ML20202D439
| ML20202D439 | |
| Person / Time | |
|---|---|
| Issue date: | 12/02/1997 |
| From: | Stolz J NRC (Affiliation Not Assigned) |
| To: | Smith N SEISMIC QUALIFICATION UTILITY GROUP |
| References | |
| NUDOCS 9712050024 | |
| Download: ML20202D439 (14) | |
Text
-.
- ' December 2, 1997 3
Mr. Neil P. Smith, Chairman Seismic Qualification Utility Group clo MPR Associates, Inc.
320 King Street -
Alexandria, VA 22314
SUBJECT:
EVALUATION OF SEISMIC QUALIFICATION UTILITY GROUP'S RESPONSES TO GENERIC ISSUES INCLUDED IN NRC'S REQUEST FOR ADDITIONAL INFORMATION By letter dated August 19,1996, Seismic Qualification Utility Group (SQUG) provided a response to six potential generic items included in NRC staff requests for additional information (RAI) to individual SQUG members implementing the Generic implementation Procedure, Revision 2 (GIP-2). A staff evaluation of SQUG's response to these six generic RAIitems was issued December 5,1996. SQUG pmvided additionalinformation in letters dated April 18, June 11, and June 30,1997. Additional discussion of the first generic item was the focus of a letter from SQUG to the NRC on November 11,1997, regarding the GIP provisions on use of Method A in Section 4.2.3 of GIP 2.
Enclosed is the staffs evaluation of SQUG's responses dated April 18, June 11, June 30, and November 11,1997, to the six generic RAI items related to the use of GIP-2. If you have any questions regarding this matter, please call Mr. Dan Dorman at (301) 415-1429.
Sincerely, Original Signed by:
Donald S. Brinkman/
for John F. Stolz John F. Stolz, Lead Project Director Project Directorate 1-2 Division of Reactor Projects - 1/11 Office of Nuclear Reactor Regulation
Enclosure:
As stated cc w/ encl: Mr. R. Kassawara, EPRI l
Program Manager j
3412 Hillview Avenue P.O. Box 10412
['
- a Palo Alto, CA 94304 =
0 DISTRIBUTION:
,\\
T) %
V, /
Central File
- Slittle RWessman PYChan A
PUBLIC JStolz DDorman GBagchl
\\., [7 d
PDI-2 R/F DOCUMENT NAME: G:\\DORMAN\\SQUGLTR 7
- r. '
To receive a copy of this document, indicate in the box C= Copy w/o attachment / enclosure E= Copy with
"\\
attachment / enclosure N = No copy ice PM:PDI 1 m E LA:PDl-10 D:PDI-2 u DDormaN JStolz)
A.VE Slit 6e r
DATE-1 % /97 34/e/97 44/ t/97 l'
OFFIC'AL RECORD COPY
,:j ]
a 7.;2 vf E M* >5~F R U P P M8
~
9712050024 971202 R
' L a J PDR REVQP EROSQUG s
pe neg$%
UNITED STATES
_y g
j NUCLEAR REGULATORY COMMISSION t
WASHINGTON, D.C. =a mg
\\,,,,,,*/
Decemoer 2,1907 Mr. Neil P. Smith, Chairman Seismic Qualification Utility Group clo MPR Associates, Inc.
320 King Street Alexandria, VA 22314
SUBJECT:
EVALUATION OF SEISMIC QUALIFICATION UTILITY GROUP'S RESPONSES TO GENERIC ISSUES INCLUDED IN NRC'S REQUEST FOR ADDITIONAL INFORMATION By letter dated August 19,1996, Seismic Qualification Utility Group (SQUG) provided a response to six potentiel generic items included in NRC staff requests for additional information (RAl) to individual SQUG members implementing the Generic implementation Procedure, Revision 2 (GIP-2). A staff evaluation of SQUG's response to these six generic RAI items was issued December 5,1996. SQUG provided additionalinformation in letters dated April 18, June 11, and June 30,1997. Aoditional discussion of the first generic item was the focus of a letter from SQUG to the NRC on November 11,1997, regarding the GlP provisions on use of Method A in Section 4.2.3 of GIP-2.
Enclosed is the staff's evaluation of SQUG's responses dated April 18, June 11, June 30, and November 11,1997, to the six generic RAIitems related to the use of GIP-2. If you have any questions regarding this matter, please call Mr. Dan Dorman at (301) 415-1429.
Sincerely,
'.)s -*-
n 4
John F. Stolz, Lead Project Directo, Project Directorate I-il Division of Reactor Frojects - t/11 Office of Nuclear Reactor Regulation
Enclosure:
As stated cc w/ enc!: Mr. R. Kassawara, EPRI Program Manager 3412 Hillview Avenue P.O. Box 10412 Palo Alto, CA 94304
EVALUATION OF SElSMIC QUALIFICATION UTILITY GROUP'S APRIL 18. JUNE 11. JUNE 20. AND NOVEMBER 11.1997. RESPONSES TO GENERIC ISSUES INCLUDED IN NRC REQUESTS FOR ADDITIONAL INFORMATION 1.
Use of Ground Response Spectra for Estimatino Seismic Demand NRC Evaluation in December 5.1996 Letter Seismic Qualification Utility Group's (SQUG's) response did not provide the requested information. During the August 28,1996, SQUGINRC meeting, the staff elaborated its concem and the primary focus of the request for additionalinformation (RAI) question.
As a result of considerable discussions on the subject, the staff agreed to clarify the question. The following revised RAI will be forwarded to affected unresolved safety issue (USI) A-46 licensees for their response:
Referrina to the in structure response spectra omvic in vour 120-dev-resoonse to the NRC's reauest in Sucolement No.1 to Generic LetteriGU 87-02, dated May 22.1992.
the followino information is reauested; a.
Identify structure (s) which have in-structure response spectra (5% critical damping) for elevations within 40-feet above the effective grade, which are higher in amplitude than 1.5 times the SQUG Bounding Spectrum, b.
With respect to the comparison of equipment seismic capacity and seismic demand, indicate which method in Table 4-1 of Generic implementation Program, Revision 2 (GIP-2) was used to evaluate the seismic adequacy for equipment installed on the corresponding floors in the structure (s) identified in item (a) above. If you have elected to use method A in Table 4-1 of the GIP 2, provide s technicaljustification for not using the in-structure response spectra provided in your 120-day-response. It appears that some A-46 licensees are making an incorrect comparison between their plant's safe shutdown earthquake (SSE) ground motion response spectrum and the SQUG Bounding Spectrum, The SSE ground motion response spectrum for most nuclear power plants is defined at the plant foundation level. The SQUG Bounding Spectrum is defined at the free fie!d ground surface. For plants located at deep soil or rock sites, there may not be a significant difference between the ground motion amplitudes at the foundation level and those at the ground surface.
However, for sites where a structure is founded on shallow soil, the amplification of the ground motion from the foundation level to the ground surface may be significant, For the structure (s) identified in item (a) above, provide the in-structure response c.
spectra designated according to the height above the effective grade. If the in-structure response spectra identified in the 120-cay-response to Supplement No.1 to GL 87-02 was not used, provide the response spectra that were actually used to verify the seismic adequacy of equipment within the structures identified in item (a) above. Also, provide a comparison of these spectra to 1.5 times the Bounding Spectrum.
Enclosure
- he staff does not consider a generic response to this question acceptable, and affected licensees should address this question on a plant-specific basis.
SQUG's Response in June 30.1997 Letter -
A comparison of the design ground response spectrum (GRS) to the SQUG bounding spectrum as a method for evaluating the seismic adequacy of equipment is included as Method A in Table 4-1 of GlP-2. The GIP allows this method to be used under the restriction that the equipment must be located at an elevation below about 40-feet above the effective grade of the building add the equipment fundamental frequency must be above about 8 Hz.
- One of the advantages of the GIP Method A is that the various effects associated with in-structure responses are inherently included in the method. The GIP approach differs from current seismic licensing criteria in that, in GlP Method A, the seismic capacity of equipment and the seismic demand vn this equipment are anchored to ground response spectra.- Further, the GIP method is not based on the performance of a single item of equipment subjseted to an artificial time history on a shake table. Instead the GlP method is based on the performance of numerous items of equipment subjected to several real earthquakes.
For these reasons, the GlP method. based on comparing ground response spectra at data base sites to safe-shutdcwn earthquake (SSE) ground motion response spectra at nuclear plants, was accepted by recognized independent experts,i.e., Senior Seismic Review and Advisory Panel (SSRAP report), including NRC staff, e.g., supplemental safety evaluation report (SSER) #2, as an acceptable method to verify the seismic adequacy of equipment installed in operating nuclear power plants, in addition to the above, SQUG also provided some recommended screening method to SQUG members to address the issue.
SQUG's Response in November 11.1997 Letter i
SQUG's position is that the Method A option continues to be available (provided special conditions are met, i.e.', the equipment is mounted below about 40 feet above the effective grade and its lowest natural frequency is above 8Hz) based on the clear rules in the GlP and it is applied property as clarified in SQUG's letter of June 30,1997, to the :
NRC on this subject. Any attempt to require use of gn!y the 120-day spectra would represent a change in the NRC's published position and would have to be justified under 10 CFR 50.109,
NRC Staff Evaluation
The SQUG-recommended three-part screening method discussed in the SQUG's June 30,1997, letter, does not address a major aspect of the staff's concem on this j
issue, in its discussion of GIP Method A contained in the June 30, and November 11, 1997, letters, SQUG failed to mention a limitation on the use of Method A contained in the GlP which is an essential consideration for the appropriateness of this method. On page 4-16 of GIP, Revision 2, in the Section entitled Advantaae and Limitations. the GIP states l
'The restrictions and limitations on use of the grggn( response spectrum for comparison
.=_
.- to the Bounding Spectrum and the GERS is based on the conditions that the amplification factor between the free-field response spectra and the in structure response spectra will not be more than about 1.5 and that the natural frequency of the equipment is not in the high energy ran9e..." Therefore, the staff position is that if the floor response spectrum in a structure is 1.5 or more times larger than the free-field response spectrum, the use of GIP Method A is not appropriate. The NRC staff will continue to ask each USl A-46 plant licensee, where appropriate, the question contained in the staff s December,
r 1996 letter.
Furthermore, with regard to the issue of using the in-structure response spectra provided by licensees in their 120-day submittals, SQUG asserts, in the November 11,1997, letter, that requiring licensees to only use those spectra would have to be justified under 10 CFR 50.109. The staff considers that these submittals constituted commitments from affected licensees regarding the quantification of the proposed seismic demand, to be used in verifying the A-46 equipment seismic adequacy at their plants. The staff expects these licenseer, to meet their commitments which the staff had accepted in its replies to the 120-day submittals. The staffs position is that licensees who deviated from their 120-day in-structure response spectra commitments, should have specifically informed the NRC of the changs so the staff would have an opportunity to determin9 the adequacy of the revised spectra. The final determination of adequacy of individud ! censee's responses will depend on the review outcome of each USI A-46 plant iicensee's submittal.
2.
Operator Actions in Difficult Environmental Conditions NRC Evaluation in December 5.1996 Letter The NRC staff recognized that...the only events which must be considered (when performing the USI A-46 program) are the SSE and loss of offsite power. However, each licensee should consider these factors on a case-by-case basis and determine what, if any, other complications they do need to address. During the August 28,1996, SQUG/NRC meeting, the staff provided SQUG with some additional clarification of the types of concems which should be considered including (1) the potential for diminished lighting due to loss of offsite power, (2) other barriers such as damaged equipment or structures which could inhibit operators cbility to access plant equipment, and (3) the potential for requiring operators to enter hazardous or unfamiliar areas to manually reset or realign equipment.
Therefore, the staff does not consider a generic response to this question acceptable, and, on a plant-specific basis, each affected licensee should address this question as requested in the original individual plant RAl.
SQUG Response in April 18.1997 Letter During application of the GIP criteria and guidelines, SQUG members were advised that diminished lighting due to loss of offsite power should be considered when planning how to safely shutdown the plant 'ollowing an earthquake. Some utilities have elected to add certain items of emergency lighting equipment to ^r safe shutdown equipment list.
Others have elected to rely upon emergency rest tools, such as flashlights and hand-held battery powered floodlights.
,~
37 i
i
_i
! L The potential for barriers such as damaged equipment or structures to inhibit an operator's ability to access plant equipment is not considered to be a significant har.ard.
Earthquake experience has shown that typical industrial grade equipment and structures are inherently rugged and not susceptible to damage which would result in the structures or equipment inhibiting operator access at A-46 plant SSE levels. Therefore, it is considered very unlikely that operators will be faced with hazardous or unfamiliar circumstances which are not covered by existing plant procedures and training.
it is for these reasons that the GIP in Section 3.2.7 of Part II, allows operator action to be used as a means of achieving and maintaining a safe shutdown condition provided procedures are available and the operators are trained in their use.
' Further, as stated in GIP Part il, Section 3.2.8, operators are expected to use normal operating and emergency procedures to shut down the plant following an SSE:
_The shutdown procedures which are associated with the use of the USl A-46 safe shutdown equipment should be procedures which are available to the operator as a result of his following approved normal and emargency operating procedures (EOPs).
The GIP goes on to say:
i l
It is not necessary to develop new procedures specifically for compliance with the USl A-46 program. Existing plant procedures can be used.
l 4
The GlP further discusses the use of normal and emergency operating procedures in L
Section 3.7, where the guidelines for performing the Operations Department review of the 1
l Safe Shutdown Equipment List (SSEL) against plant procedures and training are given.
1 Therefore, SQUG considers that the procedures in the GlP provide adequate guidance to the licensees for evaluating whether operators can safely shut down the plant following an earthquake, including unusual circumstancas beyond those associated with normal and emergency plant operating procedures. We will advise SQUG members to consider i
the NRC comments and SQUG's perspective on this issue.
l
NRC Staff Evaluation
The staff considers the SQUG Steering Group's response to this issue (i.e, advising the l'
SQUG members to consider the NRC comments and SQUG's perspective on this issue)_
to be acceptable. While the SQUG response indicates that such hazards are very unlikely due to the inherent ruggedness of plant equipment, each licensee should -
consider the likelihood of such hazards as part of their A-46 review. If such hazards do
\\
l 1
~ -
. - - = -
exist for a specific plant, the licensee shnuld discuss how that hazard has been taken into consideration.
3.
Opelator Trainina on SSEL NRC Evaluation of December 5.1996 Letter The NRC staff is in agreement with SQUG's position that existing normal and EOPs are expected to be sufficient to lead operators to use appropriate, operational equipment and systems following an SSE, and operators are expected to be trained in their use.
However, the staff expects the licensee to consider the need for (1) additional checklists or system alignment information to ensure there is adequate guidance for new or different operator actions, (2) verification that sufficient time and resources are available for planned operator actions, and (3) additional aids or guidance to manipulate plant equipment, such as relays, which are not routineh operated in such a manner.
The staff also acknowledged that plant operators should not base their actions on preconceived ideas of which systems are on the SSEL. Nevertheless, they suggested that during SSE training, the licensee should consider including a description of which systems have been selected for the SSEL and the reasons for such a selection.
Therefore, the staff does not consider a generic response to this question acceptable, and, on a plant specific basis, each affected licensee should address this question as requested in the original individual plant RAl.
SQUG Response in April 18.1997 Letter
- It is the intent of the GlP that the review by the Operations Department of the SSEL consider the need to change procedures and training to accomplish all required operator actions within the required times and available resources. Specifically, Section 3.7 of the GlP requires a documented review by the Operations Department to assure that the plant specific SSEL is a " legitimate safe shutdown path consistent with plant procedures and operator training." It further provides suggested methods to assure that operators can perform any actions expected of them in the A-46 program. We do not object to including background information on the USl A-46 program in operator training associated with response to an earthquake, however, we do not consider it necessary.
NRC Staff Evaluation
_- The staff considers the SQUG Steedng Group's response to this issue (i.e., The Operations Department review of the SSEL shall consider the need to change procedures and training to accomplish all required operator actions within the required times and available resourecs) to be acceptsble.
,n,,,.
v.
e_
..-,a.,
.m--n-..
w
--r
,- 4.
Listina and Justifyino insianificant Deviations from the GIP NRC Evaluation in December 5.1996 Letter The statement provided in the Summary Report is too simplistic and may be subject to misinterpretation. Since a precise description of what 'no significant deviation' really means is needed, the staff is, therefore, requesting each affected plant to itemize those GIP recommended evaluations / methodologies which the licensee did not follow or donated from. Affected licensees should discuss what the deviations are and why they are justified to enable the staffs assessment of such determinations. Also, a definition including the use of examples as to what is considered significant should be provided.
Therefore, ti,e staff does not consider a generic response to this question acceptable, and on a plant specific basis, each affected licensee should address this question as applicable.
SQUG Response in April 18.1997 Letter As discussed during the meeting on August 28,1996, with the NRC staff, we consider it sufficient for SQUG members to provide examples of typical
- insignificant" deviations to the GIP guidance in their plant specific response as an adequate response to this generic issue if it is raised in an RAI.
NRC Staff Evaluation
it is important that licensees provide examples of " insignificant" deviations to the GIP guidance in their plant-specific response. However, the staff is requesting that the licensees provide the worst case items which deviate from the GIP guideline that were categorized as not being significant. In addition, the staff is asking licensees to provide the definition of " safety significant" that the walkdown crew used and provide a justification of why the definition is appropriate. The staff needs this information to make a finding at to the adequacy of the information provided in response to GL 87-02.
5 Lateral Load Ductility Evaluation of Cable Trays NRC Evaluation in December 5.1996 Letter in response to the staff RAI for the Point Beach L'SI A-46 report, GQUG provided a generic answer (Letter from Neil P. Smith of SQUG to Daniel H. Dorman of the NRC, August 19,1996). The response stated that the basis foi the ductdity eva!uation procedure is in EPRI NP 7151 (" Cable Tray and Conduit System Seismic Evaluation Guidelines," March 1991). Our review found that the report did not contain the basis we are looking for, namely, a quantitative evaluation that demonstrates the ductile behavior of cable tray systems and the exclusion of lateralload evaluation specified in GlP-2 for ductile cable tray systems vs. those constructed #,th less ductility.
I n.-
I l-,
The staff reviewed additional reports related to this issue. These reports were suggested
?
by Mr. Stark of MPR Associates who authored the SQUG response to the NRC Review summaries of the three reports are as follows:
EPRI NP-7150 ("The Performance of Raceway Systems in Strong Motion Earthquakes,"
EPRI Palo Alto, CA, prepared by EQE Engineering Consultants, March 1991), contains a summary of test data (Blume, Bechtel and ANCO) and site experience data. However, there is no specific data relating to ductility except a general statement that ductile cable trays are better than non-ductile ones. Experience data does not provide a quantitative demonstration of the validity of the GIP'S, guideline regarding ductile cable trays.
The March 1,19gi, report of the Senior Seismic Review and Advisory Panel (SSRAP)
(Review Procedure to Assess Seismic Ruggedness of Cantilever Bracket Cable Tray Supports, Senior Seismic Review and Advisory Panel, Rev. 3.0, March 1,19g1), contains only an outline procedure for the cable tray evaluation, but the basis for the ductility evaluation is not provided except to refer to the test and experience data mentioned above.
The last report referenced by MPR is EPRI NP-7152 (' Seismic Evaluation of Rod Hanger Supports for Electrical Raceway Systems,' EQE Engineering, March 1991), on rod hanger supported cable trays, it discusses fatigue data for rod hangers, but there was no mention of ductility in the report.
The staff finc's that the documents referenced by SQUG and its consultants do not provide an adeouste basis for the ductility assumption and specified evaluafon approach for these ccble tray systems.
SQUG Response in Aoni 18.1997 Letter A lateral load evaluation is not included in the GlP for cable tray support systems classified as ductile, since earthquake experience data has shown that typicallateralload evaluation methods do not provide meaningful conclusions regarding the seismic adequacy of such systems. SQUG will provide, in a separate letter, an explanation to support the procedure in the GlP that the limited analytical review of supports for ductile raceway systems covers the appropriate parameters for demonstrating whether cable and conduit raceway systems in nuclear power plants are similar to those which performed successfully during earthquakes.
SQUG Response in June. 11.1997 Letter The subject RAI requested the technical Justification for the lateral load evaluation methodology provided in the GlP for cable tray supports. As you are aware from your previous review of th!s methodology, the cable tray evaluation approach is not based on-traditional first-principle analyses, instead, it is based on an extensive database of successful cable tray performance in strong motion earthquakes and shake table tests.
.,n
,v,-
,..,w r
m., -. - -
w N~
-~=-c-r~
--~"-
i
... )
The analysis checks described in the GIP process are intended to confirm that nuclear-plant cable tray systems are bounded by the earthquake and test experience. They are not intended to satisf;> typical structural analysis criteria.
NRC Staff Evaluation
i The staff's concem is with the concept that the cable tray and conduit supports in the experience data base performed well because they were ductile and the fact that without any quantMative analysis of the ductility of the supports, the criteria is extended to A-46 plants. On several occasions SQUG has indicated that it would provide information on the basis for the criteria but, to date, it has not done so. In view of this, the staff will ask licensees of A-46 plants to provide the following plant-specific information about the evaluation of cable tray and raceway supports using the ductility concept.
Seismic Adequacy of Cable and Conduit Raceways The NRC staff has concems about the way the A-46 cable trays and conduit raceways issue was being dispositioned by licensees. We issued RAI to several licensees on this issue. SQUG responded instead of the licensee because SQUG considered the RAI to be generic in nature. The staff issued a subsequent RAI to SQUG as a follow up to their response. However, the staff found that the correspondence with SQUG did not achieve the intended results in that they did not address the technical concems of the staff.
Therefore, we are restating our concems in the following discussion.
. The GlP procedere recommended performing what is called a limited analytic evaluation for selected raceways and cable trays. The procedure further recommended that when a certain cable tray system can be judged to be ductile and if the vertical load capacity of l
the anchorage can be established by a load check using three times the dead weight, no further evaluation is needed to demonstrate lateral resistance to vibration from '
. earthquakes. The staff has concems with the manner in which these simplified GIP criteria were implemented at your plant.
The GIP procedure eliminates horizontal force evaluations by invoking ductility. However, all the so called non-ductile cable trays would eventually become ductile by inelastic -
deformation, buckling or failure of the non-ductile cable tray supports and members. If this procedure was followed for eliminating cable trays for further assessment at your plant, then all the cable trays could conceivably be screened out from A-46 evaluation.
i We are requesting your response on the following items to elicit information that would support our safety evaluation of cable trays at your plant.
1, Provide the total number of raceways that were classified as ductile in your A-46 svaluation and for which you did not perform a horizontalload evaluaSon.
Indicate the approximate percentage of such raceways as compared with the entire population of raceways. Discuss how the ductility concept is used in your walkdown procedures.
2.
Provide descriptions of all the typical configurations of your ductile rac3 ways (dimension, member size, supports, etc.)
3.
Provide justification for stating that ductil6 raceways need not be evaluated for horizontalload. When a reference is provided, state the page number and paragraphs,' The reference should be self contained, and not refer to another reference. You should also provide a technical discussion why the reference you i
4 g_
reference. You should also provide e technical discussion why the reference you provided consittutes bases for the oa.,lity methodology, 4.
In the evaluation of the cable trays, and raceways, if the ductility of the attachments is assumed in one horizontal direction, does it necessarily follow that the same system is ductile in the perpendicular direction?
5.
Provide a definition of ductility in engineering terms. Provide an assessment of the maximum ductility utilized for the weakest cable tray support.
6.
Discuss raceways and cable trays that are outside of the experience data by explaining criteria used for making your safety determination, the configurations of such raceways and the number of the raceways and percentage with respect to the whole population of raceways. How are they going to be evaluated and disposed?
The staff needs this information to make its finding as to the adequacy of the information provided in response to GL 87-02, 6.
Seismic Adecuacy of Relays Mounted on Diegel Generators and Air Comoressors NRC Evaluation in December 5.1996 Letter The SQUG's generic response to the NRC question is net acceptable. First, the specific issue in question is not regarding relays mounted on diesel generators and air cornpressors, The issue is the inappropriateness of using the " rule-of-the-box" concept and the judgment based on the normal operation of the diesel generators or air compressors to justify the seismic sdequacy of devices, such as relays, mounted in the instmmentation and control cabinets anchored on the common skid of the diesel generator or air compressor.
The concept of the " rule-of the-box" applies to components in a system that has already been successfully subjected to a vibratory environment comparable with or greater than the required motion (e.g., SSE). Therefore, the " rule-of-the-box" concept can also be applicable for acceptance of the relays mounted on vibratory equipment (or in a cabinet supported on the common skid) provided it is demonstrated that the vibratory motion of the equipment (or the skid) is at least equal to the required seismic motion at that locatior., and that the relays performed all their intended functions during the periods of vibration.
With regard to relays mounted on diesel generators and air compressors the following specific questions should be addressed:
Does the mechanical vibration envelop the require d input motion (e.g., SSE) from a.
all aspects (e.g., amplitude, frequency, direction, otc.)?
b.
Do these vibratory equipment mounted relays perform all their operational safety functions (e.g., change of state) during the mechanical vibrations (i.e., during startup and normal operation) so that the relays can be considered qualified to that level?
, c.
For any of these vibratory equipment items, could an SSE occur when the equipment is vibrating? If so then the earthquake load will be an incr3 ment to the normal operational vibratory load, and the relay may need to be reviewed for the increased motion.
The vibratory motion of the skid is expected to be less severe than the vibratory equipment item itself provided the skid is rigidly mounted on a heavy foundation.
Therefore, all of the above questions will also apply for relays contained in skid-mounted cabinets with a particular emphasis on the vibration level verification, with the understanding that it is the vibration of the skid and ol of the vibrating equipment, that D
will be compared with the required input motion (i.e., item a. above). In addition, the cabinet that houses the relays may also change the vibration level and characteristics at the relay locations.
Regarding ' bad actor relays, they are so described mainly because of their low seismic capacities, or inexplicable performance characteristics. The
- ,re,' bad actor relays mounted on vibratory equipment may be expected to have -
onstrated their performance under certain conditions. But, in general, the s
. lays also should be verified following the approach for other vibratory equipment n ounted relays discussed above.
In conclusion, the staff's original RAI has been clarified and divided into three questions as described above. Any USl A-46 plant licensees who have inappropriately used the
" rule-of box" concept or exercised the judgment, based on normal operation of i vibratory equipment, to justify the seismic adequacy of component / device mounted on a vibratory equipment or mounted in a cabinet which is anchored to the common skid of the vibratory equipment, should demonstrate the seismic adequacy of the component / device by calculation and/or test data.
SQUG Response in April 18.1997 Letter SQUG considers the NRC's request for expanded testing and analysis of contact devices and relays on EDGs and air compressors as impractical, unnecessary, and beyond the scope of the USl A-46 program as agreed to by the staff when they accepted GIP-2 for N
resolution of USl A-46.
SQUG considers the existing guidelines and criteria contained in the GIP and its supporting references to be adequate for resolution of USl A-46. Earthquake experience l
data and experience with these machines in commercial transportation applications, such as locomotives and ships where significant extemalloads are applied simultaneously with normal operating loads, has not shown there to be a problem with respect to contact chatter. The design of the protective and control devices on typicalindustrial diesel generators and air compressors is very similar to those used in nuclear applications.
Based on these considerations, and the fact that EDGs and compressors routinely experience high vibration during startup and operation, it was judged that it is not i
k
4 2
... - - _ =
am.
m_
11-necessary to separately evaluate the seismic adequacy of relays mounted on EDGs and reciprocating air compressors, or in cabinets mounted on their skids, provided these devices routinely see high vibration during operation.
NRC Staff Evaluation
SQUG did not address the staffs concems as delineated above in NRC's Evaluation in December 5.1996 Letter. The staffs position on this issue remains unchanged.
Therefore, the staff considers SQUG's generic response to this question unacceptable.
The staff identified this issue during its review of Point Beach's submittal on USl A-46.
The licensee for Point Beach agreed with the staffs concem and stated that the vibratory excitation caused by the operation of the vibrating equipment might be outside the frequency range of concem for the seismic event. Therefore, the licensee intends to reevaluate the seismic adequacy of those contact devices (rely circuit breakers, contractors, mechanically actuated switches and manually actuated Switches) that were originally considered to be seismically adequate based on being mounisd on vibrating equipment. The licensee intends to provide an update to the Relay Evaluation Report as its evaluation is completed. The staff considers the licensee's approach to resolve this issue acceptable.
However, to avoid potential future misinterpretation of the GIP's provision on this item, the staff requests SQUG to convey the staffs position to other member utilities of SQUG.
i
.. REFERENCES 1.
Letter from Neil P. Smith of SQUG to Daniel H. Dorman of NRC, "SQUG Response to.
Certain Generic issues included in RAl's Sent to SQUG Member Utilities implementing USI A-46," August 19,1996.
2.
EPRI Report NP-7151 " Cable Tray and Conduit System Seismic Evaluation Guidelines,"
March 1991.
3.
EPRI Report NP 7150 "The Performance of Raceway Systems in Strong Motion Earthquakes," EPRI, Palo Alto, CA, prepared by EQE Engineering Consultants, March 1991.
4.
SSRAP Report "Deview Procedure to Assess Seismic Ruggedness of Cantilever Bracket Cable Tray Supports," Senior Seismic Review and Advisory Panel, Rev. 3.0, March 1,1991.
5.
EPRI Report NP-7152 " Seismic Evaluation of Rod Hanger Supports for Electric Raceway Systems," EQE Engineering, March 1991.
6.
" Generic issues included in NRC's Request for Additional Information (RAl)," dated April 18, 1997.
7.
"SQUG Response to NRC RAI on Lateral Load Ductility Evaluation of Cable Tray Supports,"
dated June 11,1997.
8.
" Generic issued inciuded in NRC's Requests for Additional Information on Use of GIP Method A," dated June 30,1997, 9.
- Seismic Qualification Utility Group (SQUG) Clarification of GIP Provisions on Use of Method A," dated November 11,1997.
-. -