ML20210B256
| ML20210B256 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 07/19/1983 |
| From: | Noonan V Office of Nuclear Reactor Regulation |
| To: | Knight J Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML16340C148 | List:
|
| References | |
| FOIA-86-151 NUDOCS 8609180014 | |
| Download: ML20210B256 (8) | |
Text
.
p? %q%,
4 4
UNITED STATES NUCLEAR REGULATORY COMMISSION n
1
(
I WASHINGTON, D. C. 20s55
%,...../
JUL19iw3 MEMORANDUM FOR: James P. Knight, Assistant Director Components and Structures Engineering Division of Engineering FROM:
Vincent Noonan, Chief Equipment Qualification Branch Division of Engineering
SUBJECT:
DIABLO CANYON DESIGN VERIFICATION PROGRAM SUPPLEMENT TO SAFETY EVALUATION REPORT Plant Name:
Diablo Canyon Unit 1 Docket No.:
50-275 Licensing Stage: OL Responsible Branch and Project Manager:
LB-3, Hans Schierling Review Status:
Continuing The enclosed supplement to Safety Evaluation Report (SER) was prepared by DE:C&SE, Equipment Qualification Branch; and this input covers' the staff evaluation of the Diablo Canyon Unit 1 design verification program for the following subjects:
(1) Shake table testing (2) Seismic Qualification Interface -- Westinghouse Main Control Board hfl1 Ah Vincent Noonan, Chief l
Equipment Qualification Branch Division of Engineering
Enclosure:
As stated I
cc:
V. Noonan M. Subudhi, BNL G. Knighton A. Lee G. Bagchi
Contact:
G. Bagchi Ext. 28251 A. Lee, Ext. 28280 8609180014 860708 PDR FOIA HOLMES86-151 PDR
Supplement to Safety Evaluation Report Design Verificat. ion Program Diablo Canyon Unit 1 Equipment Qualification Branch.
4.5 Other Seismic Design Verification Topics 4.5.2 Shake Table Testing 4.5.2.1 Introduction PG&E employed testing for certain Class IE electrical equipment and instrumen-tation subject to this design verification.
For Hosgri qualification, the cri-teria used are in conformance with IEEE 344-1975, Recommended Practice for J
Seismic Qualification of Class IE Equipment for Nuclear Power Generating Sta-tions and Regulatory Guide 1.100, Seismic Qualification of Electrical Equipment for Nuclear Power Plants, August 1977.
To obtain a representative sample for design verification, RLCA reviewed the list of. Class IE electrical equipment and instrumentation qualified by shake table testing.
Seven groups of items, tested at Wyle Labs, were chosen as the I
sample according to equivalent physical locations to facilitate definition of the required test.
RLCA reviewed the initial sample in two segments. The first and major segment was the verification of Myle grouping and seismic inputs for Class IE electrical equipment, which were reported in ITR-4.
The second segment was the design verification of shake table test mountings used in testing Class IE electrical equipment. This verification was reported in ITR-44. Detailed staff review of ITR-4 and ITR-44 are also contained in Appendix 4.5A and Appendix 4.58 respectively.
4.5.2.2 Verification of Initial Sample: Grouping and Seismic Inputs The RLCA review of the Wyle grouping and testing sequence is summarized as follows:
07/20/83 2
DIABLO CANYON SSER 1
+-..e.~w.
y
,.,__.w.m
%-,.,m w-
,,,9,,,,,,,,,y-..,---,,-_-.,.5
,.,,.,,,,-.,.,._*,,e
.--.g.-w w
e (1) Reviewed the test procedure Wyle used to test each of the seven groups of j
Class IE electrical equip ~ ment and instrumentation.
(2) Verified the location of the electrical equipment and instrumentation included in the seven groups.
(3) Developed " worst case" response spectra for each group. These spectra provided the highest seismic accelerations associated with the location of the group.
(4) Made two response spectra comparisons. The RLCA worst case response spectra (worst case spectra) was compared to the W91e test response spectra (test spectra). The Wyle target response spectra (target spectra) was compared to the test spectra.
The RLCA worst case spectra were generated for two time histories (Blume and Newmark) according to building, floor location, elevation, type, and damping.
The types of spectra were both vertical and horizontal. The horizontal spectra consisted of effects of East-West translation, North-South translation, East-West torsion, and North-South torsion. The RLCA worst case horizontal response spectra were developed by adding torsional. effects to the transla-tional spectra.
The test response spectra must envelop the required response spectra by at least 10%.
Both the RLCA wor t case spectra and Wyle target specta were developed to represent the required response spectra.
Four EDI reports, 2005, 1007, 1013, and 1049 were issued by RLCA as a result of design verification.
E01's 1005, 1007 and 1049 were later closed and EDI 1013 was resolved as a Class B Error. The EDI 1013 was subsequently downgraded and the equipment was demonstrated to be qualified. The resolution of these EDI's were found to be acceptable by the staff.
The staff also concurred RLCA's recommendation in that the correctness of target response spectra specified for all items shake table tested by PG&E s
07/20/83 3
DIABLO CANYON SSER 1
,cw~-
-a--n-, - -
--.~--.~,---,-.--.---.e----.
ew-,.
~ ~ - -
--m-
... ~-
r-
l and seismic service-related contractors be sub"jected to additional verification.
RLCA recommended the following four specific actions.
(1) Confirm field locations of all equipment (2) Select the applicable Hosgri response spectra (3) Develop the worst case responst spectra (4) Compare the worst case response spectra to the target response spectra specified in the testing procedures.
4.5.2.3 Verification of Initial Sample of Shake Table Test Mountings
~
For shake table testing, the test mountings were intended to simulate the in-service condition.
For testing convenience, some equipment was mounted to the shake table through an interposing fixture which was intended to simulate the dynamic and structural characteristics of the in-service mounting. Test procedures and test reports were examined to determine the mounting configura-tions and fixtures (if any) used for the test of each' item of equipment. Where the test mountings were identical to the inservice mountings, they were judged to meet criteria.
For equipment with test mountings not identical to in-service configurations, each test mounting was evaluated to determine if it adequately represented the dynamic and structural behavior of the in-service configuration.
Of the 31 electrical equipment items tested, 25 were found to meet criteria.
One was classified as an error, and one was excluded from review as being non-Class IE. The remaining four items (mountings of the main annunciator typewriter, battery charger cabinet, and the snap-lock limit switch) were not reviewed because they have been retested or replaced by the DCP and should therefore be reviewed as part of DCP activities.
07/20/83 4
DIABLO CANYON SSER 1
-.-.m 2
._.--#,-,_,-e-,
e w.
g
w.----.-.--_
-y.,
.9
-.--,--s,--99 3,,
,y
l Two E01 reports were issued by RLCA:
E0I 1119 was classified as a Class C Error because the documentation of the test mounting configuration was not sufficient to allow an evaluation of the structural adequacy of the in-service mountings. This' mounting was then qualified by DCP analysis, which was verified by the IDVP.
EDI 1118 was classified as a Deviation. The staff concurs with the above IDVP verification.
4.5.2.4 Verification of DCP Activities The IDVP verification of DCP work on shake table testing is documented in ITRs-8 and -35 and in response to IDVP concerns developed during verification for the initial sample. The results of the verification will be reported in ITR-67. The staff evaluation will be made when ITR-67 is issued.
-The DCP Internal Technical Program (ITP) of seismic qualification is conducted by checking the latest response spectra.for the DE, DDE, and Hosgri event against data used for equipment qualification. Whenever changes to the response spectra required requalification of the equipment, the equipment was requalified by analysis or testing.
Equipment identified for review comprised that asso-ciated with the engineering safety systems designed by PG&E (reference) PG&E Phase I Final Report, Section 2.3).
?
DCP reviewed the validity of the previous seismic qualifications of equipment against current spectra.
If the qualifying test response spectra did not completely envelop the current required response spectra, an attempt was made to qualify the equipment by analysis.
Otherwise, equipment modifications would be performed and the equipment would be retested.
The sample selected by the IDVP for verification of the DCP's ITP for shake table tested equipment consists of the portable fire pump and radiation monitor RE-14A, both are Design Class I equipment.
The portable fire pump represents the only shake table tested mechanical equip-ment.
Radiation monitor RE-14A, on the other hand, represents one of approxi-mately 27 categories of tested equipment within the electrical equipment and 07/20/83 5
DIABLO CANYON SSER 1
instrumentation scope.
For both sample items, the IDVP performed design reviews and test reviews of the qualification documentation.
No E01s have beeri issued to date 'for this review area.
The verification program intendert to De conducted by the IDVP is not yet complete. The staff agrees with the IDVP's conclusion that, based upon the efforts performed to June 25, 1983, the following aspects of the DCP work are acceptable and satisfy the licensing criteria.
(1) Applicable criteria have been identified and applied for shake table testing.
(2) Functional capability requirements have been specified and met.
(3) Mounting of the test specimens were either representative of the installed condition or were adequately evaluated.
l 4.5.2.5. Conclusion i
Based on.the staff review of the design verification performed to date by PG&E and IDVP, it.is concluded that for the sample of equipment items and their mountings selected, the seismic qualification using shake table testing is ade-quately performed and therefore acceptable. The staff agrees with the IDVP findings and concurs in the recomunendations as stated in Section 4.5.2.2 for additional verification'for correctness of target response spectra specified for all items shake table tested by PG&E and seismic service-related contractors.
The final staff conclusion on the overall adequacy of the shake table testing j
will be made when the IDVP verification of DCP work in this area is completed.
4.5.3 Seismic Qualification Interface - Westinghouse Main Control Board (MCB) i 4.5.3.1 Introduction l
The Diablo Canyon Unit 1 MCB was procured for PG&E by Westinghouse (W) in accordance with an equipment specification issued in 1971. W specification 07/20/83 6
DIABLO CANYON SSER 1 y--=w.-
m,
,.7--.-
-..,,,.-,.--,,-,--------,-_._.--,...---.p-
.,_-.--.,-c.
required that the MCB should ae qualified to Ig horizontal and 0.5g vertical with stresses within allowable limits, and for 2g horizontal and lg vertical stresses should not exceed the yield point. MCB was supplied by Reliance, and j
Reliance used a private consultant to qualify the MCB by analysis. This is the original analysis which predicted the lowest natural frequency of the MCB to be above 70 Hz based on a simple analytical model used.
Stresses were analyzed and shown to be well within allowable range. The axial load in one of the bracing members slightly exceeded the allowable load, and the report j
recommended addition of two bracing members to each end frame. The next phase of reevaluation was caused by the need to evaluate the adequacy of the MCB to the Hosgri earthquake in 1977. The revised seismic input at the base of the MCB produced 1.55g horizontal and 0.81g vertical; this is the original Hosgri input. These values are lower than the original qualification level of 2g horizontal and Ig vertical; thus the MCB was acceptable under 1977 Hosgri evaluation.
The final phase of the MCB qualification developed when the independent design verification program generated new floor response spectra for the auxiliary building. These new floor response spectra, referred to as current Hosgri l
spectra, indicate higher values of zero period acceleration (ZPA) for the ver-tical direction 1.45g as opposed to the qualification level ZPA of 1.0g.
It was during this final evaluation pr'5 cess that the MCB was modeled using field measurements and results of in-situ tests.
In-situ tests pointed out the existence of natural frequencies between 15 to 28 Hz, much below 71 Hz - the value of the lowest natural frequency calculated in the original qualification report.
Because of the severity of the current Hosgri spectra at the base of the MCB in the 15 to 33 Hz range, y has chosen to retest selected devices that are attached to the MCB. y has also proposed modification of the MCB.
4.5.3.2 Evaluation The original analysis modeled the MCB as a uniform cantilever beam restrained at the base by the floor.
It appears that the MCB behaves as a horizontal beam supported by rigid cantilever frames; and this type of behavior yielded the lower natural frequencies. The modification proposed is to add plates and channels.on top of the MCB in order to strengthen the beam property of the MCB 07/20/83 7
DIABLO CANYON SSER 1 y---
w-w-
.-+-.-g
-n-,-
s i.--.-,.-e.
---.___,,*-m
--w v
,..e=.---s--me m
y--
yielding higher natural frequencies. W was requested to ensure that buckling loads in bracing members of the modified MCB are below critical loads with adequate margins.
The floor spectra enveloping the DDE and current Hosgri spectra were used to generate device location spectra by transient analysis. The devices were tested on a shake table at the maximum expected level.
Some modifications such as strengthening of device mounts and restraints are expected. The seismic qualification of the auto / manual station was recently completed by shake table testing which the NRC staff witnessed; all other device testing is complete. Other non-Class IE devices are being analyzed for structural integrity. Also, modifications to the mounting of some non-class 1E devices are anticipated.
4.5.3.3 Conclusion Based on the review of the characterization of the seismic input to the MCB and the input to the devices located inside the MCB, review of the detailed model of.the MCB and correlation of the model properties with results of in-situ dynamic testing, attention paid to supports and restraints to both Class 1E and non-Class 1E devices within the MCB, the staff concludes that Westinghouse has performed a thorough investigation of the qualification program for the MCB, and that with the completion of all proposed modifica-tions, the MCB should perform satisfactorily.
Staff acceptance of the MCB is contingent upon written confirmation of completion of all modifications to the MCB including the devices with the complete qualification documentation being
(
available at a central location for staff audit.
i l
l l
i I
07/20/83 8
DIABLO CANYON SSER 1
...m....
~....mn.
..~,. m
......s..
-- -. - - _ - - _ __ _ _ _ ~ _ - - - -,
..__.-_-,.--~....m--...,... - _ - - -, _,-
-,-m
.,.,,...,my--
_m,-,,--
-