Text

Deposition of Wh Zewe on 820526 in New York,Ny.Pp 480-645
	ML20072H804
Person / Time
Site:	Crane
Issue date:	05/26/1982
From:	Zewe W; GENERAL PUBLIC UTILITIES CORP.
To:	;
References
	TASK-*, TASK-01, TASK-02, TASK-05, TASK-06, TASK-08, TASK-1, TASK-2, TASK-5, TASK-6, TASK-8, TASK-GB NUDOCS 8306290751
	Download: ML20072H804 (165)
	v • d • e

{{#Wiki_filter:_ _ _. _ Pk UNITED STATES DISTRICT COURT 480 SOUTHERN DISTRICT OF NEW YORK I _ _ _ _ _ _ _ _x GENERAL PUBLIC UTILITIES CCRPORATION, c.c. JERSEY CENTRAL POWER & LIGHT CCMPANY, METROPOLITAN EDISCH COMPANY cnd PStINSYLVANIA' ELECTRIC COM7ANY, a Pl air.tif f s, 80 CIV. 1683 (R.O.)

j..

-against-ThS BABCOCK & WILCOX COMPANY and J, RAY F.cDER)ltTT G CC., XNC., f l D Defendants. (t i i x Contie.ted deposition of' GIU NUCLEAR u s e CORPORATION by WILLIAM 11. ZEWE, taken by Defendants, pursuant to adjournment, at the offices of Davis Polk & Wardwell, E s g s,., One Chase Manhattan Plaza, New York, New York, on Wednesday, May 26, 1982 at 9:10 o'cloc'k in the forenoon, before Harvey B.

Kramer, R.P.R.,

l a Certified Shorthand Reporter and Notary Public within and for the State of New York. l m-k 3[MNhob l t j DOYLE REPORTING, INC. s I CERTIFIED STENOTYPE REPORTERS 369 LextNGTON AVENUE WALTER SHAPIRO, C.S.R. NEW Yon x, N.Y. 10017 CHARLES SHAPIRO, C.S.R. TELEPHONE 212 - 867-8220

1 481 A-2 Appe arance s: s 3 4 KAYE, SCHOLER, FIERMAN, HAYS & HANDLEF, ESQS. Attorneys for Plaintiffs <( 5 425 Park Avenua New Ycrk, New York 6 BY: LAVID KLIFCSBERG. ESQ., of Connuel g S 1 i 9 CAVIS POLK & MARDWELL, EJQS 4 j Atternsys for Defendants t l 10 One Chase Manhattan Plaza q' 11 hew York, 1Tev York i l 12 BY: ROBERT B. FISKI, ESQ. j { WILLIAM E. WURTZ, ESQ. 13 -and-RODMAN W. BENEDICT, ESQ., 14 of Counsel 15 e 16 KILLIAN & GEPHART, ESQS. 17 Attorneys for the Witness Box 886 18 Harrisburg, Pennsylvania 17108 19 BY: KEVIN WALSH, ESQ., 20 of Counsel ( Also Present: 22 JONATHAN QUINN 23 Law Assistant Davis Polk & Wardwell, Esqs. () 24 -oOo-25 p 4e. w w -,se-,- m. m--------..--. ,e,-- r.,,, ,-,-,-e .---,wm-, --n r ww w-

1 482 [ t 2 W I LLI AM H. Z EW E, having been 3 previously duly sworn, resumed and was examined 4 and testified further as follows: t ( 5 EXAMINATION (continued) (p 6 57 MR. ?IsKE: h 7 I Q Ifr. 7.aw e, going back to the early ainutes 6 ct t h e-a;cidect,

7. thin 3 you testified yesterday that 9

q both Itr. Fredurick and air. Paust wcra up there in 10 front of the connrols. b 11 g A Yes, they <dere. U 12 l' O I.' it correct trat in the tirst couple of (~') u ~' 13 minutes of the accident Mr. Faust was working on 14 controls having to do with the steam generator? 15 A Yes, he was. 16 Q Did it come to your attention in the 17 first couple of minutes of the accident that he was 18 having trouble getting feedwater into the steam 19 generators? 20 A Yes. 21 Q Did you go over to try to help him with 22 that? 23 A Yes, I did. ). 24 Q While you were over there trying to help 25 Mr. Faust get feed to the steam generators, did you

1 Zewe 483 (~h ~ (_f 2 learn that high-pressure injection had come on 3 automatically? 4 A I was aware that high-pressure injection 5 had actuated autom'atically, though I am uncertain i 6 at that time the exact time frame in relationship to 7 ) the emergency feed. V 0l c when you learned that high-prescure I 9 Injection had come on automatically, did you ye13 l ( l 10 out te verify act2ation and go to bypads? l 4 1 11 A hs I recall, I did, yes, k 12 Q And you were giving those instructions Ch 13 to whom, to which operator? 14 A As I recall, Mr. Frederick. 15 Q And then sometime after you had given e 16 those instructions to Mr. Frederick to bypass HPI, 17 did you learn that the 12's had been closed? 18 A I did learn that the 12's were shut from 19 Mr. Faust. Here again, the relationship between 20 high-pressure injection actuation and going to bypass 21 and finding that the 12 valves were shut seemed to be 22 at the very same time, as I recall. 23 I find it very hard to differentiate the p* 24 first ten or 12 minutes in time. G 25 Q It was kind of a big event when Mr.

1 Zewe 484 /~h V. 2 Faust discovered that the 12's were closed, was it 3 not? 4 A I was completely surp rised. ( 5 Q He yelled it out, didn't he? 6 A Yes, he did. 7 Q And you said something like, "Why do you j 8 have tha 12's ciceed? Why did you close che 12's?" l 9, And.hs said, "I didn't close them. They were closed." 10. A T just said, "Why are they shut?" And l l 11 he said, "That in how I found ther." And then he was

2 instructed to open then up and he was alree.dy doing 13 that.

14 Q Let me, in an effort to help you re' fresh 15 your recollection as to the sequence of events of 16 the first few minutes, show you your testimony that 17 you gave to the Nuclear Regulatory Commission 18 Inspection and Enforcement Group on April 12, 1979. 19 This has not been marked previously, so 20 we will now mark it as Exhibit 748. 21 (Testimony of Mr. Zewe to the Nuclear 22 Regulatory Commission Inspection and 23 Enforcement Group on April 12, 1979 marked 24 B&W Exhibit 748 for identification, as of this 25 date)

~. r a reduction 11 in pressurizer level and a reduction in RCS pressure. 12 Q And isn't it a fact, as we established 6 i \\_/ 13 earlier in the deposition, that in those other three 14 situations where high-pressure injection had come 15 on automatically, there had been a drop in 16 temperature of at least 30 degrees within a period 17 of one or two minutes? i 18 A As I recall, yes. 19 Q Now, looking at the instrument' panels 20 that were in front of you and the other operators in 21 the control room, did you witness a drop in 22 temperature in the reactor coolant system between 23 the time of the trip and the time when high-pressure f) 24 injection had come on? 'v 25 A Yes, we did. l

1 Zewe 520 ' ~ ' 2 Q Wasn't that drop in temperature 3 approximately five degrees? 4 A I don't remember what that drop was 5 particularly. 6 Q Well, let me show you a document, Mr. 7 Zewe, which has been marked Exhibit 279, which is 8 something called the memory trip review. 9 Do you have that in front of you? 10 A Yes, I do. t 11 Q That is a document that comes off the 12 computer, is it not? (~ ( )h 13 A Yes, it could. 14 Q Am I correct tha't this particular 15 document prints out information for a time period 15 16 minutes.before a trip and 15 minutes after a trip? 17 A Generally speaking, yes. 18 Q And someone did punch the button on the 19 morning of March 28 in order to produce this memory ~ 20 trip review, did they not? 21 A I can't read the date of this in the (_ 22 upper left-hand corner. 23 I can only assume that this is the 24 document from that morning. 25 Q I think that has been established in

1 Zewe 521 ~ 2 o,ther depositions. 3 A Okay. 4 Q And my question, pending question, is: ( 5 this document was produced from the computer at some 6 point within the first hour and a half of the 7 transient, was it not? 8 A I would assume so,.yes. 9 Q The GPU chronology filed with the NRC as 10 part of the LER on this transient places the time 11 at which the memory trip review was produced as 4:30. 12 Does that help refresh your memory as to the time? 13 A No, it doesn't. 14 Q Now, the reactor tripped on high 15 p re s sure, did it not? 16 A Yes, it did. I 17 Q That is 23 -- what is the setpoint? 18 A 2355. 19 Q 2355. And high-pressure in'jection is 20 actuated at 1640; is that correct? 21 A That is correct. 22 Q From your understanding of the system 23 on the morning of the accident, Mr. Zewe, did you ( 24 expect a drop in temperature in the reactor coolant 25 system of approximately five degrees to produce a I r ,,,-----e v-- + - - - -,. -, - - -, ww-er--.--,-w, ,y-,,,v.--w, ,w, mwv-t

v-we w*-
---=---a---+~w v'+ -*e
1 zewe 522 -(J 2 drop in pressure from 2315 down to 1640? 3 MR. KLINGSBERG: 23 4 Q from 2355 down to 16407 ( 5. A I would not. 6 Q The fact is, is it not, that the e #~ n 7 temperature drop between the time of the reactor 8 trip and the time when high-pressure injection came 9 on was no more than five or six degrees? 10 MR. KLINGSBERG: Are you asking him to 11 read the memory trip review or to tell you his 12 recollection? O 13 MR. FISKE: Let's start with the memory 14 trip review. 15 THE WITNESS: Would you read that back, 16 please? 17 (Que s tion read) 18 A I would conclude from this memory trip 19 review that the temperature in fact had dropped by 20 about six degrees. 21 Q Before you concluded, Mr. Zewe, that this 22 was not a loss of coolant accident, did you look at 23 the temperatures? () 24 A I do recall looking at the T-average 25 temperature,.though I don't recall calculating in my
1 Zewe 523 ~ 2 mind whether the T-Av at that particular point when 3 I looked at it is what it should have been for the 4 p re ssure reduction that I had indicated. {' 5 Q Let me show you a document we will mark 6 as B&W Exhibit 750. And I will represent to you, 7 Mr. Zewe, that this is a strip chart of T-Av on the 8 morning of the accident. 9 (Strip chart of T-Av on the morning of ~ f 10 the accident marked B&W Exhibit 750 for I 11 identification, as of this date) 12 Q Do you have that document in front of b N/ 13 you, Mr. Zewe? 14 A Yes, I do. 15 Q That strip chart for T-Av is displayed 16 in the control room, is it not? 17 A Yes, it is. 18 Q How far is it from the strip chart that 19 displays the pressure in the reactor coolant system? ~ 20 A Close proximity. 1 21 Q A couple of feet? 22 A Less than that. 23 Q And both of these are on the front panel, f'T 24 are they not, along with digital readings for s_J 25 p re s s ure and temperature? n. ---=,w--,,,,--nc-, n,, ,. w, nw_,,.,--n-,.,c.m.,-.,-,,,,m_,,---,,w,,,,,,..,,..,,__.,.,,., ,.,.n--, ,.,..,---,--.,a.,.,,,.,,..a
1 Zewe 524 (~) 2 A We do not have digital readings for 3 pressure that I recall. 4 Q Okay. ( 5 A We do for temperature. 6 Q where is the instrumentation for 7 pressurizer level? 8 A on the main console. 9 Q Directing your attention to the document 10 we marked yesterday, the l a'yo ut of the teontrol room, 11 Exhibit 745, do you have that in front of you? 12 A Yes, I do. 13 Q Do you see the arrows that indicate the 14 location for temperature? 15 A Yes, I do. 16 Q That is marked by a T, is it not? 17 A Yes. 18 Q The next one down is P; that is pressure? 19 A That's correct. ~ 20 Q The next one is PZRLV. Is that 21 pressurizer level? 22 A That is correct. 23 Q And are the strip charts for pressure ( 24 and temperature in the same location as the T and P 25 indications on this?
1 Zewe 525 O V. 2 A As I recall, they are right beside each 3 o t-he r. 4 Q It is fair to say, isn't it, that the J{ 5 indications for temperature and pressure in the 6 reactor coolant system, both in terms of an instant 7 reading and in terms of a strip chart, are right in 8 there, sort of side by side with the instrumentation l 9 reflecting pressurizer level? 10 A That is correct. L 11 Q Do you know whether Mr. Frederick 12 looked at any of the instrumentation reflecting fh N-13 3 temperature in the reactor coolant system in the 14 first few minutes of the accident, before he 15 throttled high-pressure injection? 16 A I don't know. You will have to ask him i 17 that question. l 18 Q We did, Mr. Zewe, for your information. I i 19 Do you know whether Mr. Scheimann looked 20 at the information for reactor coolant system 21 temperature within the first few minutes of the 22 accident, before HPI was throttled? l I 23 A Again, you would have to ask Mr. [) 24 Scheimann that. V 25 Q Did you discuss pressurizer level with l l v ,w-, ---,m ,4- -n,- ---,,,.,,-m- ,-vv- +
1 Zewe 526 (\\ \\ - 2 Mr. Frederick or Mr. Scheimann before HPI was 3 throttled below 1,000 gallons per minute? 4 A Yes, we did. 5 Q Did you discuss reactor coolant system ( 6 Pressure with them before HPI was throttled below a 7 thousand gallons? 8 A I don't recall discussing reactor coolant 9 system pressure before we throttled. 10 Q Did you discuss the reactor coolant 11 system temperature with them before the HPI was 12 throttled below a thousand gallons? (")% 's 13 A I did not. 14 Q Did it occur to you, Mr.Zewe, that before 15 you determined that high-pressure injection had 16 come on for some reason other than a loss of coolant 17 accident, you should look at the temperature to see 18 whether or not there had been overcoolingi l 19 THE WITNESS: Read that again,'please. l 20 (Question read) 21 A I had made the determination that the 22 reduction in pressure was due to the initiation of 23 emergency feed. And as I recall, when I did look at 24 the RCS temperature, the temperature was down in the 25 neighborhood of 540 to 545.
1 Zewe 527 2 MR. KLINGSBERG: I am sorry. I missed 3 something there. Could I have the beginning of 4 that? { 5 (Record read) 6 Q Now looking at the strip chart, Mr. Zewe, 7 that reflects, does it not, that RCS T-Av did not 8 reach the temperature did not reach the 9 temperature of 550 degrees for more than 20, 25 10 minutes into the accident; it didn't reach 545 until 11 approximately 30 minutes into the transient? 12 A That is what the chart shows. But if you 13 are asking me what I recall from that morning, I 14 recall the RCS as being between 540 and 545 when I 15 looked at it. 16 Q Just so we understand each other, the 17 first time you looked at it, it was 540 to 545? 18 A As I recall, that's what I read, yes. 19 Q Do I understand your prior testimony ~ 20 correctly that it was the existence of high 21 pressurizer level that led you to the conclusion 22 that you did not have a loss of coolant accident at 23 the time you allowed the operator to throttle back 24 HPI below a thousand gallons per minute? 25 A That was the primary indicator that I t-t'-'mr-

r--e v-+=-'*~'e

+r' c- - - + ' - - - - -

m--t--e->*-

'e-
=-*-
y r w'- v--
1 Zewe 528 2 used in making that determination, which was 3 consistent, based upon my previous training and 4 experience at the island and at the transient ( 5 response at B&W, Lynchburg. 6 Q I think, going back to these four 7 transients that we discussed before in which HPI had 8 come on automatically, the three of them that we had 9 previously characterized as overcoolings each 10 involved a situation where pressurizers. level had 11 dropped, did they not? 12 A That is correct. ) / 13 e Q Indeed, is n ' t it correct that in each one 14 of those three cases, pressurizer level had gone-all 15 the way down to the zero level on the scale, or close 16 to it in the case of the third one? 17 MR. KLINGSBERG: Are you talking about in 18 some particular point in time? 19 Q Isn't it correct, M r.- Zewe, th'at in two 20 of the three overcooling transients, pressurizer 21 level had dropped very rapidly down to the zero 22 level on the scale, and in the third overcooling 23 transient it had come very close to the zero level () 24 within a few minutes after the trip? i %_/ .25 A That is correct. l
5 1 Zewe 529 2 Q And from your understanding of the 3 dynamics of the way a cooldown works, as you have i 4 described it before, the drop in temperature produces ({ 5 a shrinking of the volume of the water in the 6 reactor coolant system, which causes both the 7 pressurizer level to fall and the pressure to fall; 8 isn't that correct? 9 A That is correct. '10 Q All right. And three of the situations 11 in which you had previously seen high-pressure 12 injection come on automatically had involved that i 13 kind of a situation, with a drop in temperature and 14 a drop in pressurizer level as well as a drop in 15 pressures right? 16 A That is correct. 17 Q So the existence of a high pressurizer 18 level was completely inconsistent with th$t j j 19 situation, was it not? 20 A "That" situation? 21 Q Being a cooldown. Putting it simply, a ( 22 high pressurizer level was totally inconsistent with i 23 a cooldown. 24 MR. KLINGSBERG: At what point in time? i 25 MR. FISKE: I am getting comments on both ] -,r. .--,,,e.., . - -. -,..,. ~. - - -,. - -,., _, -, - -,,,... -,, .n, ,,,,---,n_,,--,,-.---,-_-n. ,,a.. ,,.,.,,,. +.,
1 Zewe 530 [~h ('/ i 2 sides here. What was that? s 3 MR. KLINGSBERG: I said, at what point in 4 time. { 5 Q The existence of a high pressurizer level 6 was inconsistent with the three situations in which 7 you had previously seen HPI come on as a result of a 8 cooldown; isn't that correct? 9 A That is correct. 10 Q And isn't it correct that the indication 11 of the pressurizer level was in effect directly 12 opposite to the concept of a cooldown of the type 13 that had previously produced HPI actuation? 14 A Yes. 15 Q Before you concluded that HPI had come 16 on as a result of a cooldown, did you stop to 17 consider the fact that the high pressuriz,er level f 18 was totally inconsistent with that conclusion? ( 19 MR. KLINGSBERG: You are talking now 20 I about the day of the accident? / l / l 21 MR. FISKE: Yes. l 22 MR. KLINGSBERG: I didn't know that there 23 was such testimony. ( 24 MR. FISKE: I thought there was. 25 I will put it another way.
1 Zewe 531 n 2 BY MR. FISKE: 3 Q Before you concluded, Mr. Zewe, that 4 HPI had not come on"as a result of a loss of coolant C 5 accident, did you consider the fact that a hi,gh 6 pressurizer level was inconsistent with, indeed 7 contrary to, HPI having come on as a result of a 8 cooldown? i 9 A As I recall, it was inconsistent, yes. 10 Q Did you come to that conclusion back 11 at that point in the transient? 12 A Throughout that morning I was trying O \\/ 13 to evaluate on why the pressurizer level was so 14 high. I did not understand why it was, and I had 15 reached no conclusion on exactly why it was high. If 16 the indication was correct, I did not know why the 17 pressurizer level was high. 18 Q And at the time you allowed th'e operator 19 to throttle back HPI in light of the high pressurizer 20 level, at tha t point in time, before you allowed the i 21 operator to do that, did it occur to you that you 22 should check the temperature at that point in time, 23 to see whether or not there was, there had been, [~h 24 an overcooling? %.) 25 A I don't recall at that particular point r- -,.o-e--- ~ + - - - - - - - + - '
i Zewe 532 On# che' king the temperature. There were other things 2 c 3 going on in the control room and the operator said 4 that he was going to throttle, and I concluded that ( 5 it was because of the recovery of pressurizer level, 6 which I observed, and I was in complete agreement 7 with him throttling. 8 Q Now, in trying to figure out, Mr. Zewe, 9 what was causing this high pressurizer level, did 10 you think back to these three -- to the. transients 11 that had occurred before, right there at Unit 2, 12 in which high pressure injection had been automatically (D 'w ) 13 actuated? 14 A I did not. I was.trying to evaluate 15 what we had in this particular case, which was 16 unlike any of the other transients that [ had been 17 trained on or had heard about, and I was trying to 18 evaluate this because it was so particular to itself. 19 Q -Going back to the four situations in 20 which HPI had come on automatically before, you did 21 know you said that a high pressurizer level was l 22 inconsistent with the cooldown which had produced 23 three of the actuations, correct? [~) 24 A That is correct, but (/ 25 g And l
1 Zewe 533 O 2 MR. KLINGSBERG: You keep cutting him off. 3 A You are inferring that -I had these four 4 or those three transients in mind that morning, to 5 compare this particular one with those three or { 6 four that were in mind, and assuming the information 7 that I had learned from those transients I was trying 8 to apply to this particular case. 9 That morning I was trying to deal with 10 what I had and trying to use whatever knowledge 11 and training I had had up to that point with 12 everything, and it just didn't fit in. /'N (j) 13 Q Didn't anyone at Met Ed, any time after 14 these four transients had occurred in which the HPI 15 had come on automatically, try to give the operators 16 some guidance as to how in the future, when HPI 17 came on automatically, they could tell whether the I 18 HPI was coming on because of an overcooling or whether 19 HPI was coming on because of an open PORV?' 20 MR. KLINGSBERG: I object to the form 21 of the question. 22 MR. FISKE: All right. 23 THE WITNESS: Would you read that'back, 24 please? j 25 (Question read) _,. ~.. - - _ _ _.. _. _..
i Zewe 534 (~) \\~/- 2 MR. KLINGSBERG: You are assuming those 3 are the only two possibilities? 4 MR. FISKE: Those are the only four that 5 there were. 6 MR. KLINGSBERG: No, that an overcooling 7 or open PORV are the only two possibilities. 8 MR. FISKE: I think that is correct, in 9 terms of the transients that had been 10 experienced before that. g 11 THE WITNESS: I am sorry. Read that 12 again, please. 13 Q I can put it to you again,lHr. Zewe. ~ 14 A That's okay. 15 Q Didn't you and the other, operators receive 16 training from other people at Met Ed as to how, in 17 light of the four actual transients that had occurred 18 at Met Ed in which HPI had come on automatically, 19 you could tell in the future, when you had an 20 automatic HPI situation, whether that was due to an 21 overcooling or whether that was due to an opan PORV7 22 MR. KLINGSBERG: Same objection. 23 A As I recall, any of the training, {^} 24 including the training of these four cases at Met v 25 Ed or B&W, never pointed to the fact of the response l
1 zewe 535 /^% k- - 2 of the HPI and how it would be differentiated between 3 a cooldown and an open PORV. It was always put in 4 respect to a cooldown versus a loss of inventory 5 during a LOCA condition. It was always within that ( 6 frame, and even up to the point of my most recent 7 B&W training which was in January of '79, that 8 training was consistent with my previous training 9 that if I had a LOCA, which a PORV would be, I would 10 have classical LOCA symptoms and immediate actions 11 that would be taken based on loss of inventory, and 12 that an overcooling accident -- which is more'true ) 13 than a cooldown, all right it is reilly an cooldown'is j 14 overcooling accident, -because a a 15 control thing, where an overcooling event is 16 practiced and it is trained on so we do differentiate 17 between a LOCA and an overcooling condition. 18 Q Picking up on one thing you said in 19 that answer, you didn't receive any specific 20 training on'those four transients when you went to 21 B&W in January '79, did you? 22 A As I recall, we had overcooling transients. 23 No t that were exactly like those which occurred, but (~N, 24 they were similar. Steam leaks, overfeeding, and \\ / 25 also LOCA's.
1 Zewe 536 2 Q Having concluded, as you previously 3 testified, that existence of a high pressurizer 4 level was contrary to an overcooling situation, did 5 you consider during the accident, based upon the { 6 March 29, 78 HPI actuation, whether a high pressur.izer 7 level might be consistent with an open PORV? 8 A I never considered that a high pressurizer 9 level would be in any way consistent with a loss of 10 inventory which resulted from an open P RV, whether 11 it was in relationship to the March 29 accident the 12 year before or any training that I received. 13 Q You testified earlier, Mr.'Zewe, that 14 you understood -- prior to the Three Mile' Island 15 accident -- that if the PORV remained open, that 16 you would lose water from the system. 17 Do you remember giving that testimony 18 earlier in this deposition? 19 A I remember stating that I would lose 20 reactor coolant system inventory if the PORV was 21 OPen, that is correct. 22 Q All right. And in connection with that 23 understanding, didn't you understand that in order 24 ~ for the water to go out through the PORV, it would 25 have to go through the pressurizer on the way? i l .,--.-.,-c. .r,-,yy.._.---..m.-_.--- .-,,_-.,,,.r,,,,,. --..r--w. e---. c-.- - .---_mm,.,.-r 7,
1 -Zewe 537 O' 2 A PORV is located on the top of the 3 pressurizer itself. So the inventory loss would come 4 from the pressurizer. But it would be reflected i~n 5 a loss of inventory from.the system, as indicated { 6 by pressurizer level. 7 Q But didn't you understand that in order 8 for the water in%the system to go out through the l 9 PORV, that it would have to go through the pressurizer 10 on the way? t 11 MR. KLINGSBERG: He just answered that. 12 MR. FI5KE: I am not quite sure I 13 understand the answer. 14 A My understanding was that if'the PORV 15 was open, that I would lose inventory that was in { 16 the pressurizer and that, in effect, it would 17 reflect that reduction in inventory in pressurizer, 18 level. 19 Q I believe your answer earlier in the 20 deposition, Mr.-Zewe, was that you recognized that if [_ 21 the PORV stayed open, you would lose water from the 22 reactor coola'nt system. Is that what you mean by 23 " inventory"? ~ 24 A Whether it be steam or water, it is 25 inventory. Because the pressurizer has steam and
8 1 Zewe 538 2 water both, and when you open up a valve in a high 3 temperature, high pressure system, depending on the 4 downstream pressure, you will flash the water to i 5 steam. So it is just inventory. The steam and 6 water is the inventory of the RCS, and it would pass 7 that inventory. 8 And depending on what the conditions were, 9 it could be steam or water. 10 Q And the reactor coolant sys. tem remains 11 under very high pressure, correct, while this PORV 12 is open? You understood that? 13 A I did not understand that. 14 My understanding was that the' p.ressurizer 15 was used to maintain inventory in the RCS and also 16 as a means of pressure control of the RCS, in that 17 it had saturation conditions, so that if I would 18 reduce inventory from the RCS or if I would reduce i j 19 the pressure in the pressurizer which is part of l 20 the RCS, I would reduce some pressure. 21 Q But you did understand that while the l l 22 PORV is open, the reactor coolant system is still l l 23 under pressure considerably higher than atmospheric 1 ("% 24 pressure right? ( 25 A As I recall, the only time that the PORV l l l t
1 Zewe 539 O 2 cpens is at 2255 psig, and it is to reduce the 3 pressure back to 2205 psig, and then reclose, and 4 that is actually what the pressure is for the RCS 5 itself that is maintained by the pressurizer. ( 6 Q I guess my question is fairly simple. 7 Didn't you understand when you said earlier than 8 an open PORV could produce loss of inventory through 9 the open valve, that the way that would work is sort 10 of like shaking up a coke bottle and watching 11 everything go out the top, that the pressure in the 12 system would force inventory out through the hole 13 e in the top? 14 MR. KLINGSBERG: I don't think the witness 15 testified to a coke bottle. 1 16 MR. FISK3: No, that is my analogy. 17 MR. KLINGSBERG: You said, "Didn' t you. 18 say?" 19 MR. FISKE: No. ~* 20 Q Wasn't that your understanding? 21 A My understanding was, as I have just 22 stated, that the PORV would open at setpoint, would 23 close at setpoint to relieve the pressure, the 24 overpressure condition, and that's how I remember 25 that it was just used as an overpressure relief
1 Zewe 540 O( 2 valve, and that was all. 3 It was not designed to continuously 4 relieve reactor coolant system inventory and { 5 continually reduce pressure. 6 Q I guess maybe my question wasn't too 7 clear. 8 I wasn't asking you about the normal 9 operation of the JORV, what it was supposed to do. 10 My question went back to some questions 11 I asked you in an earlier part of this deposition 12 about training that you received at Met Ed after 13 the March 29, '78 incident, in which the PORV had 14 failed open. And I asked you whether or not you 15 were trained at Met Ed as to what the consequences 16 might be.if the PORV stayed open when it was supposed l 17 to shut. i l 18 I believe your answer was that you didn't 19 receive any specific training on that, but' that you i ( 20 did understand that if the PORV remained open, you 21 would lose water from the system. 22 All I am trying to ask you now is, isn't l 23 it a fact that it was your understanding that the ( 24 way that would work -- that is, in the case of an l 25 open PORV, it would stay open -- was that the I i
I A ' I i 1 zewe 541 f) 2 pressure in the sysrem would force the water out the 3 hole at the top. i ~ 4 MR. KLINGSBERG: I think the witness 5 has answered it. g 6 MR. FISKE: This is where the rather 7 clumsy analogy to the coke bottle came from. 8 MR. KLINGSBERG: I object to the form 9 of the question. I 10 MR. FISKE: Okay. ( 11 A It is true that I was aware that in order 12 to have any flow of inventory out the PORV, you have s 13 to create a delta pressure in order to have any flow. 14 The only training that I can recall or even thinking 15 about with the PORV open was that you would lose .s. 16 inventory from the system, whether it be steam or 17 water, and that it would be reflected by a reduction + 18 in pressurizer level as indicated and the pressure 19 would also go down. And whenever you woul' ' clos e d 20 the valve, the system would recever. Or in the 21 normal operation, 2255 and close again at 2205 psig. 22 Q In other words, it was your understanding 23 before the accident that even in the case of a loss ( 24 of coolant accident which involved water going out 25 the top of the system through the PORV, you would - i - s ~9 e .vir,, y '-y e ,,=--3 y-- ew---e,- y.-w -.,gy y q gm- -ww,--s---.
s 1 zewe 542 \\ 2 see a drop in the level of water in the pressurizer? 3 A I considered that an open PORV would be 4 a loss of inventory and I expected to see a i 3 5 corresponding reduction in pressurizer level due to ( 6 that loss in Inventory. 7 MR. FISKE: If you want to take a break 8 now, it.is okay. s 9 (Recess taken) 10 BY MR. FISKE: e 1 11 Q Mr. Zewe, going back to one.answerfthat 12 you gave just before the break, was it your y O s i 13 understanding, in the case of a failed open or stuck' [ w 14 open PORV, that the level in the pressurizer would a W 15 come down even during the period while water was l c 4 16 going out through the open PORV? 4 17 A I don't recall. making that differentiation, 18 while it was happening or what have you. 'But I 19 remember that I. thought that if the POR7 was open, 4 20 that pressurizer level would reflect the inventory 'i 21 loss and it would be reduced. { V, Hi e 22 I had never seen any training to the \\ 23 contrary, where any time I had a loss of inventory s ( 24 by whatever means', that pressurizer 1 vel did not 25 indicate a reduction in response to thAt loss of 3 ,\\ \\ s r x ____ _ J A\\
s n 1 Zewe 543 ,2 s inventory. 3 Q Did you have training at Met Ed or in 4 N the Navy or at B&W or anyplace that told you that s.'5 i - if you had a hole in a system that was under ~ ( 6 pressure, that the water would be forced out that 7 , hole by the pressure? / l 8 A Like I stated before, a hole in any y 9 spressurized system, in order to get flow you need 5 s 10 to have a differential pressure, and in the case of 11 a hole you are using the system pressure inside the 12 vessel or inside the pipe as opposed to the pressure i s /' 13 i ' outside, whether it be atmosphericopressure or another 14' system. ~ 15 So you need that delta pressure in order { to E 16 induce the flow or you would have no transfer of 17 inventory. j '18 Q And you certainly understood,'did you 19 inot, that water would go out a hole in the' system i 20 a lot faster if the system were under pressure than 21 if it were not under pressure? C. ?" A As I just stated, if there was no 23 prissure difference between, there would be no ! ' NY /~N 24 transfer of inventory, other than maybe diffusion or l25' so:tething of that nature. It relied on delta pressure s t
.:[.v l 5 s
s
~ Zawa 544 1 ("h \\'-) ' 2 in order to get movement or flow. 3 .Q one last questi'on on this. I believe one 4 last question. 5 ( Didn't you understand that in order for 6 the water to go through the open PORV, that before 7 it did that, it would have to go through the top of 8 the pressurizer? 9 MR. KLINGSBERG: Water from the 10 pressurizer or water from the reactor coolant 11 system? 12 MR. FISKE: Well, whatever water was 13 going out. 14 A As I recall, I never considered actually 15 having water flow out of the pressurizer. I always 16 assumed that the steam that was in the top of the 17 pressurizer would go out through the open, valve and 18 then the valve would reclose long before you had 19 actually reached the water going out, that'it would 20 only be open for that time to restore the pressure, 21 and that you would never arrive at that point. C 22 Q Going back now, Mr. Zewe, to this point 23 in time during the accident when you allowed t'he I 24 operator to throttle HPI back below 1,000 gallons 25 per minute, I think you have testified that at that
1 zowe 545 (3 2 time you had determined that this was not a loss of 3 coolant accidents is that right? 4 A As I remember, I concluded that we did 5 not have a loss of coolant accident, that is correct. { 6 Q Did you know at that point of any cause 7 for a drop in pressure down to the point of high - 8 pressure actuation other than a loss of coolant 9 accident or an overcooling, as you put it? 10 THE WITNESS: Read that back, please. 11 (Question re&d) 12 A As I can recall, those are the only two 13 transients that would result in pressure being 14 reduced to the high-pressure injection se'tpoint. 15 Q And it is correct that high pressurizer 16 level, as you understood things on the day of the 17 accident, was inconsistent with both a LOCA and with 18 an overcooling? ~ 19 A As I have stated before, yes. ~It was. 20 inconsistent with either of those two. 21 Q What did you understand would be a result 22 if you had an overcooling and you allowed HPI flow 23 to continue without throttling back? ['~Y T 24 A My understanding of what would happen? = 25 Q Yes, or what could happen.
1 zowe 546 O 2 Putting it another way, what was 3 undesirable about allowing HPI to continue at full 4 throttle if this were an overcooling? { 5 A First of all, the operator would take 6 action necessary to at least control or completely 7 stop the overcooling if he recognized that he had 8 that event. And then, if the makeup and high-9 pressure injection was on, the operator would then 10 turn toward taking control of that system in order 11 to not to overfill the pressurizer and go solid. 12 Q Do I understand from that answer,'Mr. s 13 zowe, that the reason to throttle HPI flow in an 14 overcooling is to prev'ent the system from going solid? 15 A There again, it isn't that simple. You 16 have to.look at the overcooling and combating the 17 overcooling along with the increased makeup flow I l 18 provided by the high-pressure injection. " ll right? A 19 But it is true that you would throttle the'high 20 pressure injection in order not to go solid in the i 21 RCS. Once you have recovered pressurizer level, 22 there would be no need for continued high flow rate i 23 from high-pressure injection. You would only need i j ( 24 that required to recover your desired pressurizer 25 level. And then you would cut back on it.
1 Zewe 547 O 2 Q What did you understand would be, might 3 he the adverse consequences if you had a loss of 4 coolant accident and you throttled HPI when it should 5 have been left on full? g 6 A I don't recall'ever considering throttling 7 HPI if I had a loss of coolant accident, until the a 8 point was reached that the loss of inventory out 9 the break was exceeded by the makeup capability. 10 And I would restore the inventory that.was lost 9 11 and recover the reactor coolant system inventory as 12 indicated by pressurizer level ;then I would consider O) k-13 throttling. 14 Q Putting it in its simplest terms, didn't 15 you understand that one of the reasons that HPI came 16 on in a loss of coolant accident was to prevent a 17 situation in which there was core uncovery and 18 possible core damage? 19 A I understood that the main reason for 20 the HPI was to insure reactor coolant system inventory l l 21 within its capability under all circumstances. Q Okay. And one reason to do that was i 23 to prevent core uncovery and possible core damage, ( 24 right? 25 A I don't remember thinking in terms of e r- ,,.,-,-n ---e, 4 e

,,weng,----

1 Zewe 548 2 core uncovery. It is just that if you had a LOCA, 3 you would use it to its design capability of 1,000 4 gallons per minute for the HPI, and if ever you 5 would get low enough to where you would have core 6 uncovery, your pressure would be low enough to where 7 you have other emergency feature safety systems like 8 core flood and low-pressure injection provided by the, 9 decay heat removal system to insure that the core 10 always had adequate inventory so that it could remain 11 cool. I never thought that the high-pressure 12 injection system was to keep the core covered by h) s_/ 13 itself. 14 It was the engineering safety features 15 system that was designed for all the various failure 16 modes and size of breaks that would cover the broad 17 spectrum that would afford that. Not only the HPI. 18 Q so is it your testimony that l'eaving HPI 19 on when it was automatically actuated was not 20 important in terms of preventing possible core 21 tacovery? 22 MR. KLINGSBERG: I object on the ground 23 that that was not his testimony. ( 24 A Would you rephrase that, please? i 25 MR. FISKE: Could you read it back? Then 4 y-,- - --, -,, -, ~.,, -,, -, .,wsw-- w,, -,e,e, y v-
m. 1 Zewe 549 2 I will decide. 3 (Question read) 4 Q Can you answer that question? 5 A Yes, I can answer that. But I don't 6 believe that's what I said, and it is certainly not 7 what I believe I said. 8 Q I didn't suggest you had said it. 9 A' I am sorry, but I thought from that that 10 you were suggesting that I had said that. H 11 Q No. I am asking you whether that is your 12 testimony, not whether it was. n\\_/ ' 13 MR. KLINGSBERG: It gets very confusing 14 when you say "is.your testimony." That sounds 15 like, "Is it your testimony that you have 16 previously given?" I think it is much clearer 17 to say to the witness, "Is it the fact, is it 18 the case, is it correct." ~ 19 MR. FISKE: Let's hear the question again. 20 (Question read) 21 MR. FISKE: I will rephrase that. 22 Q Isn't it correct, Mr. Zewe, that you 23 did understand that leaving HPI on when it had come 24 on automatically was important in terms of preventing 25 possible core uncovery? 1
1 Zewe 550 0 2 A As I recall, it was my training and 3 understanding that the' high pressure inj ec tion system 4 was important to be'left on during a' loss of coolant { 5 accident so that it,Jin conjunction with the other 6 engineering safety features system, could provide 7 enough reactor coolant system inventory in order to 8 keep the core cool. 9 (Continued on next page) 10 t. 11 12 13 14 i 15 i E s 16 17 18 19 l 20 21 l 22 1 23 24 25 1 ,--,--._____,-._,-m .,_. - m ,__,.__.,..,.--,,_.,_y- ,,,.,_,_-,__.,._,.-_-,.m ,--,-__._.___._-,.y_,--, ,.,-_..,c.. .-._,,p_,,
1 1
Zewe 551 -) O 2 Q What did you understand on the day of 3 the accident, or at any time up to the accident, was 4 bad about having the system go solid? 5 A From my training and experience, it was ( 6 my understanding that you would have a resultant 7 uncontrollable pressure excursion in the RCS if the 8 plant was solid; and if you would suddenly drain an 9 amount of water from the system or relieve pressure 10 in some manner from the system or heattup the 11 system or start a reactor coolant pump while it was 12 solid, that you could either have an increase or a 13 decrease in pressure that would be very rapid in 14 response to the solid condition. 15 And this was brought out many times, 16 that it was very undesirable to go solid. This had 17 been brought up while I was in the Navy, ,a t TMI, and 18 at B&W, Lynchburg, that going solid was something 19 that you should not do, that you should avoid that ~ 20 if at all possible, and that we had precautions and 21 even tech specs that applied to pressurizer level 22 and having pressurizer level too full. 23 And it was all of these experiences and ( 24 training that cautioned me, if at all possible, do 25 not take the plant solid, which meant to increase the
552 1 Zewe \\- 2 pressurizer to its fullest capacity. 3 Q what adverse consequences did you 4 understand could happen, could result, from a sharp 5 increase or decrease in pressure in the course of g 6 solid conditions? 7 A Well, the main consequence was that the 8 operator would not have control of pressure, that it 9 would be very hard for the operator to control 10 pressure very accurately, and that it may cause 11 component failure to have a rapid increase in 12 pressure and have the safety valves on the .r (_)\\ 13 pressurizer as the upper pressure limit for the 14 pressure excursions. 15 Q Did you have an understanding, before 16 the accident, as to whether it was more harmful to 17 the plant to go solid or to uncover the core? 18 A As I recall, my training and e'xperience, 19 never at any time did it pit one against the other, 20 that relationship, uncovering the core versus going 21 solid. k. 22 I believe that I knew that uncovering 23 the core is worse, in my own mind, but it was never 24 said, " Hey, weigh this against this." Never in that 25 light, in that context. l
1 Zewe 553 2 Q You did know befo,re the accident that 3 uncovering the core could produce a situation in 4 which there was an increased risk of releasing (( 5 radiation in the atmosphere? 6 A Yes, that's true. 7 Q And is it correct that you understood 8 that releasing radiation to the atmosnhere was a 9 worse situation than anything that would result from 10 going solid? ( 11 A Yes, because you are looking at two 12 different extremes, one where I have too much 13 inventory and in the other case, where I have too 14 little inventory, and I felt the only time you could 15 ever get core uncovery is if you had a very large 16 leak that was greater than the capability of the 17 engineering safety features system combined that you 9 18 had available at that particular time, and that you 19 may have some core uncovering only in that case where 20 you could not make up the inventory by low-pressure i 21 injection and high-pressure injection and core 22 flood. 23 That was the only case that you could f) 24 ever reach that condition. V 25 And I never thought, again, having too
l 1 g*** 554 2 much inventory would certainly not be as bad as 3 having too little inventory, but I was to avoid both 4 as best I could. (( 5 Q I,t is a fact, is it not, Mr. Zewe, that 6 at the time you allowed HPI to be throttled, you 7 did not think that the system was actually solid? l 8 MR. KLINGSBERG: Which throttling are i 9 you talking about? t 10 MR. FISKE: The same one we have been 11 talking about. 12 MR. KLINGSBERG: We have been talking j 13 about to a thousand and below l 14 MR. FISKE: Fair enough. l 15 BY MR. FISKE: 16 Q I am talking about the time at which yo'u i I 17 allowed the operator to throttle the HPI down to a 18 thousand. 19 A As I recall, I thought at the time that ~ 20 we were very, very close to being solid, but I had 21 not seen what I thought would be the signs of l (. 22 actually being solid: the uncontrollable pressure i 23 spike,-though I felt that our efforts in increased () 24 letdown and throttling our high-pressure injection 25 were keeping us from going that last little bit of .,---r -*y. ra ,_,,----,---,-m-,,, ,_,----y- --,--yy--r,.---.-..,--g-,,.y. ,,.r,- e--w.-- w- -.n m.e-e.,---._%.. %--4.--
555 -5 1 Zewe O 2 going completely solid. And from our indication, we 3 were extremely high, around 385 to 400 inches. So 4 I felt that our efforts were helping us from not (( 5 going completely solid. 6 But I in fact thought at the time that 7 we were close to solid, but in fact that we we re ' not. 8 Q Let me show you a document which we will 9 mark as the next B&W exhibit, 751, which is one of 10 the graphs that comes out of the LER that was filed 11 by GPU on this accident. This one reflects reactor 12 coolant system pressure, saturation pressure and 13 pressurizer level for the first 30 minutes after the 14 turbine trip. 15 (Graph from the LER filed by GPU was 16 marked B&W Exhibit 751 for identification, as 17 of this date) w 18 Q Do you have that in front of you, Mr. l 19 Zewe, Exhibit 7517 l ~ 20 A Yes, I do. 21 Q Do you see the line on there, the dotted I 22 line that reflects pressurizer level during the first ] 23 30 minutes? ( 24 A Yes, I do. 25 Q There was, as you said earlier,
1 6 gewe ~ 2 instrumentation on the front panel of the control 3 room that showed pressurizer level; right? 4 A There is a recorder, yes. ( 5 Q strip chart? 6 A Yes. 7 Q Just looking at this graph today, it 8 indicates, does it not, that at various times 9 between ten minutes and 30 minutes after the accident, 10 pressurizer level was as low at one point as about 11 350 and at other periods of time between 350 and 375? 12 A It is my judgment it shows here it was as ('S \\--) 13 low as 360 and that it varied between 370 and 375, e 14 yes. 15 Q were you aware of those pressurizer 16 levels during that period of time? 17 A I was aware of pressurizer level, not l 18 the exact inches, but I was aware that it was 19 abnormally high. And as I remember, numbe'rs 380 to 20 385 come to mind"during that period. 21 Q Now, you said I think that if the l 22 system were solid, you would expect to see sharp l l 23 swings in pressure up if there was an increase in i [\\ 24 flow or down if there was a decrease in flow; i %-] l 25 correct?
7 1 Zewe 557 /^\\ U~ 2 A I would expect to see that for a solid 3 system, yes. 4 Q And in order for there to be any kind of ( 5 a problem as a result of an increase in pressure, 6 you would have to raisc oressure to the point where 7 the code safety valses lifted, would you not? 8 THE WITNESS: Read that again, please? 9 I am not sure I understood it. 10 (Question read) t 11 MR. KLINGSBERG: When you say "you 12 wc J have to raise," you are not talking (' \\- 13 about the operators. You mean pressures 14 would rise 15 MR. FISKE: Yes, conditions would have 16 to be such. 17 A It was my understanding that,first the 18 PORV would lift; and if that could not handle the ~ 19 rapid excursion, that the code safeties would then 20 lift, and it would be around those points that I 21 would expect the problems to occur. 22 Q By "those points" you mean what? 23 A The relief valve setpoints themselves. 24 Q That was 2500 psig? 25 A As I remember, that was the code relief ~
1 1 Zewe 558 id 2 valve setpoint at that time. 3 Q Now looking at this chart that is in 4 front of you, Exhibit 751, this indicates, does it ( 5 not, that during the period of time from about ten 6 minutes into the trip to about 30 minutes, the 7 pressure gradually went down from about 1450 to 8 around 1000 and in fact was at around 1100 or less 9 for two,15 and 30 minutes into the trip? 10 A That is what the graph indicates. 11 Q And I guess you knew, of course, did you 12 not, Mr. Zewe, that it would be desirable to try to I~') (/ 13 get the reactor coolant system pressure back up 14 above the setpoint at which HPI had been 15 automatically actuated? 16 A I did know that it was desirable to 17 restore to pressure above the actuation s.e tpoin t. 18 Q During this period of time that we have 19 been talking about in the last few minutes-- this 20 maybe 20-minute period from ten minutes into the 21 accident to 30 minutes did you make efforts to 22 increase the flow into the system to try to get the 23 p re s sure up and see whether or not that increased [m 24 flow produced any significant change in pressurizer x/ 25 level?
- = _. 1 zow* 559 (1) 2 A We did not try to do that, because 3 pressurizer level was high. There was no need for 4 us to add more inventory to the system, which was { 5 indicating abnormally high to begin with. We 6 concentrated our effort on trying to hold or even s 7 reduce pressurizer level and also to try to regain l 8 pressure by heating up the pressurizer, using the 9 pressurizer heaters. 10 Q What steps did you take tottry to raise Il pressure by means of the pressurizer heaters? i 12 A By insuring that all the heaters were 'N 13 turned on from the control room, and also having 14 an auxiliary operator go out to the pressurizer 15 heater control cabinets down by the main steam lines e 16-and check to see if all the heater breakers were 17 closed and that they were all energized, and if i 18 there was any failures of the particular pressurizer 19 heater breakers -- because we had had problems 20 earlier in other events to where, due to their I 21 location in the same room as the main steam safety, 22 that we did have a reduced heater capability because 23 of the atmosphere that they were in, so I asked to () 24 have an operator check to verify that we had all 25 available heaters to us.
i l 560 10 1 Zewe 2 Q Did he come back with that information? 3 A As I recall, I asked for it to be done, 4 and I assumed that it was done. I do not recall ({ 5 feedback saying that he had found any problems or 6 that it.indeed was okay. 7 I remember assuming that it was done, 8 there were no problems or he would have alerted me 9 to it. 10 Q There did come a point somewhere in the 11 first hour or so where you came to the conclusion 12 that there wasn't anything wrong with the heaters? q _) 13 A I don't believe that during that first 14 hour period that I concluded that there was not 15 anything wrong. I still felt that there was a 16 reduced. heater capability, but not that there was 17 anything that I could correct at that point. 18 Q Didn't you understand that increasing 19 high-pressure injection flow would be a means by 20 which you could raise pressure? 21 A It was my understanding at this point 22 in the accident that increasing the high-pressure 23 injection would raise pressurizer level and would ( 24 cause the plant to go solid and I would experience 25 the inability to control my pressure response. 1-, -.. - ,-----..-.----m-,- -~ -
1 Zewe 561 m 2 I did know that it would take it solid 3 and that I would have a large pressure response. I 4 did not want to take it solid and have that ( 5 uncontrollable pressure response. 6 Q I guess my question was, Mr. Zewe, didn't 7 you understand that you had a capability within the 8 high-pressure injection system to increase the flow 9 in an amount which would raise pressure without 10 raising it all the way to the point where you would 11 lift a code safety? 12 A I believe my understanding at the time AU 13 was not that I could control that well enough to try 14 to do that. 15 I felt that I didn't have the experience 16 or training to take the plant solid under those 17 conditions and to limit the overpressure condition l 18 which I described earlier. l ( ~ 19 Q It is true, isn't it, that the way the l 20 high-pressure injection system works, you have 21 these various valves with which the operators can 22 control the flow of water into the system; right? 23 A Yes. 24 Q It doesn't have to be a thousand gallons 25 or nothing; right?
1 l l Zewe 562 2 A Absolutely. 3 Q All right. I guess my question is, 4 didn't you understand that you had the capability of { 5 increasing the flow by some amount, starting with 6 maybe a very little amount, and watching to see 7 whether that produced an increase in pressure without 8 causing the kind of pressure spike that you testified 9 about before? 10 A You are asking me ifIhadyhat 11 understanding before the accident 12 Q Or during, anytime up to and including e 13 during the accident. 14 A During the accident I did not consider 15 adding increased makeup flow to take the pressurizer 1 16 solid in the control fashion to limit the pressure 17 excursion. 18 I was dealing with what we had. We 19 already had a high pressurizer level. P re's s ure 20 indeed was low. But I did not conclude from that 21 that it was so low that it was detrimental at that 22 point. 23 I felt that things were relatively stable ( 24 at that particular time. I did not know why the 25 level was high. I felt that pressure would slowly
1 Zewe 563 L-) 2 start to recover. And I didn't feel that any other 3 action at the time was warranted. 4 Q You keep saying in your answers, Mr. { 5 Zewe, that you didn't want to take any action which 6 would cause the pressurizer to go solid. 7 I haven't really been putting that 8 condition in my questions. What I have been asking l 9 is whether you understood during the accident that 10 you had the capability in this high-pressure 11 injection system of increasing the flow of water 12 into the system in, le.t's say, a'smal1 amount, and f) s/ 13 then watching to see whether that amount produced an 14 increase in pressure without taking the pressurizer 15 solid. 16 MR. KLINGSBERG: I think he has answered l 17 that. I will allow it again. 18 A Again, I don't see any real difference 19 in how I answered it before. ~ l 20 ~ The only understanding I had was that if 21 I would increase the flow in, that I would go solid, 22 and that the controllability aspect of controlling 23 the flow to go solid, I didn't have any real training ( 24 on at that point, and I didn't know exactly how to 25 deal with it, and the conditions I had that day
I /. 1 Zewe 564 2 were such that I chose,not to do that. 3 Q For this period of time, which I will now 4 expand from about, let's say, ten minutes in the { 5 accident through the time that the first reactor 6 pumps were turned off, what actually was the flow 7 that was going into the system from the HPI pumps? 8 A I don't recall. 9 Q I am not asking you to give it down to 10 the last gallon, but was it 500, was it 250, was it 11 7507 Would you give us your best recollection? 12 A Best recollection from that day? 13 Q Yes. 14 A Approximately 140 gallons a minute. 15 Q Total? 16 A Total flow during that period of time. 17 Q And was there also flow going.in from 18 the makeup pump in addition to HPI or are you 19 including that in the 140? 20 A All the flow comes from the makeup pump, 21 whether it is HPI or normal or coolant seal k' 22 injection pumps. It all comes from the makeup pump. 23 Q And you are including all of that in the 24 140? 25 A Yes.
1 Zewe 565 .O t t '#~ 2 Q There is also a procedure, isn't there, 3 for letting water out of the system? 4 A There is a method to do that, yes. { 5 Q What is that method? 6 A We have a letdown system, which is 4 7 designed to remove water from the reactor coolant 8 system for various purposes, and then returning that 9 water. 10 Q I think you said earlier on that one of 11 the first things that was done after the reactor 12 trip was closing the letdown isolation valve. (') (_/ 13 A That is true. 14 Q 'Was that valve opened at any point in 15 the first hour and ten minutes of the accident? 16 A Yes, it was. As part of the method of 17 trying to control pressurizer level from going 18 off-scale high, the operator reopened the letdown 19 isolation valve and used this letdown path'in trying 20 to control the pressurizer level rise. 21 Q Can you tell us during this period of 22 time again, from ten minutes into the accident 23 until the time the first set of reactor coolant pumps ( 24 was turned off -- what amount of flow was going out 25 o f the system through the letdown line?
1 Zewe 566 i 2 A No, I don't know. As long as you have 3 the isolation valve open, as long as you don't open 4 up another manual throttle control valve, you will (( 5 let down at approximately 40 to 45 gallons a minute. 6 The other flow path which is par allel to 7 that is controlled by the operator on a manual basis. 8 Q Do you know whether or not that other 9 flow path was opened up during this period? 10 A At varying times it was used. 11 Q Do you know the amounts that went out of 12 the system through that path during those periods of A 13 time? 14 A I.do not. Because as I recall, the flow 15 rate when we started to take action to reduce 16 pressurizer level, the operator went,,as far as I 17 can remember, to his maximum letdown flow and then he 18 throttled back on that flow and then he closed the 19 valve and then he throttled it open again. And I i 20 really wasn't there watching the letdown flow. 21 So it really varied over this period. 22 Q All right. But you have said that about 23 140 gallons a minute was going in. () 24 A Counting normal makeup and soal 25 injection, yes. 4
l 1 Zewe 567 2 Q And various amounts were going out either 3 through the letdown line or this other path that you 4 have described; right? i g 5 A They are both letdown paths. They come 6 from the same place and go to the same place. It is 7 just that you have a pressure breakdown orifice which 8 is sized for a certain flow at a certain inlet 9 pressure and you have a bypass valve that is 10 controlled to flow around that. So it>is the same 11 path, but that is strictly under the control of the 12 operator. 13 Q Were there times during this period you 14 have just been talking about where more water was' 15 going out than was coming in? 16 A That I don't know, at any particular 17 time. Because the makeup flow, that 140,is, as I l 18 recall, an average number, but it was higher than 19 that at times and a little lower than that'. And i i 20 with the makeup flow varying over quite a range, 21 I don't really know if that is the case or not. l 22 Q Mr. Zewe, you understood, did you not, 23 that HPI is actuated by a drop in pressure below i 24 1640? l 25 A Yes. I
1 zewe 568 (~v\\ ~ 2 Q You understood that it was not actuated 1 3 by a change in pressurizer level. 4 A That is true. It was actuated by a ( 5 pressure setpoint only, automatically. 6 Q .In the course of any of your training 7 in the Navy, at B&W, at Met Ed or from any 8 understanding that you arrived at outside that 9 training, did you unders tand that when an emergency 10 safety system comes on automatically as a result of 11 a certain condition in the plant, that that 12 automatic safety system should not be defeated until 13 the condition which caused it to come on has been 14 cured? 15 MR. KLINGSBERG: What do you mean, 16 " defeated"? t 17 MR. FISKE: Interfered with.. 18 MR. KLINGSBERG: Well, that is just an 19 abstract question. t j 20 THE WITNESS: Would you repeat that, 21 please? (. r 22 (Question read) l 23 A Case 1, my Navy training in relationship () 24 to this area of engineering safety features, which I 25 assume that you mean here, is strictly a manual L
1 Zowe 569 ( / 2 initiation by the operator, based on a low 3 pressurizer level alarm and a low reactor coolant 4 system pressure alarm and a reactor trip, and it is ( 5 all three of those combined is the only time that 6 the operator does that, not on prescure, not solely 7 on level and not solely on the reactor trip, but all 8 three of them is when he initiates it. 9 My training at B&W in the transient 10 response which I had prior to operation in either 11 Unit 1 or Unit 2, whenever an engineering safety 12 feature system would actuate, I was trained to bypass U'# 13 the actuation of the system in order to take manual r 14 control to throttle if needed, based on prevent 5.ng 15 pump runout and based upon restoring RCS inventory 16 as indicated by pressurizer level. I do not recall 17 any training at B&W at any time prior to the March-18 28 accident where you would not bypass engineering 19 safety features and throttle high-pressure ~3.njection 20 other than based on the recovery of pressurizer 21 level that was either actuated by an overcooling 22 event or a loss of coolant event. l 23 And also at Met Ed, from my training and p) 24 experience at Unit 1 and Unit 2 further amplified '.v 25 the training that I received at B&W, that whenever I
i j p-9., } L t 7 i i i a 1 zewe 'y 570' r^g ~ T V 2 had automatic actuation,
t. hat I would bypass land that
,i 3 I would throttle, based on recovery of inve,ntory-3 4 based on pressurizar level indicating a rec'very of o 5 the inventory, not that I would not be allowed to y g 6 bypass or throttle based on pressure which is the J' T s 7 actuation point. 8 In all cases, it is inferred that 9 pressurizer level..and pressurizer pressure would 10 always trend in the same direction fort.an overcooling _ 1 11 event or loss of coolant accident where the HPI or 1 12 engineering safety featuies would be automatically 13 initiated.'-- ,.~ 14 Q' Youcwere trained at B&W, were you not., -t 15 -that the throttling of HPI should be governed by 8 \\ m 16 both pressure and pressurizer level? 17 'A As I ~ remembc'r, that is not c o,rre c t. My 18 r r t. in g, as I remember it, puts significance, p s 19 i U
tance, on the recovery of pressurizer level as s
/ \\ 1 20 the criteria forythrottling high-pressure injection. 'N 21 They inferred,in that, my understanding is, that s 22 level and pressdri would trend together and that the 23 most'significant point would be the recovery of 24 p ressurizer leve-l. ~ 25 g Did you over ask anyone a$B&W or at Met t 1 L
w 7s 2,* nec 'k } i g (f [ l' k Zewe 571 e. s: ><-) 1 1,- ' s 's - s ]2 Ed why, if pressurizer level was the paramount 'l' 3\\ \\ consideration, HPI was not initiated'in the first , Th4[ place by pressurizer level? w1f" A I don't remember asking that particular 5 ( '6' question. But I do remember discussing many times y;' is I ' f, $"'7 about the need for putting on the second makeup pump s
b right after an ordinary event like a reactor trip, S9 because it was generally felt that the pressurizer
? 10 level would be reduced below the cutoff point of 11 the heaters if this action was not taken, even on a I2-normal trip. ) 13 So it wouldn't seem to follow that you s_ 3 15 would want to actuate an HPI actuation or engineering 15 safety features based on the pressurizer level 16 every time. As long as you left guidance to the s 17 operator to have him initiate an increased makeup b 4 18 flow during a normal transient, that you could always 19 restore pressurizer level before it got a b'n'o rm ally 20 low and had the pressurizer heater cut off. 1 21 (Lunch recess taken at 12:54 p.m.) \\+ 22 i A 23 x., 25 s 3,. . _ _ - -, _,, ~, ~ _, ,,w. -..w..- -.y,_.-,--.,y
1 572 [\\ 2 Afternoon Session 3 2:02 o' clock p.m. 4 g 5 W I LL IAM H. Z EWE, resumed: 6 (Mr. Benedict not present) 7 MR. FISKE: Could we mark the next B&W 8 exhibit, B&W 7527 9 (Page of information that was printed 10 out by alarm printer marked B&W Exhibit 752 11 for identification, as of this date.) 12 BY MR. FISKE: A k) 13 Q I show you a document, Mr. Zewe, that 14 has been marked B&W Exhibit 752, which is a page 15 of information that was printed out by the alarm 16 printer at some point during the accident. 17 Do you have that in front of you? 18 A I do. 19 Q First of all, I would ask you if you saw 20 this page of the alarm printer at any time during 21 the first hour of the accident. 22 A I did not go over to read the individual 23 items on the pages. 24 Q Directing your attention to the fourth 25 item from the top, the number on the far left-hand
1 Zewe 573 [) 2 side of the page is the amount of the time that 3 elapsed after the reactor trip, is it not? 4 A No, I don't believe it is. 5 Q Is that the time of day? g 6 A I believe that is correct. 7 Q They just more or less happen to be the 8 same? 9 A Real close. 10 Q What does the next column i.ndicate? 11 A Status of the contact that monitors the 12 particular parameter. 13 Q What is the next column, 03467 14 A That is the individual point number of 15 the computer for that particular parameter. 16 Q The next column says "MU letdown flow 17 GPM." Do you see that? 18 A Yes, I do. ~ 19 Q That refers to the rate of letdown? 20 A That is the description of the point 21 0346. 22 Q And point 0346 is what, a letdown point? 23 A Right. The computer is set up with various 24 points, and they assign various parameters to certain 25 points, and it just happens that the computer people ,e--- -y,--- w.- ,-n m,_ ,-,n,,
1 Zewe 574 [)\\ 2 established computer point 346 as the letdown flow 3 parameter. 4 Q Four minutes into the accident, if you ( 5 would look at the chart reflecting pressurizer level, 6 pressurizer level was someplace between 350 and on 7 its way to 4007 right? 8 A Four-minute time interval? 9 Q Yes. 10 A I would say that it was around 350, yes. 11 Q And increasing? 12 A And increasing. O 13 Q And I think you said earlier that in the 14 early minutes of the accident, action was taken to 15 try to reduce the increase in pressurizer level by 16 opening.the isolation valves for the letdown lines 17 is that right? 18 A It was used for that purpose,'yes. 19 Q And in trying to counteract thii-sharp 20 increase in pressurizer level that was occurring at 21 that time, was the valve opened at a point which would 22 allow the maximum letdown flow to occur? ' 23 MR. KLINGSBERG: I am not sure that ~ 24 he said that the valve was opened at this point ~_ 25 in time.
1 Zewe 575 O 2 MR. FISKE: I thought he just did. 3 A I said the letdown was used to try and 4 combat the raising pressurizer level. ( 5 MR. KLINGSBERG: At this point in time 6 or just generally? 7 THE WITNESS: Somewhere during the 8 approximately four-minute interval time when 9 pressurizer level was at 350 inches and increasing, 10 we did use letdown flow to try to combat 11 the rise in level. 12 Q Actually, the time is indicated in the O'J 13 fourth line at the top as five minutes 35 seconds 14 after, fours is that right? 15 A Yes, it is, but you cannot conclude 16 much information from that point, because.i t 17 prints out thatit is a bad input, and the flow rate. i 18 in gallons per minute is a minus three question marks ~ 19 period question mark. The computer at that'particular 20 point did not know what the value was and it let 21 the operator know that it did not know by putting 22 that in a series of question marks. 23 Q I guess I am trying to find out from [G) 24 you what the rate of the letdown flow was when the 25 alarm printer doesn't reflect, and I guess I am just
1 Zewe 576 {J N 2 trying to point out isn't it a fact that at about 3 four to five minutes into the accident, when the 4 pressurizer level was rising, that it was during ({ 5 that period that you had the isolation valve for 6 the letdown line open to allow the maximum letdown 7 flow? 8 A That I am not sure of. I recall that 9 we were increasing the letdown flow, but I am not 10 sure if at that particular four-to five-minute 11 period if we went to maximum there or not. It may ' 12 very well have been. I am not sure if we did go to 13 max at that point. But I am sure that we had 14 reinstituted some letdown flow, but I am not sure of 15 its magnitude. 16 Q Did you tell us before lunch what the 17 maximum letdown flow is? ~ 18 A I did not. ~' 19 Q could you do it now? 20 A The maximum letdown flow that is 21 indicated is, as I remember, 160 gallons per minute. 22 Q The same page of this alarm printer 23 indicates about ten items up from the bottom, that () 24 at seven minutes 35 seconds after four, the letdown 25 flow was 71.4 gallons per min ~ute.
1 zewe 577 2 A I see that. 3 Q Looking at that, does that help you 4 remember that the letdown flow was reduced at or (( 5 about that period of time? 6 A I don't think that it improves my 7 recollection of what happened, though I believe that 8 it supports that letdown flow was a variable during 9 that period. f 10 Q What would cause the printer to print 11 out the rate of letdown flow at 4:07:357 12 A I don't recall at what parameters the ~I 13 computer was keyed to print a high, a low or a 14 normal value. I don't recall what those numbers are. 15 But typically, the computer tech would 16 put in a rate of flow that would print a high rate of 17 flow and it would alarm or a low rate of flow. Or. 18 if it was a bad input, it would be reflected. Why 19 it printed on this particular point as a normal 20 contact at point 346 when the makeup letdown flow 21 was 71.4, I don't know. That doesn't necessarily L 22 mean that that is the point either. Because as the 23 computer scans through it, the point may already ( 24 have triggered the point at which it should print 25 and it will read the current value, which may be less
1 Zewe 578 A 2 or higher than that value. 3 Q From your understanding of the way the 4 pressurizer was supposed to work, if there was a { 5 significant increase in letdown flow, would you 6 expect the level of the pressurizer to come down? 7 A Yes, I would. O Q And correspondingly, if there was a 9 significant increase in the flow of water into the 10 system, you would expect the level to g'o up; correct? 11 g y,,, 12 Q And isn't it a fact that on a number of (_ 13 occasions, during the first hour of the transient, 14 that you and the other operators took action to increase 15 the amount of letdown and did not see the kind of F decrease in pressurizer level that you would have l 17 expected under normal circumstances? l 18 A That is correct. 19 Q Isn't it also true that during that 20 same period of time, you took action on various 21 occasions to increase the flow of water into the 22 system and did not see the kind of increases in 23 pressurizer level that you would have expected to A o4 see under normal conditions? ~ 25 A I don't recall that being the case. I
1 Zewe 579 Cj')- 2 do not remember that the operator had increased makeup 3 flow to the RCS and had looked for a corresponding 4 change which he thought was normal for that change { 5 in inlet flow, or the reverse, that he had reduced 6 amount going in but was increasing letdown, looking 7 for the pressurizer to respond in a downward direction, 8 not necessarily the reverse as I remember. 9 Q Did the fact that significant increases 10 in letdown did not produce decreases in pressurizer 11 level of the type that you would ordinarily expect 12 raise any question in your mind during the accident 13 as to whether pressurizer level was, in fact, 14 accurately indicating the amount of inventory in 15 the system? 16 A Yes, it did raise that question in our 17 minds. 18 Q Now, you knew, didn't you, Mr. Zewe, ~ 19 that it was important to keep the pressure in the 20 reactor coolant system up above the point at which 21 the water in the system would start to boil? L 22 A As I recall, my training and understanding 23 was not to look at the pressure in terms of keeping () 24 up pressure so that you did no t have boiling. 25 It was more in terms of maintaining
1 Zewe 580 2 pressurizer level and insuring that you had the 3 pressure control in the pressurizer, and that that 4 would be the normal conditions. And I don't { 5 remember thinking the other way about pressure and 6 boiling. 7 I felt that the only time you could have 8 boiling in the RCS was if you had a loss of coolant e 9 inventory which was reflected by a low pressurizer 10 level on a loss of coolant accident where you 11 exceed the makeup capability with a leak and you, 12 in fact, then,had taken the p'ressurizer off-scale low. 13 That was the only time when I inferred 14 the pressure and the boiling, only in relation to 15 keeping normal pressurizer level and keeping the 16 pressure in the pressurizer. 17 Q Did you have an understanding before the 18 accident that if pressurizer level was full, that t l 19 you couldn't have boiling in the reactor c'olant o 2 system, no matter how far down the pressure dropped? 21 THE WITNESS: Read that back, please. 22 (Question read) l 23 A As I have stated, as long as I had full ( 24 system inventory, which was reflected by the 25 pressurizer level, and in this case that you have
1 zewe 581 d("N 2 stated, a full, pressurizer, I would not have expected 3 it was possible to have any boiling in the reactor 4 coolant system. ( 5 I would have also expected to have a 6 controllable pressurizer -- or reactor. coolant system 7 pressure -- which was afforded by the action of the 8 pressurizer be.ing at a normal desired level and 9 controlling pressure throughout the system. I never 10 separated the t,w o. I never thought it was possible to 11 have boiling with the full pressurizer under any 12 circumstances other than a LOCA. 13 Q Well, you knew, didn't you, Mr. Zewe, e 14 that there was a certain pressure / temperature L 15 relationship at which the water in the reactor J 16 coolant system would start to boil? 17 A From the theoretical sense, I,was aware 18 that there is a pressure / temperature reladionship 19 that water will boil, yes. But again, in relationship 20 to the reactor coolant system itself boiling, only 21 those conditions. 22 Q Did you understand that if the 23 pressure / temperature relationship reached that point () 24 at which the water would start to boil, that boiling 25 would not occur if the pressurizer was full?
1 Zewe 582 O ~ \\' 2 A All of my experience, understanding and 3 training, including the transient response at 4 Lynchburg, Virginia, supported that as long as I 5 could maintain pressurizer level which was a g 6 reflection of primary system inventory, that would 7 never be a problem. And it was never pointed out 8 to me that that was even a possibility under certain 9 conditions that that would exist. 10 Q Well, let's put it the othqr way, Mr. 11 Zewe. You knew as a matter of reactor theory -- 12 heat transfer, whatever, the basis fundamentals of 13 the training on a pressurized water reictor -- that 14 there is a certain pressure / temperature relationship 15 at which the water would start to boil, correct? 16 A In relation to the pressurizer and the 17 steam generator, certainly. Not in relationship 18 to any other portion of the RCS. ~ 19 Q You understood that when pressure in 20 the RCS reached a certain point in relation to 21 temperature, that the water in the RCS would start 22 to boil? 23 MR. KLINGSBERG: That is the exact same 24 question that you have now asked two or three 25 times. I think the witness has answered.
1 zewe 583 (n C 2 MR. FIS KE : He put a couple of 3 qualifications in. 4 MR. KLINGSBERG: Well, he is allowed to 5 do tha't. {_ 6 MR. FISKE: I want just to be sure that 7 'the answer isn't broader than the qualifications 8 he put on it. 9 A As I can remember, my understanding was 10 that I would not have boiling in the RC.S without 11 losing presurizer level. And I don' t remember 12 thinking about or even looking for saturation Ox 13 conditions in the RCS, because I never thought 14 that it was possible other than a LOCA condition. 15 So it would not lead me to draw that conclusion that 16 under this condition I would have it, so I must do 17 this to confirm it or to look for it. 18 Q Well, let me just put it this ~ Mr.
way, 19 zewe.
Recognizing that there is a particular pressure / 20 temperature relationship at which the water will ~ 21 boil, why did you understand that once the pressure 22 and temperature hav9 reached that relationship, that 23 it made any difference what the level in the pressurizer 21 was in terms of whether boiling would occur? 25 MR. KLINGSBERG: That isn' t what he said. 4 m.__._.,,__--__._._...,___-y_ ._.--,,_,,_,-,,.,_,_,,m_ _, - ~. - - ,,.,--_____,.-._,f .-.m-..... _w.-
1 Zewe 584 ~h (Y 2 A It is more the reverse of that. Something 3 must lead you to a conclusion or at least into an 4 evaluation stage, to look for this relationship (( 5 between teyperature and pressure in the RCS. 6 Something must lead you to say, " Hey, look at this." 7 All right? 8 And the basis of that has to deal with 9 pressurizer level. If you have the pressurizer 10 level, there is no need to go and even ake that 11 evaluation. 12 And hone of my training had ever said, J 13 look at this, key on this, look and see if the f 14 conditions exist. It never did that. 15 It said, maintain level. If you could 16 maintain level, you could maintain pressure, you 17 had an automatic system that was designed,to put 18 water into the core, and something had to' key you 19 to look for it. And I just didn ' t have th'a t key. I 20 didn't have that understanding. 21 Q It was your understanding, I gather, that 22 if you maintained pressurizer level, then that would 1 23 maintain pressure in the system in.a way that would 4 ( 24 prevent boiling ? 25 A Absolutely. As long as I could keep
1 Zewe 585 2 pressurizer level, I could then keep the pressurizer 3 the hottest point in the system and have a saturated 4 condition in the pressurizer which would maintain 5 the desired pressure within the reactor coolant { 6 system, and I had a means on the pressurizer to 7 vary that pressure and temperature. 8 Q All right. Now, you knew, didn't you, 9 that on the day of the accident certainly within 10 the first hour -- that even though youthad level 11 in the pressurizer, you were not keeping pressure 12 up where it was supposed to be? O,- 13 A Yes, that is true. From o$her indications s 14 and things that we had done, I had concluded in my 15 mind that there were reasons why the pressure was tkmes.was 16 low but was stable, and it I thought at i 17 showing signs of recovery. P 18 (continued on next page) ~ l 19 20 21 (. 22 l l 23 24 G 25 j i l
.1 1 Zewe 586 h v v 2 Q When you realized that the pressurizer was n t perf rming its function of keeping pressure up 3 4 where it was supposed to be, did you then consider 5 whether pressure had deteriorated to the point where ( 6 you might have boiling in the system? 7 A I felt during the accident that the 8 pressurizar was still insuring that the RCS was full 9 of water, because I still had an indicated level, 10 though it was higher than normals it as,sured to me 11 that I did have proper inventory, if not too much 12 inventory in RCS. And I still felt that the pressure D 13 I had in the reactor coolant system was' the result 14 of the pressurizer. 15 And again, I would never associate having 16 boiling conditions in the RCS without a loss of 17 pressurizer level. It never entered my mind that I 18 could have an inventory problem and indicate too much 19 inventory. 20 Q I think we may be talking about two different things here, Mr. Zewe. 21 22 You indicated earlier that you looked upon 23 pressurizer level as an indication of the capability of the Pressurizer to maintain system pressure where 24 25 it was supposed to be; correct?
587 1 Zewe OO 2 A Yes, it is used to maintain pressure at 3 desired setpoint yes s 4 Q A pressure of approximately 1100 pounds ( 5 per square, inch is not the.ind of pressure that you 6 are supposed to haver isn't that correct? 7 A No, it's not, at that particular point. 8 But I thought that the pressure that we did have was 9 a result of the action of the pressurizer, 10 Q Okay. But a pressure of ap' proximately 11 1100 psig is about 1000 psig below what normal 12 operating pressure is, isn't it? 13 A Yes, it is. 14 Under that condition, it was not normal 15 operating condition. We were operating during a 16 severe transient. The reactor was indeed shut down. 17 And we were in mode 3 18 Q sure. But you knew that to the extent of 19 approximately 1000 psig, the pressurizer was not. 20 performing its function of keeping pressure up, l i 21 A As I stated earlier, I reached the 22 conclusion, through my observations and knowledge, 23 that there were reasons why the pressure was low and () 24 was slow to recover. 25 Q Well, whatever the reasons were that I l l
588 IK3 1 g,y, (D u> 2 produced the drop in pressure, you knew that the 3 pressurizer was not bringing that pressure back up 4 to where it ought to be; right? ( 5, A It was not at that point in time that I 6 expected it to recover, 7 Q You said that for the pressurizer to 8 perform its proper function, the pressurizer should 9 he the hottest part of the system? 10 A It should contain the hottest reactor 11 coolant, yes. 12 Q And there is instrumentation on the rh ( ) \\# 13 control panel that measures pressurizer temperature? 14 A Yes, there is. l 15 Q And under normal operating conditions, l 1 16 Pressurizer temperature is hotter than the reactor l 17 coolant system temperature? l l 18 A It is the hottest portion of the reactor 1 19 coolant system itself. 20 Q Did you look at the instrumentation during 21 the course of this first hour of the transient to l 22 determine whether the pressuriner temperature 23 continued to be higher than the temperature in the ) 24 reactor coolant system? I J 25 A I do not remember comparing pressurizer
4 1 Zewe 589 ~ temperature with the rest of the RCS temperature 2 indicators. 3 4 Q What did you understand would happen if ( 5 the condition existed where the temperature in the 6 pressurizer was not hotter than temperature in the 7 reactor coolant system? 8 THE WITNESS: Would you read that back, 9 please? 10 (Question read.) t 11 A I don't recall giving it much thought 12 one way or the other, because it always assumed that i \\' the pressurizer would always be hotter except if you 13 14 had a :.LOCA. condition to where..you could not 15 adequately cool the core, at which the reactor coolant 16 system may be hotter. But I could not think of a case l l 17 where the pressurizer temperature would be less than l l ,18 the pressurizer -- than the rest of the RCS -- nor 19 do I recall any training that would point 'to that 20-fact, to check pressurizer temperature versus the 21 RCS temperature during any transient. L' 22 The only time that I can recall being 23 trained on pressurizer temperature was during a heatup. [~ ) 24 At the very stages of heating up, it specifies a %J 25 certain temperature and pressure for the pressurizer.
Zewe 590 . f~ ~ J, 2 Other than that, I don't remember 3 thinking about that possibility. 4 Q In the course of trying to develop an ( 5 understanding of how this reactor worked, when you 6 learned that under normal conditions the temperature 7 in the pressurizer was kept 40 degrees or so higher 0 than the temperature in the hot leg, did you have an 9 understanding as to why that was important? 10 THE WITNESS: Read that again, please. 11 (Question read.) 12 A My understanding was, to insure that the 1 (/ 13 pressurizer was able to maintain the desired reactor 14 coolant system pressure of 2155, 15 Q You had heaters, did you not, t; hat were 16 supposed ~to heat the water in the pressurizer up to 17 that specified temperature? 18 A The pressurizer does have electric heaters 19 that heat up the pressurizer water that is controlled 20 by adjustable setpoint controlled by the operator or 21 automatically a setpoint. 22 Q Did you have any understanding at the 23 time as to what the consequences would be to the 24 reactor coolant system if those heaters for one reason 25 or another were unable to keep the temperature in the pressurizer higher than the temperature in the hot leg?
6 1 Zewe 591 p'%U 2 A As I recall, the only time where tha.t would be a factor is on a blackout or a loss of power to the 3 4 heaters. And if you would cool down the reactor coolant system at a slower rate than that which would ( 5 6 cool off the pressurizer to ambient conditions, you could have that occur. 7 8 Q Well, you could also have it occur if 9 for some reason the heaters failed; right? 10 A That's what I am saying. A loss of power, 11 or a heater failure. It just renders the heaters ~ 12 inoperable. O 13 Q What did you understand would be the ~ 14 consequences to the reactor coolant system if, as a 15 result of that kind of a failure, the heaters couldn't 16 keep the temperature in the pressurizer hotter than 17 the temperature in the hot leg? 18 A As I recall, I never gave it any thought 19 in those terms, because we had five different banks of thirteen groups of heaters. And unless we had lost 20 21 p wer to all the heaters, it was my understanding 22 that only a group or two out of the thirteen groups 23 would be enough to overcome the ambient losses of 24 the pressurizer and maintain it hotter than the RCS, 25 and I hadn't really thought of multiple pressurizer heater failures where I would lose more than a few
Z SWE 1 592 0 2 groups or a couple of banks of heaters. Not total 3 loss which would be the case on a blackout condition 4 where you would have a reactor trip and then a loss of ( 5 heaters a l,s o, but you would start out from the condition 6 where your pressurizer would still be hot and your 7 reactor had just had a trip and had cooled down some 8 and now you have a wider margin because your 9 temperature then is roughly 550, so you have 100 10 degrees now and you would have more time to slowly 11 cool down the plant to keep i't cooler than your 12 ambient losses cooldbwn that you would have on your O 13 pressurizer. 14 Q Is it your testimony that up to the time 15 of the accident you had no understanding as to what 16 consequences might occur to the reactor coolant system 17 if the pressurizer temperature was not hotter than 18 the hot leg? 19 MR. KLINGSBERG: I don't think that was his 20 testimony. 21 MR. FISKE: That is what I am trying to 22 find out. Mr. Zewe has just given us a couple 23 of lengthy answers as to why he didn't think () 24 the heaters could fail and as to what might 25 happen in the course of a blackout.
l 1 ZeWe 593 1 As I am interested in those answers, but I am also interested in finding out, apart from what he thought one way or the'other as to whether 4 the heaters night fail,did he hhve an understanding at the time of the accident as to what if any adverse consequences might flow from the fact that the pressuriner temperature was not hotter than the hot leg. 9 A As I have stated, in a theoretical sense I certainly knew that if you had reached the saturation temperature for a particular pressure, that you would O have boiling, and that could occur anywhere that that relationship occurred. In applying that logic to the reactor coolant system, I don't remember saying that I am going to have boiling in the RCS. I always pilt it in terms of what we had and how we used it. ot in the strict g relationship of saying that if I lo s 't this, I would have boiling there. I just didn't think of it in those terms, 21 but 22 certainly did know water would boil if you reached saturation conditions. O O Mr. Zewe 24 ~ A I am just trying to put it in the light of how I thought about it in respect of operating the v-n
i 1 Zewe 594 O-P ant, and the things that we had in the plant and how l 2 3 we used them and what procedures supported their use 4 and what my logic was behind what I thought about 5 using them and what they meant, which I believe is two (( 6 different areas. 7 Q Well, Mr. Zewe, you have said, I believe 8 several times, that during the course of this transient, 9 you 1 oked at pressurizer level as being an indication 10 that the system was full of water. Right? 11 A That is correcu. 12 Q Now, didn't you know that the pressurizer 13 level coqAd not be relied upon for that kind of 14 indication unless the temperature in the pressurizer 15 was higher than the temperature in the reactor 16 coolant system? 17 A I did not think about it in those terms, 18 no. I thought that if the pressurizer was full, it 19 reflected inventory and under all conditio~ns that I 20 could think of that would reflect reactor coolant 21 system inventory, and as long as I had heaters' 22 capability, which I knew I had, I could then insure 23 that I had pressurizer pressure control afforded in ( 24 the pressurizer. 25 Q Is it your testimony, Mr. Zewe, that on
1 Zewe 595 ('O) 2 the day of the accident, you thought that pressurizer 3 level accurately reflected the system inventory, even 4 if pressurizar temperature was not higher than reactor g{ 5 coolant system temperature? 6 MR, KLINGSBERG: I don't think that was 7 his testimony, 8 MR, FISKE: That is what I am trying to 9 find out. 10 MR. KLINGSBERG: I really think you try to 11 make these things oversimplified. The witness 12 has given you answers both on a theoretical level /~ k -)' 13 and in terms of what his understanding was at 14 the time, and in terms of how the unit operated, 15 and then you go back and say "Is it your 16 testimony?", and you summarize it in an 17 oversimplified fashion. 18 I have no objection to your asking him any 19 questions you like and within limits ~ repeating 20 the questions as you are doing, but I don't think 21 it is fair to say "Is that your testimony?", 22 and -- 23 MR. FISKE: I will be happy to rephrase f~} 24 the question. I will say if the questions seem N_/ 25 to be repetitive it may only be because they
1 Z awe. y 596 2 have not been fully answered. 3 BY MR. FISKE: 4 Q But I will rephrase the question, Mr. Zewe, 5 in a form that I hope satisfies everybody. ( 6 on the morning of the accident, was it 7 your belief that you could rely on pressurizer level 8 as an accurate measure of the system inventory, even 9 though or even if the temperature in the pressurizer 10 was not hotter than temperature in the eactor coolant 11 . system? 12 A The morning of the accident I recall that (. 13 I believed the pressurizer level was an accurate 14 indication of system inventory. I did not compare 15 the pressurizer temperature with the RCS temperature l 16 to determine which was hotter. And as 2ar as I can 17 remember, I thought that under any condition that I 18 had at that point the pressurizer level wo'uld be a 19 true indication of the inventory. ~ 20 From a theoretical s candpoint, I know that 21 if you reach saturation conditions, I could have 22 boiling in the reactor coolant system or whatever. 23 I did not have an understanding that this O' 24 would cause pressurizer level to indicate erroneously. G 25 Q Let me show you a document which has been
.2 1 Z"M" 597 ) v 2 Previously marked as Exhibit 662, which is a transcript 3 f a hearing before a Congressional committee of which 4 Congressman Udall was the chairman. It was a hearing 5 that took place in May 1979. ({ 6 You did testify, did you not, before the 7 Udall committee? 8 A I did. 9 Q Let me direct your attention to page 173 10 and 174 of the transcript that you havg in front of 11 you. You can read as much as you want to yourself. 12 I am going to ask you about questions and answers 13 that appear on the bottom of page 173 and the very 14 top of 174. 15 Have you' read as much as you would like to e 16 read of that testimony? l' 17 A Not yet. 18 Q Okay. 19 A All right, I have read it. 20 Q Let me read to you some questions and 21 answers from the bottom of 173 and 174. I would like 22 to start with a question by Dr. Myers, which follows 23 a discussion of a lecter which had been sent out 24 relative to the SMUD light bulb incident. 25 The question starts: "Dr. Myers: Because ~, _... _ _.
1 Zewe 598 4 \\ V 2 t1is lettar also suggests something that might be an unambiguous instruction, where it says, 3 4 quoting, 'The pressurized level in reactor 5 coolant systemfpr' essure share of the reactor C 6 coolant system is filled,' that is saying that 7 the pressurizer water level is not full. g "Mr. Zewei That is true, our key 9 parameters are that as long as we have water 10 pressure in the pressurizer, that our main t 11 cooling system is full and the water level in 12 the steam generator is our main way of indicating 13 removal. 14 "Dr. Myers: But that is as long as the 15 temperature in the rest of the system is less. 16 "Mr. Zewe Right. At the hihhest i 17 temperature the saturation conditions occur in l 18 the pressurizer, that is true. 19 "Dr. Myers: Then you are aware that that 20 temperature condition was necessary in order for 21 thespressurizer water level to indicate a full 22 Press.ure vessel. Were you aware that the 23 temperature in' the rest of the system had to 24 less than the saturation temperature? 25 "Mr. Zewer Yes. I realized this; yes. u t --.n<,
N_ ( 14 1 Zawa 599 (~N 2 But at the time, I felt that we still had the 3 higher temperature system in the pressurizer 4 and not in the coolant system itself " 5 Now, were you asked those questions and did g 6 you give those answers to the Udall committee back 7 in May of 19797 ( g MR. KLINGSBERG: I have an objection to s g the question, in the introduction about the letter. 10 I don't know what fact was established but it 11 does say further up the page that the letter 12 was sent to Davis-Besse, and there seems to be' 13 s'o m e confusion as to whether it was sent to ~ 14 Metropolitan Edison or not, so I assume that is 15 not part of -- 16 MR. FISKE: No. That is another issue, 17 and it has been dealt with at length in other 18 depositions so that we don't have to' bother 19 Mr. Zewe about it. 4 20 The only reason I alluded to it is 21 because Dr. Myer's question which started the 22 questioning referring to pressurizer temperature 23 referred to that. So I wanted to make it clear 24 what we were talking about. 25 (Continued on the next page.) ~ .,--,-,.---.--m---,r,,-.---,,--rw y ,,-,,-~<y.-.vy,.,--
15 1 Zewe 600 2 BY MR. FISKE: 3 Q The pending question is, Mr. Zewe, were 4 you asked those questions and did you give those { 5 answers? 6 A As I recall, those were the~ questions and 7 answers, yes. 8 Q When you told the Udall committee that at 9 the time you felt that you still had the higher 10 temperature in the pressurizer and not in the coolant 11 system itself, what was that feeling based on? 12 A I don't remember exactly what I was ( 13 referring to right there. s 14 I-believe, looking at it now, trying to 15 put myself back in March 28th again, that it was by 16 more a virtue that I had heaters available and I had i 17 a pressurizer level, that it should have been hetter, 18 more so than looking at one and comparing it against 19 the other. I just felt that it had to be hotter. 20 Q Recognizing that unless it was hotter, 21 you could not rely on the level as an indication of l (_ 22 system inventory, did you take any steps to check the 23 actual temperatures in the pressurizer on the control } 24 panel against the temperatures in the reactor coolant 25 system?
16 1 Zawa 601 2 MR. KLINGSBERG: No objection to the 3 question, but I think you are linking it with the 4 " recognizing"part, which is objectionable. 5 MR. FISKE: You are making an objection ( 6 to the form? 7 MR. KLINGSBERG: Yes. g Q Yo'u can answer, Mr. Zewe. 9 THE WITNESS: Would you read that back, 10 please? g 11 Q To obviate an objection, let me ask this. 12 Did you take any steps during this period 13 of approximately the first hour of the ~ transient at 14 any time to check the instrumentation on the control 15 Panel to determine whether or not the temperature 16 in the pressurizer was in fact higher than the 17 temperature in the reactor coolant system? I 18 A I do not remember going over and comparing 19 pressurizer temperature with the temperature in the 20 RCS. l I had assigned my shift foreman and 21 ~ 22 control room operator to the duty of pressurizer 23 control of letdown and makeup. And I was relying on / 24 them to take action. But I did not go over and V) 25 verify the two, no.
-= _ - -. 17 1 Zeve 602 0 2 Q Did you ever ask Mr. Scheimann or 3 Mr. Frederick, at any time during the first hour of the 4 transient, whether pressurizer temperatures were { 5 higher than temperaturss in the rest of the system? 6 A I recall having several conversations 7 in relationship to pressurizer level and RCS pressure 8 and letdown, makeup flow. I don't recall any 9 particular -- I don't recall at any particular point 10 that I asked for that particular parame'ter, no. 11 MR. FISKE: Do you want to take a short 12 break? 13 (Recess.) 14 (Continued on the next page.) 15 16 17 1 18 19 ) 20 l 21 22 23 24 25 i
1 1 Zewa (~ 2 BY MR. FISKE: 3 Q Mr. Zewe, was the configuration of the 4 reactor coolant drain tank at Unit 2 different from 5 the one at Unit 17 ( 6 A Yes, it was. 7 Q In what respect? 8 A Unit 1 RC drain tank is a vertical 9 tank. Unit 2 is horizontal. 10 Unit I has a different des 4gn 11 capability as far as relief valve pressure goes and 12 the rupture disk. 13 Q Which one sorry. I 14 A Go ahead. 15 Q Which one has a greater capability? 16 A Unit 2. 17 Unit 2 has a greater heat removal 18 capability. ~ 19 We have already discussed their 20 difference in the indication available in the control 21 room. 22 Those are the major differences that I 23 could recall at this particular time. ("' 24 Q Was the drain tank located within the \\ j 25 reactor building? . ~..
1 604 gewe 2 A Yes. 3 Q Let me show you a document that we 4 marked previously as B&W Exhibit 663. Do you 5 recognize that as a diagram of the auxiliary { 6 building, reactor building, turbine building, Unit 27 7 A In a very basic sense, yes. 8 Q The drain tank is reflected on this 9 diagram, isn't it? 10 A Yes, it is. g 11 Q And there is a pipe that goes from the 12 pilot-operated relief valve directly to the drain 13 tank, isn't there? ~ 14 A That is correct. 15 Q That's what we have referred to a couple 16 of times as the discharge line? 17 A That is correct. t 18 Q It is on that pipe,that instrumentation 19 is contained relating to temperatures? ~ ~ 20 A That is correct. l 21 Q Going back to your understanding as of j 22 the day of the accident, Mr. Zewe, what pipes led 23 to the drain tank other than the pipe from the top O("h l 24 of the pressuriser, if any? 25 A There were pipes from reactor coolant i l _ - r ---~r
605 1 Zewe O 2 pump seals. 3 There were pipes from relief valve 4 headers from other relief valves inside the reactor ({ 5 building. 6 There was also lines that went to and 7 from the cooling water system to the tank. 8 It also connected to the vent header in 9 the auxiliary building and also to the miscellaneous 10 waste holdup tank in the aux. building'to the 11 reactor coolant bleed tanks. 12 Q Let me see if I can go over those 0 V 13 just to make sure I understand them. 14 The first one you listed was reactor 15 coolant pump seals. 16 A Yes. t 17 Q Do I understand correctly that in the 18 event there was leakage of water through the reactor 19 coolant pump seals, that would go into the drain tank? 20 A There normally is a design leakage rate 21 from the reactor coolant pump seals that do go to the 22 reactor coolant drain tank. 23 Q The next one you referred to was relief 24 valve headers inside the containment bu,ilding. 25 A Yes. There are various relief valves
606 1 Zewe 2 inside containment on the makeup system, the decay 3 heat system or core flood system, that communicate 4 water from that system to a line that goes into the ({ 5 RC drain tank. 6 Q so that in the event that any of those 7 relief valves lifted, then the water would flow into 8 the drain tank? 9 A That is correct. 10 Q Then the next one you menti'oned was the 11 cooling water system. 12 A Yes. O 13 Q How did that connect with the drain tank? 14 A The drain tank was provided with a 15 cooling water system of two pumps and a heat exchanger 16 which would circulate the reactor coolant drain tank 17 to maintain water temperature inside the RC drain 18 tank. 19 Q Do I understand that water just sort of 20-went around and around inside the drain tank? r 21 A From the drain tank to the heat 22 exchanger, to the pumps and back into the tank, and 23 then another cooling water system would cool of f f) 24 the heat exchanger to keep it cool. 25 Q All right. ___ ---~._ _____
607 1 Zewe 'u 2 A Also there were lines that would allow 3 you to pump the RC drain tank to the auxiliary building 4 which I mentioned. ( 5 g which would allow water to be pumped from 6 the drain tank into the auxiliary building? 7 A Right, to a place like the miscellaneous 8 waste holdup tank which I mentioned before, and the l 9 reactor coolant bleed tank. 10 Q So both the miscellaneous dater tank and 11 the reactor coolant bleed tank were tanks that would 12 receive water from reactor coolant drain tank? 13 A Yes, it would. 14 Q I think the other thing you mentioned 15 was some kind of a vent header in the auxiliary 16 building. How did that work? 17 A Well, the reactor coolant drain tank was 18 connected to the vent header, which is located in 19 the auxiliary building. 20 Q What was the purpose of that connection? 21 A To maintain a constant vent h e a d e r-22 pressure that was kept under control by the waste 23 gas compressors in the auxiliary building. () 24 Q Or to put it in the simplest terms, 25 did the connection between the drain tank and the
1 Zewe 608 [) \\xJ 2 vent header result in water going into the drain 3 tank or out of the drain tank? 4 A Repeat that again. (( 5 Q Did the connection between the vent 6 header and drain tank result in water going out of 7 the drain tank or into the drain tank? 8 A Water out of the drain tank into the 9 vent header and into the auxiliary building. 10 Q All right. I 11 Were there any other sources of water in l 12 d'ain tank which you were or wat,er out of the r h'-)' 13 aware of on March 28, 19797 14 A There were vent lines that would 15 communicate from the reactor coolant hot legs and 16 also from the control rod drive mechanism during 17 venting operations that vent to the RC drain tank. 18 Q That was a control procedure? 19 A Both of those were control procedures 20 only when the RCS was at cold shutdown. I 21 Q Is that the complete list? L 22 A If you would provide me with a print, I l 23 could check to see if there were any others. The () 24 actual flow diagram. There may be some other ones 25 that I have overlooked. This is as complete as I l
. _ -. = = _ _ _ 1 zowe 609 2 can re, member. 3 Q The best print I can give you is the one J 4 you have in front of you at the moment. ( 5 A Without consulting the actual print, and 6 not to confuse line's between Unit 1 and Unit 2, that 7 is as complete as I can remember. 8 Q All right. 9 I think we established earlier, Mr. zewe, 10 that at some point within the first hodr of the 11 accident you left the control room and went down to ^ 12 the turbine room. ~ 13 A Yes, I did. e 14 Q Before you left to do that, did you look 15 at the readings for the drain tank temperature, 4 16 pressure, and level? 17 A I believe I did look at it before I left 18 the control room, though I cannot pinpoint the exact 19 time. 20 Q Do you have a diagram in front of you, 21 B&W Exhibit 745? 22 A Yes, I do. 23 Q Is it correct that this diagram reflects ) 24 the location of the instrumentation for drain tank 25 pressure, temperature and level?
1 Zewe 610 <g 2 A Yes, it does. 3 Q Is that at the place on the diagram where 4 it says RCOT PTI? ({ 5 A Yes, it is. 6 Q That was what was known as panel 8-A? 7 A As I remember, that is the designator 8 for it, yes. 9 Q And it is correct, isn't it, that the 10 instrumentat..n was on the back of tha t, panel? 11 A on the front of the panel that faced the 12 wall. /~h V 13 Q Right, but in relationship to where the 14-operators were, in front of the control, in front 15 of the instrument panel in the con trol room, in order 16 to look at drain tank temperature, pressure and level, l 17 they had to go around behind that panel, did they 18 not? 19 A They had to go around, between the wall 20 that is right above figure 1, and then look at the 21 RC drain tank in the direction of the arrow. L 22 Q And you yourself, in fact, did that at 23 some time before you went down to the turbine room, , p) l (_ 24 right? l 25 A As I recall, I did.
1 Zewe ~g \\ 2 Q And is it correct that when you looked 3 at that instrumentation, you saw that the temperature 4 was high and that both pressure and level were low? { 5 MR. KLINGSBERG: I don't know what you 6 mean by high and low. 7 MR. FISKE: I would guess that maybe 8 Mr. Zewe does. 9 A As I recall, when I looked at it, 10 pressure was zero, level was off-scale Sow, and it 11 had an elevated temperature of around 200 degrees. 12 Q Which was above the alarm point? o 13 A Yes. ~ 14 Q Am I correct, Mr. Zewe, that you 15 concluded at that time that the drain tank had e 16 ruptured? 17 A As I recall, at that time I concluded Eibherthe i 18 that the drain tank had a hole in it. 19 rupture disk had ruptured, the relief valve had 20 failed open, or that the level tap or another part 21 of the tank or connecting to the tank'had broken k. 22 and had allowed the contents of the tank to go into 23 the reactor building. But I didn't know for sure s ) 24 exactly what that mechanism was, other than I believed t 25 that it had a hole in it.
1 Zewe .612 2 Q Let me show you some transcript of an 3 interview you had on March 30, 1979 with Mr. Long 4 and Mr. Reppert, which we will mark as the next (( 5 B&W exhibit. 6 (Transcript of an interview on March 30, 7 1979 with Mr. Long and Mr. Reppert marked 8 B&W Exhibit 753 for identification, as of 9 this date.) 10 Q Do you have that in front df you, Mr. 11 Zewe? 12 A Yes, I do. O 13 Q Do you recall being interviewed by Mr. 14 Long and Mr. Reppert a day or two'after the 15 accident? 16 A Yes, I do. 17 Q I direct your attention to page 3. A 18 statement about two-thirds of the way down, which l 19
starts, "I'd say maybe 20 minutes, maybe a half i
20 hour at the most." 21 Are you following along? L 22 A Yes. 23 Q If you would like to read that whole -() 24 statement, go ahead, and then I have got a couple 25 of questions. .r-,w-w-em-gvy -y9 y, my, ,..,.w---- --m-.4y mm + y-y;y-4-.-+m-,v9ag,-ww +-- --me.m --wwe9 r e-- w--s ---a -m., ,-.r, ---,w g -mw,w
Zewe 613 3 Have you read that, Mr. Zewe? A Yes, I have. 3 Q The third sentence, third and fourth 4 sentences of that particular statement
say, "I
noticed be' fore I went down that the drain tank had 6 a high temperature and zero pressure. The running pump had a very low discharge pressure, means that we had ruptured the RC drain tank." 9 Did you make those statements to Mr. .Long or Mr. Reppert back in March 19797 A As I remember, I did. 12 s,/ Q What was the running pump that you were 13 referring to in that answer? 14 A Cooling water pump for the RC drain tank itself. ~ 16 l Q And why did the fact that that pump had l 17 a low discharge pressure indicate that you had 18 ruptured the drain tanki A Well, the level was off-scale low, so I 20 didn't know how low the actual level was. But if 21 ( the pump had a very low discharge pressure, it 22 l must have had potentially zero level in the tank, 23 l l because the pur.p would have had a high discharge pressure with water in the tank to provide a suction 95 I
i 1 zowe !614 0 2 for the pump. 3 Q When the reactor coolant drain tank 1 4 temperature goes into an alarm state, what ( 5 indications are there, visual or audible, to 6 reflect that? 7 A There is an alarm window that would ~ 8 flash, an alarm that would sound on the 8-A panel. 9 The computer should also alarm, 10 indicating that point was reached. And,the 11 temperature indicator on the panel 8-A should 12 reflect the high temperature. V 13 Q Where is this alarm window that you 14 refer to? 15 A on panel 8-A itself, right above the 16 drain tank indicators themselves. 17 Q If you were standing in the center of 18 the control room and the temperature in the drain 19 tank reached the alarm stage, would you hear the 20 alarm sound? 21 A Yes, you would. The audible alarm is L 22 heard from the operator's normal position of the 23 control room. l 24 Q And I take it that you had not heard any 25 such high temperature alarm prior to four a.m. on
2 1 1 Zewe 615 2' March 28, 19797 3 A You have a horn that sounds for any 4 number of alarms. And the horn sounding would { 5 not be indicative of that high temperature but it 6 would be a whole realm of other things, that you 7 would have to investigate into what caused the 8 alarm to sound and check which of the overhead 9 annunciators had a flashing indicator. 10 Q why don't we make it simple". In between 1 11 eleven p.m. when you came on shift and four a.m. 12 when the accident started, did any alarms go off? 13 A Yes. 14 Q How many? 15 A No idea. .l 16 Q Are we talking about ten or are we i 17 talking about a hundred, or what range are we in? l 18 A Typically on shift we used to get various l 19 alarms from time to time. The overhead alarm. The 20 computer alarms. l 21 So it really varied from night to night. k. l 22 I wouldn't think that it would be in the neighborhood 23 of a hundred alarms. I would think from eleven r I ("N() 24 o' clock until four o' clock we nay have had 20 maybe, 25 maybe 25 separate alarms. ~,
1 Zewe 616 s 2 Q And as they came on during that period 3 of time, did you check out to determine what 4 particular alarm.had gone off? { 5 A A member of the operating staff would 6 have checked it out. 7 Q Had it come to your attention at any 8 time before four a.m. that the high temperature 9 alarm on the drain tank had gone on? 10 A As I recall, to my knowledge, I did not 11 think that the drain tank had alarmed that evening. 12 Q So it is fair to say, isn't it, that O 13 when you learned that the reactor coolant drain 14 tank temperature was above the alarm point, that 15 you knew that that had happened at some point af ter 16 the reactor trip? 17 A I believe I assumed that, yes. 18 Q Did you have any discussions with any do'n to the 19 of the other operators, before you went w 20 turbine room, about readings for temperature, pressure, 21 level of the drain tank? L 22 A Yes. 23 Q Did you discuss them with Mr. Scheimann? () 24 A I don't remember if it was with Mr. 25 Scheimann or not.
617 13 1 Zewe (G ~\\ 2 .Q You had a conversation with Mr. Frederick, 3 did you not, about turning off the sump pumps before 4 you went down to the turbine room? 5 A That is correct. ( 6 Q And in the course of that conversation 7 or any earlier conversation, did you talk to Mr. 8 Frederick about the pressure and temperature 9 readings at the drain tank? 10 A As I recall, I did. 11 Q Do you remember what you said to him 12 and what he said to you about it? h 4 N/ 13 A As I remember, Mr. Frederick was with 14 me at the RC drain tank panel,and we observed the 15 high temperature, low pressure, and off-scale level 16 at about the same time. And I remember discussing 17 with him about that fact, that I felt that there 18 was a hole somewhere in the RC drain t a nk,' and that 19 that is where the water came from that had gone to 20 the reactor building sump which he had ref erred to me 21 around that same time, that the operator had 22 notified him that~the reactor building sump 23 indicated full and that both sump pumps were running. 2-1 Q Did Mr. Frederick tell you that at some l 25 earlier point in the transient he had noticed that i i
1 Zewe O 2 both temperature and pressure were shown high on the 3 drain tank instrumentation? 4 A No, I don't remember that. 5 g Again referring to this period of time 6 before you went down to the turbine room, Mr. Zewe, 7 did you learn that there had been a sharp increase 8 in reactor building pressure? 9 A I was aware that the reactor building 10 pressure was above normal. I don't recall sharp 11 increase, but I do know that it was above normal 12 when I had inspected it. 13 Q Did you learn it had gone up from zero 14 to more than two pounds per square inch? 15 A What time frame are you referring to now? 16 Q The time frame before you went down to i 17 the turbine room. I 18 A Which is what time reflected there? 19 Q If it would be helpful to you,'I think 20 the GPU chronology reflects that the sump pumps were 21 turned off at 4:38, if that helps you place the time L 22 you went to the turbine room. 23 A Because I didn't recall exactly when I ( 24 left. It was a frame of about 24 after four until 25 20 to five. r- ~. m..... ... -. _.. -,. _ -. _, _. _ _ _ - - - _ -, _, _ _ _ _ _ _ ~..
bl9 15 1 Zewe O 2 Q It is correct that you didn',t go to the 3 turbine room until after the sump pumps had been 4 turned off; right? (( 5 A As I remember, that is true. 6 Q And it was within a minute or two after 7 that that you left? 8 A I believe that to be true. 9 Q Let me show you what has been marked as 10 Exhibit 616, which is a strip chart reflecting 11 reactor building pressure on the morning of the 12 accident. 13 A I have it before me. 14 Q Where was that strip chart displayed in 15 the control room? 16 A RB pressure indicated just in the panel 17 off to the operator's left. 18 Q Is there an arrow for that on Exhibit 7457 19 A Yes. 20 Q Does this chart, Mr. Zewe, reflect that 21 pressure went up to around two pounds per square L 22 inch within the first 45 minutes after the trip? 23 A Yes, it does. 24 Q At the time you learned that the reactor 25 building pressure had gone up to around two pounds,
k l Zewe 620 1 /~) l you concluded that that had been as a result of the 2 release of water from the drain tanks correct? 3 A As I recall, that is the conclusion I 1 4 i had reached. ( 5 Q Mr. Zewe, when in relation to your trip t 6 to panel 8-A did you learn that the level in the e 7 sump was high? MR. KLINGSBERG: Objection. I don't think 9 he has testified to his trip. MR. FISKE: Well, whatever. 11 A I don't recall with any accuracy, except 19 that within five or ten minutes, maybe.~ I don't 13 know. Maybe closer, closer than that. I am not sure. 14 Q And again, this is before you left to 15 go down to the turbine room? r 16 l l A As I remember, it was. 17 Q How did you learn that the water in the sump was high? 19 j A As I recall, the primary operator notified the control room that the liquid level 21 k-indicator for the RB sump was at six feet, and also 22 that the reactor building sump pumps were both 23 (~)) running. i ( 24 And the control room operator,in turn, 25
621 1 Zewe -x ~ 2 informed me. 3 Q And the control room operator was Mr. 4 Frederick? ( 5 A Yes, he was. O Q Who was the primary operator? I 7 A If I remember right, it was Terry 8 Dougherty. 9 Q You understood at the time that the 10 reading for six feet of water in the sump was as 11 high as the reading would go; isn't that right? 12 A 'It was full. /~T 13 Q And you concluded, did you not, at that 14 time, Mr. Zewe, that the water in the sump had come 15 from the drain tank? 16 A I concluded, with both pieces of 17 information on the high level in the reac, tor building 18 sump and the fact that the sumps were running and 19 the indications I had from the RC drain ta'nk -- from 20 those I concluded that the water from the drain tank 21 was the reason for the water in the sump. k. 22 Q Going back to your answer, your 23 statement that you gave to Mr. Long and Mr. Reppert, () 24 page 3, the same answer, near the bottom of page 3, 25 the statement is, "We figured that the water from the ( r i
1 1 Zewe 622 {j T s 2 drain tank was going into the sump because we did get 3 the running alarm on the computer and we had the 4 primary operator check the reactor building sump ( 5 level, and it was high." 6 Did you make that statement to Mr. Long 7 and Mr. Reppert? 8 A As I reme mb e r, I did, yes. 9 Q Now, you also by this time had observed 10 an alarm on the intermediate closed cooling system 11 monitor? 12 A The time frame of these alarms on the \\_) 13 iptermediate letdown monitors, I am not sure of the 14 exact time frame, if it occurred right whenever we 15 knew that the sump was full and we secured the pumps 16 and I noticed the drain tank. 17 I recall seeing the alerts in.on these ^ 18 particular monitors, but the exact time frame I am 10 uncertain of. 20 Q First of all, these are monitors that 21 measure radiation? L 22 A Yes, they are. The intermediate closed 23 cooling water system cools the letdown coolers, and ( 24 they are located directly outside the reactor 25 building sump cage.
1 Zewe 623 2 Q Did you see these monitors in alert or 3 in alarm or both? 4 A That day I had seen them in both alert ( 5 and alarm. I believe the first time I noticed them, 6 they were just in alert. I am not sure when that 7 was, because as I recall, these monitors have 8 indicated an alert and even alarm at other times 9 during normal operation, and I remember making a 10 comment about that, that we needed to reevaluate 11 their setpoints so that they would be more useful, 12 that the normal background radiation seemed to be 13 having them in'dicate abnormally high to the alert 14 and into the alarm range when there was no problem. i 15 Q And that is c comment that you made 16 before the day of the accident? 17 A Yes, it was. ~ 18 Q Who did you make that statement to, Mr. 19 Zewe? 20 A I don't recall for sure who it was. I 21 seem to remember discussing that with Mr. Weaver, 22 though I am not certain that it was him or if there 23 was only him. 24 I believe I discussed it with other 25 people, though I don't remember exactly who. i .----m igg ---++e-r7-g.g--w-t e

,,w e

wy w g--e- ,-e
1 Zewe 624 , 7x \\_ 2 Q My next question was going to be, other 3 than possibly talking to Mr. Weaver about it, what 4 other steps did you take before the accident to try 5 to put in motion a change in the setpoint of those g 6 monitors? 7 A As I remember, all I had asked for, for 8 it to be reviewed to see what could be done to make 9 them more useful. Because as I recall, at times it 10 was a case to where they would come into alert and 11 then clear and then come in again at another time, 12 and I felt that it would be more usefulifwewchid f) 'd 13 determine if we could not put in another setpoint 14 that would be more helpful. 15 Q For how long before the accident had this e 16 been. going on? That is, the monitors going into 17 alert under normal conditions. 18 A I don't recall the time frame, except 19 that once we got the power levels in the neighborhood 20 of about 85 percent to 90 percent range seems to be l l '21 the time at which it became a problem. In my (_ l 22 estimation, a change needed to be evaluated. 23 Q And that was for at least several weeks i p 24 before the accident? . j l 25 A Yes, at least. t i
L' Zowe 625 2 Q What to your knowledge, if anything, was 3 done to evaluate this situation during that period? 4 A I don't believe I was aware of the (( 5 current status of it, other than that it was being 6 evaluated. It was more of a nuisance thing to me, j 7 not anything that needed to be taken care of in i 8 rapid order of events, if you will, but something 9 that I felt they were looking at, evaluating, and I 10 didn't know at that particular point skactly how far L 11 along or whe-all had reviewed it. 12 Q iou understood, didn't you, that these O 13 monitorc were supposed to be a guide for you in i 14 determining whether or not there was more radiation 2 15 in the system than there was supposed to be? 16 A Yes and no. And I will clarify that. 17 It is not to be used as an indicator of 18 how much' radiation is passing through the letdown 19 coolers, but if there is a leak'in the spiral tubes 20 in the letdown coolers from the makeup letdown into i 21 the intermediate closed, the intermediate closed 22 should not have any radiation level at all, so that 23 it would reflect that we had a cooler leak -- all ) 24 right so that any levol above background that was 25 in the intermediate cooling,.one of the sources may
~. 't 1 1 Zewe 626 /~N ( ) s w' 2 be a leak in a letdown cooler. e 3 Even though$'as I stated before, that i 4 through the course of operation'through a cycle,.the ( 5 activity or radiation level of the cooling system 6 goes up -- all right you would only communicate 7 any of the radiation at any level if you had a leak. 8 So this is primarily saying if you had a 9 cooler leak, more so than an indication of radiation s 10 level being high. It just showed that'you had ~ l. a. 11 problem aNd you had the potential then of 12 contaminating a closed cooliig water system-that O \\'/ 13 should not be contaminated. e 14 Q And~the monitor was designed to help you ,.I 15 diagnose that? '/ 16 A Yes, it was. 17 .Q And y'ou, felt for several weeks before the 18 accident that the us'efulness of that monitor was 19 diminished because it used to go off even under 20 normal' conditions; right? s 21 A Yes, because we would use that as an L 22 indication. The real means of detecting whether we 23 had.a cooler leak was really used by a sample.of the (( h) 24 i n te rme dia te closed, as a firm backup that would' 25 ' confirm any of the indications that we may he.ve on o
t / 1 Zewe 627 fsn V 2 the letdown intermediate closed monitor. 3 So that' was used in conjunction with -- 4 all right a sample from the water itself, and it L( 5 was really that sample that would determine what 6 the leakage rate was.and if it had reached a problem A 7 at which we would have to take action or not, plus 8 there were other radiation monitors within the 9 intermediate closed system itself, regardless of 10 these two, that would also monitor the intermediate 11 closed. ~ 12 So you used those monitors plus the N/ 13 system monitors plus the coolant system sample. You 14 used all of those. 15 But this would -- all right as a 16 first indication may be misleading if it alarmed too 17 soon. -' ~ 18 Q Isn't it correct on the morning of the 19 accident when you saw these alarms in their s 20 abnormally high condition, that you concluded that / 21 that was because of background radiation resulting k. 22 from this water in the sump rather than from some i 23 leak into the cooling system? yt. J 24 A I did conclude that I thought that that }e t' 25 was from the location of them near the reactor T ) f l 1 t .. w. =,_
1 Zewe 628 /~h Q 2 building sump and the water that had gone into the 3 reactor building sump was contaminated and that these 4 monitors reflected that and it did not reflect a ( 5' cooler leak themselves; that is correct. 6 Q . Do I understand your earlier testimony 7 correctly that at some point during the accident 8 sequence, you saw these alarms in alert and at 9 another point you saw them at alarm? 10 A That I don't remember for s'ure. I am 11 sure that I had seen them in alert and alarm 12 sometime later. All right? Whenever I declared the -w 13 site emergency, they were in alert and. alarm at the 14 same time. ~ 15 Whether earlier in the event is when I 16 first noticed that they wero the only ones that were 17 reading in alert, I don't remember if during that 18 time, prior to everything going into alert and alarm, 19 if they were just in alert or if they were in alert 20 and alarm at the same time. I don't remember that. 21 Q Let me go back to the statement that you 22 gave on March 30. 23 About seven or eight lines up from the (A_) 24 bottom of page 3, it reads, "It just didn't make 25 sense and then the reactor building pressure was
i 1 Zewe 629 OU 2 coming up and we were up to just about two pounds at 3 this time and I figured that was from the reactor 4 coolant drain tank. At this point the only radiation 5 monitors that we had abnormally high were the 6 intermediate letdown monitors which have a very low 7 setpoint, and they are very susceptible to background 8 and they are right by the reactor building sump 9 itself.' 10 Did you make those statements on March 11 30, 19797 12 A As I recall, I did. (")N ~ \\ 13 Q Am I correct that the sentence that 14 starts "At this point" is referring to a point in 15 time at which the reactor building pressure was 16 coming up to about two pounds? Or putting it more s 17 simply, isn't it correct that the sentenc'e that 18 begins "At this point" is re fe rring to the point in 19 time before you went down to the turbine room? ~ 20 A That may be a poor use of the words 21 "At this point." 22 As I recall now, I don't think that I 23 could pin it down to at the time I noticed that it 24 was two pounds in the building. I don't recall that 25 now. I may have then, but I don't recall linking
1 Zewe 630 /~T U 2 the two pounds and "at this point," 3 It may just be a poor choice of words. 4 I don't recall. ( 5 Q Your testimony is that today, May 26, 6 1982, you don't recalls is that correct? 7 A I don't recall, no. 8 Q You are not saying that you didn't 9 remember on March 30, 1979, two days after the 10 accident. 11 A If I don't recall now what it was then, 12 I certainly could not say that I did recall it then, f-( 13 because I don't recall. 14 Q Can you tell us, Mr. Zewe, what you 15 meant when you said that these radiation monitors 16 we re abnormally high? l 17 A Abnormally high to me is whenever a 18 monitor is in alert or alarm, when I feel that it 19 should not be. But in the relationship to these 20 intermediate letdown monitors, I have stated that I 21 felt that they were at a low setpoint and that they 22 were always abnormally high when we were at power 23 level more so than what they should have been, so O(_/ 24 that is why I had asked for them to be evaluated. 25 As I remember, I was trying to make a
1 Zewe 631 (O) s-2 point here or at least to convey some information 3 showing that I didn't have any indication that we 4 had any radiation problem either outside the reactor ( 5 building or inside the reactor building. And I was 6 just trying to point out that fact at this 7 particular time. And I just mentioned those are the 8 ones that I remember did indicate an alert or alarm, 9 whichever was the case, and that nothing else had 10 been, but I I felt I knew why they were. Because 11 of the low setpoint and because of the water at the 12 g-power level. N )g. 13 Q ,What you were saying is that you had no 14 indication of any radiation problem other than 15 whatever level of radiation from the water in the 16 sump was necessary to actuate these particular 17 alarms; is that right? 18 THE WITNESS: Please read that again. 19 (Question read) 20 A No, it's not. I was trying to relate 21 that I didn't feel that we had any radiation problem 22 at all, including these alarms which I felt that, 23 even though they showed an alert or alarm, that it 24 was really not a problem. That I expected them to be 25 high due to their low background. Even the water
_=- 1 Zewe 632 0 2 from the RC drain tank I didn't expect to be ve ry 3 contaminated at all because we were still relatively 4 early in life. The reactor coolant system as I ( 5 remember had c very low activity level, and I didn't 6 feel that that will add very significantly to the 7 background reading that these monitors would see. i 8 Q Let me put the question a little 9 differently. I think you answered it, but I should 10 rephrase my earlier question. ~. 11 Am-I correct that at this point in time 12 you didn't see any indication of any level of O 13 radiation, other than whatever level of radiation 14 coming from the water in the sump had activated the 15 1092 alarms? 16 THE WITNESS: Read that back again, 17 please. 18 (Question read) r l 19 A As I thought that I had stated, I felt 20 that the level of radiation as seen by these 21 monitors, intermediate letdown monitors, were there (. 22 because of the power level t.:at we were at and their 23 low setpoints. I didn't feel that there was much of 24 a contribution due to the water that I thought at 25 some point in time went from the drain tank to the
1 Zewe 633 ~ 2 sump. '3 If I had just been operating normally 4 and if we-had not had the trip or anything else, I 5 would still have expected those radiation monitors' ( 6 to be at about the same place that they were when I 7 looked at them at this particular time. I didn't 8 know that there was any significant increase that was 9 afforded by this water once I determined that that is I 10 where the water had come from. 11 Q Now, Mr. Zewe, how did you find out that 12 the sump pumps had come on? O 13 A Mr. Frederick informed me. t 14 Q Did you learn from Mr. Frederick how 15 long they had been on? 16 A As I remember, he informed me that both 17 sump pumps were running and that the operator had 18 told him that the reactor building sump tank was at 19 its maximum six feet. 20 I don't recall the information on how 21 long they had been running. L 22 He made the suggestion that we secure 23 the sump pumps,.tnd I agreed with that based on the 24 fact that I did not want to overflow the auxiliary 25 building sump tank in the auxiliary building,
1 Zewe 634 O V 2 knowing that I did not have much freeboard left. And 3 he then informed me that he would tell the operator 4 to secure them. ( 5 Q Did you know what the capacity of the 6 pumps was the day of the accident? 7 A I don't believe I knew the exact 8 capacity, no. 9 Q Did you know that the capacity was around v 10 140 gallons per minute? 11 A Each pump? pg-12 Q. Yes. L) 13 A Not that I recall, no. 14 Q Is that number wrong? Do you know today? 15 A I don't know today, no. I w a,s thinking 16 more in.the neighborhood of 70 gallons a minute, 80 17 gallons a minute maybe at the outset. 18 Q How many pumps are there? 19 A Two. ~ 20 Q Which would make the combined capacity 21 140 gallons per minute; right? 22 A Yes. 23 Q Did you know how much water the sump Os 24 could hold when it was full? 25 A I believe at that time I had a relative
1 Zewe Oi 635 2 idea of,how much it held. 3 Q What was that idea? 4 A I don't remember at this time. But as I ( 5 recall, I had a feel for all the various tanks and 6 about what their capacity was. Right now I have a 7 hard time separating Unit 1's with Unit 2's, so I 8 really don't remember. 9 I think then I had a relative idea on t 10 how many thousands of gallons it would hold to its 11 capacity. (~T 12 Q Does 2500 gallons sound about (,,/ ~ right, in 13 terms of what your understanding was on the day of 14 the accident? 15 A Th a t,- as I recall now, would.seem a 16 little icw. 17 Q Is it fair to say, Mr. Zewe,'that 18 whatever it was, in fact you knew it at the time? 19 A Relative magnitude, yes. Not exact 20 numbers, I don't believe. 21 Q Is it correct, Mr. Zewe, that at some 22 point after you learned that the pumps had come on, 23 one of the operators looked at the alarm printer k ') 24 relative to the pumps coming on? 25 A I don't know. I don't remember.
1 Zewe 636 b(s 2 MR. FISKE: Mark the next exhibit, s 3 please. 4 (Transcript of an interview Mr. Zewe ( 5 had on April 23, 1979 with Mr. Marsh and others 6 from the Inspection and Enforcement Division 7 of the Nuclear Regulatory Commission marked 8 B&W Exhibit 754 for identification, as of this 9 datel 10 Q Do you have Exhibit 754 in. front of you, 11 Mr. Zewe? 12 A Yes, I do. i 7g(_) 13 Q This is a tranceript of an interview 14 that you had on April 23, 1979 with Mr. Marsh and 15 others from the Inspection and Enforcement Division 16 of the Nuclear Regulatory Commission. I direct 17 your attention to page 22. 18 Would you like to read the questions and 19 answers on page 22? 20 A Yes, I would. 21 I have read it. 22 Q Let me direct your attention to the 23 statement that you make beginning around, or that \\ 24 appears beginning around line 14, where it says, 25 "Well, see, that's why it, we had the sump pump 1.
1 Zewe 637 OV 2 secured because they alarmed on computer saying that 3 they were both running. And we knew at that point, 4 that something had happened, in fact, to the RC drain ( 5 tank, and that the water from the RC drain tank wa's 6 going into the sump.." 7 Do you see those two sentences? 8 A Yes, I do. 9 Q Did you make those statements to the ( 10 gentlemen from the Nuclear Regulatory Commission back 11 in April of '797 12 A As I remember, yes, I did. 1 ~ 13 Q What computer were you re ferring to when 14 you said that the sump pumps had alarmed on the 15 computer? 16 A. The Bailey 855 computer that is in the 17 control room. The alarm printer. ~ 18 Q And that alarm printer indicated, did it 19 not, the time at which each pump had started running? 20 A As far as I can remember, the only time 21 that the alarm printer prints anything on the sump 22 or on the pumps is when both pumps are running. 23 In fact, I did not review the alarm on O(_) 24 the computer that morning to determine if the pumps 25 were running. I was aware of the fact at the time .-.~. -
1 ~ Zewe 638 (~) \\J 2 of this testimony that it had alarmed on the 3 computer, but I myself did not read it. 4' Q You knew during the course of the ( 5 accident that they had alarmed on the computer, 6 didn't you? 7 A Not that I remember. I knew that they 8 were on from what Mr. Frederick had told me. Whether 9 he had told me at that point that he had noticed that t 10 they were in alarm on the computer or whatever, I 11 don't remember. As I remember, he had obtained that 12 information from the primary operator along with the 13 level in the tank, and that is my recollection of 14 how I knew about it the morning of'the accident. 15 Q But do you have a different r,e colle ction 16 about that than you did on April 23, 19797 17 A Today, and also on April 23 when this was 18 made, I was aware that there had been an alarm on 19 the computer saying that they were both running. I ~ 20 am only stating that as I remember, on March 28 I 21 did not look at the computer alarm or remember that 22 they had alarmed on the computer, because I don't 23 remember going over the alarms on the computer, and I \\-- 24 don't remember the other control room personnel 25 going through the alarm printer and picking out f
1 Zewe 639 (w/ 2 things like sump pump running on the alarm, printer 3 either. 4 In fact, I was aware of it at this point ( 5 in time when I was talking about the NRC because,I 6 had reviewed all that subsequent to the accident, and 7 it wouldn't surprise me it should be alarmed on 8 there because it is one of the monitors, but I don't 9 remember wanting to go to look at it or, in fact, e 10 looking at it or having anybody else lo*ok at it, or 11 indicating to me that they had looked at it for that gg 12 purpose. NA 13 Q Was it of any interest to you, in the 14 first 45 minutes of this transient, to find out how 15 long the pumps had been running? 16 A At that time 17 Q Yes. 18 A It was not of interest to me at that 19 particular time. Typically the pumps would start on 20 just about every shift and pump down to the aux. 21 building because we had various support system 22 leakage inside the reactor building and condensation 23 mainly from the normal cooling fans in the reactor s.,/ 24 building. We used to get a lot of condensation and 25 leakage from these fans that would be communicated to
1 ~ Zewe 640 \\_/. 2 the' reactor building sump and they would periodically s 3 pump over to the aux. building. 4 So at any particular time, I wouldn't ( 5 know at what level the reactor building sump was at 6 in relationship to full or empty or whether it had 7 just pumped or not, because normally one sump pump 8 would pump and it would pump it down and the next 9 time the opposite pump would start. And unless you 10 were at the rad waste panel in the auxdliary building 11 watching the pumps automatically start and stop, I 12 wouldn't know when it had started and stopped. And 1 13 at the time it didn't concern me much to know how 14 long they had been pumping. 15 Q You did know, didn't you, on the day of 16 the accident, that there was a way for you to find 17 out how long they had been pumping if you had wanted l 18 that in fo rmation? i 19 A only if both sump pumps were running, 20 which I knew in this case they were. t 21 Q All right. 22 A But typically one pump will start to l 23 pump and then if it can't handle it, the other one l 24 will start. I would only be aware of when they both 25 started, and I would not be aware of what the level
1 Zewe 641 0 2 was tp begin with. i 3 But from the point that I could i 4 determine when they both started from the computer ( 5 alarm and up to the time that I found out that they 6 were still running, I could determine how long they 7 -had pumped. And knowing their capacity, I could find 8 out how much water was transferred. 9 I could also use the recorded levels 10 from the tanks that these pumps pumped *.to over some 1 11 differential time period and determine, from a 12 change in level, how much water was being O 13 transferred. 14 Q You did know at the time Mr. Frederick 15 brought it to your attention that the sump was full, t 16 despite.the fact that both pumps had been running for 17 some period of time? 18 A I don't recall being aware of what that 19 time period was. It may have just started. It may 20 have been on for some time. I did not know. 21 Q If the pumps pump 140 gallons a minute, 22 finding out how long the pumps had been running j 23 would enable you to find out how much water had 24 already been pumped out of the sump before you 25 learned that it was full; correct? . _ _ - _. ~. _ _ - _. _ - -.
~ I Zewe 642 2-A I could determine the volume of water i 3 transferred, assuming those rough conditions, yes. 4 Q And that in turn would enable you to ( 5 de te rmine how much water had in fact gone into the 6 sump since the transient had started. 7 A Not correct. 8 Q Well, it would enable you to determine 9 how much water had gone into the sump since the 10 pumps started running? 11 A Since both pumps started running 12 Q Yes. 13 A -- yes. I feel a much more accurate 14 case would be to monitor the level of the tank that 15 it was going into. 16 Q And did you do that? l 17 A I did not. Only from the senses that I 18 wanted the operator to secure the pumps based on not 19 overflowing one of the tanks in the aux. building. I i 20 only from that standpoint, as I didn't want them to 21 exceed the capacity of it. Not that I was trying to 22 find out how much water was being transferred. 23 Q Is it correct, Mr. Zewe, that you did 24 not take any steps to try to find out how much water 25 had in fact been pumped out of the sump?
1 Zewe 643 0 2 A I did not. The only steps were to stop 3 the transfer over to the auxiliary building. 4 (Time noted: 4:48 p.m.) ( 5 'i WILLIAM H. ZEWE 7 8 Subscribed and sworn to 9 before me this day of 1982. 11 12 13 14 15 16 17 18 19 20 k_ 22 23 24 25
644 1 CERTIFICATE 2 STATE OF NEW YORK ) 3 33,. COUNTY OF NEW YORK ) 4 I, HARVEY B. KRAMER, RPR, CSR a Notary Public of the State of New York, do hereby certify that the continued deposition of WILLIAM H. zEwE 8 was taken befcce me on WEDNESDAY, MAY 26, 1982 cons 13 ting i of pages 480 through 643 l 10 I further certify that the witness had been previously sworn and that the within transcript is a true record of said testimony; That I am not connected by blood or marriage with any of the said parties nor interested directly or indirectly in the matter in controversy, nor am I in the employ of any of the counsel. 18 IN WITNESS WHEREOF, I have hereunto set my 19 ~ hand thiX i day of 'M b lN 1 20 21 (. l O'M W -{ c w. 23 HARVEY B. KRAMER, RPR, CSR 24 25
a 645 f- 'd IND E X WITNESS PAGE William H.* Zewe (resumed) 482 EXH I B I T S B&W FOR IDENTIFICATION 748 Testimony of Mr. Zewe to th.e Nuclear Regulatory Commissi6n Inspection and Enforcement Group on April 12, 1979 484 749 Transcript of the interview Mr. Zewe had with various individuals at TMI on April 6, 1979 486 750 Strip chart of T-Av on the morning of the accident 523 751 Graph from the LER filed by GPU 555 752 Page of information that was printed out by alarm printer ' 572 753 Transcript of an interview on - March 30, 1979 with Mr. Long and Mr. Reppert 612 754 Transcript of an interview Mr. ( Zewe had on April 23, 1979 with Mr. Marsh and others from the l Inspection and Enforcement Division of the Nuclear Regul'at o ry Commission 636 ) ~"~ -o00~ f l - -}}

ML20072H804

Text

Navigation menu

Search