ML20083C837

Supplemental Testimony of G Bockhold Re Intervenor Exhibit 18.W/Certificate of Svc & Svc List

ML20083C837
Person / Time
Site:	Vogtle
Issue date:	05/04/1995
From:	Bockhold G AFFILIATION NOT ASSIGNED
To:
Shared Package
ML20083C835	List: ML20083C832 ML20083C837
References
93-671-01-OLA-3, 93-671-1-OLA-3, OLA-3, NUDOCS 9505180491
Download: ML20083C837 (150)

Text

{{#Wiki_filter:c 00CKETED USNRC UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION gg gy.$ gj g7 Docket Nos{ OFFICE 0F SECRETAR In the Matter of ) ) 50-495ANCM-3 GEORGIA POWER COMPANY, ) et al. ) Re: License Amendment J ) (Transfer to Southern ) Nuclear) (Vogtle Electric Generating ) Plant, Units 1 and 2) ) ) ASLBP No. 93-671-01-OLA-3 SUPPLEMENTAL TESTIMONY OF GEORGE BOCKHOLD, JR. CONCERNING INTERVENOR'8 EINIBIT 18 t 9505180491 950504 PDR ADOCK 05000424 T PDR

o TESTIMONY OF GEORGE BOCKHOLD, JR. i 1 Q. WHAT IS THE PURPOSE OF THIS SUPPLEMENTAL TESTIMONY? 2 A. The ' purpose is to supplement my testimony at an evering 3 session of hearings on April 20, 1995 related to statements 4 made by Georgia Power concerning the Vogtle diesel generators 5 after the March, ' 1990, Site Area Emergency. Judge Bloch ) t 6 permitted me to supplement my testimony within two weeks t 7 concerning the "As Found" temperature settings of Calcon 8 sensors identified in Intervenor's Exhibit 18. Tr. 3904. 9 Q. HAVE YOU IDENTIFIED ADDITIONAL INFORMATION CONCERNING THE "AS o 10 FOUND" TEMPERATURE OF THE SWITCHES (SENSORS) IDENTIFIED ON 11 INTERVENOR EXHIBIT 18 WHICH WAS IN YOUR POSSESSION AT THE TIME 12 OF GEORGIA POWER'S APRIL 9, 1990 PRESENTATION TO THE NRC? 13 A. Yes, additional information was located in my files from 14 April, 1990. 15 Q. PLEASE IDENTIFY AND EXPLAIN THE INFORMATION. 16 A. On or about April 3,

1990, Mr.

Mark

Briney, an IEC P

17 Superintendent, provided me information in memorandum form 18 which addressed the Calcon sensors at diesel generator 19 locations 1TSH-19111 and -19112. These locations correspond 20 to the sensors identified in the two right-hand columns of { 21 Intervenor Exhibit 18. A copy of the April 3rd memorandum is 22 attached hereto as Georgia Power Exhibit II-64 (Bockhold L). {.

l' As is explained in the memorandum, the sensors "11" and "12" '2 (corresponding to ITSH-19111 and -19112, respectively) were on 3 the Unit 1, train "A" diesel at the time of the March 20, 1990 4 Site Area Emergency. After diesel testing on March 30th 5 during which the diesel tripped, the sensors were replaced on 6 March 31, 1990. The "As Found" data was determined before the 7 testing on March 30th by Georgia Power technicians using 8 existina calibration methods. 9 Mr. Briney's memorandum also outlines the history of the 10 sensors. After recalibration and reinstallation or, the A 11 diesel, the switch in position 1TS-19111 was observed t. ripped 12 during logic testing on March 30, 1990 and was then 13 depressurized and repressurized (i.e., " popped") to seat it. 14 The switch located at 1TSH-19112 was also recalibrated on 15 March 30th and observed with leakage (i.e., venting of 16 pressurized control air) at temperatures more than 20 degrees 17 below its set point. The "As Found" data on the 1TSH-19112 18 switch was orally given to the IIT on April 2, 1990, as well i 19 as the fact that in the logic testing one sensor was found 20 tripped and one was venting. IIT Document 168-2, April 2, 21 1990, pages 46-47. A copy of this IIT document is attached 22 hereto as Georgia Power Exhibit II 65 (Bockhold M). 23 The NRC IIT also discussed the March 30 A diesel testing and 24 the performance of the sensors in a transcribed meeting on 1 March 31, 1990 with Georgia Power engineers and consultants 2 which I did not attend (IIT Documents 150 and 116) and was 3 provided Mr. Briney's memorandum (IIT Document 152). The 4 recalibration of the sensors on March 30, the pulsing of air 5 to reset 1TSH-19111, the venting of 1TSH-19112, the quarantine 6 of the sensors, and the installation of new sensors are s 7 discussed in NUREG-1410, Appendix J, page J-14. l 8 In addition to Mr. Briney's memorandum, summary documentation 9 concerning these two Calcon

sensors, among

others, was 10 provided to me prior to the April 9, 1990 presentation to the 11 NRC. Georgia Power Exhibit II-66 (Bockhold N), consisting of 12 four pages, is a copy of this summary information.

13 Q. WERE THESE TWO CALCON TEMPERATURE SENSORS EVENTUALLY TESTED? 14 A. Yes, as reflected in NUREG 1410, Appendix J, pages J-26 and J-15 27, these quarantined switches were examined and tested for 16 "As Received" set points by Wyle Laboratories. The "As 17 Received" setpoints were approximately 189 and 162.4 degrees 18 Fahrenheit for 1TSH-19111 and 1TSH-19112, respectively. Egg, 19 Wyle Test Report No. 17133-1, page II-17, Georgia Power 20 Exhibit II-63 (Ward Exh. E). However, I did not have this 21 test information at the time of the April 9th presentation. 1 Q. . PRIOR TO APRIL 9, 1990, DID-'YOU BELIEVE SENSOR CALIBRATION. 2 PROBLEMS CONTRIBUTED TO THE UNIT 1,. "A" TRAIN DIESEL GENERATOR 3 FAILURE ON MARCH 20, 19907 ) .4 A.- 'Yes, and I told the IIT team exactly that. On April 4,

1990, i

5 I told Mr. Al Chaffee that I believed that we had to have some ~ 6 sensor calibration problems to produce the failure. IIT' 7. Document 168-1, April 4, 1990, page 22, Georgia Power Exhibit l 8 II-67 (Bockhold O). ) i ) I 9 I had information that indicated a variable response of Calcon 10 temperature sensors to. different heat-up rates used during l 11 calibration, as reflected in a document entitled "Calcon i 12 Temperature Response Test on April 4,.1990." A copy of this j 13 document, which I apparently received on April 5, is attached f l 14 hereto as GPC Exhibit II-68 (Bockhold P), consisting of two 15 pages. Mr. Chaffee was aware of this variable response from j i 1G statements by the Calcon vendor; on April 4,1990, Mr. Chaffee.. [ 17 referred to the depression of setpoints associated with 18 increased heat-up rates used during calibration. IIT Document 19 168-1, pages 13-14. I agreed with Mr. Chaffee that there 20 could have been an accumulation of effects that caused the 21 trip when I said on April.4th: e i 22 Yes, it went from sensor calibration. problem and then 23 with this phenomenon and just hitting it at the right l 24 time, and with one intermittent -- we know we had one l i r l l

.~ i 1~ intermittent problem. We~ had one sensor that had an 2 . intermittent problem associated with it. 3 We know that for a' fact, so you know, one sensor _having i 4 a calibration problem, and the accumulation of all of ~ 5 'these things could have produced the trip. It could be 6 the explanation of why there's an'80 second delay on'the 7 first one [ trip of the diesel on March 20] and'a 70 8 second delay on the second one [ trip]. In fact, that'[s] 9 what we believe right now. (IIT Document 168-1, pages 10' 22-23) i i I 7 6 i I o i l. ..?

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of ) Docket Nos. SC-424-OLA-3 ) 50-425-OLA-3 GEORGIA POWER COMPANY, ) et al. ) Re: License Amendment ) (Transfer to Southern ) Nuclear) (Vogtle Electric Generating ) Plant, Units 1 and 2) ) ) ASLBP No. 93-671-01-OLA-3 CERTIFICATE OF SERVICE This is to certify that copies of the within and foregoing " Supplemental Testimony of George

Bockhold, Jr.

Concerning Intervenor's Exhibit 18" were served on all those listed on the attached service list by express mail delivery on this the 4th day of May, 1995. / n LamberskV OUTMAN SANDERS Suite 5200 600 Peachtree Street, N.E. Atlanta, Georgia 30308-2216 (404) 885-3360 i I t

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION ATOMIC SAFETY AND LICENSING BOARD In the Matter of GEORGIA POWER COMPANY, ) Docket Nos. 50-424-OLA-3 at al. ) 50-425-OLA-3 ) (Vogtle Electric Generating Plant, ) Re: License Amendment Units 1 and 2) ) (Transfer to Southern ) Nuclear) ) ) ASLBP No. 93-671-01-OLA-3 i SERVICE LIST I Administrative Judge Stewart D. Ebneter Peter B. Bloch, Chairman Regional Administrator Atomic Safety and Licensing Board USNRC, Region II U.S. Nuclear Regulatory Commission Suite 2900 Washington, D.C. 20555 101 Marietta Street, N.W. Atlanta, GA 30303 Administrative Judge James H. Carpenter Office of the Secretary Atomic Safety and Licensing Board U.S. Nuclear Regulatory 933 Green Point Drive Commission Oyster Point Washington, D.C. 20555 1 Sunset Beach, NC 28468 Attn: Docketing and Services Branch Administrative Judge Charles Barth, Esq. James H. Carpenter Mitzi Young, Esq. Atomic Safety and Licensing Board Office of General Counsel U.S. Nuclear Regulatory Commission U.S. Nuclear Regulatory j Washington, D.C. 20555 One White Flint North l Stop 15B18 Rockville, MD 20555 Administrative Judge Thomas D. Murphy Director, Environmental 3 Atomic Safety and Licensing Board Protection Division i U.S. Nuclear Regulatory Commission Department of Natural Washington, D.C. 20555 Resources 205 Butler Street Michael D. Kohn, Esq. Suite 1252 Kohn, Kohn & Colapinto, P.C. Atlanta, Georgia 30334 517 Florida Avenue, N.W. Washington, D.C 20001 Office of Commission Appellate Adjudication One White Flint North 11555 Rockville Pike i Rockville, MD 20852 i )

GPC EXHIBTT 64 BOCKHOIE L int:r:ffica C:rrcsp:nd:nce Georgia Power al -{ i DATE: April 3,1990 RE: Calcon Temperature Switches FROM: M. S. Briney TO: G. Bockhold, Jr. ] The following 'is the information. the IIT has requested on the first two Calcon i temperature switches we wish to test. 1 TSH-19111 D/G 1A Jacket Water Main Header Outlet Temp Hi HISTORY: The original calibration test was performed satisfactorily on 3/2/90. As Found data was 206.2*F to trip. As left data was 199.1'F. On 3/30 it was calibrated again.(as well as 19110 and 19112). i As Found data was 198.56*F. to trip. During subsequent logic testing it was found to be tripped. It was " popped" by pinching off the air line, allowing the switch to depressurize, and then quickly repressurizing the switch. Since that time, subsequent engine testing showed it operating correctly. However, it was removed on 3/31/90 and replaced with a new switch. Since its removal, it has remained in storage and has not been tested. 1 TSH-19112 D/G 1A Jacket Water Main Header Outlet Temp Hi l HISTORY: The original calibration test was performed satisfactorily on 3/1/90. As Found data was 210.4*F. As Left data was 203.1*F. On 3/30/90 f t was calibrated again. As Found data was 186.2*F to i trip. It was noted at that time that a small amount of leakage was -{ present even with temperature >20 'F below setpoint. This did 'not appear to affect the calibration. The Cooper vendor verified this would not cause a problem with the D/G control system and the switch was put back on the 1A D/G. Subsequent logic testing showed. this l switch to be leaking still. It was removed and replaced with a new i switch on 3/31/90. Since its removal, it has remained in-storage and has not been tested. Calcon's proposed test for both switches: With both switches at ambient temperature (out of the temperature bath), pressure test with 60 psig air supplied through a 0.028" orifice. If one or both units leak at room temperature, we will take the top section of the switch apart to access the valve area. We will then inspect the valve and

seat, clean as

required, reassemble, and check the trip setting / response of the switches and adjust as necessary.

If no leaks are found at room temperature, we will heat up the sensor as we do during a normal calibration test in the bath to check its trip setting and overall response. If the switches do not respond correctly, we will disassemble and inspect the switch to attempt to identify the root cause of the malfunction. MSB/jbc xc: H. M. Handfinger 92 PROJECT 006216

a-g c5-W blG M GPC EXHIBIT 63 y OFFICIAL TRANSCRIPT OF PROCEEDINGS **" "" i geng Nuclear Regulatory Commission Diesel Generator Meeting

Title:

setween NRC.na ceorgia rever Docket No. ( Waynesboro, Georgia Monday, April 2, 1990 1 - 60 p i 1 ANN RILEY& ASSOCIATES, LTD. 1612 K St.N.W. Suite 300 Washington, D.C. 20006 (202) 295-3950 92 PROJECT 03522e

1 1 U. S. NUCLEAR REGUIATORY COMMISSION g i 2 e 3 i 4 ) 5 DIESEL GENERATOR MEETING ) 6 BETWEEN NRC AND GEORGIA POWER ) 7 ) 8 9 10 Main Conference Room 11 Administration Building j 12 Vogtle Electric Generating Plant i 13 Waynesboro, Georgia 14 15 Monday, April 2, 1990 16 17 18 The meeting commenced at 8:45 a.m. 19 20 21 l 22 e i 23 i 24 l 25 92 PROJECT 035219

i 2 i 1 ATTENDEES: 2 3 \\ AL CHAFFEE TOM WEBB i \\ 4 WILLIAM LAZARUS KENNETH BURR i 5 dE ALAN HERDP RAY HOWARD i 6 g ff KEN BROCKMAN KENNETH HoutES 7 RICK KENDALL JIM ROBERTS 8 PAUL KOCHERY JOSEPH D'AMICO 9 GEORGE BOCKHOLD GEORGE FREDERICK-10 ARREN LYON PHIL CUPP 11 PAUL DIETZ MARK BRINEY ~ \\. A. TRAGER 12 E MICHAEL LACKEY i 13 WILLIAM JONES VINCE STEWART 14 HERB BEACHER ERNEST TOUPIN 15 ROBERT JOHNSTON ALLEN MOSBAUGH 16 DONALD PESOUT MICHAEL HORTON l 17 SHELDON OWYOUNG JIMMY PAUL CASH 18 KENNETH STOKES INDIRA KOCHERY 19 CHARLES COURSEY i 20 l 21 l 22 t 23 j 24 25 l 92 PROJECT ~ 3 035220 i i I

i 3 1 PROCEEDINGS r r 2 MR. CHAFFEE: It's April 2 and it's 8:45 in the 3' morning and we're here -- this is the IIT at Vogtle and 4 we're here to go over the information that exists at this i 5 time on the problem with the diesel generator relative to 6 its tripping and what the troubleshooting has found and what 7 the machinery history has reviewed that is relevant to what 8 transpired in the event with the diesel generator tripping. 9 I don't know how we should do this. Do you just 10 want to have somebody kind of run through what you guys' 11 perception is on it? 12 MR. BOCKHOLD: You know, we, last night provided 1 ~ -1 13 you a root cause that the critique team is -- you know, the i 14 diesel team came up with this root cause and some key 15 lessons learned and recommandations and a statement about 16 diesel reliability and ouraanalysis of it. 17 I think the key lesson learned -- and we have the 18 Calcon vendor on the way and he's supposed to be here today 19 -- is to go through the calibration procedures so we can 20 determine what is the best methodology for calibration. We 21 have some ideas, I'm sure he'll have some ideas, and look at 22 switch reliability. That's ongoing. { v 23 I don't know if you want me to read this for the 24 record's sake or that kind of stuff or you ask specific 25 questions about the history. 92 PROJECT i 035221

I 4 1 MR. CHAFFEE: Why don't we just ask some questions 2 then. I guess one question is in one of your lessons i' 3' learned you say that "The operator training should be 4 revised to ensure that operators realize that an emergency 5 reset will overrun the high jacket water trip." I'm not 6 sure I understand what that means. Does that mean that you i 7 can actually override that particular trip feature by i 8 emergency resetting? 9 MR. BOCKHOLD: Yes, I believe that's correct. And l 10 Sheldon Owyoung is probably -- you know, the vendor expert 11 that identified that to us just recently. l 12 MR. OWYOUNG: If you have a jacket water (- 13 temperature trip, the engine will shut down on two out of l 14 three. One you have a shutdown, it'll give you emergency 15 shutdown, the controls are set up such that when you reset 16 the emergency shutdown, the unit will start up again and 17 bypass that shutdown. That way, it's the opinion of the 18 operator or the experience of the operator to determine if 19 he's going into a safe plant or not. He has all the 20 particular indications to indicate that it is a sensor i 21 malfunction or if the jacket water temperature is actually l 22 high. l 23 MR. BOCKHOLD: So that gives us an additional 24 means to bypass this particular trip. I 25 MR. CHAFFEE: And I guess up until yesterday, that 92 PROJECT 035222 .I

5 1 feature of this design, as it applies to Vogtle, was 1 2 something that wasn't recognized. J 3 MR. BOCKHOLD: That's correct, it was not 4 recognized by our operators. 5 MR. KENDALL: Does that bypass the other three 1 6 emergency trips also? 7 MR. OWYOUNG: Well it bypasses the temperature ) 8 trips and also the lube oil pressure trip. It does not 9 bypass the over-speed or differential. 10 MR. CHAFFEE: So of the four emergency trips, it- ) 11 bypasses two of them? 12 MR. OWYOUNG: Yes. 13 MR. CHAFFEE: I guess what's running through my \\ 14 mind right i.ow is the design of the system and some of the 15 features it has are somewhat complicated. I'm not saying 16 that's good or bad but it sounds like there's a lot of 17 things -- there's this one -- I'm saying a lot because the 18 first out enunciator was something that didn't seem to be i 19 well understood by the operators. I guess one of the things 20 that's beginning to run through my mind is that perhaps 21 there's a need to further define all the different features t 22 that this system has to make sure that the operators are 23 aware of those features so that if they find themselves in a 24 scenario where things aren't working right, they're better l l ^ 25 able to capitalize on them. I guess the reason you're 92 PROJECT r 035223

l \\ 6 1 i bringing this one up is perhaps in this event this might i 2 have been useful if the jacket water temperature had caused 3 the second trip as well. i 4 MR. BOCKHOLD: That's correct, in this event that 5 would have been very useful if the' jacket water -- the 6 jacket water pump was running, the jacket water temperature 7 was normal. And you know, we agree with you, we're going to 8 look at our training overall and make sure that the various 9 operators we have understand the local enunciator panel on 10 the diesel, but there are features that enable you to -- you' 11 know, the normal trips are there for a surveillance type 12 start in case you had some problem, you would shut the (f 13 machine down before you had any damage to the diesel. And 14 then when you go into the higher level events, you would 15 take out some of those trips because you definitely want the i 16 engine to run and in the ultimate, even if the engine is not 17 running correctly, you would still want the engine to run. 18 I guess an analogy that I have, since I'm a pilot I 19 and own a small airplane -- my engine doesn't have any 20 trips, it's just manual. i 21 MR. CHAFFEE: I guess as far as the event goes, my 22 understanding is from your testing is that your belief is 23 that -- well maybe I should ask you guys. As far as, you 24 know, the first trip and the second trip, what is currently } t 25 your belief of what caused those trips and why do you l 92 PROJECT 035224

7 1 believe that? (~ 2 MR. KOCHERY: Because we could recreate the same i 3 scenario. 4 MR. CHAFFEE: So what do you believe caused the 5 first trip? 6 MR. CHAFFEE: What do you believe caused the first l 7 trip? 8 MR. KOCHERY: Jacket water temperature. ? 9 MR. BOCKHOLD: You know, I would say what we know ) 10 is, we know we had a trip originally and we didn't capture i 11 all the anunciators because the people weren't there. We 12 dispatched people and when we dispatched the people and got ( 13 their statements, you know, we have the anunciator 14 indication and on the second trip what we believe caused the 15 trip is jacket water. We believe we had one failed sensor j 16 that later we got to reset. Okay? And we assume that one 17 of the other sensors failed. We had one failed sensor going 18 into that second start. -l 19 MR. CHAFFEE: Okay. Let me make sure I understand 20 this. The first trip was caused by high jacket water 21 temperature or low jacket water pressure? 22 MR. KOCHERY: We don't have enough information. 23 MR. BOCKHOLD: We don't have enough information. { l 24 MR. CHAFFEE: Okay, let me try -- I understand I 25 that what you learned in your testing was that -- if I'm 92 PROJECT 035225 i

S 1 correct -- if you had two jacket water temperature sensors 2 in trip, that you find that the timer ran out, I believe it 3 was in 70 seconds or 72 seconds, is that right? 4 MR. STOKZS: 70-71 seconds. 5 MR. CHAFFEE: Okay, now as I understand it, the t 6 first run of the diesel, it tripped 70 seconds after'it i 7 started, is that correct? 8 MR. STOKES: The first one? 9 MR. KOCHERY: The first one was at 80. 10 MR. CHAFFEE: And I understand the second one was 11 80 seconds, in the actual event. 12 MR. KOCHERY: The other way around, first one was l 13 80, second one was 70. I 14 MR. CHAFFEE: Are you sure? 15 MR. KOCHERY: Yes. 16 MR. CHAFFEE: Is that right? I thought it was the 17 other way around. I thought the shorter one was the first 18 one and it was longer the first time. t 19 MR. KOCHERY: No. 20 MR. CHAFFEE: Okay, well if you're right then -- 21 so the longer run was the first one and that one you believe 22 was caused by the jacket water pressure being low? 23 MR. BOCKHOLD: We're not sure why the first trip 24 occurred hscause we can't capture all the anunciators. You ( i 25 know, the trips may have been different from the first one 92 PROJECf 035226 e n v-=,--

8 1 to the second one. We don't know, you know, that to be a 2 fact. I don't know, I guess I'd ask Ken or Sheldon to speak l 3 up about the difference in the first and the second. Maybe 4 you can add something. 5. MR. OWYOUNG: Well without information on the t 6 first. shutdown with the enunciation, I personally don't have 7 any idea what shut down the first unit because I haven't 8 gotten any feedback as far as what enunciator points were l 9 showing on the first unit. 10 The second unit, we were told that they had a high 11 temperature jacket water shutdown, a low pressure jacket i 12 water shutdown and a low turbo lube oil pressure shutdown i 13 indication. So we target those particular enunciator points 14 and our testing proved out that if we have two temperature j 15 shutdown trips for jacket water, that the system would 16 react, that it will shutdown on those two points and the ] 17 resulting fact that we will get the other two enunciator 18 points because the lines aren't being filled. 19 MR. CHAFFEE: Oh, I see, so all four enunciations 20 were able to be reproduced by having two of the sensors 21 reject water temperature -- 22 MR. OWYOUNG: Yes. i 23 MR. CHAFFEE: And those are what was observed in 24 the second trip? i 25 MR. OWYOUNG: Yes, and we repeated the test four 92 PROJECT 035227

10 1 times and got the same results. So we're very confident 2 that_that was the shutdown on the second trip. I don't know 3 what shut down the first trip.- 4 MR. CHAFFEE: I forgot when the first -- was there 5 any reports from anybody that they thought they say any 6 particular enunciators, or that was the second one? 7 MR. OWYOUNG: It was the second one. 8 MR. KOCHERY: The second one we have a list of 9 enunciators. 10 MR. CHAFFEE: Well okay, then -- ) 11 MR. BOCKHOLD: Yeah, when we went through it, we 12 didn't have anything except for the second start. The ( 13 second start, we had clear consistent information from the 14 operators on what enunciation they had. 15 MR. CHAFFEE: And it was the second run that ran i 16 80 second -- I mean 70 seconds? 17 MR. STOKES: Yes, 70 seconds. 1 18 MR. CHAFFEE: And I guess what you found is when 19 you take two sensors out that the thing times out in 70 20 seconds? 21 MR. OWYOUNG: Well with the proper response of our 22 group 2 lockout signal going out to the engine to shut the 23 engine down and back to the panel givi.ng an indication that 24 the engine is in the process of shutting down is about 72 I 25 seconds. 92 PROJECT 035228 i

11 1 MR. CHAFFEE: Okay. So that all seems to make 2 sense for the second trip. 3 MR. OWYOUNG: Yes. 4 MR. CHAFFEE: Since the first trip, the thing ran 5 for 80 seconds, does that tend to suggest that something 6 different occurred or is that close enough within tolerance 7 that -- 8 MR. OWYOUNG: I have no idea. 9 MR. KENDALL: Did you get, from one of the tests, 10 a trip in about 80 seconds? 11 MR. STOKES: That was the same startup with jacket 12 water trip vented and we got about 79 seconds. .( 13 MR. KENDALL: That was with the jacket water 14 temperature vented? 15 MR. STOKES: Jacket water pressure. 16 MR. CHAFFEE: So just with jacket water pressure 17 vented, it took 80 seconds -- 79 seconds? 18 MR. OWYOUNG: What we did also is we wanted to see 19 if we tripped a low pressure jacket water sensor, would it 20 create a temperature indication, so we used another sensor. 21 And we ran one test and it did not indicate a high jacket 22 water temp enunciation, which I expected that to happen. We 23 needed to prove that out. 24 MR. CHAFFEE: So again, it's just speculation, but 25 it's possible that the first trip was just due to low jacket 92 PROJECT 035229

1 12 1 water pressure since the 80 seconds is close to the time it i 2 ran? l 3 Again this is all speculation but perhaps the 4 jacket water pressure caused the first trip, just the fact l 5 that the diesel tripped and started back up, then the second { 6 time, the jacket water sensor -- since we found there is 7 some e.7racticness in their operation, parhaps they chose 8 that opportunity for both of them to fail and that brought 9 in all the anunciators. 10 I guess that sounds plausible but it's not 11 necessarily what happened in terms of the first trip. Do 12 you agree with that? (~ i t 13 MR. KOCHERY: Based on the timing. 14 MR. CHAFFEE: But I guess what is true, the fact l 15 that the first run was 80 seconds, that is a sufficient l I 16 enough difference in the time that it suggests that 17 something different occurred than what happened the second i 18 time? 19 MR. OWYOUNG: I would say yes, because of the 20 timing, the four times that we recorded for the temperature 21 trips were somewhat within one or two seconds of each other. i i 22 MR. CHAFFEE: Oh, okay, I spaess I understand. Oh, i i 23 I see what you're saying -- and therefore, since the second 24 trip was due to jacket water temperature -- well why did the 25 diesel start the third time when they did the emergency 92 PROJECT 035230 i

l 13 1 start? I mean this jacket water temperature could have very A i l 2 easily prevented that start as well, couldn't it? i 3 MR. OWYOUNG: Yes, it could have. 4 MR. CHAFFEE: So I guess you just lucked out -- 5 it's just fortunate that I guess the jacket water 6 temperature probe must have fixed itself. 7 MR. OWYOUNG: Either it fixed itself or it really 8 had a high temperature and then it pulled down or whatever. 9 MR. CHAFFEE: Oh. 10 MR. OWYOUNG: I'm just speculating on that without 11 knowing what all the conditions are. 12 MR. CHAFFEE: So therefore, the point is that this e 13 ability to emergency reset could have been very important if 14 it had turned out that the jacket water temperature probe i 15 was going to prevent the next start. 16 MR. OWYOUNG: Yes. 17 MR. CHAFFEE: I see -- oh, I see what you're 18 saying -- recognizing that the operators had indication of f 19 what caused the second trip, if they had known about this 20 feature, then the logical thing would have been to emergency 21 reset to have done the third start. Recognizing you were 22 trying to ensure you got a good start, people were carefully 23 monitoring everything. 24 MR. OWYOUNG: Yes. 25 MR. CHAFFEE: Okay. 92 PROJECT 035231

14 1 MR. JOHNSTON: You would have to do the emergency 2 stop reset after the third stop. You couldn't do it before 3 that. 4 MR. CHAFFEE: Oh, you can't do emergency reset 5 until you've done an emergency start? 6 MR. JOHNSTON: Right. 7 MR. CHAFFEE: Oh, okay, and therefore, the 8 scenario is that if the emergency start hadn't worked, they 9 would have had to emergency reset to have been able to get j 10 this out or they would have found that they couldn't 11 probably -- there's a possibility they wouldn't have been l 12 able to get it started with the fourth or subsequent ones, l

{.

13 if they hadn't tried to do emergency reset. 14 Okay. Let's talk about the jacket water 15 temperature sensors themselves and their reliability and 16 what's possibly causing them to have problems. My guess is 17 in looking at the history -- is there anybody here that can 18 describe the history and what it sort of tends to suggest 19 about reliability or what might be going wrong with these i 20 sensors? l 21 MR. BOCKHOLD: Well tha way we summarized it was, 22 you know, we have problems, have had problems associated 23 with these sensors during initial startup phases of the 24 engine and during overhaul times, and we have not had 25 problems in between overhaul periods of time. You know, f 92 PROJECT i 035232

15 I 1 we've had reliable starts on the engines in between those 2 periods of time. So that's a general summary. 3 MR. CHAFFEE: Have you guys had -- you know, we 4 talked last night about the surveillance procedures and I. 5 haven't had a chance to take a look at those, I'm not sure 6 we've gotten them. 7 MR. BOCKHOLD: Herb, did Operations deliver some 8 surveillance procedures this morning? 9 MR. BEACHER: I do have a procedure from 10 Operations on my desk, I don't know the number, I can check 11 it. 12 MR. BOCKHOLD: I guess I see Jimmy Paul over 1 ("' 13 there, he could talk to you about the normal diesel 14 surveillance procedure. 15 MR. CHAFFEE: If you did the monthly surveillance, 16 and the diesel started and ran but you got a high jacket l 17 water temperature sensor malfunction? l 18 MR. CASH: We would write a work request tag on 19 that sensor malfunction and that would go into the system I 20 identifying the problem. l 21 MR. CHAFFEE: Do they have a jacket water 22 temperature sensor malfunction enunciator in the control 23 room? 24 MR. STOKES: Yes. s 25 MR. BOCKHOLD: Not in the control room. You don't 92 PROJECT 035233

i 16 1 have any-of that.in the control-room, that's local -- you I 2 do, jacket water? 3 MR. CASH: We have a complete duplicate panel, 4 local and in the control room. 5 MR. STOKES: Yes. 6 MR. BOCKHOLD: It is complete? 7 MR. CASH: Yes, sir. 8 MR. CHAFFEE: It's interesting that -- 9 MR. BOCKHOLD: You don't have -- the trip panel is 10 not the came. 11 MR. STOKES: It's the colors. 12 MR. BOCKHOLD: The colors are different? l 13 MR. STOKES: Yeah, the colors are different, 14 that's all. 15 MR. CHAFFEE: Does the procedure require you to 16 take the action you just described or is that just common 17 sense? 18 MR. CASH: No, that's our standard policy. 19 MR. BOCKHOLD: We have an overall conduct of 20 operations policy, you know, and I believe it's written 21 there. 22 MR. CASH: This would be for any piece of 23 equipment that we ran and got an alarm in that was not 24 anticipated or expected or explained, we would write a work 25 request. 92 PROJECT 035234

i 17 j .g 1 MR. CHAFFEE: Okay, that's very good. 2 The thing that's disturbing me though is like in 3 this most recent event as I understand it, they did the i 4 . maintenance work on the 1-A diesel,-they went through all 5 this testing and then it was declared operable, therefore f 6 indicating that it ought to be ready to go and that should 7 be the condition that it's going to be monitored for every l 8 month thereafter on a monthly basis. Yet after that, the ~ 9 avant occurred and the testing showed that jacket water 10 temperature probes that were in place after the diesel was 11 declared operable had a lot of problems associated with 12 them. So what I'm trying to figure out is why didn't you j [ 13 see some of that in your monthly testing -- why is-it all of 14 a sudden miraculously some time after you start the plant 15 up, for the next year or so you see absolutely no 16 indications of these jacket water temperature sensors having 17 any problems? There seems to be an inconsistency in my mind l 18 there. I don't understand why everything worked so well 19 month after month while you were in operation, yet when you 20 get down and you have your outages and you go through all i 21 your operability tests and it's all ready to go, then you 22 find out after the fact no it's not ready to go. I'm 23 getting the impression that the jacket water temperature 24 sensors are not that reliable and I'm surprised you haven't l 8'. 25 seen that in some of your monthly testing. That's what I'm i l 92 PRO.7ECT 035235 l

i 18 I 1 'being told and if that's true, then there's a discontinuity 4 2 there that can maybe be used to figure out what's going on 1 3 with these, I guess -- or maybe it makes me a little dubious 4 that the monthly maintenances are as rigorous as you say, l 5 although I don't disbelieve what you're saying because it 6 makes sense, operators tend to do that. Do you see the 7 dilemma I find? 8 MR. BOCKHOLD: The -- you know, working with the l 9 vendor, we have some switches -- we received -- I think I 10 told you we received a switch in that he calibrated on i 11 3/24/90 and we received it in and part of the repair process 12 is we recalibrated his switch and it failed. Okay, that i (( 13 switch is currently quarantined. We're going to take the 14 vendor with his switch and try to figure out if there is { 15 some difference there, what the problem is. 16 But you know, the facts show that when we've come 17 out of overhauls basically associated with engine runs 18 around overhauls, we've had problems and we've replaced the 19 switches and we've gotten good switches in that have run 20 until the next overhaul. Those facts have shown that on the 21 four engines. 22 MR. CHAFFEE: Have you ever had either of the l 23 units, when you've done your monthly surveillances, an 24 occurrence where the diesel was tripped, you know, for an I l (k_ l l 25 unexplained reason? I guess you've had in Unit 2 recently? 92 PROJECT 035236

) 1 19 1 .MR. BOCKHOLD: We've got one of the Unit 2 diesels 2 that is in a weekly surveillance but it's failure causes 3 were not associated with jacket water, and Ken can speak to 4 that. 5 MR. STOKES: Yeah, these were totally unrelated. 6 The engine just rolls slowly and did not start. I think 7 this is one that I told you about during the first 8 interview. It happened twice over a period of four starts 9 and I think we tested it about ten times since that and 10 r., *hing has shown up. 11 MR. CHAFFEE: But you weren't able to determina j 12 why it didn't start? ( 13 MR. STOEZS: No. i 14 MR. CHAFFEE: And when you say it wouldn't start, 15 what -- 16 MR. STOKES: It just rolled slowly and -- it gets 17 the start signal for five seconds, rolls slowly and just 18 didn't start. Normally it gets a pretty good kick and it 19 rolls fast and starts up in about six or seven seconds, but 20 this time it just couldn't start and immediately -- I think i 21 both times after that, within ten or fifteen minutes, we did 22 another start and it worked fine. 23 MR, CHAFFEE: Do you have any experience with 24 that, the cooper people? ~ 25 MR. JOHNSTON: We had a problem at the River Bend l 92 PROJECT 035237

20 1 station where oil had accumulated and the pilot tubing fron-i 2 the starting air distributor that feeds the signal to open 3 air start valves and in our analysis of that situation, we 4 felt that problem was unique to the in-line engine. In some l J 5 conversations I had with our people back on the west coast j 6 yesterday, we're evaluating it to see if that can al'so apply 7 to this engine, so that's under investigation. It's 8 possible there may be some oil accumulating in the pneumatic f 1 9 lines that would cause that to occur. f 10 MR. CHAFFEE: Oh. i 11 MR. JOHNSTON: But again, this is unrelated to the i 12 Calcon devices. ( 13 MR. CHAFFEE: Okay, back on the monthly 14 surveillance, you understand why this confuses me that you [ 15 could have all these problems in maintenance and do your 16 testing and stuff and then have what you're telling us is a i 17 clean record for almost an entire year or so. That seams j 18 inconsistent, so that's why I'm a ilttle -- I can't help but 19 wonder if -- t 20 MR. BOCKHOID: You know, the handling of these [ 21 switches -- and let's see, were you able to -- i 22 MR. CHAFFEE: No, no, they -- l l 23 MR. BOCKHOLD: They didn't make it over here with (((- 24 the probe? t 25 MR. BRINEY: It was delivered to Herb Beacher this l 1 92 PROJEcy 035238 i

. _ =. -21 1 morning. f 2 MR. BOCKHOLD: You know, I think handling of the t 3 probe might have something to do with it. The tip of the 4 probe is probably fairly sensitive and, you know, we can t 5 postulate and speculate and Item 1 on key lessons learned 6 and recommendations is to go attack that aspect of the 7 thing. We really can't come up with -- you know, we can 8 speculate all day on what it might be, but we don't know 9 until we go look at it. 10 MR. CHAFFEE: Maybe what we're seeing is just the 11 fact that the same sensitivity the switch has that causes it 12 not to work also results in the fact that you fait-ly quickly ( 13 after the fact find out they're not working and that of i 14 course forces you into fixing them right so that when you -- 15 MR. BOCKHOLD: Yeah, we put a new switch in and 16 that kind of stuff. We'll pass you down the switch 17 assembly. 18 (The switch was handed to Mr. Chaffee.) i 19 MR. BOCKHOLD: The rings on the -- you've got to 20 take the shield off. 21 MR. CHAFFEE: This silver part here? 22 MR. BOCKHOLD: Yeah. i 23 MR. CHAFFEE: Don't let me break this thing. I 24 MR. BRINEY: There's a set screw in the side of l ( ki. 25 it. 92 PROJECT 03$239

22 1 MR. CMAFFEE: You have to take the set screw out 2 in order to -- I don't have an allen wrench. 3 MR. BOCKHOLD: If you take that -- 4 MR. CHAFFEE: Can we get an allen wrench? 5 MR. BOCKHOLD: St tre. t 6 MR. CHAFFEE: if you take it off, there's a bunch l l 7 of little rings under there? 8 MR. BOCKHOLD: Well there's another probe, there's 9 a thinner, solid probe. At the and is a bunch of rings. 10 I'll go to the board and draw it. ) 11 MR. KENDALL: When they're removed and installed { 12 are they taken out of place with the well? (( 13 MR. KOCHERY: No, we don't take the well, just the 14 sensing device. 15 MR. KENDALL: So when you calibrate them, you just i 16 unscrew them out of the well and with rings exposed and take 17 it to the shop? 18 MR. KOCHERY: That's right, and put it in another 19 well. 20 MR. CHAFFEE: So when you get it out of the 21 warehouse and you do the cal test on it, do you take it just 22 like this or do you take this shield off? ^ 23 MR. KOCHERY: You have to take the shield off. 24 MR. CHAFFEE: You guys just asked that ((%- t 25 question. 92 PROJECT 035240

23 1 MR. KENDALL: Yes. You have to loosen the set i 2 screw and then it just lifts out. 3 MR. CHAFFEE: Why do you have to do that, why 4 don't you just leave it intact? 5 MR. KOCHERY: Because when you take the thermal 6 well you can leak. 7 MR. CHAFFEE: Say that again, I didn't understand 8 you. 9 MR. KOCHERY: It's coming from the pipe, that l 10 thing is screwed in. 11 MR. CHAFFEE: This thing is what you screw into 12 the pipe. ) ([ l 13 MR. KOCHERY: That's correct. 14 MR. CHAFFEE: Why don't you just take this whole 15 thing and put it in the thermal bath and do your test that 16 way? 17 MR. KENDALL: Because you've got the keep warm 18 pump and -- 19 MR. KOCHERY: The system is filled, when you take 20 out, water comes out. 21 MR. BOCKHOLD: A1, if you take that metal shield 22 off, the thermal well off, you would find it's head with 23 inside another probe, another metal probe. And at the r 24 bottom of the probe if you looked at it, it looked like it li 25 had kind of rings here. And if you could see inside that, 92 PROJECF 03524g

24 4 1 there's a shaft and there's some springs up above and that i 2 kind of stuff, but the shaft actually ties onto the outer 3 casing down here at the bottom. l 4 As the' device heats up, these rings expand at a 5 different rate than the center shaft. The center shaft 6 moves up and down -- well then this expansion moves this top i 7 adjusting ring up and down and over here, there's some type 8 of air switch. So when this contacts the air switch it would 9 affect the sensor. 10 MR. CHAFFEE: Does -- so what you do, the reason 11 you take this thing out of the shield is because you don't 12 use this specific shield -- f 13 MR. BOCKHOLD: That's right. i 14 MR. CHAFFEE: -- you use the shield that's in 15 place for this one. 16 MR. BOCKHOLD: Right. 17 MR. CHAFFEE: When the licensee calibrates, do 18 they calibrate this as an assembly, including the shield, or 19 do they calibrate the probe without the shield. 20 MR. BOCKHOLD: Calibrate the probe. l 21 MR. CHAFFEE: You've checked on that. i 22 MR. KOCHERY: Put in the thermal well. l l 23 MR. CHAFFEE: What I'm getting at is when they do 24 their calibration in the shop, do they treat this as a I km. 25 calibrated component? 92 PROJECT 035242

.~ _= 25 1 MR. BOCKHOLD: The way I've seen it done in the 4 2 shop is just the assembly.. i 3 MR. BRINEY: What he's asking I think, George, is t 4 that'do we need to calibrate the switch as an assembly with f 5 its associated thermal well. l 6 MR. CHAFFEE: Right, is there something about i 7 this-- i 8 MR. BRINEY: The answer is no, we not have to do 9 it. } 10 MR. CHAFFEE: You've talked to the vendor about 11 that? 12 MR. BRINEY: Yes. The vendor also -- we also had (( 13 some experiments that showed that there was no significant i 14 difference in calibration whether we used the well or not. 15 They recommended that we did because the tip of the probe is i 16 somewhat fragile. 17 MR. CHAFFEE: But there's no close tolerances 18 inside here? I mean those little rings and stuff, they 19 don't come in close contact with -- I would assume that you 20 21 MR. BRINEY: There's an air gap between the actual i l 22 probe and the thermal well. i 23 MR. CHAFFEE: But they'd want to minimize that t ((.- 24 because the goal is to get the rapid heat transfer from the ( 25 water to those little things, little rings, right? 92 PROJECT 035243

l t 26 I i 1 MR. BRINEY: Right. I 2 MR. CHAFFEE: Is the thermal well that's in the 3 plant, is there any rust or anything inside there that -- l 4 MR. KOCHERY: That's stainless steel. 5 MR. CHAFFEE: Is there -- I mean, there's no f 6 cleanliness problems, no nothing in them that's abnormal? l 7 MR. KOCHERY: No. 8 MR. CHAFFEE: Okay. And the vendor, when he does l 9 his calibration, he doesn't calibrate it with the particular l 10 thermal well that they send out, they just do the 11 calibration without? 12 MR. BRINEY: The vendor uses a thermal well, yes. 13 MR. CHAFFEE: And that's the one that they send i 14 with the instrument? i 15 MR. BRINEY: Not necessarily. 16 MR. CHAFFEE: Oh, I see. So what ends up 17 happening is this part, the one that's in place -- I guess 18 what you do is you unscrew the allen screw and you take this 19 thing and screw it down inside the thermal well and then put i 20 the allen screw in? 21 MR. BRINEY: It just slides into the thermal well, 22 it doesn't screw into the thermal well. 23 MR. CHAFFEE: Oh, and the allen screw just holds ( 24 it in place? 25 MR. BRINEY: Right. Once you unloosen the allen 92 PROJEC' 035244

.7_, .4.. 2 m 21 1 screw you can pull it straight out. 2 MR. CHAFFEE: Is there any way potentially the 3 allen screw could impact the properties of the sensor or is 4 it somehow located such that it can't have any impact on 5 that? 6 MR. BRINEY: The sensitivity seems to be at the 7 and of the probe where the o-rings are, not up at the shaft. 8 MR. CHAFFEE: Is there any way attaching the allen 9 screw could cock the sensor such that it comes in contact' 10 with the wall of the thermal well and that has an impact on 11 those little o-rings? 12 MR. BRINEY: Not that we've seen, no. t 13 MR. CHAFFEE: Okay. 14 MR. LYON: Can the air connection bring any air 15 down into the system itself, into the instrument itself? 16 MR. OWYOUNG: I would say no. The air supply to 17 that particular device is dead-ended. If it vents, it vents 18 out through the exhaust port. 19 MR. CHAFFEE: Okay. So I guess if you have not 20 found any problems with these jacket water temperature 21 probes between refueling outages, then it must be that 22 somehow the process of dealing -- i 23 MR. BOCKHOLD: -- with handling them or 24 calibrating them or, you know, maybe some of then aren't as 25 reliable as other ones. The ones we seem to get in there, 92 PROJECT l 035245 j l l i ~. 1

28 1 you know, that we discard -- the ones we finally get in 2 there are very reliable. The ones that we try to put in and 3 try to get calibrated during overhauls, we have a number of 4 problems and you know, the key element here is to -- when 5 the calcon man arrives today, we're going to go through all L 6 this and try to determine what those issues are, how come 7 we're having problems at the overhaul time frame. 8 MR. CHAFFEE: And I guess along with that is 9 coming up with a methodology of ensuring what testing is 10 necessary to ensure that once you declare the diesel 11 operable, you're not going to have any of these jacket water 12 temperature probes causing you any problems. 13 MR. BOCKHOLD: That's correct. 14 MR. CHAFFEE: Okay. I guess at this point, the 15 hope is that on Unit 2, somehow you have -- 16 MR. BOCKHOLD: Unit 2 had a different calibration 17 procedure and a special test procedure performed, associated 18 with raising jacket water temperature because of a heat i 19 exchanger concern. And so we used a different procedure 20 there and we actually used the whole system and we haven't 21 had any failures on Unit 2. So the switches we have in 22 place, we feel are reliable and as I say, we actually use 23 temperature to finally -- before the diesel was declared 24 operable for fuel movement, some -- you know, beginning of j 25 '89, and I think you have in here that we changed out some 92 PROJECT 035246

29 1 switches on Unit 2 in 12/8/88 -- I'm sorry, on 7/21/88, we i 2 changed some switches out and that's when -- that's i 3 basically six months ahead of when we needed to declare the 4 diesel operable for fuel movement on Unit 2. p 5 MR. CHAFFEE: But again, how they calibrated the 6 new ones that they put in -- 7 MR. BOCKHOLD: Those were calibrated to a 8 different procedure at that time and a special test was run 9 at that time, we believe that procedure yielded reliable 10 switches, and that special test, although it wasn't designed 11 specifically to handle the Calcon switch issue, verified 12 that those switches would be reliable and they have proved 13 reliable ever since then. (: 14 MR. CHAFFEE: Was the special test done after they 1 15 put the new switches in that needed to be calibrated or was 16 the calibration done as a result of the test where it turned 17 out the engine tripped at a lower temperature? 18 MR. BOCKHOLD: I don't know. Ken, do you remember ) 19 -- either of the Kens? 20 MR. BURR: We ran the engine and it tripped, so 21 therefore we took them out and recalibrated them and put 22 them back in and then they were okay. 23 MR. CHAFFEE: Then after you recalibrated them and 24 put them back in, how did you then ensure that you -- okay, I k.' 25 so you took them out, you recalibrated them and you l 92 PROJECT 035247

30 1 recalibrated them to a procedure -- r. 2 MR. BURR: Yes. 3 MR. CHAFFEE: That procedure that you used for 4 recalibrating them, is that different from what you've been 5 using over here in Unit 2 in terms of how you did it? 6 MR. BOCKHOLD: Unit 1. 7 MR. BURR: It's different from Unit 1, yes. 8 MR. CHAFFEE: How is it different? 9 MR. BURR: We used a different air pressure. 10 MR. CHAFFEE: Oh, really? What was the air 11 pressure used on Unit 27 12 MR. BURR: On Unit 1, we are required to reduce 13 the air pressure to 20 pounds. On Unit 2, we used 40. 14 MR. CHAFFEE: Oh, I see, the difference is how low 15 the pressure has to get to. 16 MR. BURR: Yes. 17 MR. CHAFFEE: And Unit 1 is going to vent down to 18 20 pounds and Unit 2 down to 40 pounds? 19 MR. BURR: Yes. That at that time is how we 20 calibrated it. 21 MR. BOCKHOLD: The 40 pounds -- 1 22 MR. BURR: That new procedure came after that time j 23 period, so this procedure we used on Unit 1 is a later time 24 than what we calibrated Unit 2. I- { 25 MR. CHAFFEE: Oh. So what that tends to suggest 92 PROJECT 035248

31 1 is that the calibration you're doing on Unit 1 where you've i 2 got to demonstrate it can vant down to 20 pounds -- 3 MR. BURR: I don't know what significance that 4 has, it's just that it's different. 5 MR. BOCKHOLD: The 40 pounds would tend to make 6 the switch -- 7 MR. CHAFFEE: Less likely to trip. 8 MR. BOCKHOLD: -- less likely to trip, basically. 9 The logic card that sees the pressure has to get to 40 10 pounds for that logic card to actuate that trip. Do you see 11 the rings on it now? 12 MR. CHAFFEE: Yeah, and the tolerance in this ,( 13 thing is more than I expected. 14 MR. BOCKHOLD: The tip is what's sensitive. If 15 you look at the tip, you'll see how that is connected. 16 MR. CHAFFEE: It looks like it's not a tight fit, 17 but it's a -- 18 MR. BOCKHOLO: It's got to be close, you can't 19 have a big air gap because you don't want a big time delay 20 for temperature to be sensed by the probe. 21 MR. CHAFFEE: Wouldn't it be something if all you 22 were dealing with here is the tolerance of these rings and 23 your thermocouple and the fact that you're changing the 24 manufacturing design tolerances is enough to cause them to f (,- ( l 25 behave strangely and what you're really doing is you're i 92 pgGJECT o35249 l N-.

32 1 going through enough probes until you find one that is i 2 within that tolerance so that it works properly once it's-3 put in. 4 on Unit 1, did you at one point historically 5 calibrate then down to 40 pounds and then changed to 20 6 pounds? 7 MR. BOCKHOLD: Have we got that kind of memory? i 8 MR. BURR: I don't know. 9 MR. BRINEY: In 1988, we asked for information on 10 how to calibrate the switches. That information hasn't i 11 changed since the outage of '88 and this date. This is on I 12 Unit 1 switches. I ( 13 MR. CHAFFEE: So the information you got in '88 14 how to calibrate the Unit 1 switches, that included testing 15 them to vent to 20 pounds? 16 MR. BRINEY: Correct. 17 MR. CHAFFEE: Prior to that. 18 MR. BRINEY: The reset was considered to be at 40 19 pounds. 20 MR. CHAFFEE: Whereas in Unit 2, the trip point 21 has to be down to 40 pounds and they reset at some different 22 value? 23 MR. BRINEY: The supply air pressure is at 60 24 pounds. The way we calibrate on Unit 1, the supply air 25 pressure was at 60 pounds, the switch was considered to trip 92 PROJECT 035250

33 1 as it got to 20 pounds, it was considered reset as the 2 pressure built up to 40 pounds. 3 MR. CHAFFEE: Okay, now what's the situation in 4 Unit 27 l 5 MR. BOCKHOLD: Unit 2 was a startup test procedure l C and -- do you remember? 7 MR. BURR: We basically used the 40 pounds. 8 MR. CHAFFEE: If they came down to 40 pounds, they 9 tripped, and what were they resetting it? If the pressure 10 is 60 pounds and they tripped at 40, do you have it reset at il 59 or what? 12 MR. BURR: The reset issue was not an issue on ( 13 Unit 2. 14 MR. CHAFFEE: Does 40 pounds as being a trip point I 15 for these sensors make sense? l l 16 MR. OWYOUNG: Yes, it does because we have a P-3 17 valve in there which trips off at 45 psi. Okay now -- [ 18 MR. CHAFFEE: Well then is the 20 pounds i 19 incorrect? t 20 MR. ONYOUNG: I think there's a confusion of 20 21 pounds, 40 pounds and 45 pounds and what-have-you. i 22 Basically what we sent the utility is a generic calibration t 23 procedure -- just a generic one. In our normal system we ((L 24 have an element, we call it a knock out element, in the 25 circuitry and this particular element, this other logic j 92 PROJECT 035251 l

34 1 monitors the stroke and as the stroke heats up, it starts l I' 2 venting out air. And the way this particular logic element ] 3 is setup, at 20 psi with some tolerance, this element will 4 then fire, which would create a shutdown. As this element 5 here cools down and starts blocking the pressure, at 40 psi 6 this element will then stop this output signal. So that's I 7 how we got the 20 and 40 psi increments. 8 Now for this utility and other utilities that will 9 install another device here, pressure device and that guy is 10 st a 45 psi falling, in addition to that. So that now when 11 this element falls down below 45 psi, this guy here goes to 12 vent and also vents this line down. There's other circuitry g 13 in here but in essence this right here goes to vent. 14 So that's where we're looking at 45 psi, so on the 15 second unit when these guys were calibrated, they were 16 shooting for the 45 psi falling level because it's five I 17 pounds below this valve here. i 18 MR. CHAFFEE: Is that design the same for Unit 1 19 and Unit 27 20 HR. OWYOUNG: It is the same for Unit 1. The 21 procedure that they're using for Unit 1 as far as 22 calibration, again, is just a generic calibration. 23 MR. CHAFFEE: There's no reason for the procedures t [ 24 to be different in the two units? g s-25 MR. OWYOUNG: That's correct. 92 PROJECT 035252

t 35 1-MR. CHAFFEE: Yet, in fact right now they are, is 2 that right? i 3 MR. BOCKHOLD: They were done differently,-they 4 would be done the same if we went to overhaul the Unit 2 5 engines. 6 MR. CHAFFEE: Is there any chance that the set 4 7 point that they're trying to calibrate for is having any 8 kind of impact on what is being seen here? In other words, 9 the fact that the switches keep acting -- there's been I 10 occasions where they've acted a little erratic. In the 11 diesel start you did a couple of days ago, your sensor i 12 failed and you puffed it with air, then it works fine for g( 13 awhile. Who knows when that might occur again. Does the j l 14 calibration have anything to do with that? 15 MR. OWYOUNG: I don't know, I would just be 16 speculating. There could have just been a little piece of 17 dirt in there, could have been almost anything. 18 MR. CHAFFEE: Could it simply be the fact that the 19 stickiness -- as the thing heats up, you know, the thermal 20 expansion and the tolerances in this thing is causing it to 1 ) 21 stick? 22 MR. OWYOUNG: That could be also. That's why I 23 suggested to get the Calcon expert here, the expert of that 24 device. l ) 25 MR. CHAFFEE: Yeah. Okay. Do you have any other i 92 PROJECT 035253

t 36 1 questions, Rick? l i r 2 MR. KENDALL: No. 3 MR. CHAFFEE: So basically what we've heard is I 4 that as far as Unit 2 goes, you did the calibration there -- 5 I would assume that when you recalibrated them, you must i 6 have used the thermal bath type approach, but one difference 7 there is you considered to have tripped once you got to 40 4 8 psi. I guess what's true when we watched you do the 9 calibration is that for that one that was sluggish, I guess 10 the question remains in my mind, would that one that was 11 sluggish, would it have been considered to be sluggish, l 12 considered to have tripped once it got to 40 pounds? I l( 13 guess we saw it slowly -- it starts off at 60 pounds and 14 slowly bled off and then I guess -- what pressure was it, 15 when you do the calibrations, that it begins to drop fairly { 16 quickly? It starts off very slowly and then goes very i 17 quickly -- about what pressure does it go over edge and 18 start going down quickly? 19 MR. BRINEY: Somewhere between 40 and 50 pounds, 20 it depends on the switch. 21 MR. CHAFFEE I wonder if there's some way that 22 having that different set point at 40 pounds would -- I 23 guess not. If anything, the 40 pounds seems to work better i 24 than the 20 in terms of Unit 2. i ((L f 25 MR. BOCKHOLD: Forty pounds tends, in my mind, to I 92 PROJECT 035254

. ~. 37 1 1 raise the trip set point'on these switches, therefore the i -2 likelihood that these switches would_cause an engine 3 shutdown is reduced. But, you know, we're going to work 4 with the vendor on the switches, especially the one that he i 5 sent us that he calibrated on the 24th, and we tried to I 6 calibrate and it didn't appear to calibrate correctly. It 7 might be a fluke, that one might be just bad. 8 MR. CHAFFEE: In that case what happened was you 9 received that particular item like this and the calibration 10 that you did that said it wasn't workable was you took this 11 thing out and put it in a bath and found out that it 12 wouldn't work properly, is that correct? ( 13 MR. BOCIQiOLD: Yeah, we used our standard 14 procedure on it. l 15 MR. CHAFFEE: When you put it in the bath -- 16 MR. BOCKHOLD: The standard procedure now is you 17 take that out and you put it in a bath and you raise -- you 18 know, you check its set point and it should be venting at 19 the appropriate point and then you raise and lower 20 temperature three times and you're tapping on it lightly to 21 simu' late the vibration of the machine. And that's our 22 current calibration procedure for the probee we have on Unit 23 1. 24 MR. CHAFFEE: When you actually do the calibration 25 in the bath, you actually take the probe and put it in the 92 PROJECT 035255

e 38 l' bath without it being in a thermal well? i 2 MR. BRINEY: No, we calibrate it in a thersel 3 well. 4 MR. CHAFFEE: Oh, okay. 5 MR. BRINEY: We have performed some calibration of 6 some switches outside of the thermal well with the vendor 7 present and there was no significant difference in the t 8 calibrations whether it was in the well or outside the well. 9 MR. CHAFFEE: Okay, I understand. So the one the i 10 vendor sent and said was fine, you took that component with 11 the thermal well it came with, put it -- 12 MR. BRINEY: Just like you have it in your hand. 13 MR. CHAFFEE: -- in and it turned out it didn't g 14 work. 15 MR. BRINEY: Correc's. r 16 MR. CHAFFEE: Okay. I wonder how the vendor is 17 going to explain that, that's interesting. i 18 Have you guys done any kind of a QA source 19 inspection of Calcon to look at their QA program -- before I 20 ask that -- the Calcon jacket water temperature sensors that i 21 we're talking about, those are Q-class components because 22 it's emergency? 23 MR. KOCHERY: (Nodding head affirmatively.) 24 MR. CHAFFEE: It is Q-class, okay, fine. Have you (- 25 had any kind of source inspections to see how good their 92 PROJECT 035256

l 9 5 39 1 calibration process or whatever is'in terms of.-- i ' 2 MR. BOCKHOLD: George, do you know? 3 MR. FREDERICK: I'd have to research it, I don't i 4 know offhand -- just sitting here, I don't know, I'd have to 5 go look. I should have the documents in my office. 6 MR. BOCKHOLD: We buy the switches directly fron 7 Calcon or we buy the switches from who? j 8 MR. FREDERICK: I don't know that, Georgia. l 9 MR. KOCHERY: There is a Part 21 on the calcon. 10 MR. CHAFFEE: Say again? 11 MR. KOCHERY: There is a Part 21 report on Calcon. 12 MR. CHAFFEE: I noticed in this write-up here, ( 13 there was a Part 21 here that seemed to talk about low -- I l 14 assumed from the way this was written that it has to do with 15 low pressure lube oil. But is there another one for these ll 16 temperature -- 17 MR. KOCHERY: Yes, and another one just for the 18 safety-related equipment. 19 MR. CHAFFEE: Safety related? 20 MR. KOCHERY: Yes. 21 MR. CHAFFEE: So there's a Part 21 for Calcon 22 switches for safety-related equipment? 23 MR. KOCHERY: That's right. 24 MR. CHAFFEE: And it does address itself to I 25 temperature sensors? 92 PROJECT 035257

40 1 1 MR. KOCHERY: I don't know right now, I'm trying 2 to get it. 3 MR. CHAFFEE: You're trying to get it now? I 4 MR. KOCHERY: We've got it here, I don't know 5 exactly what it is. There is another Part 21 on the lube 6 oil pressure switch also. 7 MR. CitAFFEE: The Part 21 that's on safety-8 related Calcon switches -- 9 MR. KOCHERY: That's right. 10 MR. CHAFFEE: -- do you know how old that is? Was 11 it just recently received or is it years old? 12 MR. KOCHERY: I don't know. Ken, do you know how I ( 13 old that Part 21 is? 14 MR. STOKES: (Nodding head negatively.) l 15 MR. KOCHERY: It's a lo; dora newer, I believe -- 16 okay, that's a number 89. 17 VOICE: 88. 18 MR. KOCHERY: 887 Okay. 19 VOICE: The work orders that were issued to l 20 recalibrate the pressure switches, because of Part 21, are 21 1988 work orders. l \\ l 22 MR. KOCHERY: No, I'm talking about the Calcon i I 23 Pafety-related parts, Part 21 -- 24 MR. BOCKHOID: Yeah, but he's saying we responded I ('. 25 to it by a work order. We'd have to get the paper and -- 92 PROJECT 035256

41 1 MR. CHAFFEE: It sounds like what Paul says is 2 there's two Part 21's. 3 MR. KOCEERY: Yeah, there are two Part 21's, one 4 for the lube oil pressure switch alone'and another one for - 5 - the Part 21 for the Calcon safety-related parts. There is 6 some problem with their QA problems. 7 MR. CHAFFEE: Can you have somebody pull those two 8 Part 21's for us? 9 MR. KOCHERY: Yeah, I think we can get them. 10 MR. CHAFFEE: Do you happen to remember what the 11 Part 21 says about the safety-related -- is it focusing on - 12 13 MR. BOCKHOLD: He doesn't know, he says -- g 14 MR. KOCHERY: I only know I have the number, I 15 just looked this morning, I think it's 141. 16 MR. BOCKHOLD: These words here, and I'll read 17 them to you - "These items were manufactured by California 18 Controls (calcon). The report stated that the items in 19 service that have performed satisfactorily, were considered 20 to be acceptable. A follow up report dated 5/12/88 21 identified an additional problem with lube oil." 22 It appears -- and I'm speculating now and we'd 23 have to get the paperwork -- but it appears to have said 24 hey, maybe don't accept our calibration when it comes in i k. 25 because we've had some QA problems, do your own calibration. i l 92 PROJECT 035259

42 1 If your own calibration.is good, the switch is good. I 2 guess that's what -- just from these words is what I would i 3 speculate and we'd have to see the information. Do you t 4 think that's what it was, Mark? 5 MR. BRINEY: George, I didn't go through and 6 actually read the Part 21 report. The only thing I had to 7 go by is -- 8 MR. BOCKHOLD: Is what the work order says, yeah, 9 and that's what I would speculate from what the work order I 10 says. i 11 MR. CHAFFEE: Okay. So I guess we'll get the Part 12 21's to see what they say, and Calcon is coming out to see if 13 if they can shed some light on what's going on with these 14 switches. There's a need to try to determine what has to be 15 done to make sure that once you declare operability on the 16 diesel, it's going to work. There's indication that in the 17 past -- 18 MR. BOCKHOLD: We believe the diesels are operable 19 right now and we believe that we have -- from our history we 20 have -- the switches that are in the diesel are reasonably 21 reliable to perform their function. 22 MR. CHAFFEE: No, but what I meant was that in the 23 case of this outage, you went through the outage in the 1-A 24 diesel, went through the testing, declared it operable and 25 then it failed. 92 PROJECT 035260

r 43 1 MR. BOCKHOLD: That's correct. t 2 MR. CHAFFEE: So there's something in the 3 operability testing to date that didn't work completely. So 4 whatever that flaw is needs to be flushed out. j F 5 MR. BOCKHOLD: We need to determine what the 1 6 problem is with the switches. 7 MR. CHAFFEE: Right. And -- right. So that you 8 can ensure that once you declare them operable, they -- 9 there's not any concern about there being some hidden -- 10 MR. BOCKHOLD: That the probability of two switch I 11 failure is very, very remote. 12 MR. CHAFFEE: Right. Okay, do you have any ([ 13 thoughts or questions, Rick? 14 MR. KENDALL: (Nodding head negatively.) r i 15 MR. CHAFFEE: So as far as testing on the diesel 16 generators go -- I guess you finished the UV test and that j 17 was -- was there any other problems encountered during the 18 testing? And the testing is now complete? 19 MR. BOCKHOLD: Since we changed out these 20 switches, how many starts did we do on the A diesel after 21 that? We did multiple starts? 22 MR. STOKES: Yeah. 23 MR. BOCKHOLD: A dozen? 24 MR. OWYOUNG: About a dozen. I 25 MR. BOCKHOLD: About a dozen successful starts. 92 PROJECT 035261 .~

i 44 1 MR. CHAFFEE: You.mean after you changed out-these t 2-jacket water temperature sensors? 3 MR. BOCKHOLD: Yes, we've done what -- so you're 4 saying more than a dozen, 14, 15 starts? 5 MR. STOKES: Something like that. t 6 MR. BOCKHOLD: Successful starts on that engine. 7 MR. CHAFFEE: So basically on -- what was it, 8 Friday I think it was, is that when we started this test f 9 sequence -- Saturday -- Friday I think it was. 10 MR. OWYOUNG: Friday I think it was. e 11 MR. CHAFFEE: So that the first start -- at that i 12 point, you've got a jacket water temperature sensor that I g 13 didn't work properly, that was the first start. Then you { 14 pulsed it with pressure and -- now that particular sensor l 15 has been removed 16 MR. OWYOUNG: Yes. 17 MR. CHAFFEE: And another was put in its place. 18 MR. OWYOUNG: That's right. 19 MR. CHAFFEE: Okay, and also you put another one l 20 in to replace the one that tripped? 21 MR. OWYOUNG: Yes. 22 MR. CHAFFEE: So what's in there is two new 23 sensors plus the one that was originally there before you i l 24 started the test sequence? 25 MR. OWYOUNG: That's correct. 92 PROJECT 035262

i l 45 i 1 MR. BOCKHOLD: No, let me correct the sequence. I P 2 think -- and you guys-correct me if I'm wrong -- we found 3 one that was tripped, okay? We found one that was venting 4 some. i 5 MR. CHAFFEE: Right. 6 MR. BOCKHOLD: Okay, that one that was venting 7 some may have worked perfectly well, but if they vent at.all t 8 we go ahead and pull them out of the engines and some of the 9 history here on this thing that you have might'be, gee, that' 10 little vent port there was venting, and venting some makes i 11 our technicians reasonably nervous and I agree with.them. 12 It's just better to pull the switch and put a brand new one (( 13 in that's not venting at all. 14 MR. CHAFFEE: Is the one that was venting the one 15 that was 186 degrees set point earlier on? It was the other 16 one? i 17 MR. OWYOUNG: At this time, I don't really know. i 18 MR. BOCKHOLD: I think you're confusing B engine 19 with A engine now. 20 MR. CHAFFEE: No, it was the -- well maybe, but I 21 was told that it was the A engine that they found that the 22 jacket water temperature high was 186 degrees so they re-23 adjusted it. g 24 You went into the testing with that one having ( (. 25 been re-adjusted and the other two, and it was those three iP2 PROJECT 035263

l I 46 1 sensors that you found one of them was in trip, one of them i I 2 was venting and we asked the question Saturday I think it l 3 was, well was the one that was venting, was that the one l 4 that was found at 186 degrees? Becattse we were saying 5 well, is it possible the venting was some sort of another E 6 indication that that one was going to have a problem as well r 7 as the 186 degrees indication. I guess maybe you haven't i 8 been able to figure that out. The question is which one of 1. I 9 the sensors was the one that was 186 degrees? 10 MR. BOCKHOLD: Do we have that information? l 11 MR. BRINEY: I've got the data down on my desk, 12 George, I don't have it in my hand. ( 13 MR. BOCKHOLD: We can't answer that one without 14 .st data. 15 MR. CHAFFEE: Okay, I'd like to -- 16 MR. KOCHERY: This is at the shop, venting at the 17 shop? I 18 MR. BOCKHOLD: Yeah. No, what he said is after 19 the start, we went back to see what the calibration was on r l 20 those switches and we found one at 186 degrees. His I 21 question is, is that switch --.was that switch venting, 22 because later on in the sequence of testing when we got back t 23 to A diesel, the recent change out was one was in a tripped [ configuration and one was venting. We changed the one out 24 l 25 that was venting just for not to take any chance. 92 PROJECT 035264 n

47 1 MR. KOCHERY: Okay, that 186 switch was the one t 2 venting when it was calibrated and when we put it back on 3 ' the engine, it was venting, but it didn't cause a trip or 4 alarm. 5 MR. CHAFFEE: Okay, so that -- 6 MR. KOCHERY: It's switch number 19112. l 7 MR. CHAFFEE: The reason I was curious is I was 8 wondering if perhaps the fact that the calibration had gone 9 out -- maybe there is some relation to the fact that it was 10 venting, maybe there's -- 11 MR. BOCKHOLD: There may be, may very well be. 12 MR. CHAFFEE: So then you replaced the one that t ( 13 was venting, you replaced the one that was tripped and after 14 you did that, then how many successful starts did you have? I 15 Because that's what you currently have in place. 16 MR. BOCKHOLD: Right, about 14 -- 127 l 17 MR. STOKES: I'm not sure precisely, but after 18 actual replacement of the switches maybe -- 19 MR. ONYOUNG: I know that when we were there we J 20 started at least six times. 21 MR. STOKES: Yes. i l 22 MR. BOCKHOLD: Six times plus the under-voltage 23 test plus the operability test. 24 MR. STOKES: Maybe eight times. l(w 25 MR. CHAFFEE: Okay. Now how about the jacket 92 PROJECT 035265 9 y -.-,,_y-. ,r--

48 1 water pressure low, that didn't give you any problems at all i 2 and that one was replaced before you started the test 3 sequence, wasn't it? 4 MR. KOCHERY: We didn't replace it, we calibrated 5 it and put it back, the same one. It was in calibration. 6 MR. CHAFFEE: Is that the same one that was -- 7 MR. OWYOUNG: We checked the calibration, we 8 didn't calibrate it, we checked the calibration. 9 MR. KOCHERY: That's right, we checked the 10 calibration. 'I 11 MR. CHAFFEE: Is that the one that was in place at 12 the time of the event? ( 13 MR. KOCHERY: After the event. 14 MR. CHAFFEE: So the one that was in place during l 15 the event -- during the event on March 20th, is the one l 16 that's still in there? i 17 MR. KOCHERY: Which one, pressure or -- 18 MR. CHAFFEE: Pressure. 19 MR. KOCHERY: Yeah, same switch. It was in 20 calibration. When we checked the calibration, it was in 21 calibration, so we put it back. i 22 MR. CHAFFEE: So if you have any strength in the j i 23 conviction that the first trip was caused by low jacket 24 water pressure -- to the extent that's a possible scenario, l ( 25 you still have that instrument in place. 92 5tOJECT 035266 i )

49 i 1 MR. KOCHERY: That's.right, that instrument is in I 2 place. 3 MR. CHAFFEE: I guess I would ask you whether or E 4 not -- hamm -- 5 MR. BOCKHOLD: The low jacket water pressure comes 6 in any time you get the high jacket water temperature. The l 7 second start proved that. 8 MR. OWYOUNG: Only if the two sensors are venting l 9 during a start period. 10 MR. BOCKHOLD: Well -- [ 11 MR. OWYOUNG: If the engine is running after the j 12 group 2 lockout time has elapsed, so we're running for maybe f ([ 13 15 minutes and we have a high jacket water shutdown, that I 14 will be the only indication you will get. You'll get the 15 three indications only during a startup period. 16 MR. BOCKHOLD: Okay, I understand what you're 17 saying. 18 MR. CHAFFEE: I don't. i 19 MR. BOCKHOLD: Well that alarm on the first start i your question is that alarm on the first start could -- 20 21 don't you have a concern about that alarm on the first 22 start. The answer to that question is we checked the 23 calibration of the switch, it was fine and on the second 24 start, okay, we got that alarm also. So our logic should be ((f' 25 that the root cause that we have identified here would cover i ) 92 PROJECT 035267 i - ~.

50 1 that enunciator on the first start. 2 MR. CHAFFEE: Right, except the one thing that's 3 got me a little nervous is that you know,_the first. trip 4 occurred 80 seconds after start whereas the second one was i 5 70 seconds but we said that tends to suggest that what 6 caused the first trip is different than what caused the 7 second trip, and to the extent that you believe that, then 8 as I understood it, the testing you did showed that 80 9 seconds is the time period you would expect to have if you 10 had the first trip caused by high --- low jacket water 11 pressure. 12 MR. OWYOUNG: No, the 80 seconds could also be if ( 13 you had a vibration trip or if you had a turbo oil trip or 14 any of the other trips. l 15 MR. CHAFFEE: Okay, then it still exists. The 16 thing that caused the first trip, whatever that was, may 17 still be in place. 18 MR. BURR We don't know what caused the first 19 trip and it could be the fact that it was just the first i 20 trip and you're filling extra air lines. There's a lot of 21 eomsibil), ties. 22 MR. CHAFFEE: Okay, let's assume it's part of 23 that, maybe you're filling some lines and stuff. Now the 24 next question that would be of interest is the testing (. 25 you've done to date has included an air roll of the diesel. l 92 PROJF.CT 035268

51 1 Now is there a possibility that the next time you have a i ~ 2 call upon for this diesel performance function as it has run 3 for 28 days, that you're going to find that low jacket water 4 pressure is going to cause the diesel to trip out? I 5 MR. BURR: I wouldn't think so. We've got nothing 6 pointing towards that. ~ 7 MR. CHAFFEE: Except that the first trip may have 8 been caused by that. 9 MR. BURR: It could be caused by other things 10 also. 11 MR. CHAFFEE: Right. 12 MR. KOCHERY: Also we tightened the lines. i 13 MR. BOCKHOLD: The key thing here though in this 14 discussion in my mind is lesson learned number 4, to go 15 ahead and bypass the non-essential engine trips on under-16 voltage start. 17 MR. CHAFFEE: Okay, so then I don't disagree with 18 that and if that's what you're going to do so that this 19 thing is not a problem before you declare the diesel fully 20 operable, I can understand, but that's going to be a study. 21 You're not going to do that today. 22 MR. BOCKHOLD: We plan to do that over the next 23 week or two. The design mod is in process right now, the ( 24 design evaluations are supposed to be complete and on site 25 by Wednesday. 92 PROJECT 035269

52 1 MR. CHAFFEE: Okay. What you do with the jacket i 2 water pressure sensor in terms of making sure in your minds 3 that it's operable is really between you and the Region, so 4 I guess I'll let you guys discuss that. What I'm heari') is-5 that what caused the first trip is not for certain~known. 6 It's a possibility that it was caused by low jacket water 7 pressure and certainly the timing out of that would be 8 consistent with that. I guess what we also' don't know is -- 9 do you know whether or not the 80 second run for the first 10 trip, would that be consistent with other trips or do you i 11 need -- based on the testing you've done to date, are there 12 other specific trips that could cause trip to occur within { 13 80 seconds? Do you have any tests to determine that sort of 14 thing? 15 MR. OWYOUNG: The jacket water temperature trip P 16 goes -- and the lube oil pressure trip goes througa a 17 different circuit than the other trips and my estimate is 18 that the other trips going through a totally different 19 circuit has additional time delay. 20 MR. CHAFFEE: So then you're thinking any of the 21 other trips could have brought it in in 80 seconds? 22 MR. OWYOUNG: Possibly. l 23 MR. CHAFFEE: I guess -- 5 24 MR. OWYOUNG: And again, there's no indication of ( 25 that. l 92 PROJECT 035270

) i 53 1 MR. CHAFFEE What's true than?h is the first trip l 2 was not probably the same scenario as the second one because 3 the time was different. Well then I guess the question is 4' to what extent do you need to go out and try to determine i 5 what caused that first trip. It's fortunate that'you got a j 6 good reading on the second one, but how do you rationalize 7 not taking further action to identify what caused the first 8 trip? What do you hang your hat on? t 9 MR. BOCKHOLD: We -- l 10 MR. CHAFFEE: I guess that's operability, but I 11 guess from the IT's perspective -- 12 MR. BOCKHOLD:. You know, we have checked all the t[ ' 13 sensors and we've done all the logic testing and we've done 14 under-voltage testing and we've started the diesel multiple 15 times, we have tightened air fittings and basically with a 16 team of engineers and technicians, the Operations people put 17 the A and the B diesel under a microscope and tested it as 18 much as we could technically figure out how to test it and 19 basically brought both of those engines to what we feel is a 20 state of readiness to run in an emergency. 21 We looked at and cannot get enough data to 22 definitely tell you why the engine tripped on the first l 23 trip. 24 MR. CHAFFEE: Okay. Then is there anything that ) 25 can be done from the standpoint of going back and looking at 92 PROJECT 035271 i

54 1 the diesel and the configuration it was in at the time when 2 the diesel was called upon to perform its function that can 3 be used to hypothesize the most likely failure -- thing that 4 caused the failure, recognizing that one thing that is [ 5 different between the diesel at the time of the event and 6 all the testing you did is that it was in a state of being 7 secured for a much longer period of time than was the case 8 when you did your testing. 9 NR. BURR: We also have one additional thing that 10 has changed. We had several leaks in our. tubing, the lines, i 11 that we found after the incident. These leaks have been 12 tightened and repaired so we do not have those leaks now. ( 13 MR. CHAFFEE: Do you believe that the leaks that 14 they found in all of this were significant enough that they 15 could have had any impact on the -- 16 MR. OWYOUNG: I wasn't here to see what leaks were 17 tightened up. 18 MR. CHAFFEE: But what you've told us is you'd 19-have to have leaks on the order of the.006 orifice before 20 it could have any impact on the -- 21 MR. ONYOUNG: Yes. 22 MR. CHAFFEE: Why don't you describe to him what 23 that -- I guess you guys know what that looks like -- are 24 there people here that understand what a.006 leak looks 25 like and were involved in the leaks that were identified I PROJECT 92035272 T- --*-- +m 'c-y w -e. m .ea2-i-- = y yu .>--s

i 55 1 guess on the 23rd or something -.whatever the dates were -- i i 2 to tell whether or not the leaks that was observed was 3 significant enough to have an impact on the operability of 4 these -- 5 MR. BURR: Not the operability but it may explain 6 the additional time that that first trip required. If you 7 -have additional leaks, it will take a little longer to fill f 8 lines. 9 MR. CHAFFEE: Yeah, but why wouldn't you have also 10 seen it on the other one as well. 11 MR. BURR: Because the linsa are pressurized, they 12 haven't been sitting that long. (( 13 MR. CHAFFEE: Oh, I see what you're saying. Is 14 that possible? 15 MR. OWYOUNG: It's possible. I would have to 16 analyze what lines are possibly filled and not filled. 17 MR. CHAFFEE: Well then maybe that could be an 18 avenue to try to explain that. All I'm saying is I'2 19 requesting that you don't drop figuring out what happened to 20 the first trip. I understand that more testing isn't going 21 to help, I don't disagree with that, but I want to make sure 22 that you've exhausted trying to analyze the cause of the 23 first trip through things like this. I'm not sure what else 24 there is, but to the extent possible, I want you to try to 25 get a reading on that so that we can understand what caused 92 PROJECT ' 035273 l

56 i 1 that first trip. If it turns out it's what you said, that's 2 great, then let's try to figure that out and if that's the 3 case, then we'll know that and this phantom will cease to i 4 exist. 5 MR. BOCKHOLD: All we can do is we can have the 6 team stay together today and postulate on what could 7 possibly cause the first trip, with the prints out and that 8 kind of stuff and I can commit to do that. And will commit 9 to go do that. 10 MR. CHAFFEE: Okay. I guess that's it then. 11 Anybody else have anything? 12 MR. DIETZ: One question. The circuitry for these d 13 other trips is a different circuitry of the air system to t 14 give you the trips. 15 MR. OWYOUNG: Yes. i 16 MR. DIETE: Is it worth putting one of those trips 17 in and measuring that time? 18 MR. BOCKHOLD: They did. 19 MR. DIETZ: What did that and up being in time? r 20 Is that close to the -- 21 MR. BOCKHOLD: Ken. { l 22 MR. OWYOUNG: When we did the low pressure jacket i 23 water trip, it was about 79. 24 MR. DIETZ: Which is close to the 80 that was seen 25 on the first trip. 92 PROJECT 035274

~ 57 1 MR. BOCKHOLD: Yes. I think earlier we said that, I 2 you know, one of the other trips could have caused the first ) 3 engine trip and the question I think that was finally asked j 4 was which one do you think is the most probable one. okay? 5 And we'll look at the prints but we may at the and after we 6 look at the prints and brainstorm it awhile, may only 7 conclude that one of the other trips could have caused the 8 first trip. 9 MR. CHAFFEE: So basically where we're at is that 10 either the first trip was caused by the same thing, which is 11 the air filling up which would make sense, so really you 12 don't have another phantom sensor problem, or if that's not 13 the case then with all the starts you've done, that tends to 14 suggest that the sensor is okay and that the problem was 15 somehow the diesel being secured for a longer period of 16 time, somehow allowed it to be a problem in the first. If 17 that's the case, then we have the phantom concern is well 18 the next time the diesel is called upon when it's been shut l 19 down for several weeks, are you going to then see the 20 problem that resulted in the trip. 21 I hope that this explanation that you've come up 22 is the one that'll -- 23 MR. KOCHERY: See, we don't have any more leaks on 24 the pneumatic lines. 25 MR. CHAFFEE: So there's two things then. 92 PROJECT 035275

58 1 Hopefully what's the case is the leaks in the lines combined I 2 with the time it was -- oh, I see what you're saying -- 3 hopefully the leaks in the line explain why the first trip 4 was 80 seconds instead of 70, because then there won't be 5 any phantom sensor problem or there won't be a sens'or 6 problem in the sense that maybe you'll see the problem the 7 next time the diesel has been shut down and not run for a 8 long period of time. 9 MR. KOCHERY: That's right. 10 MR. CHAFFEE: I'm with you. 11 MR. BOCKHOLD: George wants to say something. 12 MR. FREDERICK: When we ran the UV test on the 1-A i 13 diesel, when we started this test sequence, I believe it was 14 on the 29th, the first UV test we ran, that was almost the 15 exact amount of time the diesel had sat idle from the major 16 maintenance that we did until the 20th. l 17 MR. CHAFFEE: Right, except that before you did 18 that test, didn't you roll it? 19 MR. FREDERICK: No. We tightened -- the only 20 thing that happened was we tightened the fittings and I 21 believe they replaced some lube oil pressure switches. 22 MR. BOCKHOLD: The air roll doesn't charge the 23 controls is what I've been told. The air roll does not 1 /' 24 charge the controls so the controls were sitting in the same ( i 25 -- we air roll the machine to remove any condensation and 92 PROJECT 035276

59 1 where there's a concern about a jacket water leak into a 2 cylinder head and you'll damage the cylinder head unless you 3 air roll it. So we air roll it as a good practice for 4 engine integrity. 5 MR. CHAFFEE: But when you air roll it, doesn't -- 6 MR. BOCKHOLD: Does not charge the control lines 7 with air, so the control system is sitting there in sort of 8 the same time frame as the first -- well as on the event 9 day. 10 MR. CHAFFEE: But does the -- 11 MR. KENDALL: And no fittings were tightened 12 between the time? i[ 13 MR. BOCKHOLD: No, we tightened fittings. t 14 MR. KENDALL: So it's not exactly the same l 15 condition necessarily -- close but not as much air may have 16 leaked out. 17 MR. BOCKHOLD: Yeah, you know, the fitting 18 tightening. i 19 MR. CHAFFEE: Does the air roll have any impact on i 20 this jacket water low pressure, does it in any sense somehow 21 pre-prime the jacket water -- not the pneumatics but the i 22 jacket water system itself such that the -- 23 MR. BOCKHOLD: Let's ask Cooper that. Let me ask 24 it and you correct me if your question is different. Does iL 25 the air roll with the shaft driven pumps cause a temperature 1 92 PROJECT 035277

60 1 change or a pressure change that some of the sensors might ( 2 see something differently? i 3 MR. CHAFFEE: You know, when it subsequently 4 started. i 5 MR. OWYOUNG: I would say that the air roll is of 6 such short duration that you wouldn't see anything. 7 MR. JOHNSTON: I don't think you could measure 8 anything. 9 MR. CHAFFEE: Okay, well then what that would tend 10 to suggest is hopefully you're onto it, which is that the 11 leaks that were in those lines are enough to account for the 12 ten second difference and that the first trip is the same i( 13 thing that caused the second one. It'd be nice if it turned 14 out that way. 15 MR. BOCKHOLD: In our report we will put down what 16 we consider as the most probable cause for the first engine 17 failure or we will -- you know, we'll say that we can't say 18 for sure but we postulate this scenario for the first engine 19 failure start. i 20 MR. CHAFFEE: Okay. Is that it? Does anybody 21 else have anything? l 22. (No response.) l 23 MR. CHAFFEE: Okay, thank you very much. l 24 (Whereupon, the meeting was concluded at 10:00 25 a.m.) 92 PROJECT l 035278

61 1 CERTIFICATE t 2 3 This is to certify thac the attached proceedings before the 4 U. S. Nuclear Regulatory Commission in the matter of: 5 Name: DIESEL GENERATOR MEETING 6 7 Docket Number: 8 Place: Vogtle Nuclear Generating Plant, Waynesboro, GA 9 Date: April 2, 1990 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear 12 Aegulatory commission taken stenographically by me and, ([ 13 thererfter reduced to typewriting by me or under my 14 direction, and that the transcript is a true and accurate 15 record of the foregoing proceedings. 16 17 Arn 0 _ m/Ap 6 ( Q (f ~ 18 PEGGY J. WARREN 19 Official Reporter 20 21 Ann Riley & Associates 22 23 24 I(g 25 92 PROJECT 035279

1 /56! meetino Atiendene n. cord Georgia Power 2 ~ Meetung PurposeOleg/ &k Mib. ,t-ZZ 7" Exist Date/ 91 Mr. VA/ C4Ac - p% 6L-Conouctec ey h Name (Print) Title (Printi U O DepartmentrCompany liikl brl /2J G.L L-) i p e 2. gpc /sp3 %',k!I$orlod Mor-Eno Onb 312I CNG /G L b AA. M$$f xD% (/C hilenL.flodog ERviE TouPlp c?l'C REF 39'o/ 0 6 LErNo,<M bales Smr~ 5CS G u ci o s e w e. H88 Scs d/r -WW O i P' Ms' Ft-yzce or'P/6<t. 'Ala,rK S. b n n {l Is'C Lw;n tuwt (Adid) 3247 111&n6/IiC/ 6it-Paau Oupp se. O tr G a- .%&4 ":'XSEG /ar-GEORGc FfEWA/Ch Syst - 5,+c7d. %3 2 ?. 3 G./C, l d osCP& [ WA M t C<> $ a pl - 0 W f 1l31 6 Eh f $18_. C. 20 vr'lL %,wn,8x.Assdd 39/h % msg *EP &n A0lma 1%. TNil,

• EP 390/

os R4s< r dow&J

AlWl, twN

~3 D h /rPC. I Melwh i fe$ BuRe %. Pps; ed Ew. '7Aa c NMslsosoxdB%.) v i lAlr.Ab E flx.1 S whNr $/d$ $1.Sss2 $?d m / N b hW D DQbD5 Y T, yak lN

• fdkEN Ener,'/GK Ken Stokes se, Nn4 E nge,

m3 ass) &m &u Oud/naat. & Sosm a Ec e.d.. s,n tst t. 00cPent. IuDusidin (4,1 D M kf SDUT b O tic O(.,S M L f,14-70I h _ ~1.ODuS7'l W i ms komtRT "c '3 T o c b Tett E v. C i c E n W*MS CooPC A Y"OC%DO A, D.SYMkrY AFsiDr O Z~W sera X 4'2 W

1. /c

/u Bac4,- seaw M. E,,, sn94g ac/GPc {U2t Y~13 $7"lN mf3*M/m Mar. Pi.iser Annin 3ll 3 &K a 92 PROKCT 062516

Meeting Att:ndanco Rectrd GeorgiaPower A ~ .....n ~,. Date Conductec Oy 7 Hens Name ipnnn Ltie(Pnnu

1. ([ f,;r; Department / Company M

/) h M.8 [- ~ 'd% b ockmne klmC RfS30$9Ir 4 Nec-4r $6 A' m 5 bhc t ~ e Nu /nxx u P^ut t4ko cacay Pwr Ewcc. SupvR sues GPt / EwGc.SdMT l GEoccE ($ocenac> lyP/ 3ill GPC-. ~ WMR1H 1 Vm/ 5% KfA~r Dv4 Nxc/Mx?? Piu) E. D;etk. 117 M+mls+ ZAJPo E. 4 ltA6a LET~ VIEH8EA LK Nair \\ koM / / 7 />Wesa A.)BG i /.//M4&t bbMVd2s' //r- /9%w /W h C//d/3 CSS //7-W CW /VO ~ SnciA k0t W lY $$ DM UC Pohn Aohtccd h at re s b w k oo u L,a,a /co h % L tla d - a AntGHPl.S&oet w3 GPc v hhd NQ y hhd hpo b b kg O .72. Sarwet BEE Ops *Plae- %d3 d5/GM i Mahdi % i k,* Se E:y 32 M u5+c /s/e l i i 92 PROECT 062517 a --w. .-.n

/, GPC EXHIBIT H BOCKHOLD N ITSH19153 i LUBE OIL HIGH TEMPERATURE SWITCH DG1B Prior calibration was performed on 3/14/90. At that time the switch was found out of tolerance with an as found of 300 *F to trip and 199 *F to reset. It was calibrated and returned to service with a trip of 199 F and a reset of 191 *F. It cwitch calibration was checked and would not calibrate within tolerance. p. The was removed from service on 3/23/90 as the suspected cause of DG1B tri l It was i placed in storage on 3/23/90. SWITCH TWO This switch was placed in service on 3/23/90 with a trip of 203.4 *F and a reset of 198.0 *F. On 3/27/90 it was removed from service and it's calibration chec ked. As found was 203.5 *F to tria and 199.5 degrees to reset. However,it was found to be venting continuously and subsequently replaced. The old switch was placed in storage on 3/27/90. 1THS19146 LUBE OIL HIGH TEMPERATURE SWITCH DGIA Prior calibration was performed on 3/3/90. At that time it was found out of tolerance with a trip point of 211.0 *F and a reset of 203.1 *F The switch was calibrated and returned to service with a trip point of 200.2 *F and a reset of 193 degrees On 3/30/90 the switch was removed for calibration and found out of tolerance sluggishly. point of 190.4 *F and a reset of 188.0 *F. The switch also owratedIt was replaced with a new switch cali' orated to with a trip anc, a reset of 196.20 *F. The new switch was returned to service. The original switch is in storage. 1TSH19117 JACKET WATER HEATER OUT HIGH TEMPERATURE SWITCH DG1B Prior calibration on 3/14/90 was within tolerance with a trip point pf 201 *F and a reset of 193 'F. On 3/26/90 switch was found out of tolerance with a trio reset of 182.4. Further investigation determined a small leak. point of190.6 and a A new switch aso failed leak test. A third switc1 calibrated correctly with a trip noint of 200.67 $ egrees and a reset of 196.93 'F and was returned to service. The o'ld switches are l in storage. ITSH19119 JACKET HEATER OUT HIGH TEMPERATURE SWITCH DG1B Prior calibration on 3/14/90 was within tolerance with a trip point of 200 *F and a reset of 194 *F. On 3/26/90 the switch was found out of tolerance with a triopoint of 188.2 *F and a reset of 180.6 *F. Further investigation determined a small Teak. A new switch was ino wrable due to a missing gasket. A third switch was calibrated with p trip point of :.98.57 and a reset 191.90 and returned to service. Old switches are in storage. 92 PROJECT oo621s

4 i TEMPERATURE SWITCHES 1TSH 19153 LUBE OIL HIGH TEMPERATURE SWITCH SWITCH 1 DG1B Ori 'nal calibratiort 3/14 S 0. Removed un replaced as cause of ciesel tip on 3/2300 by 1 (O 19001482. h calibrati within tolerance.on was c 2ecked anc, woni not calibrate SWITCH 2 Switch was removed again on 3/26/90 by MWO ;.lM)015,11 for calibration check. It was found to vem ua.y anc, was replacec;. This switch was fun:tio koc o,n 3//',S/90. The trip and reset aspeared to ')e sa but would not i fully close upon resei Placed in storage 3/2 1TSH19146 LUBE OIL HIGH TEMPERATURE SWITCH DG1A Original calibration 3/10/90. Trio point was, b MWO 19001629 on 3/2440. SwifcN appeared sL a was replaced. Switch was an has not been checked again.placec. in storage on 2 ITSH19117 JACKET WATER HEATER OUT HIGH TEMPERATURE DG1B SWITCH 1 Original calibration 3/14/90. Calibration atte waq, maie at 13 cation of engine. Replaced due to.eakage urma subb.e testinr. Switcli was re ' ced tcE 1 w placec. in sto on 1.w u tW wi vendor and fo to o ry anc, was return gge on 3/27/90. perate sa l The replacement for the above was found to leak by The valve was replaced again and us not been cwkd placed in storage on 3/27/90. l I j i l l l 92 MtOJECT l 006219 i

1TSH19119 CACKET WATER HEADER OUT HIGH TEMPERATURE DG1B Original calibration 3/14/90. Calibration a was made at :ocation of the engine. Replaced due to bubble testing. Switch w p aced on 3/26/90 in storage at that time. switch was wi; vendor and found to be satis ctory on 3/29/90 an return to storage. ITSH 19111 JACKET WATER MAIN HEADER OUT HIGH TEMPERATURE DG1A Originall and found calibrated on 3/10/90. Ca;ibrated a(Wn o to be tri 19001683 on 3/31Mped This swites was rep acedand placed in storag not been retested. ITSH19112 JACKET WATER MAIN HEADER OUT HIGH TEMPERATURE DG1A Original calibration 3/1050. Calibeition check on 3/30/90 i and found tripped. It was re 19001863 and placed in storage on 3/31/90. placed by MWO PRO ED TEST FOR CALCON TE RATURE SWITCHES With units ou", of temperature bath pressure test with 60 PSIG air throug,B tace to.0P,8 inch orifice. :f units leak at room temperature, we wi rti.qn of temperature sensor apart "4 access valve area. ~ le anc test. pect valve and will t sen ins seat, clean as req ~ .,reasse If we find n e a cs at mgera . We will in temperature ts trip reset. switches do not co we root cause mh.e insf.ect tile entire swa to attemp,t to i en of tu ma function t [ 2 8 92 PROJECT 006220

i PRESSURE SWITCHES l 1PSL4903 ' N NORMAL TRIP 7-3 / DGIB k This sensor c(2 three (3) times Oririnal calibration 3/16/90. was removed using MWO 19001542. and then was placed in storage on 3/2 1PS4749A I LOW PRESSURE LUBE OIL DG1A 7 Original calibration 10/21/88. Sensor was removesl and che~ked on 3/30/90 by IWWO 19001433. Switch wou d not e reset, was replaced, and placed in storage. IPS4749B k. LOW PRESSURE LUBE OIL l DG1A Original calibration 10/2 On 3/30/90 Sensor was a chec1ved and checked unc.efa O 19001433. The sensor l%' rem ed satisfactory, but rep and placed in storage. /\\ 1PS4749C 4O LOW PRESSURE LUBE OIL DG1A ' (/' Original calibration 10/21/88. See 1PS4749B. h 112403G4002 PANEL LOGIC BOARD DG1B While working MWO 1900 a trip could not p uced when a sensor trip was si The logic was found c.efective, replaced, in storage on 3/27/90. PROPOSED CALCON PRESSURE SWITCH TEST dtTc ' thio"h.b E check tse switchE' we wil asemEe, inspect the i orifice. valve anc, seat area'does t ot : clean' ar, .d m) e. If t2e swite: t we wil wrform a normal calibration procec;ure to t a swi wk the trip point and erall resoonm. If the not aspond correc we will risassemble inspatc} thq swites to i attemp lo ic.entify t se root cause o malfunction. LOGIC BOARD TESTING DG1B No further testing of the DG1B logic is planned at this time. 'E PROJECr

Oosgy, i

2 r- /(tr l'b-9c GPC EXHIBIT 67 BOCKHOLDO OFFICIALTRAWKRIFTOF PROCEEDINGS g Nuclear Regulatory Commission Teleconference Between IIT,

Title:

Licensee Personnel, and Region II Personnel Docket No. Bethesda, Maryland Wednesday, April 4, 1990 1 - 67 ANN REEY& ASSOCIATES,I.TD. 1612 K St. N.W. Suite 300 i Mshinpon,D.C 20006 (202) 295-3950 M MOJECT 061973

1 1 UNITED STA'fES OF AMERICA ) 1 2 NUCLEAR REGUIATORY COMMISSION 3 4 - - - - - - - - - - - - - - - - - -X 5 In the matter of 6 TELECONFERENCE BETWEEN IIT, 7 LICENSEE PERSONNEL, AND REGION II 8 PERSONNEL 9

---

X 10 11 Nuclear Regulatory Commission 12 7735 Old Georgetown Road i

13 Conference Room 14 Bethesda, Maryland 15 16 Nednesday, April 4, 1990 17 la The above-entitled matter commenced at 10:07 a.m., i 19 when were present: 20 t 22 23 24 25 92 PROJECT 061900 l

2 1 ON BEHALF OF IIT l' 2 3 RICK KENDALL 4 AL CHAFFEE 5 HARVEY NYCKOFF I 6 E 7 ON BEHALF OF GEORGIA POWER iLICENSEE): 8 9 GEORGE BOCKHOLD 10 XEN McCOY 11 12 ON BEHALF OF REGION II: i 13 14 KEN BROCEMAN 15 16 17 18 19 t 20 21 22 23 24 25 92 PROKCT 061981

3 1 PROCEEDINGS I 2 f10:07 a.m.) 3 MR. CHAFFEE: We're going to be transcribing this 4 thing. t 5 MR. BOCKHOLD: A1, as a separate issue, before we 6 get to that, would you like a copy of the critique of Region 7 II's response? 8 MR. CHAFFEE: Yes. 6, 9 MR. BOCKHOLD: I will forward that on to you, to 10 the fax number that I got yer' iday. 11 MR. CHAFFEE: Okay. 12 Okay. Let's see. Where are we going to go with 13 this? 14 A couple of things -- since we last talked -- 15 maybe I should ask you guys first, George. Since we last 16 talked, is there anything that you guys have done, from 17 where you were yesterday, in terms of looking into the kind 18 of troubleshooting you want to, that you want to talk to us 19 about before we start giving you some of the thoughts that 20 we have? l 21 MR. BOCKHOID: You want me to recap where I 22 thought we left of yesterday? 23 MR. CHAFFEE: Okay. That's probably a good start. 24 MR. BOCKHOID: Okay. Yes. We talked about a 25 center calibration yesterday, and we told you we would send RWEcr og,4

4 1 you some more information about the various sensors before 2 we started calibration, and we sent you some more 3 information, and at that plant, the two sensors that we 4 desire to test next, and we are pulling together the rest of 5 the information on the other sensors. Our Isc people are 6 working on that, to get the best information, and we will 7 telecopy you that information. In fact, sometime this 8 morning, that information will be available. 9 After you receive that information, you've 10 committed to us that you'll review it and, in approximately 11 an hour or so, get back to us and tell us that we could go 12 ahead and start testing on the two sensors we desire to test i 13 on. 14 MR. CHAFFEE:s Yes. We'll review it, and we'll get 15 back to you, unless we have some problems with it. 16 MR. BOCIGIOLD: Okay. 17 MR. CHAFFEE: Okay. Well, in any case -- is MR. BoCKiloLD: And that talks to the things that 19 are on the quarantine list. 20 We also told you that we were going to go ahead 21 and monitor jacket water temperature on one of the engines 22 over, basically, a 1-day period of time, and then we'io 23 going to start the engine without an air roll and see what 24 happens to that jacket water temperature. 25 We expect that when we go through this, we will be l l 92 PROECT W

~ ~ ~ p 5 1 able to pick the maximum jacket-water temperature point and 11 2 a start at that particular time, we believe, will have some 3 variation in jacket water that that sensor could possibly 4 see, but at this time, we believe it would not be high 5 enough to trip that sensor. 6 MR. CHAFFEE Okay. 7. MR. BOCKHOLD: Okay? But that's not quarantine, 4 and when the procedure was approved, we were going to send 9 it to you and go ahead and start into that. 10 MR. CHAFFEE: Right. But we want to reciew it 11 before you guys do the test. We want to make sure we agree 12 with your approach and have an opportunity to comment on it. .I 13 MR. BOCKHOLD: We were to install the equipsent 14 right away, because it does not in-op the diesel when wo 15 have set this test up, and we do not want in-op our diesels, 16 except for the one change that we plan to make associated 17 with the ~~ making the under-voltage start the same as a la safety injection part for the engine - not for the 19 generator section but for the engine. 20 so, we would go ahead and install this 21 instrumentation. We do not plan to run the engine for, 22 basically, 26 hours, unless we needed it for something -- 23 you know, unless we had some type of blackout or sosething 24 that needed to comply with tech specs, and we would -- we .I 25 could receive your comments anywhere between that timeframe. 92 PROJECT 061984

6 i 1 So, do you think you can comment within that I-2 timeframe, if you have any comments? 3 MR. CHAFFEE: I'm sorry. Say the timeframe one i 4 more time. 5 MR. BoCKHOT.D Approximately 24 hours. We'll go 6 ahead and install this instrumentation on the engine. 7 MR. CHAFFEE: Yes, I think we can do that. 8 MR. BOCKHOLD And you know, the data that we plan 9 to collect is we plan to watch jacket-water temperature at 10 the high jacket-water temperature trip sensors for 11 approximately 24 hours, just to see what a daily variation 1 12 tags for that temperature. i 13 MR. CHAFFEE: And that's with the diesel secured? 14 MR. BOCKHOLD: That's with the diesel just sitting 15 there, in standby, reading to be started, fully operable. 16 MR. CHAFFEE: Okay. 17 MR. DOCKHOLD: Okay. And then we plan to, is basically, trip start on the engine and simulate the 19 starting period that occurred on the 20th. Okay? Excuse me 20 -- we'll do three starts -- simulate the timeframe that the 21 starting sequence occurred on the 20th and, basically, 22 observe jacket water. 23 We expect that the way we will pick the jacket-24 water temperature, we will expect that the sensors would see i 25 a higher temperature than an average start, let's say, but j l 92 PROJECT 061985

_.. _ _ ~ i 1 we expect that that temperature would'be, maybe, the most -- I 2 and I'm just speculating at this time -- 10 degrees higher. 3 We expect that this test will not turn up anything different 4 except for that if one of the sensors was intermittent or 5 had a poor calibration that if the right jacket-water -- if 6 we started it during the right time of the day, it might be 7 sore likely to trip. That's what we expect to conclude out 8 of this test. 9 MR. CHAFFEE: Okay. I understand. 10 MR. BOCKHOLD: In the meantime, what we're doing 11 is after the electrical system on the other train of the 12 diesel is made fully operable, we plan to go ahead and j i 13 install our modification that makes the engine trips on loss l i 14 of power or under-voltage the same as if the engine were 15 responding to a safety injection. 16 MR. CHAFFEE: So, you have completed your 17 engineering evaluation, and then you're actually going to do 18 that. 19 MR. 80CKHOLD: We have completed our engineering 20 evaluation. We're going to proceed and do that, and that 21 modification is ready to be implemented. We've got a few 22 more reviews to do today, but it would be implemented, 23 probably, this evening, on one of the engines. 24 MR. WYCKOFF George, this is Barvey Wyckoff. g 25 I do have a question on that, just to learn. a M

i 8 1 That entails two things: Number one, it bypasses 2 the trips, and that's the thing you really want to do. 3 Will this also result.in an unlimited start, or 4 will it still have the 5-second start feature? 5 MR. BOCKHOLD: It will make the engine start the 6 same way as if the engine was going to respond to a safety 7 injection, which is an emergency start, which the engine S will roll until it fires or 150 pounds is reached and 9 received. 10 MR. WCKOFF Understand. 11 MR. CHAFFEE: Does that modification -- is that 12 consistent, just some electrical rewiring, or does it result 13 in any sort of pneumatic changes? 14 MR. 80CKHOLD: It's electrical. There is no 15 pneumatic changes. 16 MR. CHAFFEE So, there are some pneumatic 17 changes. 18 MR. WCKOFF There are none. 19 MR. KENDALL: No, there are none. 20 MR. BOCKHOLD: Changes aren't in the change. 21 MR. CHAFFII: There's no pneumatic changes; just i 22 electrical. 23 MR. BOCKHOLD: Just electrical. 24 MR. CHAFFEE Okay. g 25 MR. BOCKHOID: Basically, it's moving the wires 92 PROJECT 061987 i

t I 1-that come in for the under-voltage to move them to the 4 -2 emergency spot. Basically, that's all it does. i 3 MR. BROCKMAN: Which. diesels are you planning on i i 4 doing this for? t 5 MR. BOCKHOLD: We will do this on 1-A and 1-B this l l 6 week, both Unit 1 diesels this week, and next week, we will { 7 enter 72-hour I40s and do it on Unit 2 then. 8 MR. McCOY Ken Brockman, this is Ken McCoy. 9 Imt me be sure you understand. I think the IIC 1 10 all understands, but you may not have heard all of these 11 sessions. 12 There is another portion or another design mod j 13 that we will have to complete that is a larger design mod, i 14 because it involves pulling some cables to get the 15 electrical protection grid to be the same on an under-16 voltage as they are on a safety injection. i 17 MR. 80CKHOIA: I understand the differences. 18 MR. McC0Y: Okay. And we intend to do that at the 19 next refueling outage on each unit, but everybody agrees 20 that -- at least, it's my understanding everybody agrees 21 that those particular protection devices are highly 22 reliable, and the only reason for making the second change 23 is really simplification to keep the design of the 24 protective devices on the engine as simple as possible for 1 25 operations, understanding training, etcetera. 92 PROKCT 061988

P 10 1 MR. CHAFFEE: Who is talking right now? 2 MR. McCoY: This is Ken McCoy, Vice President of 3 Vogtle, and I'm in here with George Bockhold. 4 MR. CHAFFEE: okay. Thank you. 5 MR. noCKHOLD: I understand, Ken. Thank you. 6 (Phone rings.] 7 MR. CHAFFEE: Just a second. e okay. Let's see where we're going. 9 okay. A couple of questions: In thinking about 10 what happened with the diesel, there was a couple of 11 ' thoughts that had come to our mind that I just want you to 12 hear so you can patch them into the investigation that 13 you're doing. 14 one of the things tnat still perplexes me -- this 15 Al Chaffee -- and we don't know the answer to it yet -- is 16 why the monthly surveillances have not revealed any of this 17 problem, and we only seem to be seeing it in terms of a high 18 jacket-water temperature or sensor problem coming out of l 19 these outages. I 20 So, in trying to look into that, one of the 21 questions we asked ourselves is, you know, to what extent is 22 the monthly surveillances being done somehow differently t 23 than the kind of testing that you do coming out of the i 24 maintenances, recognising that even in the case where you i 25 did your post-maintenance testing, you didn't find these P l

11 l'- problems until the event occurred, at least; your testing 2 wasn't comprehensive. 3 And one of the things that we thought about was 4 the fact that the nuclear service cooling water system is I 5 operational when you do your monthly testing, but in the 6 event, and when the diesel is called upon to perform its j 7 function, that nuclear service cooling ' water system doesn't a get powered up until the diesel has sequenced on, and so, 9 one of the thoughts we had is since it would, therefore, not 10 be available, is it possible that that further contributes 11 to the fact that the jacket-water temperature is actually i 12 higher in the case where the diesel is called upon to 13 perform -- when you have a dead bus, as opposed to the case 14 i when you do your monthly surveillances or, in fact, the j l 15 surveillances that you did coming out of the outage. 16 I don't know if that has impacted, but we thought 17 that was something you ought to think about. 18 MR. BOCKHOLD: My immediate reaction to that, if i 19 you'd like it -- let's think about it some more, but I can 20 give you my immediate reaction to it. 21 MR. CHAFFEE: Certainly. Yes, go ahead. 22 MR. 80CKHOLD: My reaction to it is on actually i 23 the 20th, you know, we started the engine the third time, 24 and it ran, okay? Each time, you know, the jacket water 4 25 probably was warning up a little bit from engine heat input. i CT 061900

12 1 NSCW had been sequenced on, but it van sequenced on a while i lf 2 after, you know -- NSCW was one of the last sequences done. 3 I believe it's about 55 seconds after the engine starts. 4 I'd have to look that up. I. 5 So, my gut reaction is that NSCW not being i 6 immediately on didn't effect the actual day on the 20th of 7 diesel starting. Also, subsequent to that, as you know, we 8 did an under-voltage test of the engine, and basically, the i 9 under-voltage test trips NSCW, so NSCW is not on, and 10 therefore, we did under-voltage on the A-Train as well as 11 the B-Train. On the A-Train, we did under-voltage testing 12 twice. On the B-Train, we did it once. So, we basically y 13 simulated the event as far as NSCW is concerned, and 14 therefore I would say that that has no effect on the diesel 15 because of the testing we've done. i 16 MR. CHAFFEE: Okay. But you'll take a look at it 17 also, right? i 18 MR. BOCKHOLD: We'll think about it some more. 19 MR. CHAFFEE: Okay. l 20 The other thing we were thinking about was trying j 21 to sort of put together -- maybe there's a combination of 22 effects that's causing this phenomena with the diesel 23 tripping and high jacket water temperature. So, again, just i 24 for your information, so you know the kind of thinking we're I 25 doing, let me throw this out and we can go from there to 92 PROJECT 061991 i l

13 1 discuss it. As I -- and this is based on our understanding, 2 and I realize that not all of what we have here is accurate. 3 My understanding is that the set point for the 4 jacket water temperature is 200 degrees plus or minus four 5 degrees. 6 MR. BOCKHOLD: That is my understanding. 7 MR. CHAFFEE: Okay. So then we said, "Let's 8 assume that the thing is calibrated, and it's at the lower 9 level of calibration so that it would trip at 196 degrees. 10 The other thing we were told, or at least wo 11 thought we were told, was that due to mechanical, you know, 12 hitting of the switch, or somebody holding in the stuff, 13 it's possible that that could effect the calibration by ten 14 degrees. 15 So, we said, "Well, maybe let's assume then that 16 when people were doing the calibrations, somehow the 17 instruments that got installed were, impacted 'so that really is they were now set down to as low as 186 degrees." Then wo 19 said, "Okay. Well, let's assume that the probes are at that 20 point. They're at 186 degrees." Then we said, "Well, you 21 know, what else can impact the calibration of these 22 switches?" 23 And then the other things we've been told by the 24 calcon fellow was that the rate of temperature change can 25 effect the set point for the Calcon switches. If I remember 92 PROJECT 061982 i

14 .1 correctly what he said, he said that if you increase the i' 2 temperature from one to two degrees per minute, you'll get a 3 certain set point. If you change that to two to five 4 degrees per minute -- in other words, doubling it -- the set f I 5 point will go down, I believe he said five degrees. 6 so we extrapolated that by the following logic. 7 We said, "Well, let's assume that the initial temperature of a the jacket water system is -- I think we said 150 degrees. 9 Let's assume that's what the keep-warm system keeps it at. 10 MR. BOCKHOLD: The keep-wara keeps it hotter than 11 that, though. 12 MR. CHAFFEE: What does it keep it at? One-13 hundred and sixty-five? 14 MR. BOCKHOLD: Yes. It's somewhere between 163 15 and'165. i 16 MR. CHAFFEE: Okay. And then let's assume that 17 you have this -- you know, they're talking about this column is of water being something close to 200 degrees where the 19 heater is? ? 20 MR. BOCEROLD: No, that's supposed to be 170 21 degrees, 175. Top, 175. 22 MR. CHAFFEE: Also that -- so where the heater is i 23 is supposed to be just at 175 degrees? 24 MR. BOCKHOLD: We're going to check it. It's 25 design number is 175 max. 02 PROJECT 061993

15 1 MR. CHAFFEE: Oh, I see. So the heater has a cut-i 2 off when the temperature gets up to 1757 3 MR. CHAFFEE What's.that? 4 MR. CHAFFEE The heater has a cut-off -- l 5 MR. BOCMOLD: One second, A1. l 6 (Pause.) 7" MR. BOCMOIA: The fact, A1, at this point, is 8 that we have a temperature controller that controls the 9 heater in that standpipe, okay, and it's really an on/off 10 type controller, and it's supposed to come on at 146 degrees 11 and go off at 170 degrees. The element that sets that 12 temperature is above the heater in that standpipe. 13 MR. CHAFFEE Okay. Well, then, that would tend 14 to suggest that you shouldn't have water in there that's 15 auch above the -- the max should be about 170 degrees. 16 MR. BOCMOID: Correct. The max should be 170 17 degrees. 18 MR. CHAFFEE Okay. Well, then let's just take i 19 that,.then. If it's 170 degree water in there -- well, let 20 se ask this question: When the diesel is secured and the 21 warm-up system is working, do you have any feeling for what 22 the temperature would be up near the sensor initially? I i 23 realize that's what your -- that's what your test is going 24 to go find out, right? 25 MR. BOCMOLD: Find that out for sure. You know, 92 PROJECT 061994

~ is 1 I would say the maximum it could be, because we've got a 90-2 gallon-per-minute pump pumping that water around, okay, the 3 maximum it could be is 170 degrees. 4 MR. CHAFFEE What would the minimum be? 5 MR. BOCKHOLD: A hundred-and-forty-six degrees. 6 MR. CHAFFEE: Okay. Well, what I'm interested in 7 and what I'm trying to find out is what's the worst case 8 rate -- what's the fastest rate of change of temperature 9 that the high water jacket temperature sensors might detect 10 when the diesel starts, because the faster the tempera' cure i 11 increases, according to the Calcon person, -- 12 MR. BOCKHOLD: We do not know that,'okay. This i 13 test that we're going to do is we're going to gather some 14 data associated with that. But let me give you another \\ 15 reaction that I have from the information that everybody's 16 given me, is that when the engine starts and its pain shaft { 17 jacket water pump starts, it's pumping roughly 600 gallons a 18 minute. Oh, 1600. A lot more than that. Sixteen-hundred 19 gallons per minute. A lot more than the numbers they had 20 told me previously. And NSCW, you know, comes on in like 55, 21 seconds, and that drives temperature down fairly quickly. 22 MR. CHAFFEE: Okay. When the shaft pumps for the 23 jacket water system kicks in, does it inject any slugs of 24 cold water from stagnant portions of the system? 25 MR. BOCKHOLD: Yes. i ECT 061 gag

17 1 MR. CHAFFEE Is it possible, then, that it sends 3 2 a cold slug through and then followed by a hot slug that i 3 causes a rapid -- so that you do have some rapid changes in 4 temperature to the -- 5 MR. BOCKHOLD There is a service bank control 6 drea associated with the -- NSCW supplies the heat 7 exchanger. l 8 MR. CHAFFEE Right. 9 MR. BOCKHOLD: So'you've got a valve there, and 10 depending upon the position of that valve is dependent upon 11 the temperature of the water flowed to the jacket area. So, 12 you know, that pump is bypassed normally, so its water is at 13 ambient. 14 MR. CHAFFEE: That's the shaft-driven jacket water 15 pump? 16 MR. BOCKHOLD: It's the section from the 17 standpipe, okay, which is -- you know, that particular is water, let's say 170-degree water, and it cups it thrcugh a 19 three-way valve that can either go into the heat exchanger 20 or bypass the heat exchanger and go back into the engine. 21 Initially, it should be bypassed. So, you know, it would 22 have a little bit of cool water in the dead leg between the 23 standpipe and that pump that would mix with the other water. 24 So immediately when that pump starts, just by reading this 25 diagram here, you should have a slight decrease in - 92 PROECT 061996 1

e 18 1 tesperature followed by the added heat input of the engine, I-2 an increase in temperature, and the three-way valve would 3 then try to control temperature. 4 MR. CHAFFEE: Oh. So the water goes through a 5 heat exchanger before it goes where the temperature probes 6 are? 7 MR. BOCKHOLD: It's bypassed first. The heat 8 exchanger is bypassed, and then it goes to the engine. Then 9 later, as temperatures change, the thermostat responds in 10 position to that. 11 MR. CHAFFEE Okay. Here's the picture I have in 12 my mind. The pump takes the suction from the standpipe. e 13 MR. BOCKHOLD: Right. 14 MR. CHAFFEE Then it jets it out and it can 15 either go through or around the heat exchanger. Then it 16 goes to the engine, and the water goes back to the 17 standpipe? 18 MR. 80CRHOLD: That's correct. 19 MR. CHAFFEE: Okay. Where in that circuit is the 20 jacket water sensors, temperature sensors? 21 MR. BOCKHOLD: The jacket water sensors are on the 22 inlet to the fan pipe. 23 MR. CHAFFEE: Oh, I see. So that what they see in 24 terms of the temperature for the trip, is that they see the 1 25 temperature of the water after it's gone through the engine. 92 PROJECT 061997 l

19 1 MR. BOCKHOLD That's correct. 2 MR. CHAFFEE: Well, what sort of things are going 3 on the engine when-it's starting up that would have an 4 impact on the jacket water temperature? I guess originally, 5 it's -- is the metal mass of the diesel originally -- it 6 must be, what, -- you have your oil warning system and you 7 have your -- 8 MR. 80CEHOLD: Metal mass in the diesel is 9 approximately 160 degrees, let's say. i 10 MR. CHAFFEE: Okay, then it sounds like then that ) 11 the rate of temperature change at the jacket water -- 12 MR. BOCKEOLD: Yes, the jacket water temperature 13 that's next to the metal, I would guess, is approximately 14 160 degrees; maybe 150 degrees. 15 MR. CHAFFEE: So then you may only be seeing 16 changes that may be only like 10 degrees. You could be 17 seeing very small changes in temperature. l 18 MR. BOCKHOLD: We're going to run these tests, 19 okay, and we're going to document what temperatures we see, 20 you know. You know, previously, we had committed to come to 21 use -- to cose up with an explanation. Basically, what we 22 believe is that we had an intermittaint failure on one sensor 23 and we probably had a calibration problem on another sensor, 24 okay? 25 You know, what we committed to was to try to 92 PROKCT 061998

I 20 1 figure out to come up with a theory on why the first trip d 2 took 80 seconds and the second trip took 70 seconds. l 3 MR. CHAFFEE Right, I understand that. 4 MR. BoCKHoLD: The difference in the timefrase 5 could be associated with temperatures that one sensor saw i 6 that was out of calibration, okay? So, we're doing this l h \\ test to try to answer that question. If the sensor had been 8 calibrated correctly, we believe that the diesel would have ) 9 not tripped. { 10 It had one intermittent failure and we really, in 11 this particular case, -- what we're saying is that we needed 12 two failures for the diesel to shut down. 13 MR. CHAFFEE: I understand. So what -- so, as far 14 as a temperature change that you would expect that these i 15 high jacket water temperature probes would see during the 16 start of the diesel -- I guess it sounds 1.ike what we're -- 17 not knowing all the facts, but it sounds like maybe a 10 18 degree temperature change? 19 MR. BOCKHOLD: That is my engineering judgment; is 20 the maximum you'd see is about 10 degrees. 21 MR. CHAFFEE: That would occur in something less 22 than a minute? 23 MR. BOCKHOLD: Well, it would occur in the l 24 timeframe of roughly 78 seconds. 25 MR. CHAFFEE: Okay, again -- and I re.alize that l 32 PROKCT 061999

i 21 I this is speculation, but if the calcon fellow said that the r 2 setpoint changes by 5 degrees if you go from basically i 3 double the heatup rate from 2 degrees to 5 degrees per' f 4 minute, then one might expect that if you also went from 5 5 degrees to 10 degrees in a minute that you might have a 6 total change in setpoint of the 5 degrees plus 5, which 7 would be 10 degrees. 8 So, perhaps the setpoint for these sensors with l 9 that rate of change might be as low, when you add the 10 cumulative up, of a 176 degrees, in which case, you may be 11 to the point, if it's 176 degrees -- and I realize that i 12 there's a lot of unknowns here, that if you had the i I 13 temperature initially at -- I think you said the warning 14 system keeps it at what, 160 degrees? 15 If, in fact, you did get a 10 degrees rise, then 16 maybe that 10 degrees rise, with all these things we just 17 talked about, could account for why you saw the diesel trip. 18 I realize that a lot of what I'm saying may not bear out, 19 but I guess, in my mind, it may be the cumulative effect of 20 all these different things that affect the setpoint, 21 combined with the fact that there is small temperature rise 22 at the start that may cause all these things to take on a 23 reality in terms of what occurred, and it may not be that a 24 sensor failed. 25 It may be the combination of the tolerance in the a2 PROJECT %

t 22 1 setpoint, the fact that the problems in doing the l l' 2 calculation. There may have been some mechanical shock. 3 Also, the rate of temperature change may reduce it some more 4 and with some sort of -- maybe as much as a 10 degree rise, 5 you could get this kind of trip. l 6 Interestingly enough, you did notice on Unit 2 f 7 when you did your testing program back in April of '88, that a you got a trip on high jacket water temperature of that 9 engine. I think you said it was 175 degrees. Again, that f 10 was when you throttled back the nuclear cooling service 11 water. Maybe, again, what you're finding here is a 12 reaffirmation for why you got that trip back then. t 13 MR. BOCKHOLD: I believe we had to have some 14 sensor calibration problems to produce this. I 15 MR. CHAFFEE: I just wanted you to hear what we i 16 had run through. I think that you would agree that there l 17 could be an accumulation of effects that are causing the 18 phenomena that are occurring. 19 MR. 90CKHOIA: Yes, it went from sensor l 20 calibration problem and then with this phenomenon and just 21 hitting it at the right time, and with one intermittent -- 22 we know we had one intermittent problem. We had one sensor 23 that had an intermittent problem associated with it. [ 24 We know that for a fact, so you know, one sensor l 25 having a calibration probles, and the accumulation of all of 92 PROKCT 082001

23 l -1 these things could have produced the trip. It could be the H 2 explanation of why t.here's an so second time delay on the 3 first one and a 70 second on the second one. In fact, that' i 4 what we believe right now. f i 5 MR. CHAFFH When we get the documented i + r 6 hysteresis and when we get a chance to take a look at this i 7 test you're going to do -- let me make sure I understand i e a this. The test you're going to do to measure how the 9 temperatures change in the jacket water system; can you 10 explain to us where these temperature sensors are~ going to 11 be for monitoring that? 12 MR. BOCKHOLD: We're going to take off the engine, i 13 one of the normal high jacket water temperature trip probes 14 and put in that well, a temperature element that can measure 15 the same temperature as a high jacket water temperature t 16 element would see, okay? We're going to just monitor that 17 and we're going to cap off the line associated with that 18 high jacket water temperature trip, so that the diesel will 19 remain operable throughout this period of time. 20 And then you will monitor it while the diesel is 21 secured. 22 MR. CHAFFH: Then you will monitor it with a 23 start in the diesel. 24 MR. BOCKHOLD: Over basically a 24-hour period, to i 25 see if there is any change in how the controls on the 92 PROKCT 062002

24 1 standpipe are working. And then we are going to pick what t' 2 we consider the highest temperature associated with those 1 3 probes, and then we are going to start the engine three 4 times and see what happens to temperature at the jacket l 5 water high temperature trip sensors as measured, and we are 6 going to measure it on a recorder. 7 MR. CHAFFEE: Oh, so, it will be on like a strip 8 chart recorder or something? 9 MR. BOC10 told: Yes. 10 MR. CHAFFEE: Okay. That sounds like a good test 11 to me. Do you have any comments on that? 12 (No response.] t 13 MR. CHAFFEE: Okay. I guess as far as the 14 procedure for doing that, is it near completion? i 15 MR. BOCKHOLD: Right. i 16 MR. WYCKOFF What are they going to do about the 17 gap between the sensor and the well,

• hat air gap that slows I

18 everything down? 19 MR. CHAFFEE: Harvey just asked a question that 20 when you put this sensor in for measuring what is going on 21 with the water temperature near the sensors, how are you 22 going to ensure that you get good heat transfer between the 23 water medium and the sensor itself? And he is presuming 24 that when you put this other temperature sensor in, that I 25 there will be an air gap. 92 PROKCT 062003

s 25 1 MR. K0CHERY: We are trying to get.a -- j 2 MR. CHAFFEE: Say that again, Paul? 3 MR. BOCKHOLD - We are' trying to get a sensor that 4 will be right against the well, or if we can't do that, we ) l 5 will consider maybe putting some sort of medium in the well 6 to make sure the heat transfer is good. I 7 MR. CHAFFEE: Okay. Is it a high-speed recorder? 8 MR. K0CHERY: Yes, there is a high-speed emitter. j l 9 MR. CHAFFEE: It is that recorder that you had 10 down at the diesel, the wide paper? 11 MR. K0CHERY: Yes. i 12 MR.-CHAFFEE: Okay. And the procedure itself is 13 in typing, so it may be out today. Are you planning as your 14 goal to try to do that test today? ) 15 MR. BOCKHOLD: ife are going to start recording the 16 temperatures today, and we are going to do the starts 17 tomorrow. 18 We had a question on the heater itself. Mas the 19 heater calibration been left unchanged since the event? Are 20 you doing anything to monitor to make sure that that heater 21 is operating properly? 22 MR. BROCKMAN We have not done that heater, we 23 haven't touched it since the event. 24 MR. BOCKHOLD: Okay. Are you going to do 25 something to check it? 92 PROKCT 082004

l 26 1 MR. BROCKMAN No, we are going to measure it as .I 2 it is right now. 3 MR. BOCKHOLD: We are going to measure it as is. 4 MR. CHAFFEE: And as far as asasuring it as is, 5 that will consist of just the measurement up by the high 6 jacket wire temperature sensor, or are you going to put some 7 sensors down by the tank? 8 MR. BOCKHOLD: We are just going to measure it at 9 the high jacket wire temperature. 10 MR. BROCRMAN: At the point of issue. 11 MR. KENDALL: Seems like you would want to be sure 12 of the initial conditions. 8 13 MR. BROCKMAN: That is right. That is the purpose 14 of measuring it at the point where the switches are so we i 15 know the effect during this 24-hour period of the heater l 16 cycling on and off as seen by the switcher. 17 MR. KENDALL: Okay. So you are saying that is whether the temperature in the tank is 190 degrees of 170 19 degrees doesn't make any difference? 20 MR. CHAFFEE: What you are saying is that you are 21 going to maintain continuity of the test all the way through 22 the diesel startup so you will see the impact if it is not 23 operated properly. 24 MR. BOCKHOLD: Correct. We will see the impact 25 the same as it was on the 20th. 92 PROKCT 062005 l

27 1 MR. CHAFFEE And if it turns out there is a (- 2 problem there, you are going to go back and try to look for 3 it in terms of if it is a problem with the heater, then you 4 will go look at that. Just a second. 5 (Discussion off the record.] l 6 MR. CHAFFEE: Maybe there is a misunderstanding. 4 l 7 We are not recommending that the test, that you go measure i 8 it at the tank. We agree that you need to take it at the 9 sensor. What Rick was proposing was that perhaps in l 10 addition -- The test needs to depend on the temperature at 11 the sensor. But what he was saying is perhaps if you were 12 to go down in addition to this and take a look at the j i 13 temperature at the heater itself, you might find out, as you 14 were proceeding through this thing, that you did have a I 15 problem with that heater, and whether or not you and up l 16 having a problem with the sensor or not. But I guess we can 17 understand why that is not absolutely necessary. 'l 18 Let me ask a different question. Does the jacket 19 water heater and its starting and stopping, is that 20 something that is on some sort of a calibration program? 21 And I guess the other that surprises me is I wouldn't have 22 been surprised that if it was it would have been something 23 that would have been part of this big, huge maintenance 24 overhaul activity. 25 MR. BOCKHOLD: I believe it is part of the PM 92 PROJECT W

28 1 program. We have to find we what the frequency is. I 2 MR. CHAFFEE2 Yes. Okay. 1 (Discussion off the record.) 4 MR. CHAFFEE: Was there any var.that, for example, 5 I see what you are saying. It is per,sible that the diesel 6 starting and the problem could be interlocked with where the 7 heater is in its cycle? In other words, if the diesel a happens to start when the heater has just finished its i 9 heating process as opposed to just before its starts? 10 MR. BOCKHOLD: We are going to look at the hottest 'l l 11 temperature in the jacket water when va do the starts on the 12 diesel. 1 13 MR. WYCKOFF: He is going to force it up there 14 somehow at the maximum. 15 MR. CHAFFEE: How are you going to do that? 16 MR. BOCKHOLD: We will look at the high i 17 temperature at the sensor. l 18 MR. CHAFFEE: Oh, I see. So what you expect is 19 going to happen as a function of time, you are going to see 4 I { 20 jacket water temperature at the sensor vary as the heater l 21 comes on and goes off. 22 MR. BOCKHOLD: Right. Varying by maybe a degree. 23 MR. CHAFFEE: And then what you are going to do is 24 you are going to time the start so it occurs when it says 25 worr.t case conditions. 92 PROKCT 082007 i

i 29 ~ l MR. BOCMOLD: Yes. i' 2 MR. CffAFFEE: Oh, okay. Well, that makes sense, 3 then. I didn't knc.w that. 4 MR. 80CM0LD: Let me make sure, I want to make 5 sure you understand. We are proceeding to do this 6 modification as far as during this temporary ood and 7 installing the temperature monitoring point as gaickly as we 8 can. . After we install that temperature menitoring point, we 9 are going to wait 34 hours and if you'have any commen:s 10 during that 24 hour 2 we would expect you to make them to us 11 and so we could adjust for any of your comments prior to the l 12 engine start. I 13 MR. CHAFFEE: We understand. i 14 MR. BOCMOLD: When we have the procedure, we will 15 fax the procedure to you. 16 MR. CHAFFEE: And that is assu: ting you get the 17 procedure to us before you start the 24-hour run? 18 MR. BOCMOLD: We are going to far the procedure 19 to you, t 20 MR. CHAFFEE: If you give it to us, we will look 21 at it. We will get your comments back to you before you 22 guys do your start. 23 MR. BOCMOLD: We will put the temperature 24 monitors in the plant, okay, because we want to start i 25 collecting data on this 24-hours as soon as we can. If you 92 PRCMECT W l

-_ _ = - 30 1 got any other comments on the procedure, we can work with I 2 you to incorporate those prior to the engine start. 3 MR. McCoY: Al, this is Ken McCoy. The thinking 4 there is that putting the, replacing the temperature switch 5 with an RTD does not have any effect on the test. If you 6 have comments, that can be done in parallel with taking ,h data. If it should change, and we wanted to take data a longer or something, we can do that. But there is no 9 exposure to going ahead and getting started on collecting 10 the data with the engine sitting there in an idle, operable 11 condition. i 12 MR. CHAFFEE: I understand. And I don't disagree I ( 13 with that. But one of my people here is talking, so just a 14 second here. I need to see what he is saying. Just a 15 second. 16 (Discussion off the record.] 17 MR. CHAFFEE: Is it possible that you could have a 18 scenario where when the temperature is highest at the -- 19 MR. KENDALL: When the temperature is the highest 20 at the sensors, could that be the point in time when the 21 temperature is the lowest in the tank? 22 MR. 80CKHOID Not with a 90-gallon per minute j 23 heat-blowing pump circulating. 4 24 MR. CHAFFEE: What is the volume of liquid in the I i 25 jacket water system? l 92 PROJECT 062009 yie., -g ,-wv. e-wm,= w-e,,- -+wr-

I 31 1 . MR. BOC1010LD: we don't have that number right 2 offhand. I'm not sure we can find that number with our 3 thrusts. 4 MR. CHAFFEE Okay. But again, the 90 gallons per 5 minute, what is the trigger for the system, because of the 6 volume of the system. 7 MR. KENDALL My thought was when the temperature 8 at the sensor is the lowest, perhaps that is when the heater 9 has been kicked on and the temperature in the tank is the 10 hottest, and then that slug gets up to where the sensor is, il plus the heat of the engine, maybe your biggest temperature i 12 gradient is going to be at that point instead of when the 13 temperature at the sensor is the highest. And if that is 14 when the biggest gradient is, that is going to have the most 15 severe effect on the temperature sensor in terms of it 16 potentially actuating at a lower setpoint. 17 MR. McC0Y: That is an issue that I think we 18 probably ought to consider. We can look at that in detail 19 while we are collecting this data. I think the point you j 20 are raising is that the rate of circulation through the 21 system with the keep-warm pump running is something we can 22 get an idea for; and if it is to the point that that might 23 affect the point where we want to do the start, we can j l 24 probably reach that conclusion over the next day or so by 25 looking at the size of that system and the rate of flow 92 PROJECT 062010 i ~

32 1 around it, and the elevations. d 2 MR. KENDALL And perhaps the temperature in the 3 tank. 4 HR. CHAFFEEt I think what Rick is saying is that 5 you may want to -- 6 MR. KENDALL: We just want to make sure you test. 7 for the worst case conditions. And just on what I am 8 hearing, I am not sure that when the temperature is highest, 9 that the sensor is going to be the worst case. 10 MR. BOCKHOLD: We will think about your comment. 11 MR. CHAFFEE And I think what he is also saying 12 is when you are collecting this 24 hours of data, seeing how 13 this interrelationship is between the tank temperature and 14 the sensor temperature, it would probably serve you to put i 15 another sensor on that tank to try to collect that data 16 during the 24 hours rather than waiting to deal with it just 17 prior to starting the diesel, and try to get some data then. la so, you're putting together procedures, some sort 19 of -- measuring device in place of one of the high 20 jacketwater temperature sensors. You're going to take a 21 look at what Rick just brought out about the interplay 22 between the heater and that hot point and you're going to 23 take a look at whether or not it's appropriate to instrument 24 that during the 24 hour run to capture that kind of data or g 25 not and you're going to give us the procedure befor'ai you 92 PROJECT 062011

33 1 start taking the 24 hour. I 2 We're going to review that procedure and if we l 3 have any comments, and if we have any comments, we're going 4 to get them back to you before you do your -- you start 5 which will happen some 24 hours after you've collected the 6 data. 7 I assume also that as reflected in the data, 8 you're going to be analysing it because you're going to want 9 100 -- telling you before you attempt to start the diesel. 10 so, we would also like, you know before you 11 actually start the diesel, to take part in understanding 12 what that data is telling you before you actually do a start i 13 on the diesel. We'd like to understand what kind of 14 phenomena you see occurring there so we will be able to 15 understand your thinking in terms of, well, this is when we 16 start to the diesel and this is what we expect to see -- 1 17 MR. BOCEBOLD: We will fax to you at the same time is that we're telling our technicians to go into the plant to i 19 go ahead and implement the changeout on the one high 20 jacketwater temperature and put an RCD in its place. At 21 that same time, we'll fax to you the procedure. So, that 22 will go on in parallel. 23 Tomorrow, we'll talk to you about the data ve have 24 and what we think on that data. t 25 MR. CHAFFEE Okay. That sounds good. We're l .2moscr osam2

i 34 1L going to get the history of the thing. 2 MR. BOCKHOLD: We'll fax to you the history of the 3 4 MR. CHAFFEE: Oh, and we did have some other i 5 thoughts on testing. Oh, okay. Several things. We'd like { 6 to get a copy of the test data. It doesn't have to be all 7 the test data but we're interested in if you have any test 8 data that was done during the startup of the diesels during i 9 startup of the plant that gives you any information on some 10 of the type of information on some of the type of 11 information you're trying to gather right now like, what 1 12 pertinent type information to try to resolve this problem. I 13 Getting the whole test data itself night be just j i 14 what we'll have to get because it's too difficult to sort ] 15 through and pick up the specific things. But, we'd like to 16 get some of that information so we can better understand how l 17 the diesels performed when the plant was starting up. l 18 In particular, we talked about this before and I ) 19 do not know if we've gotten it or not and we will check, d j 20 but, I don't think we have but the data on the unit 2 diesel 21 test that you did in April of '88. 22 We'd like to get a copy of that to see if we can 23 see from that how sensitive the diesel was to the -- you 24 were putting on it with that nuclear service cooling water 25 variations you were doing. So, that might give us some 92 PROJECT 062013

i ~ 35 1 insights in to how sensitive the system is. o 2 The other thing we were talking about up here in 3 terms of doing the testing -- 4 MR. BoCKHOLD: All I ask. Let me ask a question. 5 This is a kind of standard test procedure. It tells you 6 what to monitor and what to look at and make sure that you 7 observe normal readings. I don't think you're going to tell 8 anything from what we would pull out of document control out 9 of that. 10 MR. CHAFFEE: You mean for both the startup i 11 testing of the diesel and also the testing thEt was done in i 12 April of '88, there's no data in there that you think would i 13 help give an impression for how different temperatures in i 14 the various diesel systems were varying as a function of 15 time? 16 MR. BOCKHOLD: I think you would have acceptance 17 criteria. You would have one or two points when the engine 4 18 is fully loaded and when the engine just started. You know, 19 stuff like that, you would have some points that people read i 20 and wrote down. Okay. And you could speculate a lot but I 21 you'd have very little information. You'd have a few l 22 points. That's what I'm saying you'd have. Ken Burr worked 23 out there on the engines could say something else. j 24 MR. BURR What we were doing is setting up bypass l 25 - you'll have a lot of data -- bypass valve. I don't. 92 PROJECT 062014 l I

I } 34 1 1 remember -- l' 2 MR. CHAFFEE: I'm sorry. You're cutting in'and 3 out. Could you say that one more time? 4 MR. BOCKHOLD: What he said was that we set up the 5 bypass flow on the NSCW to jacketwater heat exchange. We 6 recorded a lot of flow information. But, we really didn't 7 record any jacketwater temperature information except for 4 that which remained within its normal limit. 9 MR. CHAFFEE: Okay. I understand. Okay. If that 10 information -- 11 MR. BOCKHOID: So, y u know, we could send you a 12 box full of data but that data, we don't think is going to 13 have auch meaning. 14 MR. CHAFFEE: Okay. Well, I don't want you to 15 send the information to use if it has nothing in there of 16 value for us to look at. However, we did talk about the 17 nachinery histories on the diesels and once we pulled it out 18 there was some very interesting information there relative 19 to what happened on the last outage on the diesels in unit 20 2. 21 Perhaps we can take a different approach. Have 22 your people gone back and looked at the data that exists on 23 the startup of the diesels on unita 1 and 2, have them take 24 a look at it and not just looking at temperature i 25 information, but, I'm assuming that they've already gone 92 PROKCT 062015 1

37 1 back and looked it from the standpoint of having problems f' 2-associated with the jacketwater temperature probes. t 3 MR. BOCKHOLD: The people that -- 4 MR. CHAFFEE Well, actually I'm not sure of that 5 because you said you didn't have your computerized system 6 back then. 7 MR. BOCKHOLD: You know, we're part of this 8 investigation, okay? And they gave you their opinion about 9 whether it would be contained in the data. If you want the 10 data, we'll gladly ship you the data. 11 MR. CHAFFEE: No. Let me ask you a different 12 question. When we talked about getting the historesis of I 13 all these probes, what you've given us now that I think 14 about it was the data as that data that existed in.your 15 computerized historesis. Is that right? 16 MR. BOCKHOLD: What we did, when you asked for the 17 data, I said that basically I'd get the appropriate people is to pull the data that you asked for after we got the diesel 19 operable. 20 It took us a while to get the diesels operative 21 but after we got the diesels operable, we then went to work 22 on pulling the data. And the thing that's readily easy for 23 us to pull was ask the computer to give us a dump of the i 24 information associated with the data that you had requested. 1 l 25 And we got that and we gave that to you. 92 PROJECT 062016

38 1 And now on various probes, we've pulled together il 2 some additional data associated with some of our other I 3 records associated with those probes and we will be 4 teletyping you that other data because we thought it was 5 pertinent tt) the_ investigation. 6 MR. CHAFFEE: Is this information on -- let's make 7 sure we're not confused here. There's two things. You were 8 in the process here yesterday of providing information on 9 the quarantine sensors. That's different. Is what you're 10 telling me is there other information? l 11 MR. BOCKHOLD: As much of the human input as we 12 could capture to be appropriately responsive, the recent l I 13 temperature sensors because we've got some people who worked i l 14 on some of the work orders and we may have been able to get 15 some human input on that and we basically tried to pull i 16 together a narrative about what t'.te people remember plus i 17 what the data showed. i l 18 MR. CHAFFEEt For the sensors that are I t 19 quarantined? 20 MR. BOCKHOLD: Yes. 21 MR. CHAFFEE: Okay. I understand that.

Now, 22 let's go back to -- I understand all of that.

But, what I'm 4 23 going back to now, is we're also interested in trying to see 24 what history exists back before your -- we want to make sure 25 that we have a complete history on problems that occurred 92 PROECT 082017 1 I l

I 39 1 with the diesel in the sensor area and the data you gave us, f 2 I think it was on Sunday, had that. 3 At that time, we were told that that information 4 was pulled from what you had in your electronic system. I 5 was under the impression that you hadn't pulled any data for 6 stuff that was prior to that and I also had the impression 7 that your electrical system did not cover the history back j 8 before -- 9 MR. BOCKHOLD: Yes. 10 MR. CHAFFEE so, it's possible on units 1 and 2 11 that during the startup testing progran you may have had 12 problems with the jacketwater temperature probe and if that is 13 occurred then that data may be out there but it wouldn't 14 have been included because it wasn't part of the search that 15 was done? 16 MR. BOCKHOLD: Well, we gave you some data on 17 temperature probes associated with the unit 2 startup and we 18 said that the unit i startup, our computerized record system 19 was not developed as well as it was for the unit 2 startup, 20 therefore, that data, I'm not sure was as comprehensive. 21 MR. CHAFFEE: so, does that mean that people went 22 back to look at that or they didn't.when they pulled this 23 stuff together? 24 MR. BOCKHOLD: Pulled it out of the computer i 25 record -- and I'm not sure you're going to get auch useful 92 PROKCT 062018

40 1 information. U 2 MR. CHAFFEE: No, but it would be a little 3 embarrassing to all of us if it turned out that during 4 startup testing, you did have something back there in unit 1 e 5 that goes right into this area and in doing all this 1 6 troubleshooting, we didn't know about it and it turned out 7 that if we had known about it, it would have shed some clear 8 light on what was being dealt with. 9 MR. BOCKHOLD: We supplied a copy of all NW0s and 10 all calibration for all calcon separate switches. I am not 11 sure what you are asking for is my problem. 12 MR. CHAFFEE: Oh, oh. What we're trying to do is ,i 13 make sure that we have the full history of what's going on 14 with these diesels at Vogel. And I want to make sure that 15 if there is something going on that happened during -- what 16 I'm saying is I'm not convinced you can rely upon the 17 memories of people of all the things that happened in a 18 diesel, even those people that were involved with it. 19 The reason I say that, you know, as we have 20 learned -- and I am not saying -- I don't think it's fair to 21 expect people to remember all that stuff. And we found out 22 in this particular case that they didn't. People didn't 23 remember all the problems they had in the water tempereture 24 probes that occurred back in the last refueling outage. 25 So, what I want to make sure is, to the extent we 92 PROJECT 082019 i

r 41 1 can and the process of developing and going into trouble- ) 2 shooting and finding this thing, that when people have gone 3 back and looked at all the history that's available to make 4 sure they have taken that into account before they go i 5 through the process of doing a lot of this trouble-shooting. 6 And I realize at this point -- it is my impression that we 7 haven't quite gotten all of that. And I want to make sure i 8 that we get all that information together. Because I guess 9 I have an opinion that, in the process of looking at some of 10 that information, it may give you insights into what's going 11 on. 12 Hopefully, what's going to happen is after you do i 13 the test on the jacketwater temperature system, that 14 combined with the other information will be the answer. 15 But, unfortunately, it may turn out that it's not. In which 16 case, you may find that there is something back in this 17 other historical information that we haven't brought to l 18 light yet that saybe has the key to what's going on. t 19 MR. BOCKHOLD You asked us in one of Rick 20 Kendall's requests, okay? You asked us for a copy of all 21 NWOs and all calibration data logs, all Calcon temperature 22 and pressure switches for both units. Okay? And what we 23 have started to comply, okay, is a list of all those NWOs. 24 We have got them going back as far as '85 and we are still i 25 working on that particular data request. 92 PROECT 062020 v,

b t 0 42 1 MR. CHAFFEE: Okay. But you realize -- we asked 2 for the data but -- 3 MR. BOCKHOLD: This will and up -- we will provide 4 sose information and if you want some more information, we 5 will go to microfilm and get as much information out of the 6 microfilm as we can get. 7 MR. CHAFFEE: I understand. But you understand 8 that it's not so much the goal that we get all this 9 information and we on the team look at all this information, 10 what's really most important is that your people on your 11 team are making sure that they have analyzed the information 12 so that they have enhanced their ability to try to detect 13 what's going on with this diesel. We are asking for it so 14 we can do the same thing. 15 MR. BOCKHOLD: Yes. What we feel is most 16 appropriate is to press on with the testing of these 17 switches as we have outlined to you. 18 MR. CHAFFEE: And we're not saying that doing that 19 testing is wrong. We are just in the procese of talking 20 about the stuff. I wanted to make sure that you understood i 21 we still want to get that information. 22 MR. BOCKHOLD: We'll get right back with you. 23 MR.CHkFFEE: Okay. 24 George -- did they just go off the phone? 25 George, did you go away or are you going 'to come 92 PROJECT 082021

i I O 43 1 back? 4 2 MR. BOCKHOIA: I'm still here. We are trying to 3 figure out how to do what you want to do, okay? And 4 understand the sagnitude of the effort that's involved and 5 how many weeks it's going to take to go back and look at 6 various things and that kind of stuff. What we have is 7 recent data and we're actively looking at the recent data 8 out of this overhaul. But when you go back to startup, our 9 engines sat in the yard for a long time and when we moved 10 them into the plant, we basically discarded a lot of the 11 sensors that were out there because of the age of them and 12 how the engine was handled and everything else. We did i 13 entire tear-downs of all four diesel engines because we had 14 to inspect these engines and go through a reverification 15 program. You know, we tore them basically down so they were 16 in pieces, okay? We put them all back together during the 17 startup program and then we brought, the sensors up to speed, 18 okay? And we had some problems with sensors and just in the 19 shakedown period, no one brought them operable. l 20 MR. CHAFFEE: Well, maybe we should start -- 21 MR. BOCKHOID: The time on the sensors was pretty 22 auch to the operable time frame on the diesels. And we are t l 23 trying to figure out exactly what else we can do and whether 24 that makes sense. l 25 MR. CHAFFEE: Okay. 92 PROJECT 062022 l

44 i 1 MR. McCoY: This is Tim McCoy. ] (I 2 You know, what you are saying is obviously logical 3 and something that, you know, we need to think about how to-4 tackle. Just quickly here -- and I don't have a complete 5 appreciation of this yet, but what Paul is telling me here 6 is there is some 4,000 NW0s that we pulled out of the index 7 on the diesel generators. To go back and pull the microfilm 8 on those things and probably each one of those is about a 9 half inch thick, or something like that, of data. It's a e 10 massive undertaking. Now, there may be ways we can sort 11 through and decide which ones we would want to look at and 12 make copies of, and that sort of thing, but we have got to l' 13 do that in a smart way. And that's what we are trying to l 14 talk about right now. 15 MR. CHAFFEE: I agree with you. 16 MR. McCoY: So the immediate reaction is that that 17 is the logical kind of thing to do but when you get into the 18 practicalities of what may be retrieved from that, and so 19 forth, I think we need to do some thinking about how to do 20 that. I also don't think that needs to be done in advance 21 of doing this testing or continuing at all. That may 22 something that we can get into after we get through this 23 effort that I think is much more likely to produce some real 24 meaningful results, getting the testa done and getting the 25 results of that. Going back and researching all the history 92 Ph0KCT 062023

{ l 45 1 at startup might be something we could do in a longer period i l 2 of time. 3 MR. CHAFFEE: Okay. I understand. I don't 4 disagree for the test you are going to do which is going to 5 show how the jacketwater system temperatures vary as a 6 function of time and doing this history search is not 7 necessary prior to doing that test. 8 MR. McCOY The people that do the history search 9 have got to be knowledgeable people that could make a 10 connection as they read things as to whether anything would 11 have a relationship. 12 MR. CHAFFEE I do think, though, that for the 13 testing of the currently quarantined sensors, I do think 14 that a record search only to the extent of going back after 15 the diesels were, you know, put together and you had the 16 sensors in and you, basically, felt that you had them ready 17 to go and then proceeded with testing the diesel, if any of 18 those testing records show that there was a lot of problems 19 with jacketwater sensor probes, I wouldn't think that it 20 would be that difficult to find that out and I would think 21 it would be important that we have that history so we can 22 factor that in as we continue to develop what testing is 23 going to be done on these jacketwater sensor probes, which 24 is something else that we just want to talk about for a few 25 minutes. We had some ideas and some things for you.to 92 PROJECT 062024

44 1 consider in terms of how to do that testing. Il 2 MR. McCOY I think we just made the list. 3 MR. CHAFFEE: Okay. 4 Again, we don't know why the jacketwater sensors 5 are not working properly but one of the thoughts that people 6 have had both at your site and here is the fact that perhaps 7 some sort of foreign material is getting into the valve and 8 the sensor and causing it not to work properly. So, one of 9 the thoughts that we had was that perhaps when you take the 10 sensor into the IEC shop to do the testing on it, you should 11 put some sort of a micron filter on the vent side of the 12 valve so that if any particles are there, if they get blown 13 out, they will be captured by the micron filter and at least 14 you will retain the particle rather than losing all the 15 evidence. So that's one thought that Rick had come up with. 16 MR. McC0Y: There's not an easy way to do that. 17 But let me give you one piece of personal data that I just 18 happened to have by coincidence. 19 I was out in the diesel room during the overhaul 20 and personally looked at the interior of the jacketwater l 21 system and the coolers because I was curious as to whether i 22 our corrosion protection scheme was working and I talked to 1 23 the people who were disassembling that and physically looked 24 at it myself and it was extremely clean, no indication in 25 the cooler that there was anything, any corrosion -- and the 92 PROJECT $5

P 47 1 opinion of the people there was it was very clean system. 2 Now, I realize you are talking in terms of micron particles 3 and that's not what I'm addressing. 4 But what I'm addressing is that I know from I 5 personal observation and talking to the people that were 6 disassembling it that they did not find any kind of 7 corrosion or material in that system. 8 MR. CHAFFEE: The switches we were talking about 9 though, as far as far as having the rest of the stuff in the 10 jacket water system itself, the switch doesn't see that. 1 11 There's an RTD well that the jacket water sensor' probe fits 12 into. What we're talking about is impurities that are in i 13 the air system that goes to the switch itself. l 14 We're not proposing that the impurities are 15 getting from the air system itself, because it's our 16 understanding that there's a 5 micron filter that prevents 17 that. What may be happening is that when the switch is 18 actually installed in place -- and again, I haven't watched 19 them do that, but when they actually install it in place 20 there's some' fittings that they take off on this particular 21 switch. 22 There's a possibility in just doing that that some 23 debris might get introduced. Also, in the threads; they put l 24 material on the threads -- at least the switch we have up t I 25 here has material on the threads and if that's the case, 92 PROKCT 062026 1

4e 1 maybe some of the material is getting in there. But the 11 2 point is, there may be some material that's being introduced 3 unintentionally as people are working with the switch. 4 Even when you take the top of the switch off to do 5 the calibration, if aamehow a particle got in there, that 6 perhaps could find its way in and have an impact on the 7 ability of the switch to work. So, what we're saying is, if~ 8 that's what's occurring, if you could put some sort of a 9 filter on the end of the switch, you might be able to retain 10 that evidence, rather than having it being lost and as you 11 go through and test the switches, finding out at the end -- 12 if that was the only thing that was causing it, it is 13 conceivable that you'd lose all that evidence as you did all 14 those tests on the quarantine and we'd have no explanation 15 for why the switches are behaving the way they are. 16 In fact, having a particle being part of the cause 17 -- you can come up with a paradign,that makes sense by the 18 fact that they have found these sensors where they just seen 19 to be venting and leaking. One might surmise that that's 20 because there is a particle in there that's presenting the i 21 seat from fully seating. It tends to cause you to think 22 that maybe there could be some sort of particle. 23 Also, in looking at the switch, the tolerances in j 24 that switch -- I'm sorry, the valve. The valve itself is t 25 really very, very small. It's only about a quarter of inch 92 P%NiCT %

C 49 1 in diameter and the actual 0-ring that seats on it, it is e 2 probably on,1y about 3/8ths of an inch in diameter. The 3 clearances around the thing are on the order of 10/1000ths 4 or 5/1000ths, very, very small, so you're not talking about 5 a very large particle and having it be introduced. If it's 6 possible to prevent the seat from being able to seat and 7 also perhaps somehow it can cause the valve itself from 8 being stuck. 9 Again, I'm not sure how hard it is, but I thought 10 you should at least hear what we thought on this so that you 11 could consider it. What are some of the other ideas that we 12 had? The other thought we had was, again, not knowing 13 what's all going on here, maybe one of the problems with 14 these jacket water tesperature probes is the fact that 15 they're somewhat erratic; in other words, they don't have 16 good reproduceability. Maybe how they perform varies from 17 one occurrence to the next. 18 The other thought that we had that we thought 19 would be prudent would be -- and we would try to resolve 20 that; whether or not that exists or not -- and that would be 21 to take one of these jacket water temperature probes and do 22 a series of tests on them where yod plot as a function of 23 time, temperature drop through the valve as well as the 24 temperature that's heating up the probe and see if you do 25 get good reproduceability. 92 PROJECT 062028,

50 I 1 If you don't get it, then that coudl be another li 2 contributor to the probles you're seeing. If you do get 3 3 good reproduceability, then you can waive that out as being 4 a factor that's contributing to the phenomenon that's J 5 occurring. Again, I'm not sure exactly how you build that 6 type of a test, but do a certain number of tests on that. e 7 You might want to do like, you know, three or five 8 on one probe to see what its reproduceability is and then do 9 one other one three or four or five times to see what it is 10 and get some sort of a feeling as to if the hysteresis or 11 the variability of the switch varies from switch to switch. 12 Again, this is in an attempt to eliminate possible 13 causes. I have the suspicion that what's going on in that 14 diesel is an accumulation of different effects. 15 MR. BOCKHOLD: Three to five times, I think -- we 16 have already done that with the calcon rep at other times in 17 the past. That would be part of our testing plan with the 18 quarantine switch. 19 MR. CHAFFEE: Okay, I'm assuming that for the l 20 tester that's being used to do the testing, you'll somehow 21 verify before it's used that it's high quality. We had L 22 talked earlier in the INC shop about whether or not the air 23 there is good enough. We were told that you guys were going 24 to go take a look at that. 25 Have you done that? Is there -- have you found g2 PROJECT W

p 51 i 1 any concerns with the air in the instrument shop? Have you i 2 looked into the possibility that as they did the calibration 3 of the switch, the air from the instrument shop is what is 4 introducing the foreign material? 5 MR. BOCKHOLD: Yes, we're using. instrument air 1 6 from the plant and.it's -- we do the INPO verification test 7 on instrument air from the plant. It's different air than a the air in the diesel but I think the instrument air from 9 the plant even has a lower dew point than the.nstrument air 10 for the diesel. 11 We have been all along checking air quality. We 12 haven't any filters in the shop on instrument air. We have 13 five microns? We believe we have good instrument air in the 14 shop. We use the instrument air from the normal instrument 15 air system that's in the plant. To feed the shop, we have a 16 filter in the shop on instrument air. We do not know the i 17 micron size of the filter in the shop. l la We will find out the dew point that plant j 19 instrument air normally runs at and we'll find out the size j i 20 of the micron filter in the shop. I 1 21 MR. CHAFFEE: Okay. .) 22 MR. BOCKHOLD: I think the instrument air in the j plant is like minus 70 degrees or something. It's real dry 23 i 24 air. I 25 MR. McC0Y: I have a kind of general recollection 92 PROKCT 082030 l

r 52 1 and I'd like to go back and look at this. In response to il 2 the bulletin or notice on the instrument air problems, we 3 looked at our instrument air quality and history and we were 4 basically in very good shape. We had not had instrument air 5 quality problems that the other plants had had. That is my 6 general recollection. 7 We can take a look at that, and I now understand 8 where you're coming from on the particles. I was complete 9 off on a different subject when we were talking about the 10 jacket water stuff. 11 MR. CHAFFEE: Okay. We had a couple other l i 12 thoughts. I think, if I remember correctly, the way the i 13 test goes it that you take the switch at ambient 14 temperature, pressurize it, see if it's leaking or not. If i 15 it's not leaking, then you would heat it up and verify that i 16 it had the proper setpoint. f 17 One question we had is; if you heated it up and it 18 starts to leak before it reaches the setpoint, again, our 19 hope would be that you have a filter in it so that if you 20 heat it up and it begins to leak, when it starts leaking, if 1 21 it blows anything else, it will have it for retention to 22 look at it. 23 MR. BOCKHOLD: We have a very hard time complying 24 with that request. The problem is that -- you see the 'l 4 i 25 button on that switch? 92PROJEM M i

l t 53 1 MR. CHAFFEE: Yes, it comes off and it's got

• (

2 threads. 3 MR. BOCKHOLD: It's where it leaks, you know, and 4 they you then would have to have something that didn't have 5 a backpressure on the switch that could somehow attach the 6 particles, and we really don't have anything set up in our 7 shop to do that. 8 MR. CHAFFEE All right, okay. I'm not going to 9 refute anything you said, but all I can tell you is that 10 it's got threads. I don't know what kind of filters are 11 available and what kind of backpressure they create. The 12 only thing I can say is that the diameter of the hole that i 13 it vents through is about -- I'd say it's almost a half an 14 inch. 15 As far as the orifice size of the actual valve 16 itself, I don't know, but I'm going to guess that the thing 17 represents about maybe 20 thousandths. As far as when the 18 valve is fully open, there's not that much clearance that 19 the air passes through. I'm sort of -- and again, I'm 20 probably wrong, but I'm not sure that the -- what you would 21 introduce in terms of a backpressure and having it go 22 through some kind of a filter -- would be as much of a 23 concern as you're addressing, but if I'm wrong, then 24 obviously you can't do it. I 25 Let's go on to the next one. As you heat the

l F 54 1 valve up, if it does start to leak, the next thought we had 1 2 was, "okay, well what do you do at that point?" because, 3 again, if it's beginning to leek, then it could mean a 4 couple of things. It could mean that maybe you do have a 5 particle in there, and now that it's heated up, there's not 6 as much pressure, and it's then leaking, so you'd like to 7 retain that particle, or maybe it just means that the valve 8 is at the point where it's sort of like beginning to simmer 9 before it begins to make its transition. 10 I guess what's true is that the valves do begin to 11 leak a little bit before they then change state to fully 12 open. Isn't that correct? 13 MR. BOCKHOLD: Yes, that's correct. 1 14 MR. CHAFFEE: So, I guess once they get to the 15 point where it's hot, is there anything they need to do 16 diffarent there? They just continue up and do the 17 calibration on it. Okay. I guess that's not a problem, 18 then. 19 So, I guess, really the only thing that we wanted 20 to add to in terms of your troubleshooting was, one, to try 21 to see to what extent you can not lose any particles if 22 they're in there that me causing the problem, and then I 23 guess the last thing that we thought you should consider is 24 -- and again it depends on the history of the switch -- if, 25 for some reason, the switch wasn't working properly before, 92 PROMCT 062033 j

n 55 1 and you then go out and do it, and you can't find anything 2 wrong with it, I think you should consider then taking it 3 apart and seeing if you can find if there's anything 4 retained within the switch that could have been a 5 contributor to why it wasn't working properly. 6 MR. BOCKHOLD: Yes. That's part of our procedure 7 here. 8 MR. CHAFFEE: Again, we're a little -- you're 9 probably a little concerned also that I'm afraid we're going 10. to go through all seven of these switches and find out, when 11 all is said and done, that they can't find anything wrong 12 with them. So we want to make sure if that occurs, it'll be j 13 as clear as possible that everything was done to try to i 14 retain any particles or find anything else that was going 1 t 15 wrong with them. i 16 The other thing that confused us from the j 17 discussion yesterday was the fact that when the-Calcon 18 fellow -- j 19 MR. BOCKHOLD: Hey, A1, you know, the thing that 20 we keep worrying about here is -- you know, I have an 4.. 21 impression in my mind -- our technicians worked with the 22 Calcon rep, and he has some experience, and he's going to i 23 say, " Gee, should you do this?" and he's going to tell us 4 a 24 what he thinks we should do, and we're going to pretty much 25 follow his advise.

.~ i 56 1 Basically, I's concerned that we're going to do d i 2 something you don't want us to do, and saybe your person 3 should be here watching. 4 MR. CHAFFEE: Yes, that -- you may be right. We i 5 may do that. 6 MR. BOCRHOLD You know, that's my concern. Maybe j 7 we should hold off testing these switches until tomorrow, I 8 the next day, or something like that, and give you the 9 information and let your person be here and see what we can 10 do and what we can't do. I'm sure we're willing, if we can 11 find a fitting, to hcok up some piece of tygon tube and try 12 to hook it to a filter, but, you know, that's a discussion i 13 that's -- you know, is it reasonable? If it's a tiny little 14 piece, is it going to get stuck some place in the tygon tube 15 and get lost? You know, all those worries go through your 16 head. 17 So, you know, if you got all these concerns, maybe 18 you should have your person here while we're doing. That's 19 the only thing that comes to mind. 20 MR. CHAFFEE Okay. I think whether or not we 21 have a person there is something that we need to consider. 22 Again, until we do that, what we're trying to do is make you 23 aware of some of tP.e considerations that we've been able to 24 think up that tre think are germane to your testing so that, 25 you knW, you the opportunity to hear the thoughts we have 92 PRCAECT 082035

i i 57 1 so you can take them into consideration in how you develop 3 84 2 your test. 3 In the dialogue, you know, if after you think 4 about then you decide they are not of merit, then, you know, l 5 we'd like to discuss them, and if they're not, then fine. 6 But, again, the approach here is to just try to maximize 7 this opportunity in terms of trying to find out what's going i 8 on with these calcon switches, and the input we're providing 9 is only in the interest of accomplishing that. 10 We recognize that we are not the experts on how 11 these switches work, but what we think is helpful is when 12 you get a number of people together and they think about 8, 13 various aspects of it, that the and product will be better 14 served because it will have had the advantaga of a variety 15 of thoughts on how to maximise the opportunity. 16 MR. McC0Y This is Ken McCoy again. I think 17 before we do anything with these switches, we need to decide l 18 on wht t is the appW'riate test plan so we don't do 19 something and then 1,6ter cone up and say, " Hey, we didn't 20 find anything," and then agree that maybe we didn't do the 21 check-out or testing properly. 22 MR. CHAFFEE And I agree with that. 23 MR. McC0Y: So, you know, there are two elements 24 here. One is that we need -- Georgia Power needs to come up i i 25 with our plan of what we think is the best way to get to the h A ftp f,0-b O qttc s/t'lr r

+ l 58 1 bottom of this. Maybe it's to send these switches to an t 2 instrument lab that Calcon agrees with and can supervise the 3 disassembly and inspection or whatever needs to be done in 4 the presence of Georgia Power and the NRC or something like 5 that. I don't know. But if this issue is critical, we l 6 shouldn't go march off and do anything until we're all in 7 agreement about what needs to be done. 8 MR. CHAFFEE: I agree with you. In fact, that's j 9 what this dialogue has been about, it's to try to move 10 further along in that direction by giving you some of the 11 ideas we had. 12 MP.. McC0Y And the kind of thing that you're 13 talking about in locating micron particles and all is not 14 something that you can do in an instrument shop in a power j 15 plant. 16 MR. CHAFFEE: Okay. 17 MR. McC0Y: And that's what's got me worried. 18 MR. CHAFFEE: Okay. Again, because what we're 19 very -- what we're pushing for to come out of all this is a 20 clear understanding of what's going on with these switches 21 because we think it's important to you as well as to the 1 22 industry to unde?tstand them because these switches are being 23 used on a lot of diesels at nuclear power plants around the j 24 country. 25 Mk. McCoY: Yes. That's kind of the thinking that g2 WU

I" 59 1 we have, too, that this is more a -- this is a bigger 2 problem than Vogel's problem, and if we're going to go into 3 a research effort on these switches, which, you know, the i 4 data we have here indicates that there's something wrong 5 with the reliability of these switches, and we are hearing 6 the same kinds of things from some of the other plants that 7 we've contacted, that there are reliability problems with 8 these switches. If that's the case, then they probably 9 ought to be taken some concerted effort to get to the botton 10 of the reliability of these switches, and it ought to be 11 done in conjunction with the vendor, you know, as a generic 12 effort, I think. 13 MR. CHAFFEE: Okay. And also realize that at 14 this point within the agency, because there is some concern 15 about the reliability of these sensors, and in fact it 16 resulted in the diesel not being able to start properly at 17 Vogel during the event of March 20th, it -- and there is 18 some concern within the agency in terms of how operable the i 19 diesele are at Vogel. 20 MR. McCOY: We understand that, and, of course, 21 we're -- that's one of the key things that we monitor from a 22 safety point of view, is the reliability of our engines, and 23 we want to be sure that we have highly reliable engines. 24 So, that's what we're trying to do, and we've -- you know, t 25 we think our testing program is good, and that our data 92 MOECT 062037

60 1 indicates our engines are at least as reliable as the rest i 2 of the engines in.the industry in recent times. So, we've 3 got to address that issue. l 4 The switches -- you know, one thing that was 5 different at Vogel was that these switches were in the 6 circuits during under-voltage, cnd we are taking -- 7 MR. BOCKHOLD: The thing that's different at Vogel 8 is we have a two-out-of three logic on these switches and 9 some other people only have a one-out-of-one logic on some -10 of the switches at their particular plants. 11 MR. CHAFFEE: Right. But in this case at this 12 plant -- ti 13 MR. BOCKHOLD: The overall reliability of Plant 14 Vogel's engine -- we meet the blackout rule for the higher 15 classification of diesels. We have the 95 percent or 95 [ 16 times out of 100, our engines will start and load and 17 perform satisfactorily. so, we meet the highest 18 classification on the blackout rule. i i 19 These switches -- we've had trouble at overhaal 20 times, but during normal plant operation and in between 21 overhaul times, the switches have been reliable. 22 MR. CHAFFEE Right. But your monthly test may 23 not be valid. 24 MR. BOCEROLD: The monthly test is valid that the i 25 engine will start and run and appropriately come up to speed l

( l 61 1 1 and that type of stuff. 2 MR. CHAFFEE: If it's had an air roll before you 3 do the test, before you have the event. 4 MR. BOCKHOLD "'ha air roll, as we talked about, 5 did not effect the pneumatic controls on the engine. So the 6 failure mode of the pneumatic controls.was basically 7 eliminate to be associated with the air roll. So the 8 surveillance testing on the diesel is appropriate. 9 MR. CHAFFEE: Well, but what about if the air roll 10 has had some impact on the temperature distribution of water 11 in the jacket water system? 12 MR. BOCKHOLD: And we're looking at that specific 13 test that we're doing to address that. We have an 14 engineering judgment at this time about that, but we're 15 going to address that in a very specific test. 16 MR. CHAFFEE: Right. And I understand that, and l 17 that's why I'm very strong on having that test be done to 18 find out what the jacket water temperature is doing, and j 1 i 19 hopefully that will eliminate the current anxiety that 20 exists that perhaps the monthly test, although it was 21 intended to be a valid test, parhaps these air rolls j 22 unintentionally has invalidated it. Hopefully, we'll be j 23 able to answer that question yes or no once you do this test 24 on the start of the diesel with it instrumented after 24 25 hours without an air roll. 92 PROKCT 082039 l

g-62 1 MR. BOCKHOLD: A1, we've had other challenges, 2 other starts on these diesels without air rolls. In fact, 3 on the 20th, Unit 2's 8 diesel, okay, started, tied in, an 4 carried its bus appropriately. 5 MR. CHAFFEE: Okay. 6 MR. BOCKHOLD: We've had a lot of others, 7 especially in the start-up time frame where the engines were i e sitting there available, and we blacked out the bus for 9 various reasons -- some personnel and start-up errors, some 10 associated with testing where they were not air rolled, and, 11 again, they started and loaded and carried the buses 12 appropriately. l 13 MR. CKAFFEE: Have all the air rolls, the ones { 14 that have been successful, have they all occurred in diesels 15 that have not yet had this big 11-or-12-day naintenance 16 thing done on them? 17 MR. BOCKHOLD: We had to do this big maintenance is thing on all four engines before we started. Because wo 19 have fewer valves engines and because of other industry 20 problems, we had to completely take the engines apart. 21 MR. CHAFFEE: That is before you ever even got 22 licensed? 23 MR. BOCKHOLD: That is before we got licensed. So 24 the first thing we did back in our Unit i days is we 25 disassembled and we took all kinds of measurements and we s2 PROECT W

( 63 1 cleaned up the engines, and then we reassembled the engines i' 2 onsite. l 3 MR. CHAFFEE And you did the same thing in Unit 4 27 5 MR. BOCKHOLD: Sare thing on Unit 2. 6 MR. CHAFFEE: Oh. I d!dn't know that.

Well, 7

then, maybe what is going on here could in fact exist for 8 both units. Well, let me ask this question then, George. 9 Is what you did in the last refueling outage on the Unit 1 i 10 diesels, is it very similar to what you did before the plant 11 was started up on both units? 12 MR. BOCKHOLD: Before the plant was started up, we 13 did a much more extensive -- 14 MR. CHAFFEE Somehow was it done differently? Up 15 to now I got the impression that this problem was associated 16 with these two maintenance activities. 17 MR. McC0Y: A1, Ken McCoy again. These DeLaval 18 engines, you know, there was an owners group that had to 19 address the generic problems of these engines back in the 20 1981 time frame, and the owners groups all agreed to do 21 these extensive outages, and they are done on all of the 22 DeLaval engines. We are not unique in that respect. There i 23 are some differences with individual licensees in the 24 construction and exact commitments and so forth, as I 25 understand it. But in general, everybody has to do these g pf OJECT W1

,L u t 64 1 maintenance outages, and DeLaval does that work at most of .4 2 these places. 3 MR. BOCKHOLD: Yes. DeLaval, and they have 4 changed ownerships. But basically the same people at each 5 of the refueling outages. The original equipment 6 manufacturer has supervised all overhaul of our engines. 7 The one thing that he didn't supervise was the calibration 8 of these switches closely until we got into this problem. 9 So, you know, we have basically, because of his supervision, 10 made sure that we did these calibrations in a very 11 consistent fashion. 12 MR. CHAFFEE: Okay. I have one more question. 13 This is on a different -- Can we move on to one more topic? 14 MR. BOCKHOLD: What did we agree to on the 15 quarantine switches? We will send you the other data? 16 MR. CHAFFEE: You are going to send us the other 17 data and you are going to consider the comments we have made 18 for incorporation in your test procedure, and then we will 19 talk again once you have considered these things, so that we 20 can nail down our agreement in the test procedure for the 21 switches. 22 MR. DOCKHOLD: Fine. We'will send you the 23 additional data. We will consider your comments. And we 24 will call back. 25 MR. CKAFFEE Okay. The last thing we had is 92 PROJECT M

l 45 1 yesterday when the calcon fellow was on the line, he talked v 2 about a switch, a high jacket water temperature switch, that 3 they found that it was finger tight. In other words, it was 4 supposed to be real tight, and it wasn't. And I guess now, 5 since we have looked at the switch a little more, and we are 6 a little more educated, we want to make sure that we 7 understood properly what he was saying. Is he there? 8 MR. BOCKHOLD: He is not here. Is he in the plant 9 someplace? We can get him over here in about five minutes. 10 Why don't you ask the question, and we will get the answer. 11 MR. CHAFFEE Okay. Here is what confused us. We 12 had the impression that e at he fc?tnd that wasn't finger-il 13 tight was where the top cylinder, which is about a one-inch 14 diameter cylinder, where it was attached to the lower, same 15 diameter cylinder which then goes down into about a half-16 inch diameter thing, and we thought what he was saying was 17 that those two things were loose and that the screws that 18 are underneath that go up there, that they had no impact on 19 that, and it was just a matter of finger-tightening the 20 thing. And with that impression in our mind, that was our 21 understanding. When we took it apart, we found that it 22 didn't work that way at all. We found that in fact there 23 were no threads there, and in fact the three screws do 24 connect the bottom and top. 1.nd so having seen that, we had 25 two conclusions. Either we atsunderstood what he said -- 92 PROKCT 082043 \\

l L-t 66 1 R. BOCIGt0LD: We will have to get him on the 2 phone. Do you want us to get his right now? 1 MR. CHAFFEE We can'do it with him calling us as 4 soon as he is available. But we want to make sure we 5 understand this, because it has an impact on our opinion of 6 his knowledge. 7 MA. BOCKH014: Okay. We will have somebody with f 8 him and call him. 9 MR. CHAFFEE: Okay. The other thing I would 10 recommend is if you guys haven't done it, you think about 11 taking one of these things and taking them apart and taking 4 12 a look at it, because it is very, it is much easier to i i 13 understand how it works by taking a look at one of these 14 actual switches and seeing how the internal mechanics of it 15 work. 16 The other thing we thought would be valuable is 17 the people that are actually going to do the testing of 18 these ought to take one of the apart to get a feeling for 19 what it looks like. Because what they are going to want to 20 do, depending on what they see is, they may want to take one 21 apart to finc. evidence, and by having taken it apart once 22 already, and seeing what it looks like internally, it would 23 help then prestrategize what the appropriate actions would 4 24 be as they get into testing the valve, whether or they 25 should take it apart, if they take one piece off, what will szImoJEU M

67 1 that do, would they lose evidence or not lose evidence, that i 2 sort of thing, in terms of trying to figure out what is 3 going on. 4 Also, as they see the thing behave, they have a 5 better mental image of what is going on internally in terms 6 of how the valve is actually performing its function. 7 MR. BOCKHOLD: We agree. 8 fGt. CHAFFEE Okay. That is all I have. Do you 9 have any'4hing more, Rick? 10 MR. KENDALLs No. 11 MR. CHAFFEE Okay. That is all we had, George. 12 MR. BOCKHOID Okay. We will get back to you 1 I 13 later today, after we telecopy you some stuff. 14 MR. CHAFFEE: Okay. Thank you, George. 15 (Whereupon, at 11:44 a.m., the teleconference was 16 ended.) 17 18 19 20 21 22 23 24 l t 25 92 PROKCT 082045 i

REPORTER'S CERTIFICATE This is to certify that the attached proceed-ings before the United States Nuclear Regulatory Commission in the matter of NAME OF PROCEEDING: Teleconf erence Between IIT, Licensee Personnel, and Region DOCKET NUMBER: It Personnel PLACE OF PROCEEDING: Bethesda, Maryland were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court report-ing company, and that the transcript is a true and accurate record of the foregoing proceedingr. Y$- m i-Rossie Sutton Official Reporter Ann Riley & Associates, Ltd. l k 92PROMCT Dameg

[ s**~ Nz4r ant GPC EXHEIT 68 waynesboro, Georgia 30830 BOCKHOLD P $$Nt b Nwlear Plent Vogtle Georgia Power l the sourfern elecitC system Calcon Temperature Switch Response Test on 4/4/90 This testing provided infonnation concerning the response of a Calcon temperature switch to different heat up rates. All data was taken with the switch in an oil bath. The temperature was varied at rates of 1*F/ min., 2'F/ min., 3*F/ min., and 4*F/ min. The attached chart shows the response of the switch. The second test was performed by stabilizing the switch in an oil bath at 160*F. The switch was then placed into an oil bath at 200*F. The response of the switch to trip, given the 40*F step change, was 3 minutes and 54 seconds. 92 PftOJECT 006222 f\\ n

l 2AL

\\

TEMPERATLRE SWIT3 RATE OF INCREASE VS TRIP POINT / / l / / A ,} ~' EN 195.4 195.6 195.3 190 186 p l l I Q l. ./,/ /l l d 1:1 2:1 3:1 4:1 .uu "3 DEGREES : MINUTES de/im ed

6. G 6 04 6

1. [5[?e cn

.}}