ML20212A521
| ML20212A521 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 07/27/1982 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML20211Q614 | List: |
| References | |
| FOIA-86-151 NUDOCS 8703030368 | |
| Download: ML20212A521 (161) | |
Text
{{#Wiki_filter:-
,g , .
NUCLEAR REGUI.ATORT CCMMISSION I L u l IILtIIe .M M Cf3 , DIABLO CANYON, INDEPENDENT SEISMIC EVALUATION OF THE CONTAINMENT ANNULUS STRUCTURE AND SELECTED PIPING SYSTEMS MEETING WITH BROOKHAVEN NATIONAL LABORATORY, TELEDYNE ' ENGINEERING SERVICES, AND PACIFIC GAS & ELECTRIC CO. ( . DATZ: July 27, 1982 PAGES: 1 thru 137 AT: Upton, New York 97o3030360
' 060304 I Dit
(([MEO 2 131
~
\
s A.LDfGUO.%*#- (REPORT 1.)*G 400 Vi vi.~.ia Ave., 5.W. 'J E ng.mn, D. C. 20024 Telepheme (202) 554-2345
..--......-....u.. . . . . . . . . , . .. . . . . , , . . . . . - . * , - ----.=,%.,g..
- ^- " _m..k
- A 1
1 UNITED STATES NUCLEAR REGULATORY COMNISSION 2 - -- 3 DIABLO CANYON, INDEPENDENT SEI$MIC EVALUATION OF THE CONTAINNENT ANNULUS STRUCTURE AND SELECTED PIPING 4 SYSTEZS 5 - -- 6 NEETING WITH BROOKHAVEN N ATION AL LABORATORY, TELEDYNE ENGINEERING SERVICES, AND PACIFIC GAS C ELECTRIC CO. 7 8 Brookhaven National Laboratory 9 Berkner Hall Conference Room - Upton, Long Island , No w York 10 Tuesday, July 27, 1962 11 The parties to the meeting met, pursuant to 12 notice, at 5:40 a.m., James Knight presiding. 13
' PRESENTa '
14 For the Nuclear Regulatory Commissions 15 H. Schierling 16 P. Norrill F. Miraglia 17 J. Knight P'. T. Kuo 18 H. Polk
- 3. Harteman 19 B. Jones 20 For Brookhaven National Laboratorys 21 N. Reich A. Philippacopoulos 22 P. Bezier ~
P. C. Wano 23 C. Miller . J. Currori ' 24 M. Subudhi - 25 ALDER $CN REPORTING COMPANY,INC, 400 VtRGINIA AVE., $.W., WASHINGTON, D.C. 20024 (202) $54 2346
._.--.~__._...;7.- . . . . - . . . . _ . . . - , . ~ , . . .
-_..,,7.-- . , . . , . . ~ - _ . - . =-
-- ~
- - . .- : . , . , _ . = - r; p o .3 -
t I
! , 5 -
/ ,'1 1 PEESENT: (Con tinued )
2 For the Independent Design Verification Projects 3 W. E. Cooper E. Danison 4 R. L. Cloud ' S. Harris - s 5 P. Anderson s
! R . W ra y ' ; 6 N. Holley - ' J. Bigga -
7 R. Cintto ' S. Chin .. . 8 c For the Diablo Projects y , 9 - .: . v, H. Friend, Me:htel to J . H och, PGCE ' B. Norton, PGCE ~ ^ 11 B. Lew, PGCE , L. Malik, URS/Blume 12 W. White, Bechtel ' N. Tressler, PGCE 13 L. Shipley, Sechtel - T. Esselman, Westinghouse (. 14 R. Locke, PCCE ' R . T ra y , PGCE 15 P'. Hirschberg, ?GCE , s 16 , - 17 18 . 19
- 20 l* '21 l
22 s l 23 . 1< 25 --- l I' l ALDERSON PEPCRftNG COMP 4/M INC. 400 vtR0lNIA AVE., S.W., WASHINGTON. O C. 20024 (202) 864 2348 t _ . _ . . . _ . - . . , . __ = _
.. _ - ~. . .
+ _ ,
~_
1 3 1 Ea2CIIRI1EE 2 MR. KNIGHTS ladies and gentlemen, we do seem 3 to be assembled. If we could, I would like to get under 4 var. 5 I am Jim Knight. I am Assistant Director for 6 Construction Engineering at the NRC. I will introduce 7 scue of the other members of the NRC staff here in a _ 8 ucrent and ask that each of the groups introduce at 9 least their principal members to the extent that they to find it desirable. 11 Just a couple of administrative details. This 12 meeting is being recorded. The court reporter is to my 13 left. And therefore, somewhat at difference with the 14 usual technical meeting, I' must ask each person'when is they speak to identify themselves so that the record to will be clear, and if in the latter part of the- meeting
. 17 a heated discussion should evolve, we vill try to let 18 each party finish their points before speaking so that 19 the reporter can keep up with us.
20 We plan to take r, break around 10:00 o' clock 21 both for the benefit of ourselves and the reporter, and 22 ve vill break at 11:30 for lunch. 23 This is an NPC meeting. Our purpose bere is 24 to provid e sn opportunity for Teledyne, as'the 25 independent reviewer on the Disblo Canyon plan, to ALDERSON REPORflNG CCMPANY, INC, 400 VIRGINIA AVE. S.W., WASHINGTON, O C. 20024 (202) $54 234$
, a- w.. : '
~-.. -1.
~" -
*J :. ' : . .= ; . * ~
,P7 * ,
.% g
. . s 1 obtain information and clarifica tion on the work
,i s 2 recently completed by Brookhaven on the analysis of the
,M :3 annulus region on t$le Diablo Canyon' Unit 1. m i 4 I think Brookhaven obviously, as we are all 5 well aware,' Brookhaven has accomplished its work as a
. 6 consultant to the NRC Staff. The Staff itself is 7 reviewing the Brookhaven report at this time. I think _
8 that it is appropriate to keep a certain historical 9 perspective in mind here, and I think it will help us in T 10 our discussions. That goes to the question of why did 11 the Staff ask Brookhaven to do this work. What was the , 12 purpose? s . h 13 ~ During the early stages of discussions that i 14 evolved after'w'e all became aware that there were errors 15 in some portions of the seisaic analysis in Unit 1, the 16 Staff felt tha t the complexit'y of the situation was j .17 sufficient that we, needed let me call it a benchmark. ( 18 We needed a point of departure. We needed some " 19 reference that would allow us to sharpen our insights as l l 20 to just exactly what the problems were and what the 21 impact of these problems would be, and in that light 22 there were two'somewhat separable questions., 23 One was -- well, one went to-the point of , i 24 whether or not thefplant-vas built in accordance with e - j 25 the application, she . legal ~requ-1-rements of the license. 1 ALDER $oN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345 i = .
. = = . -
- s. - = . - . w __ = - ==-
, _j - . u_ _._ m w __ < < 2 ul i
- i l
, 5 1 And as I said, the other, somewhat separable, was the 2 plant safe, if you want to use the simplistic word. Was
, 3 it acceptable?
I
\
4 We asked the folks at Brookhaven to take a e 5 small portion of the plant and since the discussions to . 6 that point had centered on the annulus region, we chose 7 the annulus region; sin.ce they had centered on vertical _ 8 seismic analysis, we selected the vertical analysis. We 9 asked them to take that set of conditions and to do what 10 one might call a best available analysis, without 11 ref erence to the time scope of when the analysis, 12 original analysis was done or what the techniques were 13 at that time or anything of this type, simply do the ( 14 best job that 'the technology would allow at'this time, 15 the idea being there that we would then have hopefully a l 16 point of ref erence that everyone could agree on. 17 From there we could then make judgments as to 18 whether, in fact, the procedures that had been employed 19 were consistent with the application, and in fact,. 20 whether or not the procedures that are employed and the l 21 various steps in the process yielded a result that was 22 acceptable.
'23 And I think it is important also that_we keep l
24 in mind the central thrust here of the staff's review, . 25 or re-reviav, if you will,-in_this instance. Although ALDERSoN REPORTING COMPANY,INC. 400 VIRGINIA AVE., S.W., WASHINGTON. O.C. 20024 (202) 554 2345
' ~ ~ - - ~ ~ - ~ ~ ~
3- .8 : - . ;, -- _ - , , ~__--=3-~'--~~
~
- .L ;
- _,-
F l _ 6
~
1 we use as a tool'.a certain.$ortion of.the plan, a 2 certain methodof. analysis,- overriding continually is 3 the question, given the res'ults of any specific analysis 4 or any specific r.eview, what is 'the import of that
; 5 result to the generic question, to 'the overall efficacy, i
f 6 if you will, of the process that was used throughout the 7 design of the application. That is continually.where .. 8 the thrust of the staff review is going. 9 I want to make that point to get away from the 10 narrow question of, you now, all right, we looked at 11 component Z and the stresses were acceptable. That is, 12 of course, fine, sud that is a step in the process, but 4 13 overriding continually will be the question, out of this 14 review, do we 'o'r do we not discern that the 'overall i 15 process that was used is one that should and would lead 16 'to acceptable results? l 17 We have asked Teledyne, by letter from Mr. I 18 Denton, to consider the Brookhaven report along with the 19 numerous other steps that they are taking as an 20 independent review agent. 21 Today I would hope -- and I certainly will l 22 work to see that this is true -- to concentrate upon, if " l 23 rou will, what was done, what information was made 24 available, how that information was utiliz'ed by - 25 Brookhaven, what analytical met-hods they us ed, what l ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345 l ._ y- -
,.~""_--
~ - - -
- m. ;T _ _
, e -
N 7 1 steps they took, and ulti'mately, what'results they got. 2 It.is easy in my view for us to depart quickly 3 from this central theme, and as I said, I will work - 4 diligently to keep us back on that track so that we can 5 accomplish the purpose of today's meeting. 6 There may well be other sessions, I would. 7 predict that th'ere probably will be, where we would - 8 either get into greater detail such a going through a 9 given computer run or-a' number of them, but I trust that 10 today we will set the. groundwork for those later 11 sessions, if they are-necessary. 12 To the extent that there are peripheral 13 matters that bear, let's'say, on the annulus region but-i .. t 14 are not to the point'specifically of what did Brookhaven 15 do to prepare this report, 'I would hope to put those off l 16 to the latter part of the day. As I said, I have no 17 ambition to embargo an opportunity to exchange 18 information, but I would like us to get to the central 19 point of the meeting first and be sur.e that we meet-20 that. l 21 I do not regard today, Staff does not regard l
~
! 22 today as a conclusion meeting. There is no , desire, nor 23 do I believe it feasible to end the day with some l 24 conclusion that someone was right or someohe was wrong, - 25 or some approach was best 6r seme approach was second I l s
- ALDERSON REPORTING COMPANY,INC, i 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
I 8 1 best, but primarily to have a full and complete 2 understanding at the end of the day as to exactly the 3 proce'ss and procedure employed by Brookhaven. The 4 B'rookhaven report will eventually be issued as a NUREG 5 report. I believe it is in printing now, so that other 6 copies would be available. 7 We are going to opea the meeting with Dr. _ 8 Reich making a short presentation. I think it will give 9 us a good context for further discussion. 10 The intervening parties are represented here 11 today. I would propose that toward the end of the day, 12 if they so choose, that any sta temen t th ey would care to 13 make could be made in the latter part of the af ternoon. k 14 If that is not acceptable and not ' desirable 15 from their standpoint, then perhaps during the first 16 break we can talk. 17 We have such a large group here, as I said, I 18 will ask each group to make their own decision as to how 19 far they want to extend introductions. Since the Staff 20 is the smaller contingent, I t$hink I will ask that we 21 start here at the end of the table and make our 22 introductions. . . 23 MR. SCHIERLING: I am Hans Schierling( the . 24 Project Manager for the verification ef fort. - 25 MR. MORRILL: I's- B1.114cerill, Operations ALDERSoN REPORTING CCMPANY. INC. 400 VIRGINIA AVE., S.W. WASHINGTON. 0.C. 20024 (202) 554 2345
~~~
. . . ._ .-:^~~'? "' r .
Y~ . I'"~~~~ *
'~~~
, m _
i . t t l 9 1 1 Inspector, Region 5. 2 MR. HIRAGLIA: Frank Eiraglia, Division of
; 3 Licensing, NRC.
4 MR. KUOs P. T. Kuo, Structural Engineering 5 Branch. 6 HR. POLK 4 Harold Polk , Strutural Engineering 7 Branch. - 8 ER. HARTZMAN Mark Hartzman, Eechanical 9 Engineering Branch. 10 MR. JONES 4 Brad Jones, Executive Legal 11 Director 's office. 12 ER. REICH Lawrence Reich.
! 13 ER. BEZLER: Paul Bezier. ! 14 MR.'WANGa P. W. Wang.
15 HR. MILLER: Charles Miller. 16 NE. CURRERI: John Curreri. 17 MR. ANDERSON: Paul Anderson, Robert Patt l l 18 Associates. t 19 HR. HARRIS: Steve Harris, Robert Patt 20 Associates. ' i 21 ER. CLOUDS Bob Cloud. 22 MR. DENISON: Ed Denison , Fox Associa tes. 23 MR. C00PERa Bill Cooper, GS. _ 24 MR. WRAY: Bob-Wray, TDS. -
.25 MR. HOLLEYa Ch rfs Ho-liey, . Consul tan t for l
i-ALDERSoN REPORTING COMPANY. INC. 400 VIRGINIA AVE., S.W WASHINGTON. D.C. 20024 (202) 554 2345 _ _ .,, L. ....7-..-.,,.,-,-- - - . - , - - - - - - - - - - - - - - - - - - - - _ ~----,--~-----~~;-,.---
m a , - - - - - - 10 1 Teledyne. 2 NR. BIGGS: J. Biggs. 3 NR. CIATTOs Ray Clatto. 4 MR. CHIN: S. Chin, TDS. 5 ER. FRIEEDs Howard Friend, Bechtel. 6 MR. HOCH: John Hoch, Diablo Project. 7 ER. RORTON: Bruce Norton. -- 8 MR. LEW Bob Lew. 9 HR. HALICK: L. Malick. 10 MR. WHITES Bill White, Bechtel. 11 NR. BUCHERT: -Ken Buchert, Bechtel. 12 NR. THESLER: Mike Tresler, Diablo Project. 13 HR. SHIPLEY: Larry Shipley, Bechtel. 14 ER . ' E'SSLEN AN To'd Esselman, Westinghouse. 15 MR . ', LOCKE s Richard Locke, PGEE. 16 HR. FRAY: R . F ra y. 17 MR. HIRSCHBERG Paul Hirschberg, Diablo 18 Project. 19 MR. HUBBARD: Richard Hubbard, representing 20 Governor Brown. 21 HR. DYNNER: Alan Dynner, Counsel for Governor 22 Brown. , 23 MR. FLEISCHELER: David Fleischeler - 24 representino the joint Intervenors. - 25 MR. KAUSEL: Edwa rd-:farsel. representing David ALDERSoN REPORTING COMPANY,INC. 400 VIRGINIA AVE, S.W., WASHINGTON. D.C. 20024 (202) 554 2345 w .,.--r-w--. .. --. -n -- -. n--= ,
p I I 11 1 Fleinhaven. 2 MR. KNIGHT Dr. Reich? 3 NH. REICH I would like to start with a 4 general statement. BNL has never conducted an analysis 5 to determine whether the plant, Diablo Canyon plant, was t 6 safe or unsafe, and since we have not considered this l ~, 7 question for NRC, we hold no view pertaining to it. _ 8 What we have done, we have compared our 9 vertical floor spectra with those of PGCE and their 10 contractors, and we have found places where we agreed, 11 and we have places where we disagreed. There are in 12 f act places where PGEE results were more conservative 13 than ours, and there are places on the~ structure where 14 they are not. ! 15 Let me go to the task outline as presented to l i 16 us by NRC, and from there on I will make a short comment f 17 on the results of our study, and after that we can have 18 individual talks about results, or questions. 19 Now, essentially the work on Diablo Canyon 20 started early in October, October 9, to be exact, when 21 ve were called by MRC to come to Bethesda and partake, 22 or listen in to the discussions regarding the so-called 23 ' diagram error, which were going.to be discussed.by PGEE 24 and their contractors. After that we part'icipated with - 25 NRC at the audit which wa s~ heli n San Francisco. At i ALDER $oN REPORTING CCMPANY. INC. I 400 VIRGINtA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
'S ~~P - . . _ . T- ~ 'T i._ ___R Z.'_~~'T_E_ Z T..~ Z T [; Z ~ T.T_7 -_. . _ _ _ . _ . . _ .i.
_ _ =_- .= . _. 8 I l
! 12 1 this audit, by the way, we did get information from PGCE 2 regarding the 2D model, and those of you who have had 3 the report know that we got an input card for the new 4 N' ark 7.5 ground acceleration time history, a descript!.on 5 of the input data that was used for the two-dimensional 6 Blume model. We also got a two-page summary of the 7 two-dimensional analysis, and we got a copy of the __
8 URS/Blume report which was entitled Diablo Canyon 9 Nuclear Power Plant Unit 1 Containment Structures to Dynamic Seismic Analysis for 7.5 Magnitude Hoscri 11 Earthquake, and it is dated Hay, 1979. We also got a 12 fairly good background about the structure itself, the 13 people that were there. 14 After we returned from this trip, 've gave a 15 report to NBC'on our findings, and we were requested to 16 independently develop a vertical floor response spectra 17 for Unit 1, and'we started with what we call Model A. 18 Model A is where we consider shear joints for all beam 19 and column connections, and after that we were asked to 20 do a Model B where only connections for the first and l 21 second floors wer*e considered, the other two floors l i 22 being the same as Model A. , 23 'In addition to that, we were also requested to i 24 independently analyze two piping systems wi th PGCE - ! 25 designation No. 4A-26 and 6-12, -{n addition-to this, ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) $54 2345
~ ~ ~ ~ ~ ~ '"~ "" ~ "
_ 7 TT- ~ __ ~T T T_Z ~' T
~
'6 - i I i
}
13 1 the Staff asked us to check our calculations since we 2 were doing these calculations with the BNL computer
; 3 method, or a. method which SNL had revised , to check our 4 calculations of the floor spectra, using a co puter code 5 which is available to the genera public, and see if we 8 would get the same results with that computer code as 7 with BNL. -
8 As time went on, the Staff asked us to try a 9 third model, and the difference between this model and 10 the others are that in this model the beam-to-column 11 connections for the first, , second and third floor are 12 considered rigid. The fourth floor remains as in the 13 first model. f 14 Af ter that we were told to carry o'ut a i 15 confirmatory-computer run for the original 2D model, and i ' 18 once we were completed with that, we were told to carry j 17 out for the same piping systems which we had before, 18 this time using multiple input analysis, that is, for
, 19 the piping system 4A-26 and 6-11, and we wo uld be using i 20 the spectra input from the 3D analysis, and we would 21 evaluate these systems now in accordance with ASNE class 22 2 classifications. .
23 Again , f or the 2D model, we were told -to 24 verify the floor spectra'results with spectra generated
~
25 from a different code. In thfs case.it was the ALDER $oN REPORTING COMPANY,INC, l 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345 l
. e 14 1 McDonnell-Douglas codes. In both cases it was STRUDL 2 McDonnell-Dougins.
3 After that, we were told to extend the 2D 4 study by varying the boundary conditions, taking 5 different boundary conditions to see what effect that 6 had on the results. And.then finally, .we received the. 7 exact input-output conditions for the URS/Blume 2D run, _ 8 and we used that and we were told to verify that using 9 an identical model and identical input data, comparing 10 raw floor spectra data. 11 And the final item on the task outline is the i 12 report which I think you people have. 13 Now, in terms of the drawings that were used 14 for generating 'the structure, in the report' they are , 15 listed on pages 3, 4, 5 and 7. Those are the drawings l l 16 that we used for generating our 3D model for the piping 17 systems. The drawings and all the other items that we l 18 used are shown on page 4. We had two packets sent to 19 us, and the packets, whatever they contained, are Listed 20 on page 4, from Item A to Item L. We had two different 21 ones, one for 6-11, and one for 44-26. 22 I would like to summarize some of the results 23 tha t we had. One word I want to say, that from,these 24 drawings, Model B is closest to what the structure - 25 really is, and when we did-the-piping system analysis we l ALDERSON REPORTING COMPANY,INC, [ 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345 I
~-- - -
,y.-w-~,---- y
~ ~ - - - - -
15 1 used the spectra f rom Model B that we generated. 2 Now, with respect to the 3D model studies, the 3 floor response spectra generated by these models did not il 4 agree with those which we saw or which v.ere given in the 5 May '79 Blume report. We.noted frequency shifts and l 6 difference in peak spectra acceleration magnitudes. 7 Now, this was the case for Models A, B, and C. - . 8 The URS/Blume results for the top floor were 9 consistently conservative in that they always exceeded 10 the acceleration magnitudes predicted by BNL 1. This is 11 now the case, however, for the other three floors, that 12 is, floors 1, 2, and 3. For some frequencies, the Blume 13 results for these floors were conservative. For others-l 14 they were not, and there was not any observable trend 15 that we saw. 16 As f ar as the piping analysis is conc ~erned, a 17 I said previously, the confirmatory evaluations were 18 performed for the two-pipe PGCE piping numbers 6-11 and 19 4A-26. 20 Now, we carried out these evaluations using 21 envelope response spectrum methods, independent support 22 motion response spectrum methods, and both PGCE and BN1 l 23 developed spectra. . 24 Now, the PGEE supplied spectra in the package - 25 we had also had a little package entitled New Hoscri 5 l ALDERSON REPORTING COMPANY. INC. 400 VIRGINIA AVE., S.W WASHINGTON. D.C. 20024 (202) 554 2345
'~~_ ___._:_ _ - ~ - -
'-'""7".- , _ "' 7";^ * '
_~ Z~ ~ '~ ' ~ ~ Z _ - - -
O M - 16 1 Mass Spectra. A check showed that these spectra sete 2 different than those presented in the Blume report.s t '. 3 Now, our models developed from the drawings 4 which were sent to us, which were marked as build 5 drawings, they were found to differ somewhat from the 6 PGEE models. The differences are due to the fact that 7 PGCE used design dimensions, and ther differ from the - 8 as-built dimensions. In addition, there were errors 9 aade in the modeling of the pipe bends. 10 Furthermore, for problem 4A-26, an overlap 11 procedure was used in modeling the problem. However, we 12 believe that the extent of the overlap used in the 13 problem is adequate and it meets the intent of 14 NUR EG -C R- 19 8 0, which discusses the problems'of overlap 15 modeling. 16 Results were as follows. The BNL predictions 17 of system frequencies differ from the PGCE estimate. 18 The BN1 support force values obtained using our models 19 and even the PGCE supplied spectra do not match, and the 20 differences we believe are due to the differences in 21 modeling. 22 The support forces calculated using the BNL 23 piping models and the 3D model B envelope, or -
- 24 independent spectra, exceed the PGCE calcuisted values, -
25 and the major cause for this ts that model B' spectra ALDERSON REPORTING CCMPANY,INC. 400 VIRGINIA AVE .S.W., WASHINGTON. 0.C. 20024 (202) 554 2345 _ . _ =
_ . . , _ _ -, s _ m- -_ _ w
. 17 1 exceed the spectra used by PGCE.
2 The ASNE Class 2 evaluations which were then 3 performed using the uniform response spectrum method 4 indicated noncompliance with Equation 9 for three points 5 in problem 6-11. i
. 6 Now, for problem 4A-26, this problem me t all' 7 criteria. We didn't have any problem with that. ,
z._ . 8 However, when we went to independent support response 9 spectrum methods, we found a reduction in the stress to level in problem 6-11, but an increase in the stress 11 level for problem 4A-26. For this procedure, 6-11 l 12 showed overstressing now only at one point. However, 13 4A-26 had overstressing at three points. 14 That' sort of summarizes the piping analysis. 15 Let me go to the final item, and that is the 16 results of the 2D model studies. 17 Now, we had uncertainties at the beginning l 18 because our notes that we brought home from the meeting 19 and all the reading that we did of the Blume report, we 20 had uncertainties with respect to data, various boundary 21 conditions and what was really in that model. We 22 weren't 100 pe'rcen t sure. We carried out various 23 parametric . studies for this model. None of these, 24 however, correlated well with the re sults khich were in -
~
25 the report. - - - - ALDER $oN REPORTING COMPANY,INC. 400 VIRGINIA AVE S.W., WASHINGTON. D.C. 20024 (202) 554 2345
- - . ~ . . . - - , , - - - . _ . , . _ . . . - - . - . _ . - . _ .
_ ,. . nnn - . ., . - -
&&mdi"' -
me - -- -
. _ . . _ . e_1
~
18 2 1 We requested that NRC obtain a listing of the 2 computer input-output-for the runs used to generate this 3 spectrum, and this information was relayed to us on 3 4 Epril 24. ,, 5 Once we had this information, we carried out a 6 confirmatory computer run with identical input data to 7 that used by Blume, and we,had the raw floor spectra ,._ 4 8 similar to that sent to us in digitized format, of S course. 10 Now, the broadened spectra sssociated with 11 these structural frequencies, which are in the May '79-12 report, correspond with the raw spectra values sent to 4 13 us in '82. In the lower spectra frequency range, we 14 noted that the smoothened spectra were obtained by use 15 of raw spectra values. 16 Now, another thing we found by going'over our i 17 drawings, we f ound the diff erence in the ma ss. This is l 18 the total mass per floor in these units, 19 2 i Kip-sec / foot, for each one of the floors in the Blume to model. Cn the first floor, 1.52, we found them to be as 21 a total to be 3.08. On the second floor it was 2.68. , 22
- WE found 4.S4. On the third floor it was 6.~71. We had 23 .
11.43. The isst floor where there is a concrete slab, 24 Blume had 46.84, and we had 67.75. These were the j 25 --- differencas in the masses. ALDERSoN REPORTING COMPANY,INC, i 400 VIRGINIA AVE S.W., WASHINGTON. 0.C. 20024 (202) 554 2345
- _ .' . _ l. _ _ _ _ _ . . _ . . _
_ . . , ._, _ _ _ _ _ . . . _ . . _ - . . _- . ~ .
s 4 * [ 19 1 I
.1l Tn addition to that, if you look over the 21 joining. drawings, the steel fabricating drawings show . i.i'.
3 that the member connections used in the 2D model did not 4 represent the actual field conditions in those joints, 5 and we know ' from our parametric studies tha t were f 6 carried out.that the floor spectra results are altered
! 7 by 'he t connectivity and all that. -
8 Again , as with the 3D results -- the 3D 9 results, by the way, we checked with a McDonnell-Douglas 10 computer run. They came out very well. I can show you 11 that. 12 Our results that we got, and the ones that we 13 got from the McDonnell-Douglas computer run , the STRUDL, 14 were almost on ' top of each other, and the sa'me goes for 15 the 2D results. A good match was obtained from that. 16 And'that is my summary of what we did'. 17 We can, if you want, go over individual . 18 reports. We can go over, if you want, the 3D model, the 19 2D model, piping model, if you want, whichever way you 20 would like. i 21 MR. KNIGHT I would proposed that at this 22 point, Dr. Cooper, the floor is yours, and if you would 23 set a format for the rest of the day. . 24 MR. COOPER I'will. Thank you,'Mr. Knight. ( ! 25 I have about half errough copies here of a l l ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (2021 554-2345 t . . _ . _- _
cc A 20 1 suggested format for doing this. 2 I brought twice as many as should be required 3 for this meeting and it is about half as many as we. 4 n'ee d . 5 Our objective in this meeting is to exchange 6 inf orma tion concerning the Brookhaven report which 7 pertains to our review which is being conducted in _ 8- response to the July 1 letter from Mr. Denton to 9 aysel'f. 'This is the kind of technical interrhange that to should~normally involve a dozen people sitting around a 11 table with a lot of papers. Because that format is not 12 permitted to us, we are strongly recommending a very 13 detailed procedure be followed. On each portion of the 14 .Brookhaven report which we wish to discuss,'we will 15 follow th'is~ procedures no implications regarding,the 16 other aspects of the Brookhaven report are to be drawn 17 from our failure to call upon that being discussed. The 18 only thing you can conclude is that we don't wish to i 19 disc'uss it. We think we understand it. We don't 20 necessarily agree or disagree. We just see no need to 21 discuss it for our purposes. ! 22 Dur suggestion is that we go through the 23 report, that in each case I identify a specific content 24 under discussion and quickly summarize our' concern, that - 4 25 ve give the Brookhaven folk ~s dhe-opportunity to l 1 i . Al.DERSON REPORTING COMPANY,INC. 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
. - . . . _ . , - _ , , , _ . ~ - _ , . , , _ . , . , _ . , _ _ _ . ._ , _ _ . . . . . .. . ___ _ - _ . _ . . _ _ _ _ _ . _ .
L L. >+ < 1; i
, l 21 l .F~
l 1 supplement the written report in any manner which they , l 2- vish, and we invite the NRC to make any sup ple men ta ry 3 comments they may wish; that the IDVP participants be 4 permitted to ask questions to SNL and BNL be invited to 5 respond to these questions. And, of course, that is l 6 going to involve some back and forth, I'm sure. I
! 7 Next, that the Diablo Canyon Project Team be -
4 8 invited to make any observation which they believe will 9 assist the Independent Design Verification Program in, to conducting our review; finally, that we have a circle of 11 closing comments as desired by the individuals from l 12 Brookhaven, the Staff, the project and ourselves. t 13 I suggest that at the end of going through 14 this one time, 'that we each have our individual . 15 caucuses, give us a chance to exchange thoughts on what 16 we have learned -- and we do expect to learn a lot, and 17 we would anticipate that there will be need for 18 clarification of perhaps the other portions of the 19 report that we find we should have asked to discuss. We 20 vill at some point be requesting some selected computer 21 printouts. We understand a great deal of the work which 22 Brookhaven has done on this, and obviously 1t would be , 23 ridiculous for us to say give us on~e copy of . 24 everything. We would hope that as part of'the benefit - 25 we gain from this discussi6n ,-thet we could single out i i ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTCN, D.C. 20024 (202) 554 2345
- , i c _ ,_ ~ . _ - . . T.ZT J,
, _ - - - _ - - _ . . ' "~7.' , ~.' 7 '"T
. 4 .
22 1 specific itams which will be helpful to us in the 2 future. 3 I agree completely with Mr. Knight's 4 description of the ground rules. I have summarired it 5 in too few words compared to the way he stated it. This 6 is a what meeting, not a why meeting. We consider that 7 this meeting was requested by us to obtain clarification - 8 df the report, and as far as the independent program 9 participants that are here today, that is our sole 10 purpose in being here. Other things that may be 11 presented or done at this meeting, which is properly an
. 12 NRC meeting, that is their prerogative, obviously.
13 And finally, just to make it abundantly clear, 14 no conclusions are to be stated or implied.' And I 15 expect to exercise the prerogative that if I feel - t 18 somebody is making a conclusionary type of stat ~ement, I I 17 will interrupt them, and I hope that we don't have~to do 18 that more than once. 19 3r. Knight, is this acceptable to you folks?
, 20 MR. KNIGHT: I think it it is quite fine, 21 yes.
22 MB. C00PEH Thank you. , l 23 Gentlemen, can we have lights. Are there 24 brighter lights in in the room because at 'least we are - 25 not going to have anything to pWect. I suggest that ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C.20024 (202) 554 2345 _ . , , _ , ,.. y .
' ~
_ - - . . :. . . _ _ _ _ - --= - : _ _ - -=
. +
23 1 those of'you who have a report available turn to page 2 15. 3 We did have some questions concerning the 4 earlier chapters in here, partic,ularly the assignment to 5 Brookhaven, but I think the presentation we just 6 received is more than adequate to answer all questions 7 along that line, and I suggest that it is not necessary - 8 to go through that any further. ! 9 Cn page 15, near the end of the second 10 paragraph, there is a discussion of the concrete and 11 steel structures on the so-called fourth floor, the 140 12 foot elevation, and it is noted that the concrete and
~
13 ~ steel beams on this floor act independently and not as a 1 14 composite, tha't'the concrete slab of his is'a one-way
,15 slab spanning between the crane wall and the tangential l 16 girders, and finally, that a schematic drawing of the 17 configuration is given on Figure 2.1.
18 There is, of course, further disrussion of the 19 modeling of this structure in what follows, but we don't l 20 f ully understand the manner in which the concrete / steel 21 structure was modeled on that fourth floor, and we would l 22 like to hear more about that. 23 3R. REICH: Okay. This is the reason why I 24 vant to go over the modeling portion. I think it would - l l 25 answer most of your questid~ns c - -- i ALDERSoN REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W WASHINGTON. 0.C. 20024 (202) 554 2345 _ ~ , , . _ . _ - . . . - -
~~^ '
41< _y
~
. ', _ _ 1 i
o o -~ 4 24 1 MR. COOPER: I would strongly suggest we keep 2 the discussion to those items we have questions about, 3 sir. I think it will speed us up a lot. 4 MR. REICHa It is not going to be as easy as 5 you think, but we'll try. 6 MR'. C00PERa We didn't think it would be - 7 easy. - 8 Chuck, do you want to answer that question? 9 MR. MILLEHs My name is Charles Miller. 10 If you look at page 24 of the same report, it 11 is a computer-drawn plan of the fourth floor. The 12 triangular elements that show up are triangular elements 13 used to model the floor slabs. In most regions of the 14 top floor there are tangential girders that'run around 15 connecting to columns that run around the outside of the 16 radius. For example, on the top of the -- if you look 17 at the picture from the side, the top of the elements I 18 391 and 397 is a steel girder. The concrete slab rests 19 on cranes into that. There are steel girders also in j 20 areas that look like open areas, for example, toward the 21 bottom of that figure, 331 and 327, there is a steel 22 girder where t'here is no concrete slab in th,at , 23 particular area. . l 24 So the places where there are, l'ike at the - 25 top, both concrete and steel,-is a member, a steel ALDERSoN REPORTING COMPANY, INC. 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345 x- ._, -_
. ,,_ ;. ,.y
, .. = ~ LM : - a i
- i 25 1 member that runs from 391 to 397 on the top of the page, 2 and there also happens to be a concrete slab that runs 3 between the same areas. So if you will, it is really 4 modeled as if there were two elements running between 5 the same set of nodes. So it is like one beam sitting 6 on top of another beam. The stiffness is added, but i
7 there is no composite action. _ 8 3R. COOPER: The drawing. that really 9 represents the steel structure, well, represents part of 10 the steel structure, and at least in some of the cases, 11 the lines also represent the boundary of -- 12 MR. MILLERa If you had been drawing, there 13 would have been two lines showing up because there were 14 two sets of el'enents filling the same space.' i 15 MR. COOPER: What about the interaction 16 between those two members, the steel and c.oncrete 17 members? 18 MR. MILLER: There is no connection except the 19 node. 20 MR. COOPER: There is connection at each of 21
- the nodes.
22 HR. MILLER: At each of the nodes there is , 23 connection. So the displacements are identical.at the 24 nodes, but between the nodes, other than the attempt to - 25 find the model, req uires th~a t-the- displacement be ALDER $oN REPORTING COMPANY. INC. 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554 2345
4 9- 'l_' . I1 _ 9 2 26 1 compatible. There is no other coupled with the 2 displacement. 3 ER. COOPEHs And the connection between, if
. 4 you wish, between the concrete and the steel beam can 5 carry either tension or compression.
6 HR. MILLERS That's right. It is a strictly 7 linear model. So if they wanted to open it up, it would _ 8 assume that the slab would pull the steel, and the steel 9 would -- 10 NR . COOPER : Is it just a vertical connection, 11 or are the other displacement nodes coupled? 12 ER. MILLER The degrees of freedom that were 13 retained in the model, the vertical displacement and two 14 horizontal rotations. So 'for all.those rotations they 15 were coupled. 18 On the top floor, actually, the . 17 bean-to-girder, the girder-to-column connections on the 18 top floor are always a pin . connection. So in fact, the 19 pins have nothing to do with the girder response, with 20 the steel beam response because the steel beams at the 21 top floor in all three models tha t we used were taken to 22 be a shear connection. So the horizontal rotation had 23 nothing to do with the stiffness of the beams. 24 MR. WRAYs Son.Wray. . 25 Would you summarize a,qain which of the ALDER $oN REPORTING CCMPANY,INC, 400 VIRGINIA AVE., S.W., W ASHINGToN, O C. 20024 (202) 554 2345
.. E '
s - 27 - 1 elements connecting nodes of the steel elements? 2 MR. MILLER: There are steel girders that run 3 around the whole outside radius, in Figure 2.6. If I am 4 correct, the only other places where there are steel 5 beams are places where there are openings, like starting 6 from the top of the page, going clockwise, like in the 7 region from nodes 412 to 416. Those sets are steel - 8 beams. In the regions from 385 all the way around to 9 337, that's an opening and those are all steel beams. - 10 In the region f rom 353 around to the next one over at 11 358, I guess, those are all steel beams, and also in the 12 region 368 to 372, those are steel beams. 13 So aside from the openings and the tangential 14 girders that run around the outside radius, 'there are no 15 other steel beams connecting to the slabs. 16 MR. WRAY: Thank you. 17 MR. COOPER: The next item is at the end of i 18 the first paragraph of Section 2.1. It is noted that l
. 19 certain bracing members did not ha ve to be included-in 20 the model because of -- in the sense that the motion was 21 constrained by the modeling techniques, so the members l
22 which ordinarily constrain that motion are n,ot j , 23 required. . 24 Our only question about that is 'whether or not - l 25 the mass of those omitted sembers- was included in the ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
+ - - _ , . . . . . . - ,
mat -
-- . . _; _ r_ n. _
.m _ s _.
28 1 model. 2 MR. PHILIPPACOPOULOSa First o f all, we don 't 3 say that we neglect these members, because we took care 4 of producing these members in the model in some other 5 var. We just state here that these models, these 6 particular members, they are in there to restrain 7 motion, and we analyzed the structure very carefully. _ 8 We concluded that these members were not excited when at 9 rest, so therefore they would not be included in the 10 model. 11 Most of these members are very light members, 12 and offhand, I recall a few cases that some of the 13 masses were. included, but I don't know if generally we 14 have, as a rule', included the mass in these ' members. 15 That is the answer. 16 HR. COOPERS Thank you. 17 I presume this is something that if it is of 18 later significan.ce to us, by looking at the computer 19 information we can determine that. 20 HR. PHILIPPACOPOULCS: Yes. 21 MR. COOPER: At the bottom of that page, it is 22 noted that nod'es are included a t the 17tersection of all , 23 members with additional nodes added along the span of 24 the beams and girders so that higher beam 'm odes may be i 25 obtained , and the question 7 ve -ere interested in getting ALDERSON REPCRTING CCMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. 0.C. 20024 (202) 554-2345
. . .ma . -- ._w- - - - - - - - - - - - - - - . - - - - - - - -- --- - - - - - - - - - -
~
, _a _: : -- 2: a l a__ -- '
- 29 1 a feel for where these additionaI nodes may'b'e, and -
2 perhaps the simplest way to ask ' t.'h e. . question. would be if 3 on the figures on pages 21 through'24, if.sil the floor 4 nodes are identified on those, other additional nodes. 5 HR. MILLER: All the nodes are there that we 6 used. Those are the nodes in the model, no additional 7 nodes. - 8 MR. PHILIPPACOPOULOS: I would like to nov 9 make an additional statement about it. If you look at to page 23 of the report, there is a figure No. 2.5 that 11 shova that the final element reads for the third floor, 12 the nodes that are between the radial girders are the 13 additional nodes we are talking about in this particular For example, nodes 2'03, 204, 14 section of.the report. 15 205, 210, 211, 212, and so forth, these are the nodes we 16 mention in this pa rticula r report. 17 HR . COOP ER : Good. So all the nodes are 18 here. 19 The next item we had was on the top of page 17 20 where reference is made to a ENL version of the SAPV 21 computer code. We recognize there is further discussion 22 of this modifying code on page 43 but just ask the 23 question, are there any other signficant modifications 24 other than those that are outlined on page'43. - 25 3R. PHILIPPACOP0dLOsr 7 re..you asking me if ALDERSoN REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
v; -
- ti s - -m- -t % _ . .s - - -
e . i / 1
, < ..s -
, g. , :/ ~
s.. 30 . I
~
1 there are any modifications with respect to this? If
~ t 2 you are asking if there are further modifications as 3 partoftheviewsoftheprograntogeneratofloch
~
4 response spectra, no, the answer is no. .,
; 5 LHR . C00P ER a Thank'you.
6 BR. PHILIPP'ACOPOULOS: Okay. 7 ER. C00 PERT Turning to page'18',-the first- .- 8 paragraph on that pace, just below the.' middle of the sentence, and then again in a later sentence, there is 9 10 discussion here of the use of the tributary areas in , , 11 determining the manner in which the nodal masses are to 12 be assigned, and we would appreciste some further 13 discussion of that point and Jhow it'was done..in a little 14 . acre detail. - ' 15 HR-. PHILIPPACOPOULOS: Okay. As I understand, 16 rou would like to hear some more information how the ' 17 total mass at this nodal point on this final Element 3 18 was determined . 19 First of all, we realize that we have two 20 kinds of contributions. One is all the support items by 21 the floor, on the floor, and the second is tas. 22 structure, the contribution of the structure,itself, t5e 23 weight of the structure. - 24 Now, this particIlar area here l's decliu vit- - 25 the first item, is explainfng-trov, from the equipeont ALDER $CN REPCRTING COMPANY,INC, 4rx1 VlMGINIA AVE S.W., WASHINGTON. D.C. 20024 (202) 554 2345 ___._%__ ____."-_ C: *- 'N-__, -
" ' ~ - ' ~
_- r- an- ,
* .* :l .
31 1 and other weights supported on the floor we got the 2 nodal weights.' Acain, there are two items you have to ). 3 look to to answer this question. The first item is that 4 de have two typas of contributions. The first type is 5 due to distributed loads. The second type is due to the 6 concentrated loads. Both of them were obtained by first 7 o f all, the listributed weights were obtained by the _
- 8 tributary areas, which are not necessarily the same as
; 9 the tributary arets that URS/Blume used in his model, 10 and we found what are the veights for.the uniform loads 11 hnt contributed particular node, and then multiplied 12 this particular times the model of the load, and that
,/ 13 represented the contribution of the uniform load to this
/
14 particolar node. - 15 With regard to the concentrated loads, some of 16 them there were applied on the nodes themselves. 17 Othgrs, they were between nodes. Therefore we have to 18 redistribute them to the nodes. In this process we 19 ust.d -- we treat, if the load was falling between two 20 nodes on a beam, the reactions of this beam were the
'c 21 contributions to the node.
22 I don't know if I've explained it.
;3 NR. DENISON: Edwward Denison. .
'24 What va's the cutoff f or concentrated nodes? .
25 Md. PHILIPPACOPOBLOSt _Je had a se.t in mind. j o ALDERSoN REPORTINo COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON D.C. 20024 (202) 554 2345
.+----7.-.--3
w .,
-y: _ - -
,r 4
- O' -
.O (.
, i s
? t,
-- s y -
__- _ J f 54 1 questions? 2 MR. WRAY: Yes. Ron Wray. On the topic of 3 the hangers, are the hangers modeled into the 3D model? 4 MR. PHILIPPACOPOULOSa,, Yes, they were included 5 there. The ones in our drawings. The ones that were e indicated on our structural drawings were included in 7 the model. _ 8 MR. DENISON: Do you think there are other 9 hangers? 10 MR. REICH: We can't tell you what there is. 11 We can only tell you what we know. 12 MR. WRAY: Are those obvious when you look at 13 your 3D models? 14 3R. MILLER: The only var you can see those are 15 in isometric. On the original list of elements, I think 16 the order -- element in fact has the hangers at the end, 17 the order of elements occuring of the beam elements four to by four first, then on the vertical columns on the 19 outside. 20 MR. PHILIPPACOPOULOS4 And then isola tion. It 21 was our concern to see the whole geometry of the 22 structure snd see how it was represented. Therefore, we 23 break down these first four floors. The four pictures 24 you get are the four floors. In addition,'we isolate 25 all these in separate and M'a v e *. h e m fixed. In this ALDER $oN REPoRfiNo COMPANY,INC, 400 VIRGINIA AVE., S.W., W ASH.NQToN, D.C. 20024 (202) 554 2348 [ --- _ . . _ - . , - . . _ _ _ . . _ . _ _ _ - _ _ . . . _ _ . _ _ ,
4 s . 55 i
.1" particular page of.the report, we cannot see, so under 2- this we esnnot see. t c, 3 - '. -
i
' 3 '-
NR. KNIGHTS We have reached a point of -- do
~
y'ou have one other question? 5 ER. HARRISs Steve Harris. I have a question l e regarding the hanger loads that are included in your 7 aodel. The loads you put into your model came off of _ i 8 PGCE drawings. Were they represented as design loads or 9 vere they broken down'as load components for seismic, 10 and if so, which components of those loads did you 11 include in your masses? 12 ER. PHILIPPACOPOULOS: The ones that were i 13 breaking down the loads, the distributed loads and
~ ! 14 uniform loads. Let me find the page of the* report.
15 NR. MILLERa Seventeen. l 16 ER. PHILIPPACOPOULOS: Seventeen? On Page 17, t 17 ve demonstrate what are the particular drawings. In . 18 addition, on Page 18, there is a description of loads,
; 19 and among the first ca tegory, contribution due to large i
i 20 poor mechanical pipes and support the mechanical 21 equipment, architectural, et ceters, et cetera. The I 22 latter category includes mechanical small -- et cetera, 23 et cetera, so these are the weights. , ! 24 MR. HARRIS: You are saying the weight loads. - 25 (General conversation.4 1 i ALDER 8oN REPORTING COMPANY,INC. 400 VIRGINIA AVE., S.W., WASHINGTON O.C. 20024 (202) 654 2346 l
. s
' 56 1 MR. COOPER: We don 't have any more 2 questions. If PGCE has any observations. Bob?
'3 MR. WHITEa Yes. On Table 3.2 there is the ,
4 f'requencies of the model. 5 NR. PHILIPPACOPOULOS: I couldn't hear you. 6 HR. WHITE: Table 3.2 are 'those frequencies 7 for Model B. , 8 HR. PHILIPPACOPOULOS: No. These frequencies 9 are for'Model A. I -don 't recall if we mentioned this in to our report. 11 MR. COOPER: Is there a list available for 12 Model'B? 13 , 5R. PHILIPPACOPOULOS: In fact, today I have 14 -- over there, a big presentation after Dr.' Reich did 15 his, and I have a slide here in which I demonstr. ate a 16 few of the modes for the B model. Probably I will have 17 it changed later on. ! 18 MR. REICHa I think that would clear up some 19 of.the other questions. 20 MR. C00PER4 I would suggest we take a break 21 and then look at it. 22 MR. KNIGHTa Fine. We vill do it. Could I 23 charge everyone to be back at 10:257 . 24 (Wheraupon, s brief recess was tiken.) 25 HR. KNIGHT: White ve- tre back acain together, ALDERSoN REPCRTING COMPANY,INC. 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
..n~.--...,_.-,,-~. .---n
_-.. __----..~.. _ _;_ _ -. 3 - . _
i!
.j~ '
j 57 1 once again I ask that we make an extra effort, since 2 this meeting is being reported, to identify ourselves l l 3 and, perhaps more importantly, to avoid having 4 simultaneous utterances, if you will, froa different
~
5 parties. It makes it extremely difficult for the 6 reporter. I know it is a rather stilted format than our
; 7 usual technical discussions, but please try to _
8 accommodate it. 9 And I might add, probably from my observation, to sitting up here, it would not hurt all of the speakers 11 to turn up the volume a notch or two. I think when we 12 speak a little louder most of us tend to enunciate l 13 better. That would help also. l
! 14 Dr.' Cooper? -
15 HR. C00PEB: Thank you, Jim. 16 I want to clarify something just from.the 17 comment as we vere breaking, and the clarification is, 18 ve now understand tha t Figures 3.1 and 3.2 are no t Model 19 B. Is that correct? 20 MR. PHILIPPACOPOULOS: Figures 3.1 and 3.2 i [ 21 represent modals taken from Model B. 22 MR. COOPER: They are Model B. We understand 23 You were going to show us something about frequencies 24 and mode shapes and Model B, and that is wh y I concluded 25 vrongly that those were not. _L Auggest that would be of i i ALDERSON REPORTING COMPANY,INC. I 400 VtROINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345 (- .-.- - , - _ - -. -...- - , .-.- - -.- -. . . . . - - . - . - . . .
. ~ . , _.
- " ^
W _ #84 4 k . m D- -t M l' 58
'i 1 great benefit,,to see that.
4 2 NR. REICH 4 h a ce coito to show you a > 3 different table. . i ,, . 4 MR. KNIGHT If I could say something, th a t is 5 not a table which nececsarily replaces something that is 6 in the report, but an additional tabh. ' 7 NB. PHILIPPACOPOULOS: In this slide here, wo _ 8 can see the firar 15 frequencies, from Model S. '?s you s 9 can see, the first few frequencies are higher than the , 10 Model A, and the figures number 3.1 itnd*3.2 you just . 11 mentioned, Dr. Cooper, they represent mode number 4"and 3' 12 mode number 8 -- 9, I am sorry. Excuse me. I have to 13 sake a correction on that. I think it is mode number 8,
~
14 but mode numbec 8 and 9, they have pretty close <
~
-15 frequencies. So in these frequencies you'see the 16 frequencies around it to the first decimal peint, and 17 indeed.in those two figures you can see the unbers.
18 Figure 3.1 is modo number 4, on the top lef t, upper left 19 of the figure, and figure number J.2 is -- ve,can see 20 mode number 8 on the lef t part fiture. So, these are 21 the corresponding frequencies lis*.ed in this slide.-
- 22 Thank you.
23 MR. C00 PEPS Do you.hac.*en to have anything 24 about how many modes it took you to get up'to 13 hort: ) 25 in this case? - - - i ALDERSON MEPCRflNG COMPANY. AC. I 400 VIACINIA AVE S.W WASHINGTote. O.C. 2C324 (202) 314 2346
, ,,, , .,.n.-,.-n7.-..----..--- .- 3. . . y.~ - 3 y .
- .s
--w--- :y 2- - --
_ _ - n f :lk
- .r* ,
r -
}' S .// ,
59
, 't MR. PHILIPPACOPOULOSk I think approximately
' ; 140 sodes, a pproxima tely.
- 3 -
MR. COOPERS Jim, I assume that a copy of that 4 will ~ by attached to the transcript. 5 ,~MR. KNIGHTS That's correct. We will make 8 anything that is shown here today.
, 7 MB. REICHs We have just been told to make __
w 8 this available to you, and we will try. 9 MR. KUOs I would like to clarify one figure.- 10 You just showed the slide there that has the number 11 12.4, I believe, for model number 8, or for mode number ! , 12 8. The figure on Page 75, modal number 8, you have 13 12.'987. Are they the same one? 14 MR. PHILIPPACOPOULOS: Yes. I noticed here
, '15 the modes that are just demonstrated, not -- mode number l .
to 8 has frequency of 12.9.
. 17 MR. REICHa They rounded the figures.
i. l 18 MR. CLOUDS When you include that in the 19 tcanscript, can you label it as model 37 It is not 20 labeled, and I am having a problem with labels in 21 general, but also what mode was the mode for ficure 22 3.17 .
./ 23 MR. REICH Bob, on the upper lefthartd ,
24 corner -- - 25 MR. CLOUDS Yes , I've -got it. ALDERSC% AEPoRTING CCMP ANY. INC. 400 VIRGINIA AVE S.W., WASHINGTON. 0.C. 20024 (202) $$4 2343
.....i . . .. _ . . . . _ . . . - .
7--_....,_..7.,..7._,_, 7 ,_ _
a ._ - _ . , = , 4 e y, .. r i
> 60
\"
1 MR . COOPER : We got to this point when we were
. .. D ,
t 2 sort of going around the circle to see if anybody had 3 other comments, observations, or what have you on' this 4 matter. PGCE had asked a question which led to this s' 5 presentation. Do you folks have other questions? 6 MR. WHITE: On Page 27, at the bottom of the 7 socond paragraph there, it says the top floor not _ 8 excited before. Could you amplify a little bit on what 9 before meant, not excited before, but now? 10 MR. PHILIPPACOPOULOS: At this particular 11 sentence, before is the figure 3.1 model, okay? , 12 MR. WHITE: Referring also to figure 3.1, 13 could you explain what the alpha and the beta refer to 14 there in the upper lefthand corner? 15 MR. PHILIPPACOPOULOS: Yes. Alpha and beta 16- stand for two angles. 17 MR. CLOUD: Do you have the participation 18 factors,for these modes with you? I mean, do you have 19 them on an overhead or a tra nspa re n.cy ? 20 MR. PHILIPPACOPOULOS: No, I do not have it 21 with me. 22 MR. COOPER: Okay. I think we have been ,
, 23 around. Any other comments or observations or guestions 24 on that section? -
~
25 (No response.) -- l l, ' AU:ERSON REPORMG CCMPANY, INC, 400 VIRGINIA AVE., S.W WASHINGTCN. O.C. 20024 (202) 554 2345
61 1 MR. COOPER: I suggest then we turn to Page 2 4?. We are starting here with a description of the 3 floor response spectrum from the 3D models, and we would 4 just like to make note of the fact that although there 5 are curves here and so forth, compared directly to those 6 of the URS bloom spectra, I think we all recognize that 7 those curves which appear in the figures in solid dashed - 8 lines are of a particular vintage. It is still very 9 useful for us to spend some time on this section, in 10 order that we can understand the approaches that get 11 used in some of your thinking here, and our first 12 questions rela te to the so-called methods of comparison. 13 You use three methods of comparison, and we 14 vant to make sure that we understand just what you did. 15 HR. KNIGHT 4 Jim Knight. Bill, if I might, 16 just one point with regard to your remark that the heavy
'17 dashed lines are research vintage. I want to be certain 18 that we don't lose sight that one pressing question 19 before the regulatory staff is what import this migh t 20 . have for everything of that, vintage. In other words, 21' whereas there may be -- I guess one might say that some 22 of the thing s iden tified in the Brookhaven r,e po r t a re no 23 surprise when, you know, things here are just coming l
24 together here. This is when the overwhela'ing question l 25 is, to my way of thinking,~w e i t,--f i n e , but was there ALDERSCN REPORTING CCMPANY, INC, l 400 VIRGINIA AVE S.W., WASHINGTON. 0.C. 20024 (202) 554-2345 l .. m
- . _ . ? ad e ,
?
f 62 i 1 something wrono in the process that led to this 2 difference, and wha t does it mean for other things, t 3 other areas of the plans? 4 MR . COOPER : Jim, I appreciate that, and the
! 5 reason I said it the way I did, and perhaps I was not as
! 6 clear as I should have been, is that we are eventually 7 going to be looking at some of these comparison -
e matrices, and what I.am trying to say is that we are 9 looking at them for the purpose of understanding the 10 method, not for the purpose of agreeing or disagreeing 11 with the acceptable-nonacceptable categoriration here, 12 but rather trying to understand, and that is all I 13 really meant. 14 MR. KNIGHT: In that context, your point is
- 15 well taken.
16 , WR. REICH: I would suggest that we e'xplain to 17 you the three methods that we used for comparative 18 purposes. 19 MR. COOPER: We have questions about each'of 20 the three me thods, and I think that our questions are 21 sufficiently broad with one possible exception,..our 22 . questions are sufficiently broad that if you,were to go 23 through all three of those methods by whatever mechanism 24 you would like, that would help us all. - 25 MR. EEICH: I think 7e sho uld, and therefore l ALDERSCN REPCRTING CCMPANY, INC, 400 VIRGINIA AVE., S.W. WASHINGTON, D.C. 20024 (202) 554 2345 __ .-- _ - .-.----w -- -_ _ g. , ,
~
e . 63 1 let's put some slides up, just like we just did with the 2 frequencies. Mo'v about doing that? 3 MR. COOPERS Is that acceptable? 4 ER. REICH 4 We vill have Eike do that. 5 HR. PHILIPPACDPOULOS: We have some levels for 6 these three methods. The first one we say we compare 7 based on fan cooler locations. The second is altered __ 8 over floor sections, and the third enveloped over floor 9 sections. The second and third methods of comparison 10 are based on the so-called floor sections. This is for 11 comparison. We condensed the result. Let me go through 12 each of these. 13 This is a schematic presentation of both 14 units, and here we can see the orientation of the five 15 rings comprising the URS Bloom model. The first method 16 of comparison is based on one to one correspondence. 17 Let's see, now. The circuit numbers here indicate the i 18 fan cooler locstions. This doesn't mean that this 19 particular picture represents the top floor, but all of l 20 these can be applied to all four floors. So, these are l 21 the vertical planes through the locations of the fan 22 coolers. - 23 As you can see f rom here, the fan cooler
- 24. number 1 for Unit 1 is in that location of Unit 2, and -
25 corresponds to f rame numbet S- ef-the model that Bloom l l l ALDERSoN REPORTING COMPANY,INC. 400 VIRGINIA AVE., S.W WASHINGTON. o.C. 20024 (202) 554 2345
- - - . .- . .- .n - . . -
_.p,- - 7.- , __._ ,, m
- --n------
. . . . .= _ . w ew 0 4 6 ts 1 used in the report. Now, the first method of 2 comparison, we have in mind that picture, and we find 3 what are the nodes which are very close to the se 4 locations on our 3D.model. A three-dimensional model.
5 does not depend on the rotation of the five-f an 6 coolers. Having that in mind, we found for eac'h floor 7 what are the nodes nearest to these, fan cooler number -
- 8. 1. He pick up the floor response spectra, and they ra n ,
9 and we find that at this rarticular floor what is the 10 frame that was used to represent this particular 11 location by URS Bloom. 12 So here is frame number 5. So this is the var 13 we compare floor response spectra between 14 three-dimensional model and two-dimensional' aodel f rom 15 - URS Bloom. 16 MR. COOPER: Should we ask questions'about 17 method number 1 now, or should we allow you to continue 18 all through and then come back and ask questions? Your 19 choice. 20 3R. PHILIPPACOPOU105: I will leave it up to 21 you.
- 22 MR. COOPER: All right. With respect to what 23 rou have just said to us, do you happen to have- a flimsy 24 of the model of the fourth floor, such as' Figure 2.6 on -
~
25 Page 24? Do you happen to have 7 flimsy of that l l ALDERSON REPORTING COMPANY,INC. 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
-u- -
s,_ _ _
a.
- ~
__- :w . .:- -e_ n- - : , d .
\
I' . 65 1 available? - i, 2 HR. PHILIPPACOPOULOS: -(ould you. pl'ease, repea t 3 the page number? -
~
4 HR. COOPER: Page 24, F'iqure 2.6. 5 HR. PHILIPPACOPOULOS Yes. l l 6 MR. COOPER: Now, what I as wondering is, is
, 7 it possible to overlay the figure on Page 24 with the -
8 figure on the upper left of that? I recognize the radii 9- probably won 't match, but at least we will be able to 10 see about where the radial lines come, and I would like 11 to know in more details the nodes which you would 12 consider with one of those filter locations.
'! 13 HB. PHILIPPACOPOULOS: To answer your
'14 question, nac!5 of 'these locations from our model
~
. i 15 probably would have three or four nodes, so we have to 16 chose between three or four floor response spectra. 17 First of all, we should pick up the worst case floor 18 response spectra, pick up some, among the five of them, 19 the one that has the highest peak, or a big frequency 20 shift. This is not possible always. Therefore, the 21 comparison number 1, we don't consider it a 22 representative comparison method. However, it gives you 23 a fair start in this particular model to compare 3D 24 results from oup analyses versus a Bloom d'esign. That - 25 is why we can go f arther. ~ We- have comparison number 2 i l l l 6 ALDERSCN REPCRTING CCMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345 m .. _._..w. .
,.,..- .m .
_~ :__. - 1 "
.- K_ n . e -
66 i 1 and comparison number 3. 2 MR. COOPER: I-had. hoped to get more detail on 3 number 1. , 4 4 HR. PHILIPPACOPOLOUS: Let me say something.
; 5 more on that. I have from my~.vorking papers, I have set 6 what you are .iust asking now, the Bloom model on top of l
7 the BCL model, but in that report,'ve' don.'t solve this. - 8 MR. COOPER: But if you had.'that other flimsy 9 available, you could put them both on-the. screen and 10 orient them rotationvise well enough thEt I think that 11 ve could ask a specific question. 12- HR. PHILIPPACOPOULOS: Unfortunately, in that 13 particular case we don't have this kind of figure. 14- HR. REICH: We might have ' that. floor. 15 HR. PHILIPPACOPOULOS: Well, hold on a
~
16 second. 17 (Pause.) 4, 18 MR. PHILIPPACOPOULOS: The information, Dr. 19 Cooper, that you are asking for, we do have 20 inf orma tion . We don't have slides. I can.demcastrate { 21 from here. 22 ER. REICH: We can make one up. . 23 MR. PHILIP P ACOPOU LOS :- Yes. - 24 MR. COOPER: Let's make it easy.' You are - 25 holding the page with the i)er!ccTtions down.~ Just so i i ALCERSCN REPCRTING CCMPANY, INC, 400 VIRGINIA AVE, S.W., WASHINGTCN, D.C. 20024 (202) 554-2345
~- ~ ~ ~ ~ " ~ ~ ~
Ji , ' ,~ :' .L :-~;_ - .
__ 1e _
=
67 1 everybody can see where I am looking, I am going to look 2 at the edge opposite.the perforations, and let's say 3 ve've got a fan =coler some place in the middle of the 4 t'riangular section included by nodes 391, 394, and 392, 5 391, 392, and 394. Let's say we've got a fan cooler in 6 that triangular region, in the middle of the triangular 7 region., _ 8 MR. PHILIPPACOPOULOS: I follow you now. g MR. COOPER: Then how would you use the data 10 at nodes 391, 392, and 394 to obtain the result? 11 MR. PHILIPPACOPOULOS: All right. In this 12 particular case, we have three nodes. Node number 392 13 is a mid one. You would expect to have that 14 verification t'h'en. However, at this particular floor, 15 we have a concrete slab, and the input is not amplified 16 basically for this level of the structure at this
~
17 particular part of the structure. Therefore, in this 18 case, I would say you are not going to have to pick up 19 one out of three. The three of them are distinctly. 20 apart. So the floor response spectra specifically for f 21 node number 391, they are very much similar in shape and 22 in frequency conduct. 23 Therefore, here, if you pick up 3 92, it is 24 fine. If you pick up 391, it is again fine. - l MR. COOPER: What if _3S2 were distinctly 25 ALCERSCN REPCRTING COMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON. O.C. 20024 (202) 554 2345 ( _ _ _ _ _ ~_ ~;- _
u: , . _ a 68 1 different from 391 and 394? Let's say higher at all
. 2 frequencies. Would you just use 392, or would you 3 modify it somewhat?
4 HR. PHILIPPACOPOULOS: I wouldn't modify 5 anything. I would pick up one of these floor response 6 spectra from these nodes which is mostly excited. The 7 worst case. , 8 'HR. COOPER: So you pick up the worst case of 9 the th ree . 10 MR. PHILIPPACOPOULOS: I would pick up the it worst case out of the possible candidates a s comparison 12 spectra for this particular elevation. Okay? 13 MR. COOPER How about the other floors?
~
14 HR.'PHILIPPACOPOULOS: Would you please be 15 more specific? 16 MR. WRAY: Okay. You claim that this'was 37 close to like nodes 392 and 391. You would use the 18 worst of those two spectra or three spectra. Now, as 19 you proceed down through the structure, I don 't know 20 where that -- does that angular orientation take you -- f i l 21 MR. COOPER: We assume it takes you there. We 22 assume it takes you, if it is 2, 3, . 4, 5, 5,,7, 9, 9. 23 MR. WRAY: Suppose it takes you through the 24 internal node points. - ( . c 25 3 R.. PHILIPP ACO PCULOS r -Excuse me. In order to l ALCERSCN REPCRTING COMPANY, INC, l 400 VIRGINIA AVE, S.W WASHINGTON, D.C.20024 (202) 554 2345 m .. , . . .-
_ s - e - . r 69 1 help communication on this particular model, the floor 2 number and the node number fron the figures listed on 3 the report, that will help me better to give you the 4 answer to your question. 5 HR. WRAYs Okay. The fifth floor would be 6 figure 2.5. 7 HR. PHILIPPACOPOULOS: Okay. That is floor __ 8 number 3. Now, would you please tell me the node number ? g MR. WRAY: I can 't see when I put them up 10 where it falls, but I am not so sure. Okay. In this 11 case, Figure 2.5, that angular orienta tion takes you 12 very close to nodes'283, to 286. 13 HR. PHILIPPACOPOULOS: That is correct. 14 HR. WBAI Okay. Are there situations where 15 translating down through the thickness or down through 16 the different floors where that angular orientation 17 actually can take you through, say, the interior nodes 18 on the figure 2.5. 19 HR. PHILIPPACOPOULOS: In some cases, yes. 20 MR. WRAY5 In those cases, what would you use l l 21 it to be representa tive of a third floor spectra then, 22 interior nodes? . 23 MR. PHILIPPACOPOULOS:- Basically, in _this 24 particular method of comparison, at any fl'oor, at any 25 fan cooler location, th e re~ a re -541 in the case of one, i ALCERSON REPCRTlNG CCMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTCN, D.C. 20024 (202) 554-2345
... _,_ _ =-
7 d 4 70 I where you just pick up.the one. You have some 2 candidates via floor response spectra which you want to 3 compare from 2D again. Our proced ure was to select one 4 o} these candidates which is worse in terms of spectra 5 acceleration , frequency, acceleration of frequency 6 ceiling, and so we selected the worst amount. 7 This procedure was followed throughout all _. 8 floors and all five f an coolers. 9 MR. WRAYs Okay. Just for my own 10 clarification, if you did f all in a vertical plane that 11 took you between, say, where there were radio beams and 12 column nodes in the interior nodes you would probably 13 consider the spectra from both the interior nodes and 14 the radial beam' nodes and select the worst of that group 15 of spectra. 16 MR. PHILIPPACOPOULOSs Yes. What would I 17 select in this case? The first criterion is to select-18 those nodes which are close to that plane, and then from these select the worst one. If my tangential be am . .wa s 19 20 closest, the node on the tangential beam was closest at this point, I would pick up.that one. If not, if the 21 22 radial beam wa~s closest to this plane, I would pick up l 23 nodes from the radial beam. , 24 Does that answer your question? - 25 MR. WRAYs Yes, sir.___ l ALDERSON REPcRTING CCMPANY (NC.
~
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
~
~r - . . .
, za . . - - ..
?
71
.1 MR. CLOUD 'I have a more pedestrian 2 question. The results from this method are shown on 3 Pages 56 and 75.
4 MR. COOPER: Well, let 's go through the th ree 5 methods. We would particularly like to do this. 6 MR. CLOUDS I want to get clarification on the 7 method, because it is related to your question. In _ 8 particular, these graphs are shown. My question on both 9 pages 56 and 75 are which node points do these curves
~
10 represent on those two pages, and similarly which 11 spectra. What is the reference for the UES Bloom 12 ' spectrum? In other words, I, the reader, can ' t tell 13 .w hich ones are which. So, what is the node number for 14 these spectra? 15 MB. PHILIPPACOPOULOS: The node number for 16 these spectras is not indicated in these plots-here. And 17 this is why we try to give you the correspondence f rom 18 this drawing and from the small table which were listed in Page -- let me find the page number. Yes, on the 19 20 bottom of Page 47. Can you find that page? That will 21 give you wha.t particular nodes are involved in this 22 particular -- . 23 HR. COOPER: What Bob is asking is, f_r o m your 24 explanation, I would understand that on the figures on . 25 'Pages 56 th rough 75, that mach _of these figures in your ALDERSCN REPORTING COMPANY, INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
~7-
u ,. - . _ .. u . _. u . _ _
.)
I I l 72 t 1 results represent the response at a specific node, and 2 can you tell us what node is represented on each page?
;. 3 MR. PHILIPPACOPOULOS: Okay. The original i
4 q'uestion, I think , was, what are the nodes from UBS 5 Bloca report. 6 MR. CLOUD: No, no, we asked for.both. 7 MR. PHILIPPACOPOULOS For both. Okay. If _ t { 8' you ask for both, and the first one is, you can find 9 them through that small table on the bottom of Page 47.. 10 Now, for the second one, the second for our nodes are 11 not indicated on this figure. However, they are 12 available, and you can have them. l 13 ER. CLOUDS That brings to mind another I e 14 question then. At the top of Page 48, the floor spectra l 15 obtained at these locations are shown on Pages 56 and 16 75, but in f act it should be 56 through 75. Is that 17 correct? 18 MR. PHILIPPACDPOULOS: That is the correct
, 19 spelling, yes.
I 20 3R. DENISON: How do we request a correlation 21 between the node numbers and these figures? Is there a 22 calculation? . 23 MR. MIRAGLIA: Frank Miraglia of th e ,s ta f f . 1 l 24 The test way would be to say, in regards to . l 25 Page 56 of the spectra, would yo-u plea se provide what l! ALCER$oN REPORTING CCMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
, . . . ..---.~--.--c, - --,-
7 ,3
e * -- 73 1 mode number, and we can provide that list. 2 3R. CHIN: Stan Chin, Teledyne. Before you 3 vent to an explanation about picking out three node 4 points for a particular location, you said that it was 5 the worst responsa spectra case. i. 6 ER. PHILIPPACOPOULOS: Not all the time, but 7 we are aiming to pick up the worst. _ 8 HR. CHIN: What happens where you have a node g where.the frequency is different, you have an envelope,. 10 the three node points, and you get the response 11 spectra? What happens to the shift? 12 MR. PHILIPPACDPOULOS: We did not envelope. 13 We pick up individual nodal points and compare them 14 directly to the URS Bloom spectra. - 15 MR. CHIN: But what I am saying is, say you've 16 got a radio beam, and you've got a node at mid-span 17 which happe ns to have a certain -- excited at a certain 18 frequency locstion which is not at another node point. jg It happens to be at an intersection. You are going to 20 have a spike at a certain f requency location. What do 21 you do in those cases when you actually see you are 22 going to take the worst response spectra at that node 23 point? Do you allow peaks like that to just ;o around , 24 and you don't even censider them, or how do you make . 25 that decision? - l l ALCERSON REPORTING COMPANY. INC. 400 VIRGINIA AVE., S.W., WASHINGTCN. D.C. 20024 (202) 554 2345
- --n----,--,
- - - - - ,- - - - - - . . ..--,--.--,7~,----. .
a e 74 1 MR. PHILIPP ACOPOU105 : Okay. You are asking me 2 if in the case that I have to select:between those two 3 spectra, those two different numbers, and the criterion 4 i's as big as you see it. What would I do in this case. 5 Is this your question? 6 MR. CHIN: Yes. 7 MR. PHILIPPACOPOU105: I mentioned in this _ 8 particular case if both of the nodes are pretty close, g you won 't see the frequency shif t as a very large factor 10 to distinguish them. Frequency shifts in closed nodes, s 11 keeping in closed spaces in the same beam, you won't see 12 big frequency shifts, so offhand I don't remember a 13 specific case, but I have to deal with this particular 14 problem. 15 MR. CHIN: I think you would find that you had 1C the node in the mid-span. That would be in your motion - 17 drawings, you would get the form shape or you wouldn't 18 get amplification of beams, because they are not rigid. tg So that happening, I think, could happen, because if you 20 got a node at an intersection of the column and a node l 21 at the mid-span of a tangential beam, you are going to l 22 get that frequency shift, and you are going to get a i 23 large acceleration in the middle of that radial _ beam, so 24 I am just saying, if that was the case, what would you . 25 do in that e ve n t . - -- t ALDERSoN REPORTING CCMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
--w.._ ,_ . - _ ,.
, 7 _
* - m a e .
75 1 HR. PHILIPPACOPOULOS: In that case you would 2 pick up the one in the middle, but if you have two nodes 3 tha t are on the span of the~ beam, very close together, 4 that frequency shif t is not very significan t. Okay?. 5 So, if you have to compare between a node on the column 6 and on the beam, then it is obvious you are picking up 7 the one on the beam, not the one on the node, but if you _ 8 get two of them on the span of the beam, then the g f requency shif t is not great. 10 NR. MILLERS Chuck Miller.. 'I think for this 11 particular kind of comparison, there are three kinds of 12 comparisons, and you have two comparisons that really 13 address that question where they have many modes with - 14 different kinds'of spectra, and how do_you really pick 15 the worst, but the first kind of comparison, I believe 16 all of the spectra are for a particular node. There is 17 no attempt to take part of the spectra from one node and 18 part of the spectra from another. All of the spectra 19 are from one par'ticular node. I think the node pretty f 20 auch is the node that is closest to that location 21 geometrically corresponding to that location in the 22 Bloom report. We try to get as close as we can to make 23 comparison, the other comparison, what is aim ed,at . 24 solving the problem. 'Je -have spectra 20 degrees of the . 25 spectra, and how do you pick tha worst, the average, but \ ALDERSoN REPCRTING CCMPANY. INC, l 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024,(202) 554 2345 _ ,. _. ., n..,. ,
,- .=.-,. = , - _ - - _ , ,. ;_g
I e - . 76 1 I think in this particular comparison all the spectra 2 from one node in our model compared to one in the 3 spectra. 4 MR. REICH: Mike, I have just given you one 5 copy. I gave you a copy of this page, and I think if 6 You could clarify for this particular floor shown on 7 this model. _ 8 MR. KNIGHT And could I ask one thing? If 9 you say this page or this thing, it might be good to 10 identif y fig ures. 11 HR. PHILIPPACDPOULOS: Yes. For this one we 12 are talking about Figure 2.6 in th e BLN report, and in 13 this particular case, the nodes in the middle are the
~
14 nodes selected, and I think if you. let me go to the next 15 two comparisons, most of these ansvers will be more 16 clear. 17 MR. COOPERS Could you answer this question? 18 On that picture on the board, can you tell us which 19 direction is north ? 20 No, on the other one. The one you just had up. 21 MR. PHILIPPACOPOULOS: Give se one second. 22 O k ay' . The north direction in this case is , 23 that direction (indicating). _ 24 MR. COOP ER : Vertically upwards?' - 25 MR. PHILIPP ACOPCULOS s -Yes. ALDERSoN REPORTING CCMPANY,INC. 400 VIRGINIA AVE., S.W., WASHINGTON. O.C. 20024 (202) 554 2345 _ _. ..,_,.y_m._.rr--,___ _ . _ _ _ . . _ _ ._- _ _ - _ . _ _ _ _ . .__.m.~ .
~ .
u- -
- __ ._ ._ i 4- e 77 -
1 You have everything now. 2 MR. COOPER: That is what I had hoped to see 3 earlier. Would it be possible to center the two circles-4 s'o the centers of the two diagrams coincide? 5 VOICE: Is that the right orienta tion ? s NR. COOPER: That is what I was seeking 7 earlier, just to try to get a better feel for how the _ 8 Bloom planes coincided with your model. I am sorry. I 9 can 't read them back here. So, five just about 10 coincides with the radial beam in your model. That is 11 one exam ple, a nd a t the other extreme one looks like it 12 may go right through , about halfway between two radio 13 beams. You have got both situations, which is what I 14 wanted to see. The radial' line for plane 5 -just about 15 coincides with.the 346, 344, and yet in contrast, say, 16 the radial line f or cooler number 1 is almost halfway in 17 between 376 and 379. The radial beam is terminated by l l 18 those. 19 MR. PHILIPPACOPO'ULOS: Do you see the 20 difference between those two? The pick is 2.6 here. The're is no frequency shift. There is no frequency 21 22 difference. You pick up any of these two. This is the 23 easiest choice, but we have tougher choices to make when 24 ve have more notes around and significant differences in . 25 terms of frequency and pick, Iq.u_A n o w . ALCERSCN REPCRTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON. O.C. 20024 (202) 554-2345
-_ a. .+
41 w ' F 78 l f l >
- 1. MRi.~BEICH4 I' think you .ha ve answered the 2 question. -
'. t . .
3 MR.iCdOPER: Do ' you want _ to go into the second 4 method? 5 MR. PHI 1IPPACOPOU10Sa Thank you. This slide-6 here represents the way we define the so-called sections a l 7 here. We have five sections, one through five, five _ 8 sections. Now, why we went through this kind of g comparison, obviously, from the models you have a set to number of nodal points and a set number of flow response 11 spectra, whereas in the URS Bloom model you have one 12_ _ beam representing all sections. Therefore, you have to [l 13 compare between those two. Now, the sections again,
'L 14 ~ they follow very much the orientation of the f an 15 coolers. However, again I would like to emphasize our . 16 results are independent of the five fan coolers. I want 17 to make this clear. In this particular case, the five 18 sections are defined from the lines bisecting the two, 1g three fan cooler planes.
20 In this particular rase, the first section is 21 defined by the bisecting line between the fifth fan 22 cooler plane and the first fan cooler plane, whereas the 23 other boundary of the section is defined by the_ 24 bisec tin g line or the first fan cooler in'the second. - 4 25 The rest of the boundaries-are-defined by the f.rame vall i ALDERSoN REPCRTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON, D.C. 20024 (202) 554 2345 _ w._
3
+
s . .
~
~ , 79 '
1 and the outside beams of the model. 2 Now we have two kinds of spectra tha t we are 3 comparing J with'the UBS Bloom spectrum. The first kind - 4 is the uppe'r spectra. Upper spectra are defined as, for 5 this particular spectra frequency, you operate, simply I 6 operate all the spectr,a volumes from all the nodes which 7 fall within a sectio'nj.and the spectra were genera ted by _ 8 taking the envelope.'of .all of the spectra values for its
~
e particular frequency of all the nodes falling within the 10 section, so this is afdefinition of the two other kinds 11 of comparisons. 12 MR. COOPERS Do any of your dividing lines 13 happen to coincide with a.line of nodes on your model? 14 MR.'PHILIPPACOPOULOS Let me check my notes. 15 (Pause.) 16 MR. PHILIPPACOPOULOSs Geometrically, you 17 mean? 18 MR. COOPERS Geometrically. 19 HR. PHILIPPACOPOULOS: Some of them come--out ! go very, very close. 1 21 MR. COOPER: But not completely in 22 coincidence? 'You never have the problem of how a given , 23 node is shared between two sectors, I guess. . 24 MR. PHILIPPACOPOULOS: From th e ' geometrical - i l 25 point of view, no. It is 'clese -what. section it belongs ( t s ALDERSON REPCRTING CCMPANY. INC. 400 VIRGINIA AVE., S.W., WASHINGTON. O.C. 20024 (202) 554 2345
,=,
t: --- l 80
~
1 to. 2 MR. COOPER: The node is always associated 3 with one section. You never had that problem of trying 4 t'o figure out what to do with it. One is on the line. [ 5 - ER. REICH If it shares. I 6 MR. COOPER: Yes. l 7 ER. CLOUD 4 I just want to make sure I _ 8 understand the definition of the average. For example, 9 on the fif th sector, on the fif th sector there are 10 probably more than 100 nodes. 11 HR. PHILIPPACOPOULOS: I wouldn 't say that, 12 but there are a lot of nodes. 13 HR. MILLER: Not per floor. 14 MR.' CLOUD: By t'he floor. All right, so it 15 would be fewer on a per floor basis. 16 HR. PHILIPPACDPOULOS: No. . 17 MR. CLOUDS If you add up the spectral 18 accelera tion f rom those nodes, say, 40 or 50, and you 19 divide by the 50, you get the answer. I 20 MR. PHILIPPACOPOULOS: For each particular 21 frequency of that spectra, yes, that is what we did. MR. CLOUD: For the frequency spectra? Well, 22 l 23 yes, that 's righ t. Well, how many frequencies?, i ER. PHILIPP ACOPOULOS : All the floor spectra l 24 , 25 ve generated. They have 19 8 stectra frequencias. 1 t ALCERSCN REPCRTING COMPANY, INC, l 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (2C2) 554 2345 l l . .een - e.- , n ,- ,y mn m:- --
' ' * ^
i ---: , ll i
& D 81 1 MR. CLOUD All of that was done, I presume, 2 automatically.
3 MR. PHILIPPACDPOULOS Yes, through the
-4 computer.
5 (General laughter.) l 6 MR. REICH 4 And let me add, I had to use a lot 7 of clout to get that. It isn't fun, but we had to clear _ 8 everybody off. We had to use up the entire 7600 9 capacity and clear everybody off. There were a lot of 10 sharers, and we had to kick them all off. 11 MR. CLOUD 4 Well, I agree. That is 12 substantial. 13 MR. PHILIPPACOPOULOS: Are there any further 14 questions? 15 MR. REICHa Let me add one other. word. When 16 we used the other code to verify it, the cost was ) 17 horrendous. . l 18 MR. WHITES When you are going through the 19 averaging process in various modes, what was the l 20 weigh ting function associated with each individual i 21 node? l 22 MR.'PHILIPPACOPOULOS: It was unity. 23 MR. WHITE: The weighting factor was unity? , 24 MR. PHILIPPACOPCULOS: One. - 25 MR. C00PER4 Ars these-any more questions? ALCERSON REPORTING COMPANY. INC. 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345 L
= _ - .
_ . - , yf--- - 7
__. . . ; ,, m 82 1 MR. WRAT On'the diagram you are showing up 2 there, where the ba ttery is down about like 530 or clock
~
3 basis, the lower these mean that the first and fifth 4 sections, does that read 2357 5 ER. PHILIPPACOPOULOS: 235.5.
.6 VOICEa Is it a transposition ? Should it be 7 253.57 _
8 YOICE: A rather quick calculation indicates 9 there may have been a transposition. to MR. COOPER: I think we can all take out our 11 calculators and work on that. 12 NR. PHILIPPACOPOULOS: Excuse me. I would 13 like to ask if you have any further questions. Ve have 14 got the definition of the methods we used to prepare the 15 spectra. If not, the I would like to read that. 16' MR. COOPER: Are there other questions about 17 the three methods of comparison? I think not. 18 5R. PHILIPPACOPOULOS: Thank you. 19 58. C3OPER: Let me give you a ch ance to.get 20 back to your seat, and before I identif y exactly the 1 l 21 subject we want to consider next, and we are trying to l 22 understand the method in your thinking and doing what 23 rou did in developing the comparison matrices on Pages , 24 51 th rough 5 3. We fully. understand there is some . 25 judgment involved, and we a re_ not trying to. nickel and ALDERSCN REPCRTING CCMP ANY. INC. 400 VIRGINIA AVE S.W., WASHINGTON. 0.C. 20024 (202) 554 2345 i - _.__;...__,- , ,__ _ _ _ _ , _m _ _ _ . , ,
.a: .
- - _ 'n . x .- Bu s- -
.83 1 dime you to death in what you-have done here, but we are 2 asking the question,'because when we do what it is we 3 .think that you did, we get a different answer in some 4 cases, and we are trying to understand'what your 5 thinking is, and'if we are getting a different answer i
I 6 than you, then there are obviously three possibilitiess 7 one, we don't understand, two, in the mass of data you _ 8 have; and three, the figure has to be a typo, and our 9 problem is one of understanding. to In order to do this, let me suggest that we 11 need five different pages available to us. We need 12 Pages 52 and 53, two of the comparison matrices, and i 13- really perhaps one of those would suffice, but it is 14 easy to do two'for the price of one. We also need Pages 15 76, 82, and 83, and what we have attempted to do here is I 16 choose a limited number of pages that will exercise our 17 question about this issue. 18 MR. REICH 4 Could you go over those again? 19 ER. COOPER: Yes, sir, Pages 52 and 53, a.lso 20 Pages 76, 82, and 83. We don't need the last three 21 elements., We can take them one at a time. We will 22 start with Page 76. Now, on Page 76, just to bring ) 23- everybody up to date, as I understand what you ,have here 24 is, you are concerned with floor 1, section number 1, . 25 and your matrix tables on Paggs_E2 and 53, that is the
- ALCERSON REPCRTING COMPANY,lNC, 400 VIRGINI AVE., S.W WASHINGTON O.C. 20024 (2C2) 554 2345
_w...... _
.m .- [- s _, _ l_; ,_ . __ *. , ;.n l- _ ._ , __ .
84 1 upper lefthand cor'ner of the matrix, so we are going to 2 be doing something with Page 76 which will result in the 3 inclusion in the upper lefthand corner on Pages 52 and 4 53 of either a one or a zero. 5 Let's take Page 53 first, which is your third 6 method, which is comparison of your envelope spectra 7 with your averaging spectra, and on Page 76, your two 8 lightweight lines, the higher of the two is the envelope g and the lower of the two lines is the average, so the 10 higher of the two lines is the one we should look at for 11 the entry on Page 53, and the lower two lines are for 12 the entry on Page 52. Now, that is just to bring 13 everybody up to date on what we are doing here. 14 Now, we are trying to -- you folk s are trying 15' to represent an awful lot of detail by one simple zero 16 or one, and as I say, everybody appreciates the, 17 difficulty in.doing that.'Now, the reason I chose this 18 one to start with is, we certainly can agree with you on tg Page 53 to have a zero, which indica tes tha t in this 20 comparison the top lightweight BEL line is clearly much 21 higher in magnitude than is the peak of the heavy waved 22 line, and so obviously a zero should appear by the
~
23 scheme, and although there is some difference in 24 frequency, the major difference is in the amplitude, so . 25 rou put a P there. _ l ALDERSoN AEPCRTING COMPANY. INC. 400 VIRGINIA AVE., S.W., WASHINGTON 0.C. 20024 (202) 554 2345
~w_r _.
7 ; .~_,_..__.
~
- - = - G .- & - :w . L - K :' 3. ___ :- T' _ a J .-
85
, 1 So, like I say, the fact that there is a 2 frequency difference is not nea rly as inportant as the 1 3 fact that there is a difference,-and it is the kind of 4 thing you have to go through in making this sort of i
5 comparison. Now, let us look at the lower of those two 6 lines, and to ask the question, what should the entry 7 read on Page 52 in the upper lef thand corner. 8 VOICE: That should be Page 52 also. There is 9 an error in tha t page. 10 3R. COOPER: Well, that is why I selected this 11 instance. On Page 82, next, which is the second floor, 12 Section Number 2, and so this is the one that is just 13 one step right-and one step down from the upper lefthand 14 corner we were . talking about previously, and if we look 15 at the higher of the three curves on Page 8 2, we can 16 note that you have identified in the table on P. age 53 a , j 17 zero entry indicating that your spectra represents a 18 more severe loading condition than the other spectra 19 that is there, and you have indicated there is a 20 discrepancy in both the P value and the frequency with 21 which it occurs, and we don 't have any trouble with that 22 one. 23 In the entry on Page 52, where we are t li 24 comparing the lower lightweight lines on P. age 82 with 25 the solid line, we note one e,ntry which says that in the t . l
' ALDERSCN REPCRTING CCMPANY,INC.
l 400 VIRGINIA AVE., S.W., WASHINGTCN. 0.C. 20024 (202) 554 2345 l _- . _ . -----..:.-,--n.----.---
.. . - . . .. m c.. , ..
. , . 3 ._ _ _ _
-- 4 o . ,
s v ,-
', 86 g/
1 judgmen t -you were using when you did this, tha t the 2 curves were essentially the equivalent or perhaps , 3 actually in 'some cases it will indicate that"the heavier 4 line curve is much higher than the other curves. :Your- 4 3 5 judgment'is, that would be acceptable. If you had ts. 6 make a single judgment On that whole picture, you are 7 saying that would be acceptable. __ 8 Let's turn now to Page 83, so what we are 9 saying is, we don't disagree with you. We understand on 10 that, on how you determine >Page 82. If we look at Page 11 83, then clearly 83 compared with Page 53, so we a re 12 lookina at the top, clearly , your peak is auch higher, 13 than the heavy line peak, and very properly for the .
~
14 second floor, section 4 and section 3, you have a zero 15 in your table. No problem. , r 16 Now, when you come down to the lower of your 17 pair of curves on 83 and you compare it with your 18 heavier line curve , in your table you hite a zero, a P, 19 and an F, and we were trying to see ju e': conparino the ' ! 20 lower curves on Pages 82 and 3 with the heavy line 21 curve, we are trying to see if you were using some 22 criteria where on Page 83 it poses out a zero, and on 23 Page 82 it comes out one. - 24 N o w, if your answer is, it is just a minor 25 difference in the judgment ~ b etw ee n t he tw'o,-taen we l _ ALCERSoN REeCRT'NG coa *P AM,INC. I 400 VIRGINIA AVE., S.W., WA5NNGToN. D.C. 20024 (202) 554 2345
, __ _ - - _ _ . 7 , , - :y. ,. -, -
7,_
;5$
wh ::4%.Ma i - . 9 - A l '
. , 87 i
- j. . ..
' \
1 would be ready to accept, if we had the role of
-2 acceptiny, which we don't, but we are asking the 3 question to try to understand what criteria you may have
^
, 4 established, if any, for doing this other than just the 4 5 judgment of the guy that looks at it. Ho nas got a l ,
l 6 problem. He has got to write down a zero or a one, and 7 ha cannot take forever to do it. ,_ 8 MR. REICH 4 Can you go over the question 9 again? You are talking about 83. 10 NR. COOPER: Let's look at 82 and 83. The 11 a'verage method, the comparison on 82, you awarded it a 12 onh, and on 83, you awarded it a zero. I just wanted to
,13 -~ yo.u know, they have a lot of the same similarities.
14 I an trying to~ understand'if there is something I don't i
,15 understand, I guess.
- 16. MR. BLAKE: They are on different scales, if Y 17 you look at then.
/ 18 ER. PHILIPPACOPOULOS: Why don't you compare ,- 19 between the GEE, and the other ones? Wh y d on ' t you 20 compare between 2.8 and --
I t i 21 MR. COOPER: Too much looking at pictures, and 22 too little looking at fine print letters. A question on 23 Pag e .77. There is only one curve for the B CL spectra, , 24 and the question is, is one of those parallel missing or . l 25 do the others coincide ? - i
'l l Jr s
' s" ALCERSoN REPCRTING COMPANY, INC, 7
< e
! 400 VIRGINIA AVE., S.W., WASHINGTON. 0.C. 20024 (202) 554 2345
)
I e-
, - g - , y-9, , v,-..,-- .
w T p x _:- -
]
_; 'r [ 88 1 MR. PHILIPPACOPOULOS: What page? ?. 25 MR. COOPER: Seventy-seven. ,, 1 '. 3 MR. PHILIPPACDPOULCS: Onthisparticula}r'page. 4 we have here, and let me get the page again, now, you 5 would expect it to have two lines. It just so happened v, 6 tha't for this pa'rticular section, we have basically one 7 little point. - 8 MR. MILLER: This is one of the ones that is 9 open. The columns, the 18 columns -- or about the same-i , 10 time. Some of those columns stop above the first four 4 11 levels, so there is a big section on the first floor 12 that doesn't defer in certain sections of the floor, so 13 there are not that many nodes that go into this floor.
~
14 NR. C'00PER: So you are basically'saying ther , 15 are basically the same curve,' and they both occur. ?
;\
16 MR. MILLER: They are basically both ht the 17 bottom. 18 MR. WANG: That is basically from figure 22. l 19 MR. P.EICH: Lbok at Page 21. You will see , 20 nurbers -- q uestio n:- , i h a coh 23.
- 21 MR. HARRIS: Steve Harris. 'Jould that be the
/ '
22 spectra associated with nod 13 shown on Page 21'). , 23 MR. PHILIPPACOPOLOUS: No, node number'13 is 3 24 the base of the program. It is not a node'of the -
~
25 floor. --- ALDERSON REPORTING COMPANY. INC. 400 VIRGINIA AVE S.W., WASHINGTON. O.C. 20024 (202) 554 2345 en- mn, - --
--n.=-.- - . - - ,. .- - . . . - - - - - - - - - , - - - - - . - . - - _ _ .
ja --
.-- .,_ w _-
.__ . _ . x: .-
s , . 1 l-1 89 E l 1 1 VOICEa Section 2137 2 HR. PHILIPPACOPOULOS: In the same feature i i 3 that Dr. Reich mentioned. Please look at figure 82. i I 4 This is the node that the spectrum would account'for. i 5 HR. REICHs In this case, they fell on top of 6 each'other. Do you see it on 827
- 7 ,
f
,p 8 .
9 , 10 11 12 i i' 13 1 i ' 1 14 - 1 I
- 15 I.
16 17 18 19 20 21 i 22 23 l l 24 - i 25 -. i i t
- ALCERSON REPCRilNG CCMPANY, INC,
! 400 VIRGINIA AVE S.W., WASHINGTCN, D.C. 20024 (202) 554 2345
-- .m - _ . . _ , , . . _ . . , _ _
,s - - -
, _ _7 f. . --...-- ,-_ 7_ _
-# , - _ _ - . . , . _ m .
90 1 Let's list it in the middle. Should that 2 figure show the average of all three nodes that are 3 connected 'together, or is it just one . node that has been - 4 taken? 5 HR. PHILIPPACOPOULOSk That is the only node i 6 that was taken for this particular section. . 7 MR. MILLER: Well, first, 83, all those nodes 8 are on the crane wall and the crane wall is unmodeled. 9 So they support all the interior nodes on the floor. 10 They all support tha t node . So the one on 83, it has a 11 free field. So I think probably what this is in fact is 12 a free field spectrum. So all along at that level there 13 is no application, just because -- well, for whatever 14 sort of reasons. .
.15 But 83 is the support, so the only motion to 16 see is the inputs motion. ,
17 HR. HARRIS: The node we're dealing with from 18 827 19 MR. MILLER : But that should also have the 20 input motion. 21 MR. PHILIPPACOPOULOS: Well, excuse me. 847 22 84 belongs to.section number 3. 23 MR. KNIGHT: Gent lem en , we're getting into 24 this mode where we're having conversations around the 25 room. Try to speak up, t:I to 1et a person finish their l
- ALCERSON REPORTING CCMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
=e..e**w -o Wme ww$ e* g e-m wem- w- w p w * = =e p ,e
, s - _,= - ~ _ : . _.
91
~
..t point'.
~
., 2 .
MR. DENISON: Can we identify the node right, 1.b.' the average'from 827
- 4. MR. REICH 4 Well, 81 on top, 82 and 83.
51 MR. DENISON: And does the figure in the back 6 show an envelope and an averaging of nodes 81 and 827 7 MR. REICH: No, it shows 81, 82, and 83. One 8 is on the crane wall and.on.e is on the column. 9 MR. HARRIS: Steve Harris. 10 What criteria did yo*u use in eliminating nodes 11 for averaging and off-limiting of the spectrum of the 12 structure? 13 MR. PHILIPPACOPOULOS: The criterion we used 14 was to accept all nodes except the nodes that they are 15 located on the supports of the crane wall and the nodes 16 located on-the columns. 17 MR. HARRIS 4 Ste ve Ha rris again. 18 So You are assuming there is no amplification ( 19 through the vertical columns in your structure? l ! 20 MR. PHILIPPACOPOULOS: That's true. 21 MR. HARRIS 4 And my question would be, why are 22 they modeled if we are going to assume they are l 23 n on -am plif yi ng structural elements? 24 TR. PHILIPPACOPOULOS: '4 h y we::e .they modeled? . I 25 The answer to this question I_ guess is if you den't i l
- ALDERSCN REPCRTING COMP ANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345 y _ -. ._ . , . - - _ _ . --_.-.,.---m.--.-...,. .-
^
n . . . . . i '92
~ -
l 1 model the columns -- we didn't model the crane wall from 2 the-side, and you don't model the columns; you just 3 isolate the core. You don't talk about the full model. , 4 We have to have the columns there. We said 5 there's no amplification. There's a couple between the i 6 floors, between -- the amplification is through the f 7 hole. { , 8 MR. WRAYs Ron Wray. 9 Did you ever do an averaging including the 10 crane ' wall nodes or the column nodes, just to see what 11 effect on the average curve you would get? 12 MR. PHILIPPACDPOULOS: For the average, no, we 13 did not include those nodes. For the envelope, of' i . 14, contsel'these. nodes,'ther have no effect at.all. Now, i 15 for th'e other spectra, these nodes were not included. I i 16 MR. COOPERS Are there other question,s in , 17 trying to understand the three methods and the 18 methodology which was used without going to the results 19 that were obtained, without drawing conclusions here? i l 20 (No response.) 21 M r. Knight, it's 11539. Do you want to break i 22 now or do you.want to continue? We've got a couple that 23 may be fairly lengthy on page 49. 24 MR. KNIGHTa I have been informe,d by our hosts 25 that if we break now and go to 1unch we can probably get
- ALDERSON REPORTING COMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
,n m , C ' T ~~' ' -.-.:. - :-~~~
. 93 1 through expeditiously. It would seem in the interest of 2 overall time that would be the best thing to'do.
3 Let me ask one other question. Is it feasible 4 to get through lunch? I assume 'that we all probably 5 vill stay here, eat in the cafeteria. Is it feasible to 6 get through lunch in a half-hour if we go now? Well, 7 we'll give it a shot. 8 So it would seem that we could reconvene here 9 at 12:00 -- I'm sorry. Let's agree at 12;30. 10 (Whereupon, at 11.40 a.m., the meeting was 11 recessed, to reconvene at 12:30 p.m. the same day.) 12 13 14 . 15 l 16 17 18 l 19 i l 20
. 21 l
l 22 23 24 - 25 - ALCERSON REPCRTING COMPANY,INC. 400 VIRGINIA AVE., S.W WASHINGTCN, D.C. 20024 (202) 554 2345 g, - - . - . .
- ., . , - - - , -7..,....~.--...-.--.--_,...-..;..,..
~~
- _ - 1 _ _
94 1 AFTERNOON SESSION 2 (12:38 p.m.) 3 NR. KNIGHT: We had some discussions during 4 the lunch break about the remainder of the day and 5 whether or not it would be fruitful for perhaps a 6 smaller working group to stay over tomorrow. It 7 appears, however, that the likelihood of having 8 available specific information that a small working 9 group might want to look at, such as printouts, by 't 10 tomorrow is unlikely. It's just not feasible. 11 Information is stored on tape, we understand. 12 So it would seem, therefore, that the best format is for 13 us to proceed with this discussion, to do as much as we 14 can do today,.and then at some time, I assume rather 15 shortly, when we've had a chance to digest the results 16 of today 's meeting and decide upon wha t inf ormation 17 you'd like to have, work through the NRC project 18 manager, let him know what that information is. We will l 19 ask Brookhaven to develop it. 20 And I'll kind of shoot a glance to counsel 21 here. I'll provide to all parties, if that is desirable 22 -- if it turns out being large quantities of numerical 23 data that all parties are not interested in ,' then some 24 compromise will be reached, we can assume.. But it . 25 certainly is available. _ l l
- ALCERSON REPCRTING CCMPANY. INC.
400 VIRGINIA AVE S.W., WASHINGTON. O.C. 20024 (202) 554 2345 _ . , . - - _,--.---_._s._. :_.. . . . _ _ . - - . . - , . . . . , . _ _ _ - . _ - - _ , . - . - _ , . . . - - - - _
_w: a o- . 95 i 3. 1 And I think Mr. Fleischeter has very 2 appropriately said, just good track of what is asked 3 for, so ev'aryone is aware of what is available. And 4 we'll kind of charge Brookhaven at this point to do 5 thats whatever we do develop, keep a good record of 6 it. l 7 Dr. Cooper, do you have a feel for timing this i - 8 afternoon? 9 HR. COOPER: Ey best guess is we will complete 10 between 4:00 and 4:30. 11 MR. KNIGHT All right. We'll take a break at 12 2:30. Some of our people here staying here at
' 13 Brookhaven have to check out. " We may not necessarily 14 wait for everyone to get b~ack. So I'll plan on a 15 ten-minute break at 2:30, and we can proceed af ter that 16 with a somewhat reduced contingent here when people 17 check out.
18 Are there comments or problems on the i 19 schedule? 20 MR. C00PER4 Gentlemen, page 49. 21 MR. REICH: One comment on the previous l 22 discussion. I want to clarify one point. It may have l 23 come across that we do not have floor spectra for all 24 the columns and the in tersections on each . floor. We . 25 do. I just want to point that_out. We have a spectrum ALCERSCN REPCRTING CCMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTCN. D.C. 20024 (202) $54 2345 t .
, r-,-,,. . , - - .- - - -..-.-,,- .--:..-.-. . .----- -. ~;
._ xw.w2a q . -
~
96 1 1 for each point.
, 2 MR. COOPER: But these are just -- ; 3 HR. REICH: We didn't use it in the av.eraginq 4 process. , 5 HR. COOPER:- These are not available. They're l 6 in the digitized form?
I 7 MR. REICH: Yes, they are in digitized form. 8 You 've got to ramember' tha t we looked at in the 3-D 9 models -- we have 3-D models, and I think there are 10 6,000 spectra or so, and then in the 2-D we had model A , 11 B, C, D, maybe even E. 12 There's an awful lot of data that we have, a I 13 lot of run. And the best way to keep this was on tape. l 14 So whatever you need, we'll have to be specific so that 15 we don't spit everything out, because that's an awful
, 16 lot of paper.
17 HR. COOPER: Could I ask this: Is there 18 someone our computer people could talk to down here i 19 about the possibility of tying our computers together
; 20 and transfering it that way, rather than hard copy?
I* l 21 MR. KNIGHT: Technologically, it sounds like a l l 22 fine idea. But I'm not sure. 23 ER. COOPER: Well, we'll make a request. 24 MR. REICH: We can genera te a ta.pe. l 25 MR. Cl0UDs With_ respect to your question
;
- ALCER$cN REPCRTING CCMPANY, INC, 400 VIRGINIA AVE., $.W.. WASHWGToN. O.C. 20024 (202) 554 2345
.,- w m
, p. 97 1 'about spectra of the column lines -- 2 HR. REICH: That was a clarification. 3 HR. CLOUDS With respect to your 4 clarification, the elastic properties of the columns, 5 were they considered, so that the spectra are different 6 at the column points for each elevation? 7 HR. KNIGHTa The answer is yes. 8 HR. MILLER: Chuck Hiller. 9 , The elastic properties of the columns, the
. 10 elasticity, is included in the models. If I recall 11 properly, there's probably very little difference in the 12 spectra from one end of the column to the other, because 13 the elastic properties don't make a big difference.
14 HR.. REICH: It was included? . 15 HR. HILLER: They were included, just it won't 16 make a difference. 17 HR. CLOUDS I understand. Then the 18 clarification is that on page 48, where you talk about 19 your methods B and C, and you say that the envelope 20 spectra from all nodal points, what you mean is all 23 nodal points except the edge ones. It's paragraph 3. 22 In the middle of the second column where you 23 average, the word "all" is not correct. It should be 24 all except, if I understood your cla rif tca. tion, and the . 25 exception would be all the_ columns , all the column _
- ALDERSCN REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
,_. . - ~ . - _ _ .._.,-_...--_.....-r_m.__, . . , . , , _ .
, . , . _ . .T 98 1 nodes. And that only leaves the center nodes there. .
2 MR. COOPER: And the crane vall nodes also. 3- MR. CLOUD 4 The columns and the crane vall. 4 The crane vall is rigid, right? -l 5 MR. MILLER: Yes. The crane vall is rigid. 6 So that spectra is the input spectra ~. 7 ER. CLOUDS So it's really the average of all 8 the center nodes. 9 MR. REICH: Anything that is not on the crane 10 vall or the column. 11 MR. CLOUD s There is nothing else. 12 MR. MILLER: At different locations in between 13 columns, as the girder spans between columns, there may 14 vell be a node in the middle of that girder. 15 MR. CLOUDS There are in some cases, is that 16 it? 17 MR. MILLER: Yes. 18 MR . COOPER : On page 49, in the two-sentence 19 paragraph number 4, the second of th ose two sentences. 20 Dur question is one of trying to understand what is. 21 trying to be communicated here, and let me first ask a l . 22 question. When the words " envelope spectra" and 23 " average spectra" are used in this ' sentence [ do they 24 mean the third and the second m.ethods respectively? , 25 MR. PEICH: Yes._ _.
. ALDERSON REPCATING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON. O.C. 20024 (202) 554 2345
-.nn - - . . - - . . , . . . - - _ . . - - . - . ,._a _e.... . . ~ . - - ,_
-m
7 _ 99 L 1 MR.-COOPER: Then what you are saying is, in 2 your opinion the third method is more realistic than the 3 second method. And our question is, when the person 4 wrote this sentence, when he was saying more realistic, 5 what was -- more realistic for what purpose? 6 NR. PHILIPPACOPOULOSs Realistic in the sense - 7 that they are more representative of the response of 8 this particular section of the floor, in that sense. 9 ER. COOPER: I did not hear all of your 10 words. 11 MB. PHILIPPACOPOULOS: We say here realistic; 12 we mean realistic in the sense that they are i 13 representing more accurately the results, the motion of 14 the actual s cr'ucture's particular section. , 15 ER. COOPER: You are saying, I believe, that 16 if you had the choice between applying the second or the . ! 17 third method to the solution of the problem, that you 4 18 would choose to use the third. 6 19 MR. PHILIPPACOPOULOS: That is right. 1 20 MR. REICHa That was our opinion. 21 MR. MILLER: I'd like to add something. The 22 trouble with the anvelope spectra is, say you only had 23 two nodes and one node had a peak amplificat' ion of 11 24 cycles a second and the other node had a peak , l
- 25 amplification of 10 cycles a second. When you average l
i
- . ALCERSCN REPCATING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTCN. 0.C. 20024 (202) 554 2345
{ ,,, y , - ---. - ,.. _ -. ., _ _ . , , - - , , . . , . - __ _ ,,,,...
A i-
-"v ,qwf
= , d .; .
- o o t
100 i' - i . 1 those two, what you do is you cut bo th peak s in h alf , 2 both amplifications in half. And I think you tend to 3 wash out the amplifications when you go through the 4 averaging process, unless every node in that sector has
'5 amplifica' tion'at the same frequency. Since we have 200 e
i
! 6 frequencies, about, the chances are that there are some l ! 7 small shifts from one node to the other.
8 So I think the average spectra is very 9 sensitive to the number of nodes that are included. I 10 think that is one reason why you have the arguments 11 against the spectrum, the support nodes, et cetera. So 12 all that'does is basically wash out a little more the 13 effect of the amplification.
. 14 So I. don't think- you really see what is going 15 on as well as with the average spectra.
16 MR. COOPER Your choice between the two, 17 that's what you're trying to relate. 18 I've got some people that want to ask 19 questions, but again I want to caution everyone, we are 20 trying to understand what the guy meant by this
. 21 sentence, and we're not going to get into a discussion I
22 of whether or.not we think it is appropria te. We 23 understand what you meant; that is our purpo'se here 24 today, just to reaind everybody. , 25 Okay, Bob, go ahead. ALDERSON REPORTING COMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON, O C. 20024 (202) 554 2345
" - ---N#M, , 49 ' W9,9 99D -
_r--
u _
- m., ; - = m e n m.~ . =: .3. ; , ~. pn ,, -e e 101 1 MR . Cl0U D s I justi.'want to.make s.ure I do 2 understand, because to me realistic, the reason for 3 computing response spectra in tlh'el'first place is'to 4 obtain the response of attached.equipmen't. So to say 5 " realistic," realistic to me.means that if.that 6 earthquake were to happen in the way it is defined, that 7 ve would most closely approximate the response of some 8 attached equipment using the envelope spectra.
9 And I just want to make sure. That is what 10 realistic means to me. Is that what it means to you?
. 11 Do you believe that to be the case?
12 MR. MILLER 4 Chuck Hiller. 13 Yes. I think if you go back to my example 14 again, if you.had two nodes, one with a peak at 10.5 15 cycles and one with a peak at 11 cycles, if you have a 16 piece of equipment that is sensitive to that kind of a 17 frequency it is going to see, more than likely, an 18 amplitude that is equivalent to the amplitude of either 19 one of those spectra, so they both have the same 20 amplification. 21 What the average spectra would do, in the way 22 it is made, is basically cut those peak amplitudes in 23 half, and I don't think that is realistic. 'Perhaps, you 24 know, one could say maybe if you broadened,each spectra . 25 first and then averaged them, that that might be another ALDERSON REPORTING CCMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON, O C. 20024 (202) 554 2345
. - , = .
,e =====.
'?= ~M;
- _tt ,2 a .. -
T 102 I' 1 way of doing it, which .may be more realistic. 2 But somehow you have to taka care of the case 3 where each of the three amplifications are different 4 f.requencies,.and if you taketone out of 50 nodes, you 5~ take that one amplification, divided by 50, it comes _ f. 6 down a whole lot, and the'next' node has the same 7 amplification but a very slighily different frequency.
-8 You divide that by 50 again, you. bring that down, and 9 that looks like you never had much of;an~ amplification 10 at all.
'11 As a matter of fact, you do have an 12 amplification in the whole area around 10 to 11 cycles a 13 second, but you're bringing them all down with the .
14 averaging technique, and I- think that's not, realistic at 15 all. 16 Again, there are cases where I would sort of 17 agree that if the amplifications were far apart and very 18 isola ted -- 19 MR. CLOUDS I'm not disputing, in deference to 20 my chairman, I'm not disputing you. I-just want to make 21 sure tha t -- 22 NR. MILLE'Rs I personally very strongly 23 believe the averages are a very da ngerous w a'y to look at 24 it, at the spectra. , 4 25 MR. COOPER: We are, I think, ready to go on l
. ALCERSoN AEPCRTING COMP ANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
, - ; .: - - - - ~ -
- - . . , . = . + .. -
k._ - 103 1 into the piping area. let's just pause for a minute, go 2 around the loop, and make sure if anybody wants to add 3 anything about what we have been saying here recently. 4 Brookhaven folks? 5 '(No response.) 6 Staff? 7 (No response.) 8 Us guys? 9 HR. REICH: I want to ask one question. Is to there any comment about our table 5.1? 11 ER. COOPER: No, we understand what is 12 intended to be portrayed here. Your description of how 13- you felt I think we now understand. We are eventually 14 going to want. to look at the drawing relatiye to the 15 model to make sure we understand the detail we need to, 16 but that is not something we can do today. 17 MR. HARRIS: Steve Harris. 18 I am paraphrasing now from your report, but in 19 there you talk about models A, B and C, and you 20 characterire model B as most representative of , 21 structure. I was wondering if you might be able to l l 22 comment on what kind of differences there are, if any, 23 between the structural drawings that you are" in 24 possession of and model B that might be of, si;;nifica nce 25 to the audience here. _ l l
. ALCERSON REPORTING CCMPANY, INC.
400 VIRGINIA AVE., S.W., WASHINGTCN, D.C. 20024 (202) 554 2345 l n,,, ,
- - - . - - - vm n
- y . 3. - r, , y - -- .--~~~--,-~~73---
u : s- .:__ o . I
! 100 1
1 f: i 1 (Pause.) 2 MR. REICH: No, we don 't see any difference.
; 3 MR. HARRIS: You' re sa ying that model B in 4 fact does represent all the structura1 configuration 5 drawings you have in your possession, to the best of 6 your knowledge?
l 7 HR. REICH: To the best of our knowledge. 8 HR . COOP ER : And recognizing norm al modeling 9 techniques and so forth. 10 bat? 11 MR. CIATTO: Ray Ciatto. 12 The column to facemat connection, is that a 13 pin or a fixed connection? It was a Loman connection? l l 14 ER..PHILIPPACOPOULOS. Yes. , I 4 15 MR. WRAY: Ron Wray. 16 On page 66, 67, 68, and again on 70, this 17 piece is occurring at about 12 or 13 hertz. Have you 18 people identified whether that tends to be a local type 19 of response or a general type of response? Maybe this 20 was already answered. I 'm just making sure I understand t 21 it. 22 NR. PHILIPPACOPOULOS: The answer to your
; ~
23 question is no. We have identified'this local response 24 spectra by local. , 25 MR. DENISON: Ed_Denison.
= ALCERSON AEPCRTING COMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON, C.C. 20024 (202) 554 2345
_..-.-.7.._.--..
a , . . ., , m _ 4 105
- 1 I think I've got the answer to this. Did you 2 put crane wall connections all the way up to crane vall 3 1407 4 MR. MILLER: Yes, the input motion was in all 5 supports.
6 MH. C00PERa Can we go on to piping? Page 7 125. In the second . paragraph -- and this may have been 8 answered by the introductory presentation that you made t, 9 today -- in the penultimate sentence on that second 10 paragraph, in speaking of the two PGEE piping problems 11 You say "were selected." We were going to ask a 12 question about what basis for selection, but in your 13 ear 11er' presentation, as I understand what you said, that those two . vere identified for Brookhaven to be 14 15 analyzed by staff. Is that correct?. 16 MR. REICHs Yes. 17 MR. COOPER: Thank you. 18 MR. REICH: This happened, by the way -- let's 19 go over that. 6.11 and 4.A.26, this happened at the 2d trip that Mark Hartzman identified. l 21 MR. C00PER4 Our next item is at the top of 22 page 126, and .I would also ask you to look at page 136. l 23 In talking about the operating temperatures 'and
~
24 pressures for that particular problem, you, notice 25 there's a difference between the tempera ture and ALDEASON REPCATING CCMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) $54 2345
. m, ~ .- q. - ..--.-..--7-,. .-_.7_---. 77 7. -. -..., 7 -
_ _..7
.7
__ r 106 1 pressure on the top line on page 126 and in the top line 2 of the tabular information on page 136. 3 And we looked back at the drawing which we 4 understand you to reference for these data and find 5 still ano't her set of numbers. And so our question here 6 could also pertain to the other system at the same time 7 the difficulty arises. But we would like clarification 8 as to what the actual input was. 9 NR. BEZLERs The input for the stress 10 evaluation is on page 136. 11 MR. COOPER: And I presume on page 137 also? 12 MR. BEZLER: Yes. 13 14 . 15 16 17 d 18 19 20 21 1 22 i 23 24 . l 25 -
- ALCEASCN REPCRTING COMPANY, INC, 400 VIRGINI A AVE., S.W WASHINGTON. D.C. 20024 (202) 554 2345 i ~ - ._.y _
.,.;,y _ _._ . _.
- w. . .. _ _ . . - _ ._ : _ x , = . . a 107 y
a
- 1 HR. COOPER: The reason v~e are talking here is.
2 there is one we were looking at a little differensly and 3 ve were going to drop it based on previous discussion. 4 But on page 126, the third sentence, although the node 5 numbering is consistent with the Applicant's, node
! 6 locations are not identical.
i 7 MR. WRAYs And that sentence and the sentence 8 before implies that since the numbers are not consistent
~
9 with the as-builts, one could draw the conclusion that 10 the model numbering and positions of the nodes is fairly 11 close to one another, but then the rest of th a t sentence 12 goes on to say that the node locations are not identical. 13 Si) my question is are the node locations very 14 close together.or are ther significantly different that 15 that is really not a one-for-one node comparison. 16 MR. BEZLER: Bezier, Brookhaven. 17 I would say in general the node locations are 18 relatively close to each other, except where we chose to 19 put in additional nodes. Of course, they do not exist. 20 MR. WRAY Is it possible for you to clarify l 21 tha t further? , 22 MR..BEZLER4 I would clarify it by saying tha t 23 to the extent that as-built dimensions matc$ design 24 dimensions, that is how close ve are. I vould not want ' 25 to give it a number. It is probably correct in most ) __
= ALCERSON AEPORTING CCMPANY, INC.
400 VIRGINIA AVE., S.W., WASHINGTON O.C. 20024 (202) 554 2345
E .
.-a .
. ., = -1 4 .
108
- 1 instances bht there are times when it could be wrong.
9 2 MR. COOPER: On page 127, the first full 5 3 paragra ph on tha t page, the last two sentences -- the 4 last two out of three sentences in that paragraph -- we t
]- .5 are just not really sure we understand what it is that 6 you folks are trying to communicate there and'perhaps if ; 7 rou say it in different words that is all we need. . 8 NH. BEZLER4 Are you talking about the second 9 paragraph?
10 MR. C00PER4 The first full paragraph that 11 starts with "The comments." 12 HR. BEZLER4 And you are talking about the 13 values for spring stiffness, et cetera, et cetera?
! 14 MR..C00PER4 Yes. .
t 15 HR. BEZLER: Okay. Bezier, by the way -- 16 B rookha ven. 17 In the Applicant's solution,'certain of the 18 support elements were modeled as springs -- linear 19 springs. We chose to model those identically the way 20 the applicant had modeled them, although we could not 21 corroborate.their choice of spring stiffness of those l 22 elements. Is.that clear or should I elaborate? 23 MR. C00PER4 You attempted to independently 24 verify? . 25 MR. BEZLER: I a_.ttempted to do that and my
- ALCER$CN REPCRTING COMPANY. INC, 400 VIRGINIA AVE S.W, WASHINGTON, D.C. 20024 (202) 554 2345
[ _ _. ._.,__..._ _ _ _. _..___ _ _ w _ .. _ _ , . _ , , _ _ _ . _ . . . _ . . . . , ,
4
- .- - .= _ _ ._ 1
-~
a e . 109 1 attempt was inconclusory. 2 MR. COOPER: You got different numbers? 3 HR. BEZLER: I would have used different 4 numbers. 5 HR . COOPER : But you are saying that your , 6 number and their number were such that the parts would 7 behave in a rigid manner? 8 NR. BEZLER4 That is correct in terms of the 9 overall conclusion. Either selection would have 10 resulted in a near-rigid measure. 11 3R . COOPER : Thank you. 12 The last line on that page, page 127, you note 13 that -- you have noted that the overlap requirements 14 from the NUREG, the intent of those requirements have 15 been met and continue on with a statement that other 16 recommendations regarding overlap support forces'and 17 input spectra expressed in the NUREG were not adhered to. 18 Presumably, since you did not say anything 19 further about that, are we to understand it was your 20 judgment that the lack of adherence was not significant? 21 3R. BEZLER: No. I think this becomes a 22 matter of judgment -- not of our judgment b ut other 23 Staff judgment -- as to whether it should or should not 24 be adhered to. Clearly, when the plant was designed . 25 there were no criteria so_those criteria could not have
- ALCEASON REPORflNG COMP ANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTON. O.C. 20024 (202) 554 2343
.--r._ -----3--,.-,-.--+ _
- n- ,n,.-- . .
. 7-y _; 7 - --~~;- -- ; - _ -
4 i l- 110 i 1 been held to. 2 MR. REICH 4 The NUBEG CR, by the way, came out 3 in 1980. 4 HR. C00'PER4 So you did not continue to pursue 5 this because you recognized that the NUREG document 6 itself was post the design period of the plant? l 7 HR. BEZLER: Correct. 8 MR. COOPER: Those are the last questions we 9 had on Section 7. Perhaps this is a good place to 10 pause, go around the room to make sure there is no one 11 frustrated with anything I said. Anything here? Staff 12 comment? 13 MR. KNIGHTS No. 14 MR.. ANDERSON: Paul Anderson. . 15 Just one quick question concerning the piping 18 .being still verified. If Erookhaven could mention 17 quickly why they feel verified, whether it was a Staff 18 directive and to what extent it was field verified. 19 MR. BEZLER We at Brookhaven had no field 20 verification. The plan submitted to us had design j 21 dimensions. l 22 MR. ANDERSON: So when you specified design
~
23 dimensions, that wa s the utility 's design dimensions? 24 3R. BEZLER: That is correct. 25 MR. REICH: This is the same as this morning i . _ _ _
. ALCERSCN REPORTING CCMPANY, INC, 400 VIRGINIA AVE, $.W., WASHINGTON, D.C. 20024 (202) $$4 2345 - - . , . - - . - A.._,.--.,-._ _ . . . . _ . . . -__...___.s_. .___ _,_
q ..
-t
- j. 111 l
l' i
~
1 regarding structural. 2 MR. TRESSLER: Mark Tressler of the project.
.,. 3 Paul, how did you model the overlap in your 4 anslysis?
j 5 HR. BEZLER: We used the extent of model as 6 per your model -- in other words, an identical model as
, 7 rou saw, an identical cutoff point for the model which 8 you said you had.
9 . MR. COOPER: This area, noting for piping 10 obviously we are going to be interested in taking in the 11 computer printouts and so forth eventually, but we 12 recognize we cannot do that today. 13 Other comments or questions on Section 77 I l 14 think we have.been around the loop. 15 Let us turn to 8, page 132, Item A, the words, 16 the PGCE mathematical model. We may have missed 17 something but we are not sure just -- were you given a 18 definition of the mathematical model or did you deduce 19 what the mathematical model was from the drawing, from 20 the model? 21 MR. BEZLER: No. We in fact were supplied a 22 computer input for their model. 23 MR. COOPER: Okay. Thank you. 24 Items B through 3 are things, of, course, that 25 are very significsnt, but we saw no way to really ask
- ALDERSON REPCRTING CCMPANY, INC.
400 VIRGINIA AVE., S.W WASHINGTCN, O.C. 20024 (202) 554 2345
,- . .. . -, e . y ---,,n _
y-- - - -,.n.~..
; u- . _ _ _ _ __
o . 112
- es 1 detailed questions about it without going into the 2 computer printout in most cases. In some cases we do 3 have questions identified.
4 On page 133, the last paragraph before Section 5 8.2, again we had a little problem in understanding what s 6 you were trying to communicate there and while I can 7 almost see, I know because we have had all the 3 discussion on what you did have and what you did not 9 have to work from. But could you amplify a little bit to what -- did you determine a source of this minor 11 geometric error that you mentioned? 12 MR. B EZLER4 As it regards the geometric 13 error, one node was mislocated in the PGCE model. I 14 could show you .a vuoraph of the magnitude of that 15 error. It was nodel point 27, as I recall. 16 MR. COOPER: But the systems dimension being 17 different, are they associated with the same thing, or 18 are they different? 19 MR. BEZLER: No. The systems dimension 20 there. We got into the as-built versus the design and 21 we got into this problem. One reason or one branch of 22 the problem is, if we can refer to the drawings, there 23 was some ambiguity in the dimensions. One dould select i 24 either two sets of dimensions, if you will,, and come up . 25 with locations. _
. ALCERSON REPCRTING CCMPANY, INC, 400 VIRGIMA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
_.--_ _-. - ,..,_- _ m _. - . . - - _ _ . - _ - _
..a -w_ s -. = .
_. _. _ _ _ . _ __ r __ l : ;, n a= . .
+ 113 1 HR. COOPER: You had two sets, obviously. You 2 only had a choice of two.
3 MR. BEZLER: Right. , 4 MR. COOPER: So I think what is clear is we 5 vill eventually have to determine which one of the two, 4 6 if either is the appropriate one and see what it means. - 7 Perhaps if we could see the vugraph. 8 HR. BEZLER : Of the geometric error? 9 MR . COOP ER : Yes. 10 (Slide.) 11 MR. BEZLER: That was roughly the scale. I 12 think it is the scale. 13 HR. W5AY: What type of offset are we talking 14 about to judge that in the ski direction? . 15 ER. BEZLER: I do not know. The real drawing 16 shows it straight. But I am afraid I do not know what 17 that angle is. 18 HR. HARRIS: Steve Harris. 19 From looking at Figure 7.3, tha t node point 20 appears to be the overlap region, is that correct? 21 MR. BEZLER: Right. But in our solution we 22 used that node.as part of this problem. In other words, 23 the dashed portion of the piping shown in the figure was 24 included as part of this problem. , 25 3R. BIGGS: Which of the various floor l
- ALDERSoN REPCAT;NG COMPANY, INC, 400 VIAGINIA AVE., S.W., WASHINGTCN, D.C. 20024 (202) $$4 2345
- . ~ . _ . . . . . . . -_ - -.
-7.~_...
....., _ - - , . . -- . . .y
__ a - x- .
.a . ...+ -
uo o 4 ,. p 4 . 11 f4 4
.s c
(~ . ' , 1 response s'Inctra p did you use? / , , i 2 MR.,SEZTER: wen id the analysis both with the a 3 PGCE spacted,V.the' supplied ' spectra. 4 HR.'BIGGS4 It was horizontal. 5 NR. BEZLER: No, both horizontal and j j -6 vertical. The next analysis was PGCE--supplied , horizontal spectrum -- the Model A' spectrum coupled wi,th 7 8 the BEL Model E spectrum'for-the vertilctl.
\
9 MR. BIGGS Which of the three me thod s -- , i to envelope? - * , 11 MR. EEZLER: Most analysas we used envelope. 12 We used independent support motion methods. Oh,,axcuse 13 se -- which of the spectra, you said? ,, , 14 3R..EIGGS Yes.- , , i 15 MR. BELLEBs We would use an envelope spectra te derived from the spectra a p;'1'ying to all supports, in I 17 other words, we would get the spectra that appliec to. , n
~
;i 18 each nodel point of a support and then envelope all of )I 19 those. Am I clear?
20 MR. BIGOS: No, I do not::hink so. 21 MR.. SUBUDHI We did not go through the ! 22 spectra. We went node by-node. '9e would pinpoint this 23 point, put them in the spectra -- e s 24 MR. C00 PEPS' Ihe end points werc, always 25 coincident with the spectrum mod el node. '
. AJE98CN RivCRflNC *CMPANY, NC, 400 VIRGINIA AVE., $ W.,64 MNGTON, D.C. 2C024 002) $$4 2348
, . ~ . + - - ew - . .w e -- e - +,_--~y -f - - + . . .
---.-=p 9,o _-w e _-*og ,
= _ M,, _
'i _ 2a=_ . _c : _ _ _ .=
q + (- ,,- f c; / 115
+ c e 1 1
m
. MR. SUBUDHIs Thst is right.
2 ..' MR. COOPER: Page 136, the first paragraph of I. , l F ./ 3 8!%2, we have three questions about that one little
; 1 .
- /
4 e p.aragraph and let me not take them in the sequence in 5 which they appear but perhaps in the sequence that makes i I
- I, ' 6 it .! sore rea'dily answered . -
h i ' (q ,, / 7:' First, the program, PSAFE 2, is the users
!,V '
'( . 8 manual available?. ,
9 "> HR.'BEZ.ERs No, the users manual is not in ic formal report form. 1,1 ER. C00PER4. Is t! s a modification of one of
+
- 1, the commonly used piping programs or what can you tell
/ , 12 y
J , 13 u: about it? 14 HR..SUBUDHIt 'he have a program for E-pipe. I have a meno for that, and this is an extended version of
~
! . 15 16 E-pipe.
17 ER. CLOUDS That' has been verified? 18 MR. SUBUDHI That has been verified. That
)'
! I has been used as a benchmark for all the others.
te 23 MR. CLOUDS. Okay. -
! -[21 MR. COOPER And presumably this variatiCn of 'T 22 E-pipe has been verified.
23 All right. The next question, in the last 1
., 24 line of that paragraph it is stated that t,his CCde 25 performs the stress evaluations in accordance with 1;
il l l ! l
. ALCERSON REPCRTING COMPANY,INC, 400 VIRolNIA AVE., S.W. WASHINGTON, D.C. 20024 (202) $54 2345
( .*., -- ,, _ . ,,,- 7- _ . . - - . - - - . - _ _ , _ . , . __ _ ,,,-
. ._.a 'T'.=
u + " :_ '
- .._~ ._.. .. -: . .. .
J 116
/
1 NB-3600, 1974 version. Could you roament on why you t
,f )
2 would chose to use NB-3600, 1974 version, in this ^ ' 3 evaluation? . . 4 . MR. SUBUDHI: We use'd because that is'the !' 5 thing available to us at this particular time and that 6 is the mostly classical' analysis of any piping. We use 7' it and it is not auch different from 477, except the 8 branch connections or something like that. 9 NR. COOPER: I would like to acknowledge that 10 I just got my wrist slapped for using the word'"why" and 11 I would just like to apologize to everyone. , 12 (Laughter.) ' 13 NB. COOPERS The way I wanted to word that was 14 did you mean to reference -NC-36. c , NB, NC there is,not much -' 15 NR. SUBUDHI l 16 difference in 3600 and piping analysis.- c' 17 MR. WRAY: Ron Wray. 18 In th e text it sa ys NB .
/
19 ER. SUBUDHI No. It should be NC. N3 is 1 20 Class I. I am sorry shout that. i 21 5R. HOCH: NC or NB? l 22 MR. . COOPERS So that is a typo in the text,
~
23 basically? 24 MR. SUBUDHI: Sure. 25 MR. COOPER: The final question about this-
. ALCERSON REPCRTING COMP ANY. INC.
400 VIRGINIA AVE S.W WASHINGTCN D.C. 20024 (202) 554 2345
- iP" - --
* ,~~ _ j ,ff[_OY _2- _. 2. _-
-.. _ a . :: n~- .=; - _ _
=-
,d -
a < t t 117
, 1 paragraph, if we understand the third line correctly you 2 did use seismic anchor movements in this evaluation and 3 I presume that you used it in the evaluation for all the 4 various design conditions that you mentioned. Is that ;. 5 what you intended to communicate?
I
; 6 MR. SUBUDHI That is correct.
I i 7 MR. COOPER 4 I guess just as a comment, the i 8 use of seismic anchor movements here is not a mandatory 9 requirement of the code and tha t is the reason we asked 10 the questior. the way we could ge't at it. . 11 MR. ANDERSON: Paul Anderson. 12 Could Brookhaven identify the possible 13 magnitude associated with the seismic anchor motion as 14 opposed to the. thermal and pressurization anchor motion 15 associated with the evaluation of Line 6117 l 16 MR. SUBUDHIa It is not attached to any of 17 these cubes where you can get a nozzle load, but as for 18 seismic, we had provided all the s'tructure motions from 19 PCCE's, right, and we took this press bench from our 20 anab.ysis, whatever we superimposed to find out about 21 worst possible combination we could get, which would 22 make the highest stress condition in the pipe 23 condition. 24 MR. ANDERSON. I am just trying to get a feel 25 for -- I see over an inch of movement at the exterior
. ALDERSCN REPCRTING COMPANY. INC, 400 VIRGINIA AVE, S.W, WASHINGTON, D.C. 20024 (202) 554-2345
--- y _ -
.- - n , 3 - . mp ~ ,,_ , ,._- , p r ,y
^
-- m . . - ~=_ : -e_ s
.[ :
118 1 containment anchor and .it appears to be radial-2 movement. I am just trying to get an idea. Is that 3 primarily due to pressurization? 4 MR. SUBUDHIa Tha t is right -- containment 5 pressurization and that is provided -- 6 HR . COOPER: Can you reference something that 7 states that the seismic condition should be combined 8 with the containment pressure eff ect ? I am asking what 9 was your source for havinc checked this. 10 ER. SUBUDHI -- We thought the pressurization 11 should be combined with seismic. 12 MB. COOPER: But I asked if you had specific 13 reference in the Hosgri evaluation for the requirements
~
14 of making this evaluation.- . 15 NH. SUBUDHIa I do not know. 16 MR. WRAY: Hon Wray. 17 In the same context, was there a requirement 18 to include seismic anchor motion in with seismic motion 19 stresses in any Hosgri documents? 20 MR. SUBUDHIs I d o no t- know about Hoscri, but 21 as far as NCOND is concerned, the seismic anchor 22 movement has to be considered in the evaluation, but not 23 as primary stresses combined with secondary ' stresses. 24 So when you say NC-3600, we are talking e i,g h t , nine, 25 ten. _ I l l l
- ALCERSON REPCRTING COMPANY, INC.
400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345 c--. . - - - _ - . _ . + . , , . . - - = . -: :.,
- m. *
- M.- _'
_t___# . _-. ,; e 2 - - n v~ .
! 119 j
1 MR. WRAY: So in your evaluation later on 2 where would the seismic anchor motion stresses appear -- 3 as additions to the primary stresses, or would they be 4 compared to the primary allowable, or would they be then 5 added in with the secondary effects?
] 6 NR. SUBUDHI The computer program there with ; 7 question 9 and if it satisfies in the secondary stress 8 it goes to thermal and other things, but if it does not 9 satisfy, then that goes to part of the thermal stress 10 equation 10 and check against that. And if it does not 11 satisfy that, it goes to equation 11, where equation 8 12 is combined with th a t. So it goes to the entire, ! 13 step-by-step way of doing it, as it says in NC-3600.
I i 14 ER.. COOPER: Is -the way in which this program 15 we see treats it the same as an E-pipe? 16 HR. SUBUDHI4 No. The E-pipe does not do an [ 17 SME evaluation. 18 HR. COOPER: Okay. Then is this 19 post-processing routine available that you are talking 20 about now so that we can chase it through to see just 21 wha t you did ? l 22 MB..BEZLERs It is in the form PSAFE 2. l And it is a vaila ble 'in NC-3 600. 23 MR. SUBUDHI 24 It is available there. , 25 MR. COOPER: Well, we have a lot of experience i li
. ALCERSCN REPORTING CCMPANY !NC.
400 VIRGINIA AVE., S.W., WASHINGTCN D.C. 20024(202) 554 2345
, . . _ - - . - . ~ ~
yv, , - . k -- " ^
.:. m _ _ 2 E l m
__ ._._ .I:C: -+ 120 1 with the people's^ programs interpreting the Code, and we 2 find it necessary to verify for ourselves that they used 3 the Code appropriately and this is a general statement, 4 not aimed at you folks. 5 HR. SUBUDHI: How are you going to check it 6 unless you go subroutine-by-subroutine 'and 7 line-by-line? You cannot look at the output or input 8 and say that it is checked. We have checked by hand 9 calculations at seve ral points. 10 - ER. REICH: It is different point. You are 11 going out of context. 12 MR . COOPER : On the top of page 138 you 13 mention comparison with the emergency condition equation 8 14 9. Did you also check the satisfaction of the Hosgri 15 criteria for this system? 16 HR. BEZLER: I would say no. . . 17 MR. REICH: Please clarify. I 18 MR. COOPER: These code rules were not 19 applicable to the Hosgri evaluation and I believe that l 20 equation 9 was to be checked with the allowable 249H, l l 21 and my question,really was whether yoa had checked to l 22 see if the piping did or did not -- your evaluation of 23 the piping did or did not satisfy the stated criteria 24 for the lirensing code. 25 5R. BEZLER: We used the current ASME
= ALDERSON REPORT NG CCMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTCN O.C. 20024 (202) 554-2345
u :,. . _ _ ~. . x.a ,
.a 121
, 1 cri te ria .
2 HR. REICH: .We were told to use the current 3 AS3E. 4 MR. COOPER: Well, you can understand why. 5 NR. SUBUDHIa Well, if you look at page 151, 6 which gives you equation 9 on some of the tables and the 7' allowable KSI is around 27, which is 1 8, 2.4. So that 8 verifies what number we used. 9 ER. REICH: But to answer your question, we to knew that. This is the message we had from our 11 sponsors. 12 MR. CLOUDS Can I ask one question in that 13 regard? In thise sentence --
. 14 HR..REICHs What- sentence? ,
15 ER. CLOUDS At the top of page 138. I believe 16 that the stresses did not meet the emergency limits out 17 where you had the anchor motion by the containment, 18 which would suggest that, you know, when asked the 19 question, it is your opinion. 20 The Hosgri seismic 1 cads should be added to 21 the anchor motions due to the LOCA, and both together be l l 22 -compared to emergency condition limits.
~
23 MR. C0'OPER: You can ask the quest 1on, if th a t 24 is what they did, Bob. 25 MR. KNIGHT: This is Jim Knight f rom the i
. ALCERSCN REPCRTING COMPANY. INC.
400 VIRGINI A AVE., S.W., WASHINGTCN, D.C. 20024 (202) 554 2345 _n--
c;. r- [ -. 1-
.o 122 l
1 Staff. 2 I think we really asked a different question. 3 Is it your opinion and it should be. We had asked to do 4 certain specific tasks for us so we could gain 5 insights. I should go back to my original statement 6 when I opened the meeting that Brookhaven is acting as , 7 an arm of the Staff. We asked them to do certain 8 specific things for us -- if you will develop 9 information using the best technology that was available to and a number of other things. 11 We were not trying to so much do a pass-f ail 12 for a single problem or a single couple of problems. We 13 vere doing these things to give us insights to explore 14 the overall question of the process that was used, the 15 criteria that were applied , with an eye toward answering 16 questions in a number of different forms. 17 Whether or not that question was related to 18 specifically what was required for that plant at this 19 time or whether that question was a what-if question, we 20 just wanted to be able to answer on a broad spectrum of 21 questions. I think it is a disservice -- I am not just 22 saying it to you, but just in general -- a disservice 23 for anyone to try to make more out of the ce' port than 24 that, that it was an exploratory effort on the part of 25 the Staff to garner, if you will, a rather broad ALCERSON REPoATING CCMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345 7-. ,-,p-,.mm-~,.-,---_,
. , _x n- - : . :e 1 . a_ a, _. -
. . :_ _ , - - ::n . :.
.l . .-.
-i. -123
;) -
' - 1 . spectrum of information. .
2~ 3' 3 - 4 .
.]_
,. 5
.t-t' j'- 6.
e
- - . 7
, 8 , 9 1 10 1-11 1-
. . 12 i 1 1 13 i .i - 14 . . u jt I .15 1
-16 +
17 i 18 19 - I i i-l- 20 'A 4 -
. 21 i
! 22 23 24 . i 25 i a 1
. ALDERSON REPORTING COMPANY,INC. t 4 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
~
-.n-_ g N,
..._..__'W,MW'*s9M,. - . . . - . - , - , . ~ - -- . . + " -
.w- . .s,,- - - * - --w'**- ---
m- -~=~-<-re ' - u "w f~ ,,F-e f r, e
~~
__m. _ e e 124 1 'MR. COOPER: We will do that oursel'v'.s.. e We' 2 will then use that information appropriately i n t h e - 3 context of the case when we make our decisions.' t 4 HR. CLOUDS Really what you said was tNe
~
5 emergency limit that is really for the purposes of 6 information to understand. 7 NR. KNIGHT: It certainly wasn't any what I e would say -- there was no desire or effort to somehow 9 knowledgeably modify the criteria that were to be to applied to the plant. 11 HR. COOPER: Jim, th a t is the last question we 12 had. We would like to have the opportunity to caucus, 13 and we suggest the other groups might also, but before 14 we d'o that, I.think we ought to go around the room just 15 one more time and pick up items, and we will follow the 16 sequence we had before, and ask if Brookhaven would like 17 to add anything. l 18 (No response.) 19 MR. COOPER: All right. Staff? 20 ER. ANDERSON: I had a few things. Cn Page 21 132, Section 8.1, Brookhaven mentions they considered 22 the first 15 frequencies. I was wondering if that is
~
23 associated with some engineering judgment per the 24 interests of these two lines, or is it soqething they . 25 are not derided on, or wha _t t;5e mechanism was for coming
- ALCEBSCN REPORTING COMPANY, INC, 400 VIRGINIA AVE S.W., WASHINGTCN, D.C. 200241202) 554 2345
. ,_, - .. - ,...-...,. .n . . . , _ y ,_m
~~ ..--
tw _ LM .__c . J _.. _1 ,l
* - ~
-d- ' 4 1. - -
l 1 l 125 i 4 l l
'! 1 up with 15.
2 MR. BEZLEP: I think we chose the 15 by and 3 large because it was above the cutoff range of 33 hertz, 4 a t least for problem 6-11. Problem 4-A26 goes all the I 5 var up to 47 hertz, but I think this corresponds to the 6 applicant's choice of how many frequencies'are -
! 7 considered.
8 MR. ANDERSON: And I have an additional 9 question on Page'133. The first complete p arag ra ph . 'It 10 talks about the differences between problem 6-11 using 11 high Diablo's~1ong radius elbows. And my question is, 12 are there any other as built differences between the 13 design analysis and the Brookhaven analysis other than 14 the elbows? . 15 HR. BEZLER: In problem 6-117 16 MR. ANDERSON: To be quite honest, I can't . 17 recall for sure. I would be inclined to say, yes, there 18 were some differences in dimensions, but I cannot say 19 definitely. I just don't recall any. 20 MR. COOPER: You of course did ha ve the 21 dilemma then in the next paragraph that we talked about 22 earlier.
~
23 MR. BEZLER: Yes, there clearly are 24 differences in the a s built. I believe in both there 25 were some differences, but_as I say, I just can't f
. ALDERSCN REPCRTING CCMPANY, iNC, 400 VIRGINIA AVE., S.W., WASHINGTCN, D.C. 20024 (202) 554 2345
--_._.;u._-..__7__-.__..______._,,m., . . _ _ _ , _ _ . . . , . .
;- 126 +
1 remember for certain. 2 HR. COOPER: Ray? 3 HR. CIATTO It was mentioned in the summary 4 section, I believe, that support forces computed by
- 5. Brookhaven were substantialiy
- greater than those 6' computed by PGCE, and I was wondering if there was any 7 code evaluation done of supports for the use in 8 Brookhaven forces using ASIE code or something.
9 HR. BEZlER: No, we did no subsequent analysis. 10 MR. KNIGHT: Along those lines, I do feel it 11 incumbent on me, and don't feel I am jumping on you, it 12 is something I feel we ought to keep before us. All of 13 us as engineers tend to focus on the bottom line, is it -
~
- i. 14 accept'able or.isn't it, and is it safe, or 13owever yo.2 15 vant to phrase it, but again, let's not allow the fact 16 that in some specific instance there was a variation 17 somewhere in the process that perhaps shouldn't have 18 been there. As it turned out again it is hard in that 19 instance, and that is fine. It is the broader question 20 that we really have to focus on. Did tha t , whatever 21 caused that variation from the e xpected , so to speak per 22 other areas or other reason to say it didn' t occur, it 23 didn't matter, whatever. There is always th'e broader 24 question. ,
25 MR. ANDERSON: One more question. Cn the
. ALDERSoN REPoATING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTCN, D.C. 20024 (202) 554 2345
, - . .,.- --. c-.~,.- - m ~ - - ~ ~ . - - - -
.7 ,
127
'1 Brookhav.en pipe analysis, on the input vertical spectra
. .:2~ for'model B specifically, were the spectra peaks treated ,
a- . . .
' .3 'or were..they broadened or anything?
14 HR. BEZLER: They were broadened.
'5 NR. ANDERSON: By what?
6 ER. BEZLER: Plus or minus 10 percent. They 7 are shown 'in figures -- let se find the figures. Figure 8 8.4 on Page 155. And Figu.re 8.3 on Page 154 shows the 9 other. So reference to'those shows the broadening. 10 MB. ANDERSON: I~he*ard a little bit of 11 conflict, I think. Could I get.a repeat of the answer 12 on the procedure for broadening? 13 ER. BEZlER: Plus or minus 10 percent on the-14 peak. . 15 HR. SUBUDHI: According to Reg. Guide 1.100. 16 Tha t is 122. 17 MR. ANDERSON: Let me clarify one more point. 18 On the output for the piping problems, I am still a 19 little confused about how the spectra was created. 20 BR. BEZLER: Can I describe the procedure? l 21 ER. ANDERSON: Please. 22 MR..BEZLER4 A specific problem, problem 6-11, 23 had some 15 supports, including the anchors. 'Je would .\ 24 have requested Mike to supply us with the. spectra for . 25 each node of his grid that_essen11 ally corresponded to
' ALDERSCN REPCRTING COMPANY, INC, #
400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
- --,-ry--- --. - - - - - , - - - - - - -
_.; r, a.g_. . m_._ .. .2 o e
~
128
', 1 one of those important points, connections.
2 MR. ANDERSON: The raw spectra. 3 MR. BEZLER: Right. And then we would have 4 enveloped all of the spectra and once we had an envelope 5 ve broadened it. Now, am I clear, or is that clear' ? 6 HR. ANDERSON: Yes, it is.- Thank you.
~7, HR. COOPER: Is that done by the computer, or 8 'is this something that was done so it was raw, so that 9- 16 could be available?
10 , MR. BEZLER: The broadening was done by 11- computer. The others were done by hand. 12~ MR. DENISON: Were those four locations taken 13 regardless of whether they were vertical?
~
14 NR..SUBUDHI: I think we went to the support 15 drawings.to find which point is the nearest point. 16 58. DENISON: My question was, if you have 17 , 17 support attachment points, 14 of them may be vertical 18 restraints. 19 ER. BEZLER: Yes. Excuse me. The vertical 20 supports were considered in this approach. - ! 21 MR. DENISON: And the others were not? l 22 MR. .BEZLER : If the support was purely 23 ' horizontal, it was not included in this. 24 MR. DENISON: How did you take 1.nto account . 25 the torsional response? _
- ALDERSCN REPORTING CCMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024(202) 554 2345 ___ -.+m _ _ _ . _ , - _ _ _ _ . , ___ _ , - . ~ . - - _ . , . , . _ -,y-,
. a;. -- - .e -
- w. .
w . -. ,. :i 129 i 1 MR. BEZLER: We took no account of torsional 2 response other than.if it was demonstrated in Mike's 3 model, and I don't believe it was. So we are using 4 curve vertical. 5 MR. DENIS0Na Using the horizontal?
~
6' HR. BEZLER: We used the horizontal supplied 7 by PGCE, which includes a correction for torsion. 8 MR. WRAY: Ron Wray. Do you have the envelope 9 raw spectra in your broadened or smooth spectra for to these piping problems? 11 MR. BEZLER: I do not have them with me now. 12 The report shows you only the broadened, but we would 13 have the raw spectra if you will. Yes, we do. 14 HR . -C00'PER : Eike? - 15 HR. BUCHEST: How are nodal points combined? 16 NR. BEZLER : Nodal is combined -- direction 17 components. If you look through the report, we have 18 done a number of different types of analyses. We have 19 done analyses that correspond to your efforts which 20 require using an absolute combination on the direction 21 of componen ts, and then we also went and did wha t we 22 call a hard state of the art technique, which would be 23 an SRSS combination between the directional components 24 considering three components simultaneously. Is that , 25 enough answer? - ALCERSoN REPCRTING CCMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
; , 7 .. .77; .
a- - - . -
. - .- - .. =_. . . . _ .
- o. - . . ,
130
. 1 MR. BUCHESTa Yes.
2 MR. REICH: You can see that by the way in the 3 tables. 4 HR. BEZ1ER: The tables in the report which 5 summarize the forces, I think, spell this out. 6 MR. COOPER: Are there other questions or I 7 comments? 8 (No response.) 9 HR. COOPER: Our estimate is that about a 10 45-minute caucus would satisfy our needs. 11 MR. KNIGHT: Frank'just reminded me it might 12 be an opportune time to ask members of the intervening 13 parties if they would care to say anything at this 14 point. . 15 3R. FLEISCHE1ER: Why don't we wait until we 16 get back? 17 MR. KNIGHT: I would assume the larger body is
- 18 the TDS group.
l 19 MR. COOPER: '4 ell, tfier might caucus on their l 20 own. 21 MR. KNIGHTS Forty-five minutes w as the , 22 request. 23 MR. COOPER: That is right, sir. 24 MR. KNIGHT: Okay, uS minutes it.shall be. . 25 Can we agree to meet back h.ere at 2:30? __ ! (Whereupon, s brief racess was taken.) l
= ALCERSON REPoRDNG COMPANY INC.
l 400 VIRGINIA AVE., S.W , WASHINGTCN D.C. 20024 (202) 554 2345 __ _, , _ , _ _ _ . _ _ - . . . _ - _ ~ . - . _ - _ . . - - - . _ _ .
_ _ . _ _ _ _ _1 .- m 131 1 MR. KNIGHT: Gentlemen, may we reconvene? 2 Bill? 3 MR. COOPER: We have just two more questions. 4 The first pertains to Page 16. Page 16, the third and 5 fourth lines of that page, and Ray Ciatto. 6 NR. CIATTO: This should be a fairly j 7 straightforward question. The remaining three degrees 8 of freedom which are not excited with the vertical input 9 and therefore need not be considered, does this mean 10 that those degrees of freedom were restrained, or does 11 it simply mean that those degrees of freedom are not 12 dynamic degrees of freedom, but do have deflection l 13 associated with them? - I i 14 MR.. MILLER: Those degrees of f reedom a re 15 restrained. 16 MR. CIATTO: They are restrained. ~ 17 MR. MILLER: So there is no rotation abcut the 18 vertical axis, sni there is no displacement along the 19 horizontal plane. 20 MR. CIATTO: Okay. 21 MR. COOPER: And the other question pertains 22 to Page 153. .It is a question about several figures in l 23 tha t area, but the one on Page 153 we can us'e as an 24 illustration. Ron Wray can ask that one. 25 MR. WRAY: Ihe question is exactly how I i
. ALDERSCN REPCRTING CCMPANY,INC.
l 400 VIRGINIA AVE., S.W., W ASHINGICN, D.C. 20C24 GO2) 554 2345 l 7- . _ _ , , _ _ _ . - ,
[o . 132 1 Brookhaven would go about plotting the PGEE spectra 2 shown on Page 153. 3 ER. BEZLER: We had in the information package 4 supplied to us by PGEE a digitized record, and we just 5 plotted it. 6 MR. WRAY: You say it was a digitized record? 7 ER. BEZLER: A digitized spectrum. I believe 8 we had plotted it correctly. 9 HR. 4 RAY: The reason for my inquiry is that 10 the plot does not show any broadening. 11 MR. BEZLER: As I say, we plotted it directly 12 from a digitalized. 13 ER. WRAY: That was part of the transmittal? ( 14 MR..BEZLERs Correct. . 15 MR. COOPER: Would your computer, so to speak, 18 have curve plotted? Would a computer program h. ave given 17 you a peak even though it had been broadened slightly 18 below the peak .that you showed? 19 MR. B EZ LER : It should not have. These plots 20 are hand-drawn. These are not computer plo ts. 21 MR. COOPER: But the plotted points were dense 22 enough so that in the drawing it shouldn't have missed a 23 flattening. 24 53. BEZLER Th.at is correct. - 25 MR. COOPER: We don't,have any comment on the
\
- ALCERSON REPCRTING COMPANY, INC.
l 400 VIRGINIA AVE, S.W WASHINGTCN. D.C. 20024 (202) 554-2345 ' 7_ - , .
.--,-- --- _ _ _ _ _ _ - , - . - ~ - - _. ;_; 7_ ,,
__-___---___-_7___
---.-~ ,e ,
. =+ r- --
-1 133 1 'other two questions at the moment. I don 't know if 2 Brookhaven or staff has, or if ? GEE has. If not, Jim, I 3 think we have accomplished our purpose, and I think it 4 has been a very successful meeting, and I really 5 appreciate the cooperation and the kind of answers we 6 have gotten from the Brookhaven people. We have got 7 some good answers to our questions. Everybody has been 8 most cooperative. Our biggest concern now will be to 9 find ways in which we can access the information in kore 10 detail,and look at it, but we will work that out with 11 Hans in the normal manner, and I think we have 12 accomplished what we can do here.
13 HR. KNIGHT: Well, before we break up, let's ( 14 see, David, do you have anything ? . 15 HR. FLEISCHELER Jim, I wanted to clarify one is thing for the record, and that is the counsel on my 17 right has been retained for this meeting by the 18 Governor, not by the Joint Intervenors, and I have one 19 question. 20 I would like to direct the gentleman from 21 Brookhaven to Page 11 of their report, and under the 22 paragraph entitled Results of Piping Analysis, the third 23 sentence reads, "The evalua tions were carried out using 24 envelope response spectrum methods, indepe,ndent support . 25 motion response spectrum methods, and both PGCE and BNL _,
- ALCERSoN REPCRTING CCMPANY,INC, 400 VIRGIN!A AVE S.W., WASHINGTCN, D.C. 20C24 (202) 554 2345
- . . , , - . - . .-._ -.- ..-n.,-.-,
_ __ ~ _
.- __ __,,.n.,,,
ca-. _ :- F --- _:. - :- : ' ;
'e' ~e i
i i 134 I e t-i 1 developed spectra." My question is whether- the PGEE 2 developed spectra were the spectra in the URS Bloom 79 3 report or whether they were the new Hasgri 5 mass 4 spectra. l 5 HR. BEZLER: They were the new Hasgri 5 mass 6 spectra, or they were so titled. 7 NR. FLEISCHELER: Now, to follow that 8 question, at Paragraph 3, down at the bottom of the 9 page, it states that BNL support values obtained using 10 BNL models and PGCE supplied spectra do not match. Does 11 that again refer to the new Hasgri 5 mass spectra? 12 MR. BEZLERs Correct. 13 MR. FLEISCHELERs On Paragraph 4, you refer to
~
l' 14 the -- well, let me read the whole sentence.. " Support 15 forces calculated using BNL pi, ping models and BNL 3D l 16 model P envelope or independent spectra substantially . l 17 exceed PGCE calculated values." These calculated values l 18 were derived from the new Hasgri 5 mass spectra ? l l, 19 MR. BEZLER: The PGCE calculated values were l . 20 based on the revised Hasgri spectra, or what was l l 21 entitled 5 mass Hasgri spectra. 22 MR. FLEISCHELER: That is the only question I f 23 have. Thank you very much. The other thing I wanted to 24 raise , Jim , was, in the information transfer, we would 25 request that the informatio_n c,equests from Teledyne or
?
- ALCERSON REPCRTING CCMPANY, INC, 400 VIRGINIA AVE., S.W., WASHINGTCN, D.C. 20024 (202) 554 2345
_ _m ., __, m
_ . s, a' ~' . . ~ ~ -"^
- __ _ ~'~'L'L'
'. 2 7 .? ': :L1*- ?! n' . ._ :1 135 1 from FGCE to'the staff or BNL be in writing. That way 2 we could get access so we would know what kind of 3 information is being requested and make the decision as 4 to whether it would be useful for us to request that in 5 writing.
6 NR. MIRAGLIA: Back in October, when we were 7 in the single mode for the audit, what we agreed to was, _ 8 there would be an exchange of information and there 9 would b'e a listing of the information exchanged with the 10 understanding that the parties would pick and choose and 11 identify to us what specifics out of that-list th.ey 12 might want to have or see, and if that is acceptable, we 13 can go ahead with that kind of thing. s 14 MR. FLEISCHELERa' That sounds very good, 15 Frank. Thank you. 16 ER. KNIGHT 4 Mr. Friend? . 17 ER. FRIEND: We would like to comment we l 18 th,ought this was a good meeting to enable things to be 19 discussed that will be of importance to the indepen. dent 20 programs in evaluating the report. I have a couple of 21 questions that I would like to ask. First, during the 22 day today -- well, this goes to the point, I think, tha t 23 earlier in the day somebody said that this report would 24 be officially issued as a NUREG, and my questions are as _ 25 follovs. - l ALDERSON REPCRTING COMPANY. INC. 400 VIRGINIA AVE., S.W., WASHINGTCN. D.C. 20024 (202) 554 2345
., .-,~.-- ,- a
' ~
n
~
.1
. 3 I
136 1 Ihere were a number of things that were 2 discussed today, whether there were typos discovered and 3 things like that. Will the report being modified to 4 incorporate those things before it is issued? 5 MR. KNIGHT Yes, we will -- if we can catch 6 it before it gets out, it may ha"ve an errata sheet right 7 in the report, or corrections and an errata sheet, or 8 there will be an errata sheet. 9 MR. FRIEND Secondly, your opening remarks 10 were very useful in clarifying for my mind the purpose 11 of the report and the basis for the report. Would we 12 plan to put some prefacing remarks such as that in the 13 report, or in the covering document? - 14 MB. KNIGHT: Hon'estly, it had not-been up to 15 this moment. I for one am certainly willing to examine 16 the logistics of doing that, and I sense in the. 17 discussions that we had today that there seems to be a 18 broad opportunity for misuse and misunderstanding, which l 19 is unfortunate. At this point, the best I can do is to 20 make that commitment to examine the logistics of doinc ! 21 that, and I don't know. Let me just check with the 22 staff. 23 (Pause.) . 24 MR. KNIGHTS It is somewhere ver.y close to 25 being printed, and let me _just_come back. ALDERSCN REPCRTING CCMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
- -~ - ---n ._ .a -~_n - ,-_ - ,-....--- - .-. v -
a.,2 . - Su
.n_,,. _ .. _ _
137 1 ER. FRIENDS Those .are all- the questions we 2 had. 3 MR. KNIGHTS Dick? ', 4 MR. HUBBARDs We may'have some questions on 5 the BNL report, and if so, we will send them in writing 6 to the staff for transmittal to BNL. 7 ER. KNIGHT: Any further remarks? 8 (No response.) 9 HR. KNIGHT: Thank you, gentlemen. 10 (Rhereupon, at 2:44 p.m., the mee ting wa s 11 concluded.) 12 13
's 14 -
15 16 17 l 18 19 20 21 < 22 ? - 23 24 , 25 _. i
- ALCERSCN REPCRTING CCMPANY, INC, t
l - 400 VIRGINIA AVE., S.W., WASHihGTON, C C. 20024 (202) 554 2345 t _ = ,-- . - . m- ,am _ _ . _-
x --
= n: w = - . _ - _
q . . i l 1
. .gg--.g
. - _ e.au. .o s..v.
. CT. .*. * .*. a~ ~.r"si
~ . < ,... . .og
. .". .= - w . c ..u.
.. .. .. .. s. ..e.r . .' . a , n<' .. .- .
.a s. s. .'. . . .-. < - a. .r. .. .-
I. Nuclear Regulatory Commission in *:P.e :atta." cf: Diablo Canyon, Data cf ~~cceeding: July 27, 1982 1 CCCket llu:::'c er: ' 50-275
. lace of ."Octedi~.g:' Uoton, New York t
i
. 4 .e s.
.. u.
, u. ... . s e. E .,. u. s.e .. .a .
. a gr. . e e. r s , ==a .
...u. +. -us,.....s...,~. u. ,. r 4.a-e .. c.?. .......a..
. .. .. ,.. e u. . .r .s .,,c
. .c r .. ..u. C <
s.,.<.. .. . AISRED H. WARD Cf'icial :epertar (77 ed)
.., f A--
V, , s C . . . c .4. . .- : e. ..e-(.. i, l I I 4 i 4 e 1
- a. e N 4
{ I i
~
_ , - _ _ _ - _ r r T M ----- - 2 2 T ~J ~ --
~
, _2, ---?-- '
s, , . ,
. .- -..4- -- . ' sg ,- n7 l
Dicbio Ccnysn i Meeting July 27, 1982 . held at Brookhaven National Laboratory Attendance List H. Sebierling 3 M. Reich ' P. Morrill A. Philippacopoulos F. Miraglia P. Bezier J. Knight P.C. Wang > BNL P.T. Kuo > NRC
- C. Miller H. Polk J. Curreri M. Hartzman M. Subudhi j B. Jones >
f N _ W.E. Cooper H. Friend - B J. Hoch - PG&E D E. Denison B. Norton - PG&E I R.L. Cloud B. Lew - PG&E A S. Harris L. Malik - URS/B B . IDVP*4 P. Anderson Wm. White - B L R. Wray M. Tressler - PG&E EO M. Holley L. Shipley - B p J. Biggs T. Esselman - @ R
* '**U R. Locke - PG&E S. Chin O
R. Fray - PG&E P. Hirschberg - PG&E b-
)
C ' T R. Hubbard of A. R. Dynner (G. Brown Calif. Carl. M. Cannon - San Jose Mercury News M.I.T. E. Kausel Joint Intervenors - D. Fleischeler O e en e M
;, - --o m e -r -wee. -- m- en-*- .v -- .m w re -- - - . - . -e -.--e- *-e %,-%_q=, . .m .
m ,
- ;w -6: . - . _.. . . + .-
"dPTF1 AVNE ENGINEERING SERVICES l NRC-BNL-IDVP-DCP MEETING OF JULY 27, 1982 l :.
( w {0 0lW I TES.0BJECTIVE: [74 pt, To exghange information concerning the BNL Report which pertains to j the IDVP review being conducted in response to the July 1, 1982 letter from
~ NRC (Denton) to TES (Cooper).
PROCEDURE: i On each portion of the BNL report which the IDVP wishes to discuss:
- 1. TES (Cooper) will identify the specific content under discussion and summarize our concern.
- 2. BNL is invited to supplement the written report.
- 3. NRC is invited to make supplementary comments.
- 4. IDVP participants will ask questions directed to BNL and BNL will be invited to respond.
t
- 5. The Diablo Canyon Project Team is invited to mske any observa-tions which they believe will assist the IDVP.
i 6. Closing comments will be invited from, in sequence: - BNL, NRC, t DCP, IDVP. GROUND RULES:
- 1. This is a "what", not a "why", meeting. .
- 2. The meeting was requested by TES to obtain clarification of the
~
report.
! 3. No conclusions are to be stated or implied.
l f
,wwo-*-=* p.,ses--,~= -m*.--m,y +~=+w------e.--+- ---=-.**=^*m-*m+*+-w=~=v - - - - e
-% 7
r .,s .
.m_ _ . .
,a Mu?aicL SNL BNL NEVER CONDUCTED AN ANALYSIS TO DETERMINE WHETHER l$af I
THE DIABLO CANYON PLANT WAS SAFE OR UNSAFE. SINCE WE Q. HAVENOTCONSID5REDTHISQUESTION,WEHOLDNOVIEW PERTAINING TO IT. e BROOKHAVEN NATIONAL LABORATORY l} g)l A5500ATED UNIVERSITIES, INC.(1 El1
- - . . , - - . = - . _..
~
_ a. , .c y - a < i ,
- in . - u
]
A h
! 3 ,, 7 1- o 8 WEHAVECOMPAREDOURFLOORSPECTRAWITHTSOSEOF n
PG&E AND THEIR CONTRACTORS, AND HAVE:s0UND PLACES - WHERE WE AGREE AND WHERE WE DISAGREE.I THERE ARE i L
' 1 r
^
IN FACT CASES' WHERE PG&E RESilLTS ARE'liORE CONSER-VATIVEANDTHEREAREPLACES10NTHESTRUCTUREWHERE T
; THEY ARE NOT,, i-
, ~ i ,
s 8
\
?
e i .
- /-
l , , i l l . l BROOAHAVEHATIONAL LABORATORY ]) g 3 l A5500ATED ljNIVERSITIES, INC. (I ll l 4 e e
; u m. .-,-
.y e-w
' ~ '
' ~
[ ~
. TASK OUTLINE . .? ' .
2
- 3 m (1) REQUESTED TO ATTEND OCTOBER 9, OCTOBER 9, 198r BETHESDA, MD, MEETING NMERE.
!. j { PGEE AND CONTRACTORS DI.SCUSSED THE S0-CALLED DIABLO CANYON UNIT 1. ?
E55 " DIAGRAM ERROR", h 5*" ,
"'(2) PARTICIPATE WITH NRC AT DIABLO CANYON AUDIT HELD AT PG&E SAN FRANCISCO HEADQUARTERS DURING THE PERIOD OCT.0BER 14-16, 1981.
(3) PRESENT COMMENTS PERTAlllING TO PG&E AUDIT TO NRC,
- i ._.
ll u (4) REQUESTED TO INDEPENDENTLY DEVELOP VERTICAL FLOOR RESPONSE SPECTRA FOR UNIT 1 il lll gh, CONTAiNMENTANNULUSSTRUCTURE. MODEL A (SHEAR JOINTS FOR ALL BEAM AND COLUMN Et$$
-a g i-
, CONNECTIONS), AND MODEL B (M0 MENT CONNECTIONS FOR 1sT AND 2ND FLOORS). ..
($)& REQUESTED TO INDEPENDENTLY REANALYZE PIPING SYSTEMS WITH PG&E DESIGNATIONS' O (G) NUMBERS 4A-26 AND 6-11. - a .
- 17) CARRY OUT CHECK CALCULATION F0R FLOOR SPECTRA USING GENERAL PUBLIC . COMPUTER 4 CODE, .
- i. .
0. g BROOKHAVEN NATIONAL LABORATORY l} g) l , j A5500ATED UNIVER5lilES, INC.(IllI
. TASK OUTLINE ' CONT'D)' '
- i ..
i" J (8) REPEAT. TASK (4) FOR MODEL C (WHERE BEAM TO COLUMN CONNECTIONS FOR 1sT, 2ND j AllD 3RD FLOOR ARE CONSIDERED RIGID)'. !! t . j _, (9) CARRY OUT CONFIRMATORY COMPUTER RUN FOR ORIGINAL PG&E 2-D H0 DEL. a a u. 5 P s'[m E a
$ (10)&
< CARRY OUT MULTIPLE INPUT ANALYSIS FOR PG&E PIPING SYSTEMS 4A-26 AND 6-11
?m""{* (11) USING SPECTRAL IMPUT FROM:3-D ANALYSIS AND EVALUATE IN ACCORDANCE WITH T , ASME CLASS 2 CLASSIFICATIONS. - h !! . (12) VERIFY RESULTS OF THE SPECTRA FOR 2-D MODEL WITil SPECTRA GENERATED FROM
~
STRUDL (McDONflELL DOUGLAS). 'i. 7 I m 3 r eo p (13) EXTEND 2-D STUDY = VARYING THE B0UNDARY CONDITIONS.
~
W u E !L D "c i
" " $[ (14h REVIEW INPUT /0UTPUT 0F URS/BLUME 1979 RUN.
USE IDENTICAL MODEL AND H CONPARE RAW FLOOR SPECTRA. i g- ' j [(15) _ e WRITE REPORT DETAILING RESULTS OF ITEMS 4 To 14.
-l c
it e 4 - t . BR00XHAVEN NATL 0 NIL LABORATO A5500ATED UNIVERSITIES, INC 0 L.Il n ._ 1 -
. c. _ _ _ _ _ . _Z _
. - - m RESULTS OF 3-D MODEL STUDIES I FLOOR ~ RESPONSE SPECTRA GENERATED WITH THESE MODELS DID NOT AGREE WITH THOSE GIVEN IN THE MAY 1979 URS/BLUME REPORT. BOTH FREQUENCY SHIFTS AND DIFFERENCES IN PEAK SPECTRAL ACCELERATION MAGNITUDES WERE FOUND. --
0 THIS WAS THE CASE FOR ALL MODELS (I.E., A,B AND C) EVALUATED. 8 USR/BLUME RESULTS FOR THE TOP FLOOR WERE CONSISTENTLY
~
CONSERVATIVE IN THAT.THEY EXCEEDED THE. ACCELERATION MAGNITUDES PREDICTED WITH THE BNL MODELS. s THIS, HOWEVER, IS NOT THE CASE FOR FLOORS ONE, TWO-AND THREE. FOR SOME FREQUENCIES THE USR/BLUME RESULTS FOR THESE FLOORS WERE CONSERVATIVE, FOR OTHERS THEY WERE NOT, WITH NO OBSERVABLE TREND. O 6 W MM gm,,,g; ,gyg.at U20;'moPJD Assow.re ur:mr.5mts, nic. G
- ~ , . . .
..J t . -I RESULTS OF PIPING ANALYSIS O CONFIRMATORY EVALUATIONS WERE PERFORMED FOR PG8E - ! PIPING PROBLEM NUMBERS 6-11 AND I4A-26.
I i e THE EVALUATIONS WERE CARRIED OUT USING ENVELOPE RESPONSE SPECTRUM METHODS, INDEPENDENT SUPPORT MOTION RESPONSE SPECTRUM METHODS, AND BOT}l PG&E AND BNL DEVELOPED SPECTRA.
-8 THE PG8E SUPPLIED SPECTRA WERE ENTITLED "NEW HOSGRI-5 MASS SPECTRA", A CHECK SHOWED THAT THESE SPECTRA ARE.
DIFFERENT FROM THOSE PRESENTED IN THE URS/BLUME 1979 REPORT. W 6 6 ! . _g_ snoomat wIon usamoar9 AssoanID Univa5mES, INC U
- --,.----,_,-, __._a,.---.. , . . - - . . . - . -
~
__ c.- 1 - :
'.^ _. Ja 9
RESULTS OF PIPING ANALYSIS (CONT'D) . 8 BNL MODELS DEVELOPED FROM PG&E AS-BUILT DPAWINGS WERE FOUND TO DIFFER FROM Tile PG&E MODELS, i e THE DIFFERENCES WERE DUE TO THE USE BY PGEE OF DESIGN DIMENSIONS WHICH DIFFER FROM THE AS-BUILT DIMENSIONS AND IN ERRORS MADE BY PG&E IN Tile MODELING 0F PIPE BENDS. I AN OVERLAP PROCEDURE WAS USED IN THE MODELING OF PROBLEM 14A-26', THE EXTENT OF OVERLAP USED IN THE PROBLEM SEEMS ADEQUATE IN THAT IT MEETS.THE INTENT OF NUREG/CR 1980, l l \ i _7_ BP.00)TAVDI I Ail 0 !Al LA307. ATOP.Y[3 , ASSOCIATED LJ!!IYL.P.5 FILES, l'!C. (I i
- _ - . ~_a
~
RESULTS OF PIPING ANALYSIS (CONT'D) I BNL PREDICTIONS OF SYSTEM FREQUENCIES DIFFER FROM THE PG&E ESTIMATES, O BNL SUPPORT FORCE VALUES OBTAINED USING BNL MODELS e AND PG&E SUPPLIED SPECTRA DO NOT MATCH. THE DIFFERENCES ARE PROBABLY DUE TO THE DIFFERENCES IN _ MODELING. O SUPPORT FORCES CALCULATED USING BNL PIPING MODELS AND BNL 3-D MODEL B ENVELOPE OR INDEPENDENT SPECTRA EXCEED PG&E CALCULATED VALUES. THE MAJOR CAUSE FOR THIS IS THAT MODEL B SPECTRA EXCEED THE SPECTRA USED BY PG&E. BROOKHAVEN NATIONAL LABORATORY l) g)l A5500ATED UNIVERSITIES, itK. (llli
. -8_
- ,_ n ., . -
=. - _ . - _ _ . . . - _ - , . . _ - .- -
. _ = .- ... ,,
r
$ 0 J
RESULTS OF PIPING ANALYSIS (CONT'D) I I ASMECLASS2EVALUATIONSPERF6RMEDUSINGTHEUNIFORM RESPONSE SPECTRUM METHOD INDICATED NON-COMPLIANCE ,WITH EQUATION 9 FOR 3 POINTS IN PROBLEM 6-11, WHILE PROBLEM 4A-26 MET ALL ASME CRITERIA. s ASME CLASS 2 EVALUATIONS PERFORMED USING THE INDEPENDENT SUPPORT RESPONSE SPECTRUM METHODS PRODUCED A REDUCTION IN A RESS LEVELS IN PROBLEM 6-11, BUT AN INCREASE IN STRESS LEVELS FOR PROBLEM 4A-26, i I FOR THIS PROCEDURE, PROBLEM 6-11 SHOWS OVERSTRESSING AT ONE POINT ANIj 4A-26 AT THREE POINTS. ' l}
~ ~
BROD)RAVDi ?! Ail 0:Ud U20?JJO?.YJ) l /d50CIATID U!MR.5fTIE.5, !!!C. O e _ . _ . - _ _ . _ . . - _ . . _ _ . _ _ . -.. _. _
~
RESULTS 0F 2-D MODEL STUDIES
's 0 DUE TO UNCERTAINTIES.IN SOME'0F THE PERTINENT DATA, VARIOUS PARAf;1ETRIC STUDIES FOR THIS MODEL WERE PERFORMED. NONE OF THESE, HOWEVER, CORRELATED WELL WITH'THE URS/BLUME RESULTS GIVEN IN THEIR MAY 1979 REPORT. -
8 WE THUS REQUESTED THAT NRC OBTAIN A LISTING OF . THE COMPUTER INPUT /0UTPUT FOR THE RUNS USED T0, GENERATE THE SPECTRA GIVEN IN THE URS/BLUME REPORT. 8 THIS INFORMATION WAS RELAYED TO~US BY PG&E ON APRIL 214, 1982. i
~
BPOODRVDI !!ATIO Ud U.EOP.A10?.Y,') nsoaAim U::lvf.r.smb, it:t G _.,.__;_..a_...-,_. _
, _ _ _ _ . _ _ , . . _ _ _ . , , , . , _ .._._ __m..,._.,__
. L.- ==
~ .
RESULTS OF 2-D MODEL STUDIES (CONT'D) 8 A CONFIRMATORY BNL COMPUTER RUN WITH INPUT DATA IDENTICAL TO THAT USED BY URS/BLUME YIELDED RAW FLOOR SPECTRA SIMILAR TO THOSE SENT TO BNL (IN DIGITIZED FORMAT).BY PG&E ON APRIL 20 1982 8 THE BROADENED 1SPECTRATASSOCIATED WITH THE _ STRUCTURAL FREQUENCIES PRESENTED IN THE MAY 1979 REPORT CORRESPOND WITH'THE RAW SPECTRA VALUES SENT TO US IN APRIL 1982. IN THE LOWER SPECTRA FREQUENCY RANGE, IT SEEMS THAT THE SMOOTHED SPECTRA WERE OBTAINED BY.THE USE OF RAW SPECTRA VALUES. BROOKHAVEN Nail 0NAL LABORATORY l} g)l A5500ATED UNIVERSITIES, INC.(Ill1 _ 11 _ e-- we- y -_v = _ -t e 7 g--+=- e.g e ye - , =y-= .-===---_ --e+- ;* w - --
**; es 7,
, _ me ; y e -s . . s. arm _emm
= = .
- = _- . . .- =ne b hh.'('i pga -
COMPARISON OF BHL 3D MODEL FLOOR RESPONSE Cor m [n-SPECTRA WITH CORRESPONDING PGSE SPECTRA:
}
; e FAN COOLER LOCATIONS 0 AVERAGED OVER FLOOR SECTIONS 0 ENVELOPED OVER FLOOR SECTIONS PRESENTATION OF COMPAP,ISON RESULTS BY TEST MATRIX BROOKHAVEN NATIONAL LABORATORY l} g)l A5500ATED UNIVERSITIES, INC.(IllI
,w ww= - ++ ,-p..< , --=*7 e-;-=-_----, --r*=m-- ,** *e6-. we --M_ *****'t "r- * *
- 7t'_"T ' Ni% v ~~.'
.._.____..-o.
UNIT 1 '2 NORTH TNT 2
@ O. . ,
CCWATAfAiENT CONTAMENT . O O
@ CRAPE WALL CRAME WALL -
h '
's, .
/
J- 4Cf ' I' '-'10Cf 18 0 !# ' g 1.__ eU - y]< -T(( p [ +N, -pr- e g
? ,
1
, t ,. ~.' @ @ .
A. i - O I i @ @ a __ i
- l-
- l N Un .
l h g i >. crf # 8 I 9ph j' i 4[ f O L. amR Na - 7 i Orientation of fan coolers
.. for Units I and 2 and frame -
I orientation of IIAS/Glume 20 model, d l j , W
-'- ~BRDOKHAVEN hall 0NAL lABORAIUKY l} l] l ,-
A5500ATED UNIVERSITIES, INC.(Illl l I
;q il
~
s~ __~ T. _
. - .:_ . sp: :.T
'6 e
. I -
Jb s e - A NORTH 10 s
+g e ,,c, s
s oo + O
's..s p
's
- 18. s * / 's
\,
e O s
-n d c 360
. _ . ._ . v -.. 4 s N ,
af EWLCCLQLEE--E-~~~~~~~~#Y'~ e' *
'M','~..,'
NC 2 o Z 'g *
*o '
w o ~Em e' ph- _ ,s' -
/
4 -
'id'# .
4g 2su 3 DIABLO CANYON PLANT
- UNIT 1
@ CONTAINMENT MAULUS STRU~.ILRE DIAGRAM FOR SPECTRA APPUCATION l Definition of sections for a k (APPUCABLE TO ALL FOUR F
} typical floor of the containmentgm' annulus structure Unit 1. p *
. . o t
l - BROOKHAVEN NATIONAL LABORATORY l} g3l A5500ATED UNIVERSITIES, INC.(I til
c-
., c COMPARISON MATRIX COMPARIS0!f TYPE: ATFANCOOLERLOChTIONS t
FAN . COOLER - FLOOR 1 Og 1 1 1 1 2 Op 1 O pp 1 , O pp 3 O gg O gg Op 1 Og 4 1 1 1 1 -1 l ROIE: 1 - ACCEPTABLE O - NON-ACCEPTABLE P - DIFFERENCES IN PEAK VALUES . F - FREQUENCY SHIFTS - PF - BOTH BROOKHAVEN NATIONAL LABORATORY l} g)l ASSOCIATED UNIVERSITIES, INC.(Illl
-r -
g , _
- - _ _ : - _ _ _ . = . z =.=wa - -
t . COMPARISON MATRIX COMPARISON TYPE: AVE. RAGE SPECTRA ECTION 1 2 3 4 5 FLOOR 1 1 1 1 1 1 2 1 1 O pp 1 O pp - 3 O O gg Og 1 Og PF 4 1 1 ~1 1 1 tiDIE: 1 - ACCEPTABLE 0 - NON-ACCEPTABLE P - DIFFERENCES IN PEAK VALUES _ F - FREQUENCY SHIFTS - PF - BOTH - BROOKHAVEN NATIONAL LABORATORY [][][ A5500ATED UNIVERSITIES, INC.(IllI
,w w w-m - g ir'- w
_- w . - c.. e I C'OMPARISON MATRIX COMPARISON -TYPE: ENVELOPE SPECTRA SECTION . . 1 2 3 4 5 FLOOR 1 Op 1 O gg O pp O gg j 2 . Op O pp O. p O g. O gg, t i 3 O O gg O gg Og Og PF , 4 1 1 1 1 1 Elf,: 1 - ACCEPTABLE O - NON-ACCEPTABLE P - DIFFERENCES IN PEAK VALUES . F - FREQUENCY SHIFTS - PF - BOTH
,' BROOKHAVEN NATIONAL LABORATORY l} g)l A5500ATED UNIVERSITIES, INC.(lll!
a
- s
~~v3 a- M g ;" w m -
_ --- M,_ ~- g - --W' ,9- W we
.= x : _
a c{. - RESULTS FROM 3D EVALUATION O THE FLOOR RESPONSE SPECTRA FROM THE 3D MODEL DO NOT AGREE WITH THOSE GIVEN IN MAY 1979 URS/BLUME REPORT. t URS/BLUME FLOOR RESPONSE SPECTRA FOR THE TOP FLOOR WHERE THE FIVE FAN COOLERS ARE LOCATED WERE FOUND TO BE CONSISTENTLT CONSERVATIVE. 8 FOR THE FIRST, SECOND AND THIRD FLOORS THE URS/BLUNE RESULTS FOR SOME CASES WERE FOUND CONSERVATIVE; FOR OTHERS WERE NOT3 WITH NO OBSERVABLE TREND. 8 FLOOR RESPONSE SPECTRA GENERATED BY THE BNL 3D MODEL WERE-FOUND IN GOOD AGREEMENT WITH SPECTRA OBTAINED WITH THE STRUDL~ CODE OF THE McDONNELL DOUGLAS. BR00XHAVEN NATIONAL LABORATORY l} g)l A5500ATED UNIVERSITIES, INC.(I(11
.2m - ,,- - - - - . , -yo-w med y _
r-
-- -- ~-
h . Pl,.!:;pr coneulm 3D MODEL: MODE NO. H0DAL FREQUENCIES _ 07 FREQUENCY (CPSL 1
- 9.9 2
. .11.2 3
11,4
- 4 11.8 5
1 12.0 - 6 12.1 7 12,2 j 8
. 12,9 9
13.0 10 13,1 11-13,2 12 13,5 13 13,6 14 13.8 15 13.9 MODEL B I l
~BRoomAVEN Hall 0NAl UBORATORY[} p 1
Ass 00ATED UNIVERSITIES, INC. O llI { . . - . - . _~- - - ~ ~ ~ -' a v- ' ~ a
.__: }
h w ./ r- -
$ e I
y s, - g - 1 h e N M z m cn rdD 4 xo a n, E m 2 %
,V
.3 v m m %
d' > RJ G 8m N l In g 0 @ h
'eM m b E
<r 4 L -- J g -
i l
,N -
/ In
/ n I h
/
n
* /
e 3 ~ w / T
/
r
/ U
/
/
, /
/
in / g
*s8 / ,1 m
M 4 e 9 ,~ e 9 0 G
- w{
.. l w
l'
- -}}