ML20155J808

From kanterella
Jump to navigation Jump to search
Documents Results of Joint Ofc of Investigations/Nrr Interview W/Ed Hemmila Re RCS Leak Rate Surveillance Testing During 780930-790328.Hemmila Unaware of Problem.Transcript of Interview & Leak Rate Evaluation Encl
ML20155J808
Person / Time
Site: Three Mile Island Constellation icon.png
Issue date: 09/26/1985
From: Capra R
Office of Nuclear Reactor Regulation
To: Russell W
Office of Nuclear Reactor Regulation
Shared Package
ML20151L202 List:
References
NUDOCS 8605270206
Download: ML20155J808 (6)
v • d • e


Text

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

Enclosure 16

  1. ps* ,'49'e UNITED STATES

)j E NUCLEAR REGULATORY COMMISSION r, wAsMINGTON, D. C. 20655 September 26, 1985 j.F

*..~.* _ i s

~

MEMORANDUM FOR: William T. Russell, Acting Director ~

Division of Human Factors Safety, NRR l

! FROM: Robert A. Capra, Technical Assistant

]

Division of Human Factors Safety, NRR l 3

SUBJECT:

SUMMARY

OF JOINT 01/NRR INTERVIEW WITH EARL D. HEMILA i i  :

The purpose of this memorandum is to document the results of the joint 01/NRR interview with Mr. Earl D. Hemila. Mr. Hemila was interviewed in order to obtain additional background infomation on the subject of reactor coolant system (RCS) leak rate surveillance testing at TMI-2 during the period -

September 30, 1978 to March 28, 1979. During a portion of this period, Mr. Hemila stood watch with two of the individuals (Messrs. H. A. McGovern and C. L. Guthrie) who are currently subjects of the joint 01/NRR investi-gation into leak rate testing irregularities at TMI-2 prior to the accident.

The interview with Mr. Hemila was held on March 28, 1985 in the law offices l of Killian & Gephart in Harrisburg, Pennsylvania. The interview was j conducted under oath in the presence of Mr. Hemila's personal attorneys:

Mr. Harry H. Voigt and Mr. James W. Moeller of the law fim LeBoeuf. Lamb, i

Leiby & MacRae, Washington, D.C. The interview was conducted by Mr. R. Keith Christopher, Director, Office of Investigations. Region I and me. A copy of a the transcript of the interview is attached as Enclosure 1. The page j references cited in this sumary refer to the pages of Enclosure 1 from which i the information was extracted. Enclosure 2 contains a copy of the NRR j Evaluation of TMI-2 Reactor Coolant System Leak Rate Tests Performed Between September 30, 1978 and March 28, 1979 (Updated July 30,1985).

]

After being placed under oath, Mr. Hemila was asked to describe his i employment history. Prior to joining Metropolitan-Edison Company (Met-Ed) in the fall of 1976, Mr. Hemila spent six years in the U. S. Navy. Mr. Hemila was originally hired by Met-Ed as an Auxiliary Operator (AO) at TMI-2. He ,

, served in that capacity until entering the Control Room Operator (CRO) -

training program for TMI-2. He received his Reactor Operator's (RO) license in the Fall of 1978. In 1980, he obtained his Senior Reactor Operator's ,

! (SRO) license and was promoted to Shift Foreman. In 1981, he advanced to Shift Supervisor. Mr. Hemila left Met-Ed on November 30, 1982 and is currently employed as a Senior Start-Up Consultant with Nuclear Start-Up '

Services, Ann Arbor, Michigan. See pages 2-4.

l -

NOIL: This memorandum and Enclosure 1 discuss information which is the subiect of an ongoing investigation. This memorandum and Enclosure 1 may not be disseminated outside the NRC without coordination with NRR and the permission of the EDO or the Director of 01. Internal access j and distribution should be on a "need to know basis."

! B605270206 860516 i Qa PDR ADOCK 05000320 j 11 P PM

William T. Russell Ssptember 26, 1985 O For a short time after receiving his license, Mr. Hemila was assigned to "B" shift; however, when a sixth shift (Shift "F") was created on January 1, 1979, Mr. Hemila was assigned to that shi'ft. He remained with Shift "F" until the 1tccident. While assigned to Shift "F" the members of his shift '

included: .

POSITION NAME __

Shift Supervisor: Ken Bryan Shift Foreman: Carl Guthrie CR0 Hugh McGovern CR0 Earl Hemila CRO-in-training Leonard Germer See page 5.

In describing his shift organization, Mr. Hemila stated one CR0 would normally assume the panel, one CR0 would handle " switching and tagging" duties and the third CR0 would take the log readings. The CR0s worked for the Shift Foreman, whose desk was in the TMI-2 control room. The Shift Foreman reported to the Shift Supervisor. Since the Shift Supervisor was responsible for both TMI-1 and TMI-2, he would nonnally split his time between units while on shift. Mr. Hemila believed that the Shift Foreman and Shift Supervisor were cognizant of the actions of the CR0s during the shift. See pages 5-7.

Mr. Hemila statedofthat he did not normally)have much interaction withwas the TMI-2 Supervisor Operations (James Floyd . He believed that Mr. Floyd knowledgeable about evolutions that were taking place in the plant; however, he did not deal with Mr. Floyd on routine operational matters. Hemila essentially had no dealings with other members of management such as the TMI-2 Superintendent of Technical Support (Jim Seelinger and later George Kunder), the TMI-2 Plant Manager (Joe Logan), or the Station Manager (Gary Miller). He did not recall any meetings with members of management to discuss prcblems associated with leak rate surveillance testing. See pages 7-10.

)

After reviewing copies of TMI-2 Technical Specification 3/4.4.6.2, " Reactor l

Coolant System Operational Leakage," and TMI-2 Surveillance Procedure

2301-301, "RCS Inventory," Mr. Hemila stated that the technical

! specifications required a satisfactory leak rate test be perfonned once every .

72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. Nevertheless, depending on the workload, they tried to run at

! least one test every shift. Leak rate test results that exceeded the limits of the technical specifications were discarded. Hemila said operators believed it was okay to discard bad test results since the technical specifications required a satisfactory test only once every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. See pages 11-14.

While Mr. Hemila believed that the leak rate test was a "somewhat valid" indicator of plant leakage, he did not believe it was possible for the test to accurately detennine unidentified leakage within 1 gpm. Thus, he did not

think there was anything wrong with throwing away results that exceeded the limit. He said he personally felt the plant was safe regardless of the results of the test. He did not recall ever entering the action statement,of

, _ _ _ _ _ _ _ _ _ . _ _ _ _ - - _ _ _ - _ ~ _ . _ _ _ _ . _ , ~ _a.______

~

William T. Russell September 26, 1985 t

f the technical specifications as the result of an unsatisfactory leak rate test. According to Hemila, the leak rate test was only used to satisfy the 4

administrative requirements of the technical specifications, they used other

~ '

sources'to determine whether the plant was really leaking. See pages 15-18. _ l Mr. Hemil rsaid leak rate tests were simple to run and were performed in a very routine manner through the use of the plant computer. The operator -- l performing the test would try and establish stable plant conditions prior to l initiating the test. He would also normally inform the operator on the panel j that the test was in progress. While test results sometimes exceeded the

limits, he could not recall it being too difficult to obtain satisfactory test i results. See pages 19 and 20.

i

! Despite the requirement to log the start and stop time of all. surveillance tests required by the technical specifications, only satisfactory test ,

! results were logged. While Hemila could not recall much guidance being l f given in this area,'he said that he was never directed by anyone not to log j start times as an attempt to hide the fact they were having to run so many j tests. See pages 21 and 22.

i I On their shift, Mr. Hemila said the Shift Foreman (Carl Guthrie) usually saw -

or was made aware of all test results whether they were satisfactory or

unsatisfactory. While unsatisfactory tests were discarded Hensnila said

! there was never any indication that tests were being discarded in order to hide them from the NRC. See pages 23 and 24.

! A copy of LER 78-62/1T along with an attached routing sheet was provided to-l Mr. Hemila for his review.

NOTE: During a routine inspection of TMI-2 operations on October 18, 1978 an NRC inspector discovered several bad leak rate tests lying in the i control room and that TMI-2 had been operating for an extended period of time with unidentified leakage exceeding the technical specification

! limit. The incident resulted in the submittal of Licensee Event Report l (LER) 78-62/1T on November 1, 1978.

1 The LER states in part: "This event was caused by a misinterpretation of the requirements of the technical specifications....The appropriate personnel will be instructed on the requirements of applicable sections of the T.S. and the requirement to imediately invoke applicable action statements when the provisions of the LCOs [ limiting conditions for i operation]arenotmet."

i

Despite his initials on the routing sheet, Mr. Hemila had.no preaccident recall of the incident or the LER. He did not recall being directed at any time to discontinue throwing away bad leak rate tests, and he did not recall being instructed on the proper interpretation of the technical specifications. As best as he could recall, leak rate testing practices were the same at TMI-2 from the time he received his license until the date of the
accident. -See pages 24-30.

! When unsatisfactory test results were received, the vnly follow-up action Mr. Hemila would take was to infonn the Shift Foreman of the results. It I

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

1

, William T. Russell September 26, 1985 1

l was pointed out to Mr. Hemila that if a test was considered invalid, the

procedure required the operator to note that in the remarks section of i the test and to describe the action that invalidated the measurement.

I Mr. Hemila said that action was not taken. He said the procedure was ,,

I probably not followed in this case because the test was considered such a .. ..

l, routine; everyday task that no one put much emphasis on it. jeepages30-32.

j Next Mr. Hemila was questioned about his knowledge of leak rate manipulation t by operators through the use of unrecorded or under-recorded water additions. '

1 Mr. Hemila stated that he did not know of anyone who intentionally added i watertothemake-uptank(MUT)inanefforttoinfluencetheoutcomeofa 4

leak rate test. Except for Hartman's allegations, he said that he had no reason to believe that such actions were occurring at the plant. Hamila stated that as far as he knew, if water was added during a leak rate test and 3

not properly included in the test calculation, it was due to operator error i and not done to intentionally influence the outcome of the test. See pages 34-39. .

f

! When adding water, Mr. Hemmila stated that he would dial in the amount of 4

water to be added on the batch controller. The amount of water logged in the CR0's Log and included in the leak rate test calculation would be taken off the batch controller. Mr. Hemila said that he would only use the MUT strip chart value to confirm that the amount being added was approximately what it should be. Mr. Hemila stated that he was not aware of any significant

' difference between the amount of water indicated on the totalizer compared j with the amount indicated on the MUT strip chart. See pages 39-44.

Mr. Hemila was informed that for a considerable period of time during the 1 months of December 1978 ar.d January 1979, one of the two MUT level trans-mitters (LT-1) provided a very erratic and unreliable output. While

{ Mr. Hemila could not recall any significant problem with the level trans-l mitters, he stated it would most likely have been placed out-of-service and j marked with an out-of-service sticker.. He said whether it was actually declared out-of-service or not, prudent operating practice would dictate that you would use the stable level transmitter to obtain MUT level data for leak rate tests and not use the unstable transmitter, jg, pages 44-50.

1 Next Mr. Hemila was questioned about Hartman's allegation that operators used hydrogen additions to the MUT in order to influence the outcome of Ichk rate tests. Hemmila stated that prior to Hartman's allegations being made public, he had no knowledge that hydrogen could affect'MUT level and consequently leak rate test results. He said that operators were all sensitive to adding hydrogen because MUT pressure was one of the instruments-t

- the TMI-2 Plant Manager (Joe Logan) always looked at when he was in the .

I control room. If it was low, Logan would coment about it to the operators.

! As a result Hemila stated that he would typically call up one of the j A0s at the beginning of the shift and ask him to add hydrogen sometime during j

the shift.' See pages 50-51. .

4 Mr. Hemila did not believe there was any prohibition against adding hydrogen

! to the MUT-during the course of a leak rate test, even though the procedure

stated chemical additions should not be made during the test. Mr. Hemila i reiterated that he had no preaccident knowledge that hydrogen could influence i

i i

a__.._.__ _ - _ . _ _ , . . - . . . , - _ - . . _ _ _ _ _, . - _ _ _ . _ . _ _ - , . _ . - -

4 William T. Russell September 26, 1985 i

MUT level indication. He did not believe that others on his shift including Hugh McGovern or Carl Guthrie had that knowledge either. Mr. Hemila was l asked how operators all around him could be manipulating tests by the ,

addition- of hydrogen to the MUT and he not know about it. Hemila disagreed ~ -

that it was going on all around him but conceded that it cculd have been taking place on a limited scale, and he just didn't know about it. After the ,

allegations were made public, Hemila stated he was not contacted by anyone _.

. including McGovern or Guthrie who admitted they were aware that this was going on at the time. See pages 52-54 During the next portion of the interview, it was pointed out to Mr. Hemila i that.our review of plant records indicated he was involved in 15 leak rate -

1 surveillance tests that were retained by the licensee. Of the 15 tests, all but three contained actions that were contrary to the precautions and I, limitations of the approved test procedure. It was pointed out that the

other 12 tests involved the following actions:

l Number of Tests Action -

i 3 Water added near the end of the test. The amount of water included in the test calculation was significantly less i than the amount indicated on the MUT strip chart.

i 1 Water addition that was not included in the test calculation. The test also used input from the unstable MUT level transmitter (LT-1).

3 Feed and bleed operation during the test that was not I accounted for in the test calculation. One test also included the use of the unstable LT-1.

2 Hydrogen addition during the test. One test also included the use of unstable LT-1.

2 No water or hydrogen additions; however, unstable LT-1 was l used to provide input to the computer during the test.

! 1 Water or hydrogen addition (unable to differentiate).

See pages 62-66.

When asked if he could provide any additional information to help reconcile

! the differences between his testimony and the technical evaluation, Mr. Hemila stated:

I guess I would say that it bothers me when somebody casts i some doubt upon what I have done, because I've had a lot of i

contact in the past five or six years with the NRC. I think  ;

4 you can go back and talk to any of. them, and I have a very good-professional reputation with those people. -

l Now, I can't sit here and give you a blow-by-blow account and j say'this is why this happened, this is why this happene'd. I.

can t do that. But I can tell you that there was never any water or hydrogen added to alter that leak rate on purpose, where I said: Hey, I know I've got to do this to alter the leak rate. I never did that. See pages 68 and 69.

v

William T. Russell S;ptember 26, 1985 It was pointed out to Mr. Hemila that as the date of thr accident grew nearer. the identified leakage was approaching the 10 gpm technical sp cificition limit. Mr. Hemila stated that he was never concerned about whether the~ plant should be shut down and the leakage fixed. He said it never crossed his mind. He said the attitude he had was: " Hey listen,

~ -

there's people over me, and you've got the experts, et cetera. There's people looking at this problem. I mean, I just wasn't that concerned with _..-

it. I was worried about doing my day-to-day job." See pages 71 and 72.

Mr. Hemila was informed that according to the testimony of some operators, the leakage was affecting their day-to-day job by having to constantly recirculate the pressurizer to equalize the boron concentration, make frequent water additions and frequently pump down the reactor coolant drain tank (RCOT), instead of being able to con:entrate on.other important aspects of their job. Mr. Hemila responded by stating that other than adding a little more water ttjan usual, he did not consider it a big problem. See page 72.

9,L/d. Ce Robert A. Capra, Technical Assistant Division of Human Factors Safety, NRR l

Enclosures:

1. Transcript of Interview

] 2. NRR Leak Rate Evaluation l cc: B. Hayes ~

-l K. Christopher J. Lieberman J. Goldberg ,

i l

l l

. 1 O

e m i

l i

_ _ . _ . _ _ _ _ - _ _ . _ _ . _ _ - _ _ . . _ _ _ _ _ _ _ _ _ _ ______._ _ ._,.._._. _ _ ____ . _ . , . _ _ _ . _ _ _ . _

Retrieved from "https://kanterella.com/w/index.php?title=ML20155J808&oldid=3042848"