ML20065A625
| ML20065A625 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 08/31/1982 |
| From: | Bezler P, Brown P, Chris Miller, Philippacopoulo, Reich M, Shteyngart S, Subudhi M, Wang Y BROOKHAVEN NATIONAL LABORATORY |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| BNL-NUREG-51566, NUREG-CR-2834, NUDOCS 8209130289 | |
| Download: ML20065A625 (177) | |
Text
_ _ _ _.
- BNL-NUREG-51566 Independent Seismic Evaluation of the Diablo Canyon Unit 1 Containment Annulus Structure and Selected Piping Systems Prepared by A. J. Philippacopoulos, M. Reich, P. Bezier, C. Miller, Y. K. Wang, M. Subudhi, S. Shteyngaq, P. Brown Brookhaven National Laboratory uclear Regulatory Commission
!![kDo!Ko!$0$275 P PDR
l l
NOTICE This report was prepared as an account of work sponsored by an acency of the United States Government. Neither the United States Government not any agency thereof, of any of their employees, makes any warranty, expressed or imphed, or assumes any legal liability of re-sponsibihty for any third party's use, or the results of such use, of any information, apparatus, product or process disclosed in this report, or represents that sts use by such third party would not infringe privately owned rights.
Availability of Reference Materials Cited in NRC Publications Most documents cited in NRC publications will be available from one of the following sources:
- 1. The NRC Public Document Room,1717 H Street, N.W.
Washington, DC 20555
- 2. The N RC/GPO Sales Program, U.S. Nuclear Regulatory Commission, Washington, DC 20555
- 3. The National Technical Information Service, Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publications, it is not intended to be exhaustive.
Referenced documents available for inspection and copying for a fee from the NRC Public Docu-ment Room include NRC correspondence and ir.ternal NRC memoranda; NRC Of fice of Inspection and Enforcement bulletins, circulars, information notices, inspection and investigation notices; Licensee Event Reports; vendor reports and correspondence; Commission papers; and applicant and licensee documents and correspondence.
The following documents in the NUREG series are available for purchase from the NRC/GPO Sales Program: formal NRC staff and contractor reports, NRC sponsored conference proceedings, and N RC booklets and brochures. Also available are Regulatory Guides, NRC regulations in the Code of Federal Regulations, and Nuclear Regulatory Commission Issuances.
Documents available from the National Technical Information Service include NUREG series reports and technical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.
Documents available from public and special technical libraries include all open literature items, such as books, journal and periodical articles, and transactions. Federal Register notices, federal and state legislation, and congressional reports can usually be obtained from these libraries.
Dncuments such as theses, dissertations, foreign reports and translations, and non-NRC conf erence
,erscecdings are available for purct'ase from the organization sponsoring the pubhcation cited.
Single copies of NRC draft reports are available free upon written request to the Division of Tech-nical Information and Document Control U S. Nuclear Regulatory Commission, Washington, DC 20555.
Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 Norfolk Avenue, Bethesda, Maryland, and are available there for reference use by the public. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards Institute,1430 Broadway, New York, NY 10018.
GPO Printed copy price $6.50 i
NUREG/CR-2834 BNL-NUREG-51566 Independent Seismic Evaluation of the Diablo Canyon Unit 1 Containment Annulus Structure and Selected Piping Systems Manuscript Completed: May 1982 Date Published: August 1982 Prepared by A. J. Philippacopoulos, M. P.nch, P. Bezier, C. Miller, Y. K. Wang, M. Subudhi, S. Shteyngart, P. Brown Department of Nuclear Energy Brookhaven National Laboratory Upton, NY 11973 Pr: pared for Division of Engineering Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Cummission W shington, D.C. 20555 NRC FIN A3371
ABSTRACT An independent review and development of the vertical floor spectra for the Unit 1 containment annulus structure of the Diablo Canyon Power Plant was carri A out using a detailed three-dimensional model. The developed floor spectra were then utilized for confinnatory evaluations of two selected piping syst ems. The latter were evaluated by the envelope response spectrum method, and by the independent support motion response spectrum method. ASME class 2 evaluations of the two systems were also performed. Finally, a confirmatory evaluation was carried out for the model utilized by URS/Blume for the development of the vertical floor response spectra that were reported in reference (1). Sections 1.1 and 1.2 of the report summarize the work scope and the results of the study. Details pertaining to the specific areas of the work are given in sections 2 to 8.
-iii-
Acknowledgements The authors would like to express their gratitude to the NRC staff menbers for their assistance on various phases of this work. In particular to Dr. P. T. Kuo who was our monitor in all phases of this work and to Dr. M.
Hartzman who monitored our ef forts for the MEB. Their constructive comments and advice during the course of this work is deeply appreciated.
We would also like to thank various members of the Structural Analysis Division who contributed towards the completion of this work. The assistance of Ms. E. Gilbert and Ms. J. Murray in typing this manuscript is also -
appreci ated .
-iv-
Table of Contents Abstract ........................................................ iii Acknowl edgements ................................................ iv List of Tables .................................................. vii Li s t o f F i g ur e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . n11 1.0 Summa ry of Wo rk Sc ope a nd Syn op si s o f An al ysi s Re s ul t s . . . . . . . . . . . I 1.1. BNL Diablo Canyon Work Scope .................................... 2
- 1. 2 Summary of Results .............................................. 10
2.0 DESCRIPTION
OF STRUCTURAL MATHEMATICAL MODEL .................... 15 2.1 3-D Model and Element Data ...................................... 15 2.2 Mass Distribution ............................................... 17 2.3 Member Connectivity ............................................. 18 3.0 STRUCTURAL EVALUATIONS .......................................... 25 3.1 Free Vibration Analysis ......................................... 25 3.2 Input Time History .............................................. 28 3.3 Generation of Vertical Floor Response Spectra ................... 28 4.0 VERIFICATION OF THE 3-D MODEL ................................... 43 4.1 Comparative Study ............................................... 43 4.2 Description of Results .......................................... 44
5.0 DESCRIPTION
OF FLOOR RESPONSE SPECTRA FROM 3-D MODEL ............ 47 5.1 General ......................................................... 47 5.2 Methods of Comparison ........................................... 47 5.3 Comparison with URS/Blume Spectra ............................... 48 6.0 TWO-DIMENSIONAL MODEL EVALUATIONS ............................... 96 6.1 General ......................................................... 96 6.2 Verification of Floor Spectra Results ........................... 96 6.3 Results from URS/Blume Input Data ............................... 99 6.4 Comparison Between Raw and Broadened Spectra .................... 100
-y_
l I
I s,
Table of Contents (cont'd) _
f~
7.0 DESCRIPTION
OF PIPING MATHEMATICAL MODEL ...................... 125 ' [
7.1 Problem 6-11 .................................................. 125
- 7. 2 Problem 4A-26 ................................................. 126 4 8.0 PIPING EVALUATIONS ............................................ 132 _
8.1 Frequency Comparison .......................................... 132 8.2 Results Based on Uni form Support Excitation . . . . . . . . . . . . . . . .... 133 8.2.1 Force Comparison .............................................. 134 8.2.2 Stress Results ................................................ 136 _
8.3 Resul t s Ba sed on Independent Suppo rt Exci tation . . . . . . . . . . . . . . . 133 8.3.1 Force Compari son .............................................. 139 8.3.2 Stress Results ................................................ 140 References .................................................... 165
, N. .
l t
w m
e 1
_if i .
k LIST OF TABLES Table Page 1 Task Outline ............................................ 9 3.1 Model-Cases Evaluated for Different Boundary Conditions ............................................ 26 3.2 Modal Frequencies ....................................... 30 4.1 Comparison of Ten Modal Frequencies ..................... 45 5.1 Mass Evaluations ........................................ 50 4
6.1 ~20 Model. Comparison of Modal Frequencies (SAPV versus STRUDL) ............................................... 101 6.2 2D Model. Comparison of Modal Frequencies (SAPV versus URS/Blume) ............................................ 108 8.1 Problem 4A-26. Frequency Comparison .................... 143 8.2 Problem 6-11. Frequency Comparison ..................... 144 8.3 Problem 6-11. Support Force Comparison. PG&E. Spectra .. 145 i
8.4 Problem 4A-25. Support Force Comparison. PG&E Spectra ............................................... 146 8.5 Problem 6-11. Problem 6-11. Support Comparison. Model i B Spectra ............................................. 147 8.6 Problem 4A-26. Support Force Comparison. Model B Spectra ................................. ............. 148 8.7 . Problem 6-11. Support Force Comparison. Model B Spectra ...............................................
149 8.8 Problem 4A-2A. Support Force Comparison. Model B Spectra ............................................... 150 8.9 ASME Class 2 Equation 9 Satisfaction .................... 151
-vit-
LIST OF FfGURES Figure Page 19 2.1 Schematic View of Containment Annulus Structure ........
2.2 Three-Dimensional Finite Element Model ................. 20 2.3 Model Details for Floor at Elevation 101' .............. 21 2.4 Model Details for Floor at Elevation 106' .............. 22 2.5 Model Details for Floor at Elevation 117' .............. 23 2.6 Model Details for Floor at Elevation 140' .............. 24 3.1 Modal Shape at Frequency 11.8 cps ...................... 34 3.2 Modal Shape at Frequency 13.0 cps ...................... 35 3.3 Time History Record of Input Excitation ................ 36 3.4 Evaluation of Int.egration Time Step. Node 152 ......... 37 3.5 Evaluation of Integration Time Step. Node 153 ......... 38 3.6 Evaluation of Integration Time Step. Node 157 ......... 39 3.7 Evaluation of Integration Time Step. Node 163 ......... 40 3.8 Evaluation of Integration Time Step. Node 166 ......... 41 3.9 Evaluation of Integration Time Step. Node 281 ......... 42 4.1 Floor Response Spectra Comparison: SAPV (BNL) versus STRUDL (Mc Donnell Douglas) .......................... 46 5.1 Orientation of Fan Coolers for Units 1 and 2 and Frame Orientation of URS/Blume 2D Model .................... 54 5.2 Definition of Sections for a Typical Floor of the Containment Annulus Structure Unit 1 ................. 55 6.1 URS/Blume Mathematical Model of Containment Annulus Structure ............................................ 97 6.2 Simplified Model of Containment Annulus Structure ...... 98 6.3 20 Model. SAPV (BNL) versus STRUDL (Mc Donnell Douglas)
Node 20 .............................................. 102 6.4 2D Model. SAPV (BNL) versus STRUDL (Mc Donnell Douglas)
Node 41 .............................................. 103 6.5 2D Model. SAPV (BNL) versus STRUDL (Mc Donnell Douglas)
Node 42 .............................................. 104 6.6 2D Model. SAPV (BNL) versus STRUDL'(Mc Donnell Douglas)
Node 43 .............................................. 105 6.7 2D Model. SAPV (BNL) versus STRUDL (Mc Donnell Douglas)
Node 44 .............................................. 106 6.8 20 Model. SAPV (BNL) versus STRUDL (Mc Donnell Douglas)
Node 45 .............................................. 107
-viii-
LIST OF FIGURES (Cont'd)
Figure Page 6.9 2D Hodel. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 8 ..................................... 109 6.10 2D Model. SAPV (BNL) versus URS/Blume Floor Response i
Spectra. Node 9 ..................................... 110 6.11 2D Model. SAPV (BNL) versus URS/Blume Floor Response
- Spectra. Node 10 .................................... 111 6.12 2D Model. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 11 .................................... 112 6.13 2D Model. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 12 .................................... 113 6.14 2D Model. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 19 ....................................
Il4 6.15 2D Model. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 30 ....................................
115 6.16 2D Model. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 41 .................................... 116 6.17 20 Model. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 8 ..................................... 117 6.18 2D Model. SAPV (BNL) versus URS/Blume Floor Response Spectra. Node 9 ..................................... 118 6.19 2D Model. Raw versus Broadened URS/Blume Floor Response 5pectra. Node 10 .................................... 119 6.20 2D Model. Raw versus Broadened URS/Blume Floor Response Spectra. Node 11 .................................... 120 6.21 2D Model. Raw versus Broadened URS/Blume Floor Response Spectra. Node 12 .................................... 121 6.22 20 Model. Raw versus Broadened URS/Blume Floor Response Spectra. Node 19 .................................... 122 6.23 2D Model. Raw versus Broadened URS/Blume Floor Response -
Spectra. Node 30 .................................... 123 6.24 2D Model. Raw versus Broadened URS/Blume Floor Response Spectra. Node 41 .................................... 124
-ix-
LIST OF FIGURES (Cont'd)
Figure Page 7.1 Isometric for Problem No. 6-11 .........................
128 7.2 Schematic View of Problem 6-11 ......................... 129 7.3 Isometric View of Problem 4A-26 ........................ 130 7.4 Schematic View of Problem 4A-26 ........................ 131 8.1 Problem 6-11. Envelope for 2% Damping ................. 152 8.2 Problem 4A-26. Envelope for 2% Damping ................ 153 8.3 Problem 6-11. Envelope for 2% Damping. Model B ....... 154 8.4 Problem 4A-26. Envelope for 2% Damping. Model B ...... 155 8.5 Problem 6-11. Model B. Group 1. Spectrum ............ 156 8.6 Problem 6-11. Model B. Group 2. Spectrum ............ 157 8.7 Problem 6-11. Model B. Group 3. Spectrum ............ 158 8.8 Problem 6-11. Model B. Group 4. Spectrum ............ 159 8.9 Problem 4A-26. Model B. Group 1. Spectrum ........... 160 8.10 Problem 4A-26. Model B. Group 2. Spectrum ........... 161 8.11 Problem 4A-26. Model B. Group 3. Spectrum ........... 162 8.12 Problem 4A-26. Model B. Group 4. Spectrum ........... 163 8.13 PG&E Horizontal Spectra for Problems 6-11 and 4A-26 .... 164 l
l l
-x-
p-1.0 Summary cf Work Scope and Synopsis of Analysis Results l This report describes a study performed by the Brookhaven National Laboratory for the Diablo Canyon Plant of the Pacific Gas and Electric Compa ny. This study involves seismic evaluations for the containment annulus structure and piping of the Unit 1 portion of the plant. These evaluations are based on the 7.5 M Hosgri earthquake which is the input used in the latest stage of the seismic assessment of the Diablo Canyon Plant.
1.1 BNL Diablo Canyon Work Scope Initial work assignments BNL involvement with the Diablo Canyon review was initiated on October 9, 1981 when members of the Structural Analysis Division were requested by NRC to attend a meeting in Bethesda, Maryland where the so-called " diagram error" made by the Pacific Gas and Electric Company (PG&E) of California in the de-sign of the Diablo Canyon Nuclear Power Plant, Unit I would be detailed by company officials and contractors. As a result of this meeting, an audit of the utilities ef forts was scheduled and BNL representatives were requested to attend. The purpose of this audit which was subsequently held at the PG&E headquarters in San Francisco, California during the period October 14-16, 1981, was to review with PG&E the extent of the errors, their possible effects and to assess the validity of the corrective measures being adopted to requa-lify t"e plant design. The technical areas reviewed during the audit pertain-ing to the annulus structure whose design was affected by the diagram error were: derivation of the floor response spectra, electrical equipment and instrumentation installations, mechanical equipment and ventilation, piping systems and their various supports, conduit and cable tray supports. BNL presented its comments to NRC regarding this audit, late in October of 1981.
First set of analytical tasks At the beginning of November 1981 NRC requested the Structural Analysis Division of BNL to independently develop vertical floor response spectra for the Unit 1 contaiment annulus structure of the Diablo Canyon Plant. In
! addition, piping systems with PG&E designation numbers 4A-26 and 6-11, located
- in the annulus area were to be independently reanalyzed. For the generation
! of the floor spectra, BNL was requested to develop and utilize a three-dimensional model that closely resembled the actual structural configuration of the containment annulus. Results from this analysis would then be used as t
input for the two piping systems. For the analyses of the latter, uniform support excitation methods would be applied. Final instructions under this dssijnment included an additional check of floor spectra results with a computer code available for use to the general public.
l In order to carry out the above work, it was necessary that all required data be assembled and reviewed. Thus the first order of work was to review all information and data made available during the San Francisco meeting with P G&E. From this effort it became apparent that a good deal of required information was not available. Essentially, we had the following:
a) an input card deck for the Newnert 7.5M Hosgri ground acceleration time history b) a description of input data used for the two-dimensional Blume model c) a two-page summary of the two-dimensional analysis d) a copy of the URS/Blume report "Diablo Canyon Nuclear Plant Unit 1 Containment Structure, Dynamic Seismic Analysis for 7.5 M Hosgri Earthquake", May 1979.
In addition from our discussions at the San Francisco audit, we had a general understanding of the structural and piping system layouts. This, however, was not adequate for the detailed analysis required for the BNL revi ew. In view of this, we immediately requested that the needed data be made available to us.
On November 20, 1981, the following PG&E prints were transmitted to us:
Dwg. 469355 Rev. 3 Civil-Deadloads Annulus Platform Containment Structure EL 140' & 130' Dwg. 469356 Rev. 3 Civil-Deadloads Annulus Platform Containment Structure EL 117' Dwg. 469357 Rev. 3 Civil-Deadloads Annulus Hanger Franes Containment Structure EL 106' Dwg. 469358 Rev. 3 Civil-Deadloads Annulus Hanger Frames Containnent Structure EL 101' These allowed us to model the inertial contributions from the equipment, piping, etc. that are located on various floors of the structure. Mi ssir.g however, was information regarding the distributed weights of the various structural members comprising the annulus structure and a detailed description of member connections. Information needed for the piping evaluations was also l acki ng.
On December 23, 1981 we infomally received a package of data from PG&E which was formally transmitted to NRC on January 22, 1982 describing piping problems 6-11 and 4A-26. Specifically, the data transmitted for problem 6-11 corresponded to the applicant supplied list shown below:
ITEM A: Seismic Analysis 6-11 Isometric Drawing Design Review Isometric - PG&E Dwg. 437989 (Safety Injection, Loop 1&2)
ITEM 8: Hanger Details and Summary Sheet ITEM C: Piping and Insulation Specifications ITEM D: No Valves Involved with Analysis 6-11 ITEM E: Horizontal and New Yertical Hosgri Response Spectra Seismic Analysis 6-11 ITEM F: Seismic Anchor Movements (Blume Report)
ITEM G: Design Change Notices will be sent as they become available ITEM H: File 33 Forms (Hanger Load Tabulation)
ITEM I: Contaiment Penetration Allowables l ITEM J: Schematic Diagrams and Piping and Instrumentation Diagrams (P&ID's) l ITEM K: Pressures, Themal Modes and Anchor Movements
( ITEM L: No Additional Dynamic Loading Conditions for Analysis 6-11 l
A corresponding list for problem 4A-26 was also transmitted. On j January 15, 1982 the following structural drawings were received.
Dwg. 438281 Rev. 9 Annulus Platfom Framing EL 117'-0" Containment Structure
Dwg. 438282 Rev. 10 Annulus Platfonn Framing EL 140'-0" Containment Structure Dwg. 443039 Rev. 3 Structural Steel Modification of Annulus Platfonn and Frames Contairinent Structure Dwg. 447245 Rev. 6 Annulus Hanger Frames EL 101' & EL 106' (Containment Structure)
Dwg. 447245 Rev. 4 Concrete Outline and Reinforcing Annulus Platform EL 140' Containment Structure Based on the information received to that date a three-dimensional model of the structure was developed. Unfortunately the structural drawings received in November 1981 and in January 1982 did not contain all details pertaining to the type of connections (i.e., shear or moment) used between all the structural ment >ers comprising the annulus structure. Similar questions regarding the member connectivities used for the URS/Blume 2-D model given in the May 1979 report, previously cited, also prevailed.
In discussing these uncertainties with our technical monitor, it was decided to model the structure with shear type joints for the beam to column connections. This model is referred to as 3-D model A. Subsequently in the middle of February 1982 we were also asked to make an additional computer run considering moment connections for the beam to column framings for both the first and second floors. (This is 3-D model B). We also officially requested a confirmation from PG&E via NRC for the actual framing connection details.
By the end of February 1982, work on the floor response spectra using the above two models was completed. In addition the piping systems were modeled l
and computer runs were made for mode shapes and frequencies. Furthe rmore, spectral results from 3-D model A were utilized to compute the responses of the piping systems. A comparative check using the STRUDL code for spectra generation was also made. Preliminary results of the above evaluations were presented to pertinent NRC personnel in a meeting held in Bethesda, Maryland on March 2,1982.
Second set of analytical tasks At the close of this meeting we were asked by NRC to extend our original work scope by including the following tasks:
(1) generate floor response spectra for a third 3-D model, (i.e. ,
model C) where beam to column connections of the first, second and third floor are taken as moment type.
(2) Carry out a confirmatory computer run for the original PG&E 2-0 model using data from the San Francisco meeting.
(3) Carry out independent support excitation analyses for the two piping systems (using spectral imputs from 3-D analysis) and evaluate piping system response in accordance with ASME Class 2 classifications.
The above tasks with the exception of the piping runs were completed by March 15,1982. Specific results pertaining to items (1) and (2) above were discussed with Dr. P. T. Kuo and Dr. M. Hartznan of NRC at a meeting held at BNL on March 17, 1982. It was pointed out that the floor spectra correspond-ing to 3-D models A, B and C were completed and that results pertaining to the two-dimensional model did not match with those reported by URS/Blume. We felt that the reasons for the mismatch were due to the boundary conditions, degrees-of-freedom assigned to the nodes and integration parameters (i.e.,
time steps) used by URS/Blume. It was decided that BNL should vary the above parameters in an attempt to obtain a better match with the Blume 2-D model.
In addition, it was agreed to also verify the results of the spectra from the 2-D model with spectra generated by the STRUDL-DYNAL code.
~6-
With regards to the 3-D results, just as we were deciding at the close of the March 17th BNL meeting which of the 3-D models (i.e., A, B or C) most closely represented the actual structure, a courier arrived with the following PG8E structural steel fabricators drawings:
DC663243 - 465-2 DC663368 - 1-1 DC663243 - 466-2 DC663368 1 DC663243 - 467-2 DC663368 2 These were the missing joining drawing details that we requested back in January. Frm these we readily ascertained that 3-D model B most closely resembled the actual field conditions. Thus the forthcming piping evalua-tions would utilize the spectra and displacements generated from this model.
Third set of analytical tasks In the week following the meeting our efforts were concentrated on the verification studies of the 2-D model and the completion of the various piping analysis tasks. It soon became apparent that although parameters of the 2-D model were varied, none of the parametic solutions correlated well with the URS/Blume results given in their May 1979 report. Considering the lack of agreement between the solutions and the uncertainties in some of the para-meters we requested that NRC obtain a listing of the cmputer input / output for the runs used to generate the spectra given in the URS/Blume report. This information was relayed to us by PG8E on April 24, 1982.
In reviewing this latest infonnation for the two-dimensional model, it became obvious that the distributed masses of the steel members were not included for the input member properties (i.e., the code input showed mass density to be input as zero). Secondly, only eighteen modes were considered, whereas the prirnary floor masses are lumped at the mid-spans of twenty beams cmprising the five frames. The degrees-of-freedom were mixed 1.e. , nodes along the crane wall had one degree-of-freeds (vertical translation) whereas the others had two. This was not at all clear either from the discussions at San Francisco nor from the Blume report.
With respect to the boundary conditions, it also was apparent from submitted data that only beam to crane wall connections of the concrete floor at elevation 140 feet were shear type. All other connections were taken to be rigid. This seems to be different from the statement given on page 11 of the URS/Blume May 1979 report.
Using the information received from PG&E on April 24, 1982, BNL generated
" raw" floor spectra which agreed with those presented in the submittal. How-ever, these raw floor spectra are not consistent with the broadened spectra presented in the May 1979 report. Additionally, the version of the post-processor used for the " raw spectra" submitted to BNL on April 24, 1982 is dated 12/80 (program JAB /FLS PEC VER 1.0, page 24), which seems to be inconsistent with the report publication date, May 1979.
All tasks requested by NRC to date described in this section are summarized in chronological sequence for convenience in Table 1. All of the analytical tasks were completed by the first week of May 1982. The body of this report presents the pertinent results of the study. A summary of the conclusions is presented in the next section.
l l
l l
l l
{
l l
l
Ta ble 1 Task Outline (1) Requested to attend October 9,1981 Bethesda, MD meeting where PG&E and contractors discussed the y so-called Diablo Canyon Unit 1 " Diagram Error".
a e ME i (2) Participate with NRC at Diablo Canyon audit held at PG&E San Francisco headquarters during the "c9{m period October 14-16, 1981.
- Present comments pertaining to PG&E audit to NRC.
(3)
(4) Requested to independently develop vertical floor response spectra for unit I containment annulus
- structure. Model A (shear joints for all beam and column connections), and Model B (moment connec-ygy, tions for 1st and 2nd floors).
Eu%s <
Requested to independ'ently reanalyze piping systems with PG&E designation numbers 4A-26 and 6-11.
EE}0 (7) Carry out check calculation for floor spectra using general public computer code.
1 y (8) Repeat task (4) for Model C (where beam to column connections for 1st, 2nd and 3rd floor are considered rigid),
m (9) Carry out confirmatory computer run for original PG&E 2-D model .
]uuI y% m <
$E"m3 (10)& Carry out multiple input analysis for PG&E piping systems 4A-26 and 6-11 using spectral imput from
= (11) 3-D analysis and evaluate in accordance with ASME class 2 classifications.
,(12) Verify results of the spectra for 2-D model with spectra generated from STRUDL (McDonnell Douglas).
,I,_ (13) Extend 2-D study varying the boundary conditions.
i" E % B "o (14) Review input / output of URS/Blume 1979 run. Use identical model and compare raw floor spectra.
e E (15) write report detailing results of items 4 to 14.
n
1.2 Summary of Results In the previous section a chronology of the work effort was presented in order to acquaint the reader with the actual task sequence. In this section a similar chronological order will be followed. First, results for the three-dimensional finite-element model floor spectra evaluations will be given.
These will be followed by a synopsis of the results obtained for the two confi rmatory piping system analysis. Finally, a summary of the findings for the confinnatory two-dimensional model studies will be presented.
Results of 3-D Model Studies As mentioned, at the request of NRC a detailed three-dimensional model of the containment annulus structure was developed. Because of uncertainties pertaining to structural member connections, three variations (i.e., A, B and C) of this model were. analyzed. Based on the latest information made avail-able to us during the 3rd week in March, Model B most closely represents the actual structure. The pertinent results are as follows:
(1) Modal shapes for this structure are primarily local in nature, with relatively small sections of the structure excited for a particular frequency.
This localized characteristic of the modes, however, decreases in going -from model A to model C. For model B there are mixed cases, some modes affecting only local sections of the structure while others affecting greater portions of the structure.
(2) 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. This was the case for all models (i.e., A, B and C) evaluated. URS/Blume results for the top floor were consistently conservative in that they exceeded the accelera-tion magnitudes predicted with the BNL models. This, however, is not the case for floors one, two and three. For some frequencies the URS/Blume results for these floors were conservative, for others they were not, with no observable trend.
The differences in the results can be attributed to:
(a) the lower mass values used by URS/Blume (see discus-sion pertaining to two-dimensional model results)
(b) differences in structural member connnectivity (c) the fact that a single mass may or may not adequately represent all individual structural parts comprising a section of a floor.
(3) The BNL 3-D model floor response spectra results were also verified with the STRUDL-Mc0ONNELL-D0UGLAS computer program. Good agreement for both modal frequencies and floor spectra were obtained.
Results of Piping Analysis Confirmatory evaluations were perfonned for PG&E piping problem numbers 6-11 and 4A-26. These piping systems are connected to the first, second and third floors of the annulus structure. The evaluations were carried out using envelope response spectrum methods and independent support motion response spectrum methods using both PG&E and BNL developed spectra. The PG&E 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. At sme frequencies the new results are higher for the piping systems and at others they are lower. Presumably, these spectra were developed frm a new model of the annulus structure. A summary of the conclusions are:
- 1) BNL models developed from PG8E as-built drawings were found to differ from the PG&E models. The differences were due to the use by PG&E of design dimensions which differ fra the as-built dimensions and in errors made by PG&E in the modeling of pipe bends. Also an overlap procedure was used in the modeling of problem 4A-26. The extent of overlap used in the problem seems adequate in that it meets the intent of NUREG/CR 1980.
- 2) BNL predictions of system frequencies differ from the PG&E estimates, however these differences are not large.
- 3) BNL support force values obtained using BNL models and PG&E supplied spectra do not match. The differences are probably due to the differences in modeli ng.
- 4) Support forces calculated using BNL piping models and BNL 3-D Model B envelope or independent spectra substantially exceed PG8E calculated values.
The major cause for this is that Model B spectra greatly exceed the spectra used by PG&E.
- 5) ASME Class 2 evaluations perfonned using the unifonn response spec-trum method indicated exceedance of service level D stresses at 2 points in problem 6-11, while problem 4A-26 satisfied service level D requirements.
l
- 6) ASME Class 2 evaluations perfonned using the independent support response spectrum methods produced a reduction in stress levels in problem 6-11, but an increase in stress levels for problem 4A-26. For this procedure, problem 6-11 shows slight overstressing at one point, while 4A-26 still meets requi renents. Its possible that independent support input excitation analyses based on the time history methods could produce results which would, depending on phasing, satisfy service level D requirements.
Results of 2-D Model Studies As mentioned in the previous section, due to the uncertainties in some of the pertinent data, various parametric studies for this model were performed.
None of these, however, correlated well with the URS/Blume results given in their May 1979 report. We thus requested that NRC obtain a listing of the computer input / output for the runs used to generate the spectra given in the URS/Blume report. This information was relayed to us by PG&E on April 24, 1982. A summary of the conclusions of this portion of the study is given bel ow.
- 1) 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 PGaE on April 24,1982 (see previous section regarding conputer method).
- 2) The broadened spectra presented in the May 1979 report generally correspond with the raw spectra values sent to us in April 1982. In the 1ower spectral frequency range however, it seems that the broadened spectra were obtained by use of mean raw spectra values.
- 3) A detailed review of the PG8E supplied input / output data resulted in the following findings with respect to the 2-D model described in the May 1979 repo rt.
(a) The weights used in the model do not correspond to those shown in the PG&E drawings submitted to BNL during the period November 1981 -
March 1982 (for details of drawings see previous section). For example, for the third floor the total mass used in the model is 6.71 Kip-sec2/ft, while the value from the corresponding drawings is approximately 11.4 Kip-sec2/ft.
(b) The computer printout sent to us (in April 1982) indicates zero mass density input for the structural members. It could be that member weights were meant to be included in the values of the lumped masses. However, if this were the case, then, again taking the third floor as an example, the total mass of all items supported by this floor as obtained from the corresponding drawings is approxi-mately 7.54 Kip sec2/ft. This figure which does not include the mass of the structure itself already exceeds the value used in the URS/Blume computation.
(c) A review of the steel fabricators drawings show that the member connections used in the 2-D URS/Blume model do not represent the conditions indicated in these drawings.
I (d) With respect to item (c) above, it should be noted that parametric studies carried out at BNL with the 2-D model showed that the floor spectra results can be significantly altered by member connectivity.
- 4) As with the 3-D results, the 2-D BNL SAPV results were verified with a STRUDL-McDONNELL-DOUGLAS computer run.. A good match for the modal fre-quency's and the floor spectra were obtained.
4 2.0 Description of Structural Mathematical Model The nadel utilized for the containment annulus of the Diablo Canyon Plant Unit 1 is described in this section of the report. The four floors of the structure are supported from the crane wall and by columns located just inside the containment. The widths of the floors are about 16 feet. The floors span circumferentially most of the 360' of the annulus space located between the containment and crane wall.
The first three floors (at elevations of 101',106' and 116') are steel frame structures consisting of girders spanning radially from the crane wall to the columns and tangentially between columns. Intermediate support beam frames are located between columns. The fourth floor (at elevation 140')
consists of the same type of steel structure with an 18" concrete slab supported on the steel framework. The concrete and steel beams on this floor act independently and not as a composite. Because of the slab dimensions, the concrete slab will behave as a one-way slab spanning between the crane wall and the tangential girders at the outer column line. A skematic drawing of this configuration is shown in Figure 2.1 2.1 3-D Model and Element Data A 3-D model was developed to treat the vertical response of the struc-ture. Preliminary analysis showed the crane wall to be much st1ffer than the columns and therefore it was not necessary to model this wall with shell elements for vertical analysis. In addition it was concluded that the vertical input will not be amplified in the vertical direction through the rigid crane wall. Thus, the seismic excitation applied at all of the supports of the annulus structure located on the crane wall will be the same, (i.e.,
the 7.5 M Hosgri earthquake scaled to 0.5 g for the vertical analysis).
Furthermore, each floor contains bracing members designed to restrain a rigid body rotation of the floor about a vertical axis. Since this motion will not be excited by a vertical input, these members are not included in the model.
Nodes are included at the intersection of all members with additional nodes added along the span of the beams and girders so that higher beam modes may be obtained. At each node (other than support nodes) the vertical displacement and rotations about the two horizontal axes are unrestrained.
The remaining three degrees-of-freedom will not be excited with a vertical input and therefore need not be considered. They were restrained in the computer runs.
Details pertaining to the finite element idealization of the structure are shown in Figures 2.2 through 2.6. The numbers shown on these diagrams are nodal numbers. A general view of the 3-D finite element grid is shown in Figure 2.2. The eighteen columns, located about 20 degree intervals around the structure, may be seen in this figure. Note also the hangers which connect various parts of the first, second and third floors. The framing plans for the first through fourth floors are shown in Figures 2.3 through 2.6 res pectively. The triangular elements on the fourth floor are used for the discretization of the 18 inch thick concrete slab.
Details of the 3-D model were developed from the following drawings submitted to BNL by PG&E:
Dwg. 438281 Rev. 9 Annulus Platform Framing EL 117'-0" Containment Structure l
Dwg. 438282 Rev. 10 Annulus Platform Framing EL 140'-0" Containment Structure Dwg. 443039- Rev. 3 Structural Steel Modification of t Annulus Platform and Frames Contaimnent Structure i
Dwg. 447245 Rev. 6 Annulus Hangar Frames EL 101 & EL 106' (Containment Structure)
Dwg. 447245 Rev. 4 Concrete Outline and Reinforcing Annulus Platform EL 140' -
Containment Structure The computer runs were made with the BNL version of the SAPV finite element computer code.
The steel section properties were taken from the AISC Steel Design Handbook. Steel is taken to have a Young's Modulus of 29,000 ksi and a density of 0.2836 pound per cubic inch (490 pound per cubic foot). The concrete is taken to have a modulus of elasticity of 4150 ksi and a Poisson ratio of 0.17. The concrete density is 0.0868 pounds per cubic inch (150 pounds per cubic foot).
2.2 Mass Distribution The structural weight is generated initially in the SAPV program based on the member geometry and density. Other weights are input as nodal masses.
The basis for the calculation of the lumped masses was obtained from the fol-lowing drawings transmitted to BNL by PG&E:
Dwg. 469355 Rev. 3 Civil-Deadloads Annulus Platform Containment Structure EL 140' & 130' Dwg. 469356 Rev. 3 Civil-Deadloads Annulus Platfonn Containment Structure EL 117' Dwg. 469357 Rev. 3 Civil-Deadloads Annulus Hangar Frames Containment Structure EL 106' Dwg. 469358 Rev. 3 Civil-Deadloads Annulus Hangar Frames Contairinent Structure EL 101' i
The above mentioned drawings contain information regarding point or concen-trated weights and distributed weights. Amongst the first category are contributions due to (a) large bore mechanical pipes and supports, (b) me-chanical equipment (c) architectural platforms and ventilation ducts and equipment. The latter category of weight includes such items as (a) mechan-
< ical small bore piping, (b) electrical conduits, trays and equipment and (c) grati ngs. In order to obtain the nodal masses used in the model, based on the items mentioned above, it was necessary to subdivide the total area into tributary areas for each of the nodes in the model. The uniform loading was then multiplied by this area to determine the resulting lumped mass. The concentrated masses were then assigned to the node in whose tributary area they fell . The resultant lumped masses were then totaled for each of the floors and compared with the total weight on the drawings to verify the computation.
2.3 Member Connectivity Member connections were modeled as either shear (pinned) or moment connections. All beam to beam and beam to girder connections were modeled as shear connections. All connections to the crane wall were also modeled as shear connections. The girder to column connections were modeled in accor-dance to the three different models used in the study (see work scope descrip-tion),i.e.,
Model A: All girder to column connections treated as shear connections, Model B: Girder to column connections in floors one and two fixed with the top two floor connections pinned, Model C: Girder to column connections in floors one, two and three fixed with the top floor connections pinned.
': ,, e G e Td- , ,
,., fjd9: g ,,
-g/~. m.,P- -- 7~.. t p R$..ce.
- j
( R.. .m-~ w-i l : .- - g, ~~ -
i.:(f_y,r- , &.. - , '
(,,-
., _ .; l
~ ' '
jj 1 % b =t hu m' W[. fsu wu,% g'
~.[y ;
I .; e .e, 25" ,-
T fp @ 'jTj W '<~ '?- ~: -
. ., M - . I
- i. .,
]' '
g l, t y ,
. n . .,
rg,,. / i. ! Y \ J p,s. '
(f l \- t 4 e '~
//
ei e //e// +
, 't iw.l l s ;4
' " l ..
I s
J. ~
g Q.:. l i
g / sky [s' g
, . s. ,
I ) in l i : !, i /
4 .. 4 ,
%l
~ . 7. _ L' -
!" .g F Td' 7 qk!
j g , 1
,"/. .. -
,, A..
_. s- .
M) '
' f. !
3"1 ( l i ')
/ '
L ' c In.
n
,,4
\
b....,,_. .-
esa '
w f l
a a am d y du n' Figure 2.1 - Schematic View of Containment Annulus Structure b
f 3!ae t.3 C ANTO 4 P'. ANT. CONT atm gg i JCEFOMO PePE 18:15 2 AL N s 30 00 SEYan 3 00 V
/ N/
k MM
?
@\ / 76 N ! '
w, A
N/
% \
1 l
I
,C j
k/
j
\
/ '
' [f N
/ ,,
(IN N
- h x
N s
'n R 7,N /
(hy
_Nhhv
\ a s -
N s
y .
s-J I M
O 4 u
,- f ( "\
i l
Figure 2.2 - Three-dimensional finite element model
w
.e E
E .
e a
se
=
e-4 I")
- m se N C
M .,O.
g 4J c
4 c " -
L
. 8 C
W T L 9
- Q m %
44 R . '"*
p T 'P" V
=
g
. 3 9 5 3 a.
g "2 3 A '
3
O5 4 N
3 *
~ s
. < a R
gE
= 4
>7 M
C i
l
3 5 8
7 g L #l g
8 7 1 o 4 g 4
- 9 G 8 1 2 G 0
1 1
I I 8 1 8
1 8 R
1 3 3 1 6' 7
g 0 4
1 1 1
, n 6 o 7 7 i 1 t a
v e
3 l e
7 [4 1
5 7 t a
1
_ r g
2 o 7
g
{
7 4 3 o 2 2 l 2 L f 1
r 8 g o S f 6_h 1
2 6 5 s 1 2 l i
9 a 2 t 1 8 e 7
8 7 D 6 1 2 6 l 5
6 1
1 e 4 3 d 1
6 4 1 0 , o
_ 3 1
3 1 0 M 8 9 1 3 -
6
= 4 9 3 X 5 1 A 2
- 1 T
8 E e 1
3 H r 4
5 1 7 4
1 T u g
1 5 4 .
7 8 g 4 i 4 4 0 F 5' 4 g
1 9 1 a 2
4 Y 1
5 5 A 8 4 4 T E
H T
0 Z =
Z A
T E
H T
kY
283 287 294 27 384 288 2 91 294
'I 289 2 92 295 297 8
pSS g3 296 73 301 36 34 304 27R 345 3 0 1
265 3gg SW 3 g 3 3
55 317 116 115
, 57258 ru Y
195 94 $93 92 88 97[9g 01 47 et 4
88 4 344 5
4 209 3 R37 23 8 7 234 816 5 233 223 4 232 29 226 h 21 225 g WA z. 4. TETA Y= 0. TETA X* 90. '
Figure 2.5 - Model Details for floor at elevation 117'
I 394 3 397
.. 4 3
3
%. 9,
, 4.,
446 5
499 412 2 4 5
gg 416 69 4
- 171 4 2' s s' . .. ...
r.;
n y, y',
12.
328 pJ -
~
l T34 s 21 l gg.
24 ,
N
~
yy 333 3S7
\ 1 M9
- 348 334 3 343 34g Y x 33 7
33 Z
pg*
41 fh ,
/
MA 2 g, g4 y,
- WA X. , ,
figure 2.6 - Model Details for floor at elevation 140' '
e e d - _
4 vds
~
3.0 Structural Evaluations The seismic evaluations perfonned for the containment annulus structure
_ required (1) a free-vibration analysis to obtain the modal characteristics of the structure, (2) a time history analysis to detennine absolute acceleration response records at various locations along the structure and finally (3) the development of the floor response spectra. All structural evaluations were, as mentioned, carried out for the vertical direction only. A comparative study was also made in order to verify the computational -scheme used in the BNL analysis. Followi.ng.the development of the floor. response spectra at the majority of the nodal points of the 3D model, representative floor response spectra were computed for the four floors of the structure. In addition the necessary floor response spectra and displacements were computed, in order to
~
supply the' input for the selected piping systems.
It s ould be noted that due to the uncertainties regarding the boundary conditions of the structure, all structural evaluations were repeated for three dif ferent types of connectivity between the structural members.
Each lof the above items are discussed in detail in the paragraphs that follow.
3.1 Free-vibration analysis The modal characteristics of the structure i.e., modal shapes and frequencies were obtained in the vertical direction. Three cases were studied depending on the boundary conditions utilized for the mathematical model.
f /
~ ~
The 'first case, i.e. , model A, corresponds to a pin-connected model in which all member connections are considered as shear type. The other two cases, i.e. , models B and C, involve beam-to-column connections which are moment type. Specifically in the second case, the beam to column connections of the first and second floor at elevations 101' and 106', respectively, are considered to be moment type, while in the third case, the beam to column
(-
connections of the first, second and third floor (el. 101',106' and 117') are moment type. Table 3.1 summarizes the cases used in the structural evalua-tions. With exception of the different types of boundary conditions applied, all three models have the same characteristics, i.e., stiffness and mass as those described in paragraph 2.1. As mentioned in the work scope section, this classification of the structural mathenatical model became necessary, due to the uncertainty regarding the boundary conditions encountered during the course of ' this work.
Table 3.1 Model-cases evaluated for different boundary conditions Model case Model characteristics A All member connectior.,
are shear type B beam-to-column conections for the floors at elevations 101' and 106' are moment type C Beam-to-column connections for the floors at elevations 101', 106' and 117' are moment type.
In terms of stiffness, it appears that the structure is characterized by weak elastic coupling. From the modal analysis performed for all models, i.e., A, B and C, it was found that many of the modes are local in nature.
This result was expected due to the physical arrangement of the structural components of the containment annulus structure. Coupling effects betwen the floors are mainly introduced by the presence of the columns, the crane wall and the various hangers.
1 Frm the obtained modes, it was concluded that they are either of very localized type or are spread over sections throughout all the floors. A typical localized mode which excites basically a small portion of the struc-ture at low frequency (11.8 cps) is depicted in Figure 3.1. The dotted lines indicate the undeformed shape of the structure. It is observed that in this mode only some steel beams of the top floor (el.140') participate whereas the concrete part is not excited. These beams are located at an opening arrangement in the floor. From Figure 3.1 it is concluded that most of the activity at this modal frequency occurs at the second floor. Another mode obtained at a frequency of 13 cps is shown in Figure 3.2. Again the dotted lines indicate the undeformed shape of the structure. This mode is spread over a larger portion of the structure than that of the one discussed previ ou sly. The third floor is the most effected at this modal frequency whereas all other floors also participate over a larger section of the structure than in the previous case. It is to be noted that a small part of the concrete slab of the top floor not excited before (Figure 3.1) is now (Figure 3.2) excited in this mode.
Three free-vibration analyses were performed corresponding to the three model types A, B and C. The number of very localized modes decreased from case A to C. While it was felt that modes up to 20 cps may be sufficient to compute structural responses based on the modal superposition technique, we nevertheless, at NRC's request included higher modes. The number of modes required to cover a frequency range up to 33 cps was found to be very large, due to the fact that the structural modes are close-spaced. At very close frequency values, different parts of the. structure are excited. Specifically
l a total of 158 modes were required to reach a frequency of value 33.12 cps.
The corresponding modal frequencies are given in Table 3.2. These values are for model A. Frequencies for model 8 are given in Table 3.2.1.
3.2 Input Time History The input acceleration time history used for all structural evaluations is the Newmark 7.5 M Hosgri earthquake scaled by 2/3 for 0.5 g peak value.
The characteristics of the digitized record for this input are:
peak acceleration: 0.50 g time interval : 0.01 sec data points : 2400 The input acceleration record is shown in Figure 3.3. This figure is taken from page 46 of reference 1.
3.3 Generation of Vertical Floor Response Spectra In generating the vertical floor response spectra for the containment annulus structure of the Unit 1 Diablo Canyon Plant the time history method was used. First the vertical absolute acceleration time histories at the nodal points of the structure were evaluated. For this purpose, the free-vibration characteristics of the structure (modal shapes and frequencies) described in paragraph 3.1 were utilized to uncouple the equations of motion of the structure. The time histories of the vertical nodal accelerations were obtained by snploying the modal superposition technique. Modal damping equal to 7% of the critical was employed in these evaluations.
It should be noted that for unconditionally stable schemes of numerical i ntegration, i .e. , Wil son-0 method, a time step at = T/10 (T is the shortest period of interest) is generally adequate for accurate. response calculations.
A time step equal to 0.001 seconds was used for the integration of the modal equations. The accuracy of the results obtained by utilizing this integration time step was further investigated. Particularly, a finer time increment at =
0.0005 seconds was also used to campute structural responses. The latter were then compared with those previously computed with at = 0.001 seconds. Fran this conparison it was concluded that the integration time step of 0.001 seconds is sufficient. Typical results of this comparison are demonstrated in Figures 3.4 to 3.9. Finally, all structural responses were computed for a total duration of 15.00 seconds. Essentially, the peak responses occured within this time duration.
Following the evaluation of the nodal absolute accelerations, the floor response spectra were generated by employing standard techniques. For this purpose, the absolute acceleration time histories were used as input to a single degree-of-freedom system and acceleration spectra were computed. These spectra were developed for 2, 3 and 4 percent equipment damping. Thus the spectral damping values used, concur with the values used by the Pacific Gas and Electric Company (see Ref. [1]).
The above procedure for the generation of vertical floor response spectra, based on time history analysis, was repeated three times in order to cover all models i.e., A, B and C, (see Table 3.1). Each time, the floor response spectra at more than two-hundred nodal points for the above three spectral damping values were computed. Approximately two-thousand spectral curves were generated under the present study.
l 1
Table 3.2 MODAL FREQUENCIES MODEL A Mode Freq. Mode .- Freq. Mode Freq. Mode Freq.
No. (cps) No. (cps) No. (cps) No. (cps) 1 6.59 21 12.19 41 14.34 61 16.60 2 6.71 22 12.24 42 14.39 62 l6.62 3 7.71 23 12.35 43 14.48 63 16.71 4 7.75 24 12.95 44 14.81 64 16.92 5 8.52 25 13.10 45 14.83 65 17.05 6 8.76 26 13.25 46 14.85 66 17.15 7 9.39 27 13.28 47 14.98 67 17.28 8 9.45 28 13.45 48 15.28 68 17.39 9 9.66 29 13.49 49 15.39 69 17.63 10 10.13 30 13.56 50 15.58 70 17.81-11 10.40 31 13.67 51 15.62 71 17.83 12 10.73 32 13.82 52 15.82 72 17.91 <
13 11.07 33 13.85 53 15.90 73 18.11 l 14 11.22 34 13.87 54 15.94 74 18.18 l
15 11.54 35 13.90 55 16.07 75 18.27 16 11.65 36 13.95 56 16.12 76 18.28 17 11.87 37 13.97 57 16.12 77 18.36 18 12.07 38 14.01 58 16.19 78 18.45 19 12.14 39 14.20 59 16.25 79 18.50 20 12.19 40 14.22 60 16.45 80 18.65 l
l L
\
_. . _= - . .- .
Table 3.2 (cont'd)
MODAL FREQUENCIES MODEL A Mode Freq. Mode Freq. Mode Freq. Mode Fr eq.
No. (cps) No. (cps) No. (cps) No. (cps) 81 18.80 101 21.30 121 23.92 141 27.06 82 18.84 102 21.52 122 24.13 142 27.10 83 18.88 103 22.04 123 24.24 143 27.69 84 18.93 104 22.17 124 24.29 144 27.93 85 19.07 105 22.30 125 24.38 145 28.14 86 19.11 106 22.45 126 24.54 146 28.42 4
87 19.41 107 22.50 127 24.56 147 28.43 88 19.44 108 22.63 128 24.78 148 28.88 89 19.61 109 22.69 129 24.94 149 28.96 90 19.73 110 22.74 130 25.12 150 29.21 91 19.82 111 22.79 131 25.17 151 29.50 92 19.87 112 22.97 132 25.36 152 29.50 93 19.94 113 22.99 133 25.38 153 29.89 94 20.01 114 23.09 134 25.51 154 30.66 95 20.16 115 23.25 135 25.91 155 31.03 1 96 20.34 116 23.42 136 26.27 156 32.18 97 20.37 117 23.44 137 26.56 157 32.32 98 20.59 118 23.48 138 26.60 158 33.12 99 20.66 119 23.60 139 26.91 100 20.83 120 23.67 140 27.02
Table 3.2 (cont'd)
MODAL FREQUENCIES MODEL B Mode Freq. Mode Freq. Mode Freq. Mode Freq.
No. (cps) No. (cps) No. (cps) No. (cps) 1 9.91 21 14.46 41 17.14 61 19.10 2 11.18 22 14.50 42 17.30 62 19.39 3 11.40 23 14.66 43 17.41 63 19.42 4 11.79 24 14.82 44 17.59 64 19.60 5 11.99 25 15.39 45 17.74 65 19.65 6 12.14 26 15.53 46 17.80 66 19.77 7 12.18 27 15.71 47 17.92 67 19.89 8 12.97 28 15.84 48 18.04 68 19.97 9 13.04 29 15.93 49 18.10 69 20.16 10 13.17 30 15.98 50 18.15 70 .20.25 11 13.21 31 16.12 51 18.26 71 20.30 12 13.52 32 16.22 52 18.27 72 20.37 l
13 13.63 33 16.26 53 18.36 73 20.40 14 13.83 34 16.31 54 18.50 74 20.61 15 13.89 35 16.45 b5 18.60 75 20.72 16 13.89 36 16.57 56 18.71 76 20.83 17 14.00 37 16.60 57 18.80 77 21.28 18 14.20 38 16.63 58 18.88 78 21.31
( 19 14.35 39 17.00 59 18.93 79 21.49 20 14.39 40 17.07 60 19.00 80 21.80 l
Table 3.2 (cont'd)
MODAL FREQUENCIES MODEL 8 Mode Freq. Mode Freq. Mode Freq. Mode Freq.
No. (cps) No. (cps) No. (cps) No. (cps) 81 22.06 101 24.67 121 28.43 141 32.56 82 22.23 102 24.78 122 28.53 142 33.30 83 22.38 103 24.98 123 28.93 84 22.49 104 25.17 124 28.95 85 22.62 105 25.19 125 29.17 86 22.71 106 25.47 126 29.57 87 22.85 107 25.70 127 29.67 88 22.98 108 26.03 128 29.98 89 23.09 109 26.65 129 30.05 90 23.20 110 26.75 130 30.18 91 23.26 111 26.97 131 30.47 92 23.33 112 27.10 132 30.51 93 23.42 113 27.30 133 30.69 94 23.67 114 27.38 134 30.79 95 23.91 115 27.42 135 31.11 96 24.13 116 27.50 136 31.79 97 24.28 117 27.72 137 31.96 98 24.39 118 27.78 138 32.05 99 24.53 119 27.95 139 32.29 100 24.55 120 28.16 140 32.45 ar - car ,, en
..'....h,2*E'...n..J' sa ss , ura, o.,
wttenssclM',y;m s.sm N#j
- 8\ /TAN! \ .'
s M~r 4 -- _. !
y F t
' / i r
/ <
N4 y/t 7
/ /
s \
' s ,
NW
'e (
b -
\ M y'
- e7 J
u ,, ,1 _
.L d^
,W
,_3_i f l <_ _ J U" 31 - Modal shape at frequency 11.8 cps Clre 50 e m,0 % PL net. Comt al%n t%
W6 F8E2iNCY 12.567 18:15 2 at.N J0.00 SE tas 0.00 2FLECTION SCAlg retica, g g .ca s g@
cfsyl\ f r
N f '
~
& f N
\\ / \
\
\ {
/ '
k./
7
, - l 2 t
!'A y
~
@(\<
~
[ [< m___ /
4
/
-l 5
5 :, - fg,~
i ~ y,
~ %, _~ 2My_ " '
Figure 3.2 - Modal shape at frequency 13.0 cps L
5 h /.00 7 T '
m , :
o.60 a i o I l
0
- k. h " ~
Y\
[f ^
i
-~~fb I pif
? % ;
( }
O 8 N '
7 - 0.5o l l !
e j I
l 0 i u -/.00 ' l O.o 2.4 4.6 7. 2 9.6 le.O I4.4 /G.8 I 9. 2 21. 6 24.C TIM E IN JE C O Nf3J Figure 3.3 - Time history record of input excitation.
EFFECT'0F INTEGRATION TIME STEP 2.0 PEPCENT CAMPfNG NCCE:152 i;ie .,y si ,
6 .ii i i ,iiiimi ..it iii,4 33 ; i iiii... i3 iii iij,iiii, dl
.- TIME STEP = 0.001 SEC a h I' TIME STEP = 0.0005 SEC '
i e i; '
1 A\ h L\.
y l ,
E
,_ - J{ l, E
w.
I O I
d u A u 2
- L
/
o , , ,1 ,,,,,, , ,, ,,,i,i,,i, ,,, ,,,,,,,,,,i , ,,i,r,i , , ,,,,,,,i,,,,,,,,i.i,i,,
4 2 .4 .6 1 2 4 6. 10 20 40. 60.
FREQUENCY (CPS 1 Figure 3.4 - Evaluation of Integration Time Step. Node 152 T
8 h
~
EFFECT OF INTEGRATION TIME STEP
^
. 20 PERCENT CAf1 PING NCCEyl53 i .1 i ,ii.3 i iiiigi, , , i i i ;ii n h i i i i; i ; ;<ii;i; iii iiio, - ,3 , , ,,,,i , i 3
-- 3 TIME STEP = 0.001 SEC
- ----- TIME STEP = 0.0005 SEC Ni i't 4
l gl 4
i O
4 'f
-p *
\
1 O
3 4^ i t M rg
. ta J l ' l t
u ,
. y-
' )
~
A . .
l
~
hY YhM iititierit,i.! ! tilit t t ,it ,irinitisiit , ! t ,tsi ! l , ei,,,,,iti,,,t ,t ,,,t,I , ,
o
- L 2 .4 6 1. 2 4 6. 10 20. 40. 60
. FREQUENCY (CPS) i Figure 3.5 - Evaluation of Integration Time Step. Node 153 I
E l
i
EFFECT. OF INTEGRATION TIME STEP i
2.0 PERCENT CAMPING NOCE=157 0 .is .. 1;. i; ii i . 6i,.i i i .
- iiiiii .ii i ij.i , i 1 . . ;i; i . . i i 6 . . .., , , ,
a
.s 7
TIME STEP = 0.001 SEC
- o * -----
TIME STEP = 0.0'005 SEC 1
, 1';
W I a.
h \, h 8 \/ t y
Z / j'
, o .
i c:
M t.d s
j I-l!
r
_J '
J LL.I '
L.) b .i 2
/
e i i eiiiii,Iii,ii i! , !>fif '! r i iiniitiiiif i! r i itil ! ! ii > ii! iiii i f i ) i !i! ! !
l
.t 2 .4 6 !. 2 4. 6. 10 20 40. 60 4
FREQUENCY ICPS1 4
Figure 3.6 - Evaluation of Integration Time Step. Node 157 1
4
. , _ . . . - - . . , , , _ . , _ . _ _ _ . . . _ - , ___,,, , , , , _ _ _ . , , . . . , . . _ _ . _ _ . . . _ . . - . . . . . , - ,,,.m,,_ , _ _ , , ,.x,,.. _ , , , , _ , . . . _ _ ,
4 EFFECT OF INTEGRATION TIME STEP i
J 20 PERCENT ORMPING N00E:163
. . . . . . . . . , ....i .
. j . ,
o.ii,j.........i..... . , , , . , - i j ,........im.j . . .,,, , , ,
, TIME STEP = 0.001 SEC TIME STEP = 0.0005 SEC i T I 9
, ) l 1
- I 3
9 I e 1
- . h l I z - ' '
i I,
C o t
% [1 i C !~
1 r 1
, W
) ~ h' i.' L.J !
U i U
C [.
i
. s l
g t t a s 11 t t rI f r 1 el l t i!! lf f I ! i1 1 : i i 11 f l f I r! l t I t !f I 1 I f ? : t rrisartlI t i rl t I t l t lt l { f
.I 2 .4 6 1 2 4. 6 10 20. 40. 60:
I FREQUENCY lCPS1 l
t
- Figure 3.7 - Evaluation of Integration Time Step. Node 163 1
i l
t l
l l
i I
EFFECT OF INTEGRATION TIME STEP 2.0 PERCENT DAMP!NG NCCE=166 1
io, iii i i ,i., . i
. . 1
...iini j i ,6 ...,
ii i i i ; ,i i , , i TIME STEP = 0.001 SEC TIME STEP = 0.0005 SEC i I 4 l
~
a,i <,
'J3 I l
J
'f1 ,
5~ #
V 't 9 C
Q:
-I y b '
U l u y '
C 1 J
- AA 1_
~ ,
l i
o i r i s i i r iil , s i r l it , ! ,lif ! t ir i iir! ii,1 ! ,t.!,t ! ! ,,,,,,,,t,,,,, , j !.,,, y9l
.I 2 .4 6 1 2. 4. 6. 10 20 40 60.
FRECUENCY (CPS 1 Figure 3.8 - Evaluation of Integration Time Step. Node 166 EFFECT OF INTEGRAT10N TIME STEP I
l 20 PERCENT CAMP!NG NCCE:281 2 ,,, ;, i ,;,.i,i 4 i iiiin.i. .i ;4 , - . . . , , , a n o. .;, , i , ,,,
1 {
I I i !
_. el ,
= TIME STEP = 0.001 SEC g i TIME STEP =-0.0005 SEC ;lj l
. . I I, n l
,: ,i z b \\
,i Eo t
5 \
D ,,
t
\\
d .
11 H '! hl e i vif w Od[
e v
,,,, ,, , ,,,~ /,, , , , ,, ..., ...,,,,,,ar, -
.s n. 2 .. s. no. 20 ao. so.
.n 2 ..
FREQUENCY ! CPS 1 l
Figure 3.9 - Evaluation of Integration Time Step. Node 281 i
r' i
t 1
i l
r
- 4. 0 Verification of the 3D Model 4.1 Comparative Study For the BNL studies reported here the SAPV(2) caputer code was uti-lized. As originally developed this code did not include some of the features required to perfonn floor response spectra generation. The BNL version of the SAPV code includes special features which have been incorporated over the years into the original SAPV code. Specifically for floor response spectra generation the following options are available:
- 1) option for computing absolute accelerations
- 2) restart option for the integration of modal equations
- 3) option for generating response spectra For the eigenvalue extraction the subspace iteration method was applied.
Due to the large amount of modes involved, it was necessary to modify the portion of the SAPV code dealing with.the eigenvalue solution. These modi fi-cations were made for the CDC 7600 cmputer system at the Brookhaven National Labora to ry. As mentioned before, a total of 158 modes were required to reach the frequency of 33.12 cps. These modes were taken from a computer run in which 190 modes were evaluated.
The original SAPV code also had to be modified in order to include the option of caputing absolute acceleration responses required for the genera-tion of the floor response spectra. In addition, due to the combined require-ment of the large number of modes together with small integration time step, a restart option was introduced into the code. All response records were evaluated for fifteen thousands time steps, with a time interval equal to 1 msecs. Finally, a spectrum generator was incorporated into the code. This generator is similar to that used in the SIM code (3).
I At NRC's request, a task dealing with computer code verification was undertaken. The objective of this task was to verify the BNL SAPV version with a different computer code available in the public domain. After di scus-sions between NRC and BNL, it was agreed that the STRUDL computer program of McDonnell Douglas Automation Company be utilized for this purpose.
Floor response spectra were generated by using both the BNL SAPV version and the STRUDL McDonnell Douglas computer codes. The results obtained by both codes were compared. Details of this task are given in the next paragraph.
4.2 Description of Results As mentioned, a conparative computer study was made in order to verify the results obtained from the BNL 3-D model of the containment annulus struc-ture. The results obtained from BNL's version of SAPV were compared with those obtained from the McDonnell Douglas STRUDL-DYNAL computer code.
For this purpose input data of the 3-D model were prepared according to the input requirements of the STRUDL code. Computer runs were made to evalu-ate the modal shapes and frequencies. Camparative results for the first ten modes from both codes are shown in Table 4.1. As can be seen, the results are quite close.
Additionally, floor response spectra were generated by both codes and campared as shown in Figure 4.1. The solid line represents the STRUDL-DYNAL results while the circled points represent results obtained from SAPV. As can be seen from the figure, the results compare very well.
Table 4.1 i
Comparison of Ten Modal Frequencies Computer Codes Mode No. SAP (BNL) STRUDL (MCDONNELL DOUGLAS) 1 6.58 6.45 j 2 6.71 6.55 3 7.71 7.52 4 7.75 7.91 J
5 8.52 8.36 6 8.76 8.54 4
7 9.39 9.18
] 8 9.45 9.33 9 9.66 9.49 10 10.13 9.87 4
t 4
j I
1 i
f i
l DIABLO CANYON PLANT l
UNIT 1 l CONTAINMENT ANNULUS ST3UCTUPE SAP VERSUS-STRUCL SPECTRA 2.0 PERCENT CAMPING r, .
.. , .i..m ;. . i , ;ii, .o ... . .i .. ,,i i
o
> c, 1 e j i
- $/ Le l 7 !
if \ i ] N '3/\j
'i STRUDL (ficDONNELL j DOUGLAS) hft 3 ! ' SAPV (BNL) j 1
=
r i.
Kl 3 hh,
- e l
i E A / %
C t /sv. o C
- / O
/ Tg
- ,A ![
i l WA z n l ~%
)
1 l
I j
,1 . . . . ., , . l , ,, . , ,, ..l
., 2 ., .s z. <. s. to. n. n. sa.
RECLENCY iCPS:
Figure 4.1 - Floor response spectra comparison: SAPV (BNL) versus STRUDL (McDonnell Douglas)
?
5.0 Description of Floor Response Spectra from 3-D Model 5.1 General As mentioned, the BNL work was concentrated on the development and evaluation of a detailed three-dimensional model that closely resembles the actual structure. In fact, three variations of this basic model (i.e., models A, B and C) were evaluated. For each model floor response spectra were generated at the majority of the nodal points for three levels of equipment danping (i.e. , 2, 3 and 4 percent).
Given the large volume of data produced and the lack of modeling correspondence between the BNL and the URS/Blume models, three different methods were utilized for conparative purposes. These are discussed below.
5.2 Methods of Comparison Comparisons were made by the following methods:
(1) floor response spectra from the 3-D model most closely corresponding to the vertical planes defined by the locations of the five fan coolers were compared directly to those of the URS/Blume spectra. This is re-presented schematically in Figure 5.1. This figure shows the orienta-tions of the fan coolers for Units 1 and 2. It also shows the orien-tation of the frames used in the URS/Blume 2-D model. As can be seen, this orientation corresponds to the Unit 2 configuration. This, in fact, is the so-called diagram error. Due to the error the following corre-spondence can be determined:
Unit 1 URS/Blume Fan Cooler No. Frame No.
1 5 2 1 3 2 4 3 5 4 The BNL noor spectra fran model B obtained at these locations are shown on pages 56 thru 75. In these figures the dashed line shows the corre-sponding URS/Blume broadened spectra for 2% equipnent damping. When ever the dashed line becomes dotted this indicates that the values of this portion of the curve cannot be scaled.
(2) The second camparison involved the development of average spectra over sections of the floors. These sections are the same for all four floors of the structure. Each section is formed by the bisectors of the angles defined by the actual locations of the fan coolers. Sections of a typi-cal floor corresponding to the Unit 1 of the plant are shown in Figure 5.2. As can be seen, the five sections of the floor are numbered in accordance with the fan coolers, i.e, section one corresponds to fan cooler one, etc. In order to obtain the average spectra for a given section the floor spectra for all nodes located on that floor in that section were simply averaged for each spectral frequency and for each danping value. These spectra, for model. B are shown on pages 76 thru
- 95. In these figures the dashed line shows the corresponding URS/Blume broadened spectra for 2% equipment damping. Whenever the dashed line be-comes dotted this indicates that the values of this portion of the curve cannot be scaled.
(3) For the third comparison envelope spectra from all nodal points for each floor and each section, as described in (2) above, were developed. These results are plotted together with the corresponding average spectra and presented on pages 76 thru 95. In these figures the dashed line shows the corresponding URS/Blume broadened spectra for 2% equipment damping.
Whenever the dashed line becomes dotted this indicates that the values of this portion of the curve cannot be sc 'ed.
5.3 Comparison with URS/Blume Spectra The comparison matrices that follow summarize the results obtained for each of the three comparison methods discussed in Sectio'n 5.2. If the number one (1) appears in the matrix the URS/Blume results are more conservative and are thus acceptable. If the URS/Blume results are not acceptable (i.e., not
conservative), then a zero (0) will appear at the corresponding element of the matrix. Furthennore for each zero there will be either a subscript P or F.
The former indicates differences in peak values and the latter in frequency.
( If both subscripts appear both frequency shifts and peak magnitude differences
! were apparent.
By inspection of these comparison matrices the following conclusions may be drawn:
(1) URS/Blume floor spectra for the fourth floor are conservative in all methods of compari sor.
(2) URS/Blume floor spectra for the fourth fan cooler location and for all floors are conservative, for the first and second methods of comparison.
(3) Comparisons with the BNL average spectra indicated a greater conservatism in the URS/Blume results than the other two types of comparisons.
(4) The least conservative results were obtained in comparing URS/Blume spectra with the BNL envelope spectra. In our opinion envelope spectra are more appropriate than average spectra.
(5) In all comparison methods the URS/Blume results for the third floor were found to be nonconservative.
(6) As noted in the matrices, frequency shif ts of the dominant spike of the floor response spectra between BNL and URS/Blume were found. In comparing the mass values used by the URS/Blume in their five-frame model of the containment annulus structure, with those obtained from the draw-ings transmitted by PG&E, it was found that lower mass values were used.
Table 5.1 summarizes the results of this comparison. Total mass values are given on a floor-by-floor basis. It's possible that some of the differences are due to the lower mass values used in the model.
l
Table 5.1 MASS EVALUATIONS (K-SEhT)
URS/BLUME DATA [RTA FROM DRWS. TRANSMITED TO BNL ELE % TION FRAME FRAME F ME FR ME FRAME TOTAL STRUCTUR TOTAL CONCENTR DISTR. TOT L 10\ O.21 O tR DL6 A fb 0 90 \.if l. l1 R99 l.66 lA0 4.06 l0b O. O 999 f.M #.6A Q10 'Ab8 2.4% f.10 1. 6% t.9s +.94 ll7' l.19 Lp1 1.94 1.99 1.9# 6.7] 1.#4 4.90 1 54 t.Sq 1143 14 0 O.7L '1.4f 4 19 10.4 4 10.9f ib.bi t#.11 4 64 4') 60 1 4.19 47.77
COMPARISON MATRIX COMPARISON TYPE: AT FAN COOLER LOCATIONS FAN COOLER 1 2 3 4 5 FLOOR 1 Op 1 1 1 1 2 Op 1 O gg 1 O gg 3 PF FF F P 4 1 1 1 1 1 NOIE: 1 - ACCEPTABLE O - NON-ACCEPTABLE P - DIFFERENCES IN PEAK VALUES F - FREQUENCY SHIFTS PF - BOTH
-s1-
COMPARISON PATRIX COMPARISON TYPE: AVERAGE SPECTRA i
l ECTION
- 1 2 3 11 S FLOOR 1 Op 1 1 1 1 2 1 1 O gg 1 O gg 3 O gg O gg Op 1 Op
'4 1 1 1 1 1 EDIE: 1 - ACCEPTABLE
! 0 - NON-ACCEPTABLE P - DIFFERENCES IN PEAK VALUES
^
F - FREQUENCY SHIFTS PF - BOTH
s
?
COMPARISON MATRIX ;
COMPARISON TYPE: ENVELOPE SPECTRA I
SECTI0il 1 2 3 4 5 Fl.00R 1 Op 1 O pp O pp O pp 2 Op O pp Op 0 Opp F
3 O pp O pp O pp Op Op
(, 4 1 1 1 1 1 EIE: 1 - ACCEPTABLE 0 - NON-ACCEPTABLE P - DIFFERENCES IN PEAK VALUES F - FREQUENCY SHIFTS PF - BOTH
6 i
UNIT 1 J NORTH LNT 2 CONTAMENT CONTAMENT t> 4D cwe at cRue wt
\, *
- , ,r
~, l ,r Sn L? ..
Y ~'. . [. W 2* .
p ' %y J - <
- j\ -- g--- -O_
,4 4s
,o
/% @ @ A, r
]
O O so O
f
'I %
4 FAN CNER NQ Figure 5.1 - Orientation of fan coolers for Units 1 and 2 and frame orientation of URS/Blume 20 model.
Hm g N
0 d$ 3 g N
d f I l ll 4'
- 5 5
gu 3
4
' v f$,
jNq, l +
e
+ /
o s i t to/ ~*
~
7i p 2 ~.
2 a /
' 's % / .
d
?
.I O
' , <Qi 'N 's '
~
l
/ 'Ns ,c, .w o
f ,e' o,- - 5o$
4k :::
- )g,s N
/ %g i _
- 00 at40% 85g br ,
q oc.
v .$N *#o .~
a1 tji e
k J
i i
DIABLO CANYON PLANT l UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 1 FAN C00LER N 0. : 1
- EQUIPMENT DAMPING = 2 3 4 PERCENT 4
1 l e ;,;,,,,,,,,;,;i ; i,,,,ii ;,,,,,,,y ,,i,i...,i 1 .,,,,,,,,,,,,,,,iizi i ,
i 1 .,
~ ~
i d
o BNL SPECTRA 1 l i
i yRg/BLUME SPECTRA o (ZA DAMPINC) ; I
. - i j 9 k i
i e M l
p I
z o .
i
- . I a ,
! c x !
tu d
c; - i b
o !
i c ; i l l
~
bb 0 W ,
,t}- W '~ ' R ,
' m_-
iisintirmit it iIi!i f f
.? ! ' i! ' ! I sii ir! i' ! i l i! ! ! r i a i e in i!
.i .: .. .. i. 2 i '.
. s. to. o. .o. w.
< FREQUENCY [ CPS) 4-
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 1 FAN C00LER N 0. = 2 EQUIPMENT DAMPING = 2 3 4 PERCENT
.........,,,,,, , , . . ..., ,, ...........i..,,,,,,,,,,
.......,77,,,,,,,,,,,
a l l BNL SPECTRA '
l yRj/BLUME SPECTRA :
(2A DAMPING) l4 , I hN c n no n a 5,= f p\\
h
..f \.
7 ,
g D
~
o n i iiiiiist r i iit t 3 tit,I t I i ni,: i i!! i e i,! i1 i tilit I t iiini, intiiist i 1 itniet t t
., 2 .a 6 1. 2 4. 6 10 20. 40. 60.
FREQUENCY (CPS 1 f I DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R 1 l FAN C00LER N 0. : 3
- EQUIPMENT DAMPING = 2 3 4 PERCENT 1
i t
(
- 1 i i risis; i a i j i g l 6ij j j i iri33I6ll 8 eiIl 3 l 8 43 h:k k k I I ' 3 ' l I 3 41 5 3 3 34 ' & 3 i ' i3i i i l a '
j BNL SPECTRA URg/BLUME SPECTRA $
(24 DAMPING) Ih ,
l p.3
- 3
$ A ~ 1 /
v> A \
! c.5 skk h ) M'- JA /', I i"
g da f T %%f N
I
\
\ d- $1 ) \
d / f9
\ .
c / k d'
.~
/LAN ,m t
'l i'l 'l l f ! ! L l" l 8 O ' 8 I f 1III L I I E ! I
., ., .. ..' i. 2.
.. .. io. re. .o. so.
FREQUENCY (CPS 1 l
4
+ , ~ . ., .- --- -
r---.-------Tw -ee r--v-- 7v.- av
DIABLO CANYON PLANT UNIT 1 ;
CONTAINMENT ANNULUS STRUCTURE FL00R: 1 FAN C00LER N 0. : 4 EQUIPMENT DAMPING 2 3 -4 PERCENT
,,,,,,,,,i....i ,4 i,ii4 4 ....oi,,i,,,,i ,4 ,i,,,i i i ........i...; , i , i ,i,i i i lit BNL SPECTRA 7 -----
/BLUME S ECTRA DAMPING lh i ( ,
. / .
hW r
s.
y -
,e g awt - ;
cr i i o c4 nA
, ,,,,m,,, ,, ,i ,,,i.e.iii ,,,,,,o ,,, , ,, ,r,i ! ' ,,, ,,,,a ,,,,, ,,,i,i,ii.l
., 2 .. .s n. 2 .. s. so. 20. no. so.
FREQUENCY (CPS 1 5
1
1 DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 1 FAN C00LER N 0. = 5 EQUIPMENT DAMPING : 234 PERCENT
, , , ,,,,,,i ,,,,j , , , i , j,i , i , , ,,,,,,y ,,,,i .,,,,i.,,i ........;p,,i,i.i.ii
! ?
I BNL SPECTRA f k I l URS/BLUME SPECTRA )/ F 8 l
(27. DAMPING) $h 4 , f I 4
0 g-
$0KNbV \
- g qy pp 1 l = .
, i th.v f .
b w 2 i Tj I
! B -G i
a 2 , i i i iiti!iiiil ,I i !t !i! I f iif iti:f!! i! ! ! !! !ii ! ! 2 i rieiisitiit il i! r ! i!il I 1 2 .4 6 1 2 4. 6 10. 20 40 60.
FREQUENCY (CPS 1 DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 2 FAN C00LER N 0. : 1 EQUIPMENT DAMPING = 2 3 4 PERCENT
,;,,o..i ..ri , i ,,,,,i i i , , ,,,,,,,i ,,,,, , , ,,i,i , ; ,,,,,,,,,,,,,,;,,,i,i,,ii 7 g 0
BNL SPECTRA '
g
@ /BLUME S ECTRA V) g DAMPING
=- c .i 5 J r.,t a r \
d r \
, $ b. N .* _> h s
/gy
~ '
o i t i nii n'tn iini i l i fifil f I r is istrittiini! i1 i !i!it I f rinivitt iiril r ! i liiif I t
.I 2 .4 6 1 2 4. 6. 10. 20 40. 60.
FREQUENCY (CPS 1 DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE i FL00R: 2 FAN C00LER - N 0. - 2 EQUIPMENT DAMPING = 2 3 4 PER' CENT
,,,,,,,,c,,,,,, ,,,,,,o,; .......,....;,,,o,3,; ,,,,,,,..;,,,,, ,,,,,;,;,,
i o A
- BNL SPECTRA 7 ----- URS/BLUME SPECTRA (2% DAMPING) jf , k
~
r A
/oMI A
"?
f
/dh' Ii. E'I
\
S.
\
g ~~ A,A \
~
lI / \
8 1 In .
f "d% ,
, ,t i, ,,ir , ,, ,i.... ! i ,,,,ui ,, ! ,i,,,iet ,,,,,,,,a ,,,,, t ,t,t,i ii
. .: .. .. i. 2 .. s. io. ro. .o. so.
FRECUENCY (CPS)
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 2 FAN C00LER N 0. -
- 3 EQUIPMENT DAMPING = 2 3 4 PERCENT 2 ,;
.m... -, ,,,o.i, ,, m.., ...;..;,o,....m.......o...i,.,.
I 7 BNL SPECTRA l.
URS/BLUME SPECTRA It a (2% DAMPING) gl /
7 * ..
9 i o /
a z i F IL
- b. AS A h, in Y
!~
g-
. cMYMN7\
qp V " V yy /
, ,, , . , , , , , i . _.._
,1 mi a ii,+ii .1 m m mi . . . . , . , iii.iii
.i ., .. .. i. 2 .. 5 io. 2o. .o. so.
FREQUENCY (CPS 1 4
s DIABLO CANYON PLANT UNIT 1
- CONTAINMENT ANNULUS STRUCTURE FL00R= 2 FAN C00LER N 0. : 4 EQUIPMENT DAMPING 234 PERCENT i i i i i ii ni s iiii ii,,,iiiii . .. .. 3,,i., , , , , , , , , ,
<D
,,,,,,,,,,,,,,,,,j,,,,,,
s :
a
\ !
k i T h i 33 .
4,'1 ,
BNL SPECTRA 1 S p !
4 -----
URS/BLUME SPECTRA i I
d, -
(2% DAMPING) fl} 4, 4 .
g y. -
O f f
( I i i : J ,
, T l LLJ u lss
! d- ', V i 8
l 4 y' 7
" f ,
l d
'l 24 3M l V o iiniin mitiit ii !iti f ! t iiri: r:! ,,,it ,t i!,i,i t t ,,,,,, ,,!,,,,t , ,,i ,,, i , i
.1 2 .4 6 1. 2 a. 6 10 70. 40 60.
FREQUENCY [ CPS) l l
i 1
l
. _ __ _ _ _ -- --r
DIABLO CANYON PLANT l UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R= 2 FAN C00LER N 0. : 5 EQUIPMENT DAMPING : 234 PERCENT
........,i,,,,i , i i . i . i i i ,,,......i....i,i,i.i.iii , , , , , ,,,,; , , , , i , i , i , i ; i .
BNL SPECTRA
- l o
yRj/BLUME' (4A DAMPING) SPECTRA)0 ,0 i . I h \ r ! \
. i4 g- p'g"VWBih {
$ I e, ri$s I y
o ,,,,,,,,i....i ,i,i,i.,,i ,,,,,,,,,i,, ,, , ,,i,i1 , i! ,,,,,,,,,i....i,,,,,,,,iil
.i . .. 5 i. 2. 4. .. io. 2a. .o. so.
FREQUENCY (CPS 1 l
l
1 l
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 3 FAN C00LER N 0. : 1 EQUIPMENT DAMPING : 234 PERCENT
,,,,,,,,,i.:i , , , ,,,,; , i ,,,,,,,,,j,,,,,,,,,,,, , ; ,,,,,,,,;,,,,;,,,;.,,,,,
3
- s l
~ i
?' (
. BNL SPECTRA A
/BLUME S ECTRA f}
z URkDAMPING (2 l J n y,
r g ~. A1 h
\ hA tpv l i
\
b v '
E \ :
- e. l\nu m s t e,\ d t h W a i)-i \
r
~= pd %
/
o iiiiiiiintiiii! 1 ! , !i!i! ' I ii1 1 , r,ii! ,,i:1 ! i i ,,; i! ,,,,,,,,,, ,,i,,,,,,,,,,
.I 2 .4 -6 1 2 4. 6. to. 20 40 60.
FREQUENCY ICPS1 4
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 3 FAN C 0 0.L E R N 0. : 2 EQUIPMENT DAMPING 234 PERCENT
. , , , ,,,,,,,, , i.3,; i i .......,3.., ,,,33,3 i, .,,,,,,11, ,,i ,;,;,,y , i I
P!
- BNL SPECTRA l)
N -----
ygg/BLUME SPECTRA
= (Za DAMPING) I 0
=
l }
i a
)! / '
1 E b Iti 1
- t. Ah h e '- n. o li bh I b.' -
'I. : i j
- d.' 7 f I W .
,l1 , ~ ,, , , , . . , ,, , , 1 i i , i
.i 2 .. .. 1 2 .. .. ic. 20 3 so.
FRE:'JENC Y l CPS)
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 3 F A N- C00LER N 0. : 3 EQUIPMENT DAMPING = 2 3 4 PERCENT
- ,,,,,,,,,,,,,-, , , , ,,,,,,,,,,,,,,,;,,,,,,,,,,,,,,,,,,,;,,g,,,,,,,,,,,
o g
l
- ~~
BNL SPECTRA f{
3* -----
URS/BLUME SPECTRA L (2% DAMPING) (h l, 4 l 9, / 61 ! !
f 1
= l
- 9. Ah Af l A SM( \/ I
. fft t 4
j< vu l
f ,! ,# 1 ! e ij a 2 if -
' i i ir it.' ! * ! .is! , ,!
.I 2 .4 -6 1 2 4. 6. 10 20 40. 60
-RECUENCY (CPS) i l
l l
! i
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R 3 FAN C00LER N 0. : 4 EQUIPMENT DAMPING 234 PERCENT
,F .
,,,,,,,,,,,,,,, ,,,i,,,,, ,,,,,,,mi .,,,, ,,,,,,,,,,,,,,,,,7,,,,,1,3,,,,,,,,,,
i \
i \
1 -
BNL SPECTRA / i
/ i ygg/BLUME SPECTRA / I (ZA DAMPING) j ) 4 d(# l j e* ,,r aU- H j i
} 1 i A
'yq)'w
. . A l,-
y, -
,a f) u
\
o , ,,,,,..., ,,,, , ,,,,ii, ,,,m, i ,,i,i ,,,,! ,, ,,,,,,,,,,,,,,,,,,,,,,,ii
.i 2 .. .. i. 2 .. s. io. ro. .o. so.
FREQUENCY (CPS 1 DIABLO CANYON PLANT UNIT 1 CONTAIMMENT ANNULUS STRUCTURE FL00R: 3 FAN C00LER N 0. : 5 EQUIPMENT DAMPING : 234 PERCENT 2 ,,1 . . .,,,,,i,;,,,,,,,,,,,,,,,,c,,, ,,c,,,,,., ;, . , ,- ,,,,, , ,
' l BNL SPECTRA ;
~
URS/BLUME SPECTRA 1 0
(2% DAMPING) a
- a .-
4
~
AA A \ u sM <
!) - 0, 4y)
\;\
~
wgt, i '; W
. . ,. m . . . m. ,. ., . ., . _ , , , ,, - . m , , ,
l 2 4 6 l. 2 4 6 10 20 40 60.
bCUbbbI (Chbj DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: a FAN C00LER N 0. : 1 EQUIPMENT DAMPING = 2 3 4 PERCENT
. r~t , , , , , , , , , , , , , , ,
,,,,,,,,,,,,,7.......,
/ i
/ i i /
g .
Ih i ; / /1 A i
9 BNL SPECTRA I
. {1r e ----
/ I S, ga/BLumEDAMPING sgECTRA ([/k}, ._ , .
lq o, /,
8~ $Q ' [* ' V\f*j c p 5
g~ Yllj -
U
" f o[j m,A -
, ,, ,1 .,,i .,1 t ,, , , ,a ., .t .
,, a 1 , , i , i,i !1 3 i ,
.i .: .. .a i. 2. .. s. ' o.
i 2o. .o. se.
FRECUENCY (CPS}
s
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: a FAN C00LER N 0. : 2 EQUIPMENT DAMPING : 23J PERCENT F1
- f. S.
,ii.,,,,;,,,,i,,i;,,,j; 1 < > >1 ii : ii; 1 i ii,;;;, ii,,,,iiiiij ,
- i; ; ;ji f .
I \
l \
- b
\
~
\h
] j'
- i},' ~
li b\
o BNL SPECTRA l i\
i 9 -----
URS/BLUME GPECTRA
' [j? "
i E,~
(2% DAMPING) jh ,kf ] _ ) y j $ . 'f I M .f((~f Y i a~
g ou Ai j j
f
" )
.a , nv l
t l
, ,! l i
o i. , 1
.4 6 2 4. 6 13 20 43 60.
2 1 FRECLENCY iC S}
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNUL'JS STRUCTURE FL00R: 4 FAN C00LER N 0. : 3 EQUIPMENT DAMPING = 2 3 4 PERCENT 1
1 1:mi i i.,i i,i.;,i i ; >>,io, .iii,, ,1 3,i i -
l1
, , 6, , , o i ,i.,i,j,,,iiiia
/ \
~
BNL SPECTRA / \
/BLUME S ECTRA / .
~
o DAMPING jh 4 ;/ /1 l 0
fl
!.J '
/)
(v e p;qr
/ al ,
s n$A t, d Aff '
, 2p +iw y
a ' ' ' 1 ! i i1i i i i1. ,
1 ! ii: !i! ! ! i i i i it' i i: I ! 1 !,!1 I I f
.L .4 6 1 2 a. 5. !0 20 40 60 CRECUENCY (CPS 1 DIRELO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 4 l F AN C00LER N 0. : 4 EQUIPMENT DAMPING 234 PERCENT 1 <1 r i, 1
- 1 ,1 ji ,i ; 2 i i; ,;,i;;
f i ;i giir;i :i , ii,1 i ; ;
I ,
/
- i i / ,
l ! ! \
- i ! il / i
. I I !; i 4/ \
g ~l BNL SPECTRA i ,
i h -----
URS/BLUME S ECTRA /J
[ d j 'l 1 ,s/\$,
.i h [
e m-
- (2% DAMPING
....::......, - l^[
I .
! t fs?s J
j !
[
C (4 *
,~ Ch .}sf\ <
ps
/ l 1
{
. , , , ,i . . . . . ,,, , , . .i: . .
.i 2 .4 6 4 2 a. 6 13. 2% *L 60 FRECUENCY CCP51 1
1 i
i DIABLO CANYON PLANT
! UNIT 1 l CONTAINMENT ANNULUS STRUCTURE i FL00R: a i
FAN C00LER N 0. : 5 EQUIPMENT DAMPING : 234 PERCENT l
,,,,,,,,,i..., ,,,;,,,,,i ; , ,,,,,,,,,,,y ,,,,,,,,i , , , , , , , , , , , , , , , , , , , , , ,
l \
? \ I' \ ;
l .
Jh 4 ,
/r i
~
) o BNL SPECTRA .d ,F- ; ;
I 7 - : j
"? -----
g/BLUME SPECTRA y i t ,
! 9,- a DAMPING) ,
, t .
q h ,l } ; {
g p. y q 7,j r t a ,
w -
A d,/I i <
1 C l v -
1 I e s- --
a ii si ri! e i it ! 1 f: !,1 ! ! 2 e2 r+, 1' e! ! ! i': f ! ! it i:iii! ,,,! ' ,t i!,t ; y l
.I 2 .4 .6 L. 2 a. 6. 10 20. 40. 50 1
FRECUENCY (CPS) .
t i
d
! i
- - - , - - - . . . . c. - , , - . , ,-. , . . - - - , _ _ - . , , , , - .
DIABLO CANYON OLANT UNIT 1 CONTAINMENT RNNULUS STRUCTURE
- L00R
- 1 q
v c._ r
. m T. i L, N. N. ps. -
- i.
EOUIPMENT DAMPING : 2.0 RERCENT 2 ~ ,..
l l o l 2
q" BNL SPECTRA URS/BLUME SPECTRA
= (2% DAMPING)
E*
C" CC \
. I !b Y' l QN\\
l p)v8UY .
I l' V* MI< vv s*s \b c*
3 hN i
' ' ' it i
,l 1, e i l i i .1 i ! 11
, 2 .4 6 1 2 4. 6 '.1 10 20 40 60 l FRECUENCY ICPSi DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 1 SECTION N 0. 2 EQUIPMENT DAMPING : 2.0 PERCENT
,,,,,,,,,,,,,,i ,;,,,;,;i; ,,,,,o.,i ,,,,,,,,,;;,ii; , ,,,,,,,. ;,,,,i;,,; .
a -
~
~
. in , i BNL SPECTRA j p
$3 e i /
4 -----
t)Rg/BLUME SPECTRA , j /
(4A DAMPING) / %
%_/,
S ,,~
/ d V
l \
5 i Y r 'i \ 1 S V \
! \
e v vw a r 1 i i t i t il i i i i ? i! !iti! ! I i iiiirri! iii! ! i9i!if iI iiieinsistiiii! i t i liliI i !
.I 2 .4 6 1. 2. 4 6. 10 20. 40 60.
FREQUENCY (CPS 1 l
DIABLO. CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R 1 SECT ION N 0. = 3 EQUIPMENT DAMPING = 2.0 PERCENT m, m , , , . . i , , m.,, , ,, i,u , , , ,m.i ,i., , i , ,,, , ,
I i 1 BNL SPECTRA l
=
~
1 URS/BLUME SPECTRA 9 (2% DAMPING) '
e i
l f
~
B
) 1
- d. .s 6n n1 A . ~
fmk'fy 1 Wf--- S,f k
, , .7 l
o i iiii .! > ! !1 ! ?'! 1 t :! !: ! 1 1 3 1 1 : 11:i! iirt i f i !iti! I I
.I 2 .4 6 1 2 4. 6. 10, 20 so . 60 FREQUENCY (CPS 1 DIABLO CANYON PLANT s ,
UNIT 1 CONTAfNMENT ANNULUS STRUCTURE FL00R: 1 SECTION N 0. : 4 .
EQUIPMENT DAMPING = 2.0 PERCENT
,,,,,,,,iii...i,,1 i i i;; i i ,,,,,,,,,i....; , , , ,,i,i i i ,,,,,,,,,i....i , , , i ,i,i ; ,
- a L
(
al
?. lii i 1 A. E' v? ~ BNL SPECTRA '
4 l /
5, 2
/BLUME S ECTRA /
DAMPING / di
' h#1' $ ,
h i i g y.
E
$ i d P .. I\
N
^
/
o ,,,,,,,,,i,,,,,,, ,1 i,,i! ,,,,,,m.i ,,,,i , ,i,,.iii ,,,,,,,,,,,,,,i , , ,i i., i i
.i .: .. .s i. z. 4. s. io. 2a. .o. so.
FREQUENCY ICPS1 0
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE .
FL00R: 1 , . _.
SECT ION N 0. : S -
EQUIPMENT DRMPING 2.0 PERCENT
, , ,,,,,i ,,,,, , , i , ; , , , , m , , , ;i , ; ,
mi ...i,i,i,,4 _,
- .I li n : -
o l 7 BNL SPECTRA ,
9 7{
Q -----
/ '
URS/BLUME S(>ECTRA .
- (2% DAMPING >
}
E .
i
[ ,
r
,I
$ , m~
E
~
AA Nd[ k bA 1 i 1%= -uQQ,M,\ m\
? u L.
v v1
_! ' I >! <!i!iI ! ! i l i !
e i1 1 erriil r r t : ' ! ! it :! .I iii1 2 intf ! r !>! W 2 4. 6 10- 20 40 q50.
1 2 .4 6 1 FREQUENCf [!?S1
- 7 -
A k'
,. E
(
i; '
i ~
DIABLO CANYON PLANT n
UNIT 1 C0bl.TAINMENT ANNULUS STRUCTURE
~- ,- FL00R= 2 SECTION N 0. = 1
,~.
EOLIIPMENT DAMPING : 2.0 PERCENT i
i i i isiiiii i i i ii , i i iij i ; i i i i v i i i ii i i ii i i j i j ii ,i , iiiiiii ,,,ii i i i i , j ij ij O.
V 2 BNL SPECTRA 5
^c b '
d -
URj/BLUME SPECTRA
- ( -
3
- f, (2A DAMPING) 3?
&- i
- C
< s' , -%
I. - - : ;,
d I '
s
. A i nd n> b.
~
p t r ~' v
- e.-
\Q 4Ja i .p.
/
Le .. cm i n 1 1mir'l- t in! ' i !ift ! ! ' 'n ini'ini!iriil 1 ! >1ilif f I i i i i i iiiil i n t il i 1 iIitil t t
& 2 .4 .6 t. 2. 4. 6. 10 20. 40. 60.
< FREQUENCY (CPS 1 13 .
i 4
j-A y
DIABLO CANYON-PLANT UNIT 1
, CONTAINMENT ANNULUS STRUCTURE l FL00R= 2 SECT ION N 0. : 2 EQUIPMENT DAMPING 2.0 PERCENT l
i,,,,,,,,i... , , , i ii., ; ; ,,,,,,i....,,,,,i i.,,, , , , , , , , , , , , , , , , , , , , , , , , , ,
4 y BNL SPECTRA o ' I i
c5 4 URS/BLUME SPECTRA (2% DAMPING) jt 1
j
\
) /
- ?. A d '
V + y\ d
)i Af \
4 wt w
/
o invisirl,iiit it i f ifit t t i iirir!r er: t 1 9 i!itif I f .i ,rii.,t,,,it i! ,t2 1,f f i 4 2 .4 6 1. 2 a. 6 10. 20 40 60.
FREQUENCY (CPS 1
_~
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE C L00R= 2 5ECT ION N 0. 3 EQUIPMENT CAMPING : 2.0 PERCENT aj 1 ,,
- i,,,,;;,,; i .; zii i ..i,ii >'>>i- >>i i.....i . . i l
l.
i
- i j !
BNL SPECTRA l l
URQ BLUME / SPECTRA b i (2/o DAMPING) l 0 -
T
'? .
E 'i
= !
i
!\
li E -i i j i t
i
l 0 i l }l n '
, INlf \
Y 'i bq/ -A iL l,hfslr ,:. [' '
3
~t v 'y s i I
\ '
i t
i g &,
l \. N j _
,i f ,
I o , , i,i ! i
.i 2 .. .. i. 2. .. s. io. 20 .c. so.
CREQUENCY (CPS:
1 i
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 2 SECT ION N 0. : 4 EQUIPMENT DPMPING = 2.0 PERCENT t-1 e ,,,,, ,,,i ,, ,
. i .. ,,, , :,;. , , i:i,, ,, ,,,,,, p i i, i i,i;,,,
. 1 l : ;
4 .
j i ! i
=-
T ! i 9 A I 2' BNL SPECTRA ! !
l z 11 -
l l 2- ----- URS/BLUME SPECTRA ' . !
a ,
(2% DAMPING) p ! } l E. . /)
a i I w I /
u I , / 4 l
,, i, . \
~ y7 w y -
i l , ;
j _ r,
, 1
(,W w I j
~
/
~
, , , , ,, ., .i ,, , , ,i4 1 , a ,1 .,,i,..:! . ! . i
.i .. .. , z. .. s. io. a. .o. se.
FRECUENCY fCPS}
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R= 2 SECT ION N 0. : 5 EQUIPMENT DAMPIi!G : 2.0 PERCENT
- , ,, m , , ,, , , ...o , . . ...., ;, . . . . . , - , , , ,,,,, ; ,
2 2 4 m
5, BNL SPECTRA .I 5 ----- yRj/BLUME SPECTRA t (24 DAMPING) I r e1 o
u a . .
1 e
{.7 lh )
_f g
~ -
~
mt c , ,, ., ,, ,
1 i . 1, , ,. ...i.. ,, .,1 . . , , ii,, , i 1 2 .4 6 1 2 4. 6. 13 M. 40 60.
FRECUENCY { CPS) t DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 3 SECT ION NC.: 1 EQUIPMENT DAMPING : 2.0 PERCENT j
,,,,q -
, a,
.,,,;,o,,] -
, , a, , , , o ,1., , ,
er
- e n
[ ,
i I E l BNL SPECTRA
. , l en ~
d ----- URS/BLUME SPECTRA l 5
(2% DAMPING)
/
n i
{
t j \
. h g *i g \
. i
~
l ah g. ,- r,) u] \%a -
\,
v~
~
W 'py
- s"
-b_
i l
' 'ii ' ! i1! i1 iI i ! i!.! ! l o 1 iii>
4 2 .4 6 1 2 4. 6 13 2G. 40 60.
coECUENCY (C S1 s'
' . --,,7n,.
N l'
- _ i
, y .i i
.i CONTAINMENT ANNULUS S RUCTURE c L, F - _
- e. - v r r -
r I
r L
nM v i
% nv.
s 7_
R O. M . p i lv Y r.-- a Orpr; s -
rni i .O
_am M. cuF) Ti s .
r_ . ve si mcJ j - , , , i'. i, , , , , j ., i . . i i
C l
i i i l a =i ' ,l s
I ,!
-a 2;
' i <
e - l li i
C i f I
=
t r
Bf1L SPECTRA Iil I i i J -----
URS/BLUME SPECTRA t;
i r
(2% DAMP 1f1G1 i !:i. '
W i ,
l >! 7_,
.i r- ,
f g j
- n. l i , .
t s,
mo !I g l ; /\ l% n t, sO Q , L !
~,
-},
W - . --y
% ~
-n y-i oj ! I '
.i : a 4 , 2 .. s. : n. .: s:
c,: :r Ay :Lc.
- .v- $ . l
u-uvL. 4 : _> 4s-ns m
u.: n .p. ,t , cne UNIT 1 CCNTAINMENT CNNULUS STRUCTURE nP a o *~- - -
s, ' O
- c rw T s _
+
i.
.' N:
L 4 4 v ==
^ ,*pur.:- s cyc s.,or. - _
_7 . e3 O c ovwr
'_,s r sl i 7
_us .- _i. .
I E E I t i 4 i t i EI l $ i I f l t s l Ii e l
m i !
I i l
i ,
i i 1
l I .
l
! l' l
l 1 't I
w I ! f ;) :
Ja; 1,. o
= ,
l > j i
~
BNL SPECTRA Il
!,i o 8
\
= ; I I i
-- o Z !
URS/BLUME (2% DAMPING)
SPECTRA
.l li y' f i
3 il, l,t ,: !
I b ! .
fjll 5fl f,
, 3 . 3 ,m ! u v
Y
~ . . - ,
y1
, i /,w.
t ' 'Q l
l
/y'
/ :
' ' I
! h e l
- 4. S. 13 20 80 60
.4 6 2
. 2
- ^-.D"( .~D"a
- i. - +
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS-STRUCTURE FL00R: 3 5ECT ION N 0. : 4 EQUIPMENT CAMPING : 2.0 PERCENT l
I I e, ,,;, .... , ,,;. i , .,
3 . . . , , ,,,,i, , ,,,,, ,3., , ,
.,,ii,, i ,
l BNL SPECTRA ! i h ! i ygg/BLUME SPECTRA i! l (Z72 DAMPING) !'
9- j
{;l 4 1 i
e' iV .i i
= l 1 i e- i s
x i :
A i i i h~
hii[th' !
l l
?L N ' \)
r WUY \\\\
f*
\
' 1 t l Af4'
& A o I i t. i a t + * ' ' l i' .* ! ,t1 ! t ! 1 , tit,t t t 1 r: :! 1 1
.I 2 .4 .6 2. 4. 6.
1 10 20 40. 60.
CRECUENCY iCPS1
, -[
L, M LLq,U L Fi 4 '
s o v ,Ui; i _:' F p's -
UNIT 1 CONTAINMENT ANNULUS STRUCTURE 0
F Lnnc - - - -
0 t r -
1 r s}
sj o.
v v
EQUIPMENT DAMP:NG : 2.0 PERCENT i , , , , ,i ; , , m. ,, , , i ,,, , ;i;m ;i ,,, . , , ,, , i i
- i o
. I l !'1 l j
, I
' l t y el .
1 j
= .
! j
.n !
5_
,jI l
G "I
=
BNL SPECTRA -
.a> u I c j URS/BLUME SPECTRA lP,,.I 1l I h li S i
a (2% DAMPING) -
4 !
H i i
}-
1 E 'r. ',l lrI l 0 il i !
i ! g i l . ,
- s . ,- ;\
y es l
_,tb! r
-/>l
-s
~
~
'm f-v , ~
ci t > '
t 4
- t ! . ! I i 2 .4 5 2 4. 6 O ?L 41 61 r;c
.c v6 e6
-, <"Oq' w s J L
DIRBLO CANYON PLRNT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 4 5.E' C T ICN N 0. : 1 ECUIPMENT OAMPING : 2.0 PERCENT i 4i . ,,44 , 4 ,
! i 5 i ini ii.ii ,,1 ,.;,4 iiinngiii;,i3 i3,;i, ; ;
l li.
- k 2 ll b .
i
/i J
/ j E l
! / i
=
-} k A */ ' a, A i i
9 e *1 BNL SPECTRA j r;
- Y 't
- I 2
1)Rg./BLUMESPECTRA / i. i , k .
I, id E, (4, DAMPING) '
, M n, t
. . i .
C
%, / ,
[
'( I 4
I L.) u J i t .
Se \
^
A .
%l1
/ !
a i r! ' ? I , ' '
!l il i iir f,13 t e i
.I 2 .4 -6 1. 2 4. 6. 10- 20. 40 60.
CRECUENCY iCPSI
,12, , - svns 2: - -
- u. ta . .- 1 4 m _o eIfi3I. T.
i l.
,O. Nw'
~
r n N m..i. M..M.c . ct4v T, i . v ! , C:
_ s., R,., ~ s.0m! "L Dm! im RC
= n m ,,
_ m a .. =. _- -
- r -
- 9 y.
- n _
m - _ . 2 0 _
_7 n_i;. : p s.;i
., _ ocvrv. _
u . .i r.. _ o.-
c p r r s1 -
._.m
/ ~\
,,. -e ui , i . i i
> 0 \ ,
s e \ \
l 0 t i l Dl ! l l '
\ !
i: ,
I I i e r
\
i l
. . ? !! _
E ~!
BNL SPECTRA lI! A ,d d, ' ~ lll l D \ . A , r\' >
d i -----
URS/BLUME SPECTRA / J i -
i !
g ~
(2% DAMPING) /l )
. . 1 i
E f@# n
\. l
{ ,
Z
[.(
i i
k I
- i c
y i , l l
I f:,
w" -
a
. I ,+
- , j ! ,
I . ,t ! Q' A ,
I
,1 'a '
t / $L,es o< -
j l
1
,j ' 1 .!
t i' f ' + .! l
. 2 .4 6 e 2. a. 6 :C. M. 43 80.
- ; ;w.' w .e-w
- . n. ' v -L- frOQ\ w -
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R: 4 SECT ION N 0. : 3 EQUIPMENT DAMPING 2.0 PERCENT
- st iririi sig ii6 si i ; i 4 i;rj j;i;iiij.;;,3 ; ; j i; ; i 1 i i rii.; .1 2
, i ; j ij1 ; ; ;
f .
_2
/ ; ;
~
f \
l' s.* ~
BNL SPECTRA A1
< i! ( $\
{ > L-( l ;
5 - - - - -
/BLUME S ECTRA 4'
= URkDAMPING (2 / '
i lh') ,
ge w .
C '
h b (
d~ bl \
S- # (
" by A l'%
/
o i i 2 11i it! i i t ! r ! i t ' ' ! ! t iiii it if i! ! ': ! ! ! i ii ,i i1 ! ! !.!,! ! i 4 2 .4 6 1 2 a. 6. 10 20 80. 60 l
FRECUENCY lCPS1 l
l _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _ - _ _ _ _ _ _ _
I DIABLO CANYON PLANT UNIT 1-CONTAINMENT ANNULUS STRUCTURE F L-O O R : 4 SECTION N 0. : 4 EQUIPMENT DAMP-INC- = 2.0 PERCENT
, . . . m u ,- .1 in_,,,,.,q,, . . ; , m m ,- ...q
. . 3 . i .) , , . . . . m , ,i ,..3.,i.i.i, u
'\
, e I 4
m ,
hn\\,
[ BNL SPECTRA ,
O C -----
s URS/BLUME (2% DAMPING) SPECTRA.I i
-. Ah . A b 9 I wm yt v
! ! t i!: ! ' i 1 , i it! i it i l'r !ti! ! t ,it ,ri,sitiiir i i? i!,iri ! .
c, i: it iirri! i r 1 1
.I 2 .4 6 1 2. 4. 6 10 20 40. 60.
FRECUENCY [ CPS}
DIABLO CANYON PLANT UNIT 1 CONTAINMENT ANNULUS STRUCTURE FL00R= 4 S E C T-I O N N 0. : 5 .
EQUIPMENT DAMPING = 2.0 PERCENT mi
,,,,,,,,,,,..., , , , i .i., i i ,,,,,,,,,,,,,,i,,,,,,,i,i ,,,
, , , , ....... , , ,,,,,, , , i.
i
, I O BNL SPECTRA i
- I o
/BLUME S ECTRA DAMPING jy j
Z N i .
5 u' /d k WWu A \ A i '
": A' % !
~
/
o iiii iiii!i iirt ' i ti!,! ! ! iiii !? iist i ! !ifit i! iirrii 11 i n ii! i I itifi! ! I l
.I 2 .4 6 1. 2 4. 6. 10 20 40 60.
FREQUENCY (CPS 1
6.0 Two-Dimensional Model Evaluations 6.1 General The floor response spectra that were generated by URS/Blume and reported in Ref. [1] were obtained from a two-dimensional model of the contaimnent annulus structure. This model, shown in Figure 6.1, taken from page 54 of Ref. [1] is made up from five franes that correspond to the locations of the five fan coolers located on the concrete floor at elevation 140 ft. These franes all share a conmon inside column which represents the crane wall . A schematic view of this structural idealization is shown in Figure 6.2.
i The properties given to this model were obtained on the basis of tribu-tary areas assigned to each frame. These areas were defined by the lines that bisect the angles between adjacent frames and by the exterior circumferential I beam s. Stiffness and mass calculations for the model were gotten by use of the above defined boundaries. Based on these assumptions the stiffness assigned to beams comprising the five frame model represent are equivalent stiffness of all actual radial steel girders. The masses were lumped at the midspan of each beam. Raw floor response spectra for this model were calculated using 2, 3 and 4% equipment damping.
A copy of the computer printout containing the input data used by URS/Blume for the two-dimensional model was transmitted to BNL. These data were employed to generate floor response spectra using the BNL version of the SAPV code. Computer runs were also made to verify the floor response spectra obtained from the BNL code using the STRUDL code of the McDonnell Douglas Automation Co. Details pertaining to these tasks follow.
6.2 Verification of floor spectra results At NRC's request, BNL verified results of the floor response spectra generated from BNL's SAPV version with those obtained from the STRUDL McDonnell Douglas Autc. nation Co. computer code. For this purpose the input i-
b 2 T CON TA ink 1&N T ts k --~
z
'~ f\m) %
LEGEMt3 k . ./ M' A @ - SLEMEN 7 NUM /S G/t fI h' b \
\
2 - NOne nuMSee
.& _ - . _EEQM e 3 _ 0 - M ASJ POIN T hf NOTE *
\ ey# WI; M.,s.+
w g, l/ Noo e s 1. ,. is , n . 4 0 t s:
M(\ p ARE ALONG f 09 JTRUCTURE
- ) ! AND ARE COM MON 70 A LL
~ 4.L '
- FIVG FRAM & S I
8LAN \
Nm.*
4G 4' 40 47 A to 4S AS 40 sn 4C 40 p 45 42
-C 5' 55 O
Ms so c:
go
^
g
^
WGt cr
-o - -f G4 43 1
5 5 5 3 5 5 5 5 5 5 g 5 LO EMI JC. b A8
'~ "-
47 a, p ." 11 fry, y LSI M Sf a V 54 3,4 53 e,
~ ' ~
58
~
6 Sr 6 58 0 51 p 5!
E 75 _ 'ii
.IG [1, Ti H -- E u E '13' IS 4 5 58 W 3' ~is
' ~
~ **I'~
g M --
li is '51 !! ?' '- ts - 4' M 4 5 ~ !a
- 7. " Ti
~ ~
17 T 7I n T.
- w 7'76 to ril
- w 27 7tl or -
it T
I#
_s g g 14 g is y [j,-
. _s_. _ . . .,_ _o _
- G 87
.E ~L1 .I * -
/- -
FRAM E / FRAM E 2 FRAME 3 FA AMG 4 FRAMES G L E VA TIONS Figu're 6.1 - URS/Blume mathematical model of containment annulus structure.
r.
a NORTH d\
rj(m3 NI -
FRAMtNG PLAN CRANE m LL FRAME (Typ)
'l
(
t f
amming , ,
, , ,, %g LUMPED MASS Figure 6.2 N @ CONTAWMENT ANNULUS STRUCTURE
data of the two-dimensional model of the containment annulus structure given to BNL during the San Francisco meeting (10/14/81) were used. It was also assumed that top floor to crane wall connnections were shear type, and that all other connections were rigid. In addition all nodal points were assigned two degrees-of-freeds, i.e., one vertical translation and one rota' tion in the plane of the frames. A free-vibration analysis was perfonned by use of both computer codes. A total of thirty modes were computed. These results are shown in Table 6.1. Fra this table it can be seen that the eigenvalue solution results obtained by use of both computer codes are in good agreement.
Furthennore, floor response spectra were generated for 2% equipment d ampi ng . Results fra these evaluations are given in Figures 6.3 to 6.8. For cmparative purposes the spectral curves obtained from the STRUDL and BNL SAPV codes for each nodal point are shown plotted together in the same figure. In these plots the solid line represents the floor response spectrum obtained from the STRUDL code of McDonnell Douglas Co. The dashed line represents the corresponding spectrum obtained from the BNL version of SAPV code. From these plots it can be seen that the results obtained from the two caputer codes are in good agreement.
6.3 Results from URS/Blume Input Data When the actual input data used by URS/Blume for the two-dimensional model became available, computer runs were made using these input data. Floor response spectra were generated and capared with those transmitted to BNL.
In these evaluations the following parameters were taken to be identical as those used by URS/Blume.
(a) One degree-of-freedom, i.e., vertical translation, was assigned to all nodal points attached to the crane wall. For all other nodal points two degrees-of-freeds were used, i .e. , one vertical translation and one rotation in the plane of each frame.
(b) Totally eighteen modes were used in the eigenvalue solution.
(c) The modal equations were integrated with a time step equal to 0.003 seconds. The total acceleration response time histories were computed over a duration equal to 15 seconds.
Results from the eigenvalue solution obtained by BNL together with the URS/Blume results are given in Table 6.2. From this table it can be seen that these results are identical. In addition, computer runs were made to compute floor response ' spectra. The latter were then compared with the corresponding raw spectra transmitted in digitized fann by URS/Blume. Figures 6.9 to 6.16 denonstrate the results of these evaluations. The solid line represents the URS/Blume raw floor response spectra whereas the dashed line represents the corresponding BNL evaluated spectra.
- Fram these figures it can be concluded that the spectral curves obtained
, by BNL are in good agreement with those camputed by URS/Blume. It should be i noted, however, that the number of spectrum points calculated by'URS/Blume is equal to 97. The corresponding BNL floor spectra were computed at 198 l spectral frequencies. Thus the small differences shown in the figures can be i attributed to the fact that the BNL spectra is more refined.
a I
l 6.4 Comparison between raw and broadened spectra A final task, under the 2-D model evaluations, involves a camparison between the URS/Blume broadened floor spectra and the corresponding raw spect ra. The fonner are reported in Ref. [1] whereas the latter were trans-mitted to BNL in digitized fonn. Spectral curves at 2% equipment damping were compa red . Results of this camparison are shown in Figures 6.17 to 6.24. In these figures the solid line represents the raw spectral curve whereas the dashed line represents the corresponding broadened spectra.
By inspection of these figures the following may be concluded:
(a) The broadened spectra are generally in good agreement with the l corresponding raw spectra.
(b) At low spectral frequencies, it can be seen that the broadened spectra values are only averages of the corresponding raw spectra.
l -100-
Table 6.1 2D MODEL. COMPARISON OF MODAL FREQUENCIES (CPS)
Mode No. STRUDL SAPV (McDONNELL DOUGLAS) (BNL )
1 10.64 10.65 2 11.03 11.05 3 11.45 11.46 4 11.92 11.92 5 13.16 13.19 6 16.04 16.07 7 19.42 19.45 8 22.86 22.89 9 24.13 24.18 10 24.38 24.43 11 24.70 24.73 12 24.97 25.01 13 26.55 26.58 14 26.64 26.67 15 27.09 27.11 16 28.47 28.50 17 29.44 29.49 18 30.05 30.08 19 30.37 30.41 20 32.32 32.36 21 33.33 33.36 22 34.04 34.06 23 34.20 34.22 24 37.64 37.66 25 40.89 40.92 26 41.14 41.17 27 77.16 77.25 28 150.02 150.01 29 150.32 150.40 30 150.43 150.50
-101-l L
i 20 *E* CENT ::MP!NG
, .: .4 6' t. 2 4. 5 10 20. 40. 53
, , i i ai . . . . i t e .
STRUDL -(MCDONNELL DOUGLAS) !
l 1
SAPV (BNL) 4 ,
a
-. o ; ;
i d. J r- ; i :
= t Ab di(si i ;
4, l 9E1/ 'I [ 'ig }ii' / 3 hij i
= ,
! t .
J '.
ii ., ,
+ ! V j i /! !
li E a
( !f It I i "i 1 t :. I I
= t M i jf 1 2 . Ii1 S D I i
, J, l ,% p s.
6.,.
- i. 1
- u. e. ; ;- i ; , i j 2 -
-; .,I \ l i
'.n '. i 4
- " N'. s . .
j
- w. .
I i
i i
,,, ,,s,! i!l3 ! 8 ' ! I- * ' ' 'i
} 1 r
- , 2 ., 5 . 2 .. s. ia. - m. 0 so.
FRECUENCY ICPS1 N00E NC 20 A
Figure 6.3 - 20 Model. SAPV (BNL) versus STRUOL (McDonnell Douglas).
Node 20 i
}
i
-102-
I l
I i
i J
l i
2C PERCENT CAMP!NO
.z .- .s t. z. w.
c
.. ,. 2,. . ,e.
i !
' i
! l
! i i '
, :1 l j i!t t
i-lif i
_6. i i' r '
.5 f
ll l i STRUDL (MCDONNELL DOUGLAS)l r 1 1 3 1 ----- SAPV (BNL) ; .t{
l !
- , ! iu
' -l ! I, E '
I -
!, j
=
w I
A S , ,,; t .,;
k(q\s./ ui ,
5 gll j
fY ';' l $-\".
N. ,
l l N'.
ifu[Ig 1 .ex l
W
'i 4 l
/ i ;
= , . .. i , ., . ... , ,, . , , , . .
. .: .a .e s. : .. s. n. 2o. ,o. so.
FREQUENCY (CPS 1 NCCE NO.4i Figure 6.4 - 2D Model. SAPV (BNL) versus STRUDL (McDonnell Douglas).
Node 41 1
-103-
. - - . . - . , --., ,n . . - . - ., , - - - . - , _
, - - _ - . ,a.- , , ., ,- _ , , - ,, ,
2.0 sE9ENT ::Mp !?.G 2 .4 3 2 4. 5 .c. 20 40 50 i $
l i
l t i t !'
i
\!
l fl!
l11 T l
- , t !
l 33 ii j ; STRUDL (MCDONNELL DOUGLAS) !- i
=
SAPV (BNL) , \\ ,
Z x
=
i.
i Ia A/
i i{
I i
I
_ i t
- ! A S /!i wltj-k I ! l 5, i lyd[/ ' '
! v' Ih 6
s.S
[\ * .
/
,i i -t i I i ]
2 .4 6 2 8. S. 13- 20- dC- 50-E00ENCf ' CPS. NCCE NO.42 Figure 6.5 - 2D Model. SAPV (BNL) versus STRUDL (McDonnell Douglas).
Node 42
-104-
l i
l l
I l~
l 20 PERCEMT car 1 PING
. I' 2 .4 6 t. 2 4. 6. 10 20 40. 50.
i.isit , ,.ia iia 4 j , 4 . iaiii.i
- iI & 64ia 4 4 >i' ii i;i i e i:6 e i D
A N
4 _ 1
- E t.
e 9 d ' f
- ' ((
l
! = .
STRUDL(MCDONNELLDOUGLAS)f J ()
hj
~.
SAPV (BNL)
I d !
I w
i f q
, 4 h ,f
= '
IpV ' u :.,
\
/
f
/ -
o r r!iiiil i *i!r9it ! ! ,e t rie ..tiinit e ! t, fit i t i. i,,,2!, ,1? e e.!., es
. 2 .4 6 1 2 4. 6 10 20 40. 60.
FRECUENCY (CPS 1 N00E N0.43 Figure 6.6 - 2D Model. SAPV (BNL) versus STRUDL (McDonnell Douglas).
Node 43 k
-105-
2.0 *ERCENT C @ Pl:.3
.4 6 : 4. 5 13 20 40. 53 1 2 1 1 i i . . i i i .j i . , , i 1 4,
i i i
i l j
d !
l !
i l\l l i
1 1
> I l 2 '
i T l i i
., I z
d ,, l l ih l l
~
_ . STRUDL (MCDONNELL DOUGLAS) i lI Il I
= ,
5 '
SAPV (BN,L) d. )l l
=
X l
f\ A l i
b
' i AA . A d N A ) l> St !
), ,
~
\
\ \)Y}\/ # ' ~ l
) .
WM ,
i /*
I V {'
l I e r 1! r
' s .!,' iI ,6 ii .i ! i.i ! l 1 i ! : r 3
2 4. 6 ;3 20 40 60
. 2 .4 6 1 FREQUENCY lC?S: NCCE NC.44 Figure 6.7 - 2D Model. SAPV (BML) versus STRUDL (McDonnell Douglas).
Node 44
-106-
l I
20 ?ERCENT CAMPING 2 .. 6 t. 2 .. 5 10 20. .o . 50 i i ii e ,a i i.I ii t il i i ,iii4 s 6: 'iii 2
. ii e st i e
'! l
! l 1,
\h '
I
= ! !j
~'l
} . STRUDL(MCDONNELLDOUGLAS)Il E* -----
SAPV (BNL) j 5
-a I L .Ab \
i~ A\ k da$ j h.
l *** . f i
\'
R \d. \t 3 - ~s,,
A
\ !'.^
- i. . .'
,- t j-y i- w s
/
, , , , , , ,,, , . , , , , , . , , , , , , , , , , , , . , , , . . . , , ,, m .,,- ,,, ,! ,i.,,i,,
.i . .. .. i. 2 .. s. io. ro. .o . so.
FREQUENCY (CPS) N00E NO.45 Figure 6.8 - 2D Model. SAPV (BNL) versus STRUDL~ (McDonnell Douglas).
Node 45 l
-107-
-. ~ .
_ 3 4
s.
, w .s-~
- i,, .
~
- ?
.4 '
l' _.
Table 6.2/ '
20 MODEL. COMPARISON OF MODAL FREQUENCIES (CPS)
' O Mode No. SAPV (BNL) URS/Blume 4 s" ..
I 13.20 13.20 ,
2 13.56 13.56 -
['
~
3 13.81 13.81 .
)
4 14.17 14.17 h : I
-> 1 0 '
16.01 5 16.01 &'
6 16.31 ./ 16.31 [
.. <?
, c-7 19.86 19.86 .
8 23.57 23.57 >
9 24.89 24.89 ,
s 10 25.35 25.35
+
11 25.73 25.73 %.
12 25.80 25.80 -
13 27.73 27.73 14 28.09 28.09 15 28.19 28.09 Ic 16 31.13 31.13 +.
w 17 31.72 31.72
~-
18 32.12 32.32 .; <
k
,3 7 s
~
S
, P'
-108-
. -n I
a,
. - 1 , i.
a ;
A N I (
~
N ,
2.0 PERtENT Cep!No
- -s a. z.
- t a. s. a. n. ,a. ,,.
i ' '
/ f , 'i ;
- 4
- -
[e g
m- 1
\
, y
_c. .
. t
/ !
i
- t. .
I l
.. ' . -'; l, , '., ,li r, w -
i1 3 ji .
_1 .
I, ii i hi
~
- 3 l - URS/BLUME t i
- +
f i ? /!j 1 '(/l (l l '11 >
5 l _----- SAPV (BNL) )/ Ul' ,iA. i i
'.k s: - - - i t if E i l p n l' 6
W I' j o x , s
,u y! <
,1 lYY i
. li !
-,- d
-cg ,
,m ,
- , i 'l
- f, u. 'l
- i :
i.\,
4 ,%
- l ;
i i ,
^
" I !
t
! i
. m. ., , i,,. - . , l ,1 ,,,,,,, .., . .,, !, ,, , , , ,,
- s t > n. z. a. s. ,a. n. ,a. ,,.
- EOUENCY !CP51 NCOE NO.8 Figure 6.9 . 20 Model . SAPV (BNL) versus URS/Blume floor response spectra.
Node 8 4
/*~
. +
Nr g%
f me w:,
l _109 5 \+. w- ,e
- a >
i
) 2.0 PE* CENT CAPP!NG
, 2 .4 .a i. 2 4. 6 to. 20 so. so.
1 . , i 2 ,:4 i, i io,. .., ,.i., , , . 1; .
, 4 , , ., .,,
R I !
4 a
- .i ,
\\
r,
- U; s
I '
Si URS/BLUME i
. ! : 4 D l ll SAPV (BNL)
i
~..
! L. i ad n q ,
' 'f ',*}
f {(
=
s t
i w I ..a j i }I i C i t.J ji I
= ,
7 l, '4\.
e.
1 '
i /
,.[
,/
$ ea i e ! i : 9: it i , ,,iri et i! i i !,t t e :,,iinitii1 .f a + _!.!
a i it
.I .2 .4 6 1. 2 4. 6. 10 20 40. . '30 FREOUENCY (CPS) NCDE N0.9 Figure 6.10 - 29 Model . SAPV (BNL) versus URS/Blume floor response spectra.
Node 9 1
4
-110-
20 PE9 CENT ^AMPING
.i 2 .4 .4 1 2 4. 6. 10 20 40 60
, 1. , , , ii .
m si >>i > , . . , , <>>>ui . >i.,iiiii,,
I i,;-1 I
S; 6 i
I h i j
l i !
r
,,. a
~, i i e i I
' ' j .i 3 URS/BLUME dd;
.a A
i
' l'-i A 1 +
d -----
SAPV (BNL) j 1jI - i' ,1 ,
,- ~.
i
\
j
, t ,
n
' l E i i I ]i g i i
=
O hh. g l
4
/ il 1
% ! f 8 b~
o s$ .
'M .,
l CC t I
/ i. !
,~e s
.,s a i riefr e it i ! titif f I i resis.iti if il i1itif f i einii sii!in,it i! i?,e,f I f
.I 2 .4 6 1. 2 4. 6. 10 20 40 60 FREQUENCY (CPS 1 NOCE NO.10 Figure 6.11 - 2D Model . SAPV (BNL) versus URS/Blume floor response spectra.
Node 10 i
-111-
1
\
sl I
f 2C *ERCENT CAf"P!NG
' 1 . .4 .6 5 2 s. s. ;0 20. so. go.
I i- 6 e i.ij 4; ei e , , ,6 , , i 1, , ,
,g i l i i i i e
i 4
, l
' i t :
! 'f '
T a[
O f ['i ,I '
O l l
- i .!
I Ji el i i i
~
~! URS/ELUME I , b i ,
= l j i fi ! o
I SAPV (BNL)l l ; l i,
. '\\
b E .{!
- ~
e
- I V ,j 4 , filj l il f i f
~_,
, ." l i[
. j C - .o "
r II 'i p .
.~-,
l
/ j l
1
, . ,,! . , ! , 1 '1 r! , ,,,,a, ii. , , . . , , ,, ,,,,,o, ,,,,,,1 ,,, ,,t 4 2' .8 6 t. 2. s. 6. 10. 20 40. 60.
FRECUENCf ICPS1 N00E NO.I1 Figure 6.12 - 2D Model . SAPV (BNL) versus URS/Blume floor response spectra.
Node 11 4
-112-
2.0 *E9 CENT CAMP!NO
. 2 .4 4 6 2. 4. 4. ;0 20 40. 60.
s . . . .
.l' *+6- ri . .ie i. . i . . .g a; '
I i 1 I 4
A
' l- !.
.t j .
-- ..~;
li .h i !
, 3 i .
E i URS/BLUME pfE - k i
}l{
.1 \ f
] ! ----- SAPV (BNL) lj. % '\{
!' 'l I
=
E x.
\-
\
i aj b,
l k)lI iI I
- ~- l l
- w. i W
i
./ il
- t i
/ O' 's \. l r/l k ;,
i l
.i ,, . , ,. . . . . , , ,. ..,,,.,a , , , . . . . , ,,,..r. ,. ...>,. ,,I
.i 2 .. .i . 2 .. s. 10 n. 0 .0 FREQUENCY (CPS) NCDE NC.12 Figure 6.13 - 2D Model . SAPV (BNL) versus URS/Blume floor response spectra.
Node 12
-113-
l L
2.3 PERCENT CAMPING
.t 2 4 5 1. 2 4 6. 10 20 40. 50 '
6; .; , 6 ii6 ,,, . .4ij 4 'iil 6 iiii;6 6 e ii 8.is6 i, 4 i : ia i i ,
a s
! l t 1 ,
I 1e
=
r ci .
l I, r
b i
~ ;
URS/BLUME ,,h ;
1
, , l h!
g =i SAPV.(BNL)
))
i i
g i pi
-- ,n ,
I i
- a C l I,
E x -
" l', 1 i ;
% 'y .g H ,. tl \L d j
'\
i ..
b pl f i *
,e
,* . ,i' if , t.ti! ! ! iriinnive!iirrf i! i 8 ? In ' ! I r irt i t l' '
'f 't ' f f f a ii l .g 2 4 6 L. 2 4 6 10 20 40. 60 l FREQUENCY (CPS 1 NOCE NO.19
! Figure 6.14 - 2D Model . SAPV (BNL) versus URS/Blume floor response spectra.
Node 19 i
4 -114-t
- --, - , . . . - , . . .- v.- - - .
-3 -7 - , - - , , n r-
2.0 PE4((Ni C MP!NG
. 2 .. .s t. e. to. :o. .o .
- 4. so.
i ,
. .i. ", >>,. , ..izi .i i
y- i i l' T j c 4. fiP 7 li i 9
a- , .
f' .f .'.{ !
i i p
= ! A E ;\ ,
URS/BLUME
>\
j A>h -- -
\\
E ! ,, I I l.
k !
2 -----
SAPV (BNL) '
f '11 i
$l i '
%{Q ,
lff) ,
~! I s/ 7 I - I l p i
l
} -1 i l
o ,
, , .,1.. .. ,,,mmi . , , , ,,.1 . ,, , ,, ,,. .l
.i 2 .. .. i. 2 .. . to. :c. .a. so.
FREQUENCY (CPS: NCCE N0.30 Figure 6.15 - 2D Model. SAPV (BNL) versus URS/Blume floor response spectra.
Node 30
-115-
1 1
l 4
d
(
b
- j. I i
2.0 PESCENT CMP!NG
.g 2 .. 6 4 2 .. 6. 10 20 0. 60
.uu > o , , t i>, e4 > . > > u>, o . i i;. . . :
i I
i =;
i i j I i
r i URS/BLUME'
= l.
SAPV (BNL) e i 3
rl
=
_a_
- n c l Z &
.- s l On j 5 i A4 .A h bA
- 5 x, i y w r ,
\ g i
-e i! i,;1 iI ! ! !t 8 * ' r it ? f f f il r *1 if e ' I i! 1 ! I r ii: 1ieieI # - t I r I !?ri 8 !
3 i .i 2 .. .s i. 2 .. s. io. 2o. .o. so.
FREQUENCY (CPS 1 NOCE NO.41 i
Figure 6.16 - 2D Model . SAPV (BNL) versus URS/Blume floor response spectra.
Node 41 i
i
-116-
2.0 PERCENT CAMPING
.n 2 .4 - 6 g. 2 4 6. 10. 20 43 63.
1
.i ., i i+,,ii i'isi'6 i i6'a's 8 i '''''''' ' i'I d l' I
l URS/BLUME RAW SPECTRA y 2 -----
URS/BLUME BROADENED i ii
, SPECTRA. i i
= I q
i :-
- ' j i
\ l. . i :
w : -
) 1 'A r f / -
j i E ] f}.N f ',i !
n e c i q',' c x
i !
l u ; l d i (I ,
/ t I d ti u C '
' h l
\\
l l
\
s l l
. . . . . , . 1 , ,, ,, , ,,
....it......... ,i..- l
., ., .. 6 i. 2 .. s. ic. ro. .e. so.
FREQUENCY (CPS 1 NCCE NO.8 Figure 6.17 - 2D Model. Raw versus broadened URS/Blume floor response spectra. Node 8
-117-
l 1
2.0 PERCENT C@ PING 4 2 .. 6 i. 2 .- .. 6 lo. Zo. .o. 6o.
, . . , - ,; , , , , i 1 ..y, y , . - .i ,. . . .,. . ,. o , 4 . . . . . , ,
i 1 a
! l
- , Ii i l .
~
5: f \ j\ r s T f \ A }, t' r l s
0 $ ;
- d. , ' )
, e h' t '
i- ! 1 i'
q gj-
= -
c
< y e i
4 I n
L L i 1
E 1
\,
- a , ,
d2 u lv b u it
= s
\
URS/BLUME RAW SPECTRA
.' s URS/BLUME BROADENED SPECTRA i
1 g f 8 I
! I I !I k I I 2 l f I f ' ff 1 I I ! ! f I I l l t f l 11!l ' I I ! l l l f9 f $ I
.i 2 .. .6 i. 2 .. s. io. 2o. .o . so.
FREQUENCY (CPS 1 NODE N0 9 4
I Figure 6.18 - 2D Model . Raw versus broadened URS/Blume floor response spectra. Node 9
-118-
2.0 PESCENT CF.MPING
.i 2 .4 .8 1 2 4. 6. 10. 20, 40. M.
o . . ,.. ., ;. ,
....,,,g. ..i. , , i.,.,,,
,: i i
=
1 =
/,I !
l <,
- ,_ li .._] e ii
~.
O
, ,, a 11,
, v, j\ n t
l l '
T
= e I f' \ ;\
1 / l 'i Y
y 1 \
,t i. l
s
- , ? l \,I1 * (
= 1 'g W i, '
= '
l l lj 1 e i' i 1 9 I / "A i
~0,
. O "9
- b: if '
I ./J $
i a' URS/BLUME RAW SPECTRA s li URS/BLUME BROADENED SPECTRA i
,i.
,i ,, . ., ., ,,,, , , , , , , , , , , . . , , , , .. .. ..... , , , ,, .
.i 2 .. .s i. 2 .. .. 13 20. .a. w.
FRECUENCY [ CPS! NODE NO.10 Figure 6.19 - 2D Model . Raw versus broadened URS/Blume floor response spectra. Node.10 L
-119-4
I
\
l i
l 1
i i
20 PERCENT OGMP!NG 2 .4 .a i. 2 .. s. io. ro. e. so.
.i i ,,,:,i. ."o .i,,
. ..... ., .;....o . ., .
,.i,.i,,
l i
- ' URS/BLUME RAW SPECTRA URS/BLUME BROADENED SPECTRA F;h l)i r a! ' 1' I t
=
l I j
\
i
-} '
Is'
\ ,
d& '
h )
=
/kI
/
1 b I I' dJ ,'
i E m; l ] l by T l e i 1 w [
w I J l-1 lI M f I
-)-
yI j i v
/
i l' (
' \ ,
l '
l i
.irii,.] e ,,,,,e 9 , ,,, ,, 11, ,
,, r
, ,,,,p r {
3 ' 'It t! 'I ! f r 'i! ! t ,,i 9
.4 6 2 4. 6. 13 20- 40. 60
+1 2 t.
FREQUENCY ! CPS) NOCE NO.11 Figure 6.20 - 2D Model . Raw versus broadened URS/Blume floor response spectra. Node 11
-120-
F 2.0 PERCENT CAMP'NG
+1 2 .4 6 1. 2. 4. 5 to. Zo. 40. So.
! o .... ,,i .m . u . :
,1
,!,mm , , . . . - , , ,
l
! il
. ! I 1 il r ~! ,
I a i l l 3 I T f\
a 1 - s I h/ l lj
}
.s .
- I z
o \,' '
2 l 1 1
3E i \
L /
a ~ .l u E l 5 1
! I
,; URS/BLUME RAW SPECTRA 'Ds '
s l -----
URS/BLUME BROADENED SPECTRA l l
. , ,,,,, , . . . . ,i,......., ,,,, . ,,,,,,,L,...,,,, .
. . . . ,,i
.i 2 .4 .s i. 2. 4. s. io. ro. 4c ., so.
FREC'JENCY t C?S ! NCDE NO.12 Figure 6.21 - 2D Model . Raw versus broadened URS/Blume floor response spectra. Node 12
-121-
_ . - _ . _ . . . - . __ _ . . . _ . __ . . . . . _ . _ . . .. . .. _ = = . _ . . .
i i
i l
20 PERCENT CAMP!NG
., ., .. 6 .. 2 .. 6. ... m. . 6.
5
- i . . i e 'i . . ,'.~i ae is 4 6 .
e . i Ej i
- i i , . Il , .\
- ~u ; ;
E i 1 i/ j , g -,
! .1 sgi fl l
< / t 1 F
~ ,\ _
~
' *i 3
/
l / g b '
[ t C 'd i
! C" v i
?
9
- t 1
.i C + T C I N
\
{ URS/BLUME RAW SPECTRA ,
, ----- URS/BLUME BROADENED
! SPECTRA i
t- I i! t ! *i, ,,,i ,,,,,,,,. ,, ,, , ,
s' 2 8 i , t 4
,e ,,t e i.e,; i; g,, , ,
1 2 .4 6 l. 2 4, 6 10 M. 40. 60.
FREQUENCY (CPS) NOCE NO.19 Figure 6.22 - 2D Model . Raw versus broadened URS/Blume floor response !
spectra. Node 19 i
i -122-
23 PERCENT CAMP!NG
.i 2 .. 8 4 2 .. 5 10. 20 .Q. 50 l 'il i i e6 , 2
, e i. ii 4; i- ei a
, .6, , ,4 a.
i .
I si ,
f URS/BLUME RAW SPECTRA
,. - i, URS/BLUME BROADENED SPECTRA fi 9 l a
~
a; l
Ij'i
= i l q 4 i b . i !
l ,
}" 'J< VL '"
~
~
( l l
[ '
'i
.! -, ,,1
, , . .,,,. 1 . , , ,1, ,, . .
,,,i,! , , ,, ,, ,, , ,.! , . ,
, 2 .. .. i. 2. .. e. io. .m . .o. ca.
~
FRECUENCY [ CPS) NODE N0 30 Figure 6.23 - 2D Model . Raw versus broadened URS/Blume floor response spectra. Node 30
-123-
i l
t l
i 4
1 l
i J
l i
i I:
2.0 PERCENT ORP!NG
$ 1 2 .* .6 s. 2 s. s. so. 20. so. so.
i 2 , . . , , , -.u. :
. iz , , , , , . .y .,, , , , , , ,
! 1 l '
l r, ul URS/BLUME RAW SPECTRA i ji l i i l ' t i E i URS/BLUME BROADENED ' '
7 I SPECTRA I L
- ) I .
\
.: I l i .,
_- s:
\
.I f'
, ' \
G m
\ o i I I 2 '
h
, E u-
,' i i
/ L I
, i u
(< h . INA 1
&z 4x r (,
\t
! (
i
< ~
y
- l :I t ' 'iti' ' t ,, 1 1, i t , , , : ! ',i,t,I f r 1 1 , , , , , , , , , ,, ! , , i , , , , .
i .I 2 .4- 6 1 2 a. 6 10. 20. so. 60.
~ FREQUENCY ( CPS ) NOCE NO.41 1
i
- Figure 6.24 - 20 Model . Raw versus broadened URS/Blume floor response spectra. Node 41 I
-124-
7.0 Description of Piping Mathematical Model There are approximately 60 piping problems associated with the annulus structure of the Diablo Canyon Unit 1 power plant. Of these some 25 problems required reanalysis by PG&E or Westinghouse as a consequence of the diagram-error. That is for 25 problems the response spectra defining the seismic loadings were affected by the diagram error. As a result of the reanalysis PG&E indicated that pipe stresses increased by 4-5% whereas support loads increased by as much as 100%.
During the October audit of the plant approximately 10 piping problems were given either cursory or detailed reviews. Although inconsistencies in analysis procedure were noted the review indicated that the piping systems would not experience significant stress changes when analyzed with the corrected spectra. Following the audit two PG&E piping problems in the annulus region were selected for detailed confirmatory analysis. These were problems identified by PG&E problem nos. 6-11 and 4A-26.
7.1 Problem 6-11 The problem designated 6-11 is part of the safety injection system entitled " Safety Injection RCS Loop 182". It consists of 12 IN, SCH 40 stainless steel pipe run between elevations 102' and 110' and extending over an arc of 130 of the annulus region. It is bounded by two fixed anchors and is supported from annulus steel at approximately 13 points. Figure 7.1 shows the PG&E piping isometric diagram with BNL finite element model node points for the problem. A pictorial representation of the annulus structure and the pipe are shown in Figure 7.2.
The finite element model of the system consists of 58 pipe elements, 75 nodes and 16 boundary elements. Of the boundary elements five are snubbers while the remainder are rigid supports. The pipe weight density was taken as 8.708 lb/in and includes the weight of pipe, fluid and insulation. The system
-125-
design temperature and pressure were taken as 400 F and 600 psig respec-tively. The material modulus of elasticity, Poisson's ratio and coefficient of thermal expansion were 27.7 x106 psi, 0.3 and 9.35 x 10-6 IN/IN/ F respectively. All of the above values are consistent with the information provided for this problem.
The node numbering for the pipe elements of the model are identical to those used by the applicant. Where it was found that additional points were required in the BNL model to represent the systen, these additional points were given the prefix A (i.e., 29A). Although the node numbering is consis-tent with the applicant's, node locations are not identical. The BNL loca-tions are consistent with the as-built dimensions of the system and SD bends, where appropriate, while the applicant's node locations correspond to design dimensions and long radius elbows.
7.2 Problem 4A-26 The problem designated 4A-26 is part of the coolant water system entitled "CCW Supply to Reactor Coolant Pumps, Part 3". It consists of 6 IN, 4 IN and 2-1/2 IN SCH 40 steel pipe run between elevations 101' and 110' and extending over an arc of 60* of the annulus region. It is bounded at two points by anchors while the remaining end terminates in a three-way rigid, translational support (node 25). It is supported at 13 points from the annulus steel.
Figure 7.3 shows the PG&E piping isometric diagram with BNL finite element model node points for the problem. A pictorial representation of the annulus structure and the pipe are shown in Figure 7.4.
The finite element model of the system co6tists of 55 pipe elements, 84 nodes and 20 boundary elements. Of the boundary elemerits five have been taken as flexible elements (horizontal supports at nodes 16, 19, 29, 32 and 43) to confonn with the applicant's model of the system. The pipe weight densities were 2.623,1.639 and 0.656 lb/in for the 6, 4 and 2-1/2 IN pipe respectively
-126-
and includes the weight of pipe, fluid and insulation. The system design temperature and pressure were taken as 125'F and 500 psia respectively.
Unlike problem 6-11, the design pressure and temperature f'or this line could not be detennined from PG&E supplied data. A conservative value for the design pressure was thus used. In any case this pressure value will have little affect on the overall stress results. The material modulus of elasticity, Poisson's ratio and coefficient of thermal expansion were 27.8 x 106 psi, 0.3 and 6.25 x 10-6 IN/IN/ F.
The conments regarding node numbering and dimensions for problems 6-11 apply to this problem as well. The values used for the spring stiffness of the flexible support elements were identical to those used by the applicant although attempts to independently verify these values were inconclusive. In any case these values are sufficiently large so that the supports will appear rigid.
As mentioned above one branch of this problem does not terminate in a fixed anchor. Instead the finite element model includes some pipe from an adjacent problem and it is obvious that a structural overlap procedure was used in the evaluation. As indicated by the dashed portion of Figure 7.3 the overlap extends over two supports of the adjacent problem. A review of other PG&E supplied information indicates that the complete overlap includes all the pipe to the anchor at node 1 and the support at node 16. The overlap region then includes six support points and one anchor with rigid restraints acting in all three coordinate directions. A somewhat better choice of the overlap for this problem would have included the next two restraints from the adjacent problem in this problem. However, the overlap employed does meet the intent of the requirenents regarding overlap extent expressed in NUREG/CR 1980.
Other recommendations regarding overlap support forces and input spectra expressed in NUREG/CR1980 were not adhered to.
-127-
e
' .0 lt 1 NORTil y M N g*'et 2 Z X 4 +*, 30. (H) 6 GiLO8AL COORDINATE - 4'%, ,g
- 3 4 361 6Y6 TEM s, o
{ t *I D
1 wp ,
& M 1(H)
'7*N' , o o d g;, j e
(wpu! a
.m ,
s 4 68 m' W Sid T'1(N) 'N n,4 W wwq sz, g g
\ 47
- 4. .
49 EL. lO7 '-G '/o.
g
, , 2*,39'-3* >c 4(-
SS M '
j k {6 s > c 71 hc dd g 43 \ *7,02*-34"(H)
TfW S g*,,fg'.oy (w) i (W [W y 12 3
3.
- ##4[ #
GE i / %,o 22 I'I
- cT3Y( I -
9g to '
3* te*,32'- 14"(H)
D II 31
) 32 ano i:
w m
- NoDn mint ANALYSTS No. G -11
_ oggeogg REV.No. b 11-30-01 g [g -
rixtD swD NOTG:
X - PINNED END. SiAN REVERGAL REQulRED
% X- ePRINCn HAN6ER OR Tt DIABLO CANYON NUCLEAR Q -
unir 2 8Awa R uNif I CONFihuRATION)
FOWER PLANT UNIT I 4 '2.
f%{ E REF. DHCt No. 437989 "
M. 43i $N.h.5$$E,# $AFETY tMJECTION RC6 Figure 7.1 -
A.. v o n t,. w ,.
v
. t.'"/5(.?;2.':'obu*m Lo0P No. I J 2, PART- 3
O E,*h.. ! !
- fff , : . ',,.'[
k ,"
J
,/
- x.T
_L lj@
i h,{nua w .k %)g..~gT-fr.w w '
l j
j '5 % ,
l '".. ...
- % y,,.
.ll@., .-gd,s*,y ,. / I k- pp
/ l l .
k V' I' l," Y- $.; l,lI l .fj.'
. . gr t i ,
s _
.a- ]-_T~t i
- I-f, , .,,_
) /j .;- l 1 l
h pgg, p,.f' t
' # *' /r $
.y l, M' e -( /
1 x
e I
\.
- J- ,-
g .,5.y. :
s
- l ,/. .
j ___
'r
" ., g .a g ,l ,
kuwp'i I' d' '\ .
!c ) Q
[Q\
\
-:,C ,
- -.O '
.!, ! /
^
, .... .. - e g
~
,j
$ f< l '. .
se s fk v ,
[
g{ jhj
~
N :
' %+g -
.) b _.. l.8 / _ / .. v , .
,, / 'R 1 fsL. udr, l ( J ,
N; .- C, ~t.
W A ...... m
- .I N. ,.,
! E i;
-129- g Figure 7.2
WM , (@
v g g LEGCND
, y .' -- stooAL POINT
O L 73 WEL.IO9' "'(9 a
, < ct t l* g3 cuaveo neueen
.q j W,.,lb (V M PINNeO END 14 22 /' N & Fl.reO END f ,,' W 4g >
' k pe se -
- seurs unr v^uO f /6 fAA &
y // A' s RG.{ E. REF: ONG. HO. d46495
?s,,\' ' %
f at .gA ,y o >r
~
/ '% # so ~ u sit D %
7
$9 0 74 4
Mb ll 8s #3 g
< 1*>Y 13 'og, 3r, f Old Et. fo ,
t y
^%
v,,
h / #, S h
/ b& Y o
\
1t [ $1 '
p 6
$6 Ag yy & 4, v'Mlak g y OR g gen 30 gD e
1 'o . Y\ %
ORIGIN AY M. R I is ., , ,
i ggg au .N, # GLOBAL COORDtAJATE A S YSrEM N 127-
,n m:=e ~=. = , .
Ig, &$ y, ,
5'bI.555?$;15$.?.0Esa
" r tu u. . p. 4.s, t21 .5 ?cN 'R sY s
35 -
c a et 31 $ q* O , DIABLO CAANOkt NUCLEAR POWER PLAAIT UNIT NO,/ ANAlTSO 4 A 26 vf CCW SUP. TO REACTOft " 'Y ''#'
Figure 7.3 COOL Aur pyypS DART 3 o.= a . mc . nsoc.4, es e,.c.em
.qd, e e e p.qg
, , - - --t --- 9 ,'
- ,, rp/ , -
",.. N -.., ;,;
1 l l . ,h f . ..,gm v : , ,
s
. . . .p ,
l ?g ,
W m . g< .* *y,_l--
a -,
/r ~,
l '
s2.l A \
-. .l< a I a l -x.
- p/ s m-. - tw a k I i t.'h ~ -
D
/
h )g =Ea.,,<,, ,
, ~'
- I :
l[ '..
a
.- Q fl Tj I
.fl ..' .
4 -c _egd_y,y-9__b.N r
/p i
'p y. - -q$ .;-TQ.JI.. Y, e-f "2p,!
-l j', 5 '
! : '.. g p g. n.r '
t U / l I l E' Oi i [f _,
d g,// /
Tl
~ *
.l
$I .%Y / l
- W~~ --- . ,
. .i a
p - "..'i 't '
/ ,, ' ,,', @
Ji A. i,4.'(
/ l i .
$ g; .
~
b h ', s, i
j
...--:.~- ,
l, i /
g
^ ,.
e
_gg_/. ~,% 7h
.C _,
4;$ y
- 1
~. -'
L ,/ ... '4
!j , -
$ / y A..,,in 5 i : )
k N \,$ /
N.
' %.e'~ .r ..
'b' ' ' ~ ~ ~ . . . ,
~-~~
/ --(: .i hh e
4:f$/
-131- Of(
Figure 7.4
8.0 Piping Evaluations A number of evaluations were made for the two piping systems. These included:
(a) A frequency determination for problem 6-11 using the PG&E mathematical model (b) frequency determinations for both problems using BNL mathematical I
models (c) envelope spectrum evaluations for both problems using BNL models and PG&E supplied spectra. These were perfonned for X-Y inputs with no clustering, Y-Z inputs with no clustering and X-Y-Z inputs with clustering (d) envelope spectrum evaluations for both problems using BNL models and BNL developed spectra for annulus structure models A & B for all the calculational modes mentioned under item (c)
(e) multiple independent support response spectrum evaluations using the BNL models and BNL developed spectra for the annulus structure model B considering only X-Y-Z inputs with clustering (f) ASME class 2/3 evaluations for both problems using BNL models and the annulus structure model B envelope spectra (g) ASME class 2/3 evaluations for both problems using BNL models and the annulus structure model B individual support spectra.
In this report only the essential and final results are reported. These include. frequency comparisons, boundary element force camparisons and selected stress results from the ASME evaluations. When possible, comparisons are made to PG&E results.
8.1 Frequency Comparisons Using the BNL developed finite element models the first fifteen natural frequencies for each piping problem were determined. These results are presented in the second column of Table 8.1 for problem 6-11 and in the second column of Table 8.2 for problem 4A-26. The first column of each table is
-132-
simply the mode number while the third column of each table presents the corresponding PG&E estimates for the natural frequencies.
As can be seen from Table 8.1, the natural frequencies predicted with the BNL model of problem 6-11 are all consistently higher than those predicted by the applicant. This was expected and is due to the applicants decision to model five bends, shown on the as-built drawings as SD bends, as long radius elbows. The substitution of a long radius elbow for a 50 bend has a softening effect on a systen. To verify this presumption a subsequent BNL evaluation of problem 6-11 natural frequencies using the applicants model produced frequency results which were nearly identical to those predicted by the applicant.
Referring to Table 8.2 no consistent trend is obvious when the natural frequencies predicted with the BNL model of problem 4A-26 are compared to the applicant results. For most modes the BNL estimates are low while for some they are high. These results were surprising as again a SD bend was modeled as a long radius elbow in the applicants model and higher BNL estimates for the frequencies were expected. A detailed review of the applicants model for this problem revealed that a minor geometry error existed in the model and systen dimensions were different. A further review of the as-built drawings for the pipe run indicated that sufficient ambiguity in dimensions existed to allow both models to be correct.
8.2 Results Based on Uniform Support Excitation Seismic evaluations of each problem were made using uniform, envelope spectrum, response spectrum methods. This mode of evaluation correspond to the method currently used by applicants to qualify their piping and is the method used by PG&E to qualify all the annulus region piping. The BNL studies included evaluations based on PG&E supplied horizontal spectra (see Fig. 8.13) and PG&E supplied and BNL developed vertical spectra for annulus structure models A and B. These were performed for X-Y and Y-Z inputs with no cluster-ing (matching PG&E procedures) and for simultaneous X-Y-Z inputs with cluster-
-133-
ing using a cluster factor of 0.1 (per Reg. Guide 1.92). Although initially only dynamic evaluations were made, the final calculations consisted of l complete class 2 evaluations.
I Figures 8.1 - 8.4 show the broadened vertical envelope response spectra for the two problems. Each figure shows both the BNL developed spectra and the corresponding PG8E spectra. Figures 8.1 and 8.2 show the BNL spectra developed using the model A configuration for the annulus steel while Figures 8.3 and 8.4 show the BNL spectra for the model B configuration. The PG&E supplied horizontal envelope response spectra were used as the excitation for the two horizontal directions for each problem. Referring to Figures 8.1 -
8.4 it can be seen that the BNL developed vertical response spectra for both problems and both models exceed PG&E supplied spectra. '
8.2.1 Force Comparison A reasonable caparison of the results from only the dynamic evaluations can be made by comparing the boundary element forces predicted with the BNL models to those predicted by the applicant. Since a large number of evalua-tions were made, only a selected few comparisons will be presented. These are shown in Tables 8.3 thru 8.8.
The results for each evaluation type are presented in tabular fonn in a consistent format. The first column identifies the piping model nodal point to which the support is attached and also the global direction in which it acts where H indicates a skewed horizontal support. The remaining columns present both BNL and PG&E predictions of the boundary element forces. Each of these columns has a heading and footnotes which define the mode'of the camputation. The variables in the mode of computation include; the type of calculation (unifann or independent support motion), the directions of the input excitation (simultanenus in X-Y, Y-Z or X-Y-Z), the summation rule used between directional components and the use or lack of clustering between modes. Lastly only the results for a representative number of supports are presented in each table.
-134-
At the beginning of the study the only spectra available were those provided by PG&E. Using these and the BNL models of the piping, unifonn envelope response spectrum analyses were made. The results of these analyses are shown in Table 8.3 for problem 6-11 and Table 8.4 for problem 4A-26.
Results are shown for X-Y excitation alone, Y-Z excitation alone (the modes of calculation used by the applicant), and for simultaneous X-Y-Z excitation with clustering (current industry practice).
Referring to Table 8.3 no consistent trend is apparent between the PG&E and BNL results for the same calculational procedure. The largest forces are predicted for Y-Z excitation with either the BNL or PG&E calculation providing the greater result. If these maxima are compared to the last column, again no consistent trend is evident. Clearly the PG&E results cannot be considered conservative but are of the same magnitude as would be predicted using current calculational procedures.
Referring to Table 8.4 the BNL results for Y-Z excitation for problem 4A-26 consistently exceed the corresponding PG8E estimates. Further if the BNL maxima are compared with those predicted using current methods, the last column, they are low for the most part. A judgement based only on these results would indicate a deficiency in the PG&E evaluations requiring further investigation.
The latest BNL estimates for the support forces based on uniform support excitation were developed using the annulu's structure model B spectra. Tables 8.5 and 8.6 summarize these results for the two problems. Again the corre-sponding PG&E results are presented. Referring to both tables it can be seen that the support force predictions based on the annulus structure model B spectra are all greater than the corresponding PG&E results. In addition if all the BNL results are compared, columns 2, 4 and 6, the predictions based on current practice are always the greatest for problem 6-11. Referring to the spectra drawings, Figures 8.3 and 8.4, the great disparity between the BNL
-135- ,
~
spectra and the PG8E spectra produce the differences noted here. Given the magnitude of the support force increases a reev'aluation of all support designs would seem warranted for these problems.
8.2.2 Stress Results Stress evaluation in accordance with ASME criteria for class 2 piping and ,
uni form support excitation were made for both problems. Only the envelope response spectra and seismic anchor movements for the model B annulus steel configuration were used for these evaluations. The BNL computer code PSAFE2 was empl oyed for this purpose. This code performs the stress evaluations in accordance with ASME-BPVC Section Ill, Subarticle NC-3600. .
The input for the ASME evaluation runs include the temperature, pressure and seismic anchor moiements as well as the excitation response spectra.
These input parameters were taken from the PG&E supplied data for both prob-lens with the exception of the vertical seismic anchor movement displacements which were determined from time history analyses using the model B annulus .
steel structure configuration. Regarding seismic anchor movements only those displacements greater than 1/16 in. were considered as any displacements ,
smaller are within the clearance gap incorporated at supports.
The pertinent input parameters for problen-6-11 were:
Tsupe rature = 350 F Press ure = 350 psi node X Y Z Sei smic 8 .1325
~
Anchor 37 .186 s Movements (in) 55 .154 59 1. 2 .45 .17 Service level D allowable stress, Sh = 15.3 x 2.4 = 36.72 Ksi
-136-
,> a
~
y% ;
r .
As indicated the only point experiencing borizontal displacements is point 59.
This point is affixed to the containment vessel which expands under the ef fects of containnent pressure. The vertical displacement of point 55 was taken negative as this yields the most unfavorable displacement pattern vi s-a-vis system stresses.
'The pertinent input parameters for problem 4A-26 were:
Temperature 125 F Pressure 165 psi node Y V Seismic 6 .0651 3 Anchor 16 .1216 Moyenents (in) 25 .1325 f' 48 .0892 55 .0892 Service level D allowable stress, Sg = 15.0 x 2.4 = 36 Ksi As can be seen there are no horizontal seismic anchor movement displacements for this problem. All such dispiacements were less than the 1/16 in clearance gap dimension. Again the displacement of one point (55) was taken negative to yield a most unfavorable combination. The sign for points 6,16 and 25 were not varied as time history results for the annulus structure showed these points to exhibit in phase motions.
The solution for problem 6-11 showed that the system satisfied the service level D allowables at all points except two. Overstressing occurred
-137-
at nodal points 58 and 59. These results are summarized in Table 8.9 along with the corresponding results from the independent support motion c omputation. In general the component of stress caused by the seismic dynamic loading daninated and produced the overstress conditions noted.
The solution for problem 4A-26 showed that the system satisfied the service level D stress allowable. Again the seismic dynamic stress components doninated but were not great enough to cause an overstress condition. On the basis of current design practice this system is acceptable.
8.3 Results Based on Independent Support Excitation Although the design of piping systems based on uniform support excitation represents current industry practice, there is a growing trend to adapt independent support excitation analysis methods for the evaluation of the seismic dynamic components of response. Analysis by these methods are thought to be more realistic and provide safe designs while reducing the level of i
conservatism inherent in them.
For the sake of completeness, solutions for both problems were developed using the independent support excitation methodology. It was anticipated that these solutions would show reductions in stress levels possibly eliminating i the overstress conditions in problem 6-11.
i The first phase of these analyses was to develop response spectra for each support point in each problem. These were prepared from the time history records developed for the model B annulus steel structure configuration.
Following a comparison of these spectra the supports for each problem were segregated into excitation groups such that all supports within a single group exhibited essentially equivalent excitation spectra. This accomplished, the 138-
'I
PSAFE2 computer code was used to perform the independent support excitation dynamic analyses and the subsequent class 2 evaluations.
For each problem it was found necessary to separate the supports into four excitation groups. Figures 8.5 - 8.8 show the broadened vertical re-sponse spectra for each group for problem 6-11. Figures 8.9 - 8.12 show the equivalent spectra for problem 4A-26. On each figure are also noted the sup-port node numbers of supports belonging to that group. As can be seen, the spectra for the first group in each case are essentially ground motion spectra while the spectra for the fourth group in each case exhibit the highest peaks and are associated with only one support. Each support group for each problem has associated with it horizontal excitation spectra identical to those provided by PG&E and used in the uni fonn support motion analyses. These PG&E provided spectra are depicted in Figure 8.13.
The independent support excitation evaluations were perfonned only with the spectra developed for the model B annulus steel structure configuration.
The evaluation included simultaneous X-Y-Z excitation with SRSS summation over directional components, clustering between modes with a cluster factor of 0.1 and absolute summation between excitation group contributions. The pertinent results are summarized in the following sections.
8.3.1 Force Comparison in the unifonn support excitation section comparisons were made to the PG&E results. In this section comparisons will be made only to the corre-sponding BNL unifonn support excitation solutions with simultaneous X-Y-Z inputs. Table 8.7 presents just this data for problem 6-11. Table 8.8 prewnts the conparisor. for problem 4A-26.
A review of both tables will indicate that the support forces predicted with independent support excitation are all typically greater than the forces predicted with unifonn support excitation. Only for a few nodes in problem 6-11 is the reverse true. This is contrary to expectation and seems to
-139-
contradict the basic reason for using independent support methods, a reduction in the level of conservatism.
Two reasons for these unanticipated results may be advanced. First, in order to be conservative and lacking a regulatory position, the group contri-butions have been summed using the absolute sum rule.
By adopting this procedure the assumption has been made that the group responses are completely out of phase. A less conservative procedure would have been to use the SRSS sum rule thereby assuming random phasing between groups. Second, independent support motions can excite dominant antisynetric modes in a system while uniform support motions will not. These antisymetric modes can contribute significantly to response.
A review was made of the mode shapes and participation factors associated with each problem. Indeed it was found that the fundamental modes for each problem exhibited strong antisymmetric characteristics. These are excited by the independent support inputs producing large participation factors and consequently large response for the individual group excitations. Secondly, each problem was rerun using the SRSS summing procedure between groups. This produced a marked reduction in overall response providing results which more closely corresponded to the uniform support motion results.
In any case the use of independent support motion analysis methods do not change the basic results. The BNL predicted support forces all exceed those predicted and used in design by PG&E.
8.3.2 Stress Results Stress evaluations in accordance with the ASME-BPVC Section III, Subarticle NC-3600, for class 2 piping and independent support excitation were made for both probl ems. The BNL computer code PSAFE2 was used for this purpose.
-140-
The inputs for the independent support motion stress evaluations were identical to those used for the uni form support motion evaluations with the l exception of the input spectra. That is the temperature, pressure, deadweight l and seismic anchor movements are all identical for the two analysis approaches and the reader is referred to the uniform support motion section for this data. The results are shown in Table 8.9.
A review of the stress data for problem 6-11 presented in Table 8.9 shows that the independent support motion evaluation resulted in a net reduction of stress for the high stressed points. The stress reductions at these points are on the order of 1/3. The overstress at point 58 is completely alleviated.
The stress at point 59, the right side anchor point, remains slightly above the allowable stress level.
The reduction of stress level at the peak stress points for problem 6-11 coincides with the reduction of support forces noted in the preceding section.
Although the majority of support forces showed an increase with independent support excitation some of those associated with point 59 showed a decrease.
Obviously those support reactions must contribute strongly to the stress.
For problem 4A-26 independent support excitation produced higher dynamic loads throughout the pipe run. This was shown by the marked increase in support forces presented in the last section and is evidenced in Table 8.9 by the increased stresses. These are points which satisfied the stress criterion when uniform support excitation was used (uniform stress results are also presented in the table). For this problem the use of the independent support motion analysis method penalizes the design, although the stresses remain below the service allowable, l
-141-
To summarize the overall stress results:
For problem 6-11 a possible stress condition exists at the right most anchor (point 59).
For problem 4A-26 all stresses satisfy the service level D allowable.
-142-
Table 8.1 PROBLEM 6-11 FREQUENCY COMPARISON Mode Frequency Frequency (Hz)
BNL PG&E 1 7.24 6.96 2 10.14 9.23 3 14.58 14.19-4 15.99 15.31 5 17.20 15.75 6 17.99 17.61 7 22.28 22.17 8 23.63 23.14-9 27.86 27.10 10 29.21 27.55 11 29.51 28.85 12 31.55 30.33 13 34.02 32.73 14 34.78 32.93 15 35.12 33.77
-143-
Table 8.2 PROBLEM 4A-26 FREQUENCY COMPARISON Mode Frequency Frequency (Hz)
BNL PG&E 1 5.05 5.46 2 14.63 14.49 3 15.67 16.54 l 4 18.29 18.43 5 20.37 19.39 6 22.60 21.66 7 23.64 24.50 8 28.67 31.35 9 32.20 34.23 10 34.83 37.87 11 37.96 38.06 12 43.55 38.68 13 46.78 40.95 14 47.42 42.17 15 47.73 43.87 l
l l
t
-144-l
Table 8.3 PROBLEM 6-11 SUPPORT FORCE COMPARIS0N PG8E SPECTRA X-Y Input Y-Z Input X-Y-Z Input Node BNL PG&E BNL PG&E BNL 1X 618 1240 1831 1751 2112 Y 77 81 89 98 96 Z 792 1084 1383 1326 1810 XX 7498 8315 9452 10760 9032 YY 54434 68809 34422 68809 70172 ZZ 6258 8697 9700 13616 8139 SY 364 425 382 425 424 6f 769 790 784 790 876 13H 3304 3313 3627 3113 4621 18H 240 303 497 214 306 29H 3071 3014 3783 3918 3393 32H 4984 4399 5142 4704 4691 55H 5060 4222 5895 6564 7018 59X 1691 1980 2330 2306 2483 Y 5797 5269 4885 5056 4571 Z 555 394 719 776 894
- XX 82437 84987 81644 100328 98951 YY 16917 11931 26100 32503 30068 ZZ 614000 596118 535615 601012 494778 X-Y and Y-Z, ABS between directions, CF = 0.0 X-Y-Z, SRSS between directions, CF = 0.1
-145-l
[
Table 8.4 PROBLEM 4A-26 SUPPORT FORCE COMPARIS0N PG&E SPECTRA X-Y Input Y-Z Input X-Y-Z Input Node BNL PG&E BNL PG8E BNL
! 1X 163 115 215 114 249 Y 35 31 31 30 38 f
Z 156 128 222 149 256 l l
l XX 2239 1896 2910 1854 3379 YY 2236 1678 1650 1424 1997 ZZ 1898 1528 2671 1500 3067 6Y 208 129 225 187 251 16H 237 188 510 416 555 Y 198 112 204 164 220 32H 102 60 193 122 214 29H 32 24 36 20 45 Y 30 19 41 16 45 43H 136 108 289 257 313 60X 26 16 34 24 37 Y 22 57 41 20 40 Z 37 54 70 56 69 XX 479 709 900 872 873 YY 1075 702 832 1310' 990 ZZ 465 236 655 577 731 X-Y and Y-Z, ABS between directions, CF = 0.0 X-Y-Z, SRSS between directions, CF = 0.1 l
-146-
Table 8.5 PROBLEM 6-11 SUPPORT FORCE COMPARIS0N MODEL 8 SPECTRA
~
X-Y Input Y-Z Input X-Y-Z Input Node BNL PG&E BNL PG8E BNL 1X 613 1240 1785 1751 2042 Y 92 81 -108 98 108 Z 823 1084 1362 1326 1760 XX 10231 8315 12538 10760 10928 YY 60162 68809 39494 68809 71723 ZZ 10631 8697 14184 13616 10943 SY 394 425 4 19 425 453 8Y 801 790 b31 790 905-13H 7230 3313 7234 3113 7429 18H 328 303 263 214 361-29H 8005 3014 8275 3918 7551 32H 13164 4399 12924 4704 12065 55H 8576 4222 9384 6564 9651 59X 3103 1980 3711' 2306 3485 Y. 17529 5269 16559 5056 15864 Z 933 394 1100 776 -1174 XX 187862 84987 183764 100328 186089 YY 33732 11931 42047 32503 41677 ZZ 1871788- 596118 1777017 601012 1698506 X-Y and Y-Z, ABS between directions, CF = 0.0 X-Y-Z, SRSS between directions, CF = 0.1
-147-
Table 8.6 PROBLEM 4A-26 SUPPORT FORCE COMPARISON MODEL B SPECTRA X-Y Input Y-Z Input X-Y-Z Input PG&E BNL PG&E BNL Node BNL 1X 203 115 247 114 294 Y 47 31 43 30 53 Z 180 128 241 149 276 XX 2719 1896 3263 1854 3850 YY 3086 1678 2490 1424 3280 ZZ 2271 1528 2983 1500 3422 6Y 278 129 289 187 334 16H 247 188 515 416 559 Y 254 112 268 164 296 32H 121 60 203 122 225 29H 36 24 40 20 48 Y 43 19 50 16 55 43H 144 108 294 257 316 60X 32 16 35 24 47 Y 78 57 95 70 106 2 63 54 84 56 89 XX 965 709 1188 872 1315 1255 702 1632 1310 1961 YY 368 236 671 577 737 ZZ X-Y and Y-Z, ABS between directions, CF = 0.0 X-Y-Z, SRSS between directions, CF = 0.1
-148-
l
- Table 8.7 PROBLEM 6-11 SUPPORT FORCE COMPARIS0N MODEL B SPECTRA Node 'Jnif. Sup. Mot. Ind. Sup. Mot.
X-Y-Z Inputs X-Y-Z Inputs IX 2042 5012 Y 108 211 Z 1760 4108 XX 10928 22131 YY 71723 152704 ZZ 10943 19715 SY 453 1030 8Y 905 2233 13H 7429 9076
, 18H 361 644 29H 7551 6994 32H 12065 9205 55H 9651 12844 59X 3485 4828 Y 15864 7366 Z 1174 1567 XX 186089 169037 YY 41677 51424 ZZ 1698506 826802
-149-
l Table 8.8 PROBLEM 4A-26 SUPPORT FORCE COMPARIS0N MODEL B SPECTRA Node Unif. Sup. Mot. Ind. Sup. Mot.
X-Y-Z Inputs X-Y-Z Inputs IX 294 439 Y 53 64 Z 276 439 XX 3850 5858 YY 3280 3695 ZZ 3422 5327 6Y 334 495 16H 559 1167 Y 296 449 32H 225 452 29H 48 59 Y 55 90 43H 316 667 60X 47 70 Y 106 183 Z 89 167 XX 1315 2275 YY 1961 3442 ZZ 737 1496
-150-
l Table 8.9 ASME CLASS 2 EQUATION 9 SATISFACTION Element Node Pressure De ad- SAM Earthquake Total Allowable
- # ksi weight- ksi ksi ksi ksi ksi 6-11 45 57 2.975 2.081 3.592 20.199(U)* 28.847 36.72 10.906(I)* 19.554 36.72 45 58 2.975 3.104 7.137 28.980(U) 42.197 36.72 14.726(I) 27.943 36.72 46 59 2.975 4.256 10.762 37.914(U) 55.907 36.72 18.754(I) 36.747 36.72 4A-26 21 11 .584 .343 15.059 8.056(U) 24.043 36.0 17.011(I) 32.998 36.0 21 12 .584 .339 8.726 9.322(U) 18.972 36.0 19.812(I) 29.462 36.0 22 16 .584 .934 3.158 12.763(U) 17.438 36.0 l 26.849(I) 31.525 36.0 29 16 .584 .934 .3.158 12.763(U) 17.438 36.0 l 26.849(I) 31.525 36.0
- Note: U = Uniform Support Excitation l I = Independent Support Excitation
-151-f I
2 b
6 6
j E
& 6 .
G s _
P
- 4 .
^ -
~
~
3 -
G N
\
\' 2 I
\
P i\
M [i i I\\
A ,i l \
D 1
[ 1 0
2 h F L b
O j, E -
rlI fb )2 P
O
,,8Ij!! if b (H 1 L
E V
I,Ih Y C 8 N d NE e r
E b U u g
Q i E F 1 R 1
- b ,il' F
6 - 2 M -
E L
B O .
R .
P .
0 0 0 0 O O' 0 O. 0 0 0 0 O 2
1 1
1 0 1
9 & 7- 6 5 4 3 2 1 8$hv
'Y
II.0 -
10.O-9.0-- PROBLEM 4A-26 ENVELOPEOF 2XDAMPlNG r
8.0-- ij, l \s.----
7.0--
\i G
v i
\,
2 60-9 ,
MODEL A I i b 50-
, s .
i G $
i' l \i Pc 4.0--
I
\
\
ha- ) i
/ \
7---- - s :
2.0-- I ,
,1
, i PG &E
- s. .
i,o_. '
/p- -
og)-' : : '::::' , . . . . . . . .
2 3 4 5 6789ICf i 5 d 55755b' 2 3 4 B 67Bsd FREQUENCY (HZ)
Figure 8.2 l
t
12&- !,- l I
I II.O- PROBLEM 6-11 ENVELOPE OF 2/DAMPlNG l [,
10.O- f I i i
9.0--
I 1
MLB go.. 1 I
m 87.O- I i
8- aO- I i 1
' j h / \'
b l t i
du 5.0-i
< /
I
,/ 'g 4.O- I
/
,. . J \
3.O-
,.____/
/ \'
t
, 't I G&E 2O. /j ('~,,
/,,/,
~
1.O-
,,/
0 . . . . . . .
4 6 6 j g4b2 I b 3 4 b b 78bb 2 34 $ 5 t bhIO 1 2 3 FREQUENCY (HZ)
Figure 8.3
11.0 -
10.0-- Ui.
PROBLEM 4A-26 ENVELOPE OF 2XDAMPING , 1 I \
9.0-- 1 i l
8.O-- I I
i
! i r $
7.O- I j n ,
o v
,I i e a
~
z o ,f W LB 5.0--
w i i
, J -
i
_ w l i G U l
)
Y' W4.0-- ,
/ j
, t
,. . / \
i 10 -
,-- a
/
s "G&E
/
/
2D-- l *k=g~ ~.'-
,/
/ ,'
l.0-- -
/
/
f 1 0 1 t I A B f A E F 0 t I I E A I I 8 I I I I I O g 8 4 66tegg 2 3 4 567890' 2 3 4 $ 6 7 89102 FREQUENCY (HZ)
Figure 8.4
00 25 A2 43, 1
5A 0 9 i 0 S 2, 1
E _
D9 GO2 .
NN i0 P 6 _
DO .
P P
e 40 _
X U 2S i 0 2
i 0 1
1
_ 1 P U
6 P
.)S O R
_ 1 i
(C G
. Y 5
/ '
4C B N L 8
E E e U D r Q O u E M g R i F F 2
I 1
1 6
M E
L
. B
' 1 O
R P
_ ' 6 4
2 0 6 0 4 8 2 6 4 3 3 2 1 l O
nT9 x m.& 5 bsay
.GT
ll l' 9
4, 1
4, B,
I M, I 0 3
1 0
1 S
E D
O 0 GN N
I 6
I P
M A T R 0 DO I
4 P
P
%U 2S g 0 2
g 0 1
2 i P U
. )S O 6 P R G
I (C 6
.Y B 8 kC
/ N l E
e r
/ E U
Q D
O M i u
g E F R
,F 1 1
6 M
E L
B O
R P
~
2 8 4 "
7 4 a O j
OioMDY z9 bu 6%
N 'T llll\il
72 -
f 2 X DAMPING SUPPORT NODES 8,25,37 O
'O M
L .,
y 48-z 9
5 x 24-
/
O ' ' ' ' ' ' ' '
1 .2 A .6 L 2. 4. 6. 10. 20. 40. 60. 100.
FREQUENCY (CPS)
PROBLEM 6-11 MODEL B GROUP 3 Figure 8.7
(
I4 -
i 12 -
l 2g DAMPlNG D ~
SUPPORT NODES 1,55 O
e Q
w8 -
- i m.
TA
?
g6 -
E s
w k4
/
2 -
0 ' ' ' ' ' ' i '
- 4. .2 A .6 L 2. 4. 6. 10. 20. 40. 60. 100.
FREQUENCY (CPS}
PROBLEM 6-il MODEL B GROUP 4 Figure 8.8
l'l 0
6, 3
4 2
3, 1
S E
D i 0 O 0 G
N N 1
PI T M R 0 A PO D
i 6
P
/ U 2 S '
0 4
' 0 2
1 1
01
[
1 P
' 6. P )S U O
(C R G 8 9
.Y e 4C r N B u g
E U L i
F Q E E D R O
.F M 2
6 2
A 4
' 1 M
E L
B O
R 6 P
' 4
' 2 6 0 4 8 2 6 0 3 3 2 1 l O'g d W z 9 h s "t >Pc
?
, lll
5 5
7 .
3 0
, '01 9
2 9
1 0
S 6
E D
O .
%U 2S
' 0 2
' 0 1
2
- 6. )S P P U
g (C O 01 R
.Y G 8 4C e N 8 ru E
U L g Q E i D F E
R O
.F M 2
6 2
A 4
' L M E
L B
O R
' 6 P 4
2
- - - _ ~ - - - - ,
4 8 2 4 8 2 6 -
5 4 4 2 1 1
O g o H0q, M z 9 b s $ M
- b,
P E ~
2/*. DAMPING SUPPORT NODES 6,16.25 O
e Q
M i m.
b 48 -
b 8
s '
J g /
o ,
<g ,
.. s O ' ' ' ' ' ' ' ' ' ' ' '
. .2 A .6 1. 2. 4. 6. 10. 20. 40. 60. 100L FREQUENCY (CPS)
PROBLEM 4A-26 MODEL B GROUP 3 Figure 8.11
'# .r
~
k
- 12 -
,=
/ % _
s.
~
_. z- 10 - '- , ,
,e 2% DAMPING
^
SUPPORT NODE 48 8 -
O i
Q w J e m bb i
6 -
g -
34 w
U
- I 2 -
O ' ' '
,. .2 A .6 L 2. 4. 6. 10. 20. 40. 60. 100.
FREQUENCY (CPS)
PROBLEM 4A-26 MODEL B GROUP 4 Figure 8.12
G
, PG &E SUPPLED HORIZONTAL SPECTRA.
- PROBLEM No 6-11
= PROBLEM No 4A-26 a-~5 A.0-- '
I i
o
/
s'
+50 -
/ ~~s' s
0 ,/
/ I b
//
Man-U /
/ \ _
s s f g
/ 'N
'ss a
h
- * * - a = I a a a _a_ a
$ 6 2 6 M1 a a I L 2" . 3. 4. 20 4t1 50 ~
ab. ' ' ' 10 0 FREQUENCY (HZ)
FIGURE 8.13
5 1 References
.I
([1] "Diablo Canyon Nuclear Plant Unit 1, Containment Structure, Dynamic
' Seismic .*.nalysis for 7.5M Hosgri Earthquake", URS/ John A. Blume &
I Associates, Engineers, May 1979.
- - [2] "A. Structural Analysis Program for Static and Dynamic Response of 1+
i Linear Systems", K. J. Bathe, E. L. Wilson, F. E. Peterson, University of Cali fornia, Berkeley, Cali fornia (USC 1976 version).
l
[3] " Soil-structure Interaction Methods - SIM Code", C. A. Miller, NUREG/CR-1717, Vol . II. December 1980.
4 I [4] "PSAFE2-Piping Analysis Program", M. Subudhi and P. Bezler (1981 version) to be published.
l l
r
-165-
U.S. NUCLEAR RGOULATORY COMhMOSION tr-7 n NUREG/CR-2834 BIBLIOGRAPHIC DATA SHEET BNI.-NUREG-51566
- 4. TITLE AND SueTITLE Mew vedume Na,if appralenew/
- 2. (Leare bleik/
Independent Seismic Evaluation of the Diablo Canyon Unit 1 Containment Annulus Structure and Selected 3 RECIPIENT'S ACCESS 10N NO.
Pininc Evstame 7 ^uTNoRai A. J. Philippacopoulos, M. Reich, P. Bezler, S. DATE REPORT COMPLETED C. Miller, Y. K. Wang, M. Subudhi, S. Shteyngart, P. Brown "4"y j I Y'^b2
- 9. PEEFORMING ORGANIZATION NAME AND MAILING ADDRESS (include Isa Codel DATE REPORT ISSUED Brookhaven National Laboratory "" ^"
Upton, New York 11973 Auoust l1 982
- 6. (Leave Nanal
- 16. ABSTR ACT 000 words or sessf An independent review and development of the vertical floor spectra for the Unit 1 containment annulus structure of the Diablo Canyon Power Plant was carried out using a detailed three-dimensional model . The developed floor spectra were then utilized for confimatory evaluations of two selected piping systems. The latter were evaluated by the envelope response spectrum method, and by the independent support motion response spectrum method. ASME Class 2 evaluations of the two systems were also performed.
Finally, a confimatory evaluation was carried out for the model utilized by URS/Blume for the development of the vertical floor response spectra.
- 17. KEY WORDS AND DOCUMENT ANALYS13 17a DESCRIPTORS 17b. IDENTIFIERS /OPEN-ENDED TERMS
- 18. AVAILABILITY STATEMENT 19. SE CURITY CLASS (Thns report) 21 B.O. OF PAGES Unciaccifind Unlimited dncTassON*d"*#'# s NRC FORM 335 (7 77)
UNITE] STATES sourf a ctasstatt Postaa a n es Pa,o NUCLEAR REGULATORY COMMISSION lis44C WASHINGTON. D C. 20565 nasu O C neua m. m OFFICIAL BUSINESS PENALTY FOR PRIV ATE USE. SX10 1
l 1209$5076b77 i Angg US teht ALM UIV 0F 1100 FUL10Y 4. PubL1LAllONS Mbl bd e*Dh NURLb Lury LA 212 N A $r61 h6 I Ot. UL 40$$$
i
{
l
__