ML19305A278
| ML19305A278 | |
| Person / Time | |
|---|---|
| Issue date: | 01/25/1979 |
| From: | Kennedy R, Neumark N Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-1599, NUDOCS 7903120483 | |
| Download: ML19305A278 (46) | |
Text
v.
\\?
) }, t,5
}
b
=..
,r!
DATE ISSUED:
1/25/79 by b MINUTES OF THE ACRS SUBCOMMITTEE ON FLUID DYNAMICS
,1 (j CJ SAN FRANCISCO, CALIFORNIA NOVEMBER 28-30, 1978 M /d 77 The Fluid Dynamic Subcommittee of the ACRS held a meeting on November 28-30, 1978 at the Barrett Motor Hotel, San Francisco, California. The main purpose of the meeting was to discuss the status of the NRC Review of the Mark II containment system and the present status of the use of SRSS methods of load combination for LOCA and SSE events.
Notice of this meeting was published in the Federal Register on November 13, 1978.
Copies of the meeting notice, meeting attendees, and the schedule are included as attachments A, B, and C, respectively.
No requests for time to make oral statements were received from members of the public and no written statements were received.
The Designated Federal Employee for the meeting was Dr. Andrew Bates.
TUESDAY NOVEMBER 28, 1978 INTRODUCTORY STATEMENT BY SUBCOMMITTEE CHAIRMAN Dr. Plesset, Subcommittee Chairman, convened the meeting at 9:00 a.m.,
indicated the purpose of the meeting, and that the meeting was being conducted in accordance with the provisions of the Federal Advisory Committee Act and the Government in the Sunshine Act.
Dr. Plesset reviewed a series of questions which were raised but went unanswered at the November 17, 1978 Zimmer Subcommittee meeting (attachment D).
7903120483
4 s
Fluid Dynamics 11/28-30/78 HETHODS FOR COMBINING LOADS Mr. J. Knight reported on the status of the NRC staf f considerations on the use of SRSS methods for load combinations.
GE filed a report, NEDE-24010-P, with the staff in 1977 in order to lay a foundation for acceptance of SRSS methodology in Mark II containments.
The Staff, the applicants, GE, and consultants have expended a considerable amount of time dealing with the use of the methodology.
The Staff has been attempting to develop a rigorous founcation for acceptance and have found, that in dealing with the immediate problems of licensing the lead Mark II plants, such an approach is beyond its grasp. Recently the Mark II Owners Group, through Drs. Newmark and Kennedy, have presented a simplified criteria for the use of SRSS methodology.
The Staff is in the process of reviewing the criteria, which are based upon work done in combining components of seismic responses.
The proposed criteria include:
- 1) SRSS can be used if each of the time functions in either the loads or responses are similar to earthquake ground motions; cad 2) SRSS may be used if the cumulative distribution function (CDF) of the combined response time function meets the following:
a)~
SRSS represents at least 50% of the non exceedence probab'ility (NEP) and b) 1.2 SRSS represents at least 85% of the NEP.
Mr. Knight indicated that, in general, the staff considered the methodology a good approach in that it related SRSS dynamic load combinations to previously established earthquake engineering practice.
The preliminary NRC Staff comme.sts on the two criteria were presented (attachments E &
F). One of the basic Staff feelings is that the SRSS method must be applied to the characterization of responses rather than loads, until such time as better information is available to indicate that the response will follow the loading.
In response to a question, Mr. Knight indicated
i n
Fluid Dynamics s-
../28-30/78 that the criteria apply to each of the loads that are to be combined.
The dynamic loads would be combined by SRSS and then they would be combined with the other static or quasi - static loads.
The detailed methodology and detail for doing this would have to be worked out by the Staff and could take some time.
Mr. Knight indicated that the implementing criteria for use of the SRSS method would have to be very explicit, and in actual practice would require a great deal of communication between the Staff and the plant designer.
This work will require a major effort by the Staff over several years.
A lot of the questions asked by the Subcommittee and consultants regarding detailed implementation of the criteria have yet to be addressed; at this point in time the Staff is still looking at the basic philosophy of the combination criteria.
Basic to the Newmark and Kennec:y proposal is the thesis that the loads are conservatively defined and that the margin is established at the load level. The load combination method should then preserve the margin established at the individual load step.
The NRC Staff in their preliminary review of the Newmark and Kennedy proposals set down some criteria (attachment G) that they believe should be met by the loading functions.
The list may not be complete and some items could be removed if they do not prove to be important.
In response to a question, Mr. Knight indicated that the dynamic load combination method would only be allowed for low probability events; it would not be used for SRV loads alone (in combination with another low probability load it may be used). Whether or not it would be used for the OBE is a point for further discussion.
The 84% NEP value in the criteria is based upon work in earthquake engineering and the number might have to be altered for other types of loadings.
Other areas under exploration involve the number of loads that could be combined by SRSS and the extent to which it wculd be employed (structures, piping, primary system, etc.).
w Fluid Dynamic.
11/28-30/78 PRESENTATION ON SRSS BY THE MARK II OWNER GROUPS AND THEIR REPRESENTATIVES Mr. R. Klause, Stone & Webster, presented a brief description of the LOCA and SRV loads imposed upon the Mark II containment structures and the significiant load combinations.
Loads involved with the LOCA include pool swell loads, drywell pressurization, water impact loads, water drag loads, and steam condensation loads on the downcomers vents and other pool structures.
Also involved are tne loads on the reactor vessel and supports due to the rapid pressurization of the space between the vessel and shield wall.
The SRV loads are associated with air bubble oscillations and steam bubble condensation loads.
Mr. Klause explained the method by which the structural engineers design the structures.
As an example the load response from the SRV's discharging is calculated at a number of points. The amplified response spectrum is determined at each point of interest and peaks are broadened by 15%.
The individual response spectra are then combined into an overall re-sponse spectrum which is applied over the entire structure of interest.
The combined response spectra would then be used as an input to piping systems or other components which might be attached to the containment.
Mr. Klause then reviewed the load combination and acceptance criteria for the Mark II containments (attachment H).
Acceptance criteria for the overall loads are based upon meeting ASME service levels B or C.
Dr. Newmark reviewed the criteria that he and Dr. Kennedy developed for deciding that dynamic loads can be combined by SRSS (attachment I).
The Newmark and Kennedy approach is based upon their experience in earthquake engineering.
The basic probabilistic considerations for combination of earthquake responses were developed in the early 1950's by Drs. Rosenbluath, y
e
r i
Fluid Dynamics 11/28-30/78 Goodman, and Newmark.
The mathematical derivations were based upon normal and log-normal distribution functions.
It was found that if each of the individual quantities to be combined had a certain level of exceedance probability measured in terms of the number of standard deviations above the mean, and all of those probabilities are the same, then the combination by the use of SRSS has the same proDability of exceedance.
If all of the individual loads are cnaracterized as the mean plus one sigma value then the SRSS combinaticn produces a co bined load at the mean plus one sigma level. When response spectra for earthquakes were derived for the NRC by Blume and Newmark, they based their work upon the mean plus sigma values of the measured responses to a large number of earthquakes.
This was decided to be a proper degree of conservatism when the rest of the structural design conservatisms were considered.
Based upon this earlier work for earthquakes, Drs. Newmark and Kennedy decided that dynamic functions which are similar to earthquake loading functions could also be characterized in their combination by SR55. The characteristics that make the dynamic functions similar to earthquakes include similar time histories, rapid variation with time; also the func-tions should be uncorrelated and the functions should have a zero mean value.
(If they do not have a zero mean then the offset needs to be treated as a quasi-static load.)
The two criteria were developed help to assure that the loads will be similar to earthquake loads.
The first is applicable to load functions that are clearly similar to earthquakes, the second criterion applies to loads where it is not as clear that the loads meet criterion 1.
Dr. Kennedy presented the Subcommittee with additional detailed data on the application of the Newmark and Kennedy criteria to the Mark II containment loads.
Previous work was done by GE in NEDE 24010-P where e
Fluid Dynamics 11/28-30/78 291 actual Mark II structural components with actual load combinations were studied.
Various assumptions were made with respect to the time phasing and the probability density function of the loads, and the assumptions were then varied.
For these 291 cases GE found that there was an 86% NEP for the SRSS values derived. There was also a 96% NEP value for 1.2 times the SRSS combined responsa.
At a 95% confidence level there was a 30% NEP on SRSS and a 90% NEP value on 1.2 SRSS.
The GE report also found that there was little additional reliability provided by the use of absolute combinations vs. SRSS.
If one has conservative load definitions and a correspondingly high load reliability, the additional reliability provided by absolute combination vs. SRSS is not large.
Studies were also done on the dynamic reserve margins in structures.
It was concluded that the dynamic margins are generally greater than the static margins in ductile structures if the same code allowable stress level are applied and elastic analysis is employed.
The dynamic margin was in general at least 50% more than the static margin.
(This was verified for simple structures and loadings, but not proven mathematically.)
This margin is available where the loading excites a dynamic amplification of the response in an elastic analysis.
If the structural response is quasi-static then this additional dynamic margin does not exist.
This topic is explored further in the Dynamic Margin Report EDAC-134-240.7R.
This dynamic reserve margin can in some cases provide additional margin to dynamic loadings and load combinations over and above margins that might be available in the load definitions.
Dr. Kennedy indicated that the basic conservations in structures should be included in the load definition rather than in the combination method.
Use of the SRSS method maintains a consistent level of conservatism in the load combination and does not add conservatism over that which t
F
r Fluid Dynamics il/28-30/78 exists in the load definitions.
If the load NEP values are not equal (i.e., one is 50%, the other 84%) then the NEP value of the combined load will lie between 50% and 84% depending upon the relative magnitude of the individual loads (if one load dominates its NEP value will dominate the combined load).
One then has to have acceptably high NEP values for each of the loads that are to be combined.
One area that might require some analysis is the validity of the assumption of normal or log-normal loads and responses.
Real systems may have somewhat different distributions.
Dr. Kennedy then examined the combination of SRV discharges and the OBE in light of the SRSS combination criteria.
In practice the OBE values tend to be such that they will probably not occur during the life of the plant, and will not occur more than once.
The SRV discharges are conser-vatively characterized by the load phasing and frequency as well as the number discharging.
Foe the combination of these events one would hold to stress levels that we would accent during the life of the plant.
(service level B).
Dr. Kenndy reviewed the basis for the statement that SRSS maintains a consistent level of nonexceedance probability (attachment J).
If the dynamic functions meat either one of the two criteria developed, the use of the SRSS method for load combination could be used.
Dr. Zudans emphasized the need to show that the probability distribution functions for the various loads to be combined do indeed allow combination by SRSS methods without a reduction in the overall nonexceedance probability.
Mr. Subramanian, GE, reviewed how the Newmark - Kennedy criteria have been applied to the Mark II loads. A draft copy of a report has been submitted to the NRC.
Loads reviewed include the SRV plus OBE, em
E 1
Fluid Dynamics o-
.I/28-30/78 LOCA chugging plus SRV plus SSE, and all SRVs plus the SSE. The individual loads were compared with the two criteria of Newmark and Kennedy.
Mr. Subramanian indicated that all but six load combinations meet the requirements of criterion 1 or 2, the six that did not were just outside the bounds of criteria 2-a (NEP (SRSS) F 50%).
A small change is the load combination over SRSS would allow the criteria to be met (attachment K).
Dr. Catton indicated that he had some reservations concerning the ability of the various loads to meet the zero mean value criteria and would like to see more information on the confidence levels in the load definitions.
Mr. Klause reviewed the service levels that the various load combination must meet.
For those loads that require a service level C the Mark II Owners must prove functional capability of the system.
Dr. Rodabaugh from Battelle reviewed the work that he had done on functional capability for essential Mark II piping.
Loss of functional capability was defined as more than a 5% reduction in the ficw area of the pipe.
The ASME Codes do not address functional capability.
ORNL also did some work for the NRC on functional capability.
As a conclusion to the work done it was found that some of the ASME Code requirements might not assure functional capability. A NUREG report (CR-0261) was issued which contains some recommendations on functional capability.
The report indicates that in some cases substantially more restrictions on allowable loads are needed to assure the functional capability of piping systems designed to service level C.
The results are based upon single analytical models for piping elbows and tees and some experimental data on pipe elbows.
The functional capability criteria would be imposed upon the Mark II plants for the various load combinations.
....,,...s....
Fluid Dynamics 3-
.I/28-30/78 Mr. Sobon summarized the key points presented during the day by the Mark II Owners Group. One of the key areas that remains to be resolved with the NRC staff is the acceptability of the use of the SRSS method for load combinations.
Conservatisms exist all through the design process and ii t.he criteria developed by Newmark and Kennedy are used the basic conservatisms in the load definition will remain in the load combinations.
WEDNESDAY NOVEMBER 29, 1978 NRC REVIEW 0F THE MARK II LOAD DEFINITION AND ACCEPTANCE CRITERIA Mr. R. Tedesco, NRC staff, indicated that the Staff and their consultants would be updating the Subcommittee on the status of the Mark II review.
The Staff has now completed its review of the load definitions proposed by the Mark II Owners and has issued a list of acceptance criteria.
Many of the acceptance criteria agree with the Owners proposals, however some place some additional margin on the load definitions.
The review at the present time is directed to the lead Mark II plants; further down the line, additional work will be done which may result in better resolution of loads for the intermediate plants.
The Staff considers the load definitions that have been developed to be bounding loads.
The Staff report on the Mark II plants is contained in NUREG-0487.
The Mark II Owners Group has taken exception to 5 of the approximately 25 acceptance crite. established by the NRC.
The exceptions are to the pool swell elevation, small structure impact loads, asymmetric pool swell, SRV bubble phasing and frequency, and submerged drag loads.
The Staff will be meeting with the Owners and expect to resolve or complete the review of the exceptions by the beginning of January 1979.
Modifi-cations to the design of most of the Mark II plants have already been made as a result of the loads identified in the Mark II review (attachment L).
C
Fluid Dynamics 11/28-30/78 In answer to a question that developed at the Zimmer Subcommittee meeting on November 17, 1978, Mr. Tedesco indicated that Zinner had cut the flanges off the bottom of their downcomers rather than to run additional tests to justify the acceptability of loads with flanges.
REVIEW 0F SRV LOAD AND POOL TEMPERATURE LIMITS - Mr. N. Su The SRVs are required to function for various anticipated plant opera-tional transients, for operation of the Automatic Depressurization system in the event of a small break LOCA, and they may operate in a category of inadvertent blowdown.
The Mark II plants each contain 11 to 15 SRVs. Actuation of the SRVs compresses air contained in the line and the steam released pushes the air ahead of it into the suppression pool.
Loads develop from the air bubbles expelled and then from the steam which is expelled.
Various types of quencher devices on the pipes at the discharge end can mitigate the loads produced.
Some of the devices can experience unstable pool oscillations at elevated pool
. temperatures.
The Mark II Owners have all committed to the use of a quencher device which has been shown to allow stable discharge up to 200*F.
In-plant tests will be conducted at Zimmer to verify the loads; tests are ongoing at Caorso and Tokai and that data should be available soon.
The discharge lines also contain vacuum breakers to prevent fluid from being pulled back into the pipe at the end of SRV discharge as the remaining steam condensers.
Single and multiple discharges from one or more SRVs have been considered with the actuation occurring both in phase and out of phase.
The Mark II plants were originally designed for the ramshead type quencher, and since all changed to the T-quencher which produces considerably lower loads, SRV discharges do not appear to be a serious problem for the Mark II plants at the present time.
Fluid Dynamics 11/28-30/78 The Staff indicated that at the present time they have not included fluid-structure effects for the SRVs; they have been included for chugging in the LOCA event.
Two areas of disagreement on the SRVs involve the bubble phasing to be used and the frequency range to be considered.
The Staff is asking for in-phase bubbles, the Owners want to take credit for differences in pipe length and valve set points.
The Staff wants a range of frequencies considered; the Owners want to use the calculated frequency.
Dr. P. Huber, M.I.T., reviewed the confimatory programs relevant to the SRV air clearing loads. The confirmatory program has consisted of small-scale and large-scale tests and in-plant tests will be or are being made.
Tests at 1/2 scale were conducted at NUTECH and full-scale tests at KWU.
The tests covered the entire range of expected operation of the quencher devices.
Preliminary indications are that there have been no surprises in the first tests conducted at Caorso.
REVIEW OF THE LOCA RELATED MARK II CONTAINMENT LOADS Mr. C. Anderson, NRC Staff, reviewed the load on containment resulting from a LOCA. These include the drywell pressurization vent clearing, air bubble oscillations, pool swell, pool impact, jet loads, drag loads on submerged structures, and chugging loads to the steam bubble condensation.
The chronology of the loadings was reviewed (attachment M), as well as the NRC acceptance criteria for the various loads (attachment N-1 through N-6).
The present acceptance criteria is based upon tests conducted to date and the information presented by the Mark II Owners. Additional infomation and work is expected to allow some relaxation of criteria for the longer term plant schedules.
Fluid Dynamics 11/28-30/78 The Mark II Dwners presented a Pool Swell Analytical Model (PSAM) which has been used to obtain pool swell height and velocities as well as bubble pressures, wetwell air compression, and pool acceleration.
The NRC has reviewed this model and has found portions of it acceptable and has asked for some additional conservatisms in the areas related to the pool elevation, and velocity.
The Staff has also asked for a bounding approach to the asymmetric flow from the downcomer, using an assumption of all steam flow from one half of the downcomers and all air flow from the other half during drywell pressurization.
The Owners plan to present additional information on steam air mixing in order to have this criteria relaxed.
The Owners believe that the asymmetry could be reduced to about 7% of the Staff assumptions.
The Staff also added some conservations to the upward diaphram loads by specifying a modified correlation.
In response to questions from the Subcommittee and consultants, the NRC Staff and Mark II Owners Group indicated that SSE effects that might change LOCA loads (due to pool sloshing and different downcomer submergence depths) have not been considered in load combinations or load definitions; however, GE indicated that it has been at least thought about qualitatively and the Owners don't see any problems arising since there are counter balancing effects.
The NRC Staff has also specified a method for flat pool impact on structures that the Owners are taking exception to as being overly conservative.
The NRC Staff specification on vent lateral loads was also increased for downcomers with higher resonance frequencies. This was based upon higher loads observed in some of the foreign test facilities.
The tests SMM-
Fluid Dynamics 11/28-30/78 examined downcomers of various sizes and various pool temperatures as well as steam mass fluxes.
Dr. Catton indicated that he would be interested in seeing a plot comparing steam mass fluxes into the pool vs. time, with pool temperatures vs. time and the loads produced by the chugging.
The loads on the wall for the chugging were based upon the downcomers chugging in phase with uniform and asymmetric loadings on the walls.
The loads are applied at a frequency designed to maximize the effects on the walls.
The work on chugging loads has been based on tests with one downcomer at full scale; data is becoming available on multiple downcomers and it shows the single vent data to be conservative.
The NRC also looked at submerged structure drag induced by fluid flow in the containment.
Some modifications to the Owners Group proposals were also made in this area.
In general, this involves the use of accelera-tion drag coefficients rather than standard drag coefficients.in accelerating fluid fields.
Dr. C. Economus, BNL, reviewed in detail the pool swell associated loads. The NRC has asked for a 10% margin on the pool swell velocity and elevation and changes in the Owners proposal for the wetwell compres-sion. The Owners have responded with a counter proposal for the elevation and the pressure.
The two groups are still discussing this area; it appear's that the Owners Group's last proposal will be acceptable.
It was reported that the Japanese are in the process of building a full scale, 7 vent facility to study pool swell.
Fluid Dynamics 11/28-30/78 On the subject of asymmetric loads during the pool swell the Owners have argued that sonic flow of steam will promote mixing with air and that deflector plates at the downcomers will prevent direct flow of steam to the downcomers.
The Staff believes that the arguments have merit but have been unable to quantify the degree of mixing at this time.
Dr. G. Maise, BNL, reviewed the pool swell impact loads in detail.
The BNL model is based upon an emperical model that has been compared to Navy data where structures were dropped on to flat fluid surfaces.
They specified a shape for the impulse load and the time of action and amplitude were based upon test data.
A 35% margin was added to the maximum pulse to account for data scatter.
Dr. Plesset indicated that fluid-structure impacts had been studied prior to this work and that analytic solutions to the problem exist.
He suggested that it would be more appropriate to use a correct physical approach to the problem.
The Owners Group has taken exception to the NRC criteria.
The Owners have argued that the pool surface will not be flat when it hits structures and thus will not produce as high a load.
The present NRC Staff position is that it is now up to the Owners to propose a more acceptable criteria or model for the impact loads.
Dr. Economus reviewed in detail the chugging loads on the downcomers.
The impulsive loads due to condensation of the steam bubbles is specified as a function of the frequency of the downcomer.
The Owners Group based the loads from multiple vents on a probabilistic method of combination where a Monte Carlo technique is used to determine loads down to an
~4 exceedance probability of 10 In response to Dr. Catton, it was indicated that the loads had been looked at and found to be random in direction. The Staff has found the procedure to be acceptable with minor modifications which the Owners have accepted.
Fluid Dynamics 11/28-30/78 Dr. Scanlon reviewed the loads at the pool boundry produced by the collapse of the steam bubbles.
Part of the problem in this area was the separation of the fluid structure interaction so that test data could be used to develop a pure forcing function for dirsimilar structures at the various reactors. A large part of the problem had to do with the 4T test structure and the way in which the data was obtained at the vessel walls which were flexible.
It is agreed that the measured loads on the 4T facility were conservatively large; however, there is still the problem associated with determir.ing the proper frequency of the pressure loads without the wall interactions.
The Staff feels now that the loads are conservative but discussion is ongoing on the frequency issue.
Dr. G. Bienkowski reviewed the submerged structure drag loads.
Fluid motions in the pool during SRV actuation and LOCA's produce drag on structures in the pool.
Both velocity and acceleration drag is considered.
The analysis methods are fairly straightforward and subscale tests are being conducted.
The Staff has taken issue with some of the proposed models the Owners presented.
It is felt that steady state drag data may not be conservative, and that induced fluid motion from fluid tests need
~
to be included. The NRC Staff acceptance criteria take into account these extra considerations through modifications to the Owners original proposals.
Se Owners have presented some counter-proposals and modifications to the NRC criteria. he NRC expects to reach agreement with the Owners after some further discussion.
Mr. R. Cudlin, NRC Research, reviewed the NRC Research programs which are related to the Mark II containment issues. Studies are being conducted at UCIA and MIT on air steam venting in pools of waters.
(Dr. I. Catton did not participate in the pertion of the meeting dealing with the UCIA work.)
Fluid Dynamics 11/28-30/78 The MIT work involves investigation of the scaling laws for air venting, steam consideration chtsging, and some simplified fluid structure inter-action work. Similiar work is beirg conducted at UCIA on air and steam venting phenomena.
Lawrence Livermore Laboratory is working on the PELE-1C cceputer code which couples a two-dimensional incompressible fluid code to a struc-tural shell code for fluid / structure interactions. Be basic code has been developed and they are in the process of doing some benchmark cal-culations.
Tests were conducted at Marviken where integral dynamic loads, pool swell phenomena, art 3 enndensation phenomena were measured. h sts will be done in the near future at the GKS facility in Germany with 1, 2, or 3 full size vents in order to measure condensation loads. Japan will be conduct-ing tests at 1/6 scale and full scale. Se full scale tests should start this spring and the facility includes 7 vents.
Dr. W. Butler summarized the NRC Staff w rk on the Mark II containment review.
Se Lead Plant Program has been extpleted with the issuance of NUREG-0487 and the acceptance criteria contained in it.
L e intermediate plant program will develop additional material which should allow relax-ation of some of the criteria developed for the lead plants which wre based on conservative bounding loads. Se Staff believes that the appli-cation of the criteria to the Zinner plant and the final clean up of several exceptions 2irmner is taking should be cleaned up by the end of the year, subject to the acceptability of the SRSS approach.
Shoreham and IaSalle will require additional Staff work due to the additional exceptions they are taking to the acceptance criteria.
Dr. Plesset thanked the p rticip nts and recessed the meeting.
Fluid Dynamics 11/28-30/78 THURSDAY, NOVEv3ER 30,1978 Mr. B. Sobon, GE indicated that the Mark II Owners Group and its representa-tives could review their ongoing programs that are intended to define and refine the loads associated with the LOCA and SRV discharges in the Mark II conta inments. S e lead plants have taken conservative Lounding approaches to the load definitions. Se Intermediate and Iong Term Programs will re-fine the load definitions and renove excessive conservatisms. Se areas of work involve SRV loads and quencher design tests with multiple value actua-tions.
Mr. A. Smith reviewed the intermediate SRV quencher load program. GE is working on improved models for the loads and will revise and up-date their present models with new data as it becomes available frca various in-plant tests at Montecello, Caorso, and Takai. Also included will be data from European tests. A better model for the attenuation of the pressure pulses to vessel walls will be included in the work. hst data will incluae single and multiple valves with simple and multiple actuations.
Mr. M. Davis, GE, reviewed the preliminary test results from Caorso. Tests have been or will be run with 1, 2, 3, 4 and 8 valves at various pressures for various lengths of time. Multiple valve actuation tests will be done also. Se containment is extensively instrumented for presure and tempera-ture. Single valve tests were run in August 1978, preliminary analysis of the data indicates that the loads were as expected and the design is conservative. Pool boundary loads are perhaps a factor of 2 below the loads predicted by the DFFR methodology.
Mr. D. Both, PP&L, reviewed the results of T-quencher tests sponsored by PP&L and run at Karlstein, Germany. Tests were conducted in May of 1978 and the test report should be sulxnitted in February 1979. Se tests were conducted with long and short pipe runs in the Susquehanna SRV line con-figuration. Both air and steam discharge tests were run. Test results show loads that range from 31% to 51% of the DPER calculated loads for
Fluid Dynamics 11/28-30/78 ranshead type devices.
Iocal high pressures observed in some te-'s account for the 51% value; most of the data was around 30-35% of the DFFR load definition.
Mr. H. Brinkman, Cincinnati Gas & Electric, reviewed the SRV quencher tests that Zimmer will conduct during start up operations. Zimmer was originally designed for ramshead type loads for the SRVs and they used conservative LOCA loads. Se plant evaluations were done with out the use of SRSS in the load combinations. Se T-quencher has replaced the original ramshead device and it is expected that the resulting loads will be considerably smaller, and will also provide pool stability at elevated temperature. te Zimmer tests are aimed at proving that the quencher loads are less than the ramshead loads. Wey intend to demonstrate the conservatisms in the analysis of the original plant design. Tests will be conducted with long and short lines, not and cold lines, and single and multiple discharge.
Rey will also do a turbine trip at full power. W ey will measure pool pres-sure and temperatures as well as plant structural responses.
Mr. A. Smith reviewed the program that will better define the IOCA related laternal downcomer loads, chugging loads, and the subnerged structure loads. Multiple vent tests will be done at Creare at 1/10,1/6, and 5/12 scale in order to obtain data on multiple vent interactions and loads.
Additional tests on drag loads are being conducted. We results from the current and future tests as well as previous tests will be used to improve the present models for the IDCA related loads. It is intended that the Intermediate Program load definition tasks be completed by early 1980.
Dr. Plesset thanked the meeting participants and adjourned at 11:15 a.m.
For those who want a more detailed review of the meeting a complete transcript is on file at the NRC Public Ibcunent Rocra,1717 H Street, N.W.,
F
Fluid Dynamics 11/28-30/78 Washington, D.C., or from ACE Federal Reporters, 444 North Capital Street, N.W., Washington, D.C.
A complete set of the viewgraphs used is on file with the record copy of the minutes.
O e
H yesq,
,e 6
NOTICES
[7590-01-M]
The Subcommittee may meet in Ex.
NUCLEAR REGULATORY eeutive Session, with any of its consul.
COMMISSION tants who may be present, to explore and exchange their preliminary opin-ADVISORY COMMITTtt ON REACTOR SAFE
- ons regarding matters which should be Considered during the meeting and to G UA R D S, SU& COMMITTEE ON FtOID DY*
formulate a report and recommenda.
NAMICS tions to the full Committee.
gg;,y At the conclusion of the Executive The ACRS Subcommittee on Fluid Session, the Subcommittee will hear Dynamics will hold a meeting on Nov.
presentations by and hold discussions s1th representatives of the NRC Staff, 28, 29, and 30,1978, at the Barrett the General Electne Co., the Mark II Motor Hotel. 501 Post Street, San Owners Group, and their consultants, Francisco. Calif., to discuss and review pertinent to the above topics. The the design basis and construction of Subcommittee may then caucus to de-the mark II Boiling Water Reactor termine whether the matters identi-tBWR) Containment System. The fled in the initial session have been Mark II load defirution load accept.
adequately ecvered and whether the ance criteria, and load combination project is ready for review by the full methods will be discussed. Notice of Committee.
this meeting was published September In addition, it may be necessary for 21 and October 20,1978 (43 FR 42826 and 49081, respectively).
the Subcommittee to hold one or more In accordance with procedures out.
closed sessions for the purpose of ex-lined in the ProcRat Rzcasm. Oct. 4, ploring matters invohing propnetary information. I have determined. In ac-1978 (43 Fe, 45926), oral or written
, cordance with Subsection 10(d) of statements may be presented by mem.
Pub. L.92-463, that, should such ses-bers of the public. recordings will be permitted only during those portions sfons be required, it is necessary to of the meeting when a transcript is close these sessions to protect propri-etary information (5
U.S.C being kept, and questions may be 55Tb(cX4)). '5 R h
.!T S a b a asked only by members of the Sub.
Further Information regarding committee, its consultants, and Staff.
Persons desiring to make oral state-topics to be discussed, whether the meeting has been canceled or resche-ments should notify the Designated duled, the Chairman's ruling on re-Federal Employee as far in advance as quests for the opportunity to present practicable so that appropriate ar.
oral statements and the time alloted rangements can be made to allow the therefore can be obtained by a prepaid necessary time during the meeting for telephone call to the Designated Fed-auch statements, eral Employee for this meeting, Dr.
The agenda for subject meeting Andrew L. Bates, telephone 202-634-shall be as follows:
3267, between 8:15 a.m. and 5 p.m.,
Tuesday. Wednesday, and Thursday, No.
EST.
vember 28. 29. and 30,1sTs-8 30 man. untu the conclusion of business each day, Dated-November 6,1978.
yogy g, goygg, Advisory Committee Management Officer.
(FR Doc. 78-319oo F11ed 11-s-78; 835 aml FIDttAt REGt1 Tit. Vot. 43, No. 219-MCNDAY. NOVEM8tt 13,1978 ATTACHMENT A
ACRS SUBCOMMITTEE MEETING ON FLUID DYNAMICS SAN FRANCISCO, CALIFORNIA ATTENDEES LIST NOVEMBER 28, 1978 ACRS NRC M. Plesset, Chairman A. Hafiz H. Etherington, Member J. D. Carlson T. Theofanous, Consultant C. I. Grimes I. Catton, Consultant K. R. Wighman Z. Zudans, Consultant C. J. Anderson H. Sullvan, Consultant F. Schauer l
L. Yao, Consultant S. N. Hou
~
T. Wu, Consultant J. O'Brien K. Garlid, Consultant A. Bates, Staff
- CINCINNATI GAS &
T. Eaton, Fellow ELECTRIC H. C. Brinkmann
- Designated Federal Employee BECHTEL ISHAM, LINCOLN & BEALE J. Gallo D. M. O'Connor L. C. Hua J{
yandt PUBLIC SERVICE C0 0F OKLA V. L. Conrad G. H. Shah J. B. West H. Hiraoka S. LEVY INC.
SARGENT & LUNDY E. D. Fuller G. T. Kitz K. J. Green NSC MHB TECH ASSOC S. W. Tagart, J r.
D. Bridenbaugh BATTELLE-COLUMBUS LILC0 E. C. Rodabaugh A. W. Wofford BURNS & ROE INC H. Chau T. Jozkowski NUTECH - MKI STONE & WEBSTER W. J. McConaghy N h in GENERAL ELECTRIC R. L. O'Mara M. E. Urata S. B. Mucciacciaro L. J. Scbon R. P. Klause A. R. Smith T. Trocki PHILADELPHIA ELECTRIC CO.
ATTACHMENT B
Attendees List (Cont'd) NOVEMBER 28, 1978 PENNSYLVANIA POWER & LIGHT COMMONWEALTH EDISON D. F. Roth G. R. Crane E. M. Mead ENGINEERING DECISION LAWRENCE LIVERMORE LABORATORY ANALYSIS C0 F. J. Tokarz R. P. Kennedy P. D. Smith URBANA, ILL - CONSULTANT Nathan M. Neymark NOVEMBER 29-30, 1978 ACRS NRC M. Plesset, Chairman P. Huber, MIT H. Etherington, Member C. Anderson, DSS /CSB T. Theofanous, Consultant J. Kudrick, DSS /CSB I. Catton, Consultant B. Tedesco Z. Zudan:, Consultant T. Su H. Sullivan, Consultant W. Butler L. Yao, Consultant R. L. Cudlin T. Wu, Consultant A. Hafiz X. Garlid, Consultant W. R. Butler, DSS /CSB A. Bates, Staff
- F. Schauer, DSS /SEB T. Eaton, Fellow C. I. Grimes, DOR
- Designated Federal Employee STONE & WEBSTER PRINCETON UNIV /BNL C. C. Lin R. L. O'Mara R. H. Scanlan S. B. Mucciacciarc G. Bienkowski CFE BROOKHAVEN (BNL)
T. Zazueta G. Maise EBASCO A. A. Sonin, MIT/NRC H. S. Yu C. Economos, NRC BURNS & ROE LONG ISLAND LIGHTING C0 T. Jozkowski H. Chau EDS NUCLEAR W. J. Museler D. M. Witt A. W. Wofford
Attendees List (Cont'd) NOVEMBER 29-30, 1978 SRI INTERNATIONAL blI G. R. Abrahamson J. M. Healzer PENNSYLVANIA POWER & LIGHT E' "'.
oth A. R. Smith BECHTEL POWER CORPORATION W. M. Davis H. H. Safwar R. B. Johnson.
J. A. Weyandt K. G. Hazifotis D. M. O'Connor M. E. Urata L. Hua L. V. Sobon NUTECH A. F. Deardorff C
N. Kra hnawamy PHILADEPHIA ELECTRIC C0 RDH ENGINEERING H. W. Vollmer R. D. Hoagland MHB wppsg D. G. Bridenbaugh R. Guizar CINCINNATI GAS & ELECTRIC UNIVERSITY OF CALIFORNIA -
H. C. Brinkmann LLL D. M. Norris, Jr.
A4 LS c.-
Proposed Agenda ACRS Meeting Fluid Hydraulic Dynamic Effects Subcommittee San Francisco, California November 28, 1978 9:00 AM 9:00 - 10:30 I.
Methods for Combining Loads (NRC/J. Knight) 90 min.
10:30 - 11:30 II. Mark II 0.G. Corments on Load Combinations Acceptance Criteria (Mark II 0.G.)
60 min.
11:30 - 12:30 LUNCH 12:30 - 3:00 II.
(Continued) Mark II 0.G. Comment on Load Combination Acceptance Criteria (Mark II 0.G.)
150 min.
November 29, 1978 9:00 AM 9:00 - 9:20 I.
Introduction (NRC/R. Tedesco) 20 min.
9:20 - 10:50 II. SRV Related Hydrodynamic Loads A.
SRV Loads Overview (NRC/T. Su) 30 min.
B.
Air Clearing Loads 1.
Acceptance Criteria (NRC/T. Su) 20 min.
2.
Confirmatory Programs (NRC-MIT/P. Huber) 30 min.
C.
Pool Temperature Limits (NRC/T. Su) 10 min.
10;50 - 11:00 BREAK 11:00 - 11:45 III. LOCA Related Hydrodynamic Loads A.
LOCA Loads Overview (NRC/C. Anderson) 45 min.
11:45 - 1:00 LUNCH 1:00 - 2:45 III. (Continued) LOCA Related Hydrodynamic Load S.
Pool Swell (NRC-BNL/C. Economus) 30 min.
. 1.
Pool Swell Elevation and Velocity 2.
Asytmetric Loads C.
Impact Loads (NRC/BNL/G. Maise) 45 min.
1.
Small Structures 2.
Gratings D.
Steam Condensation Loads 1.
Vent Lateral Loads (NRC-BNL/C. Economus) 15 min.
2.
Chugging - FSI Considerations (NRC-Princeton /R. Scanlon) 15 min.
IV. LOCA/SRV Submerged Structure Drag Loads (NRC-Princeton /G. Bienkowski) 75 min.
A.
Jet Loads B.
Air Bubble Drag Loads 4:00 - 4:10 BREAK 4:10 - 4:35 V.
Related NRC Research Programs (NRC/R. Cudlin) 25 min.
4:35 - 4:45 VI. Conclusions (NRC/W. Butler) 10 min.
4:45 - 5:00 VII. Comments (Mark II OG) 15 min.
November 30, 1978 9:00 AM 9:00 - 10:00 Mark II 0.G. Comments on Hydrodynamic Loads Acceptance Criteria (Mk II OG) 60 min.
10:00 - 10:10 BREAK 10:10 - 11:30 Mark II SRV Test Programs (Mk II OG) 80 min.
A.
NUTECH B.
KWU C.
Zimmer Inplant Tests
. 11:30 - 12:30 LUNCH 12:30 - 3:00 Inter:r.ediate Program Tasks (Mk II OG) 150 min.
A.
Dynamic Lateral Loads B.
Chugging Loads 1.
Ringout Removal 2.
Multivent Model 3.
Test Programs C.
Submerged Structure Drag Tasks 1.
Analytical Programs 2.
Test Programs D.
SRV Tasks 1.
New Four-Arm Quencher Load Methodology 2.
Temperature Limits
WLj o QUESTIONS FROM NOVEMBER 17, 1978 ZIMMER SUBCOMMITTEE MEETING 1.
Calculation of thermal stresses in the dry well floor -- calculation of film coefficient (overpressure calculation versus thermal stress calculation).
2.
Location of series isolation valves on pipes penetrating containment.
3.
Vulnerability of control rod lines during pipe breaks.
4.
Pool temperature history during LOCA -- Maximum temperature during steam condensation phase.
5.
Flanges on Zimmer downcomer.
6.
30 ft. long downcomer without bracing.
7.
Effects of adjacent pipes -- all tests with single vents.
8.
Asymmetric pool swell loads.
9.
SRV bubble phasing and frequency.
- 3. STAFF COMMENTS :
.i I. CRITERION IS SIMPLE MAY ACHIEVE HIGH NOM-EXCEEDANCE PROBABILITY (NEP) IN tvf0ST CASES.
HOWEVER, IT APPEARS To SE EASED ON ENGINEERING JUDGEMENT WITHOUT PATA 6 ASIS TO VERlFY THE CLAIMED 843 NEP.
- 2. PARAMETERS USED MAY NOT BE ADEGUATE TO ENSOR E JUSTIFlABLE SRSS APP' ICATloN IN ALL CASES.
PARAM ETRic STUDIES OR NUMERICA L EXAMPLES ARE SUGGESTED.
- 3. COMBINATroN OF RESPONSES SHOULD SE SOLELY BASED ON THE CHARACTERISTICS OF THE RESPONSE TfME FUNCTIONS,
NOT THE LOADING TIME FUNCTIONS EXCEPT EARTH GUA KE.
- 4. DEGREE OF CORRELATION AMONG TIME FUNCTIONS ARE DIFFICULT To DETERMINE BY ONLY OSSERVATIoN.
5, MANY RESPONSE TIME FUNCTIONS MAY NOT HAVE ZERO fr MEAN.
NO GuipANCE IS PRoviDED.
, \\ll
H.
FOR CRITERION 412 :
A.
SUMMARY
OF THE CRITERION :
SRSS MAY BE USED IF CDP OF COMBINED RESPONSE TIME FUNCTIONS M EET THE For LOWING -
1.
SRSS REPRESENTS AT LEAST 50 7, (I) NEP.
2 I.2 SRSS REPRESENTS AT LEAST 85 7. (t) NEP.
- 13. STAFF COMMENTS -
I.
THE INTENT OF THE PROPOSED cRtTERIA IS To Aceleve 84 % NEP.
IT MAY SE IN ERROR. To use SRss BASED O H B o la (t) N E P.
2.
THE PURPOSE 4 6 ASIS TO ADD REQUIREMENT OF 85 7o Lt).NEP AT I.2 SRSS IS UNCLEAR SINCE THE PROPOSED VALUE TO BE USED IS SRSS, NOT l.2 SRSS.
- 3. CDF CURVE MAY NOT BE UNIGUE.
GUIDANCE TO ENSORE CURVE VALIDITY 15 NEEDED,
NRC INTERIM ACCEPTANCE CRITERIA FOR SRSS I.
EACH OF THE RESPONSE TIME FUNCTIONS TO BE COMBINED
~
SY SRSS SH00LD MEGT THE FOLI.OV)iNG CHARACTERISTICS A.
FUNCTION IS RAPIDLY VARY (NG WITH TIME S.
DURATION OF THE STRONG MOTION PORTION is SHoRT C. FUNCTION CONstsTS OF A FEW DISTINCT HIGH PEAKS
[N RANp0M APPEARANCE D.
RESPONSE
IS NOT ASSOCIATED WITH NORMAL PLANT OPERATroN EVENT E. PHAslNG RELATIONSHIP AMONG FUNCTIONS TO SE COMSINED ARE RANDOM.
F.
RESPONSE
IS CALCULATE.D ON LINEAR ELASTIC BASIS 2.
SRSS MAY BE USED IN COMBINING RESPONSES TO THREE z
PERPENDICULAR COMPONENT LOADIN GS JF THE COMPONENT LOADINGS ARE RELATIVELY UNCORRELATED,
- a. OnOO mni SE..USED IN COM6LNING MODAL RESPONSES
?ROVIDED THAT THEIR FRE610ENCIES ARE NOT CLOSELY SPACED AS DELINEATED IN REGULATORY Gusoe I.92
- 4. SRSS MAY BE USED IN COMBINING TWO RESPONSES IF THE SRSS LEVEL REPRESENTS A NON-EXCEEPANCE PROPASIL EG GA L OR GREATER TH AN 84 7o.
IF THE PRO 6ASILITY is LESS THAN 84 70.
THE LGVEL AT 84 7.
MAY SE USED 5.
FOR COMBINING MORE THAN TWO RESPONSES, SRss IAhT (5E USED IN COMBINING ANY Two 0F THE. RESPONSES, PRO VIDED ALL RESPONSE TIME FUNCT(otis HAVING CHARACTERISTICS AS DELINEATED IN ITEM L.
THE REMAINING RESPONSES HAVE TO SE COM6tNED in ASSOLUTE SUM.
- 6. SRSS MAY SE USEP To COMBINE MORE THAN TWO RESPONSES IF THE NON-EXCEEPANCE PPsOSASILITY EQUAL OR GREATER THAN B4 7 CAN SE DEMONSTRATED.
7.
SRs3 MAY SE USED LF ET CAN _SE VERIFIED THAT THE DYNAMIC MARGINS AT THE SPECEFIC STRUCTORAL LOCATIQi CAN ADEGUATELY COMPENSATE FOR THE UNGERTAINTY OF S RSS AS INpicATED IN THE EXCEEDANCE PROBA SI LIT'(.
- 8. SRss MAY SE USED IF IT CAN SE VERIFIED THAT THE OVERALL PEslGN MARGINS AT THE SP5ciFIC STRUCTURAL LOCATroN CAN APEQUATELY COMPENSA TE FOR THE UN CERTAINTY OF SRSS AS INDICATED IN THE EXCEEDANCE pro BA St LiTY.
- 9. SRSS MAY 66 USED IF 1T CAN BE VERIFIED THAT THE' OVER ALL STRESS RESOLTANT IS FAR GREATER THAN THE DYNAMIC PORTION OF THE STRESS ANP THE UNCERTAINTY PORTIor) INDOCED BY USING 3RSS CA N 85E COM PENSA TED
[513 USTiFiASLE MEANS.
(0. SRSS MAY BE USED TO COMSINE RESPONSES OF LOCA $ SSE '
IN THE MAlfi LOOPS OF THE LIGHT WATER PLANT.
FOR MARK 11 APPLICATIONS LOAD ACCEPTANCE I3I SRV "., SRVr r OBE SSE IBA *3I I
U CASE N
DBA CRITERIA X
ADS 1
X X
B 2
X X
X B
3 X
X X
C(4) 4 X
X X
C(4) 5 X
X X
X C(4) 6 X
X X
X(2)
C(4) 7 X
X X(2)
C(4)
(1) USE SBA OR IBA WHICHEVER IS GOVERNING..
(2) LOADING DUE TO DBA/SBA/IBA IS DETERMINED FROM RATED STEADY STATE CONDITIONS.
(3) N-NORMAL LOAD CONSISTS OF PRESSURE, DEAD WEIGHT, THERMAL. &
FLUID REACTION LOADS.
(4) PIPING FUNCTIONAL. CAPABILITY SHOULD BE ASSURED. SERVICE LEVEL LIMITS HIGHER TilAN THE LEVEL SPECIFIED IN THIS TABLE MAY BE USED, PROVIDED PIPING FUNCTIONAL CAPABILITY 15 DEMONSTRATED.
(5) SBA,lBA AND DBA SHALL INCLUDE ALL EVENT INDUCED LOADS WHICHEVER g
ARE APPLICABLE, SUCH AS POSSIBLE ANNULUS PRESSURIZATION LOAD, POOL (ig-7 SWELL LOAD, CONDENSATION OSCILLATION LOAD, CHUGGING LOAD, ETC,
NU 'E - i 2 October 1978 CRITERIA FOR COMBINATIONS OF EARTHOUAKE AND/OR OTHER TRANSIENT RESPONSES Preamble.
l The intent of the methods proposed f,or combinations of transient, dynamic responses is to achieve a non-exceedance probability of approx-imately 84 percent for the peak combined response of the system, com-ponent, or element considered.
This goal is achieved by compliance with any one of the following criteria, or any alternative method that meets the intent stated above, provided that the intensity of loads or acceler-ations fer each input are conservatively represented (approximately at the level of the 84th percentile, or the mean plus one standard deviation, of the expected input ;ntensity).
1.
Criterion.
Dynamic or transient responses of structures, components and equipment arising from combinations of dynamic loading or motions may be combined by SRSS provided that each of the dynamic inputs or responses has characteristics similar to those of earthquake ground motions, and that the individual component inputs can be considered to be relatively uncorrelated; i.e., the individual dynamic inputs or responses considered are either from independent events or have random peak phasing.
This similcrity involves a limited number of peaks of force or acceleration (not more than 5 exceeding 75 percent of the maximum, or not more than 10 exceeding 60 percent of the maximum), with approximately zero mean and a total duration of strong motion (i.e., exceeding 50 percent of the maximum) af 10 seconds or less.
Explanation.
Since earthquake motions in various directions produce respc..ses which are combined conservatively by the use of SRSS, the descriptions of dynamic or transient inputs are based on those applicable to carthquake
uctocer.5, mo g _ t.
motions.
The coefficient of correlation for these is less than 0.4, ard the pattern of peaks is based on Table 2 of Circular 672 of the USGS describing earthquake ground motions for use in the design of the Alaska Oil Pipeline.
The probability distribution for the responses to earth-quake motions is based on the concepts underlying U. S. NRC Regulatory Guide 1.60, where the standard deviation is 30 to 40 percent of the median value.
It has been proved some decades ago that modal responses to earthquake motions may be conservatively combined by SRSS methods with the same degree of conservatism as that of the motions.
If each of such responses is considered to be at the level of mean plus one standard deviation, the SRSS value is also at this level.
For the same reasons, responses from the three component directions of earthquake motions may also be conservatively combined by SRSS methods.
2.
Criterion.
When response time-histories are available for all' multiple dynamic loadings being combined, SRSS methods may be used for peak combined response when CDF calculations, using appropriate assu.ptions on the range of possible time lags between the response time-histories, show the following criteria are met:
There is estimated to be less than approximately a 50%
i a.
conditional probability that the actual peak combined response from these conservatively defined loadings exceeds approxi-mately the SRSS calculated peak response, and b.
There is estimated to be less than approximately a 15% con-ditional probability that the actual peak combined response exceeds approximately 1.2 timer the SRSS calculated peak response.
i'
/
7
@ #1$ n.. h, bf.hht.
Nathan M. Newmark Robert P. Kennedy
/
e
&LA 7 -- u BASIS FOR STATEMENT THAT SRSS COMBINED RESPONSE MAINTAINS SAME PROBABILITY OF NON-EXCEEDANCE AS EXIST FOR INDIVIDUAL RESPONSES ASSUMPTIONS STATISTICALLY UNCORRELATED RESPONSE COMPONENTS ARE OUTPUT OF LINEAR DYNAMIC SYSTEM SUBJECTED TO STOCHASTIC INPUT FORCING FUNCTION.
- MEAN RESPONSE IS ZERO CENTERED.
FOR LINEAR SYSTEMS THIS CONDITION IS AUTOMATICALLY SATISFIED IF MEAN INPUT IS ZERO CENTERED.
RESULTS VARIANCE OF TOTAL RESPONSE EQUALS VARIANCE OF INDIVIDUAL RESPONSES.
2 N
2
{
o
=
o T
i=1 i
MEAN EXTREME-VALUE RESPONSE OF EACH COMPONENT, Rj, (WITH 8
GIVEN PROBABILITY OF EXCEEDANCE) FOR A NARROWBAND SYSTEM (LINEAR STRUCTURAL SYSTEM WITH LESS THAN 20% DAMPING) IS PROPORTIONAL TO STANDARD DEVIATION.
{
I R
c
=
3 o,
- THEREFORE, r " \\! E R R
r-t BASIS OF CRITER10tl 1 e ftANY STUDIES HAVE S110WN THAT FOR LOW DAMPED LINEAR -
STRUCTURES SUBJECT TO EARTHOUAKE MOTION
= 50%
P RT 2 RTSRSS -
2 1.2 RT 5 15%
P RT SRSS.
e RESULTS OF THESE STUDIES CONSIDERED APPLICABLE FOR OTHER EARTHQUAKE LIKE TRANSIENT RESPONSES
- EARTHQUAKE LIKE RESPONSE REQUIRES:
- RESPONSE COMPONENTS FROM INDEPENDENT EVENTS OR RANDOM PHASING
- LIMITED NUMBER OF NEAR PEAK EXCURSIONS
- LIMITED DURATION
- APPROXIMATELY ZERO MEAN f
- -3 BASIS OF CRITERION 2
- CDF CALCULATIONS BASED UPON APPROPRI ATELY DEFINED PROBABILITY DENSITY FUNCTION FOR TIME PHASING OF RESPONSES CAN BE USED TO DIRECTLY SHOW:
1 50%
R 2
RT P
_T SRSS-s 15%
R 2 1.2 RT P
_T SRSS-
. CONSISTENT WITH EARTHQUAKE RESPONSES WHICH ARE BEING COMBINED SRSS CURRENTLY.
t Value at P(0.5)
No.
Cases P(SRSS)
P(1.1ISRSS)
P(1.2ISRSS)
SRSS Value 1.
RHR WETWELL-CBE +
0.43
- 1.0 1.0
- 1.0 SRV bub (Mc) 2.
MAIN STEAM-CBE +
0.46 0.84 0 98 1.005 SRV bdg (Ma) 3 MAIN STEAM-0BE +
0.44 0.98
- 1.0 1.004 SRV dis (Mb) 4.
MAIN STEAM-CBE +
0 38 0.98
- 1.0 1.007 SRV dis (Mc) 5.
CONTAINME!TI AT 0 35
- 1.0 1.0
- 1.0 STABILI2ER TRUSS EL SSEns + SRV all 6.
C0tiTAINMENT AT 0.23
- 1.0 1.e
- 1.0 STABILI2ER TRUSS EL SSEew + SRV all C. V, SUBRAMANIAN 11/28/78 e
/
( ~
- . s ' D~ "1. '
I t
a TYPICALIhRKII IbDIFICATIONS i
Ritic STIFFEt4ERS - STEEL CONTAINPENT (WPPSS-2) i-
- SRV DISCHARGE DEVICE PMSHEAD/0UENCHER (ALL)
PEROUTE SRV LINE (Au_)
- l REACTOR SUPPORT PEDESTAL (SO4E)
IhlN VEta BRACING (SOME)
'( I _
?
li EaulPtBU AND PIPING IN SC (Au.)
1
- DRYWEU_ STEEL FRN41NG NO SUPPORTS (SCME)
I I~
- PRif%RY SYSTEM no E0P Ecutprern (ALL) l_
- SC INSTRUMEtHATICI - POOL TEMPERATURE (SR) l_
_p
- SRV TESTS (SacE)
- Remove IhmCasR FLAtEES (SOPE)
=
w-LOADSLM%RY P00L SWEU. LmDS LotD OR PHENOMENA SPECIFICATICN BASIS SUBMERGED BOUTEMRY V.C. '
33PSIOVERPRESSURE CONSERVATIVE ON BASEMAT CALCULATION, 4T
- ASYBT'ETRIC VENT FLCw PSA'if%vMIN BOUNDINGCALCULATION m
BUBBLE PRESSURE ON P0otBOUtHRRY l
DIAPHRAGM UFWARD LOAD CORRELATIONOFTEST REGRESSION A'!ALYSIS RESutTS AB, AP, AV, 4T,EPRI i
. VS,VD
=
O
\\
til-3 LDAD SMTMRY IMPACT LOADS
~
LDAD OR PHENCNEM SPECIFICATION PASIS
- SmLL STRUCRJRES VELOCITY - PSN1 + 10%
PSTF, CONSERVATIVE IMPULSE - PSTF CALCULATIONS PttsE DURATION - FLAT Po0L ACC0arr FOR TARGET PSI ADDITIONAL 35% MARGIN
(
- GRATING STANDARD DRAG X CONSERVATIVE I
D(NAMIC/ IMPACT CALCULATION
~
MULTIPLIER
iv-y LaADSLM%RY LATEPAL LMDS ifAD OR PHENOMENA SPECIFICATION BASIS
- SINGLE VENT 8.8 KIP F 2 7HZ FOREIGN LICENSEE, 8,8 x y 74F 14 HZ t
4T DYNAMIC F
- 14 HZ ANALYSIS l
MULTIPLEVENTS PROBABILISTICAPPROACH FOREIGN LICENSEE RAf0Cri PAGNITUDE, DIRECTION SIfGlLTANEOUS CHUG INDEPEf0ENT _
0% AIR-hTA BASE w-w mu
gC LDADSUtnRY PcotBOUtHMRY CONDEttSATION LmDS LDAD OR PHENOMENA SPECIFICATION BASIS HIGH STEAM Fwx SINUS 0ImL PRESSURE FOREIGN LICENSEE,
> 32 ts/SEC/FT 4,4 PSI-PTP 4T,ikVIKEN 2-7Hz DATA CCNSERVATIVE SYMMETRIC Lmo APPLICATION CONSERVATIVE UNIFORMBELGVVENTS k
(
MEDIUM STEAM Flux SINUSOIDALPRESSURE ftREIGN LICENSEE, 8-]2 LB/SEC/FT 7.5 PSI-PTP 4T,thRVIKEN 2-7 Hz DATA CONSERVATIVE SYMERIC Lmo APPUCATION CONSERVATIVE UNIFORMBELOWVENTS k
_y Lmo SitT%RY CHUGGINGlCADS e
LMD OR PHENOMENA SPECIFICATION 3 ASIS taiGcuM51_
REPRESENTATIVE 4T 4T, FOREIGN LICENSEE,
~
4 8 LB/SEC/n PRESSURETRACE
&RVIKEN
~
UNIFORMBELOWVENTS 9;
SYNCHRONIZED CHUGS
(
(
UNIFORM l_MDING 44,8 PSI,-4.0 PSI 20-30Hz ASYMMETRICLO?
- G
+20 PSI,-24 PSI PERIPHERAL VARIATION OF AMPLITUDE PAXIMUM f%X/ MIN OPPOSED 9
e Y-