorandum is not susceptible to Lowe's peculiar rephrasing --

that cooling of the core has been restored or reinitiated. Tr.

28,159-163 (Lowe). Lowe's artful phrasing of Thorpe's assessment of the status of the plant at that time downplays the seriousness of.the accident and casts doubt generally on Lowe's credibility.

290. The Board also finds Lowe's story incredible because it appears that there were general discussions about the pres-sure spike, hydrogen burn and hydrogen build-up in the reactor ,

building during the afternoon of March 29, fully eight hours prior to Lowe's alleged revelation. The first meeting of the Task Force was held at 3:30 p.m. on March 29 in the processing center at Unit 1. According to Lowe's calendar of activities for the early days of the TMI-2 accident, the Task Force was divided i n to two "t eams". The Events Analysis Team was composed of the following persons:

Richard Wilson, Chairman Ed Wallace Donald Reppert l Gary Broughton l

George Kunder James Moore Lee Rogers 105 l

The Recovery Planning Team was composed oft Ron Williams, Chairman -

Thomas Crimmins William Lowe D. Klingeman (not present)

Robert Long TMIA Mailgram Exh. 18 at 3.  !

291. As discussed in section III, supra, Wilson opened the meeting by stating that the company had assumed there had been core damage in the range of $20 to $30 million, and that a one-year outage was anticipated. Although different individuals attending the meeting have differing recollections about the meeting, what is striking is that the pressure spike, hydrogen ourn, or production of hydrogen to flammable limits was discussed in some manner.

292. Abramovici, for example, stated a concern that there 0'

-was a hydrogen build-up in the reactor building to four percent which led to a discussion about hooking up a hydrogen recombiner.

c TMIA_Mailgram Exh. 32H; JME l-C(50) at 12-13. See also JME l-C (78) a t 128 (Floyd); JME l-C (140) a t 76 (Plumlee); JME l-C(22)-

a t 34 (War ren). -

293. It is clear that the only method for hydrogen produe-I tion up to four percent of the total containment volume in two l

days is through a zirconium-steam reaction. Tr. 28,198-200 (Lowe). Therefore, the predicate to any discussion about hydro-t gen build-up in the containment to flammable limits would be an understanding that a zirconium-steam reaction had occurred, leading to oxidation of the zirconium cladding and the production 106 l-

of significant amounts of hydrogen.

294. Crimmins, in an answer to a TMIA discovery request, stated that he remembered that the reactor building pressure

trace was viewed and discussed at the Task Force meeting, but discounted as spurious and due'to an instrumentation problem.

l TMIA Mailgram Exh. 32F. Clearly if the pressure trace had been s viewed and discussed at the March 29 af ternoon meeting, Lowe, I

who attended that meeting, should have instantaneously inter-preted the spike to indicate a hydrogen burn.29 j 295. Crimmins, in answering a questionnaire about his awareness of communications about the pressure spike, containment 1 i

sprays and hydrogen burn on the first three days of the accident, l

. stated-the-following

I cannot remember such details.

The subjects were ini-i tially discussed by me and other technical support per-sonnel (R. Williams, W.W, Lowe & others) on the af ter-noon of March 29 and into the evening. The discussions were initiated by a briefing by George Kunder on the afternoon of March 29 and were the subject of evaluation

) and analyses and extensive open discussion with all involved parties (Met Ed, GPU, NRC snd others from

.29 The Board believes that Met-Ed si te personnel had the same capabilities collectively to. interpret the pressure spike correctly at the time it occurred as did Lowe. In fact, some engineers state so explicitly. For example, Porter stated that he believed the pressure spike was real the first time he saw it; he also stated that he' confirmed the reactor building pressure by

-reviewing the reactor coolant system pressure which would indicate a contemporaneous spike downward. TMIA Mailgram Exh. ,

32C at At t. 3. Similarly Bensel stated he reviewed the alarm printout immediately af ter seeing the pressure spike to determine whether or not the spike was in fact a real increase in pressure.

TMIA Mailgram Exh. 32E. Therefore, at least two Met-Ed lead engineers were capable-in the time period of the accident of correctly interpreting the pressure spike to indicate a real increase in pressure and hydrogen burn.

107

thot time en.

1MIA Mailgram Exh. 19 at 11.

296. Crimmins' answer seems to indicate that by the initial meeting of the Task Force on Thursday afternoon the spike was generally understood to indicate a hydrogen burn and core damage.

297. Runder also recalls that the pressure spike was dis-4 cussed at the first meeting of the Task Force. Tr. 29,998-999 (Kunder) In testimony in this hearing, Kunder remembered that in a side discussion Broughton showed him ei ther the original or a copy of the pressure trace and suggested that one possible cause of the spike was a hydrogen explosion. Tr. 30,001-007 (Broughton); JME l-C-3 7 a t 5 0- 51; JME l-C-80 at 74-75. Kunder testified at a prior time that as a result of these discussions he asked Bensel to de-energize electrical equipment in the reactor building. JME 1-C (118) at $2. This is corroborated by a 9 :30 p.m. entry in Seelinger's notes which indicates Bensel carrying out such an instruction. TMIA Mailgram Exh. 2 at 7.

i

-298. In addition, Abramovici testified that because of a concern that hydrogen levels in the reactor building may have reached four percent, the design limit at the time, the group discussed consulting with Atomics International in order to hook up a hydrogen recombiner. TMIA Mailgram Exh. ' 32H st 44-48. An entry in the Unit 2 control room log verifies that the hydrogen recombiner was started up at 8:55 a.m. TMIA Mailgram Exh. 16 at 630, 30 -

- Licensee presented testimony f rom James Hentle of Atomics International to demonstrate that licensee did not set up the hydrogen recombiner until Monday, April 1, and that Atomics (continued on next page) 108

i i

l 299. Mr. Henrie's memory of the events seemed very poor.

We accept Abramovici's testimony over Henrie's in light of the fact that Abramovici worked for GPUSC and was at the site during the entire period in question. Certainly he would know more about GPU's arrangements to secure and start a hydrogen recombi-ner than someone who is located on the Wes t coast and called to TMI-2 as a consultant.

300. Moreover, Dieckamp's notes of March 30, 1979 verify that licensee made ef forts to contact Atomics prior to the time Henrie claims. TMIA Mailgram Exh. 27 at 3. See also JME l-C (78) at 128 (Floyd).

301. Broughton and Wilson remember no discussion about the pressure spike or hydrogen burn at the Task Force meeting at 3:30 p.m. on March 29. Tr. 31,159 (Broughton); 31,530-531 (Wilson).

302. The Board finds from the testimony of Abramovici, Kunder and Crimmins, that there was discussion of the pressure spike and hydrogen burn at the March 29 afternoon meeting of the Task Force. Further the pressure spike trace may have been reviewed at the meeting. This conclusion is compelled by the consistent testimony of these three central individuals who came l from different corporate organizations. The Board also con-eludes that the group discussed setting up of a hydrogen recombi-ner to deal with hydrogen greater than the containment-building design limit of four percent. There appeared to be general TTootnote continued from previous page)

International personnel were not asked to come to the site until March 31. T r. 3 0,010 -014. Presumably licensee had no serious concern about hydrogen in the reactor building, licensee argues, until that time.

109

t understanding by members of this group, including Lowe and Kun-der, that the only means to produce within two days hydrogen greater than the containment design limit was by means of a zire-water reaction. Tr. 30,075-077 (Kunder); Tr. 28,197-200 (Lowe).

Therefore we find that members of the Task Force, including Lowe, Broughton and Kunder, determined as early as 3:30 p.m. on March 29, that the pressure spike was caused by a zirconium-water reaction which produced build-up of hydrogen to flammable limits.

303. We do not find credible Licensee's claim that Lowe was the first-to understand the pressure spike during the late even-Ing of March 29 not credible. We find further that that this theory fails to support Dieckamp's claim that the " thrust" of his mailgram is correct.

F. Information Available to Dieckamp Prior to May 92 1979.

304. The Board finds that several_ operator interviews available to Dieckamp prior to May 9 indicated that the opera-tions staff interpreted the pressure spike to be an explosion and in response took steps to repressurize the reactor, a serious departure from the strategy they had previously employed. These interviews indicate to the Board "some evidence" that the spike was properly understood to have been caused by a hydrogen burn.31 31 Licensee has argued that it was difficult for operators to make the leap in reasoning from the fact that the pressure spike was real or an explosion of some sort to the conclusion that it

! was ceaused by a hydrogen burn. Dieckamp ff Tr. 28,316 at 14-15; 28,676; 28,849-850; 28,857. See also Moseley ff Tr. 29,816 at 4, which states that it was beyond-the range of credible operator knowledge to infer the production of hydrogen to flammable limits within~10 hours of the event initiating the accident.

(footnote continued on next page)

I i 110 l

l

305. First, ' Walter "Bubba" Marshall's notes of dictation he took f rom operator McGovera at 3:00 a.m. on March 29 indicate McGo-vern understood the pressure spike and actuation of containment sprays to be indicative of a real increase in pressure. They.show also that operations personnel changed the strategy for stabilizing the reactor to a repressurization evolution irmnediately af ter the pressure spike. These notes, at JME l-C (1) at 2:

1400 Had a load of 2-32A & 2-42A loss of radiation area monitors and Rx building pressure spike that went off scale on narrow range meter-definite spike straight up, straight back down - had meter - definite spike straight up, straight >

back down - had full Rx building (Spray pumps & BS-VPS, DH-V8's) isolation and cooling. Someone secured spray pumps, shut BS-Vi's and DH-V's (Hugh did) and unisolated equipment for br'. ding.

! 1500 Took the panel...PZR heaters on at this time with RC-

R-V2 open. Shut stopped MU-P-IC and shut MX-VI6C. About 10; minutes later ThA loop came back down again. PZR level dropped to 395. At this time, isolated RC-RV2. By shutting RC-V2. PZR level started to drop again in about 5 minutes, and dropped rapidly to 150".

306. In this interview McGovern observed the pressure spike i

and contemporaneous actuation of the containment sprays and did not question, that it was a real increase in pressure. Moreover, McGovern recounts to Marshall actions which were taken a short time later to draw a bubble in the pressurizer. The only preceding event which could have triggered this abrupt change in

. T7ootnote contTnued from previous page)

However, as Dieckamp himself testified, there is no other good explanation for an explosion er reactor building pressure spike to 28 psi and simultaneous actuation of containment sprays other than a hydrogen burn or explosion. Tr. 28,964-966 (Dieckamp). Further, Moseley testified on cross-examination that .

"beyond the range of credible operator knowledge" did not meant it was impossible for an operator to interpret the spike to indicate a hydrogen burn, but only that it would be a " rare" operator who would do so. Tr. 29,853.

111

,,r., ,,. . . . _ _ . _ _ , _ _ _ . . ,

strategy would be the pressure spike.

307. Since we know of no other plausible explanation for a sudden pressure increase of this nature other than a hydrogen burn or explosion, the Board considers this interview to be "some evidence" that Marshall properly interpreted the pressure spike at the time it occurred.

308. Moreover, operator Craig Faust describes the spike in the following way in a company interview on April 6, 1979:

TEAM: What about the reactor containment bldg. spike?

FAUST: We had probably had some sort of explosion because that's what it looked like - shock waves.

TEAM: Did you hear anything?

FAUST: No. I didn't.

TEAM: Did it affect the pumps?

FAUST: There d i dn' t seem to be any change in anything.

When the pressure dropped right of f, we s topped the bldg.

spray pumps, we didn't think we needed them.

TEAM: The spray pumps came on. automatically?

'l FAUST: Right.

JME l-C (8) at 5-6.

309. The Board believes that Faus t's descript ion of the spike as "some sert of explosion" and " shock waves" indicates that he understood the spike was a real increase in pressure and similar to an explosion. It appears that the only cause of such an explosion could be the combustion of hydrogen or noncondens-able gas. Therefore, we believe that this interview constitutes se evidence that the spike was properly interpreted as a hydro-112 l

l

gen burn cnd dsmanstrative of core damage. 32 .

G. Conclusion -

310. We find Dieckamp should have known that the statements in his mailgram were, and are today, false. Moreover, wi th minimal investigation, he would have discovered that licensee personnel properly interpreted the pressure spike as a hydrogen burn and and in response changed to repressurize to stabilize the reactor. The Board finds unconvincing licensee's arguments in support of the "thrus t" of Dieckamp's mailgram.

311. Licensee has continued to defend the accuracy of the mailgram. The Board believes that Dieckamp and licensee's in-transigence on this point, given the extensive evidence presented of Met Ed and GPU Service Corporation awareness of the signifi-cance of the pressure spike on March 28, 1979, refleets poorly on management competence and character.

l V. DIECKAMP SHOULD HAVE CORRECTED THE MAILGRAM ONCE HE DETERM1NETTWilt ETATEMERYB g gisllq !dkitYFilSE.

312. Licensee argues in its Proposed Findings that "given the fact that the mailgram was accurate when sent, that its thrust remains a reasonable conclusion today, and that all subse-quently adduced contrary evidence was fully known by all concerned, it was certainly unnecessary for Mr. Dieckamp to inform the mallgram recipients that the prefatory phrase 'there 32 Dieckamp testified that even though Faust, in this interview, indicated he understood the spike was real and was a type of explosion, Dieckamp was "not able to go beyond that in terms of any significance he might have ascribed to i t ." Tr. 28,674.

However,-Dieckamp could not come up with any other plausible explanation for an explosion other than the actual one -- a hydrogen burn. Tr. 28,964-966. See also JME I-C(ll4) at 2.

113 1

is not evidence' was no longer l i t e r al." Licensee's Proposed Findings of Fact and Conclusions of Law (January 28, 1985), at 94.

313. None of the statements contained in the above conclu-sion i s true. As discussed in Part II, supra, the Board believes Dieckamp did possess evidence, and in fact positive knowledge, ,

that Met-Ed and GPU Service Corporation personnel properly interpreted the pressure spike as a hydrogen burn on the first day of the ac., dent. Therefore Dieckamp knew at the time he sent it that he was making false and inaccurate statements in the mailgram.

314. Even if he did not know on May 9, 1979 the statements were false, the reality is that on May 9, 1979, persons such as Chwastyk, Mehler and Illjes had properly interpreted and re-sponded to the pressure spike at the time it occurred. If Dieckamp had done any investigation he would have discovered this substantial-evidence at that time. See Section IV, supra.

315. Moreover, Dieckamp's statement that the " thrust" of the mailgram is correct although the words may be literally false makes no sense to this Licensing Board. Licensee and its top l management are obliged to be meticulous i n fully disclosing all l material information within their possession to the NRC in order I

for the NRC to carry out its mission to protect the public health and safety. That responsibility cannot be any more important than during an accident. Thus licensee's failure to provide informa-j tion about the pressure spike and hydrogen burn to the Commission i,

on March 28 seriously compromised the Commission's ef fectiveness in carrying out that mission. This fact licensee does not appre-l 114 t

l

'&- %a- Am"-9 em~esv ey e- y -w-,,wm.g- -

, s-,n.

y y wm-,---.- e -- e w ,g,-m-v- -wg,w.,--,,4 - - - -

elate, even today, after multiple investigations, inquiries and hearings on this matter. -

VI. CONCLUSION.

316. The Board concludes that Dieckamp's mailgram of May 9, 1979 contains false statements since licensee personnel did properly interpret the pressure spike and containment sprays to indicate a hydrogen burn and core damage. Moreover, the highest levels of licensee management, including Dieckamp, were aware of the pressure spike and its significance by early on March 29.

L Yet licensee f ailed to disclose this information to the NRC or to Commonwealth of Pennsylvania authorities until March 30. The Board therefore f!nds false Dieckamp's claim in the mailgram that there was no withholding of information.

317. The Board has also found from the extensive record developed in this hearing, that Dieckamp knew that his mallgram was' false at the time he sent it since he was informed during the af ternoon of March 28 by Herbein, Miller and Kunder of their

, understanding of the pressure spike, and later on the morning of March 29 through Keaten of the GPUSC group's evaluation of the hydrogen burn.

318. Even if Dieckamp did not know on May 9, 1979 that the s tatements in his mailgram were f alse at the time he sent it, he should have known given the extensive information available to l

the company that licensee employees understood the pressure spike '

to be a hydrogen burn and responded by changing to a repressuri-zation evolution. A minimal investigation would have uncovered

-the simple fact that the mallgram contained false statements.

115 s

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

319. We conclude licensee management has demonstrated ex-tremely poor character in continuing to defend false statements in Dieckamp's mailgram of May 9, 1979. Licensee's refusal to accept reporting responsibilities to the NRC demonstrates a basic lack of integrity which prevents this Board from finding licensee management has adequate integrity and competence to operate TMI-l safely.

Respectfully submitted,

/ '

~M W7tf (N.L ~~

j oanne Doroshow .

The Christic Institute 1324 North Capitol Street Washington, D.C. 20036 s.) p JI . Art A' _ 'c Mt w sM Ly rie Bernabel Geo rge Shohet Godernment Accountability Project 1655 Connecticut Avenue, N.W.

Sulte 202 Washington, D. C. 20036 Attorneys for Three Mile Island Alert DATED: FEBRUARY 8, 1985 116 9---- v w- - -m. +94.-gw-. c.w ygy-p-p--7,w-- 3.+.--p--.----..- -,. -, ,. ,yagw,, . , - - p +,,,wn- --,.cy,-v --.,w.--.ym,- . - , - - --.g

( TABLE OF CONTENTS I. INTRODUCTION ........................................... 1 II. Is the Dieckamp Mailgram False or Inaccurate............ 7 A. Control room operators and management understood the significance of the s occurred................. pike at the time it

. . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1. General awareness of pressure spike and related events by TMI personnel................. 9

2. Interpretation of the spike by Joseph Chwastyk and Brian Mehler.......................ll

3. Interpretation of the spike b and John Herbein.............y Gary Miller . . . . . . . . . . . . . . . . . . . 27

4. The pressure spike and hydrogen burn were common knowledge amon

. March 28.............g site personnel on

. . . . . . . . . . . . . . . . . . . . . . . . . . . 39

5. Awareness of GPUSC engineers of pressure spike

(, and hydrogen on March 28 and 29, 1979.......... 52 B. Conclusion concerning false statements in the "Dieckamp Mailgram"................................ 62 III. -Dieckamp Knew at the Time He Sent the Mailgram on May 9, 1979 that the Mailgram Was Inaccurate and contained Material False Statements.............................. 64

! A. The purpose of the mai1 gram........................ 64

(-

B. Dieckamp's knowledge of the interpretation of the pressure spike by plant and corporate staff.... 66

1. Miller-Herbein discussions with Dieckamp....... 66 C. Broughton-Keaten communication with Dieckamp....... 76

1. Broughton reported to Keaten................... 77 L 2. Keaten reported to Dieckamp.................... 81 l

l' 3. Dieckamp's awareness of core damage on March 28

.and March 29, 1979 demonstrates his awareness i(

of.the hydrogen burn........................... 82 D. Conclusion......................................... 86 l

r

.v.,__.y * ,. ,g ,

.9-,9 ..,

a - -n.-., . . - yem., -

. _.. , - =. . .. . - _ _ _ _ _ - . . , . . .

1 1

1

('

IV. Assuming that Dieckamp Did Not Know the Statements in His Mailgram Were False at the Time He Sent the Mail-gram He Should Have Known They Were False.............. 87 A. Dieckamp's unduly restrictive definition of the issue before this board............................ 88 B. A complex technical analysis is needed to determine the pressure spike was caused by a hydrogen burn...................................... 90

, C. Dieckamp's analysis of Chwastyk, Mehler and Illjes'

. interviews......................................... 93 D. The IE investigation and report on information flow during the TMI-2 accident..................... 98 E. Lowe's alleged discovery of the significance of the pressure spike late on March 29, 1979............. 103 F. Information available to Dieckamp prior to May 9, 1979.............................................. 110 G. Conclusion........................................ 113 V. Dieckamp Should Have Corrected the Mailgram Once He Determined that Statements in the Mailgram Were False. 113 VI.

Conclusion............................................ 115 i

Appendix A: Acceptance Criteria for Emergency Core Cooling Systems for Light Water Nuclear Power Reactors (Fed. Reg.)

SUPPLEMENT TO LICENSEE'S APPENDIX B: List of Exhibits SUPPLEMENT TO

! LICENSEE'S APPENDIX C: TMIA Adopts Licensee's Appendix C and.

, Supplements

)

If s

i UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION Before the Atomic Safety and Licensing Board In the Matter of )

)

METROPOLITAN EDISON COMPANY ) Docket No. 50-289 SP

)

(Three Mile Island Nuclear ) (Restart - Management Phase)

Station, Unit No. 1) )

)

)

I hereby certify that a copy of the Corrections and Additions to TMIA's Proposed Findings of Fact and Conclusions of Law of the Dieckamp Mailgram Issue has been served this lith day of February, 1985, by mailing a copy first-class, postage prepaid to the following:

Service List Administrative Judge Docketing and Service Section (3)

Ivan W. Smith, Chairman Office of-the Secretary Atomic Safety &' Licensing Board U.S. Nuclear Regulatory Commission U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Washington, D.C. 20555

[_ .

-Atomic Safety & Licensing Board Administrative Judge Panel Sheldon J. Wolfe U.S. Nuclear Regulatory Commission Atomic Safety-& Licensing Board Washington, D.C. 20555

.U.S. Nuclear Regulatory Commission Washington,-D.C. 20555 . Atomic Safety & Licensing Appeal Board Panel Administrative Judge U.S. Nuclear Regulatory Commission Gustave A. Linenberger, Jr. Washington, D.C. 20555 Atomic Safety & Licensing Board U.S. Nuclear Regulatory Commission

• Jack R. Goldberg, Esq.

Washington, D.C. 20555 Office of the Executive Legal Director L

U.S. Nuclear Regulatory Commission l Washington, D.C. 20555 1

t l' A, Thomas Au, Esq. Joanne Doroshow, Esq.

Office of Chief Counsel The Christic Institute Department of Environmental 1324 North Capitol Street Resources Washington, D.C. 20002 505 Executive House P.O. Box 2357 Michael F. McBride, Esq.

Harrisburg, PA 17120 LeBoeuf, Lamb, Leiby & MacRae 1333 New Hampshire Avenue N.W.

• Ernest.L. Blake, Jr. Suite 1100 Shaw, Pittman, Potts & Trowbridge Washington, D.C. 20036.

1800.3 Street, N.W.

Washington, D.C. 20036 Michael W. Maupin, Esq.

Hunton & Williams Mr. Henry D. Hukill 707 East Main Street Vice President Post Office Box 1535 GPU Nuclear Corporation Richmond, VA 23212 P.O. Box 480 Middletown, PA 17057 Ellyn R. Weiss, Esq.

William S. Jordan, III, Esq.

TMI Alert Harmon, Weiss & Jordan 315 Peffer Street- 2001 S Street, N.W.

Harrisburg, PA 17102 Suite 430 Washington, D.C. 20009 Mr. and Mrs. Norman Aamodt R.D. 5 TMI-PIRC Legal Fund Coatesville, PA 19320 1037 Maclay

'~ Harrisburg, PA 17103 Ms. Louise Bradford TMI Alert 1011 Green Street Harrisburg, PA 17102

/ 3 b %d b u h .-

LyngeBernabei.

• Hand Delivered h

1

,o i as ,

/

t

..