ML20151L198

Comments on Stier Rept Re Leak Rate Testing at Facility. Comparison of NRR & Stier Rept Technical Evaluations of Leak Rate Tests Encl.Results Provided.Comparison of Overall & Individual Conclusions Expected in 3 Wks

ML20151L198
Person / Time
Site:	Three Mile Island
Issue date:	09/20/1985
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 8509260250
Download: ML20151L198 (60)

Text

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___

h 1 s

f /ys tatuq'o g UNITED STATES

[ 3 ,., g NUCLEAR REGULATORY COMMISSION

, -E WASHINGTON, D. C. 20555 September 20, 1985 E

MEMORANDUM FOR: Harold R. Denton, Director Office of Nuclear Reactor Regulation FROM: William T. Russell, Acting Director Division of Human Factors Safety i Office of Nuclear Reactor Regulation l

SUBJECT:

STIER REPORT ON TMI-2 LEAK RATE TESTING Background -

As directed by the Commission, NRR has been working with OI during the past 18 months on investigations of 10 currently licensed operators in order to determine their involvement, if any, in improper activities associated with leak rate testing at TMI-2 prior to the accident. To date, reports have been completed for eight of these individuals and have been forwarded by you to the Director, 01 for his review and possible referral to the Department of Justice. The remaining two reports are scheduled for completion by October 4, 1985.

During the course of our work with 01, we have perfomed a technical evaluation of leak rate tests performed during the last six months of operation of TMI-2. Our test evaluation has been used extensively to question operators and has been the principal basis for our conclusions concerning the involvement these individuals may have had in improper leak rate testing.

By letter to you dated September 9, 1985, GPU Nuclear Corporation (GPUN) forwarded a report prepared by Mr. Edwin H. Stier entitled "TMI-2 Reactor Coolant Inventory Balance Testing," (Stier Report). The report addresses leak rate surveillance testing at TMI-2 prior to the accident.

l The purpose of this memorandum is to provide you with my comments concerning l the technical evaluation of individual leak rate tests contained in the Stier Report and to compare Stier's results with NRR's evaluation. The enclosed report and attachments address areas of agreement and disagreement between our technical evaluations. It does not, however, address Stier's conclusions regarding overall leak rate testing practices or his conclusions on individuals still employed within the GPU system.

Sumary of Comparison The Stier Report covers the entire one year period during which leak rate tests were performed at TMI-2 (222 tests). The NRR evaluation only addressed tests perfomed during the last six months of operation prior to the accident (161 tests). Therefore, for purposes of comparison, only the latter time period is addressed.

go72GO250 '

l 4

-2~ September 20, 1985 For the majority of tests (116 or 72%), the Stier and NRR analyses are in i complete agreement. For some tests (18 or 11%), both Stier and NRR agree that the tests are invalid but for slightly different reasons. NRR also concluded that several tests (21 or 13%) were invalid while Stier felt that there waf insufficient evidence to reach that conclusion. The. reverse occurred for only 6 tests (4%).

The differences in technical evaluation methods utilized by Stier and NRR account for some of these differences. However, the principal difference is that Stier's analysis, performed by MPR Assoc. Inc., did not include feedback from operator interviews. We found that it was necessary to revise and update our evaluation as a result of operator interviews in order to account for methods and techniques used by some operators to manipulate test results. I This we believe is the principal reason for our conclusion that several j additional tests are invalid.

The results of the comparison is shown below. Tests were considered invalid

- if they involved actions which directly violated the TMI-2 Technical -

Specifications, the precautions and limitations of the approved test procedure, or were performed with known faulty instrumentation supplying data to the plant computer. In some cases these actions were perfomed by operators with the intent to influence the test results. In other cases, operators unknowingly or negligently performed these actions. For these reasons, not all invalid tests identified in the tables below represent willful manipulation or intentional falsification of test results.

OVERALL RESULTS Total Tests NRR Invalid Stier Invalid 161 112(70%) 97'(60%)

INDIVIDUAL RESULTS i Name Current Total No. NRR Invalid Stier Invalid l

License of Tests Tests Tests

• R. R. Bocher Waterford 3 14 12 86% 8 57%

J. R. Congdon TMI-2 28 19 68% 16 57%

M. V. Cooper San Onofre 2/3 30 21 70% 16 53%

C. C. Faust TMI-2 20 8 40% 11 55%

E. R. Frederick TMI-2 25 12 48% 14 56%

C. L. Guthrie TMI-2 19 13 68% 12 63%

T. F. Illjes TMI-2 18 13 72% 13 72%

H. A. McGovern TMI-2 14 10 71% 7 50%

A. W. Miller TMI-2 25 18 72% 17 68%

• D. I. 01 son Waterford 3 19 15 79% 13 68%

Messrs. Booher and Olson resigned from Waterford-3 and are no longer licensed effective August 23 and June 28, 1985, respectively.

4 September 20, 1985 Conclusions Based upon our review to date, the technical differences do not appear to be significapt to our overall conclusions or our conclusions or recommendations regarding individuals. As a result, we have neither revised our technical evaluation based upon our review of Stier's Report nor discussed these technical differences with either Stier or GPUN. Our comparison of overall and individual conclusions will be completed in about three weeks.

&$ T. Russell,!:

T. =-)_D William , Acting Director Division of Human Factors Safety Office of Nuclear Reactor Regulation

Enclosure:

As Stated cc with enclosure:

B. Hayes .

G. Cunninghani;* -

cc w/o enclosur'e:

W. Dircks T. Murley, RI R. Martin, Rly J. Martin, RV

Seotember 20, 1985 Enclosure COMPARISON OF NRR'S AND STIER'S TMI-2 LEAK RATE TEST EVALUATIONS I

Introduction Between September 30, 1978 and March 28, 1979 numerous reactor coolant system leak rate surveillance tests were performed at TMI-2. Of these, 161 surveillance test sheets were retained. Former TMI-2 Control Room Operator (CRO) Harold Hartman alleged that many of these tests were intentionally manipulated by operators in order to produce results within the limits of the TMI-2 Technical Specifications.

These test records and other plant data have been reviewed independently by Mr. Edwin Stier, under contract to GPUN, and by NRR. Each review

, concentrated on determining whether the tests were performed in accordance -

with the approved test procedure. Both evaluations concluded, that over 60".

of all tests retained were performed and accepted as valid even though conditions existed which were contrary to the precautions and limitation of the procedure. While these technical evaluations were able to identify test deficiencies or operations performed during the test that were contrary to )

the requirements of the test procedure, they can not identify whether these actions were done intentionally.

The technical evaluation of several tests involved subjective judgement, based upon limited plant data, in reaching a conclusion whether a test was valid or not. In other cases, differences in evaluation methods between NRR and Stier caused differences in results for some tests. The purpose of this document is to summarize where the two technical evaluations agree and were they disagree and to identify the reason for the differences.

NRR Evaluation In 1983 and early 1984, a technical analysis of leak rate surveillance tests performed during the last six months of operation of TMI-2 was performed by NRC, Region I. This analysis was done as technical support to the Department of Justice (00J) in its criminal proceeding against Met-Ed. (Note: Various portions of this analysis were provided to GPUN by D0J on December 5, 1983 and on February 15, 1984 and are identified as Exhibits 1 and 2 in Volume IV(L) of the Stier Report.)

As a result of a Comission meeting on May 23, 1984, NRR was directed to work with 01 to determine what role, if any, currently licensed operators had in improper activities associated with leak rate testing at TMI-2. NRR used the material contained in the NRC/D0J analysis as a starting point for its evaluation; however, based upon information obtained during interviews with former TMI-2 CR0s and Shift Foremen, the original NRC/00J test evaluations have been extensively revised. References in the Stier Report to "NRC allegations of manipulation" for individual tests are based upon this NRC/D0J analysis. This report uses the July 30, 1985 NRR evaluation for )

comparison to the Stier Report. A copy of this evaluation entitled "NRR Evaluation of TMI-2 Reactor Coolant System Leak Rate Tests Performed 09/30/78 to 03/28/79," is included as Attachment 5. The reference materials supporting the evaluation, are not included.

f Stier Evaluation On September 9, 1985 NRR was provided a copy of the report entitled "TMI-2 Reactor Coolant Inventory Balance Testing." The report, dated September 5, 1985, was prepared for GPUN by Edwin H. Stier. In conducting his investigation, Mr. Stier engaged a Washington, D.C. based engineering firm, MPR, Associates Incorporated (MPR), to perform the technical analysis of leak rate tests and other plant data. An individual evaluation of each test is provided in the MPR_ report which is included as Volume IV of Stier's Report. Volume III(A), Table 1, of the Stier Report contains the "Overall Leak Rate Test Summary," which represents Stier's conclusions based upon the MPR analysis. This summary table was used in comparing the results of.the NRR and Stier evaluations. Where test evaluation results are different, the MPR report was reviewed in detail to understand the basis for the difference.

Test Identification

~

For each test retained, an NRR identification number was assigned chronologically, beginning with test #1 on September 30, 1978. In addition, five tests performed between October 16 and 18, 1978, indicating unidentified leakage in excess of the TMI-2 Technical Specifications, are numbered 12A through 12E. A total of 161 tests are included in the NRR evaluation.

The Stier evaluation included 222 tests during the period March 22, 1978 through March 28, 1979. Stier also assigned a unique sequence number to each test; however tests are listed in reverse chronological order, beginning with test #1 on March 28, 1979.

Because our purpose is to compare the two evaluations, only tests retained during the period September 30, 1978 through March 28, 1979 are included in this document.

l l

o Summary of Comparison As will be explained in more detail below, many factors must be considered in determinfng whether an individual test is valid or invalid. For the majority of tests, the Stier and NRR evaluations are in complete agreement. For some tests, both Stier and NRR agree the tests are invalid but for slightly different reasons. For other tests, NRR considers the test invalid while Stier has evaluated the test as valid, and vice versa.

OVERALL

SUMMARY

TOTAL NUMBER OF TESTS 161

.NRR AND STIER EVALUATIONS AGREE 116 (72%) ,

• Valid Tests.................. 43

• Invalid Tests................ 73 NRR AND STIER EVALUATIONS DIFFERENT 45(28%)

• Both agree tests are invalid. 17 (slightly different reasons)

• NRR Invalid /Stier Valid...... 21

• Stier Invalid /NRR Valid....... 6

• Not in NRR evaluation......... 1 Attachments 1 through 4 of this report provide the following additional information concerning individual leak rate tests:

Attachment 1: Identifies the NRR and Stier evaluation category for each of the 161 tests.

Attachment 2: Discusses the 17 tests that are considered invalid by both NRR and Stier but for slightly different reasons and the 1 test not included in the NRR evaluation.

Attachment 3: Discusses the 21 tests that are considered invalid by NRR but valid by Stier.

Attachment 4: Discusses the 6 tests that are considered invalid by Stier but valid by NRR.

i .

i .

a Summary By Evaluation Category For ease of comparison, tests have been divided into nine evaluation

} categorids. These categories come from the NRR evaluation. Depending upon

! plant conditions, status of key instrumentation, and plant evolutions during i the test period, some tests fall into more than one category. The table i below identifies the number of tests in the NRR and Stier evaluations which fall into each category. A detailed definition and discussion of each category follows the table.

i l Summary by Evaluation Category

NRR Stier Category Number * % Number * % i

~

- No apparent problems (0K) . . . . . . . . . . . . . . . . . . . . . . 50 31% 66 40%

i Bad level transmitter or instrument used (LT).. 42 26% 41 25%

t i Water addition (not included)(WAN).. . ... ... .. . . 22 14% 5 3%

j Water addition (partially included)(WAP)....... 17 11% 22 14%

l Fecd & bleed operation (FAB)........,,.....:..,, 15 91 9 61 i Hydrogen addi tion (H) . . . . . . . . . . . . . . . . . . . . . . . . . . 12 7% 14 9%

Unstable plant conditions (USPC)............... 10 6% 11 7%

! Water or hydrogen (can not differentiate)(WorH). 7 4% 0 0%

1 i ResultsgreaterthanTSlimit(1GPM)............ 6 4% 6 4%

3

• Note: NRR figures include 14 tests that fall into two categories and 3

! tests that fall into three categories.

Stier figures includes 12 tests that fall into two categories. I i l Code Discussion of Evaluation Category j OK Test appears to have been conducted in accordance with the requirements of Surveillance Procedure 2301-301, "RC System Inventory." The surveillance test sheet, CR0's and Shift Foreman's Logs, and makeup tank strip chart have no unusual characteristics or features that call i into question the validity of the test. In addition, the actual change in MUT level during the test was consistent with the expected change caused by differences in pressurizer level (Lp) and average RCS

temperature (Tave) during the test. -

USPC Test was conducted- during unstable plant conditions (not in steady state). SP 2301-3D1 directs that the test be performed at least l once per 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> during steady state operation when in Modes 1 through

4. The procedure cautions the operator to maintain the reactor coolant l

l system (RCS) 7.nd makeup system in a steady state condition during the test by avoiding changes in valve line-ups, coolers-in-service, pumps-in-service, etc. Power level changes should be minimized and the l operator should avoid additions or removal of water from the RCS and l makeup system during the test. For the most accurate detemination of 1

RCS leak rate, the initial and final conditions of reactor power, RCS

> temperature, pressure and pressurizer level should be identical.

! LT Unstable or out-of-service instrument or MUT level transmitter was used to provide input to the plant computer during the test. At TMI-2 there are two level transmitters (LT-1 and LT-2) that provide MUT level

indication. The output of one of the level transmitters drives the MUT level strip chart in the control room while the other level transmitter provides MUT level indication to the plant computer. When the selector switch was positioned to LT-1, the strip chart recorder would be driven by the output of LT-1 and LT-2 would provide automatic input to the

_ plant computer. When the selector switch was changed to LT-2, the

• I opposite would occur. All but one of the retained leak rate tests i during this period at TMI-2 were conducted using the plant computer.

{ MUT level is one of the key plant parameters used in the leak rate

calculation. During portions of October 1978 through January 1979, operators at TMI-2 experienced significant problems with either one or

• both level transmitters. In many cases, the output of one level transmitter was so erratic, that the output could not be relied upon to i provide meaningful input to the plant computer to calculate test

] results. ,

H Tests during which hydrogen was added to the MUT. In order to limit

the oxygen content in the RCS and to provide an increased net positive

! suction head for the RCS makeup pumps, a hydrogen overpressure was j maintained in the MUT. When hydrogen pressure decreased near the low l end of the operating band, the'CR0 would add hydrogen to the tank.

1

! Theoretically, the addition of hydrogen should not have affected MUT 4

level; however, because of the design of the MUT level detection

instrumentation system at TMI-2, water could collect in the

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

condition, the resultant water slug or " loop seal" could cause a temporary increase in the MUT indicated level without actually adding water to the tank. Thus, the addition of hydrogen at the appropriate i

time (after the computer collected its initial data and before the final data were taken) could affect the leak rate results in a i

nonconservative manner (e.g., the calculated leak rate would be less i

than the actual leak rate). According to Hartman, this "cause.and effect" relationship was common knowledge among operators. Hartman and i four other CR0s have admitted adding hydrogen to the MUT during leak i

rate tests to influence the outcome of the test.

I This method may have been used more times than are reflected in the analyses. First, hydrogen would only have an affect only if sufficient water were present to fonn a loop seal. Second, water might collect in the reference leg of one LT and not the other. Thus, causing one LT to I

j l

4 I

i

}.-

j  !

. l l

react to the addition and have no affect on the other LT. Third, hydrogen additions were not always logged, or if logged were not always logged at the time the addition was made. Thus, identifying the j addition of hydrogen on the MUT strip chart is difficult.

WAN Water additions were made to the MUT during the test and not included -

in the leak rate test calculation. SP 2301-301' instructs He operator to avoid the addition and removal of water from the RCS and makeup system during the test; however, if the inventory must be changed, it must be accounted for by including the amount of water edded in the

test calculation. Normally water is added to the MUT in a manner i that is easily detectable on the MUT strip chart (batch addition). If however, water is added in small quantities over a period of time (jogged addition), the water addition may be masked by normal

! oscillations on the MUT strip chart.

j - According to Hartman, jogged additions were made to the RCS during the' i performance of leak rate tests and the amount of water added was not 4

included in the test calculation in order to influenca or manipulate

, the outcome of the test result. Both the NRR and Stier evaluations 4

identify batch water additions that were not included in the calculation. The NRR evaluation also identifies possible jogged water additions. Stier does not identify jogged water additions in his i evaluation.

1 WAP Water additions were made to the MUT during the test and were included in the leak rate test calculation; however, because of two Tactors the i test results showed lower leakage than would otherwise be the case j without the water addition, i

l The first factor was a error in the test procedure itself. SP 2301-301 i did not convert the water added to the same temperature as the RCS. As i

a result, the water added to the MUT was " undervalued" and tended to reduce the total and unidentified leak rate as calculated in the test.

(' lote: This error was only one of several that contributed to 1

baccurate results.)

d

} The second factor involved MUT level instrument errors. Because adding water to the MUT compresses the hydrogen gas in the top of the MUT, y water additions to the tank shortly before the computer obtained its  !

j final data set could have the same result as a hydrogen addition. If i an operator were aware of this "cause and effect" relationship, he *

, could take advantage of this phenomenon to manipulate test results by following the procedure exactly as written. For example, the addition 4

of 150 gallons.of water to the MUT during the last 10 minutes of a leak 1

rate test should cause the MUT level to increase approximately 5 inches l (30 gal / inch); however, if sufficient water had collected in the loop seal, the indicated level on the MUT strip chart might actually

! increase 6 inches (180 gal). Once the final data were read by the computer, including the 6 inch rise in MUT level, the operator would -

, enter the 150 gallon water addition (as read off the totalizer on the j batch controller) into the computer as an operator-cause change. Thus, i

f 4

.--,.-,,-.-.,,,,,,n,-...,..,,. ,,, ,,.,-,-,__,,_,,,,._n.+,,-,_n.,_, ,--~-,.n,-. ,,,,,,n.,,,- ._,n,--,rw-m,,~,,, .--, - - w,,

j  :

3 when the computer calculated the total and unidentified leak rate, the i results would be 0.5 gpm less than the actual leak rate [(180 gal - 150 '

gal)/60 minutes]. Note: While operators used 30 gal /in as a rule of i thumb, the actual value was approximately 30.8 gal /in at normal MUT

{ conditions.

i NRR does not believe that operators were aware of the procedural error

, in the calculation; however, based upon the testimony of one CRO, who was aware of the MUT level instrument inaccuracy and used this

mechanism to influence leak rate tests, others may also have been aware of the phenomenon and used it to their advantage.

l The NRR and Stier evaluation of tests in this category differ to some i degree. Stier considers the effect of both the calculational error and i and MUT instrument error in his evaluation. Stier classifies these j tests as " water additions included in the calculation." In addition,

- Stier used a different method ("MUT slope off-set) than NRR to arrive *

, at a figure for the amount of water added during the test. While either method should produce the same result during periods of low leakage, the Stier method is more accurate during periods of high

! leakage. NRR used the " peak-to-peak" change in MUT level indiction

! (assuming 30 gal / inch) to determine the amount added. This method was i used by NRR in calculating water additions since some operators stated i they used this method to confirm the amount of water added. During i periods of high leakage, the NRR method actually underestimates the i

amount of water added since it does not account for the water leakage during the addition period.

i F&B Identifies tests during which feed and bleed operations may have

, occurred. Feed and bleed operations were used routinely to increase or decrease the boron concentration of in the RCS. Based upon the limits j and precautions of SP 2301-301, feed and bleed operations should not j have been conducted during leak rate tests.

Feed and bleed operations, like many others in the plant, do not always

have a distinctive trace on the MUT strip chart. At times water was added (feed) and then removed (bleed) from the RCS in that order or the i reverse order. Since water could also be added and removed from the j system at the same time, these operations might be undetectable on the j strip chart if the rate of addition and removal were equal and the 1 operation was not logged in the CRO's Log. While feed and bleed

{ operations could be used to used to manipulate test results, none of -

the operators interviewed by NRR have admitted using this method.

WorH This category only appears in the NRR analysis.- NRR believes that j tests in this category may have been altered by either a hydrogen addition or a jogged water addition; however, based upon the similarity .

in MUT level traces that could be produced by either of these

)

operations, it was not possible to detemine which was more probable.

~

! 1GPM The six tests in this category all occurred between October 16 and 18, i 1978. Each of these tests showed results greater than the technical l specification limit for unidentified leakage. Five tests (12A-12E)

were not signed (except 128) by the CR0s performing the tests and were not approved by their Shift Foremen. Test 13 was not signed by a CRO, I I

instead the Shift Foreman signed as the operator and the Shift Supervisor approved the test. These tests were associated with an incident *. hat occurred on October 18, 1978 resulting in the submission of LER 78-62/1T by Met-Ed on November 1, 1978.

Conclusions Based upon NRR's review to date, the technical differences do not appear to be significant to our overall conclusions or our conclusions or recommendations regarding individuals. As a result, we have not revised our technical evaluation based upon our review of the Stier Report.

~

e m

Attachment 1 NRR AND STIER EVALUATION CATEGORY FOR EACH TMI-2 LEAK RATE SURVEILLANCE TEST 8

PERFORMED BETWEEN 09/30/78 AND 03/26/79 NRR TEST EVALUATION STIER TEST EVALUATION No. Evaluation No. Evaluation 5EPTEMBER 1978 1 OK 161 OK OCTOBER 1978 2 OK 160 OK 3 OK 159 OK 4 OK 158 OK 5 OK 157 OK

• 6 F&B , 156* OK *

- *7 F&B 155* OK 8 OK 154 OK

• 9 USPC 153* USPC/Poss F&B 10 USPC 152 USPC 11 OK 151 OK 12 WAN/Poss F8B 150 WAN/F&B 12A 1GPM 149 1GPM 12B 1GPM 148 1GPM 12C 1GPM 147 1GPM 12D 1GPM 146 1GPM

• 12E 1GPM/F&B 145* 1GPM 13 1GPM 144 1GPM

• 14 WAN 143* OK 15 OK 142 OK 16 OK 141 OK 17 H 140 H

• 18 WAh 139* USPC 19 OK 138 OK 20 OK 137 OK 21 OK 136 OK 22 OK 135 OK 23 H 134 H

• 24 WAN 133* OK 25 OK (No Surveillance sheet) ---

• 26 OK 132* USPC 27 USPC 131 USPC

• 28 WAN/USPC 130* USPC 29 F&B/USPC 129 USPC/Poss F&B 30 OK 128 OK 31 LT/USPC 127 LT/USPC

• Identifies tests evaluated differently by NRR and Stier.

• • For tests evaluated as WAP, the first number in parentheses is the time in minutes before the end of the leak rate test that water was added and the second number is the difference in gallons between the amount included in the calculation and the amount shown on the MUT strip chart (NRR using " peak-to-peak" and Stier using " slope off-set).

NRR TEST EVALUATION STIER TEST EVALUATION No. Evaluation No. Evaluation

' NOVEMBER 1978 32 OK 126 OK

! 33 LT 125 LT 34 LT 124 LT 35 LT 123 LT

• 36 LT/H/WAN 122* LT/H 37 LT 121 LT 38 OK 120 OK 39 H 119 H 40 USPC 118 USPC 41 Th Inst 117 Th Inst DECEMBER 1978 l

1 42 OK (No trace on strip chart) 116 OK (Strip chart pen out of ink) 1 - 43 LT/F&B 115 Poss LT/Poss F&B 4 44 LT 114 Poss LT i 45 LT 113 Poss LT i' 46 OK 112 OK 47 LT 111 LT 48 LT 110 LT i 49 OK 109 OK

{ *50 LT/WAN 108* OK (MUT oscillations evident) 51 LT 107 LT 52 LT 106 LT 53 LT 105 LT 54 OK 104 OK 55 OK 103 OK

'56 LT/WAN 102* LT 57 OK 101 OK 58 LT/F&B 100 LT/Poss F&B 59 LT 99 LT 60 OK 98 OK 61 LT 97 LT

'62 WorH 96* OK 63 H 95 H 64 LT 94 LT 65 OK 93 OK l 66 LT 92 LT

! 67 OK 91 OK ,

! *68 OK 90* LT

! *69 LT/H 89* LT 4

70 LT/USPC 88 LT/USPC 71 LT/WAN 87 LT/WAN

• 72 LT/WAN/F&B 86* LT 73 LT 85 LT 74 LT 84 LT 75 LT 83 LT 76 LT - 82 LT  ;

• 77 LT/WAN/USPC 81* LT/USPC

• 78 WAN 80* OK 79 OK 79 OK 80 LT 78 LT

! 81 OK 77 OK 82 LT 76 LT

NRR TEST EVALUATION STIER TEST EVALUATION No. Evaluation No. Evaluation s JANUARY 1979 83 LT 75 LT 84 LT 74 LT 85 LT 73 LT 86 LT 72 LT

• 87 LT/WorH 71* LT 88 OK 70 OK 89 OK 69 OK 90 LT 68 LT 91 LT 67 LT

• 92 LT/WAN 66* LT 93 WAP (23 min-40 gal)** 65* WAP (23 min-9 gal)**

94 WAN 64 WAN

. 95 OK 63 OK

• 96 LT 62* OK (MUT levels do not match)

FEBRUARY 1979

• 97 F&B 61* OK (Poss F&B or mislogged entry)

• 98 WAN 60* OK 99 OK 59 OK 100 OK 58 OK 101 OK 5) OK 102 OK 56 OK 103 OK 55 OK 104 OK 54 OK 105 OK 53 OK 106 F&B 52 Poss F&B 107 OK 51 OK

• 108 USPC 50* OK 109 OK 49 OK

• 110 F&B 48* OK

• 111 F&B 47* OK

• 112 F&B 46* OK (Unusual dip in trace) 113 OK 45 OK j 114 WAN 44 WAN (Poss F&B) .

115 WAN 43 WAN f 116 OK 42 OK 117 OK 41 OK

• 118 OK (30 min-4 gal)** 40* WAP (30 min-25 gal)**

119 H 39 H 120 H 38 H

• 121 WorH 37* OK 122 WAP (13 min-24 gal)** 36 WAP (8 min-12 gal)**

123 H 35 H 124 H 34 Poss H

• 125 WorH 33 Poss H

• 126 WorH 32 Poss H 127 OK- 31 OK 128 F&B 30 Pass F&B i 129 WAP (4 min-50 gal)** 29 WAP (4 min-59 gal)** '

• 130 WorH 28* OK 131 WAP (17 min-26 gal)** 27 WAP (4 min-31 gal)**

• 132 F&B 26* WAP (3 min-10 gal)**

133 WAP (4 min-50 gal)** 25 WAP (4 min-23 gal)**

NRR TEST EVALUATION STIER EVALUATION No. Evaluation No. Evaluation i MARCH 1979 134 H 24 H

• 135 -WAN 23* OK

• 136 WorH 22* OK 137 WAP (3 min-48 gal)** 21 WAP (3 min-128 gal)**

138 WAP (2 min-32 gal)** 20 WAP (2 min-70 gal)**

139 WAP (10 min-182 gal)** 19 WAP (10 min-94 gal)**

140 WAP (1 min-30 gal)** 18 WAP (1 min-113 gal)**

• 141 WAP/F&B (3 min-22 gal)** 17* WAP (3 min-33 gal)**

142 WAP (8 min-59 gal)" 16 WAP (8 min-35 gal)**

• 143 WAN 15* OK

• 144 WAN 14* OK

• 145 WAN 13* OK ,

_ 146 WAP (5 min-25 gal)** 12 WAP (5 min-126 gal)** *

• 147 WAP (15 min-100 gal)** 11* WAP/Poss F&B (15 min-193 gal)**

• 148 WAN 10* Poss H 149 WAP (19 min-36 gal)** 9 WAP (19 min-39 gal)**

150 WAP (24 min-33 gal)** 8 WAP (25 min-8 gal)**

151 WAP (1 min -130 gal)** 7 WAP (6 min-82 gal)**

152 H (48 min-3 gal-0K)** 6 H (48 min- -19 gal-0K)**

153 WAP (7 min -100 gal)" 5 , WAP (7 min -31 gal **

• 154 OK (10 min--6 gal-OK)** 4* WAP (20 min-71 gal **

• 155 OK (17 min- -1 gal-0K)** 3* WAP (17 min-69 gal **

• --- No Test 2* WAP/Poss F&B (22 min-86

• 156 OK (49 min-17 gal-OK)** 1* WAP (49 min -51 gal)**gal)**

Identifies tests evaluated differently by NRR and Stier.

• • For tests evaluated as WAP, the first number in parentheses is the time in minutes before the end of the leak rate test that water was added and the second number is the difference in gallons between the amount included in the calculation and the amount shown on the MUT strip chart (NRR using " peak-to-peak" and Stier using " slope off-set).

Key OK No apparent problems 4

LT Bad level transmitter or instrument used during test WAN Water addition (not included in test) i WAP Water addition (partially included in test due to instrument error)

F&B Feed and bleed operation during test H Hydrogen addition USPC Unstable plant conditions WorH Water or hydrogen addition (unable to differentiate) 1GPM Results greater than TS limit l

1 I

l

Attachment 2 COMPARIS0(i 0F TMI-2 LEAK RATE TESTS WITH DIFFERENT EVAlbATION RESULTS BY NRR AND E. STIER BOTH AGREE TESTS ARE INVALID BUT FOR DIFFERENT REASONS NRR STIER COMMENTS

1. 9 USPC #153 USPC Chart times agree. Plant definitely unstable.

F&B? NRR agrees with possible F&B.

1. 12E 1GFM #145 1GPM Chart times agree. Both agree that results for F8B? unidentified leakage (1.32 gpm) exceeds TS limit and was rounded off to 1 gpm. NRR believes

~

possible F&B at 0640 with approximately 40 more .

gal added to the system than removed. Stier considers trace deflection similar to others during the day. NRR considers 0640 MUT trace the same as the one at 1915 which Stier classifies as possible F&B.

1. 18 WAN #139 USPC Chart times agree within 1 min. Stier considers test invalid due to large change in Lp during test (+7"). NRR agrees; however, MUT level does not decrease as much as expected due to leakage and changes in Lp and Tave. In addition, MUT trace indicates a possible 50 gal of boric acid at 1258.

Trace is the same as logged additions of boric acid at 0700, 0730, 0830, 1125, and 1209. All but 1209 addition were 50 gal. The 1209 addition was 100 gal.

1. 28 USPC #130 USPC Chart times agree. Both agree test invalid due to WAN large oscillations in MUT and Lp. NRR does not agree, however, that 330 gal addition was not made in test period because of time change from DST to EST. Computer time had to have been reset along with clocks and strip chart since MUT level during test only matches EST.

1. 36 LT #122 LT Chart times agree within 3 mins. Both agree LT H H bad and H addition. From strip chart it appears WAN approximately 150 gal of 500 gal addition stated at 2210 was added during the leak rate test. The addition is not discussed in Stier evaluation.

1. 56 LT . #102 LT Chart times agree. NRR agrees with Stier that the WAN rise in MUT level could be caused by bad LT alone without water addition.

1. 69 H #89 LT Chart times agree. NRR believes either H addition LT? or possible switch of LT near end of test. Stier believes bad LT, not H addition.

f 5

NRR STIER COMMENTS

1. 72 LT #86 LT Chart times agree. NRR agrees unstable LT on WANI computer. It also appears approximately 100 gal USPC of water may have been added as part of a F&B operation which comenced approximately 11 mins before end of test (0015) based on change in slope of MUT and large number of F&B operations around

- test. Only LT problem discussed in Stier evaluation.

1. 77 LT r81 LT Chart times off by 22 mins. NRR agrees unstable USPC USPC LT and USPC; however, based upon chsnges in Tave, WAN Lp and MUT level, it also appears small addition of 20-30 gal may have taken place during test.

- #87 LT #71 LT Chart times agree. Both agree unstable LT used -

WorH for test. There is a marked change in slope of MUT level approximately 15 min before end of test which is not discussed in Stier evaluation.  !

1. 92 LT #66 LT Chart times off by 5 mins. Both agree unstable LT WAN used for test. Stier says trace at 0218 not

characteristic of batch addition. NRR agrees; however, MUT level slope is off-set by 5" (150 gal) between beginning and end of test, i

• 125 WorH #33 H? Chart times agree. Agree with Stier that H addition possible. Jogged water addition less probable. Stier's conclusion is supported by j operator testimony.

1. 126 WorH #32 H? Same connents as test #125.

1. 132 F&B #26 WAP Chart times off by 15 mins. Test appears to be

started during F&B with 122 gal being bled and 210 gal being added. The leak rate calculation included 162 gal. Thus, more water (48 gal) was included in the calculation than reflected on the strip chart. Stier does not discuss possible feed and bleed; however, Stier concludes if water addition not made during test, the unidentified leak rate would have been = 2.264 gpm.

l #141 WAP #17 WAP Chart times agree. Both agree that rise in MUT by l F&B addition 3 min before end of test is higher than l

amount logged (due to instrument error). However, it appears test was begun during F&B operation.

I F&B is not discussed in Stier evaluation.

1. 147 WAP #11 WAP Chart times agree. Analyses the same except Stier F&B? believes second addition during test is part of F&B Operation. NRR agrees.

NRR STIER COMMENTS

#148 WAN #10 H7 Chart times agree. Stier confirms MUT tank level 8 is effected during test, but believes it could be j a possible H addition. Not logged. Performed by Hartman's shift. Either H or jogged addition would be consistent with his testimony.

---- #2 WAP NRR did not have this leak rate test thus, it was F&B not evaluated. Stier analysis appears correct.

Test begun during an unlogged F&B. During the F88 about 180 gal of water was added and not accounted for in the test calculation. In addition a 300 gal 1

water addition was logged 22 mins before end of

^

, test. MUT level indication shows 350 gal. Only

- 300 gal included in calculation. Stier used slope off-set to arrive at 386 gal added during test.

J l

i f

- - - - - - . - . _ . -- .~. - -. - - = . -.- .. - - ~ . -

j -

I i

Attachment 3 COMPARISON OF TMI-2 LEAK RATE TESTS WITH DIFFERENT EVALUATION RESULTS BY NRR AND E. STIER l NRR EVALUATED TEST AS INVALID /STIER EVALUATED TEST AS VALID

! NRR STIER CON 9ENTS J

J #6 F&B #156 OK Chart time off between NRR and Stier by 3 hrs and i 27 mins. MUT chart error is large and rapidly changing. Few additions to use for reference.

1 Both analyses have uncertain of chart time. If I chart timing error on part of NRR, F&B may not be during test.

1 ,

1. 7 F8B #155 OK Chart times agree. FSB appears on chart used by -

Stier, but is not discussed in his evaluation.

i #14 WAN #143 OK Chart times agree. Stier says MUT deflections typical. NRR. agrees MUT trace does not reflect batch addition; however, based upon expected

! change in MUT level due to changes in Lp and Tave,

} approximately 100 gal may have been added during 1 leak rate test.

J l #24 WAN #133 OK Chart times off by 17 mins; however, addition in j question is within the test period for both 4

evaluations. NRR believes MUT trace at 1213(60 l

gal) addition is exactly the same as the 25 gal

additions logged in CR0's Log at 1800, 1830 and

2116. Addition is not discussed in Stier's i evaluation.

1 .

j #50 LT #108 OK . Chart times agree. Stier states good transmitter

WAN used for test and MUT fluctuations normal. NRR i does not agree. MUT strip chart off-set by 2"

during test caused by switch of LT shortly after

] start of test and/or jogged water addition of 60 i gals. LTs switched 2010 4 2130 chart time.

i #62 WorH #96 OK Chart times agree. Stier states because bad j transmitter was on strip chart, MUT level chart

can not be used to confirm addition; however, the

unusual trace apparent between 0630 and 0640 i (chart time), interrupting consistent LT-2 trace, j is not discussed in Stier evaluation.

1. 78 WAN #80 OK Chart times agree within 1 min. Stier says MUT trace fluctuates too much to confirm water j addition. MUT should have decreased 2" during j test. Instead level increased .67". The change 1 in Tave and Lp do not account for net difference.

NRR believes possible 50 gal addition during test.

l i

E ___

i I

1. 96 LT #62 OK Chart times disagree by 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />. Both analyses have chart time uncertainties due to plant shutdown, new MUT chart roll and few good

' reference marks. NRR evaluated as LT due to difference between LT-1 and LT-2 of approximately 6" during test. Stier notes that MUT levels do not match well (2") but considers test normal.

1. 97 F&B #61 OK Chart times agree. Water addition of 300 gal logged in CR0's Log at 0100 but addition is not evident on MUT strip chart. NRR believes addition may have been part of a F&B operation. Stier notes possible F&B or logging error on test evaluation sheet but apparently considers test normal Overall Leak Rate Sunmary, (Vol !!!(A),

Table 1). *

1. 98 WAN #60 OK Chart times agree. MUT level increases by .2"

! during test, while NRR calculates that it should have decreased 3.5". The slope of MUT level is off-set near end of test. Stier believes this is caused by starting on low point of MUT level fluctuation and stopping on high point in fluctuation.

1. 108 USPC #50 OK Chart times agree. NRR believes USPC caused large changes in MUT level indication during middle of test, caused either by power level changes or F&B.

Stier believes traces are typical of the day.

1. 110 F&B #48 OK Chart times off by 5 mins. Possible F&B operation near start of test. May have had small effect.

This fluctuation can be seen on MUT strip chart but is not discussed in Stier evaluation.

1. 111 F88 #47 OK Chart times off by 10 mins. NRR believes test was started during FSB operation. Stier shows test starting after F&B. NRR used strip chart error of 1 hr 30 min. Stier uses I hr 40 min.

1. 112 F&B #46 OK Chart times off by 5 mins. Possible F88 operation near middle of test. Amount appears equal and may i

not have affected test. Stier notes unusual dip in trace during test but apparently does not consider it a F&B operation.

1. 121 H #37 OK Chart times off by 1 min. NRR believes Possible H l addition. MUT level trace exhibits the same characteristics as test #120, which was a confirmed H addition. Stier believes trace fluctuations are typical of the day.

4

1. 130 WorH #28 OK Chart times off by 10 mins. Possible WorH addition based on change in slope of MUT during test compared with period just prior and just after test. Stier believes trace flattening is similar to patterns before and after the test and thus, are not unusual.

NRR STIER COMMENTS

1. 135 WAN #23 OK Chart times agree. NRR believes possible small I jogged water addition based on change in slope of MUT. Stier believes flattening similar to other periods during the day and are not unusual. NRR agrees that trace is similar, but since the times of discarded leak rate tests are unknown, leak rate tests could have been run during periods where similar flattening occurs.

1. 136 WorH v22 OK Chart times agree. NRR believes possible jogged water addition or H addition due to 3" rise in MUT level near end of test. Stier believes trace is i similar to others that day. NRR disagrees because trace does not return expected level following

• 4

. fluctuation.

1. 143 WAN #15 OK Chart times agree. Jogged water addition of 80 gal beginning before end of test. Stier believes trace similar to other times of day. NRR does not agree. The trace remains approximately 2.5" above expected slope. Test was performed by Hartman's shift. A jogged water addition would be consistent with Hartman's testimony.

1. 144 WAN #14 OK Chart times disagree by 20 mins. Jogged water addition of 100 gal beginning in middle of test.

Stier believes trace is typical of normal fluctuations and can not confirm jogged addition.

Test was performed by Hartman's shif t. Jogged additions on this shift would be consistent with Hartman's testimony. The Stier analysis notes ,

indicate that the test may have begun before completing a 300 gal addition that was started just prior to the beginning of the test (the portion added during the test would be approximately 60 gal).

1. 145 WAN #13 OK Chart times disagree by 5 mins. Jogged water addition near end of test. Stier believes trace

, is typical and can not confirm jogged addition.

Test performed by Hartman's shif t. A jogged addition would be consistent with his testimony.

1 i

i i

l I

4 j o i

i -

l Attachment 4 i COMPARIS0N OF TMI-2 LEAK RATE TESTS WITH DIFFERENT EVALUATION RESULTS BY NRR ,

} , AND E. STIER l l

^

STIER EVALUATED TEST AS INVALID /NRR EVALUATED TEST AS VALID

! i l NRR STIER COPO4ENTS l #26 OK #132 USPC Chart times agree within 2 mins. Stier notes i Large increase in Lp during test which accounts t for higher than expected slope for MUT. MRR l agrees changes may have affected test; however, j change in Lp is 5.3" vice 7.3" used in Stier evaluation. MUT level decrease is consistent with j . changes in Tave and Lp during test. -

1. 68 OK #90 LT Hand calculation performed on 12/19/78 with data

, from 1815-1917 on 12/18/78. NRR agrees that i source of data most likely from computer since

, fonn "DS-1" was used to record data. At 1815-1917 l on 12/18/78 unstable LT-1 was providing input to i

computer. Source of data was not noted by NRR for this test.

i f #118 OK #40 WAP Chart times agree. Unlogged water addition during 4

middle of test. NRR used " peak to peak" method to i calculate 210 gal addition. The calculation 1 included 206 gal or (4 gal error). Stier used ,

"slopeoffset"tocalculate231galaddition(25 gal error). With no water addition. Stier i calculated unidentified leakage would be 1.7 gpm. t

1. 154 OK #4 WAP Chart times agree. Test contains two 100 gal water additions logged at 40 mins and 10 mins before end 1 of test. The last addition may be mislogged. It

{ appears be have been added 20 mins prior to end of test. Leak rate test calculation included 206 gal

! or 6 gal more than added. Rise in MUT level 1 indication agrees with logged amount. Stier used j slope off-set to arrive at 277 gal added. NRR

! does not agree with slope off-set for this test since slopes are greater during test than prior to or after test. Without this water addition Stier calculated unidentified leakage would be 3.19 gpm.  ;

j #155 OK #3 WAP ' Chart times off by 3 mins. Test contains 300 gal ,

water addition logged 40 mins before end of test.

Leak rate test calculation includes 301 gal or 1 4 gal more than added. Rise in MUT level indication i

! agrees with logged amount. Stier used slope i

off-set to arrive at 370 gal added. NRR agrees in i this case.that slope off-set method better accounts for leakage during time of 300 gal addition since leakage was 8.5 gpm and addition ,

, took approximately 5 min.

f I

I'

NRR STIER COMMENTS

1. 156 OK #1 WAP Chart times off by 5 mins. MUT level rise shows

320 gal water addition. Logged as 300 gal addition 11 minutes after start of test. Leak rate calculation inc?udes 303 gal or 17 gal less than MUT indication. Stier used slope off-set to arrive at 354 gal added during test. NRR agrees 1

~

for the same reason as discussed in Test #155 above.

D 4

l l

I i

I 1

1 I.

I I

i

. Attachmont 5 l

.i Updated: 07/30/85 i NRR EVALUATION OF TMI-2 REACTOR COOLANT SYSTEM LEAK RATE TESTS PERFORMED 09/30/78 TO 03/28/79 j l

BACKGROUND l On March 23, 1984, the Commission directed NRR to review 01 investigative materials concerning falsification of reactor coolant system leak rate surveillance tests at TMI-2 and refer back to 0! those matters which required

further investigation. As a result of that review, NRR determined that follow-up investigation by 01 and further evaluation by NRR was needed in the case of 10 currently licensed operators.

) A joint 01/.NRR investigation and evaluation of these operators is undemay to

- determine whether their current licenses should be revoked, modified or -

suspended under 10 CFR 55.40 or other enforcement action taken under 10 CFR 55.50, due to their possible involvement in preaccident leak rate testing irregularities at TMI-2.

l PURPOSES The purposes of this evaluation are: (1) to provide the necessary background a material to support the joint 01/NRR investigative interviews of the I operators under investigation; and (2) to serve as a technical basis for ,

judging the credibility of answers provided by these individuals concerning i

! their involvement in or knowledge of preaccident leak rate testing irregularities at TMI-2.

CONTENT During the period September 30, 1978 to March 28, 1979, 161 leak rate surveillance test records were retained by the licensee.- Each of these tests was studied in detail to detemine: (1)whichindividualswereinvolvedin l each test and (2) whether evolutions occurred during these tests that were

! contrary to procedural requirements. The results of this review and evaluation are sumarized in the enclosed 11 tables as follows: )

! Table No. Description

1 Involved Individuals i 2 Days Tests Were Required 3 Tests By Time Of Day And Frequency Per Day 4 Tests By Month 5 Questionable Tests By Shift And By Supervisor / Operator 6 Overall Evaluation Of High Leakage Period By Shift

. 7 Overall Evaluation of High Leakage Period By Individual

} 8 Test Index For Individual Operators i 9 Test Approval Index For Shift Foremen / Supervisors 10 Leak Rate Tests By Evaluation Category i 11 Individual Test Synopsis I

i I l l

)

1 i REFERENCE MATERIAL 3

In perfoming this evaluation, raw plant data and infomation from previously j issued neports and analyses on leak rate testing practices at TMI-1/2 were

)

used. The plant data and other reports relied upon in preparing the evaluation include:

! 1. Computer and hand-generated reactor coolant system leak rate surveillance test sheets;

! 2. The Control Room Operator's Log;

3. The Shift Foreman's Log;

! 4. Make-up Tank Level Recorder Strip Chart; l ,

5. Joint 01/NRR interviews, conducted between September 1984 and June 1985,

of 15 former Shift Foremen and Control Room Operators at TMI-2;

6. August 15, 1984 Office of Investigations Report 1-83-010 "TMI-2 Alleged Falsification of Leak Rate Surveillance Test Data;"

) 7. June 13, 1894 Report "TMI-1 Reactor Coolant Inventory Balance Testing "

prepared for GPUN by E. H. Stier (Stier Report);

! 8. April 16, 1984 Office of Investigations Report 1-83-028, "TMI-1 Possible l

Falsification of Reactor Coolant System Inventory Leak Rate Tests;"

9. February 1984 NRC Region I Technical Analysis of TMI-2 Leak Rate Tests.

(ChungAnalysis);

10. June 6, 1983 Memorandum from T. T. Martin (Region I) to H. H. E. Plaine with attached exhibits,

Subject:

Hartman Allegation Summary; l

11. September 1980 "Results cf Faegre & Benson Investigation of Allegations i By Harold W. Hartman, Jr. Concerning Three Mile Island Unit 2." (Faegre &

i Benson Report).

l The reference material identified above is located in Room 5201 of the Air Rights III Building, under the custody of W. T. Russell, Acting Director.

Division of Human Factors Safety, NRR.

I UPDATE OF INFORMATION

! As noted in the " Background" section of this report, NRR's initial

' involvement in this evaluation began in March 1984. As new information has 4

been learned it has been incorporated in the evaluation and the tables updated accordingly. The date of the latest revision ir contained in the

! upper right corner of each table.

CONTACT W.T. Russell,DHFS,NRR(X24803) j R.A.Capra,DHFS,NRR(X24821) i l

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

. Table 1 Updated: 07/30/85.

- Involved Individuals FRESc.NT-STATUS OF FORMER SHIFT SUPERVISORS, SHIFT FOREMEN & CONTROL ROOM OPERATORS AT THI-2 DURING THE PERIOD 09/30/78 TO 03/28/79 INDIVIDUALS STILL LICENSED g Fomer Position / Shift Present Position A. W. Miller Shift Foreman D Manager, Plant Operations TMI-2 Shift Foreman F Foreman, Radwaste Operations TMI-2 C. L. Guthrie Plant Operations Manager TMI ~

H. A. McGovern CR0 F T. F. Illjes .CR0 B Shift Supervisor TMI-2 J. R. Congdon CR0 C Shift Forenan TMI42 A Instructor V'TM1. (Corporate)

E. R. Frederick CR0 C. C. Faust CR0 A Instructor !!! TMI Instructor III TMI M. S. Coleman. CR0 D D Control Room Supervisor Waterford 3 D. I. Olson CR0 Control Room Supervisor Waterford 3 R. R. Bocher CR0 E M. V. Cooper CR0 C' . Control Room Supervisor San Onofre 2/3 INDIVIDUALS NO LONGER LICENSED Name Former Position / Shift Present Position B. A. Mehler Shift $upervisor C Radwaste Operations Manager TMI-1 Shift Sepervisor E Radwaste Operations Manager TMI-2 B. G. Smith W. H. Zewe Shift Supervisor A Superintendent, Titus Station W. T. Conaway Shift Foreman B Radwaste Support Manager TMI-2 C. D. Adams Shift Foreman C Safety Review Engineer TMI-2 K. R. Hoyt Shift Foreman E Radiation Area Recovery Supervisor TMI-2 J. J. Chwastyk Shift Supervisor B Real Estate Broker G. R. Hitz Shift Supervisor D General Physics K. P. Bryan Shift Supervisor F- Self-Employed F. J. Scheimann Shift Foreman A Energy Consultants, Inc.

J. M. Kidwell CR0 B Unknown C. F. Mell CR0 (Trainee) B Bechtel, Inc.

M. D. Phillippe CR0 (Trainee) C Quality Assurance Rep.-Nuclear.

Waterford 3 L. O. Wright CR0 D Seif-employed H. W. Hartman CR0 ,

"E Self-Employed '

J. J. Blessing CRO,(Trainee) ,E United Energy Services F Nuclear Start-Up Services E. D. Hemila CR0 Auxiliary Operator, Maine Yankee CR0 (Trainec) F L. P. Germer m

. ~ . _ . _ , .,

Table 2 Updated: 07/30/85 Days Tests Were Required DETERMINAfl0N OF WHEN LEAK RATE TESTS WERE REQUIRED BETWEEN 10/01/78 & 03/28/79 Date Time Plant Status Days TS Applicable

• Days TS Not Applicable

• 10/01/78 0000 Mode 1 10/08/78 1745 Enter Mode 5 7 10/09/78 2245 Enter Mode 4 1 10/10/78 1930 Enter Mcde 5 1 1C/12/78 0355 Enter Mode 4 3

~

~

11/10/78 1516 Enter Mode 5 28 11/20/78 1625 Enter Mode 4 11 11/23/78 0240 Enter Mode 5 2 12/01/78 0030 Enter Mode 4 8 01/16/79 0500 Enter Mode 5 46 01/30/79 0205 Enter Mode 4 14 03/28/79 0400 Accident 58 Total: Days in period 179 141 38 Note:

• TMI-2 Technical Specification 4.4.6.2, " Reactor Coolant System Operational Leakage,"

required the performance of a Reactor Coolant System Water Inventory Balance Test (leak rate test) 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 state operation. The Technical Specification was applicable in modes 1, 2, 3, and 4. It was not applicable in Modes 5 or 6.

gee w ,e, y

Table 3 Updated: 07/30/85 Tests By Time of Day and Frequency Per Day Breakdown of the time of day tests were conducted Shift on Which Tests Were Completed No. of Days No. of Tests Mid Day Swing Month Year Above Mode 5 Retained 2300-0700 0700-1500 1500-2300 October 1978 26 35 12 12 11 November 1978 11 10 4 1 5 December 1978 31 41 13 11 17 January 1979 17 14 8 2 4 February 1979 28 37 15 8 14 March 1979 28 23 19 2 2 Total 141 160 71 36 53 (44%) (23%) (33%)

Breakdown of the number of tests run per day for the days the surveillance procedure was applicable (above Mode 5)

No. of Days Month Year Above Mode 5 None 1/ Day 2/ Day 3/ Day 4/ Day 5/ Day October 1978 26 2 16 7 0 0 1 November 1978 11 5 2 4 0 0 0 December 1978 31 4 14 12 1 0 0 January 1979 17 6 9 1 1 0 0 .

February 1979 28 5 11 10 2 0 0 March 1979 28 7 19 2 0 0 0 i

. Total 141 29 71 36 4 0 0 (21%) (50%) (26%) (3%) (0) (<1%)

Notes: 1. The data above covers the period 10/01/78 through 03/28/79.

2. The data above reflects only leak rate surveillance test.s retained by l the licensee. Many more tests were conducted; however, the data was discarded.

g e +.w.s ya,- a y~ - ,,- ,

l

Table 4 Updated

07/30/85 Tests By Month NOVEMBER 78 DECEMBER 78 JANUARY'79 FEBRUARY 79l MARCH 79 OCTOBER 78 NONE MODE 4

• 42 NONE NONE 134 1 2 3 33 43/44 83 97 98 135 136 2 4 32 3 5 34 35 45 NONE 99 100 137 NONE 46 84 101 102 138 4 6 7 5 8 36 47 48 85 86 103 104 105 139 37 ~ 38 49 87 88 89 106 140 6 NONE 7 NONE NONE NONE 90 107 108 NONE 50 91 109 110 141 8 MODE 5 1745 NONE 40 51 52 NONE 111 142 9 MODE 4 2245 39 53 54 92 112 143 10 MODE 5 1930 MODE 5 1516 55 56 57 93 113 114 NONE

- 11 MODE 5 MODE 5 58 NONE 115 144 12 MODE 4 0355 MODE 5 MODE 5 59 60 94 116 117 145 146 13 9 MODE 5 61 NONE 118 119 147 14 10 MODE 5 62 63 95 120 148 15 11 12 16 12A MODE 5 64 65 MODE 5 0500 121 122 149 MODE 5 66 67 MODE 5 123 150 17 12B 68 MODE 5 NONE NONE 18 12C ** MODE 5 16 MODE 5 NONE MODE 5 124/125 126 151 19 15 69 MODE 5 NONE NONE 20 17 MODE 4 1625 18 NONE 70 MODE 5 127 152 21 41 71 MODE 5 NONE 153 22 19 73 MODE 5 128 NONE 23 20 MODE 5 0240 72 74 MODE 5 NONE 154 l 24 21 MODE 5 76 MODE 5 129 155 25 22 23 MODE 5 75 26 -

24 MODE 5 77 78 MODE 5 130 131 NONE MODE 5 132 NONE 27 25 26 MODE 5 79 28 27 MODE 5 NONE MODE 5 133 156 ACC.

MODE 5 XXXXXXXXXXXX XXXXXXXXX 29 28 29 MODE 5 NONE 30 30 MODE 5 80 MODE 4 # 96 XXXXXXXXXXXX XXXXXXXXX 31 31 XXXXXXXXXXXX 81 82 NONE XXXXXXXXXXXX XXXXXXXXX

• • 120/12E/13/14

• MODE 4 0030 fMODE 4 0205 Notes: 1. The surveillance procedure for leak rate testing is applicable only in Modes 1 to 4 and steady state conditions. It is not applicable in Modes 5 and 6.

2. The leak rate test numbers are shown in the block corresponding to the date the test was completed.

3. Test numbers to the left represent tests completed on the mid-shift (2300 to 0700), tests done on the day-shift (0700-1500) are shown in the center and tests done on the swing-shift (1500-2300) are shown to the right.

Table 5 Updated: 07/30/85 Questionable Tests By Shift And By Supervisor / Operator Breakdown by Shift 01/01/79 to 03/28/79 09/30/78 to 03/28/79 Shift Shift Control Room Shift %/# Questionable Tests %/# Ouestionable Tests Supervisor Foreman Operators D 93% 14 of 15 73% 24 of 33 Hitz Miller Olson, Wright &

Coleman B 82% 14 of 17 76% 26 of 34 Chwastyk Conaway Illjes, Kidwell &

Mell E 82% 9 of 11 78% 21 of 27 Smith Hoyt Bocher, Hartman &

Blessing gg C 73% 8 of 11 70% 30 of 43 Mehler Adams Congdon, Cooper &

, phillippe

.F 57% 8 of 14 74% 23 of 31 Bryan Guthrie McGovern, tennila

& Germer A 43% 6 of 14 46% 13 of 28 Zewe Scheimann Faust & Frederick Total 73% 54 of 74 69% 111 of 161 Notes: 1. Questionable tests include the following:

a. Water additions not accounted for in the leak rate calculation.

b. Water additions not fully accounted for in the leak rate calculation, (i.e., under-recorded water additions).

c. Hydrogen additions to the make-up tank (MUT) during the performance of a leak rate test.

d. Feed and bleed operations during the performance of a leak rate test.

e. Tests performed during unstable plant conditions, (i.e., not in steady state).

f. Test for which unstable instrumentation was used to provide input to the computer during the performance of a leak rate test. l

2. Most leak rate tests were done by members of the same shift. However, some tests l were performed with members of other shifts. In this situation, the leak rate test was included in the shift statistics for each shift involved. Thus, since some tests are double counted, the columns in the table above do not add up to the totals shown at the bottom of the table.

3. Shift F was not establsihed until 01/01/79. Thus, leak rate tests performed by shift F personnel prior to this date were done with other shifts.

_ __ )

i Table 5 (continued)

Questionable Tests By Shift And By Supervisor / Operator Breakdown by Supervisor For the period 01/01/79 to 03/28/79 For the period 09/30/78 to 03/38/79 -

Name Shift  % Questionable # Questionable # Tests % Questionable # Questionable # Tests I Hitz D 100% 2 2 80% 4 5 5mith E 100% 1 1 100% 2 2 I Hoyt E 90% 9 10 85% 17 20 Conaway B 86% 12 14 79% 22 28 Miller D 86% 12 14 72% 18 25 Adams C 73% 8 11 68% 28 41 Guthrie F 55% 6 11 68% 13 19 Scheimann A 50% 6 12 40% 8 20 Mehler C 0% 0 0 50% 1 2 Hutchison h/A 0% 0 1 0% 0 1 Notes: 1. No'rn: ally the Shift Foreman approved leak rate surveillance tests. Occasionally, the Shift Supervisor signed the test sheet. B. Mehler, G. Hitz and B. Smith were  !

Shift Supervisors.

2. R. Hutchison was not on shift rotation during this period. He was a licensed SRO on TMI-1/2 assigned to Shift Supervisor training.

Breakdown by Operator For the period 01/01/79 to 03/28/79 For the period 09/30/78 to 03/28/79 Name Shift  % Questionable # Questionable # Tests % Questionable # Questionable # Tests Bocher E 100% 8 8 86% 12 14 Hartman E 100% 4 4 80% 12 15 Wright D 100% 4 4 64% 9 14 Olson D 92% 11 12 79% 15 19 l Illjes B 88% 7 8 72% 13 18 l Congdon C 88% 7 8 68% 19 28 Coleman D 88% 7 8 62% 13 21 Blessing E 83% 5 6 82% 9 11 Cooper C 80% 4 5 70% 21 30 Kidwell B 73% 8 11 75% 15 20 Mell B 71% 5 7 69% 9 13 Hemmila F 70% 7 10 80% 12 15 McGovern F 60% 6 10 71% 10 14 Frederick A 43% 6 14 48% 12 25 Faust A 42% 5 12 40% 8 20 Germer F 33% 2 6 64% 9 14 Phillippe C 0% 0 1 57% 4 7 i

Note: An operator is considered to have been involved in a leak rate test if he was the individual who ran the test and signed the surveillance sheet or if he was the CR0 assigned to the panel (i.e., in charge of plant evolutions) and signed the CR0's Log.

Therefore, in most cases there are two operators involved in each tests; however, in some cases the same individual signed both the surveillance test sheet and the CR0's Log for the period during which the test was conducted.

1 1

Table 6 Updated: 07/30/85 Overall Evaluation of High Leakage Period By Shift Period: 01/01/79 to 03/28/79 Evaluation of leak rate tests performed during the period:

Number of Percent of Tests Tests Retained Total number of tests conducted during period....................... 74........ 100%

Total number of tests wi th no apparent probl em. . . . . . . . . . . . . . . . . . . . . . 20. . . . . . . . 27%

Total nuinber o f ques ti onabl e tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54. . . . . . . . 73%

Characteristic that makes test Questionable:

Water additions (partially included in the calculation)...........17........ 23%

Water addi tions (not included in the calculation) . . . . . . . . . . . . . . . . . 10. . . . . . . . 14%

Unstable instrument providing input to computer. . . . . . . . . . . . . . . . . . . 9. . . . . . . . 12%

Feed & bleed operations during test............................... 8........ 11%

Hydrogen additions during test.................................... 6......... 8%

s Hydrogen or water addi tions (unable to di fferentiate) . . . . . . . . . . . . . 6. . . . . . . . . 85 Unstable plant conditions......................................... 1......... 1%

Results showing unidentified leakage greater than 1 gpm. ........ . . 0. . .. .. .. . 0%

Note: Three tests (#87, #92 and #141) are included twice since they fall into two evaluation categories.

Breakdown of leak rate test evaluation by shift:

Shift "A" "B" "C" "D" "E" "F" Supervisor Zewe Chwastyk Mehler Hitz Smith Bryan Foreman Scheimann Conaway Adams Miller Hoyt Guthrie CR0 Frederick Illjes Condgon Olson Booher McGovern CR0 Faust Kidwell Cooper Wright Hartman Hemmila Evaluation Category CR0 ----- Mell Phillippe Coleman Blessing Germer No apparent problem 8 3 3 2 2 6 Water additions 1 1 0 12 0 3 (partially included)

Water additions 1 3 1 1 5 0 (not included)

Unstable Instrumentation 3 4 1 0 1 1 providing computer input Feed & bleed operations 0 1 2 2 2 2 during test Hydrogen additions 0 3 2 0 0 1 during test Hydrogen or water additions 0 2 2 0 1 1 (unable to differentiate) 0 -

0 0 Unstable plant conditions 1 0 1 Results showing unidentified 0 0 0 0 0 0 leakage greater than 1 apm i

l l

i

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

Table 7 Updated 07/30/85 Overall Evaluation Of High Leakage Period By Individual Period: 01/01/79 to 03/28/79 u

2 Control Roem Operator F F I K M C' C P 0 W C B H B M H G a r 1 i e o o h I r o o a 1 c e e u e 1 d 1 n o i s i 1 o r e g m r s d j w 1 g p 1 o g e h t s o m m t e e e d p 1 n h m e m s v i e r s 1 o e i t a r a i e 1 r i 1 n r p n n n r a c p g n k e Total

-j Evaluation of Test Shift A A B B B C C C D D D E E E F F F Number s No apparent problem 7 8 1 3 2 1 1 1 1 0 1 0 0 1 4 3 4 20 Water additions (partially 1 1 1 1 - - - - 11 4 6 - - - 3 3 0 17 included in calculation)

Water additions (not 1 1 1 2 2 -

1 - - - - 5 3 2 - - - 10 includedincalculation)

Unstable instrument providing 3 3 2 1l 3 1 - - - - -

1 - 1 - -

1 9 input to computer Feed & bleed operations - - - 1 1 1 - - 1 1 - 3 1 1 2 2 - 8 l during test Hydrogen additions - - 2 2 -

2 1 - - - - - - - - 1 1 6 Water or hydrogen additions - -

1 2 - 2 2 - - - - - -

1 1 1 -

6 (unable to differentiate)

Unstable plant conditions -

1 - - - 1 - - - - - - - - - - -

1 Results showing unideitified - - - - - - - - - - - - - - - - - O leakage greater than 1 gpm humber of Questionable tests 5 6 7 8 5 7 4 0 11 4 7 8 4 5 6 7 2 54 Total Tests 12 14 6 ,11 7 8 ,5 1 12 4 8 8 4 6 10 10 6 74 Notes: 1. Tests were included under an individual's name if.he either signed the test as the operator performing the test or was the CR0 who had the panel and signed the CR0's Log for the period during which the test was conducted.

~

2. Rows and columns do not add together to provide totals since tests may involve one or more operators. Also, three tests (#87, #92 and #141) fall into two evaluation categories.

e

i Table 8

' Updated: 07/30/85 i Test Index For Individual Operators Shift CR0/* Test Number (Top Row = Questionable; Bottom Row =No Apparent Problems)

A Faust .

10, 12A, 74, 83, 84, 85, 98, 149 4G /, 8~ 20, 21,"22, 25, 81, 102,'105, 113,.118, 154, 155, 156 .

A Frederick 10, 12, 12A, 64, 73, 80, 83, 84, 85, 98, 108, 149 48% 20, 21, 22, 25, 81, 102, 105, 113, 118, 127, 154, 155, 156 B Illjes 9. 28, 31, 47, 56, 69, 90, 91, 115, 123, 124, 125, 147 72% 11, 22, 32, 46, 117 B Kidwell 23, 31, 34, 47, 56, 71, 76, 87, 110, 114, 115, 119, 124, 125, 147 75% 11, 16, 100, 104, 117 B Mell 9, 23, 28, 45, 90, 91, 92, 110, 114 69% 16, 32, 100, 104 C Congdon 6, 12D, 12E, 18, 24, 29, 36, 39, 48, 53, 58, 80, 86, 108, 112, 120,

. 68% (continued) 126, 130, 134 -

2, 5, 8, 15, 19, 26, 42, 49, 154 C Cooper 6, 120, 12E, 13, 14, 18, 24, 29, 36, 37, 39, 43, 44, 48, 51, 59, 62, 70% (continued) 120, 126, 130, 135 2, 5, 8, 15, 26, 42, 49, 55, 95 C Phillippe 27, 43, 80, 82 57% 19, 79, 89 D Olson 7, 14, 33, 66, 93, 122, 129, 131, 133, 137, 139, 140, 141, 142, 146 79% 3, 30, 38, 88 0 Wright 12B, 35, 36, 70, 78, 133, 140, 141, 142 64% 1, 30, 38, 54, 60 D Coleman 7, 12B, 33, 35, 50, 70, 93, 122, 129, 131, 132, 138, 146 62% 1, 3, 46, 54, 57, 60, 79, 88 E Bocher 120, 17, 75, 77, 94, 96, 97, 128, 143, 144, 145, 148 86% 4, 68 E Hartman 120, 40, 52, 61, 63, 72, 75, 77, 94, 128, 143, 145 80% 4, 65, 67 E Blessing 34, 41, 63, 72, 96, 97, 136, 144, 148 82% 67,127 F McGovern 50, 64, 73, 74, 106, Ill, 121, 150, 151, 153 71% 99, 103, 109, 116 F Hennila 45, 58, 69, 71, 76, 106, 111, 121, 150, 151, 152, 153 80% 99, 101, 107 ,

F Gemer 12, 17, 34, 40, 41, 52, 61, 96, 152, 64% 65, 101, 103, 107 109 Notes: 1. * - Indicates percentage of questionable tests 09/30/78 to 03/28/79.

2. Tests are numbered #1 (09/30/78) through #156 (03/28/79). Test #83 was the first test performed after 01/01/79.

3. Shift F was not established until 01/01/79. Thus, shift composition was different prior to 01/01/79.

Table 9 Updated: 07/30/85 Test Approval Index For Shift Foremen / Supervisors Shift Approved By/* Test Number (Top Row = Questionable, Second Row =No Apparen't Problems)

E A Scheimann 10, 12A, 83, 84, 85, 98, 108, 149 (40%) 20, 21, 22, 25, 30, 46, 102, 105, 113, 118, 127, 155 8 Conaway 9, 23, 28, 31, 45, 47, 58, 69, 71, 76, 87, 90, 91, 92, 110, 114, 115

~

(79%) (continued) 119, 123, 124, 125, 147 4, 11, 15, 32, 100, 117 C Adams 6, 12D, 12E, 14, 18, 24, 27, 29, 36, 37, 39, 43, 44, 48, 51, 53, 59, (68%) (continued) 62, 80, 82, 86, 111, 112, 120, 126, 130, 134, 135 2, 5, 8, 15, 19, 26, 42, 49, 55, 79, 89, 95, 154 C Mehler 13 (50%) 68 D Miller 12B, 33, 35, 50, 66, 70, 93, 122, 129, 131, 132, 133, 138, 139, 140,

, (72%) (continued) 141, 144, 146 .

1, 38, 54, 57, 60, 88, 156 D Hitt 7, 78, 137, 142 (80%) 3 E Hoyt 52, 56, 61, 63, 64, 72, 75, 77, 94, 96, 97, 128, 136, 143, 144, 145, (85%) (continued) 148 65, 67, 104 E Smith 12, 148 (100%) None F Guthrie 12C, 17, 34, 40, 41, 73, 74, 106, 121, 150, 151, 152, 153 (68%) 81, 99, 101, 103, 107, 109 h/A Hutchison None (0%) 116 Notes: 1. * - Indicates percentage of questionable tests 09/30/78 to 03/28/79.

2. Normally the Shift Foreman approved leak rate surveillance tests. Occasionally, the Shift Supervisor signed the test sheet. B. Mehler, G. Hitz and B. Smith were Shift Supervisors.

3. R. Hutchison was not on shift rotation during this period. He was a licensed SRO on TM1-1/2 assigned to Shift Supervisor training.

I 1

rw

Table 10 -Updated: 07/30/85

. Leak Rate Tests By Evaluation Category Breakdown of the number of leak rate surveillance tests exhibiting questionable characteristics.

Number 1 of Tests

. Characteristic that makes test questionable of tests Retained Test with no apparent problem 50 31s pstable instrument used as input to computer 42 261 sater additions not included in calculation 22 141 ,

Water additions not fully included in calculation 17 115 Feed and bleed operations during test 15 95 Hydrogen additions 12 75 Unstable plant conditions during test 10 65 Water or hydrogen addition (unable to differentiate) 7 41 Results greater than 1 gpm which were retained. 6 41 Note: A total of 161 tests were retained during the period 09/30/78 and UT/Ili/79. Of these 14 fall into two categories and 3 fall into three categories. ,

Tabulation of leak rate tests by evaluation category. The Shift performing the test is indicated beside the test number. If more than one shift was involved, the shift letter for each shift involved is listed.

s No Aoparent Unstable Wster Additions Feed & Bleed Water or Hydrogen Problem Instrument Diot included) Doerations Additions W "31 B Iz F/A/L 5 G OZ G

• 2C 33 D 14 C/D 7 D '87 B 30 34 F/E/B 18 C *12E C 121 F 4 E/B 35 D 24 C *29 C 125 B 5C '36 C/D *28 8 *43 C 126 C 8C 37 C '36 C/D '58 F/C/B 130 C 11 8 41 F/E '50 D/F '72 E 136 E 15 C '43 C *56 B/E 97 E 16 B 44 C '71 B/F 106 F Leak Rate Exceeds 19 C 45 B/F *72 E 110 B T5 Limit of I som 20 A 47 8 *77 E 111 F/C lZA A 21 A 48 C 78 0 112 C 12B D 22 A/B '50 D/F '92 B 128 E 12C E/F 25 A 51 C 94 E 132 D 12D C 26 C 52 E/F 98 A *141 D *12E C

! 30 D/A 53 C 114 B 13 C l 1 32 B '56 B/E 115 8 Hydrogen 1 38 0 *58 F/C/B 135 C Adattions 42 C 59 C 143 E 17 F/L 46 D/8/A 61 E/F 144 E/D 23 8  ;

49 C 64 F/A/E 145 E '36 C/D '

54 0 66 0 148 E 39 C 55 C *69 B/F 63 E 57 0 *70 0 Water Additions *69 B/F 60 0 *71 B/F tonly parttally 119 B 65 F/E '72 E included) 120 C 67 E 73 F/A 93 p L ZJain 123 B 68 E/C 74 F/A 122 D i ;13 min?124 8 i 79 C/D 75 E 129 D 1 4minh 134 C 81 A/F 76 B/F 131D(,17 ming 152 F 88 0 *77 E 133 D L emin' l i

89 C 80 C/A 137 D [ hin Unstable 95 C 82 C 138Dq 2 min) Plant 99 F B3 A 139 D 1,10minJ Conditions 100 B 84 A 140 D L lainj s s 101 F 85 A *141 D L 3 min i 10 A . I 102 A 86 C 142 D f 3minh' 27 C l 103 F '87 8 146 D 5 min l *28 B

• 90 B 147 8 15 min) *29 C 104 B/E 105 A 107 F 91 B

'92 8 149 A 1 . mnJ 150 F L24 min;I I(19i *31 8 40 F/E 109 F 96 F/E 151 F f loin i '70 0 113 A 153Fd7minj '77 E 116 F 108 C/A 117 B (time shown is 118 A the time before

=

127 A/E end of test that 154 A/C water was added) 155 A 156 A/D *1dentifies tests that fall into more than one category.

I

~

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

Table 11

, Individual Test Synopsis .

This table provides a synopsis of each of the TMI-2 leak rate surveillance tests which were retained by the licensee and conducted between the period September 30, 1978 and March 28, 1979.

e

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

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 1 Test No. Date Start 5 top Test /CR0 Approved Coments (times are clock time)

'l 09/30 1734 1834 Coleman Miller No water or hydrogen additions.-

,1914 2014 Wright 2 09/30 2323 0023* Cooper Adams ho water or hydrogen additions.

0153* 0253* Congdon *10/01/78 3 10/01 1929 2029 Olson Hitz No water or hydrogen additions.

0004* 0104* Coleman *10/02/78 4 10/02 0728 0828 Hartman Conaway No water or hydrogen additions. l 1258 1358 Bocher l 5 10/02 2305 0005* Cooper Adams No water or hydrogen additions.

0505* 0605* Congdon *10/03/78 i

6 10/04 0512 0612 Congdon Adams Feed and bleed operation of 160 gal 1445 1545 Cooper during test. Net effect on test appears negligible. No log entry.

Not included in calculation.

7 10/04 2034 2134 Olson Hitz Feed and bleed operation of 200 gal 0930* 1030* Coleman during test. Net effect on test appears negligible. No log entry. ,

Not included in calculation.

• 10/05/78 )

1 8 10/05 0029 0129 Cooper Adams No water or hydrogen additions.

1339 1439 Congdon 9 10/13 1116 1216 Mell Conaway Unstable plant conditions (startup 1121 1221 111jes in progress). Note: Unidentified leak rate = -8.5 gpm.

10 10/14 1504 1604 Frederick Scheimann Unstable plant conditions (startup 1509 1609 Faust inprogress).

11 10/15 1140 1240 111jes Conaway No water or hydrogen additions.

1140 1240 Kidwell 12 10/15 1928 2028 Germer Smith Water addition of 150 gal at 1940.

1928 2028 Frederick 14o log entry. Between 2000 and 2015 large decrease and then increase in MUT level caused by shifting of FW demands or feed & bleed. With gross leak rate of 2.5 gpm, MUT 1evel should have decreased 5" during test.

Instead MUT level increased 2".

KEY 5 tart /5 top time: Top entry is clock time the test was started and ended. The second entry is the MUT strip chart time that corresponds with the leak rate test clock time.

Test /CR3: Top name is the individual who signed the leak rate surveillance test record as the operator performing the test. The bottom name is the individual who signed the CRO's Log for the period during which the test was performed (i.e., panel operator). l Approved: The name of the SRO who signed the leak rate surveillance test record as app' roving the test. Usually test approval- was signed by the Shift Foreman, i

"? (beside test no.): Evaluation results questionable.

{

l 1

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 2 Test No. Date Start Stop Test /CR0 Approved Coments (times are clock time) l2A 10/16 1935 2035 Frederick

• Scheimann* Unsigned, unapproved leak rate test 1927 2027 Faust indicating unidentified leak rate of 2.56 gpm. Not filed with surveillance tests.

• Assumptions based on shift composition.

12B 10/17 1328 1428 Goleman Miller

• Signed by Coleman but not approved 1323 1423 Wright leak rate test indicating unidentified leak rate of 2.07 gpm.

Not filed with surveillances tests.

• Assumption based on shift composition.

12C 10/18 0514 0614 Hartman* Guthrie* Unsigned, unapproved leak rate test

~

0509 0609 Bocher indicating unidentified leak rate of 1.77 gpm. Note on test: " rounds off high but is corrected by leak rate 10/18/78 7:35:27 start time, ie into action statement at 5:13:02 out of it at 7:35:27." Not filed with surveillance tests.

• Assumptions based on shift composition.

12D 10/18 0736 0836 Congdon* Adams

• Unsigned, unapproved leak rate test 0731 0831 Cooper indicating unidentified leak rate of 1.29 gpm. Handwritten note on test: "0K by roundoff."

Not filed with surveillance tests.

• Assumptions based on shift

' composition.

12E? 10/18 0859 0959 Congdon* Adams

• Unsigned, unapproved leak rate test 0854 0954 Cooper indicating unidentified leak rate of r

1.32 gpm. Handwritten note on test:

"OK by roundoff." Possible feed and bleed operation at 0940. Net effect i approximately +40 gal. No log entry.

Not included in calculation. Not i filed with surveillance tests. l

• Assumptions based upon shift composition.

13? 10/18 1018 1118 Adams

• Mehler* Unidenfified leakage 1.02 gpm.

1013 1113 Cooper Handwritten note on Test: " Net unidentifed leak-rate rounded off to nears (sic) whole number."

• Test signed by SF and approved by SS.

Not filed with surveillance tests.

14? 10/18 1221 1321 Olson Adams No water or hydrogen addtions 1216 1316 Cooper -apparent on strip chart; however, due to gross leak rate and decreases ,

in MUT level, Tave and Lp, l approximately 100 gal of water may have been added. Not included in calculatic t Not filed with sur-veillance tests.

i THI-2 Leak Rate Tests (09/30/78-03/28/79) Page 3 Test No. Date Start Stop Test /CR0 Approved Conenents (times are clock time) 15 10/19 0746 0846 Cooper Adams No water or hydrogen additions. ,

0736 0836 Congdon 16 10/19 1849 1949 Kidwell Conaway No water or hydrogen additions.

1844 1944 Mell Note: Power increased from 61% to 63% during test. No apparent effect on test.

17 10/20 0213 0313 Bocher Guthrie Hydrogen addition. CR0 Log states:

0203 0303 Germer 0214 vented MUT. 0230 added 18#

hydrogen to MUT. 0308 added more hydrogen. Effect uncertain. MUT level decreases -6", pressurizer level increases +10.5" during test.

Slope of MUT generally -2"/Aour.

18 10/21 1239 1339 Cooper Adams Water addition of 50 gal, at 1258.

1205 1305 Congdon No log entry. Not included in calculation.

19 10/22 0755 0855 Phillippe Adams No water or hydrogen additions.

0730 0830 Phillippe/

Congdon 20 10/23 0405 0505 Frederick Schiemann No water or hydrogen additions.

0345 0445 Faust 21 10/24 0311 0411 Faust Scheimann No water or hydrogen additions.

0255 0355 Frederick 22 10/24 2230 2330 Frederick Scheimann No water or hydrogen additions.

2222 2322 Illjes/

Faust 23 10/25 1627 1727 Kidwell Conaway Hydrogen addition logged at 1717.

1619 1719 Mell Initially level decreases -2" then

~

increases +3". Net effect appears negligible.

24? 10/26 2007 2107 Cooper Adams Possible water addition of 60 gal. at 1959 2059 Congdon 2013. No log entry. Not included in calculation.

25 10/27 0440 0540 Frederick

• Scheimann* No leak rate surveillance test sheet. ,

0432 0532 Faust Only record of test is log entry in  ;

Shift Foreman's Log. Completed test '

0540.

• Assumptions based on shift composition.

26 10/27 1703 1803 Congdon Adams No water or hydrcgen additions. I 1655 1755 Cooper 27 10/28 1605 1705 Phillippe Adams Invalid test. Large oscillations in 1605 1705 Phillippe MUT level indication. Results not reliable.

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 4 Test No. Date Start Stop Test /CR0 Approved Connents (times are clock time) 28 10/29 0132 0232 Mell Conaway Invalid test. Large oscillations in s0132 0232 111jes MUT level indication. In addition, water addition of 330 gal. made during test. Logged in CR0 log. Not included in calculation.

29 10/29 1520 1620. Cooper Adams Invalid test. Large oscillations in 1520 1620 Congdon MUT level indication. In addition, feed and bleed operation of approximately 800 gal. perfonned during test. No log entry not included in calculation.

NOTE: LT-2 TAKEN OUT OF SERVICE AT 0130. LT-1 VENTED BY I&C AT 0913.

30 10/30 1131 1231 Olson Scheimann No water or hydrogen additions.

1131 1231 Wright Note: Difference between LT-1 and LT-2 is 0".

31 10/31 0341 0441 111jes Conaway Unstable plant conditions. MUT level 0336 0436 Kidwell decreases -7.32" and pressurizer

level increases +10.52". Note:

Difference between LT-1 and LT-2 is 13".

32 11/02 0332 0432 111jes Conaway No water or hydrogen additions.

0340 0440 Mell Note: Difference between LT-1 and LT-2 is approximately 0".

33 11/02 1829 1929 Coleman Miller Invalid test. Note: Difference 1839 1939 Olson between LT-1 and LT-2 is 14".

34 11/03 0626 0726 Blessing Guthrie Invalid test. LT providing input 0646 0746 Kidwell/ to computer is pegged at 99".

Genner 35 II/03 1658 1758 Wright Miller Invalid tests. LT providing input tc 1708 1808 Coleman computer pegged at 99".

NOTE: LT-1 AND LT-2 ISOLATED FROM 1955 TO 2105 (WRITTEN ON MUT STRIP CHART).

36 11/05 2215 2315 Congdon Adams Water and hydrogen addition. Test 2227 2327 Wright / begun during a 500 gal. water Cooper addition. 150 gal. added during test. Log entry for 500 gal.

addition at 2210. Not included in calculation. Hydrogen added at 2230

' (some.effect). Hydrogen addition logged in CR0 Log at 2230. Note: ,

Both hydrogen and water additions (

done on Wright's shift (Shift D).

Note: Difference between LT-1 and LT-2 is 15". l l

l l

TMI-2LeakRateTests(09/30/78-03/28/79) Page 5 Test No. Date Start Stop Test /CR0 Approved Coments (times are clock time) 37 11/06 0204 0304 Cooper Adams Invalid test. Difference between p216 0316 Cooper LT-1 and LT-2 is 15 .

NOTE: 1615 LT-1 AND LT-2 RECALI-BRATED AND BACK IN SERVICE (CR0'S LOG).

38 11/06 1931 2031 Olson Miller No water or hydrogen additions.

1945 2045 Wright 39 11/09 0409 0509 Cooper Adams Hydrogen addition 0415. Logged in 0415 0515 Congdon CR0 Log as a late entry (small effect 2" or 60 gal.)

40 11/09 1954 2054 Hartman Guthrie Invalid test. Plant evolut.fons in

~

2009 2109 Genner progress effecting MUT level indication.

41 11/22 0217 0317 Germer Guthrie Invalid test. Tha indicates $$$ and 0157 0257 Blessing Tave indicates -17.2'F.

427 12/01 1701 1801 Cooper Adams Unable to read MUT strip chart.

2111 2211 Congdon Plant heatup in progress.

LT-1 UNSTABLE PERIOD BEGINS. LT-1 REMAINS UNSTABLE THROUGH JANUARY 11,

! 1979.

I 43 12/02 1632 1732 Phillippe Adams Invalid test. Feed and bleed 2042 2142 Cooper operation during test. No log entry.

Not included in calculation. Note:

Unstable LT-1 used as input to computer to test.

44 12/02 1918 2018 Cooper Adams No water or hydrogen additions.

2330 0030* Cooper Note: Unstable LT-1 used as input to computer. *12/03/78.  ;

i 45 12/03 0510 0610 Mell Conaway No water or hydrogen additions.

0920 1020 Hemila Note: Unstable LT-1 used as input to computer for test.

12/03 0945* 1045* Coleman* or Miller No surveillance test sheet, only log 1355* 1455* Wright

• entries in CR0 and SF Logs. Note:

Olson Unstable LT-1 used as input to computer for test.

• Assumptions based on shift composition.

46? 12/04 0628 0728 Coleman Scheimann No water or hydrogen additions.

1038 1138 Illjes/ Note: Stable LT-2 used as input to

, Coleman -computer.

__ _ _ . _ _ _ _ _ _ _ _ . _ _ . _ _ . _ _ _ . _ _ _ _ _ _______J

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 6 Test No. Date Start Stop Test /CR0 Approved Comments (times are clock time) 47 12/05 0507 0607 Illjes Conaway No water or hydrogen additions. .

0917 1018 Kidwell Note: Unstable LT-1 used as input to computer for test.

48 12/05 1841 1941 Congdon Adams No water or hydrogen additions.

2251 2351 Cooper Note: Unstable LT-1 used as input to computer for test.

49 12/06 1905 2005 Cooper Adams No water or hydrogen additions.

2305 0005* Congdon Note: Stable LT-2 used as input to computer for test. *12/07/78.

12/07 0720 1000 Bocher* or Hoyt* MUT strip chart has handwritten note:

1120 1400 Hartman* " Switched to LT-1 for Leak rates."

Blessing No surveillance test sheets retained.

Assume test results discarded.

• Assumptions based upon shift composition.

l 12/07 1740 2000 Olson* or Miller

• MUT strip chart switched to LT-1.

2140 2400 Coleman* No surveillance test sheets Wright retained. Assume test results discarded.

• Assumptions based upon shift composition.

12/07 2320 0035**Illjes* or Conaway* MUT strip chart switched to LT-1.

0235** 0350**Mell* No surveillance test sheets Kidwell retained. Assume test results discarded.

• Assumptions based upon shift composition. **12/08/78.

12/08 0035 0135 Illjes* or Conaway Leak rate test results logged in CR0 0350 0450 Mell* Log at 0130 and in SF Log at 0135.

Kidewll Results: Unidentified leakage.

.6998 gpm. Note: Unstable LT-1 used as input to computer for test.

• Assumptions based upon shift i composition.

50 12/08 1617 1717 McGovern Miller Water addition of 60 gal. at 1622. 1 2005 2105 Coleman Offset of MUT strip chart 2". No log entry. Not included in calculation. ,

Note: Unstable LT-1 used as input to I computer for test. l i

51 12/08 2334 0034* Cocper Adams No water or hydrogen additions.

0339* 0439* Cooper Note: Unstable LT-1 used as input to computer for test. *12/09/78.

52 12/09 1149 1249 Hartman Hoyt No water or hydrogen additions.

1549 1649 Genner Note: Unstable LT-1 used as input te computer for test.

l \

4 THI-2 Leak Rate Tests (09/30/78-03/28/79) Page 7

Test No. Date Start Stop Test /CR0 Approved Consnents (times are clock time) 12/09 ** ** Wright

• or Miller ** Leak rate test results logged in f* Coleman* CR0 Log at 1655 and in SF Log at 1800

Olson Results

Unidentified leakage

.2607 gpm. . Note: Unstable LT-1 used as input to computer for test.

• Assumptions based upon shift composition.

53 12/09 2252 2352 Congdon Adams No water or hydrogen additions.

0252* 0352* Congdon Note: Unstable LT-1 used as input to computer for test. *12/10/78.

54 12/10 1820 1920 Coleman Miller No water or hydrogen additions.

2220 2320 Wright Note

Stable LT-2 used as input to

. computer for test. -

55 12/11 0029 0129 Cooper Adams No water or hydrogen additions.

0429 0529 Cooper Note: Stable LT-2 used as input to computer for test.

I 56? 12/11 0650 0750 Illjes Hoyt Possible water addition of 60 gal. at 1050 1150 Kidwell 0745. No log entry. Not included in calculation. MUT level should have decreased -2" during test; however, level increased +2". Note: Unstable LT-1 used as input to computer for test.

57 12/11 1659 1759 Coleman Miller No water or hydrogen additions.

2059 2159 Coleman Note: Stable LT-2 used as input to computer for test.

58 12/12 1010 1110 Hensnila Conaway Water addition of 330 gal. About 1410 1510 Congdon 210 gal. added during the test as part of a feed and bleed operation.

No log entry. Not included in '

calculation. Note: Unstable LT-1 used as input to computer for test.

59 12/12 2255 2355 Cooper Adams No water or hydrogen additions.

0255* 0355* Cooper Note: Unstable LT-1 used as input to computer for test. *12/13/78. i 60 12/13 1950 2050 Coleman Miller No water or hydrogen additions. )

2350 0050* Wright Note: Stable LT-2 used as input to computer for test. *12/14/78.

12/14 0145 0245 Cooper

• or Adams MUT strip chart switched to LT-1 for 0545 0645 Phillippe* 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> period between 0145 and 0245.

Congdon No surveillance test sheet retained.

-Assume test results discarded.

• Assumptions based upon shift composition.

'__ _ _ _ _ _ _ _ -_. _ _ _ . ~ _ . _ _ . _ _ , _ _ _ . _ _ . _ _ . . _ _ _ _ _ _ _ . , , _ _ . _ . _ _ _ - _ _ _

~

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 8 Test No. Date Start Stop Test /CR0 Approved Connents (times are clock time) 12/14 0354 0454 Cooper

• or Adams Leak rate test results logged in CR0 0745 0854 Phillippe Log at 0530 and in SF Log at 0454.

Congdon Results: Unidentified leakage.

.6995 gpm. During test period approximately 350 gal. feed and bleed operation occurred. Note: Stable

- LT-2 used as input to computer to test.

61 12/14 1944 2044 Gemer Hoyt No water or hydrogen additions.

2344 0044* Hartman Note: Unstable LT-1 used as input to computer for test. *12/15/78.

l 62? 12/15 0208 0308 Cooper Adams Possible small water addition or 0608 0708 Cooper hydrogen addition between 0230 and 0240. Note Stable LT-2 used as input to computer for test. Two other leak rate tests were probably run first between 2320 and 0150 based upon switching of level transmitters.

63 12/15 1822 1922 Hartman Hoyt Hydrogen addition at 1845. Logged in 2020 2120 Blessing A0's Log. Most of initial effect decayed off by end of test. Net effect appears to be +1.5" in MUT level indication. Note: Stable LT-2 used as input to computer for test.

64 12/16 1328 1428 McGovern Hoyt No water or hydrogen additions.

1730 1830 Frederick Note: Unstable LT-1 used as input to computer for test.

65 12/16 2004 2104 Gemer Hoyt No water or hydrogen additions.

0020* 0120* Hartman Note: Stable LT-2 used as input to computer for test. One other leak rate test was probably run first between 1855 and 1955 based upon switching of level transmitters.

• 12/17/78.

66 12/17 0220 0320 Olson Miller No water or hydrogen additions.

0220 0320 Olson Note: Unstable LT-1 used as input to computer for test.

67 12/17 1528 1628 Hartman Hoyt No water or hydrogen additions.

1525 1625 Blessing Note: Stable LT-2 used as input to computer for test.

TMI-2LeakRateTests(09/30/78-03/28/79) Page 9 Test No. Date Start Stop Test /CR0 Approved Connents (times are clock time)

, 68 12/18 1815 1915 Morch Merler No water or hydrogen additions.

1815 1915 Booher Hand calculation.

69 12/20 0809 0905 111jes Conaway Hydrogen addition 0855 (3.5" or 100 0809 0905 Hemmila gal effective). No log entry. Note:

Unstable LT-1 used as input to computer for test. Possible switch of level transmittees at end of test.

70 12/21 0259 0359 Coleman Miller Invalid test. Plant heatup in 0259 0359 Wright progress. Note: Unstable LT-1 used as input to computer for test.

, 71 12/22 0810 0910 Hemmila Conaway Water addtion of 200 gal. at 0820.

s 0810 0910 Kidwell No log entry. Not included in calculation. Note: Unstable LT-1 used as input to computer for test.

72? 12/22 2326 0026* Blessing Hoyt Possible water addition of 100 gal.

2326 0026* Hartman at 0015. Appears feed and bleed operations may have started prior to the end of the test. No log entry.

No included in calculation. Note:

Unstable LT-1 used as input to computer for test. *12/23/78 73 12/23 2048 2148 Frederick Guthrie No water or hydrogen additions; 2048 2148 McGovern however, by using Unstable LT-1 for input to computer for the test MUT level increased +.542" instead of decreasing -2". Net effect approximately 100 gal.

74 12/24 1736 1836 McGovern Guthrie No water or hydrogen additions. CR0 1736 1836 Faust Log shows water added at 1830 (i.e. i during test); however, log entry is in error. Water added at 1730.

Note: Unstable LT-1 used as input to computer for test.

75 12/25 0120 0220 Hartman Hoyt No water or hydrogen additons.

0120 0220 Booher Note: Unstable LT-1 used as input to computer for test.

76 12/25 0752 0852 Hemmila Conaway No water or hydrogen additions; 0752 0852 Kidwell however, by using unstable LT-1 for input to computer for the test, MUT 1evel increased +1.5" instead of decreasing -2". Net effect

.approximately 85 gal.

l l l

, , TMI-2LeakRateTests(09/30/78-03/28/79) Page,10 Test No. Date Start Stop Test /CR0 Approved Consnents (times are clock time)

~777 12/26 0249 0349 Bocher Hoyt Possible small water addition 1 0227 0327 Hartman of 20 to 30 gal during test.

' Unstable plant conditions for test.

Change in pressurizer level of -13.9" i and change in MUT level of +4.4" during test. Note: Unstable LT-1 used as input to computer during test. Signficant level difference between LT-1 and LT-2.

i 787 12/26 0916 1016 Wright Hitz Possible water addition of 50 gal.

0900 1000 Wright during test. MUT level should have decreased -2" during test; however, level increased +.67". Change in i Tave and pressurizer level do not j

- account for net difference. Note:

Stable LT-2 used as input to computer for test.

79 12/27 0810 0910 Coleman Adams No water or hydrogen additions.

0825 0925 Phillippe Note: Stable LT-2 used as input to l computer for test.

12/28 1140 1240 Phillippe* Adams

• MUT strip chart switched to LT-1 for 1200 1300 or Cooper 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> period between 1140 and 1240.

Congdon No surveillance test sheet retained.

Assume test results discarded.

• Assumptions based upon shift composition.

80 12/30 0615 0715 Phillippe Adams No water or hydrogen additions; 0635 0735 Frederick /. however, MUT level should have Congdon decreased -2" during test. Instead level increased +3.95". Note:

Unstable LT-1 used as input to i

computer for test. Significant difference in readings between LT-1 1 and LT-2. Net effect approximately I

+193 gal. accounting for the large )

negative leak rate of -3.8 gpm. LRT was started on midshift (Frederick CRO) and was completed on day shift by shift C.

81 12/31 0516 0616 Frederick Guthrie No water of hydrogen additions.

0537 0637 Faust Note: Stable LT-2 used as input to computer for test.

l l

. _ , _ . _ . . _ _ _ _ _ . . _ , _ _ . , _ . _ _ _ . _ . . _ . . - . . _ _ . _ , _ . _ _ _ _ _ _ . . _ - _ . _ , . , _ , , . . _ .. _ ._,,.._._-m._,,,,__,_.,,____m,.,. ,%_m.._,_m.

l TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 11 Test No. Date Start Stop Test /CR0 Approved Connents (times are clock time) 82 '12/31 1225 1325 Phillippe Adams No water or hydrogen additions; 1245 1345 Phillippe however, MUT level should have decreased'-3" during test. Instead level increased +2.2". Note:

Unstable LT-1 used as input to computer for test. Significant-difference in readings between LT-1 and LT-2. Net effect approximately

+185 gal, accounting for the large negative leak rate of -2.4 gpm.

83 01/02 0103 0203 Frederick Scheimann No water or hydrogen additions; 0118 0218 Faust however, MUT level should have

., decreased -3" during test. Instead

, level change was approximately 0".

Note: Unstable LT-1 used as input to computer for test. Significant difference in readings between LT-1 and LT-2. I!et effect approximately

+94 gal, accounting for the negative leak rate of .45 gpm.

84 01/04 0302 0402 Frederick Scheimann No water or hydrogen additions; 0320 0420 Faust however, MUT level should have decreased -3" during test. Instead level decreased only .2". Note Unstable LT-1 used as input to computer for test. Net effect approximately +75 gal.

01/04 1215 1315 Olson* or Miller

• MUT strip chart switched to LT-1 for 1245 1345 Wright

• 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> period between 1215 and 1315.

Coleman No surveillance test sheet retained.

Assume results discarded.

• Assumptions based upon shift composition.

01/04 1520 1745 Congdon* or Adams

• MUT strip chart switched to LT-1 for 1550 1815 Phillippe* 2 two-hour periods as noted. It Cooper appears 2 leak rate tests may have been run on LT-2 (stable) then one 1910 2120 testonLT-1(unstable)andthentwo 1940 2150 moreonLT-2(stable). No surveillance test sheets retained.

Assume results discarded.

• Assumptions based upon shift composition..

85 01/05 0254 0354 Faust Scheimann No water or hydrogen additions; 0309 0409 Frederick however, MUT level should have decreased -3" during test. Instead level decreased only -2.1". Note:

Unstable LT-1 used as input to computer for test. Net effect approximately +40 gal.

1

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 12 Test No. Date Start Stop Test /CR0 Approved Consnents (times are clock time) 86 01/05 1723 1823 Congdon Adams 'No water or hydrogen additions.

1738 1838 Congdon Note: Unstable LT-1 used as input to computer for test. Significant difference in readings between LT-1 and LT-2. Net effect approximately

+74 gal.

87? 01/06 0247 0347 Kidwell Conaway Possible water or hydrogen addition 0302 0402 Kidwell at 0335. Note: Unstable LT-1 used as input to computer for test. The

! combination of water / hydrogen addition and the use of LT-1 provide i

a net effect of +165 gal. accounting for the low gross leak rate and

_ negative unidentified leak fate of

.50 gpm.

i --

01/06 0510 0610 Illjes* or Conaway* MUT strip chart switched to LT-1 for 0535 0635 Mell* 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> period between 0510 and 0610.

Kidwell No surveillance test sheet retained.

Assume results discarded.

• Assumptions based upon shift J composition.

88 01/06 1029 1129 Coleman Miller No apparent water or hydrogen addi-1044 1144 Olson tions. Note: Stable LT-2 used input to computer for test.

I 89 01/06 1920 2020 Phillippe Adams No apparent water or hydrogen i 1935 2035 Phillippe additions. Note: Stable LT-2 used as input to computer for test.

90 01/07 0341 0441 Illjes Conaway No water or hydrogen additons;

0356 0456 Mell however, MUT level should have decreased -3" during test. Instead i

, level decreased only .89". Note:

Unstable LT-1 used as input to computer for test. Significant difference in readings between LT-1 and LT-2. Net effect approximately 1

+84 gal. accounting for the low gross leak rate and negative unidentified i leak rate of .22 gpm. l 01/07 0835 1445 01:;on* or Miller Between 0835/0935,1035/1135 0850 1500 Wright

• and 1345/1445 the MUT strip chart Coleman was switched from LT-2 to LT-1. No surveillance test sheets retained.

Assume results discarded.

, *Assumptions based upon shift composition.

]

I

1 l

I

i

~

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 13  !

l Test No. Date Start Stop Test /CR0 Approved Coments (times are clock time) 91 01/08 0321 0421 Mell Conaway No water or hydrogen additions; 0336 0436 111jes however, MUT level should have decreased -3" during test. Instead level increased +1.2". Note:

Unstable LT-1 used as input to computer for test. Significant difference in readings between LT-1 and LT-2. Net effect +108 gal, accounting for negative values for gross and unidentified leak rates.

Significant difference between LT-1 and LT-2.

92? 01/10 0148 0248 Mell Conaway Possible water addition of 150 gal.

0203 0303 Mell at 0218. No log entry. Not included in calculation. Note: Unstable LT-1 used as input to computer for test.

Significant difference between LT-1 and LT-2.

NOTE: LT-1 REPAIRED.(CANNOT DETERMINE WHETHER LT-1 OR LT-2 USED AS COMPUTER INPUT FOR REMAINDER OF j TESTS.)

a

93 01/11 2153 2253 Coleman Miller Water addition of 240 gal, at 2230.

j 2208 2308 Olson Log entry of 200 gal. 200 gal.

included in calculation. Water added j 23 minutes prior to end of test.

94 01/13 0937 1037 Hartman Hoyt Water addition of 117 gal. at 0950.

0952 1052 Booher Logged in CR0 Log at 1000. Not included in calculation.

01/13 1330* 1430* Hartman** Hoyt 1430 CR0 Log entry: "Pertermed Booher another good leak rate." No surveillance sheet on file.

• Based upon log entry. ** Based upon shift composition.

1 95? 01/15 0024 0124 Cooper Adams No apparent water or hydrogen 0044 0144 Cooper additions; however, MUT level should decrease -4" during test. Actual decrease is only -2".

96 01/30 2206 2306 McGovern* Hoyt No water or hydrogen additions.

2206 2306 or Hemila* Test was started on shift F and

& Blessing completed on shift E. Difference Gemer** & between MUT LTs is approximately 6".

l Booher** , Test signed by Blessing. *Assump-tions based upon shift composition. ,

i **Booher & Germer both signed CR0 log '

during test period.

l l

L_ _ .- . _ _ _ . - - - - - _ _ _ . - - - _ - - - - - . - _

. l TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 14 Test No. Date Start Stop Test /CR0 Approved Comments (times are clock time) 97? 02/02 0055 0155 Blessing Hoyt Water addition of 300 gal. at 0100 is 9225 0325 Bocher logged in CR0 Log. Water addition is not evident on MUT level trace. It is possible this was a feed and bleed operation. No water additions are included in the calculation.

98 02/02 1431 1531 111jes* or Scheimann Possible water addition of 110 1601 1701 Kidwell* & gal. at 1525. MUT level should have Faust decreased -3.5". Actual level Mell & decreased .2". Test started on Frederick shift B. Frederick relieved Mell at 1500.

• Assumptions based upon shift composition.

~ 99 02/03 0517 0617 McGovern Guthrie No water or hydrogen additions.

0647 0747 Hemila 100 02/03 1032 1132 Mell Conaway No water or hydrogen additions.

1202 1302 Kidwell 101 02/03 2349 0049* Hemmila Guthrie No water or hydrogen additions.

0119* 0219* Genner *02/04/79.

102 02/04 1445 1545 Faust Scheimann No water or hydrogen additions.

1615 1715 Frederick .

103 02/05 0312 0412 Genner Guthrie No water or hydrogen additions.

0442 0542 McGovern 104 02/05 0835 0935 Kidwell Hoyt No water or hydrogen additions.

1005 1105 Mell 105 02/05 1837 1937 Frederick Scheimann No water or hydrogen additions.

2007 2107 Faust 106 02/06 0025 0125 McGovern Guthrie Invalid test. At 0045 MUT 0155 0255 Hemila level drops -270 gals. then slowly increased +210 gal, during the remainder of the test. Change is either caused by bleed and feed or by control rod movement (rod control in manual.)

-y -, -. -.,,.w,, -

.r --

I dd TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 15 Test No. Date Start Stop Test /CR0 Approved Comments (times are clock time) 107 02/07 0150 0250 Hemmila Guthrie No water or hydrogen additions.

6320 0420 Germer 108 02/07 1352 1452 Cooper

• or Adams ** Test invalid. Unstable plant 1522 1622 Phillippe* Scheimann** conditions caused by power level Frederick **& changes or feed and bleed operation.

Congdon** Test performed on C shift, approved by A shift.

• Assumptions.

- ** Frederick signed test /Scheimann approved test. Congdon and Adams on watch during test.

109 02/08 0100 0200 Germer Guthrie No water or hydrogen additions.

0230 0330 McGovern ,

110? 02/08 2050 2150 Mell Conaway Possible feed and bleed operation 2220 2320 Kidwell (100 gal.) near start of leak rate test. May have had small effect on test results.

111 02/09 0220 0320 McGovern Adams Feed and bleed operation started 0350 0450 Hemmila before test begun; however, more water appears to have been bleed from j the RCS, contributing to a higher leak rate than would otherwise be the case. Test is invalid.

112 02/10 0841 0941 Congdon Adams Feed and bleed operation 100 gal.

1011 1111 Congdon between 0900 and 0910. Amount appears equal and may not have affected test results. l l

113 02/11 0241 0341 Frederick Scheimann No water or hydrogen additions. 1 0411 0511 Faust 114 02/11 1808 1908 Mell Conaway Water addition of 300 gal. at 1938 2038 Kidwell 1825 logged in CR0 Log. Actual amount appears to be 180 gal. from MUT strip chart. Addition is not included in calculation. RCDT was pumped down during test. Not included in calculation.

115 02/12 2120 2220 Kidwell Conaway Water addition of 150 gal. at 2125.

2250 2350 Illjes Not logged in CR0 Log or included in l calculation. RCDT pumped down during test. Not included in calculation.

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 16 Test No. Date Start Stop Test /CR0 Approved Coments (tines are clock time) 116 02/13 1236 1336 Hutchinson* Hutchinson* No water or hydrogen additions.

1406 1506 McGovern Guthrie* *R.S. Hutchinson was off-shift SR0 on THI-1/2 assigned to Shift Supervisor Training. Guthrie was SF but did not sign test.

117 02/13 1842 1942 _Illjes Conaway No water or hydrogen additions.

2012 2112 Kidwell 118 02/14 0530 0630 Faust Scheimann Water addition of 210 gal. at 0600.

0700 0800 Frederick Not logged in CR0 Log but 206 gal, are included in calculation.

119 02/14 2046 2146 Kidwell Conaway Hydrogen addition at 2135 (2" or s 2216 2316 Olson*/ 60 gal.effect). Not logged in Kidwell* CR0 Log but logged in A0's Log.

• Kidwell relieved Olson at 2130 thus hydrogen was added on Kidwell's watch (ShiftB).

120 02/15 2026 2126 Cooper Adams Hydrogen addition 2100 (2" or 60 2156 2256 Congdon gal effect). Logged in CR0 Log.

" Pressurized MUT" written on strip chart (Adams' handwriting according toCongdon).

121? 02/16 0253 0353 McGovern Guthrie Hydrogen or water addition at 0325.

0423 0523 Hemila (2"or60 gal,effect). MUT shows same characteristics as test 120. No CR0 Log entry. Not included in calculation.

122 02/16 1203 1303 Olson Miller Water addition of 210 gal. at 1250.

1333 1433 Coleman Logged in CR0 Log and included in calculation as 204 gal.

(Reconstruction shows 234 gal, effect) Beginning of shift D pattern of adding water near end of test - (13 min. before end of test).

123 02/17 0411 0511 111jes Conaway Hydrogen addition 0500 (2" or 60 gal.

0541 0641 Illjes change, only 1" or 30 gal.

effective). Not logged in CR0 Log but is logged in-A0 s Log.

124 02/19 0001 0101 Kidwell Conaway Hydrogen addition 0100 (2" or 60 gal.

0131 0231 111jes but only 1" or 30 gal. effective during test). No log entry.

125 02/19 0136 0236 Kidwell Conaway Hydrogen or water addition at 0200 0306 0406 Illjes (1" or 30 gal. effect). No log entry. Not included in calculation.

t 1 ,

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 17 Test No. Date Start Stop' Test /CR0 Aporoved Consnents (times are clock time) 126 02/19 2128 2228 Cooper ' Adams Hydrogen or water addition at 2145 2258 2358 Congdon (2" or 60 gal, effect). No log entry. Not included in calculation.

127- 02/21 0836 0936 Blessing- Scheimann No water or hydrogen additions.

1006 1106 Frederick 128 02/23 1107 1207 Hartman Hoyt Feed and bleed operation of 150 gal.

1237 1337 Booher . between 1130 and 1155. Water addition of 150 gal. logged in CR0 Log at 1135. An additional 150 gal.

was added between 1155 and 1200 as

~ part of a second feed and bleed. Net

-' effect of both feed and bleed s , operations +160 gal. accounting for low gross leak rate. Not included in calculation.

02/25 ---- ----

Hartman*/ Hoyt* CR0 Log entry for 1334 " Late entry RC Booher* leak rate 2301-301 Long Fonn Calcu-Blessing lation perfonned sat. Unidentified

.87 gpm."

• Assumptions based upon shift composition.

129 02/25 2002 2102 . Col'eman Miller Water addition of 200 gal, at 2117 2217 Olson ,

2058. CR0 Log states 150 gal.

added at 2058. Calculation includes

- 150 gal. (50 gal difference). Note:

^

Water added 4 mins. prior to end of test.

130 ~02/26 0038 0138 Congdon Adams Water addition or hydrogen addition 0143 ,0243 Cooper at 0045 (3" or 90 gal. effect). No log entry. Not included in

calculation.

131 02/26 1839 1939 Coleman Miller Water addition of 180 gal. at 1922.

1954 2054 Olson ,_ CR0 Log states 150 gal, added at 1935. Calculation includes 154 gal.

(26 gal, difference). Note: Water was added 17 mins. prior to end of

., j test.

Miller 132 02/27 2150 '.2250- Coleman Test begun during feed and bleed 2255 2355 Colernan operation approximately 122 gal.

bl ed. Water additions of 60 gal. and i 150 gal. at 2205 and 2230. CR0 Log states 162 gal. at 2245. Calculation includes 162 gal. (48 gal.

' difference). However, since water bled was not included in calculations, the overall leak rate was lower than calculated.

l

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

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 18 Test No. Date Start Stop Test /CR0 Approved Consnents (times are clock time) 133 02/28 1909 2009 Olson Miller Water addition of 150 gal. at 2005.

2159 2259 Wright CR0 Log states 100 gal. added at 2005. Calculation includes 100 gal.

(50 gal, difference). Note: Water added 4 mins. prior to end of test.

134 03/01 0041 0141 Congdon Adams Hydrogen addition at 0130 (effect 0435 0535 Congdon on test results uncertain). CR0 Log states hydrogen addition at 0130.

Note: hydrogen addition 11 mins.

prior to end of test.

135? 03/02 0146 0246 Cooper Adams Possible jogged water addition of 0701 0801 Cooper 30-60 gal, between 0152 and 0215.

' 1367 03/02 1935 2035 Blessing Hoyt Possible jogged water addition 0255 0355 Blessing or hydrogen addition at 2025 (3" or 90 gal effect). No log entry. Not included in calculation. Note:

addition made 10 min. prior to end of test.

137 03/03 0238 0338 Olson Hitz* Water addition of 200 gal. at 0335.

0238 0338 Olson CR0 Log states 150 gal. added at 0335. Calculation includes 152 gal.

(48 gal. difference). Note: water added 3 min, prior to end of test.

• Hitz was shift D SS.

138 03/04 0142 0242 Coleman Miller Water addition of 270 gal at 0240.

0212 0312 Coleman CR0 Log states 238 gal. added at 0240. Calculation includes 238 gal.

(32 gal. difference). Note: water added 2 min. prior to end of test.

139 03/05 0320 0420 Olson Miller Water additions of 100 gal, and 210 0350 0450 Olson gal. 0405 and 0410. CR0 log states 128 gal. added at 0410. Calculation includes 128 gal (182 gal.

difference). Note: water added 10 min. prior to end of test.

140 03/06 0321 0421 Wright Miller Water addition of 210 gal. at 0420.

0351 0451 Olson CR0 Log states 180 gal. added at 0420. Calculation includes 180 gal.

(30 gal difference). Note: water added 1 min. prior to end of test.

TMI-2 Leak Rate Tests (09/30/78-03/28/79) Page 19 Test No. Date Start Stop Test /CR0 Approved Comments (times are clock time) 141 03/08 0306 0406 Olson Miller Water addition of 250 gal. at 0403.

6306 0406 Wright CR0 Log states 228 gal. added at 0403. Calculation includes 228 gal.

(22 gal. difference). Test was begun during the later half of a feed and bleed operation. Note: water 'added 3 min. prior to end of test.

142 03/09 0323 0423 Wright Hitz* Water additions of 180 gal. and 60 0323 0423 Olson gal. at 0425 and 0420. CR0 Log states 180 gal. added at 0415.

Calculation includes 181 gal. (59 pl. difference). Note: water added 8 min. and 3 min. prior to end of test.

• Shift Supervisor-Shift D.

143 03/10 0351 0451 Hartman Hoyt Water addition (jogged) of 80 gal.

0351 0451 Booher starting at 0440. No log entry. Not included in calculation. Note:

addition started 11 min. prior to end of test.

144 03/12 0132 0232 Booher Miller */ Water addition (jogged) of 100 gal.

0132 0232 Blessing Hoyt* starting at 0150. No log entry. Not included in calculation.

• Miller signed test sheet; however, test done on Hoyt's watch and logged in SF log by Hoyt.

145 03/13 0200 0300 Hartman Hoyt Water additions (jogged) of 30 and 60 0140 0240 Booher gal. at 0230 and 0250. No log entries. Not included in calculation. Note: last water addition started 10 mins. prior to end of test.

146 03/13 1105 1205 Olson Miller Water addition of 330 gal. at 1200.

1050 1150 Coleman CR0 Log states 305 gal. added at 1200. Calculation includes 305 gal.

(25 gal. difference). Note: water added 5 min. prior to end of test.

J i

i

. , _ , . . .,___,_._,._.____..._.m. . , . . _ _ _ _ _ . _ . - _ , . - . . _ . . _ . . . _ _ , - _ _ _ _ _ . _ _ _ . . , . . _ _ _ _ _ _ _ _ , _ _ _ _ , ~ , _ _ _ . . , - . , _ , . . , . _ , _ _ , _ , , _ _ _ _ _ _ _ . _ .

TMI-2 LEAK RATE TESTS Page 2 (09/30/78-03/28/79)

Test No. Date Start Stop Test /CR0 Approved Coments (times are clock time) 147 03/14 1205 1305 111jes Conaway Water additions of 390 and 210 gal.

1200 1300 Kidwell at 1235 and 1250. CR0 Log states 300 I and 200 gal. added at 1230 and 1250.

Calculation includes 500 gal. (100 l

- gal difference). Note: last water l addition 15 min, prior to end of test. The 500 gal addition was 1 included in the calculation in the l wrong step, i.e., water deletions not I water additions. l 1

148 03/15 0450 0550 Booher Smith

• Water addition (jogged) of 100 gal. ,

, 0445 0545 Blessing Hoyt* starting at 0525. No log engry. Not '

included in calculation.

• Smith (SS) signed approval space on test sheet; however, Hoyt (SF) recorded results in the Shift Foreman's Log.

149 03/16 2009 2109 Faust Scheimann Water additions of 120 gal. and 120 1944 2044 Frederick gal. at 2045 and 2050. CR0 Log states 200 gal. addition at 2050.

Calculation includes 204 gal. (36 gal, difference).

l 150 03/17 0248 0348 McGovern Guthrie Water addition of 240 gal, at 0323.

0223 0323 Hemila CR0 Log states 200 gal added at 0323. Calculation includes 207 gal.

(33 gal. difference).

03/18 0847 0947 111jes* Conaway* Strip chart marked at 0847., "84" LR I 0822 0922 or Mell* 0 0847. Strip chart marked at 0947, l Kidwell "LR out 71." Possible hydrogen  ;

addition and feed and bleed operation during test. No log entry. No test results retained.

• Assumptions based upon shift composition.

151 03/19 0058 0158 McGovern Guthrie Water addition of 220 gal. at 0100.

0053 0153 Hemila Only 90 gal. was added during test.

i Water addition of 240 gal, at 0157.

CR0 Log entries inicude 200 gal at 0100 and 200 gals at 0157.

Calculation includes 200 gal. (130 gal, difference). Note: last water addition 1 min. prior to end of test.

152 03/21 0114 0214 Hemila Guthrie Water addition of 210 gal at 0126.

0114 0214 Genner ~CR0 Log states 200 gal, added at 0126. Calculation includes 207 gal.

Hydrogen addition at 0205. CR0 Log states h l Effect2{drogenadditionat0205.

or 60 gal. Note: hydrogen added 9 min. prior to end of test.

TMI-2 LEAK RATE TESTS Page 2 (09/30/78-03/28/79)

Test No. Date Start Stop Test /CR0 Approved Consnents (times are clock time) 153 03/22 0300 0400 McGovern Guthrie Water addition of 60 gal. at 0311.

0255 0355 Hemila No log entry. Water addition of 240 gal. at 0353. CR0 Log states 200 added at 0353. Calculation includes 200 gal. (100 gal. difference).

Note: last water addition 7 min.

prior to end of test.

154 03/24 0540 0640 Congdon*/ Adams **/ Water additions of 100 gal. at 0600 0538 0638 Faust

• Scheimann** and 0630. CR0 Log states 100 gal.

Frederick at 0600 and 0630. Calculation includes 206 gal. (difference of +6 gal). Note: last water addition made 10 min, prior to end of test.

• Congdon signed test; however, test was done on Shift A (assume Faust ran test). ** Adams approved test and logged in SF log; however, test was done on Shift A. {

155 03/25 0525 0625 Frederick Scheimann Water addition of 300 gal, at 0608.

0523 0623 Faust CR0 Log states 300 gal. added at 0608. Calculation includes 301 gal.

(difference of +1 gal.)

156 03/28 0134 0234 Faust Miller */ Water addition of 320 gal, at 0145.

0134 0234 Frederick Scheimann* CR0 Log states 303 gal. added at 0145. Calculation includes 303 gal.

(17 gal, difference).

• Miller signed test; however, test was done on Shift A. -

F R RESULTS OF NRR'S INVESTIGATION AND EVALUATION OF TEN LICENSED OPERATORS INVOLVED IN TMI-2 PREACCIDENT LEAK RATE TESTING IRREGULARITIES .

, O U.S. Nuclear Regulatory Commission Office of Nuclear Reactor Regu'lation f *"*%,

e S 3 e

\...../

O- .

O .

t RESULTS OF NRR'S INVESTIGATION AND EVALUATION OF TEN LICENSED OPERATORS INVOLVED IN TMI-2 PREACCIDENT LEAK RATE TESTING IRREGULARITIES U.S. Nuclear Regulatory Commission .

Office of Nuclear Reactor Regulation -

pancoq f ,U/,$

G a _.