ML20003C851

From kanterella
Jump to navigation Jump to search
Forwards Info for SER Review.Info Will Be Incorporated Into FSAR
ML20003C851
Person / Time
Site: Shoreham File:Long Island Lighting Company icon.png
Issue date: 03/16/1981
From: Novarro J
LONG ISLAND LIGHTING CO.
To: Harold Denton
Office of Nuclear Reactor Regulation
References
SNRC-544, NUDOCS 8103180582
Download: ML20003C851 (12)
v • d • e


Text

{{#Wiki_filter:. fgg LONG ISLAND LIGHTING COMPANY MN#"# SHOREHAM NUCLEAR POWER STATION P.O. BOX 604, NORTH COUNTRY ROAD e WADING RfVER, N.Y.11792 o0 March 16, 1981 SNRC- 544 l / f p, ( $y2 '$,\ gn . Mr. Harold R. Denton, Director . 9' _I*.n-P. l Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission g jg$2 C m Q ((. Washington, D. C. 20555 l --+ C"7 ,<p.,7, SER REVIEW Shoreham Nuclear Power Station - Unit 1 M-{N f ' Docket No. 50-3.2

Dear Dr. Denton:

The enclosed information reflects the understandings we have reached with members of your staff addressing their concerns related to the review of the Shoreham docket. This information will be formally incorporated into the FSAR at a later date. Very t oly yours, lTW J. P. No arro Project Manager l Shoreham Nuclear Power Station RAH:mp cc: J. Higgins fool s 

                                                                                                  , 11 81031803TrA               6

1 4 Information Enclosed With SNRC-544 dated March 16, 1981 i

1. - Commitment To Scram Discharge System Modifications
2. Response to NRC Issue RSB-10 Regarding Failure of Feedwater Heater E
                                                                         . . .       . . . . ~ *
                                                          ~
3. Response to NRC Issue RSB-11 Regarding Use of Non-Safety Equipment In Shaft Seizure Accident

^

4. Revision to FSAR pages 3.7-51, 3.7-52, and 3B-21.

i 4' 1 .I t a ' . O I-4 + 1 f t! 't s-3 J f 1

Scram Discharge Modifications (SDV) Several actions are being taken to modify the SDV to assure that SNPS will comply with the latest NRC requirements. The specific actions to be taken are: (a) Six ,new level instruments will be added to the IV, for a total of twelve, thereby providing full redundancy and diversity of level monitoring and scram initiation. (b) All level instruments will be relocated and repiped directly to the IV rather than being connected to vent and drain lines. (c) A second air operated vent valve and drain valve will be added to the SDV to provide redundancy of SDV isolation during a" scram. (d) SDV piping design and installation will be reviewed closely to assure that adequate volume, proper venting and draining and protection against thermal expansion and dynamic pressure effects are all provided. (e) . Additional surveillance test procedures will be provided to assure operability of the level instruments, vent and drain valves and overall system. D

w' 

 .i        *
                    =

4 , c . . . Shorehan Response to Issue RSB-ll-Failure of Feedwater Heater In the unlikely event that a drop in :feedwater temperature in excess of 100% F could occur and assuming no operator action, the decrease 

             -in MCPR and increase in reactor power would be effectively the same as in the case already analyzed for a 100 F drop. This follows from the fact that in,both cases a scram from neutron flux occurs and because of the fact thac the time constant for rhe fuel is very small relative to the time to a neutron flux scram. Furthermore, the surface heat flux would correspond to the steady-state power level value, and in both cases this would be the same; namely, the scram value. The only difference is that the scram might occur marginally sooner for the greater temperature drop. After scram, no further increases in power, etc., occur.

P00R ORGINAL

x , 

   .c   ,c-  -
                                                                         '     ~~

SHOREHAM RESPONSE TO ISSUE RSB-12 USE OF NONSAFETY EQUIPMENT IN SHAFT SEIZURE ACCIDENT The recirculation purp seizure event is considered to be an extremely unlikely event and as such falls into the category generally classified as an accident. The event is evaluated as a limiting fault. The potential ef fects of the hypothetical pump seizure " accident" are very conservatively bounded by the effects of the D3A-LOCA. This is easily verified by comparison ftheMoevents. In both accidents, the recirculation driving-loop flow decreases extremely rapidly. In the case of seizure, stoppage of the pump occurs; for the DBA-LOCA, the severance of the line has a similar, but more rapid and severe influence. Following a pump seizure event, water level is maintained, the core remains submerged, and this provides a continuous core cooling mechanism. However, for the DBA-LOCA complete flow stoppage occurs and water level decreases due to loss of coolant, thus resulting in uncovery of the reactor core and subsequent overheating of the fuel-rod cladding. Also, complete depressurization occurs with the DBA-LOCA, while reactor pressure does not significantly decrease for the pump seizure event. Clearly, the increased temperature of the fuel cladding and the reduced reactor pressure for the DBA-LOCA both combine to yield a much more severe stress and potential for - cladding perforation for the DBA-LOCA than for the pump seizure. Th.erefore , , it is concluded that the potential effects of the hypothetical pump seizure accident are very conservatively bounded by the effects of the DBA-LOCA and a specific core performance analysis or radiological evaluation is not considered.necessary. However, to be completely responsive to the NRC question, the following narrative is provided to show the impact of not taking credit for nonsafety-grade equipment to terminate this event:

1) Level 8 Turbine Trip
                   - The FSAR analysis of the pump seizure event assumes that the vessel water level swell due to pump seizure will cause high. water level (Level 8) trips of the main turbine and the feedwater pumps, and indirectly initiates a reactor scram as a result of the turbine
                    . trip. The' FSAR Subsection 15 discusses the Level 8 trip function and shows that a turbine trip will eventually occur even in the event of failure of the nonsingle failuia-proof turbine trip signal circuitry. .In the case of the pump seizure without an L8 trip, the event is less severe than the analysis in the FSAR with the L8 trip for-the- following reason: A pump seizure, should it occur, would result in core flow reduction which reduces the core power and .

surface heat flux due to tha effect of the negative void reactivity coefficient. . Hence, the surface heat flux existing when the turbine trip occurs is lower because the turbine trip occurs later. Therefore, a loss of Level 8 trip would result in a lest. severe event consequence from the fuel than that currently depicted in Subsection 15. 9 

                                                                   -                         ___.m

~; .

2) Main Turbine Byoass System - \

1 As a result of the fiRC's concern rese nting rentivity effects of , pressure transients, GE and tne fiRC met on liowecber 20 and 21, 1978 for a comprehensive review of turbine trip and load reject transients without bypass. The principal conclusion of that meeting was that the cost limiting BWR transient event which takes credit for nonsafety-grade equipme'nt is the feedwater controller failure. Analysis indicates that a 2CPR increase of approximately 0.08 applies to this transient without a functioning main turbine bypass system. For recirculation pump seizure with a failure of turbine bypass system, the increase of aCPR would be less than that for the feedwater controller failure for the following reason: As this event occurs, the reactor power drops significantly within the first 2 seconds due to decreased core flow. Therefore, by the time of turbine trip, the reactor power is at a low level. The core power is the main parameter which relates to the fuel thermal limit. The effect of failure of the main turbine bypass systea to stop the steam flow retains pressure on the core but contributes only a small positive reactivity feedback. This is a secondary effect of much less significance than the reactivity decrease due to fluid flow decreasing 

      .               through the core.

The increase of . core power is more severe for feedwater controlTec, failure (increasing) event than for a recirculation pump failuree because 'it occurs at a higher power level. "

3) Relief Function of Safety / Relief Valves, The contribu' ion of MCPR from taking credit for the relief function rather than the safety function of safety / relief valves is not significant because the MCPR always reaches its lowest value before opening of the relief valves.

Analyses of recirculation pump seizure where coolant flow rate drops rapidly have shown that MCPR does not increase significantly before fuel surface heat flux begins dropping enough to restore greater thercal margins as the plant intrinsically responds to the reduced flow rate. The effect of not taking credit for nonsafety grade equipment is a ACPR increase of 0.08. Therefore, the MCPR for pump seizure event is. still well above the safety limit. ' 

                                                               ?DDRORINAL

P' w .-' 52s-)- SNPS-1 FSAR 3.7.3.5B .sientficant Dvnar: tic Reecense Modes 

                         .                                                                 'i When the : natural- frequency of a component is unknown, it may be analyzed by applying a static force at the center of mass.              In order to conservatively account for the perscibility of more than one significant dynamic mode, the static L rce is           calculated as 1.5 times the mass times the maximum acceleration from the                         <

response spectra of the point of attachments of multi-span l structures. For simply supported structures, the peak spectrai l acceleration is used. 3.7.3.6B Design Criteria and Analytical Procedures for Pipinq { See Section 3.7.3.6A 3.7.3.73 Basis for Computing Cbmbined Response The two horizontal <v=nennents and one vertical component of ground motion are accounted for in the following manner: two set of seismic results are obtained. First the maximum value of one horizontal the horizontal component of the earthquake isj ssumed to act in direction q;i,mua.taneouar with the vertical component, and the loads are ocanputed for this evvahination. ~Next the .mnvimum value of the horizontal ccr.sponent of the earthquake is assumed to act narnaMienlar to the direction previously assumed and imul ith the vertical component, and loads are computed or a combination. The larger of these two loads , at each point in the system is used for design. l i This method of analysis is based on the fact that the , seia nlogist specifies the ==v4 mum resultant value of the t horizontal cumponent of the earthquake when specifying the I horizontal component of the DBE. ' This method- conservatively assumes that the horizontal ~ and vertical components of the earthquake response occur simultaneously. 3.7.3.&B .Naptified Seimmi c Response l see Section 3.7.3.8A l l 3.7.3.9B Use of . Simplified Dynamic Analvsis Pbr equipment and piping supplied or analyzed by GE, a simplified dynamic analysis is .not used. 3.7.3.10B Modal Period variation See Section 3.7.3.10A 3.7.3.11B '1brsional Effects of Eccentric Masses s Torsional effects of eccentric masses is discussed in ' l Section 3.7.2.1.6.2B. d ___ . . _ _ _ 3.7-$2

Th ~ IsLawri sum 1 Of 2 

       '.,,                    .                                                       M s Bob Halsdyna
  • LIICO Project Office SwS-1. m (516) 92M700 I-09 {

f.g ram s.r. xista.s.k v.5/4 - sar. j 

            .T        .-                                                                                                 -'
          '.Tg                      33-1.90       Innervice Inpsection of Prestressed Concrete Containment Structures With Grouted Tendens (11 /74)                                          <
   !         N; J Prestressed           concrete  is not used in the containments therefore,
            .f.          -

Regulatory Guide 1.90 does not apply. 

                                                                                                                            ;         r   \

6+ 35-1.91 The Evaluation of Exolosions Postulated to Occur on . N- Transportation Routes Near Nuclear Power Pla'rt e  ! i 

              *T                                  Sites (1 M S)                                                          . :.

T. The guidance provided in Regulatory Guide 1.91 was used to h evaluate the suitability of the site with respect to erplosion Le?' c; hazards. b .. Reference Section 2.2 , 

        )y%,,

N' 3B-1.92 Combination of Modes and Spatia _1 components in Seismic ResDonse Analysis f12M4) t. hily i

g. f. The methods of Regulatory Guide 1.92 were notAemployed in the g' design of the plant due.to the advanced stage of design and 'l, A issued. g.g. ,,.f j.  ;

g, for balam o f pied e*=ereun+.rconstruction at the time the guide was 5: . 

                     ." 4 The seismic analysis employed a square root of the sum of the j           '
            ' .-                   sa. cares combination for all medes including                   thoee   within      { d 9                 10 percent of each other in frequency spacing.                                             ;

b$ 4-7[8~ g 

                                 -The guide           also proposes that the results from the simultaneous application of two orthogonal horizontal earthquakes plus a                              j-I g~                         vertical earthquake be combined using a square root of the sum of                        .i y                      the squares combination to obtain design values.                                        2, 4  -

For Seismic Category I structures, syste::s, and components (other i. l- 

          ,"F ."

! than balance of plant piping) the two horizontal components and 1, 74 one vertical component of ground motion are accounted for in the  ;. 

            , D-U                  following manner. The calmalated maxi:nza responses due to one                            i-borizontal earthquake are combined with the mvinum responses due                        7
              'T                   to the vertical . earthquake by the           absolute sum method.         The            i.

aprimure responses due to a second horizontal earthquake  ! J orthogonal to the first are also combined with the responses due e E Y. to the maximum vertical earthquake by the absolute sum method. I: ' the larger of the locls resulting frca these applications of  : 

        ,h. "(@~ - horizontal and vertical earthquakes is used for design.                                                   l.

t For balance of plant piping the zndividual modal respor.se due to i 

       ,[ g
  • the vertical earthquake is added absolutely to. the SRSS combination of' the .:axi:: rum responses d.2e to the two orthogonal Y.' horizontal earthquakes. - *
          -M;                                                                                 ,

Reference section 3.7 l. 

            ~

1 

          *g                ,;.                                                                                             [

g .- 3 

     !              i

_ __ ' j 1. . 3B-21 . Revision 16 - Aoril 1979 l

I I I 

    .. .                                                                                                               1 TELECOPY 3-10-81             2 of 2 Bob Haladyna i

l INSERT A However, when considered with the modelling techniques, damping, allowable stresses, and prescribed earthquake spectrum, the methods used to combine the effects of three spatial components of ground motion produce reasonable, conservative design loads. END OF INSERT A INSERT B All piping and squipment analyzed or supplied by General Electric are evaluated to the requirements of Regulatory Guide 1.92 with respect to the conbination of modal responses. For details, see the discussion in 3ection'3.7.3.3B. END OF INSERT B 

                                                                                      #   -*M  W  e =,          gap,,

e... G 

                                                                                        . 'N*N           +      we W

h

SNPS-1 FSAR

3. For fatigue evaluation, one-half percent (0.005) of y these cycles was conservatively assumed to be at the peak load 4.5 percent (0.045) at three quarter peak.

The remainder of the cycles have negligible contribution to fatigue usage. The DBE has the highest level of response. However, the encounter probability of the DBE is so small that it is not necessary to postulate the possibility of more than one DBE during the 40 year lif e of a plant. Faticue evaluation due to the DBE is not necessary since it is a faulted condition und thus not required by ASG Section III. The OBE is an upset condition and therefore, must be included in fatigue evaluations according to ASME Section III. Investigation , of seismic histories in SARs of many plants show that during a 40 year life it is probable that five earthquakes with intensities one-tenth of the DDE intensity, and one earthquake with approximately 20 percent of the proposed DBE intensity, will occur. Therefore, the probability of even an OBE is extremely low. To cover the ccabined effects of these earthquakes and the cuzmilative effects of even lesser earthquakes, one OBE intensity earthquake is postulated for fatigue evaluation. Table 3.7.3B-2 shows the calculated nu ber of fatigue cycles and o the n= har of fatigue cycles used in design. 3.7.3.2B Basis for Selection of Forcing Frecuencies All treguencies in the range of 0.25 to 33 H= are considered in the analysis and testing of systems and components. 3.7.3.3B Square Root of the Sum of the Squares 3 The square root of the sum of the squares (SRSS) cx:mbination of L I modal responses is defined mathematically as: 

          -N                                  n C                         R=      ,[   (Rt )2
          -.)

1=1 E where: R = Combined Response 

 -         Y                   Rg = Response in the ith mode l-                              n = Number of Modes considered in the analysis 4       si           3.7.3.4B   Procedure for Combining Modal Responses p                           when the response spectra method of modal analysis is used, all

] modes are combined by the square root of the sum of the squares. j 3.7-51 l l:.. 1____ - _ _ _ - _-_____-___ _ _ _ _ _ _ _ _

_./- - I ms aT 

             ~         '
                                       ,f                                              C_.                    .

DELETE EXISTING SECTIONS 3.7.3.38 + 3.7.3.48 4 ADD NEV SECTIONS 

        .,I

[ ' 3.7.3.33 COMBINATION OF MODAL RESPONSES l 4 - R 

        'l, In a response spectrum nodal dynamic analysis, if the modes are not R

I closely spaced (i.e., if the frequencies dif fer front each other by more than 10 percent of the' lower frequency), the rodal responses l-f. are combined by the square-root-of-the-sum-of-the-squares (5R55) 

    <                                 method as described in Subsection 3.7.3.3.18.

j If some or all of the modes are closely spaced, a double sus method, as described in Subsection 3.7.3.3.28, is used to evaluate the conbined p li response. In a time-history method'of dynamic analysis, the vector I sua at every step is used to calculate the corebined response. The I use of the time-histories analysis method precludes the need to 4 consider closely spaced modes. ! i 3.7.3.3.15 SQUARE ROON OF THE SUN OF THE _5QUARES HETHOD I f. , i h .*

i. Mathematically, this SRSS sethod is expressed as follows:

i u. 

 "~

n 1/2 

   'f                                           R =[R )2   g k:I         )                           .

where i( 

 ) I R = Combined Response th mode Rg = Response in the f n = Number of Modes considered in the analysis 1

a 1 i l i ' e _ __. _. . . _

3.7.3.3.28 DOUBLE SUN METHOD q

i. This method is defined mathematically as:

[ [N N h1/2 R= h Rh Ns ks 1 where i. h; R = Representative maximum value of a particular response of a given

elevent to a given component of excitation
   ,1 Rk = Peak value of the response of the element due to the Kth      mode h                  N = Number of significant modes considered in the modal response j                           combination 2, = Peak value of the response of the elaent attributed to s th mode f

Also. 

                                                  ~

[ k5

  • M 1 +

I \ 

                                 %         Ek "k
  • E'"s s
                                         ~

in which: ~ d l 4 a . f W , 2

f. "k * "k I~E 2[2 k k k U

E

  • E
  • t d"k i

where I l 4

h. wg= Nodal frequency in the k th mode . _

1 a th Sg'

  • Damping ratio in the k mode
  ;.'                     tg= Duration of the earthquake 1 i 1
                                                                        .g.

l l ', I}}

Retrieved from "https://kanterella.com/w/index.php?title=ML20003C851&oldid=2603349"