ML20155J710

Partially Withheld Results of Joint Ofc of Investigations/ NRR Evaluation of Di Olson Re Leak Rate Testing Irregularities.Denial of Role in Leak Rate Test Manipulation Questioned.Evaluation Encl.W/O Encl 2

ML20155J710
Person / Time
Site:	Three Mile Island
Issue date:	12/14/1984
From:	Russell W Office of Nuclear Reactor Regulation
To:	Harold Denton Office of Nuclear Reactor Regulation
Shared Package
ML20151L202	List: ML20151L198 ML20155J582 ML20155J591 ML20155J599 ML20155J625 ML20155J636 ML20155J641 ML20155J649 ML20155J653 ML20155J677 ML20155J684 ML20155J692 ML20155J704 ML20155J710 ML20155J713 ML20155J726 ML20155J745 ML20155J753 ML20155J770 ML20155J778 ML20155J792 ML20155J808 ML20155J825 ML20155J835
References
NUDOCS 8605270171
Download: ML20155J710 (22)

Text

, . . - .. ._

o,, UNITED STATES g NUCLEAR REGULATORY COMMISSION

[; a wasumorow.o.c.aossa December 14, 1984

(.... j!

i 4

MEMORANDUM FOR: Harold R. Denton, Director .

Office of Nuclear Reactor Regulation

'(

THRU:

Hugh L. Thompson, Jr., DirectorDivision of Human Factors Safety,

~

FROM: William T. Russell, Deputy Director Division of Human Factors Safety, NRR

SUBJECT:

RESULTS OF JOINT NRR/0I INVESTIGATION AND EVALUATION OF l DENNIS 1. OLSON

Reference:

s (01) 1.'MemorandumfromS.J.Chilk(SECY)toB.B.Hahect:

and W. J. Dircks (ED0) dated April 2, 1984, Su -

1 Staff Requirements-Discussion of Pending Investigation-

! TMI 4

2. Memorandum from H. R. Denton (NRR) to B. B. Hayes (01) l i dated May 3, 1984,

Subject:

NRR Review of 0!

Investigation Materials Concerning Hartman Allegations of Falsification of Leak Rate Data at TMI, Unit 2 The purpose of this memorandum is to document the resultstof the Joint 01/NRR-investigation and evaluation of Mr. Dennis I. Olson, currently a Senior Reactor Operator (SRO) at Waterford Generating Station, Unit 3 and to provide a recomendation regarding whether his current SRO license should be revoked.

l modified, or suspended under 10 CFR 55.40 due to his involvement in

- preaccident leak rate testing irregularities at TMI-2.

Background

As a result of a Comission meeting on March 23, 1984 NRR was directed by Reference 1 to review OI investigative materials concerning falsification of reactor coolant system (RCS) leak rate tests at TMI-2 and refer back to 01 those matters which required further investigation. The results of NRR's review was provided in Reference 2. The revdew determined that follow-up investigation by 01; and further evaluation by NRR was needed in the case of .

NOTE: This memorandum and Enclosure 1 discusses information which is i the subject of an ongoing 01 investigation. This memorandum and Enclosure 2 discusses information that is maintained in the NRC's Privacy Act System of Records (NRC-16). This memorandum and .

enclosures may not be disseminated outside the NRC without coordination with NRR and the permission of the EDO or the Director.

01." Internal access and distribution should be on a "need to know" basis. ,

B605270171 860516 PDR ADOCK 05000320 g[f P PDR ,

1

Harold R. Denton December 14, 1984 sevenlicehsedoperators. Of the seven individuals identified in Reference .'

2, four are currently licensed at TMI-2, two are licensed at Waterford 3 and one is licensed at San Onofre 2 and 3. At a follow-up Commission meeting on '

May 23, 1984 NRR proposed joint OI/NRR investigations and evaluations of -

f these individuals. Subseque..tly, NRR issued letters to the seven individuals under 10 CFR 50.10(b) requesting additional information regarding current performance. Based upon investigation into the individuals' past involvement in improper activities at TMI-2 and an evaluation of the individuals' subsequent performance, NRR would recommend what action, if any, should be taken against the identified operators.

Past Involvement in TMI-2 Leak Rate Testing Irregularities On the morning of November 15, 1984, a joint OI/NRR interview of Mr. Dennis I. Olson was held in New Orleans, Louisiana. The interview was conducted under oath and in the presence of Mr. 01 son's personal attorneys. ,,

The purpose of the interview was to determine Mr. 01 son's role in improper activities associated with RCS leek rate surveillance testing at TMI-2 prior to the accident on March 28, 1979. At that time, Mr. Olson was a licensed Reactor Operator (RO) and served as a Control Room Operator (CRO) on shift "D." A detailed summary of Mr. 01 son's interview is included as Enclosure 1 to this memorandum.

V Throughout the interview Mr. Olson did not appear to answer key questions in a straightforward or candid manner; While some of his responses associated with leak rate surveillance test problems at TMI-2 were consistent with statements made to the NR'C by other-licensed operators, many of his responses regarding his personal involvement in or knowlege of leak rate test _

manipulation do not appear credible.

During the interview, Mr. Olson admitted that leak rate tests were routinely run at least once per shift although the Technical Specifications required the test be run only once every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. At times, Mr. Olson stated, the test had to be run several times before a satisfactory result was obtained (i.e., unidentified leakage within the Technical Specification limit of 1 gpm). Only satisfactory leak rate test results were logged in the CR0 Log.

. Tests which exceeded the Technical Specification limit were not logged.

These actions were contrary to TMI Administrative Procedure 1012, " Shift Relief and Log Entries," which required the start and stop times of all surveillance tests to be logged in the CR0 Log.

Mr. Olson stated that prior to an incident on October 18, 1978, which resulted in the licensee submitting LER 78-62/1T, all bad leak rate test i results (i;e., unidentified leakage in excess of the Technical Specification limit of 1 gpm) were thrown away. Only satisfactory test results were

retained. These actions were contrary to TMI-2 Technical Specification 6.10,

" Record' Retention," which required that records of surveillance activities A required by the Technical Specifications be retained for a period of at least (v ) five years. Following the October 18, 1978 incident, Mr. Olson stated that

1 l

Harold R. Denton December 14, 1984 I he never threw away unsatisfactory leak rate tests. He stated he would -

l continuetirerunthetestuntilheobtainedasatisfactoryresult. He would

then staple the bad leak rate test sheets to the good test sheet and turn it -

in to his Shift Foreman. However, the leak rate test records retained by the i licensee include only satisfactory test results. No records of bad leak rate test results were included with any of the acceptable tests. Mr. 01 son's

= statement that bad leak rate tests were retained is not consistent with

! statements made by other operators including Mr. Mark S. Coleman, one of i three CR0's on Mr. 01 son's shift. Mr. 01 son's statement is also in conflict l . with the Department of Justice's (D0J) Statement of Facts that was read into l

the record as part of the criminal trial settlement (USA v. Met-Ed).

, Following the October 18, 1978 incident, Mr. Olson stated that he was made j aware of the proper interpretation of Technical Specification 3.4.6.2

" Reactor Coolant System Operational Leakage Limiting Condition for

Operation." As stated earlier, the Technical Specification required that this test be performed at least once every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />; however, it was operating policy at TMI-2 to run the test at least once every shift. If a leak rate test were run anytime during the 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and the results showed unidentified leakage in excess of 1 gpm, the operator was required by the j Technical Specification to enter the Action Statement. The Action Statement specified that leakage be reduced to within acceptable limits within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> i or the plant was to be in hot standby within the next 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and in cold i

shutdown within the following 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />. Based upon a review of plant records, there were several occasions where Mr. Olson was on watch and leak

! rate tests were likely conducted (since a valid leak rate result had not been I obtained for more than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />); however, no leak rate test record exists i for these periods. This would indicate that any leak rate tests conducted '

j during these periods did not meet the limits of the Technical Specifications.

l Centrary to Mr. 01 son's statement that he understood the requirement of this l Technical Specification following the October 18, 1978 incident, the evidence does not support that Mr. Olson complied with the requirement of the Technical Specification (i.e., entering the Action Statement when leak. rate test results exceeding 1 gpm were obtained).

Mr. Olson also stated that he was unaware that hydrogen additions.to the make-up tank could affect make-up tank level indication and, thus, favorably influence leak rate test r'esults. This statement is in conflict with the

! statement made by Mr. Coleman, a member of Mr. 01 son's shift, who stated it I was comon knowledge among the CR0s on their shift that the addition of

hydrogen to the make-up tank would favorably affect leak rate test results.

I Mr. Olson Jtated tnat he never added water to the make-up tank for the purpose of altering leak rate test results. If water were added during a test, it was done for legitimate operational reasons and these water i

!* additions were properly compensated for in the leak rate calculation. A

{

technical analysis of Mr. 01 son's leak rate test does not support his statement. It is apparent from the analysis that Mr. Olson was. aware that l

2 Harold R. Denton December 14, 1984 l

adding water to the make-up tank late in the test would produce the same -

effect on Rake-up tank level indication and, thus, leak rate test results as '

adding hydrbgen late in the test. All leak rate tests performed by Mr. Olson ,

between February 27, 1979 and the date of the accident (eight tests) include --'

water additions during the last several minutes of the tests. Although all shifts had the same operational considerations as Shift "D," none of the other five shifts exhibited this consistent pattern of water additions just prior to completion of the test.

Current Performance l

/

~

1 l

. ,_ *l 1

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

i l

() Harold R. Denton December 14, 1984

.. ~

Conclusions During the period Mr. Olson was licensed as a Control Room Operator at TMI-2 prior to the accident, he admitted he was involved in some activities associated with leak rate testing irregularities. However, he denies that he was involved in other actions including leak rate test manipulation or falsification. The weight of evidence, including statements by other operators, technical analysis, and the Department of Justice's Statement of Facts, strongly suggests that Mr. Olson was not truthful in answering questions regarding his role in leak rate test manipulation. The specific areas where Mr. 01 son's version of the facts differ from other evidence are:

1. That he never threw away unsatisfactory leak rate test ~results following the October 18, 1978 incident;

. 2. That he operated THI-2, without violating Technical Specification 3.4.6.2, " Reactor Coolant System Operational Leakage Limiting Condition for Operation;"

3. That he was not aware that hydrogen additions to the make-up tank O 4.

could alter make-up tank level indication and consequently influence leak rate test results in a favorable manner; That he was not aware that water additions to the make-up tank late in the test could produce the same effect on leak rate test results as hydrogen additions; and

5. That he never added water to the make-up tank for the purpose of altering leak rate test results.

I I

.I 1

4 I

• a-I i

1 l

4

j .

I t

t Harold R. Denton December 14, 1984

.. ~

t .

40.T.[,~_D_A, - -

William T. Russell, Deputy Director l Division of Human Factors Safety i

Office of Nuclear Reactor Regulation i

~

Enclosures:

! As stated ~

l cc: B. Hayes l i K. Christopher -

I I

1 i

I f

l l

I

! l l .

! . 1

! 1 1 . _

l

- , . , - , . - - - - - -,..,,,,--,.--.----.,,----n,-..--,.-,-_n6,,-,e-,,,,_,,,----,--,,- .,,,,-,,,,,,,nn~.,rs-,,-ne,_-,-n--

O NOTE: This enclosure discusses information which is the subject of an ongoing 01 investigation. This enclosure may .not be disseminated outside the NRC wit.heut coordination with NRR and the permission of the EDO or the Director, 01.

Internal access and distribution should be on a "need to know" '

basis. _

Enclosure 1 PAST INVOLVEMENT IN TMI-2 LEAX RATE TESTING IRREGULARITIES  :

1 I. Background l On November 15, 1984, a joint Office of Investigations (OI)/ Office of Nuclear l 4

Reactor Regulation (NRR) interview of Dennis I. Olson was held in the Federal

Building, New Orleans, Louisiana. The purpose of the interview was to determine Mr. 01 son's role in improper activities associated with Reactor Coolant System (RCS) leak rate surveillance testing at Three Mile Island, Unit 2 (TMI-2) prior to the accident on March 28, 1979. Present during the

. interview representing the NRC were: Keith Christopher, Director, Office of Investigations, Region I; William Russell, Deputy Director, Division of Human j Factors Safety; and Robert Capra, Technical Assistant, Division of Systems Integration. Representing Mr. Olson at the interview were his personal l attorneys: William Marcoux of the law firm LeBoeuf. Lamb Leiby and MacRae,

! Washington, D.C. and Jane Penny of the law firm Killian & Gephart, Harrisburg, Pennsylvania. Mr. Olson was placed under oath for the interview.

O A copy of the transcript associated with this interview is provided as

'Q Attachment 1.

t Mr. Olson is currently employed by Louisiana Power & Light Co. as a Senior Reactor Operator (SRO) at Waterford Generating Station, Unit 3. Mr. Olson has been employed at Waterford 3 since approximately May 1981. Prior to his current employment, Mr. Olson was employed by General Public Utilities (GPU) and Metropolitan Edison Company (Met-Ed). Mr. Olson was originally hired by Met-Ed in March 1971 after serving 8 years in the U.S. Navy's Nuclear Submarine Pro Health (NIH) gram in Bethesda, and 11 months Maryland. as Mr. an employee Olson worked at thefor National a period Institute of of -

, approximately 5 1/2 to 6 years as an Auxiliary Operator at TMI-1. He moved over to TMI-2 as a CR0 trainee when the initial set of operators were selected for THI-2. Mr. Olson received his Reactor Operator's (RO) license in June 1978 and served as a Control Poom Operator (CRO) until he left TMI-2 i for his current employment in May 1981.

i The interview with Mr. Olson concentrated on the period September 30, 1978 through March 28, 1979. During this time period, Mr. 01 son's shift consisted of the following individuals:

SHIFT "D" Shift Supervisor: Gregory Hitz

. , Shift Foreman: Adam Miller CRO: Dennis Olson O Lynn Wright -

CRO:

CRO: Mark Coleman r-,,- - - - - ---,-,,y--, - - , - - ,,,py- ,y,wer,- w -,--o-wmmy,,-ws , -.pge m g _ _ w-,, , ,_ . _ ,.,, , ,.yg,ww.,-,p-,-yey- yw-p,-,,,w+ng++.+m,,,p.e-e,,,e -

O 1 Messrs. Wright, Coleman and Olson were previously interviewed on this -

subject on March 28, 29 and 31, respectively, during an NRC investigation -

i into the."Hartman Allegations." Based upon these initial screening '

interviews, a follow-up interview with Mr. Coleman was conducted on April 102 1980. A copy of the report of this latter interview of Mr. Coleman is included as Attachment 2.

A sumary of the technical analysis of the leak rate tests involving Mr. Olson is provided in Section II of this enclosure. A sumary of the interview with Mr. Olson is provided in Section III of this enclosure. The sumary in Section III includes citations to the page numbers of the transcript in which the sumarized information was discussed. Overall con-

, clusions regarding Mr. 01 son's involvement in improper activities associated j with TMI-2 leak rate surveillance testing is contained in Section IV of this l enclosure.

II. Sumary of the Technical Analysis A technical analysis of the THI-2 leak rate surveillance records was performed by Dr. Jin Chung of the Office of Inspection and Enforcement, Region I. This analysis was originally done as support to the Department of

, Justice. The results of Dr. Chung's analysis were utilized in questioning O Mr. Olson during his interview. While a more detailed sumary of the leak rate tests at TMI-2 involving Mr. Olson, along with the surveillance test sheets, CR0 Log extracts, and the make-up tank MUT strip charts for the questionable tests, are provided as Attachment 3 to this enclosure, a brief sumary of the results of the technical analysis will help the reader understand the basis for the questions posed to Mr. Olson during the interview. .

A review of the test records show that of the 156 leak rate surveillance tests retained by the licensee during the period under investigation, Mr. Olson was involved in 20 of the tests. During these 20 tests Mr. Olson was either the individual who signed the surveillance test record or he was the operator on watch who signed the CR0 Log for the period during which the test was conducted. Of these 20 tests, the analysis shows that 14 tests involved possible water or hydrogen additions to the MUT during the one hour time period in which the leak rate tests were conducted. The table below shows the breakdown of water / hydrogen additions.

i Tests on file involving Mr. 01 son.............. 20

. Tests involving hydrogen additions.............. 1 Tests involving unrecorded water additions...... 3 i

Tests involving underrecorded water additions.. 10*

• Eight of these tests involve water additions during the last 10 minutes of 4he leak rate test. The other two tests involve water additions 23 minutes and 17 minutes prior to the end of the test.

1 O.

. - - .---,,,-.---,-,-.,--,w-,,-,,,---,---.cy-,~- -

,v---,m - v-w , e, y~.%w,c~,,+ww,

As discussed in other reports involving THI-2 leak rate test investiga- '

tions,_the addition of hydrogen, theoretically should not alter MUT level -

indication; however, because of the configuration and environment of the MUT

level detection instrumentation system at TMI-2, water could collect in the __.- ,

i low-pressure (dry) reference leg of the level transmitters. Under this

! condition the resultant water slug or " loop seal" would cause a temporary increase in the indicated MUT level when hydrogen pressure was increased in ,

the MUT (hydrogen addition) without actually adding water to the tank. Thus, the addition of hydrogen at the appropriate time (after the computer collected its initial data and shortly before the final data readings were i taken) could affect the leak rate results in a nonconservative manner (e.g., ,

the calculated leak rate would be less than the actual leak rate). Because

! adding water to the MUT compresses the hydrogen gas in the top of the MUT, i water additions to the MUT shortly before the computer obtains its final data

! set would thus have the same effect as a hydrogen addition.  ;

The leak rate test surveillance test procedure at TMI-2, SP-2301-3D1, ,

cautioned the operators to avoid the addition and removal of water from the ,

! reactor coolant and make-up systems during the test. However, in the event 1

such a change was required for operational reasons, the procedure provided a

, mechanism by which these changes could be accounted for in the computer i calculation. Before the computer completed its calculations, a series of questions were asked of the operator performing the test. One of those t questions asked the operator to enter operator-caused changes in RCS inventory. ,

An operator aware of the instrumentation inaccuracy (without needing to know l it was caused by the " loop seal" effect described above) could manipulate

! test results by following the procedure exactly as written. For exagle, the i p addition of 150 gallons of water to the make-up tank during the last 10 minutes of the leak rate test could cause the indicated level in the MUT to increase 180 gallons. Once the final data was read by the computer (including the 180 gallon increase in MUT level) the cperator would enter the 150 gallon water addition into the computer as an operator-caused change.

3 Thu.s, when the computer calculated its value for gross leak rate and total l unidentified leak rate, the leak rate result would be 0.5 gpm (30 ga11ons/60

! minutes) lower than the actual leak rate.

From February 27, 1979 through the date of the accident, Mr. Olson was involved in eight leak rate tests. All eight tests show water additiens were 1 made to the MUT during the last 10 minutes of the leak rate test. In each

) case, the water additions were properly entered into the CR0 Log and .

accounted for in the computer calculation. However, in each case the amount 4

i indicated on the MUT strip chart and, thus, the final value of MUT level recorded by the computer was signfficantly more than the amount entered into the CRO Log and compensated for in the computer calculation.

!O i

i

_ . . . . . ._.--,._.-.a..-..--.- _ _ _ __ _ - . _ - - - _ _ . ~ _ _ . - _ _ _ _ _ _ . _ . - . . _ _ . _ . , _

l

. i i

_ .. s i '

III. Interview Sumary .-

l After being placed under oath, Mr. Olson was asked to describe his employment .  ;

history prior to joining Louisiana Power & Light Co. Mr. Olson stated that ~ l he had served eight years in the U.S. Navy and 11 months employed by the j

! National Institute of Health in Bethesda, Maryland prior to joining Met-Ed in ,

March 1971. Mr. Olson spent a i as an Auxiliary Operater (AO) prior pproximately to being assigned 5 1/2 as to a6 CR0years at TMI-1 trainee at working j 4 TMI-2. Mr. Olson obtained his Reactor Operator's license in June 1978. See pages 3-6. During the majority of the period under investigation, Mr. Olson stated he was assigned to Shift "D." Shift "D" consisted of the following licensed individuals:

l SHIFT "D" l

j Shift Supervisor: Gregory Hitz '

Shift Foreman

Adam Miller "

2 CRO: Dennis Olson CRO: Lynn Wright CR0: Mark Coleman Mr. Olson stated that he did not recall being interviewed by Mr. Christopher j O and Mr. Martin (Region I) on the "Hartman A11eflations" on March 31, 1980.

See page S. Mr. Olson was asked to describe h's association with the

management structure at TMI-2. Mr. Olson stated that he would see both j Mr. Logan (TMI-2 Plant Superintendent) and Mr. Floyd (TMI-2 Supervisor of Operations) on a regular basis in the Control Room but did not recall 1 l spending any time with them ciscussing operational problems. He stated that as operators they also did not interface very closely with the Shift

^

Supervisor. Operational problems would normally be brought to the attention i of their Shift Foreman. See pages 9-11.

1 Mr. Olson was provided a copy of TMI-2 Surveillance Procedure 2301-3D1,

! " Reactor Coolant System Inventory" and a copy of TMI-2 Technical Specification (TS) 3.4.6.2 " Reactor Coolant System Operational Leakage i Limiting Condition for Operation." Mr. Olson stated that leak rate tests i

were routinely done by using the plant computer. The test duration was usually one hour. At the end of the test the operator would have to provide j information to the computer to account for operator-induced changes to the

RCS such as water additions to the MUT or pump downs of the Reactor Coolant j Drain Tank (RCDT). See pages 13-14. Mr. Olson stated that the Technical Specifications required that a leak rate test be run at least once every 1

- 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. -When a " bad" leak rate test result (i.e., unidentified leakage in j excess of the Technical Specification limit of 1.0 gpm) was obtained, a

. Mr. Olson stated he would take .it to the Shift Foreman, let him know the result,-and go back and run another leak rate test. See page 14. He stated 4 that another leak rate test was run since the TechnicaTSpecification Action

Statement required them to reduce the leakage to within the acceptable limits l (Liniting Condition for Operation) within four hours. See page 15.

\

l .

l i

1 I

( i NOTE: Part b of the Action Statement reads as follows: '

,_ "With any Reactor Coolant System leakage greater -

_ 'than any one of the above limits, excluding PRESSURE BOUhDARY LEAKAGE, reduce the leakage rate ._.-

within limits within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> or be in at least HOT STANDBY within the next 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and in COLD SHUT- 1 DOWN within the following 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />." '

l Mr. Olson stated that there were times when several leak rate tests had to be

• run before a good leak rate test was obtained. He could not recall how many

. leak rate tests would have to be conducted on an average before a satisfactory result was obtained. See pages 15-16. When asked what was done with the bad test results, Olson saTiIthey were initially thrown away.

Mr. Olson stated that following an incident on October 18, 1978 in which an NRC inspector found several bad leak rate test results " lying around the l l

Control Room," and the inspector questioned why the plant was not shutdown, '

Mr. Olson was given direction not to throw bad leak rate test results away. )

See pages 16-18.

NOTE: The incident referred to by Mr. Olsen occurred on October 18, 1978. During a routine inspection of TMI-2 operations, the inspector discovered that TMI-2 had been o>erating for two or three days with l unidentified leacage exceeding the Technical Specifications. The incident resulted in the sub-mittal of Licensee Event Report (LER) 78-62/1T dated November 1, 1978.

Mr. Olson stated that from that incident on, all bad leak rate tests were l kept. He explained:

j "If I was the one doing the leak rates or I was the one that took them off the computer, and there was a bad one and a good one, as best I can recall, they both got stapled together and attached to the sheet and got submitted with them both on there. I know that I never threw any away after that. I can't vouch for what anybody else did." .S_ee,pages ee 19-20.

A copy of LER 78-62/1T along with a routing sheet with the typewritten names and handwritten initials of control room personnel was provided to Mr. 01 son for his review. The routing sheet was used to indicate which individuals had seen the LER. Mr.-Olson confirmed that he had seen the LER before and that the initials on the routing sheet beside his typewritten name were his. See pages 21-f3. LER 78-62/1T states in part:

"This event was caused by misinterpretation of the requirements of Me technical specifications. The appropriate personnel will be instructed on the requirements of applicable sections I of the T.S. and the requirement to imediately invoke appli-cable action statements when the provisions of the LCOs ,

(limiting condition for operation) are not met." See paSe 21.

, l l

Mr. Olson~was asked if he was ever given any instruction on interpretation of the leak . rate Technical Specification following this incident. As best he could reca-11, Mr. Olson had been told the proper interpretation of the Technical Specification. See page 26. '

NOTE: Prior to this incident, the majority of operators at TMI-2 indicated that since a leak rate test was re-quired only once every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, they had the full 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> in which to obtain a satisfactory leak rate ,

regardless of how many bad test results were obtained  !

in the interim. That interpretation was the l

" misunderstanding" referred to in LER 78-62/1T. '

Mr. Olson could not recall whether from the date of the incident (October 18,1978) to the date of the accident (March 28, 1979) the Action Statement of this Technical Specification had been invoked. See page 26.  !

When asked if it was harder to get good leak rates as the date of the ,

accident approached, Mr. Olson indicated that they were "having to add an  ;

awful lot of water towards the end" and he believed it was harder to get leak rate test results within the limits of the TS. See page 27. Mr. Olson indicated that they tried to run a leak rate test at least every shift. See page 27. He stated that when he was the individual "who put the leak rate O into the computer," he tried to keep the other individuals on watch informed by advising them that the leak rate test was being run and to let him know if they were going ^to do anything that would affect the test. At the end of the test he would ask if anyone added water to or pumped-out water from the system. If they had, he stated he would enter that amount into the computer.

See page 28.

A copy of TMI Administrative Procedure (AP) 1012. " Shift Relief and Log Entries," was shown to Mr. Olson. He was asked if this procedure required that all surveillance tests be logged in the CR0 Log. Mr. Olson confirmed that the procedure required both the start and stop times of all surveillance tests, including leak rate tests, to be logged. However, during the period under investigation, Mr. Olson stated that only satisfactory test results

! were logged.. He did not recall the requirement to log the start time of all

tests. When asked if all leak rate tests were not logged in an attempt to deceive anyone about how many leak rate tests were being run, Mr. Olson j stated that "as far as he was concerned, it was not done to deceive anybody."

See pages 28-31.

t Since Mr. Olsen had stated that following the incident on October 18, 1978,

all unsatisfactory leak rate surveillance test sheets were stapled to the satisfactory test sheet. He was then asked if he knew why none of the leak rate test records that were retained by the licensee contained any i

unsatisfag_ tory leak rate test records. He could not provide an' explanation.

He stated:

f l

1

________._._______.._.-_._._.,_._.m__,.. . . _ , _ _ _ _ . _ _ . _ - _ _ . _ . _ _ _ _ _ , _ . _ _ - . _ _ , _ _ . _ - _ . - , -

l

\

l "At..the end of the day, all of the completed surveillances went s in a box, and they were picked up in the morning. I guess th~e -

guy's title was just surveillance coordinator...they were taken out, reviewed and they were brought back up and put in the file." ~

See page 33.

Mr. Olson was asked if he had difficulty obtaining satisfactory leak rate test results. He indicated that he could not really recall. He knew he had "run a few;" however, it was not something that he " dreaded" doing every time he came on shift. He stated that there was no pressure on his shift to get a good test result. When asked if his Shift Foreman or Shift Supervisor ever pressured him to get a good leak rate because they were approaching the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> limit Olson responded:

"I can't remember anybody saying anything like that. 'Get a good leak rate now. I want one.' No. We didn't work under that king of atmosphere." See page 35.

Mr. Olson indicated that he felt no more pressure to get a good leak rate j test result than any other test. He felt it was a matter of routine. See page 35.

O When running the leak rate tests, Mr. Olson stated he did not have to refer to the procedure since initiating the test was a simple matter of typing the proper code into the computer. However, he stated he was familiar with the procedure including the " precautions and limitations." Mr. Olson read the portion of the procedure where it stated:

" Avoid addition and removal of water from the reactor coolant, and make-up systems during this test. The following operations should not be conducted during the test: .

1

a. Make-up or chemical addition to the make-up system. l

b. Sampling of the RCS or make-up system. l

c. Venting or draining of the RCS or make-up filters. i

d. Changing purification demineralizers or make-up l filters in service. ,

~

e. Boration or deboration." See page 37. l Mr. Olson also read from another section of the procedure were it states:

\

"If changes to the RCS inventory must be made during the  ;

performance of this test, they must be accounted for using ,

Data Sheet 4. Operations such as adding water to the Make-up  !

Tank or sampling the RCS may be accounted for in this manner. l however, these should be avoided if at all possible." See i pages 37-38.

Mr. Olson was asked if he believed that the leakage calculations that were generated by the coniputer were representative of actual leakage. Mr. Olson responded saying:

l

4

.. s "I.dgn't know. I often thought that a leak rate was like -

trying to look for a needle in a haystack. I'm trying to look for one (i) ' ' ~,

thousand (68,000) gallon of water out of sixty-eightgallons of water and th to do." See page 41.

Other than performing the leak rate surveillance test, Mr. Olson could not ,

recall any other means that he utilized to determine whether plant leakage was within the Technical Specifications. See page 41. Mr. Olson indicated that he was " comfortable" with the leak rate surveillance test as it was written and implemented. He stated that he never expressed concern to any of his supervisors or other operators about the unreliability of the program.

See page 42.

Althcugh he assumed that the other operators on his shift had also thrown

away bad leak rate test results, Mr. Olson stated that he could not remember ever discussing problems pertaining to leak rates with any of the other operators on his shift. He further stated that he did not believe that the

reason leak rate tests were thrown away was to hide them from the NRC. See -

pages 44-45.

In discussing hydrogen additions to the MUT, Mr. Olson stated that hydrogen was added to the reactor coolant system to scavenge oxygen. He said there was a control (toggle) switch on the panel to open the hydrogen valve to add hydrogen to the system (make-up tank). He also indicated that there was a time period at TMI-2 when hydrogen could not be added from the Control Room.

During this period, the CR0 would direct.an Auxiliary Operator to perform the addition locally. See page 48. Mr. Olson claimed that he would only add hydrogen when it was recorsnended or directed by chemistry personnel. See page 48. Mr. Olson could not recall how long a typical hydrogen addition would take (e.g., he could not recall whether it took seconds or minutes).

See page 52.

Mr. Olson stated that he did not consider hydrogen additions to tne MUT as

" chemical additions" that were precluded by the procedure. He was then asked if he were conducting a leak rate test and he needed to add hydrogen, would he add hydrogen while the test was being run or would he wait until the test was completed. He stated "I don't really know what I did." Mr. 01 son reiterated that he only added hydrogen when chemistry personnel told him to add it. It was pointed out to Mr. Olson that the technical analysis of the

- leak rate surveillance tests in which he was involved supported his statement j that hydrogen was not added on a regular basis during his leak rate tests.

I NOTE: The analysis shows that 20 leak rate tests involving j Mr. Olson between September 30, 1978 and March 28, 1979,

' - were retained by the licensee. Of these tests only one

!, (February 14, 1979) indicated that hydrogen may have been added during the test. However, the hydrogen addition

! teck place at 2130, which was the same time Mr. Olson was being relieved by Mr. Kidwell according to the CR0 Log.

.g.

Mr. Olsen acknowledged that he was aware that formgr THI-2 CR0 Harold Hartman ' -

had made allegations that hydrogen additions to the MUT were one of the mechanisms-used by operators to manipulate leak rate test results. See page 55. When questioned about these allegations, Mr. Olson indicated that -

he had never added hydrogen to the MUT for the purpose of altering a leak j rate test. He stated that he was not aware of any operators adding hydrogen l for this purpose. Mr. Olson claimed that he did not have an understanding at the time that hydrogen additions could affect leak rate test results or that he had ever heard from anyone that hydrogen could affect test results. See page 58.

, He was asked if Mr. Coleman or Mr. Wright ever told him that they manipulated test results by the use cf hydrogen. He responded that he had not. However, he was aware that Mr. Coleman admitted to the NRC that he had added hydrogen for that purpose. When asked if he believed Mr. Coleman was telling the truth. Olson stated "I don't know. I have never known Mark to lie."

Mr. Olson stated that he had never talked to Mr. Coleman about hydrogen additions and that he had no knowledge at that time that Mr. Coleman may have been manipulating leak rate test results. g pages 59-60.

i NOTE: In an NRC interview with Mr. Coleman on April 10, 1980, Mr. Coleman admitted that he had added hydrogen to the 1

l O make-up tank in order to affect leak rate test results.

He had observed MUT level indication rise during hydro-gen additions. Mr. Coleman stated that it was common j knowledge on his shift that the addition of hydrogen would favorably affect the test results. Mr. Coleman also stated that a hydrogen addition would only have the desired result if it were added near the end of the test. Mr. Coleman also stated that there were a lot of

] problems obtaining acceptable test results.

Mr. Olson was then questioned on water additions. He was asked if he ever  ;

recalled a case where he added water to the make-up tank and the amount that he added, based upon the totalizer, was different from the amount of level l

increase he saw on the make-up tank level indicator. Mr. Olson stated that he could not really remember, but believed that was possible. See page 62.  !

Mr. Olson indicated that if he had to add water during a leak rate test, he i would log the amount in the CR0 Log a~nd account for that amount in the computer calculation. See page 64. He stated that he was'not aware of any instance where water ad3TEiens were made and were not recorded either in the

CR0 Log or included in the RCS leak rate calculations. See page 67. He stated that he was not aware of any individual operator who made unrecorded water additions for the purpose of manipulating leak rate test results. See page 67.

I f

l l

4

) -

! ThefoljowingpostulatedscenariowasdiscussedwithMr.Olson.

"Let i assume for a minute that because of some problem with -

i the makeup tank level indicator, that upon adding water to the 4

makeup tank, it compresses the gas bubbles on the top of the tank which increases the pressure in the tank... and causes you to have a higher apparent level than you have an actual level in the tank. And if that is postulated on the order of thirty percent (30%) difference in change. That is, if one actually added a hundred (100) gallons but the makeup tank level indicator showed a hundred and thirty (130) gallon increase. Because of the way the computer was set up, it

would automatically read the indicator off the makeup tank.

But as the operator making the correction for the water addition which occurred during the test, if you used the number off the totalizer, you would only report that you had entered one hundred (100) gallons.

The tctalizer being the more accurate indication of what water i was actually added, the operator could, in his own mind.

conclude that that was a proper logged entry.

,O But the computer would have been fooled because of the instrument inaccuracy due to the effect of the increased hydrogen pressure.

, Now, that scenario that I have just described would result in an apparent thirty (30) gallon increase which was not a real water addition. That thirty (30) gallons over the period of one (1) hour is equivalent to a half a gallon change in the unidentified leak rate, thirty (per minute 30) divided by sixty (60) minutes.

So that if an operator were to become knowledgeable of this effect through some mechanism, a water addition that is made

toward the end of the test, if that individual were to use the totalizer when the makeup tank level indicator was reading l inaccurately, if you were to make a water addition, the 1

! apparent increase would be greater than the amount of water you-actually added. You would have an under recorded water 4

addition." See pages 69-71.

i Mr. Olson-was asked if he would consider the operator in such a situation to be more a victim of circumstances or one who was. willfully attenpting to manipulate the results. He responded that he would consider the operator to be a victim of circumstances. See page 71.~ Mr. Olson stated that he did not know whether the operators knew such a situation could exist. Mr. Olson O st&ted that he was aware of the inaccuracies between the two instruments; however, he claimed that he never tried to take advantage of the inaccuracy i

a, i

l to affecf a leak rate test. See page 72. Hecouldnotrecallwhenhebecame?

4 aware of-the discrepancy between the batch controller (totalizer) and what l was being provided as input to the computer (MUT level indication). See page 73. ,

I Mr. Olson was then provided a copy of'the technical analysis of the 20 leak l rate tests in which he was involved during the period September 30, 1978 and March 28, 1979. (Attachment 3 to this enclosure). After Mr. Olson reviewed i the document, it was poin'ted out that the analysis shows 14 of the~20 tests j involve water and/or hydrogen additions during the test. All tests performed i

after December 17, 1978 involved some type of addition (14 tests). One test

! involved a hydrogen addition. Three tests involved water additions which i were neither recorded in the CR0 Log nor included in the computer calculation. Ten tests involved the addition of water to the )iUT where the i amount of water indicated on the MUT strip chart was significantly higher than either the amount logged or the amount included in the test calculation.

i All ten water additions were made during the latter half of the test, with eight being made during the last ten minutes of the test. See pages 73-81.

After reviewing the data, Mr. Olson stated that he never did anything to

manipulate the outcome of a leak rate test to the best of his knowledge. He

was asked to explain why water had to be added on a routine basis so near the end of the test. Mr. Olson responded.

"The only possible explanation that I can give would be in order to have to move the control rods. If they are right at

the edge of the band that you are supposed to maintain." See page 83. .

A detailed discussion of Mr. 01 son's explanation was then pursued. This

, , discussion included a review of RCS boron concentration and control rod

! position at the beginning and end of his shifts to see how much control rod

. position and boron concentration had changed during his watch. In most cases

. the amount of water added during the leak rate test was in range of 150 to 200 gallons. Mr. Olsen was asked if he believed a 200 gallon addition of water to the make-up tank would make a significant change in the boron concentration. Mr. Olson stated that the volume of the reactor coolant system was between 65,000 and 68,000 gallons and that 200 gallons isn't going

to change the boron concentration any significant amount. However, Mr. Olson stated that he could not think of any other reason for the water additions. -

See pages 83-88. It was also pointed out to Mr. Olson that if he needed to make these additions to control the plant it would seem that other operators

would also have to exhibit a similar pattern of additions. However, a review

of the data indicated no other shift'had exhibited such a pattern. Mr. Olson

responded by saying

O

i "I understand what you are s'aying and I understand what this shows. .And it looks very incriminating. But I never did anything in that plant to affect the outcome of the leak rate.

I've got no reason to lie. I've got nothing to hide. I do _e

. not cheat at my job. I never have and I never intend to. I can't explain all these. I really can't." g page 90.

., Mr. Olson was asked if it were possible that one or both of the other

) individuals on shift with him could have been aware of this phenomenon and been manipulating leak rates using this method without him knowing it. He

responded:

, "Anything is possible. I wouldn't turn around and say, ' Yeah, he did it. I didn't know anything about it it. I can't do

that. I don't know that. But anything is possible. You

know, my name is in the log book or my name is on the leak rate." See page 94.

! It was also pointed out to Mr. Olson that in one particular test on March 3, 1!r79, where 150 gallons of water were added 3 minutes before the end of the 4

test. Mr. 01 son's name appeared both on the leak rate surveillance sheet and on the CR0 Log. See page 96. The boron concentration at the beginning of

the shift was 1051 ppm and 1047 at the and of the shift. The controlling groups (Groups 6 and 2) were at 93% at the start of the shift and 965 at the end. It does not appear that the rods were near the upper or lower end of the limit in either case. Water was added during the test when the water level in the MUT was in the middle of the operating band. Thus, it was l pointed out that there appeared to be no technical reason for a water i addition 3 minutes before the end of the test. In this particular test, the i amount of water added according to the level increase in the MUT was i

! 200 gallons. The amount logged in the CR0 Log was 150 gallons and the amount i included in the computer ca culation for leak rate determination was 152 gallons. This unaccounted for 48 gallons would lead to undercomputed leak rate for unidentified leakage of almost .8 gpm. See pages96-100, l

Based upon the data reviewed with Mr. Olson, he was informed that his rationale of why water additions were made near the end of a leak rate test did not appear credible. See page 108. Mr. Russell explained to Mr. Olson that the issue of concern and the thing that regulators must rely on is the candor and the truthfulness of the reports that are made to the NRC or made to the licensee. The NRC has to be able to count on the individual doing

{ what is right. Not coloring it to cast the blame someplace else or to put i

some distance between himself and what went on but a willingness to accept j -responsib'111ty for his own action. M page 107.

l . Mr. Olson was asked if it was possible that before February 1979 when j hydrogen ^could be added from the Control Room, operators who were I manipulating leak rate test results, used hydrogen additions to the MUT for

{ that reason. However, during the February / March time frame, when hydrogen e,-----------e-,,-,-ww,,,,-+,-,w,- -,-.-aw,----- r,,,--ww,,,-,,-,e-----,,,,--- ---,---,-------,---wan- -,r---,,,--,r-,, ----ve-~,w,-ew-,. -mv,-----wrn-,,-

l could not~be added from the Control Room, they could not use that same ~

phenomenon so they had to go te something else like a water addition to

! produce the same effect. Mr. Olson acknowledged that that possibility could

exist. See page 113. r i

! Mr. Olson was asked if he had ever gotten together with the other members of I

his shift to discuss how they should answer questions regarding the leak rate

investigation to make sure they answered the questions the same way.

Mr. Olson stated that they had not. He also stated that he had never i

- discussed answers to any que'stions with anyone from GPU. His snly discussion

. regarding leak rate matters was with his counsel. See pa Mr. Olson i also stated that he was worried about himself; ther M re,ge he 116.

did not feel the i need to protect anybody in GPU or any of his friends. See page 117.

Mr. Olsen was again asked if it were possible that an individual on his shift, wanting to manipulate leak rate test results, could do so without 01 son's l knowledge. Mr. Olson indicated that was possible. He stated, however, that l he was " amazed" when he heard about Mr. Coleman's admissions. He said he had j not heard about it until a couple of weeks before this interview. See -

! pages 117-118.

When operators were experiencing trouble getting leak rate tests, Mr. Olson O was asked, what action, if any, the Operations Department was taking to correct the situation. Mr. Olson stated:

...when we would get a bad leak rate, Adam (Miller) would get ahold of the auxiliary operators and say, 'Go out and start

! walking down the lines and look at packing glands, various i

valves, you know, make-up pumps, isolation valves, anywhere you might see an indication of boron.... Which would indicate that you've got a leak. If it's leaking, catch it in a

, container for a specific time period, measure it and then we l can input that as identified leakage." Se_e,pages e 120-121.

With respect to the follow-up 3ction stated in the November 1, 1978 LER regarding the need to correct computer input errors, Mr. Olson stated that there was not much that he could recall being done other than a modification to make the RCDT level input automatic instead of manual. See pages 122-123.

He stated that he was not personally concerned enough about T e procedure or the computer program to mate a point of discussing it with his Shift Foreman

! or Shift Supervisor. See pages 124-125.

The interyiew was concluded after approximately three hours.

j IV. Conclusions Throug'hout the interview Mr. Olson did not appear to answer key questions in 1

a straightforward or candid manner. While some of his responses associated with leak rate surveillance test problems at THI-2 were consistent with 1

\

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

w*

t

! statements made to the NRC by other licensed operators, many of his responses '

regarding.his personal involvement in or knowledge of leak rate test mant -

pulation do not appear credible. During the interview, Mr. Olson relayed the following information: _

l 1. Leak rate tests were routinely run at lease once per shift. At times 1 the tests had to be run several times before a satisfactory result was

' obtained (i.e., unidentified leakage within the Technical Specification limit of 1 gpm).

2. Only satisfactory leak rate test results were logged in the CR0 Log.

Test results which exceeded the Technical Specification limit were not logged. Note: These actions were contrary to TMI Administrative Procedure-" TOT 2, " Shift Relief and Log Entries " which required the start j and stop times of all surveillance tests be logged in the CR0 Log.

3. Prior to an incident on October 18, 1978, which resulted in the licensee '

submitting LER 78-62/1T, all bad leak rate test results (i.e.,

unidentified leakage in excess of the Technical Specification limit of 1 gpm) were thrown away. Only satisfactory test results were retained.

Note: These actions were contrary to TMI-2 Technical Specification i T.'TU, " Record Retention," which requires that records of surveillance

! activities required by the Technical Specifications be retained for a

period of at least five years.

l 4. Following the October 18, 1978 incident, Mr. Olson stated that he never

! threw away unsatisfactory leak rate tests. Mr. Olson stated that he

would rerun the test until he obtained a satisfactory result. He would then staple the bad leak rate test sheets to the good test sheet and

! turn it in to his Shift Foreman. Note: The leak rate test records retained by the licensee include only satisfactory test results. No

! record of bad leak rate test results were included with any of the acceptable tests. Mr. 01 son's statement that bad leak rate tests were 1

not thrown away following this incident is not consistent with

! statements made by other operators including Mr. Coleman, one of the j three CR0s on Mr. 01 son's shift. Mr. 01 son's statement that bad leak

! rate tests were retained is also in conflict with the Department of l Justice's (DOJ) Statement of Facts that was read into the record as part l of the criminal trial settlement (USA v. Met-Ed). ,

l

! 5. Following the October 18, 1978 incident Mr. Olson stated that he was made aware of the proper interpretation of Technical Specification )

i ,

i 3.4.6.2, " Reactor Coolant System Operational Leakage Limiting Condition l j for Iberation." Note: While the Technical Specification required that

a leac rate test ne performed at least once per 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> during steady i state operation, it was operating policy at TMI-2 to run a leak rate l test"every shift. If a leak rate test was run anytime during the 72 hcurs and the results showed unidentified leakege in excess of 1 gpm,

l

!O 1 the. operator was required by the Technical Specification to enter the ~

Action Statement. The Action Statement required that leakage be reduced.

l to Within limits within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> or the plant was to be in hot standby 4 withTn the next 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and in cold shutdown'within the following -

30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />. Based upon a review of plant records there were several 3

occasions where Mr. Olson was on watch and leak rate tests were likely l

conducted (since a valid leak rate result had not been obtained for more  :

i than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />); however, no leak rate test record exists for these l

} periods. This would indicate that any leak rate tests conducted durin these periods did not meet the limits of the Technical Specifications.g '

Contrary to Mr. 01 son's statement that he understood the requirement of this Technical Specification following the October 18, 1978 incident, there is no evidence that would indicate Mr. Olson complied with the requirement of the Technical Specification by entering the Action

)

, Statement when leak rate test results exceeding 1 gpm were obtained.

6. Mr. Olson stated that he was unaware that hydrogen additions.to the make-up tank could affect make-up tank level indication and, thus,

) -

favorably influence leak rate test results. Note: Mr. 01 son's statement is in conflict with the statement made by Mr. Coleman, a l member of Mr. 01 son's shift, that it was comon knowledge among the CR0s l on their shift that the addition of hydrogen to the make-up tank would l favorably affect leak rate test results.

i i

7. Mr. Olson stated that he never added water to the make-up tank for the  ;

purpose of altering leak rate test results. If water were added during

! l a test, it was done for legitimate operational reasons and these water  !

] additions were properly compensated for in the leak rate calculation. '

j Note: The technical analysis of Mr. 01 son's leak rate tests does not i support his statement. It is apparent from the analysis that Mr. Olson was aware that adding water to the make-up tank late in the test would produce the same effect on make-up tank level indication and, thus,1eak ,

l' rate test results as adding hydrogen late in the test. All leak rate tests performed by Mr. Olson between February 27, 1979 and the date of the accident (eight tests) include water additions during the last 10 minutes of the tests. Although other shifts operated THI-2 under the same operating conditions, none of the other five shifts exhibited this consistent pattern of water additions during the last few minutes of -

leak rate tests.

! In summary, the weight of evidence, including technical analysis, j statements by other operators and the D0J's Statement of Facts, strongly j suggests that Mr. Olson was not truthful in answering questions i regarding leak rate testing irregularities at THI-2. The specific areas i

l were Mr. 01 son's version of the facts differ from other evidence are:

I 1. ThatJe never threw away unsatisfactory leak rate test results after the October 18, 1978 incident.

lo i

f

l

2. That'he operated TMI-2 without violating Technical Specification 3.4.6.2, " Reactor Coolant System Operational Leakage Limiting Condition '

fo_r Operation." -

3. That he was not aware that hydrogen additions to the make-up tank could._. -

4 alter make-up tank level indication and consequently influence leak rate test results on a favorable manner. '

i 4. That he was not aware that water additions to the make-up tank late in j the test could produce the same effect on leak rate test results as

hydrogen additions.

5. That he never added water to the make-up tank for the purpose of altering leak rate test results.

i i

1 l

I l

l i

a-I 1

l i

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