ML19269E026
| ML19269E026 | |
| Person / Time | |
|---|---|
| Site: | Maine Yankee |
| Issue date: | 05/16/1979 |
| From: | Harold Denton Office of Nuclear Reactor Regulation |
| To: | Gilinsky V, Hendrie J, Kennedy R NRC COMMISSION (OCM) |
| References | |
| NUDOCS 7906230101 | |
| Download: ML19269E026 (27) | |
Text
pp,e pa "'og UNITED STATES
- [?-
'h NUCLEAR REGULATORY COMMISSION
'g
- 7. -
WASHINGTON, D. C. 20555 us ' C 379 MEMORANDUM FOR: Chainnan Hendrie Commissioner Gilinsky Commissioner Kennedy Commissioner Bradford Commissioner Ahearne FROM:
Harold R. Denton, Director Office of Nuclear Reactor Regulation j
.A THRU:
J. e V ossi Executiv or
SUBJECT:
STAFF REQUIREMENTS FROM THE MAY 3,1979 COMMISSION BRIEFING ON THE STATUS OF MAINE YANKEE 1.
The staff completed its review of the final set of PSTRESS/SH0CK 1 to NUPIPE-SW code comparisons of May 3, 1979 and concluded that all requirements of the show cause order of liarch 13, 1979, has been met with respect tc the Maine Yankee facility.
The staff has also reviewed the licensee's response to IE Bulletin No.79-06B (Review of Operational Errors and System Misalignments Identified During the Three Mile Island Incident) and finds this response satisfactory.
2.
Secretary Chilk's memorandum to Lee V. Gossick dated May 4,1979, requested that staff provide responses to two questions raised during the May 3,1979 briefing. The Comission requested:
a.
A list of t.he seismic design margins built into the Maine Yankee facility that leads the staff to conclude that the Maine Yankee facility structures could absorb a.2g level of ground motion.
b.
An estimate of the time and effort that would be required to complete an evaluation of the Maine Yankee seismic design under the NRC's current seismic design criteria set forth in Regulatory Guide 1.60.
Enclosure (1), Discussion of Conservatisms in Maine Yankee Seismic Design, and Enclosure (2), Estimate of Manpower and Cost for Seismic Reanalysis, are the staff's responses to these requests.
2250 041 7 9 0 6 23 0/bl 4
The Commission 3.
On the basis of the review discussed in paragraph 1 above, the staff concluded that the requirements of the show cause order had been met and that continued shutdown of the facility was not required. On May 9, 1979, however, the licensee reported the earlier versions of SHOCK 1 known as SH0CK 0 :nay have computed piping natural frequencies incorrectly. The licensee is reviewing whether or not this latest information is significant and will report the results of its review. The staff will review the information submitted and inform the Ccmission if this information significantly effects our evaluation and our recommendation regarding restart of the facility.
4.
Also enclosed is a draft of the Order which I propose to issue, upon satisfactory resolution of the SH0CK 0 piping natural frequencies computation matter. The order, which will terminate the show cause order of March 13, 1979, which led to the current suspension of operation of the Maine Yankee facility, together with the NRR Safety Evaluation of the actions taken by the licensee in response to the show cause order, are forwarded at this time for Comission review and comment.
I will inform the Commission when the additional review of the SH0CK 0 piping natural frequency matter is complete and I am prepared to issue the order.
The order contains the staff's conclusion that the facility could withstand seismic events in excess of the current design basis seismic event, even though it is beyond the scope of matters addressed in the original show cause order. While the Office of the Executive Legal Director has no legal objection to the inclusion of this finding in the order, it points out that the discussion of seismic conservatism is legally separate from the issues addressed in the show cause order, the responses to that order, and the proposed order permitting resumption of opera on of Maine Yankee.
l G"
Harold R. Denton, Director Office of Nuclear Reactor Regulation 1.
Discussion of Conservatisms in Maine Yankee Seismic Design 2.
Estimate of Manpower and Cost for Seismic Reanalysis 3.
Proposed Order Tenninating Suspension of 2250 042 Operation of Maine Yankee
4 DISCUSSION OF CONSERVATISMS IN MAINE YANKEE'S SEISMIC DESIGN _
While increasing the SSE seismic input from 0.lg with a Housner spectrum to
~
between 0.13 to 0.29 with a Regulatory Guide 1.60 spectrum may seem to be a large percentage increase in seismic input, the inherent resistance of a facility properly designed to 0.19 should, in general, provide adequate resistance to the relatively low seismic input of be: ween 0.13 to 0.2.
For example, 9
is much less severe than going the impact of increasing from 0.19 to 0.29 from 0.259 to 0.5g.
Tnis is because nuclear plant designs are based on As an various combinations, of loads with seismic loads as only one part.
example, of the 85 piping runs analyzed at Maine Yankee, all of the peak stress points would be less than 50% of ultimate strength even if the seismic stresses are doubled from the 0.19 level. Only six of the runs would have peak stresses greater than current allowable stress limits, even though eleven runs would have peak stresses exceeding the more conservative criteria in the FSAR. Of the six runs with peak stresses over current allowable stress limits, it is likely that these stresses would be less than the actual material yield stress.
Seismic design of nuclear power requires interaction between these principal (1) definition of the seismic hazard, in terms of intensity and endeavors:
characteristics of shaking, and (2) design of structures, systems and components to resist the defined seismic shaking.
The definition of seismic hazard invloves consideration of the geology and seis-mology of the region, observed ground motion, and observed effects of earthquakes.
The information available for historic records, measurements recorded in more recent years, and insights that can be gained from analyses and damage assessment 2250 043
following earthquakes have been synthesized to arrive at the engineering methods we use to define the seismic hazards for nuclear power plants, dams and other public structures.
The seismic input, once defined, is used in a mathematical process to determine how the structure would vibrate in response to the seismic shaking.
Throughout this process very complex natural phenomena and the response of complex structures and equipment are idealized so that the principles of applied mechanics and mathematics can be employed to determine the response of each of the major portions of the structures and equipment. To compensate for these idealizations the engineering practices involved in the seismic design for nuclear power plants establish a conservative design quantity at each stage in the analytical process (see the attached list of conservatisms).
The final design, resulting from compounding of the conservatisms in each step, is therefore also conservative.
For plants of the Maine Yankee vintage, conservatisms in the seismic analysis and design for structures, systems and components are generally found in the following areas:
(1) Elastic dynamic analyses are performed using conservatively low damping values.
(2) Multiple-directional seismic input, with each horizontal component having equal intensity, is considered in design of plants. Actual earthquakes are typically stronger in one direction.
2250 044
-3 (3) The OBE is selected at one-half of the SSE and controls the design in many cases, rather than the SSE, due to the substantially lower allowable stresses for the OBE.
(4) Loading combinations consider other loadings (dead weight, live loads, pressure loads, etc.) in addition to the seismic loadings.
Seismic loading is therefore only a part of the total loading and in fact, loadings other than seismic may govern designs.
A sizable increase in seismic stresses may be only a small addition to the total stresses.
(5)
In the design of structures and equipment, it is convenient to assure that all elements of the structure or equipment are designed to stress levels well below the actual strength of the materials so that any permanent deformation is very small. This approach obviates the need for complex and costly inelastic analyses.
Inelastic behavior would significantly reduce structural response prior to failure.
(6)
Stress limits, whether clastic or inelastic, are based upon material behavior under static loading conditions.
Since dynamic loads contain a limited amount of energy, the margin (between the stress limits and failure) under dynamic loads is grenter than under static loads if clastically calculated peak response is compared to the stress limits with strain rate effects neglected.
2250 045
(7) The design of the structural elements is such that tnetr capacity usually In Maine exceeds the seismic requirements called for by the analyses.
Yankee, orthogonally spaced reinforcing steel was used in the containment wall with additional diagonal reinforcement at large penetrations. Much of the actual structural design is controlled by the availability of standard structural members, such as beams and piping sections, so that larger sizes than are needed are oftcn used.
In-situ (8)
Engineering codes specify " code minimum strength" for materials.
strengths are usually higher.
Additional conservatisms for major mechanical components and piping can be found in:
When the floor response spectra are developed for the design of components (1) located at different locations in the structure, the peaks in the individual floor response spectra are broadened in order to predict conservative responses.
Where the system has multiple supports, maxirrum response spectra is usually (2) applied to all support points.
When calculating the seismic loads for components, conservatively established (3) values are applied several times (first, to major structures, then to the intermediate structures and finally the equipment themselves).
Even identically designed redundant systems may not always experience (4) similar seismic excitation due to different mounting locations, with different structural filtering effects.
Thus, a loss of a redundant component may not mean a loss of function for the system.
50 046
The end result of the conservatisms employed in the analyses, followed by the conservatisms resulting from standard design practices, is structures and components with seismic capability well in excess of the established design goal.
This is the reason that the record is replete _ with cases where well-engineered structures, even those for which no specific seismic design standard was invo' sed, A number of have withstood major carthquakes while remaining fully functional.
The Esso re-plants of various kinds have been subjected to large carthquakes.
finery in Managua, Nicaragua is a good example.
Another example is the pump stations in the Exxon pipeline in Italy, subjected to the Friuli earthquakes.
These are structures.that were designed by ordinary codes, with perhaps the seismic design coefficient of the order of.05 to.08.
The carthquakes that 9
occurred had accelerations that were measured of the order of.35g in Managua and perhaps more than that in Friuli.
The Esso refinery was able to continue operating with r.o damage to any of the equipment while the pump stations on the Exxon pipeline were able to continue operating without damage to the equipment.
For these reasons and taking into account the use of orthogonally spaced reinforcing steel in the containment wall, the staff judgment is that the major structural components of the Maine Yankee facility will likely remain functional even for an increastd range of seismic input of from 0.13 to 0.29 level, it is unlikely that the seismic event would initiate Even at the 0.29 For minor mechanical and electrical equipment, where the a serious accident.
fragility.is likely lower, loss of function is not expected to be sufficient to prevent plant shutdown when all plant systems and available corrective actions are considered.
2250 047
6-The likelihood of the SSE is presently judged to be on the order of 10-3 or 10-4 per year for the 0.13 to 0.2g range, decreasing with the higher values.
The confidence in the judgment that major structural compo'nents will likely remain functional increases at the lower SSE range.
The NRC will be further reviewing the subject of seismic design capability of That effort will assist the all operating reactors within the next few months.
staff in determining whether and when additional seismic re-evaluation is needed at any operating facility, including Maine Yankee.
2250 048 e
~
CONSERVATISMS IN SEISMIC DESIGN Seismic Design for Ground Motion I.
Enveloping response spectra and time histories Conservative OBE (usually controls design)
Seismic Analysis and Design Method II.
Structures, systems and components a.
Elastic dynamic analysis (inelastic behavior can significantly reduce response spectra)
Damping values Multi-directional earthquakesLoading combinations (seismic o Additional conservatisms for piping and major components b.
Peak widening of floor response spectra System Redundancy Generic Qualification for Many Plants Use of maximum and widened response spectra for multiple supported systems Multiple applications of damping values Structural and mechanical resistance factors III.
Allowable stress from Code Dynar.ic resistance of materials 26 day concrete strength Ductility to failure Minor attachments absorb energy Redundancy in structural elements Use of standard size pipe and equipment Quality Assurance IV. Seismic Experience to Date Inherent resistance shown for large industrial facilities Nuclear plant resistance shovin in Japan Other loads (wind and pressure) influence design 2250 049
ESTIMATE OF MANPOWER AND COST FDR SEISMlfRIMIATT'ill An estimate of the amount of resources needed to perfonn a seismic reanalysis for a nuclear power plant is directly depehdent on the extent of the reanalysis, i.e., the requirements for the reevaluation.
A first approach would be to have the licensee reanalyze all safety-related features of the plant and have the staff review the reanalysis.
The staff estimates that this would take up to 21/2 years or even longer.
The range of time depends primarily on the actual scope of items to be analyzed, the number of modifications to be designed, efficiencies, and resource levels.
For example, for Diablo Canyon, the entire plant was essentially analyzed in 21/2 years after the basic criteria was set, including about 10 manycars of staff review, under considerable schedule pressure to get the plant started.
Other examples of reanalyses, such as Humboldt Bay, Unit 3, and San Onofre, Unit 1, have taken much longer without the same schedule pressure.
A shorter time, perhaps one year, can be considered an optimistic schedule for a more abbreviated approach similar to that being done by the staff for certain of the SEP plants.
It should be noted, however, that the result of the staff's seismic review of SEP plants may result in the requirccont for a more detailed seismic reanalysis by the licensee.
Such an approach could be similar to the SEP review of Dresden 2.
There the staff's consultants are reviewing the existing design over a period of several months and will determine what areas need further detailed reanalysis.
The detailed analyses could then take approximately 1 year.
For Dresden 2, wnere the previous design bases is not considerably changed by present requirements, this approach will reduce the amount of detailed reanalysis. However, this may not follow for other plants, where the seismic input might be increased more significantly.
Based on these limited experiences, several additional months following reanalysis could be required to complete any needed physical modifications (i.e., if there are numerous modifications but installation begins early in the process as the needs are identified.)
The staff's cost estimates are approximate, but it appears that the licensee's total analysis cost would be between $5 million and $15 million depending on the approach used the staff's resources could vary from a couple of manyears of effort to 10-15 manycars of effort.
2250 050
n.
~
7590-01 UNITED STATES OF AMERICA NUCLEARREGULATORYCOMNISSION
~
In the Matter of MAINE YANKEE ATOMIC POWER COMPANY
)
Docket No. 50-309 (Maine Yankee Atomic Power Station)
)
TERMINATION OF ORDER TO SHOW CAUSE_
I.
The Maine Yankee Atomic Power Company (the licensee) is the holder of Facility Operating License No. CPR-36 which authorizes operation of the Maine Yankee Atomic Power Station (the facility) at power levels up to 2630 megawatts thennal (rated power). The facility, which is located at the licensee's site in Lincoln County, Maine, is a pressurized water reactor used for the commercial generation of electricity.
II.
Because certain safety related piping systems at the facility had been analyzed with a computer code which incorrectly summed earthquake loads algebraically, the potential existed for compromising the basic defense in depth provided by redundant s'afety systems in the event of an earthquake.
Therefore, by Order of the Director of Nuclear Reactor Regulation (the Director) for the Nuclear Regulatory Commission (NRC), dated March 13, 1979 (44 FR 16506, March 19,1979), the licensee was ordered to show cause:
2250 051
7590-01
. (1) Why the licensee should not reanalyze the facility piping systems for seismic loa _ds on all potentially affected safety systems using an appropriate piping analysis computer code which does not combine loads algebraically; (2) Why the licensee should not make any modifications to the facility piping systems indicated by such reanalysis to be necessary; and (3), Why facility operation should not be suspended pending such reanalysis and completion of any required modifications.
In view of the importance to safety of this matter, the Order was made innediately effective and the facility was required to be placed in the cold shutdown condition and remain in that mode until further Order of the Commission.
III.
The facility is currently in the cold shutdowa condition.
Pursuant to the March 13, 1979 Order, the licensee filed a written answer to the Order by letter dated April 2, 1979.
In this response the licensee stated that it has reanalyzed all potentially affected safety systems for seismic loads using an appropriate method which does not sum loads algebraically and these reanalyses indicate that two piping restraints need to be modified to account for base plate flexibility.
These modifications have been completed.
Technical 2250 052
T 7590-01
. support for these conclusions was provided in -the " Interim Report by Stone & Webster, April 1,1979", " Containment Spray Piping Analysis of Pipe Supports 11-51 and 11-53, April 2,1979", and the licensee's submittals dated April 3, 12, 13, 19, 27 and May 2, 4 and 5, 1979.
Based on the above, the licensee concludes there is no basis for continued suspension of facility operation as contemplatea vy the Order, and proposes:
(1) That the Director modify or rescind so much of his Order of !! arch 13, 1979, as requires the continued shutdown of the facility.
(2) That the Director grant to the licensee such other and further relief as is proper in the circumstances.
The tiRC staff has reviewed the licensee's submittals. This review included an evaluation of the codes which compute pipe stresses resulting from the facility's response to an earthquake.
The means by which piping responses are combined in the codes that are currently a basis for the facility de[ign are surwarized below:
flVPIPE-SW This code combines intramodal* responses by the square root of the sum of the squares (SRSS) and combines intermodal* responses by SRSS or absolute suin for closely spaced nodes.
- Modes are defined as dynamic piping deflections at a given frequency.
Intramodal responses are the components of force, moment and deflection within a mode.
Intermodal responses are the components of force, moment and deflection for all modes.
7590-01 -
PSTRESS/SH0CK 3 In this code the intramodal responses are calculated by adding the absolute value of the responses due to the vertical earthquake component to the root-mean-square of the responses due to the two horizontal carthluake components. The intermodal compenents are calculated by the root-mean-square method.
PSTRESS/, SHOCK 1 One of four versions of this code was reviewed.
In this version the largest nodal response is added (absolute sum) to the root-mean-square value of all other modal responses.
Intranodal re-sponses due to multi-directional earthquake excitation were not calculated since the code only produced responses parallel to a given earthquake component excitation.
Because this code is not equivalent to current practice, the NRC staff requested that the licensee denonstrate the conservatism of pipe stress as deternined by this code.
This was done by reanalysis of certain piping systems using currently acceptable methods.
STRUDL-SHAKE This code conbines intramodal responses by absolute sum and the internodal responses by SRSS.
2250 054
14 :.
I The NRC staff has determined that an algebraic summation of responses was not incorporated into any of the.above listed codes.
The NRC staff has further concluded that these codes provide an acceptable basis for the facility piping design.
The "RC staff reviewed the inherent seismic conservatistas in the facility design.
Methods of analysis, material properties, actual earthquake characteristics, construction practices and actual seismic experience were considered.
The NRC staff has concluded that the facility could withstand earthquake ground motion in excess of that to which the facility was originally designed.
The NRC will be further reviewing the subject of seismic design capability vf all operating reactors including Maine Yankee, within the next few months.
That effort will assist the staff in determining wheth.er additional seismic re-cvaluation is needed at any operating facility.
Modifications on two piping supports (H-51, H-53) for the containment spray system were determined to be necessary as a result of the re-
[
analysis.
The modifications consisted of welding two stif fners to each support base plate to reduce the base plate flexibility.
The modifications were completed in accordance with the Yankee Operational Quality Assurance Program (YOQAD-1 A) and are acceptable.
21250 055 e-
7590-01
. Based on the NRC staff's Safety Evaluation dated May
, 1979, the staff finds that, in accordance wi~th the Order of March 13, 1979, the licensee has reanalyzed all potentially affected safety systems using an appropriate piping analysis which does not combine loads algebraically and has made those modifications to the facility piping systems indicated by such reanalysis to be necessary.
The licensee's answer to the Order did not request a hearing.
The New Hampshire. Legislative Utility Consumers' Council petitioned on April 2,1979, to be permitted to intervene in any proceeding which might arise from the Show Cause Order, but did not request a hearing.
No other person requested a hearing.
IV.
Accordingly, pursuant to the Atomic Energy Act of 1954, as amended, and the Commission's Rules and Rcgulations in 10 CFR Parts 2 and 50, IT IS DETERMINED TilAT:
The public health, interest or safety does not require the continued shutdown of the facility, AfiD IT IS IIEREBY ORDEREL THAT:
Effective this date the Order to Show Cause of March 13,1979, and the proceeding thereon are terminated.
FOR Tile fiUCLEAR RLGULATORY COPb11SS10!i liarold R. Denton, Director Office of Nuclear Reactor Regulation Dated at Bethesda, Maryland this day of May 1979.
22150 056
M UNITED STATES j
= > "
M NUCLEAR REGULATORY COMMISSION
)1 i WASHINGT ON, D. C. 20555 e s y' /a SAFETY EVALUATION BY THE OFFICE OF NUCLEAR, REACTOR REGULATION MAINE YANKEE ATOMIC POWER COMPANY MAINE YANKEE ATOMIC POWER STATION DOCKET NO. 50-309 Introduction On March 13, 1979 the Commission issued an Order to Show Cause to Maine Yankee Atomic Power Company (licensee) requiring that Maine Yankee (facility) be placed in cold shutdown and the licensee show cause:
(1) Why the licensee should not reanalyze the facility piping systems for seismic loads on all potentially
.affected safety systems using an appropriate piping analysis computer code which does not combine loads algebraically; (2) Why the licensee should not make any modifications to the facility piping systems indicated by such reanalysis to be necessary; and (3) Why facility operation should not be suspended pending such reanalysis and completion of any required modifications.
The licensee's response to the Order, dated April 2, 1979, stated tha t a 11 affected safety systems have been reanalyzed using an appropriate piping analysis method, and that no modifications are necessary as a result of these reanalyses. Therefore, the licensee requested that the Ordcr be modified or rescinded such that the facility could be restarted.
In support of this request the licensee provided information by letters dated April 2, 3, 12, 13, 19, 27 and May 2, 4 and 5,1979.
In the letter of April 13, the licensee indicated that two piping resfraints needed to be modified as a result of the reanalyses to account for base plate flexibility. On April 19, the licensee reported that these modifications had been completed.
Discussion The Stone and Webster (S&W) PSTRESS/ SHOCK 2 computer code for pipe stress analyses sums earthquake loadings algebraically and is unacceptable for reasons set forth in the March 13, 1979 Order to Show Cause.
This code was used in the seismic analyses of certain safety and nonsafety related systems at the facility.
The licensee has identified the seismically analyzed (Seismic Category I) systems at the facility including those analyzed with Sil0CK 2.
It has also identified the other methods of seismic analysis used for other Seismic Category I systems.
Furthermore, the licensee has summarized the results of the reanalyses of Sit 0CK 2 safety systems and has provided support for the acceptability of the analysis methods used on the remaining Seismic Category I systems.
2250 05,7
_2 We have evaluated the facility's safety related systems, the results of seismic reanalysis,and the methods of pipe stress analysis currently in effect for the facility.
Eval ua tion _
1.
Systems The licensee has stated that the risponse to Question 1.3 of the Maine Yankee Final Safety Analysis Report (FSAR), submitted February 9,1971, is the complete list of structures, systems and components that were designed to the Seismic Category I requirements.
Verification has also been provided by the licensee that the Seismic Category I piping systems identified in respanse to Question 1.3 of the Maine Yankee FSAR include all of the piping systems required to assure:
(a)
The integrity of the reactor coolant pressure boundary; (b)
The capability to shutdown the reactor and maintain it in a safe shutdown condition; and (c)
The capability to prevent or mitigate the consequences of accidents which could result in potential offsite exposures comparable to the guideline exposure of 10 CFR Part 100.
Portions of the foilowing systems were identified by the licensee as having been either analyzed with SHOCK 2 or analyzed by static seismic methods which were verified by SHOCK 2.
High Pressure Safety Injection Residual Heat Removal Containment Spray Low Pressure Safety Injection Primary Component Cooling Water Stear Generator Feedwater Chemical and Volume Control Primary Vents and Drains Waste Gas Disposal Boron Recovery Fuel Fool Cooling Fire Protection Auxiliary Steam Auxiliary Condensate Return High Pressure Drains (Secondary)
A total of 39 SHOCK 2 analyses (Computer runs) v.ere performed.
Piping associated with these analyses and the methods cf rear-lysis are identified in the Enclosure to this rifety Evaluation (SE).
2250 058
s.
. Nineteen of these 39 analyses have been identified by the licensee pcrtaining to safety related piping.
He have reviewed the asinformation submitted and agree with the licensee's identification of piping which is safety related.
The licensee has completed the reanalysis of all 39 Sii0CK 2 analyses.
2.
Verification of Analysis l'ethods We have reviewed the acceptability of the analytic methods which are currently a basis for the facility piping design. The licensee has identified the following computer codes / analysis rnthods as applicable:
PSTRESS/SliOCK 1 (4 Versions - Initial 3 Versions sometimes referred to as SHOCK 0)
STRUDL
- SHAKE (Combustion Engineering)
Static Analysis Methods PSTRESS/ SHOCK 3 fiUPIPE - SW PSTRESS/ SHOCK 1 f the PSTRESS/SHgCK 1 The licensee has identified four (4) versions o computer code.
Documentation on only the last version of this code was available for our review; however, earlier versions are expected to use similar and definitely no more sophisticated methods of analyzing seismic loads.
The licensee has stated that this version of SHOCK l combines the intermodal responses by the so-called " Navy Method". This consists in taking the largest absolute modal response and adding the root-mean-square value of all otiier modal responses.
Intramodal responses due to multi-directional earthquake excitation were not calculated since the code only produced responses parallel to a given earthquake component excitation (i.e., the responses were considered uncoupled).
A review of the code listing has confirmed these statements.
2250 059
_/
. Some safety systems of the facility were analyzed with each of the four versions of the SH0CK 1 Code.
Because this computer code only considers one direction earthquaEe excitation, it is not considered equivalent to current analysis techniques.
A comparison of the results of each of the four (4) versions of PSTRESS/SH0CK 1 and the fiUPIPE Code was conducted by the licensee using " typical" piping problems.
The problems consist of different size piping, elbows, tecs and reducers.
The licensee reported that the general stress distribution of both codes was similar and PSTRESS/ SHOCK 1 gave comparable results. The licensee concluded that although the PSTRESS/ SHOCK 1 is not equivalent to current practice, it is suitably conservative to insure that the piping systems meet the allowable stress levels.
We have' reviewed the piping configuration and results of the comparative analyses of fiUPIPE and each version of the SHOCK 1 code.
We have determined that the problems analyzed produce respresentative comparisons. We have also determined that although SHOCK 1 is not equivalent to current practice, the resulting stresses are at least consistent with the results as obtained from f;UPIPE and in many cases are conservative.
In addition the code comparison did not take credit for the alternative application of the " Robinson Fix" (i.e., adjusting the response spectra peak instead of increasing all analysis results) which would provide additional conservatism to the SHOCK 1 stresses in this com-parison. The " Robinson Fix" was described in Amendment 35 to the flaine Yankee FSAR.
Therefore, we conclude adequate assurance has been provided that systems analyzed with SHOCK 1 will withstand the design basis earthquake.
Although this assurance regarding SH0CK 1 systems has been provided, the licensee has, by letter dated !<ay 2, 1979, stated its intent to reanalyze all SHOCK 1 systems using a method of analysis verified against current criteria.
By doing this,the licensee will not only upgrade the facility's seismic design analysis but also greatly improve the records and accuracy of numerical results associated with the facility's seismic design analysis for possible future The licensee has also stated that following its review of use.
the details of this program and by June 1,1979 a schedule for completion of this reanalysis veill be proposed.
We agree with the licensee that this reanalysis is appropriate.
2250 060
. STRUDL
- SHAKE The licensee has provided the fol_ lowing description of the analysis technique used by Combustion Engineering (STRUDL
- SHAKE Code):
"The dynamic seismic analysis of the reactor coolant system main loop and pressurizer surge line piping was perfonned utilizing 3 dimensional mathematical models subjected to unidirectional support rotion response spectra. The six components of force or moment at a particular piping location were determined separately for each significant mode of response for a single direction of excitation.
The separate modal responses for each component of force or n,cment were then combined on a root-sum-square basis to define the total force or moment response to a single d, rection of excitation.
The loads due to each horizontal i
carthquake were added, manually, to the loads due to the vertical earthquake by the absolute sum method.
The larger of the two loads thus calculated was employed in the stress analysis of the piping system. "
We have reviewed the analysis technique of Combustion Engineer-ing.
The procedures are in compliance with the plant FSAR and conservatively combine (absolute sum) both the spatial com-ponents from each of two independent earthquake directions and the contribution of each mode (SRSS).
We find this technique acceptabl e.
Static Analysis Some of the safety related systems at the facility were analyzed using static analyses techniques. The licensee submitted documentation (letter dated April 12,1979) detailing the basis for static analysis technique use in the design.
Generally piping 6 inches in diameter and smaller was designed using the static irethods unless the criteria.for support placement could not be met, then a more rigorous dynamic analysis was performed.
Some piping larger than 6 inches in diamter was analyzed using the static methods if the geometry and support configurations were sufficiently sinple to make the static analysis methods practical.
The major constraint on applying static methods to larger piping was one of economics in that a dynamic analysis typically would result in fewer restraints at a Lore optimum spacing and supports for larger piping were sufficiently more costly to warrant less conservative but more expensive analysis techniques.
2L250 061
n
. The analysis technique used at the facility is outlined in Amendment No. 35 to the FSAR and the procedure-vias submitted in detail in the report, "Non-dynamic Seismic Analysis of Piping and 12, 1979.
Supports by Stone & Webster at Ma~ine Yankee" submitted April The procedure states that the piping frequencies will be designed to be a minimum of 1.5 times the peak resonant frequency of the amplified response spectra by locating seismic supports at appropriate span lengths.
Orthogonal responses will be decoupled by including supports at elbows, tees and concentrated rcasses.
The piping systems were designed considering a horizontal static load of (1.3) X (22 X peak ground acceleration) acting concurrently static load caual to two-thirds the horizontal with a vertical The method of equivalent analysis outlined in this procedure valu.
e has been reviewed cgainst the NRC's Standard Review Plan 3.7.2 and is acceptable.
PSTRESS/ SHOCK 3 The licensee has stated that in this code the intramodal responses are calculated by adding the absolute value of the responses due to the vertical earthquake component to the root-mean-square of the responses due to the two horizontal earthquake components. The intermodal com-A review of ponents are calculated by the root-mean-square method.
the code listing has confirmed these statcents.
A confirmatory analysis was performed by an NRC consultant, Brookhaven National Laboratory (BNL), of a typical piping design problem in the Maine A problem (no. 803) has been submitted by S!M Yankee plant.
together with the corresponding solution obtainod by using PSTRESS/
SHOCK 3.
This problem has been analyzed by BNL using a different code (EPIPE), and the results have been submitted to the NRC staff.
A comparison of the solutions indicat'es that various quantities of interest such as frequencies, displaccments, forces, and stresses, a;mear to differ by not more than 10";.vhich is within the accuracy of the In addition, hand calculations were perforced with the analyses.
PSIRI:SS/ SHOCK 3 results as a, check on the nodal cu3ination methods.
We find that the SM! results have been cdequately confimed by BNL and are therefore acceptable.
NUPIPE - SW The licensee has stated that this code calculates intramodal and interudal responses according to the provisions in Rcquiatory Guide 1.92.
A review of the code listing by the staf f has confirmed this Additional documentation has also been submitted to be the case.
by the originators of this code (Nuclear Services Corporation) providing detailed information on the methods of modal combinations.
fhis information has been revicued and also provides reasonable confirmation of the statements made by the licensee.
A confirmatory analysis has also been performed by our consultants on the piping A comparison of the solutions again indicates problem listed above.
that the various quantities of interest listed above again dif fer by Therefore, the use of this code is acceptab)e.
not more than 10%.
2250 06
3.
Reanalysis Methods and Results The safety related piping systems at the Maine Yankee nuclear plant have been reviewed to detennine the method of an'alyses.
Nineteen (19) computer stress probicos of safety related piping have been identified where the analysis used an algebraic intramodal sumation of responses to carthquake loadings. The problems where an algebraic intramodal response combination technique was used in the design have been reevaluated using the criteria in the FSAR.
The rcevaluation included a static analysis technique, and a dynamic computer analysis using either the PSTRESS/SH0CVs 3 or NUPIPE programs.
A static analysis techniaue was employed for reanalysis of some lines 6 inches in d tuneter and maller. The static design procedure is outlined in a report titied "Non-dynamic Seismic Analysis of piping and Supports by Stone & Webster at Maine Yankee" submitted April 12, 1979. The acceptability of this procedure has been discussed in Section 2 of this SE.
The dynamic analysis technique incorporated a lumped mass response spectra The floor modal analysis using the PSTRESS/ SHOCK 3 or NUPIPE programs.
response data used in the reanalysis included the " Robinson Fix" criteria.
The " Robinson Fix" criteria required the peak resonant frequency acceleration values to be a minimum of (22) x (peak ground acceleration) and the peaks to be broadened by + 10% of resonant frequency. The piping systems were modeled as three dimensional lumped mass systems which included conside of eccentric masses at valves and appropriate flexibility and cra tions stress intensification factors. The dynamic analysis procedures meet the criteria specified in the plant FSAR and are acceptable.
The piping support designs for affected system piping were inspected by the licensee to v o ify the "as built" configuration. As noted in NRC Inspection Report 79-05 issued. April 12, 1979, differences were found to The exist between the "as built" configuration and the support drawings.
~
differences noted resulted from the use of draw;ngs which had not been Subsequently the licensee has updated to include installation changes.
verified that updated drawings which do reflect the supports as installed, were used in the support design calculations.
The support designs were reevaluated in cases where the original support design loads were exceeded as a result of piping reanalysis. The support reevaluation included the consideration of local stresses at regions of discontinuity and base plate flexibility considerations.
Modification 2250 063
of two supports was determined to be necessary to acccunt for base plate flexibility.
These modifications consist of adding a stiffener to the base plate of each hanger and have been completed.
Loads on attached equipment nozzels were also checked and verified to be either below the initial allowable _ values or verified by the equip-ment manufacturers to be acceptable.
The design and analysis of the supports and attached equipment are in accordance with the criteria specified in the plant FSAR.
The pipe break criteria for Maine Yankee wre reviewed and determined not to be altered by this reanalysis.
Pipe break considerations were required for High Energy Lines outside of the containment structure and break loca tions were detennined by inspection and their proximity to safety related systems.
The pipe break considerations are outlincd in a report titled " Supplementary Report on Effects of a Postulated Break in a High Energy Piping System Outside the Containment" dated September 1973.
The pipi Ag systans and supports tere designed to the allcuable limits of ANSI B31.1 for the gross properties and to the limits of ANSI B31.7 Appendix F for local stress considerations per the FSAR criteria.
The safety related piping systems, supports and attached equipment, where the original analysis used an algebraic intramodal response sr ution technique, have been reanalyzed with acceptable nethods which do not use an algebraic intra:;odal respo:.se technique.
The pro-cedures used in the reanalyses and their results have been reviewed cgainst the criteria in the plant FSAR and found acceptable.
C gclusion_
The licensee has demonstrated that pSTRESS/ SHOCK 2 is the only method of analysis used for the facility's safety related systems which combines seismic loads algebraically.
Safety related systems analyzed with Shock 2 have been reanalyzed with an acceptable dynamic code or with tatic analysis techniques as permitted by the FSAR criteria. The resul ts of these reanalyses have shown that the subject sys tws sill will.utand the design basis earthquake.
The reevaluation of supports performed by the licensee for the subject piping considered base plate flexibility. As a result stif feners were added on two supports in the contair.mnt spray system.
We revia ed the acceptability of the analysis techniques which are currently a basis for the facility's piping design. We have determined that the application of these techniques, at Maine Yankee, assures that safety rela ted systems can withstand the design basis earthquake and that there is reanonable assurance that the facility can operate withnut endangering the health and safety of the public.
2250 064
_9_
Based on the cbove, we conclude that the rcquirements of the Order have been mat for liaine Yankee and therefore the Order end its restriction on facility operation should be terminated, Dated:
2250 065 O
1
,1lAINE YA?iMEE SHOCK II REANALYSIS - !! ASTER LISTING ETICl O S u!'O TP 3 RCSCLUTIC:i ITE.i. A: SAL.C A'.,:.PRTY.
DESCRIPTICN LI:4E tiUl:2ERS
. ItSM 3. TM #~
. SA..
4.................____
1. 16 A 10. 1. CHA OING PUMP RCHST SUCTICM
. 10"-CH.1 12SA11.
31B 14
. R,ERUM - 5HCCH III 2. 14S. IS. 1. CHARCINC PUMP RCH3T SUCTIC!!
. 10"-CH.2
. 10SA11.
313 '
14
. RERUM - SHOCK III 3. 003 21.
- 1. RESIOUAL HE AT R E!:CVAL SUCTICM.12".RC.29 / 14".RH_1.1C73
. 30AL32A. 16
. RERUN - SHCCH III 4
790.
28 1. CCNTA!!UtENT SM AY L LCH HEAD. IC".CS.11112
. 1103
.32A 16117. RE;UM - SHCCH III (CRIGIHAL HA3 PART SHCCK I)
. SAFETY Iri)ECTICM SUCTION 14".CS.13114140 14"-CS.15116L17
. 16"-R H -314 5. 720. 25. 1. PRIltARY CCHP CCCLI!iO SUCTICN 20"-PCC-17
. 117E 34A 20
. RERUM - SHCOM III 16"-PCC-10119 6
70. 17. 1. RHR SUCTICM SAFETY VALVE PPG 4" nH- !.35
. 146A. 30A132A. 13116. RERUM. SHCCM III 3"-CRL-19?i200
.........._+.._......+............________........._,__....___.....................+_......+.......+.....___....................
7. 126. 12. 2. RCHST HEATER RE R N N
4".PL-22 13CA 32A 6
. NCH HAM 3 CALC
- 0. 127. 12 2. RCHST HEATER INLET LINE 2"-PL.21 13CA 32A 6
. NEN HAND CALC i
9. 134. 12. 2. DEMItt HTR STCR TK HTR RETURN 3"-HCPR.6 13CA 12A 4
. NEW HAND CALC 10. LN :. 12. 2. CEllIN HTR STCR TH HTR IttLET 1.5"-HCPR-5 12A 316. NEH H MO CALC 11. 11. 15. 2. NCRMAL CHARCING LINE 3". Cit-41
. 132A 31A 9L14. RERUM - HUPIPE 12. 140. 17. 2 SEAL HATER INJECTICM HEACER 3"-CH-56
. 12SA 31A
. 10114. RERUM - NUPIPE N 13. LN :. 28. 2. PCCH TO PRESSU1IZER CUENCH 1.5".PCC 161 34A 20
. NCH HAND CALC N
. TANK CCCLER LT1 T.A 20
. NEH HAN3 CALC c>
14. 77. 31. 2. PCD; READER TO CCCLANT DEGAS. 3".PCC.34 IFIER CCCLER & CONDENSER 15. LM 4. 36. 2. PCCH HEACER TO RAD HASTE HX 6".PCC r&7
. 117J 34A 20
. KERUM - NUPIPE g
@ 16. L?!
. 36. 2. PCCH TO HASTE EVAP CONCENSER 3".PCC.56
. 117J 34A 20
. RERUM. trJPIPE 34A 20
. liCH HAND CALC 17. LN n. 36. 2. PCCil TRCM CHG PP SEAL LH3 CLR.1"-PCC-307 sjj 10. LM :. 36 2. PCCil TO liASTC CVAP OISTIL CLR 1"-PCC.50
- 34A, 20
. !(Cil HAND CALC s
,_, 19. LM a. 36 2. PCCil FRCM CCRCN RECOVERY EVAP 1.S"-PCC-117 34A 20
. RERUM - HUPIPC
....._........_e.........._.....o.__...e._.._.......eese.._.es._...e..*.e.eem........en__es.....s...._me....e........_.._........ee..sy.m..e
e e
290 0322 8
0 f
s i
e I
4 b
8 e
8 8
8 u
8 3
I 7
Z 8
3 0
c, e>
s.
a.
O 8
a e
cc 8
u i,.
s t
u z
e a,
8 td W
bJ LJ W
kJ LJ RJ IJ U
LJ B
O
. ~
r Ex CL rL f.
8 P
LJ LJ kJ L>
LJ kJ h h h h CL.
U.
H H
H H
H g ('s 8
O ' CL r.
C.
C.
L-O f-0 e
ru F'
F F F rt rt t
tt r-c, i
s H
H H
H H
_J H
e, e,4 e.
i U
> a :>
U p
i
~
1
?
? F1 p:
U p
i
- )
e a
s e:
rs c.
r.
rt rt r
i Z
k Z
Z k
a O
e :J, e
8 r;
eO i
i k i Z
Z Z
s
~ *- ~ --
O g
0 8
I I
I 3'
4 8
8 8
8 0
oc O
O 9
O L1 I
f) e sa B
Z l
I I
I B
Z l
0
~ ?
?>
,Y
- f. > f f>
f 7
A H
'a
~ **"
y' 8
f >
2 a
a.
a
- 3 8
x Er
- C tr c:
C tc cr s
(*J to g')
n e
8 r.
a a 3
i 0
L' kJ te LJ bJ LJ bJ t
Z Z
e
.cc rt tr t r.
cr Z
u a
W
'd I
f Z
I'-'
Z II IC C4 ts Cr 8
P4 3
8 tJ w
w Li oJ n.s
- t.
- ta e
0 1
Z 0
Et E
Cf f1 l'
L:
O CL O
g e. + e e a e e e e e.. e e o e
e o e +
e e e e e e e e e o
e e e., e.. e o
e e o e o e e g H
q 4
t L._
0 I
0 I
e l
8 8
to 8
C-0 (L 8
4%
h h
tt It c
O r*
I CJ oc ' O O
O M
M C
H t
O I
b f-t tz a
0 m
t u.
8 ei to ta ta ta (2
ta t2 E
e4 v4 e4 to tJ CJ C
C3 e4 8
- e. J S
1 4
a I
.J e4 8
- ^
PA 8
3 e e + e o e e e e o e e e.. e o e e e + e e e e
e e e e e e o e e o e o e e e e e e.. e e e g c-4 0
0 E
8 17 M
8 8
0 I
Z e*
I Ee 4
9 0
O O
e e.i s
t 4
er et st es et et s
O 4
8 r3 I
f *^
t CN c-c.1 r) - C.1 r.i t
H H
fa t-r-
r~
c-c-
eo N
t--
e 4
e'
.i I.
h h
h h
h M
h r3 i
C e4 h
P'n M
h M
M M
tJ Z
P4 4
U s at s
<t e"
8 (1
e h
9 L.
I h
h h
Pe M
f 0
LA U
8 I
)
F(
o e + e. e e e o e e o e o e e e e e e + e e e e
e e e o e o e e e e e e e o e e
e o e e e e e a tr rf 9
S i
0 43 F-t 8
8 8
H U
8 11 6
9 4
Z e"
0 hi I
I 4
a w
oc o
o o
o cs r
o a
w 3
ti e
4 Z
sy a
r-c-
8 h
h h
(J tJ tJ h
M CJ IT LJ 94 I (1 I
h M
1 s-4 r4 H
r4 e4 e4 e-4
- 4 e4 5
L) 9 LC
()
34 g
e e + e e.. e.. e e e.. e o
e e e + e e e e e e e e e e e e e o e e o e e e e e e o e e e g H
)-
0 7" E
e4 H
i l7 8
O Z
41 0
t B
Z t
O
,J 8
I 8
H E
Y t-i:
8 o
8 C
3 3
L1 t
8 B
0 Z
hs a
I 0
- e 9
kJ 4
e tc 8
b
- t. t ti l
i 4
s er
>c t.J d eO I
I s
< ;. s?
Z ti e
r b; ;.
s rr s
I N5 r
a
. >- O e Li e
u eJ s
t4 I
I H
M M
e e4 es gt f
c
- f. 4 a ti a
O 3
0 e
L1
>- LJ H
H e
o s ;.
et 0
8 O
h O
M
.f f
8 8
fJ O
t O
C1 kJ e4 f
CC e
r-e4 to 11 5.
I e-4 6
e4 8
I 5
>- O 4 2,J Z >-
- t-1 t
30 t-e e
a a
t-tc e
O t'
M e4 fJ H
t-tv s
c.
cr e
e Z
- 4 e-4 9
9 Z 9
O C)
.i.i
.J CJ
()
8 O
E 5
0 f
B O
C3
.I
- g -
H O
I D
OtJ
<.'1.
1.J I
t-C1 2
8 3
41 ' O L1 L) p
- a t'
.c A
u 8
r; u ta sH e
tt e
a oc t-r-
e a
I t-r tr er e
C) 4 C) 1 H
e a
a O
L1 s
2 e
a rr to O
=I e
i e
L1 m
e s
e e
O t
a s
s e e
e 9
O a
u e<t-Z oc I
e M
e4 t*
M W
M P1 H
I r4 N
fJ N
P')
M s-4 (J
tJ ei r4
-CJ 8-P-
F-w e
a e
e N
e 8
- )
Zr 99 g
e a + e e e e o e e e e o
e e o
e e o e + e e e
e e e o
e e o e e o e o e e e
e o e e e e o e e g g,
t o :1 I
t 9
C*
8 OO 8
8 La b-(3 s
6-rr es F
rs y F-3 (j
p4 g; B
Y t4 8
0
[L 8
La O
,)J I
fr W
b O
Z U
I
=
C7 8
8 rT T
ri O
- f. i
.Y 8
'C F-ra e'
~~
!^
11
!s (1
1 I-
- r' I
I O
C L1
)-
ti i
t-Or ri F-L3 p
- ^-
til
>t tP La I
3 L8 b-3:
0 0
H L8 t,
Z Z
9
- I~
U U
- )
C3 Z
f" ! )
0 O
M I' O
9 I
fL' J
['
LJ O
8 C"
e+
O Cr O
H 48 'J 8
>4 4
s 8
8 L8 CL t.t F-I hk U
t-01 P
rt
.J
> - ()
4 7
11 s f
f O
7 O
.t
- r Z
8 F-L8 LJ "r"
8
- C : Y e
e
=1 O
tt
=:
H H
O s
t-T F
PO rc 7
in L#
t-t c
9 8
W LJ b
b.
[Y
.J a
1 O
C.
P
<'
C O
8 U
Lil Z J
U Pi O
3 (m b I
CJ 4
rr 3.-
Z t
.)
O O ;)
K us 3
Z n.-
g 9,
0 t) 8 O
O L
O F-O Li f
8 t-
=% r.
O O
u Le Le C#
a es -
I B
Z O
O t4 8
7 Z
La P
='
t-4 LJ
- I C
O Cr F
P4
)-
O t # 6J O () H et X
8 I'*
t 8 P-E P4
.J 14.
C-O I
O Lf C.
C.) Lo8 O
O
.J k.
.J 8
ZZCc
- rk 8
.J CL L
O 1
tv
[J
<t 8
~~
V)
O Cf F-P -
C)
P r.t of
>- ()
3 H **
F-00 C_
tu t-H (t U 1
r_ ". r: O 9 r-t O
(1 L6J O
L tt I
c)
O O
L1 C.
t-y
_J uo a
- e4 uJ eJ 8 e4 1
El Z
t-O r)
.C f
t U I
U U
- t' ts e
c F-F L J,,
ar O
Ls a
(L r, > - p -
3 L)
B O
ht p
td O
t)
I b.
3 P-H O
O O
[r O LJ g
5 E
.)
O (3 Z
Z p'
O U F-I O O
.1.I I
C of
.J I'
at d e of Cf
()
02 O
6J <*
I LJ ts eis u
t I
O O
A 7
e-
>4 i
w H
O Z
~-)
f)
>4 t) c C,
C;
_.3.
8 t-t-Cr c:
O L8 9
.)
O LJ F-p-
- s2 3
O eT bJ O
F M
rf O
.J kJ C.
EL
.J B
- T = '
8 I
LJ O
O r3 i
O O
O ZH UJ F-U 7H e
1. J b >-
0 I
C.
- I C)
Z
.J ff 9
O F-b-
8 8
ed O
O g;
.I o
et t
Y M
O fr O
<~
4 54 L8 O b-t t J i s F* P "I
C)
Oa 9
re tv O
tr r O e
c: c: LJ LJ 0
0
.I
>- =i fr fr" of 8
- ^
bJ O
- t. t s
)
D O
OO tr a O
O e4 H
OM e
- u. L.
0 0
k.
Z O
L;
le Z
O O
I
-~*1 Z
P Q F-O C
C3 b.
OO G
>->a' s
>-p-L) (1 B
8 Z3 3 e e e o e e o e e e e e o
e e e e e e o + e o e e e o e o e e e e
e e
e e e o e o e o e
e a e e I (J LJ e s e
0 h.
h.
Z "'"
B I
8 >*
I h
M M
M M
M
- M M
I It
(*
4
~f C
(?
C (f
CJ 8
r) O I
OOZL ib I
9 8
8 e +
e e o e o o e o e e e e. e e o e +
e o e e e o e e e e e e e e e
e e * * *
- e e e e o e s 3 g g a 8 (C 8
6 g
8 c.
I 8 O s'
8 Ch O
O O
O - O O
O 8
N f4 fJ Ch Cs CN O
O O
e4 O
M I
H fJ b) 3 g
0 *! tl f CJ P1 PT F1 P1 P1 P)
P1 8
- 1 e4 ei fd f4 (J
F)
F1 F1 F1 Il P) 1 8 se e f 1
8 e
e +
e e e e e e e e o e e e e e a e e +
e e e * * * *.
e
- * * *
- e o e
e o e e e o e e e e i e
8 9
0 0 O t
8 32 I
LJ 9 at I
n tJ r'
c c%
o o
o e
o eJ LA O
f'-
(N M
M L't c*
N 9
>i f
7 3
8 Zft B t-e4 tJ tJ (J
t1 Pt a
f r)
O D
h ti M
tJ CJ ta tw
(*
Z I
F 0
9 I
e' 1 *4 8
et
,t e4 ri H
B J
U e4 ei ei e4 ri r4
,e
.)
I
>4 I
r' 9 e e + e e... e o e e o e e e e o e e + e o e e o e o e e e e e e
o e
e e e o e e e e e * *
- I
()
I 8
64 8
I 0 6# l' 8
O e4 eJ h
e U
9 rw s
O CN O
ef eJ M
L1 M
r*
O CN g
gc 0
I F-I fJ CJ N
f4 tJ CV 4J N
8 sa cJ n
P4 r1 n
M F)
M p.)
p)
Pi e
gg 8 7:
8 4
1 l'
0 I Pe t
e V'
a-