ML20210P189

Rev 0 to EA-AIR-BRP-97-027, RAI - Regarding Rerating of Portions of Post Incident Sys,A-BRP-97-027

ML20210P189
Person / Time
Site:	Big Rock Point File:Consumers Energy icon.png
Issue date:	08/12/1997
From:	CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:
Shared Package
ML20210P186	List: ML20210P182 ML20210P189
References
EA-AIR-BRP-97, EA-AIR-BRP-97-0, EA-AIR-BRP-97-027-R0, EA-AIR-BRP-97-27-R, NUDOCS 9708260468
Download: ML20210P189 (12)

Text

{{#Wiki_filter:-. EA-AIR-8RP 97-027 BIO ROCK POINT NUCLEAR PLANT ENOiNEERING ANALYSIS WORK SHEET Titk = RAI Regarding Rerating of Portions of the Post incident System, A BRP 97 027. Analyst Review Method (x) Tech Rev Rev Description By Date Alt Calc Det Resiew QualTed Date /<[F7 S O VSW 3 BAB 8/8//97 K A Operability y IPRRODUCrlON In support of the post incident system pressure rerste, analyses were performed to determine the response of VSW.3 to the limiting 0 63 ft2 steam line break. In these analyses VSW 3 was analyzed with [ Reference 1) and without [ Reference 2] insulation. With insulation the valve temperature rise at the time of spray initiation was calculated to be 34. Without insulation the valve temperature rise at the time of spray initiation was calculated to be 394. Per review of these analyses the NRC Staff requested the following additional information:

1. Provide an additionalanalysis or otherju.stifcation that demonstmeedthat the win was operable in the uninsulated conditionfor the licensing-basis accident conditions.

2. (fapplicable, provide the timeframe in which the win was desermined to be Inoperable.

To address these issues the FHSR 1980 CPCo COtREMPT analyses input data were used. The bcensmg-basis accident calculation (FHSR Figure 3 4 trace 5) assumed an initial containment temperature of 1004. The valve body and fluid initial temperatus es were assumed to be 70f which is ansistent with the plant licensmg analy s and plant operational data. These boundary conditions 6fn from the conservative p assumptions considered in the Reference 2 calculation. To determine the valve body tempercure response the Uchida and forced convection film coefficient boundary conditions were calculated usag the PC CONTEMPT LT/28 code [Refesence 3] based on the SAFE mass / energy blowdown data. With the boundary conditions defined, the Thermal Lag Analysis (TLA) code [ Reference 4) was used to determme the VSW 3 teruperature profile. RESULTS ' Assuming no insulation the peak valve metal temperature was calculated to be 1644 at about 1200 seconds This is below the rated temperature of 1754. Figure 1, page 12, presents the calculated thermal profile. The valve slow hestup without insulation is the result of the valve cast iron body large thermal inertia coupled with the fluid boundary condition. The temperature begins to decay at about I400 seco wis..- CONCLUSIONS Given the many analyses conservatisms and that the peak temperature was calculated to be less than 1754 -. VSW 3 without insulation is and has been operable w hen considering licenang-basis accident conditions. page 1-of-12 vsw3.med '9708260468 970820

• PDR-ADOCK 05000155 P

PDR;

' NR-BRPc97-027 t r ATTACllMENTS (1) CONTEMPT LT/28 Input data. [2) CONTEMPT LT/28 Output data. [3] Calculated U for TLA Input. [4} TLA Input Data. [5] VSW 3 location. [6] VSW 3 piping isometrics. [7] VSW 3 valve geanettydata. REFERENCES [1] EA SC 97 007 BB41," Post incident Pressure Retste*, BABrogan,3 26 97. [2} EA SC 97-007 BB42,' Post incident Pressure Rerate', BABrogan,3-28-97. [3] EA C BRP 96-93512, ' Analysis of 0.63 A2 Main Steam Line Break Using the CON TEMPT LT/28 Canputer Code',10 28-96. [4] EA C BRP 96-935-14, refer to SoAware Quality Assurance Plan Tiamal Lag Analysis SoAware, JROabor, 10 28-96. [5] D.C. Slaughterbeck, ' Review ofliest Transfer CoeDicients for Condensing Steam in a containment Building Following a less-Of Coolant Accident', Idaho Nuclear Corporation IN 1388, September 1970. [6] P.M. Donnelly to NRR Docket 50155, License DPR 6, ~ Big Rock Point Response to Genene Letter 88-20 Individual Plant Examination for Severe Accident Vulnerabilities,' dated May 5,1994. [7] D.P.lloffman to G.R. Burger," Big Rock Point Plant IIcat Addition (Bru/br) to Lake Mk,higan', October 4,1982, page 2-of-12 vsw3.med i

cm AIR-BRP 97 027 CONTAINMENT TEMPERATURE LEAD ' First, using a set of hmiting primary system mass and energy loads, the contamment pressure ard temperature are calculated De 19f4 CPC 0 6319 input data set were employed _ This is the licensing basis analpis as reported in the FilSR. METilODOLOGY Environmental Quahrwation of Electrical Equipment guidelines were used in this analysis: Regulatory Guide 1.89 and the Final Rule for Environmental Quahfication of Electrical Equipment important to Safety for Nuclear Power Plants (iOCFR50.49) refer to NUREG 0588, ' Interim Staff Position on Environmental Qualification of Safety Related Electric 31 Equipment', which provides guidance on the methods which are considered appropnste for qualifying the equipment in different areas of the plant. Appendix B of this NUREO establishes the nethods to determ'ae the containment environmental response. First, using a set oflimiting primary system mass and energy loads, the containment pressure and temperature are calculated. As rcww.ic.ded in NURE04588, BRP used the CONTEMPT LT/28 code to perform the contamment load analysis employing the BRP FilSR licensing analysis of record input data (i.e., Figure 3-41980 Aerojet analysis). CalculatiswtCmtaimncotCmditimu ne 0 63 ft2 steam line break was detenmned to be the most limiting case. The temperature, pressure, and atmosphere energy data are used to determine the Uchida film coefficients discussed below. Component Thermal Respses Pet 10CFR50.49 paragraph f(4): (f) Each item ofelectric equipment important to safety must be quahfied by one ofthefollowing methods: and (4) A nalysis in combination with partial type test data that supports the analyticalassumptions and conclusions. The following provides a description of the methodology for safety-related component thermal response As stated in NUREO-0588: The heat transfei rate to component should be calculated as follows-a Condensing lleat Transfer Rate q/A = ha(Ts Tw) where q/A = component surface heat flux. haor Ua = condensing heat transfer cocmcient is equal to Jchida correlation. T. = saturation temperature (dew point). T. = component surface temperature. page 3-of-12 vsw3.med i l

EA-AIR BRP 97-027 4 ..m bl-Convective llent Transfer A convective heat transfer coemeient should be used when the condensing heat flu.x is calculated to be less than the convective heat flat During the blowdown period, a forced convection heat transfa correlation should be used. For example: NU = C(Rey where. NU = Nusselt number, Re = Reynolds number, C,n = Empirical constants dependent on geometry and Rep >lds number. AAer the blowdown has terminated or reduced to a neglibibly low value, a netwal convection heat transfn corretatson is acceptahic. 'Howewr, use of a naturalconwesion coeficient must befwllyjustified whenewr used

• nis analysis conservatively assumes forced convection aner condensation has stopped i

page 4-of-12 vsw3. mod

!-97 027 Anahsis ORIGIN. 1 Set the array initialization to 1. Data. READPRN(ser e) CON 1Y.MPT LT/28 output ol' time versus t:Aalenergy in the atmosphere. Attachment 2. . Detal :: READPRN(ser_pt) CONTEMPT LT/28 output of time versus total pressure and partial pressure. Attachment 2. Data 2 '.a READPRN(UCIUDA) From Table 5, pg 65, NUREGER 3716. March 1984. time :: Data *'* Q = Data Ptot :: Detal*3* Pp steam ::Detal** T., = Detal** 9" g i.: 1. length (T,,,) j := 2 Mt' m) u h, : Data 2*2" NUREGER 3718, Table 5 page 65. ratio. Data 2*3* NUREGER 3718. Table 5 page 65. masa 3_ Vcont.: 9.128910' ft!. The Uchida values are calculated by computing the air mass fraction: PP nir Itot-Pp steam. Pp air f29) 1 - l Determin? tion of air to-steam ratio. calc ratio :: PP steam (183 ratio.h, calc,,i;o] 0.1762 uchida, linterp(mass Btu,br-ft2-F. h u i :: I.. length (time) max (huctuda) = 5.8810' Figure 2 UchidaCorrelation 100 i G 4 {w a t 0 1000 2000 time (882) Write to Glc the calculated data: %21TEPRN(time).: time. Time, seconds WRITEPRP(h_uctuda) +:huchida UcMhdh page 5-oM2 vsw3.med

=. -- EA AIR BRP 97-027 F,orced ceas eetion Propert) Deflaitions Steam READPRN(Steam) Read the data into a 53 x 1I matrix the steam Saturated property data, Reference Steam Tables by J lt Keenan et al.,1%9. Steam i = READPRN(Steam _,1) Read steam specific heat, thermal conductivity, = Prendt! oumoet values. Air :READPRN( Air) Read air property data. Temp Steam '* Press

• Steam"*

Derme the pressure and temperature vectors. Steam _cp :READPRN(St cp) Read 6e constant pressure specific heat data far saturated and superheated steam. Define the Specific Volume, A3/lb,vecton: Water,v f : Steam Evap_v gg.: Steam"" Steam _v, Steam ** <2> Derme the Enthalpy, Btu /lb, vecton g. Steam"* Water _h g : Steam"* Evap_h rg : Steam"* Steam _h Define r Vat transfer property, vectors (density, spccific heat, conductivity, viscosity, and Prandtl number): T air : Air ** p,;, : Air #3" cpair : Air"" p,;, :: Air ** kair : Air" Pr ir.: Air

• a Derme steam heat transfer property vectors (specific heat, conductivity, viscosity, and Prandtl number):

Tsteam : Steam g"* cp meam : Steam ;"" ksteam : Steam g"* p steam : SL'** I Pr team.: Steam g* s Forced Convection for a flat plate (* Principles of flest Transfer', by F.Kreith, Fourth Printing,1961): i i h =b 0.664 Pr Ret 3 2 cL pvD Re La p L = characteristic length of the component, assume 11 Forced Convection Steam Evaluation - Per NUREO 0588 AppendixB 3 25 M BD w Blowdown velocity, A/sec. V, 4 page 6-of-12 vsw3.med

EA-AIR BRP 97 027 where: v ' = velocity, ft/sec. L := 1 ft (assume local characteristic length). M BD = blowdown rate,Ib/hr. V = cmtainment volume, M. cent Blowdown : READPRN(Blowdwn) Refer to EA C BRi' % 935-10. M BD :Blowdowg2> 3600 blwdn :81 wdown '# t let: j = 1.. length (M gg} T, alinterp{ time.T,3,tblwda) Calculate the gas temperature using i g s blowdown data peints. At 0.01 see the SAFE results (refer to EA-C BRP 96-935 10) 25 M BD reports a blowdown value of 1767 lb/sec, j v=j ctc. cont H mix =linterp{Tstemn H steam,T atmos)Pmix linterp(Temp, Steam _v.Tatmos) j j j g mix.:liderp(Tsteam,ksteam,T atmos) mix :=linterp(Tsteam,Pr team T ats;os) k Pr s j j j j P mix.'V 'l 'k - 1 I j Remix

• hsteam ::

3 0.664-{Prmix}3 { Remix} Bru/hr-ft2 F j j j Figure 3 Steam Forced Convection 10 y g q Average value: (5 length (tblada) { h steamj I~I = 4.53 Brutr ft2 F length (tblwda) o so too tso 200 tame (wa) page 7-of-12 vsw3.med

EA-AIR-BRP-97-027 l** Forced Conseetion Air Es aluation Using the above equations, re-evaluate the forced convection heat transfer coefficient considering only air: H mix _ linterp(T a r*H sir,T atmos) :linterp(T air P air,T atmos) P mix j j s tinterp(T sir,Pr ir.T atmos) slinterp(T air,k,,,T atmos] kmix Pr mix a j j P mixjV'l Q j t f Remix ' h 3 0.664 (Prmix) -(Remix) Btatr ft2 F j air;

• g Figure 4 Air Forced Convection 18 i

i i Average value: g length (t blwda) h air; 5 3"I = 6.65 Btultu-ft2 F length (tbiw&) I o o so too im 200 tuns becs) From the alme, a nommal value of 6.7 Btuhr ft2 F (compared to the estimated aw: rage values of 4.5 and 6.7 Btu /hr ft2-F) will be used for forced convection. Since both air and steam exhibit the same relative magnitudes, volumetric analysis of the mixture is not required. VALVE T:IERMAL CONDUCTION ANALYSIS A thermal conduction model was developed to measure the thermal lag of various components as a function of time. The code was designated as the Thermal Lag Analysis (TI) Code (Refer to C-BRP 96-93514). The key features of this modelinclude: 10 Variable thickness and properties of multiple layers. Air gap allowid. '.mplicit solution. Slab representstion. Cylinder represenistion. Vanable Boundary temperatures and heat transfer coeflicients. Execution of the TLA model requires the creation of an input file (named conduct.dat refer to Attachment 4)to represent a specific wmycnent transient temperature response, The file allows the user to specify material properties, heat sink dermitions, boundary conditions and other miscellaneous items, Note the thermal conductivity value of the water segment was increased to -inailate mixing and natural convection in the pipe. page 8-of-12 vsw3.med a--mmim

EA-AIR-BRP-97-027 Decause the VSW 3 Mall Valve Company specification data were not available, Powell Class 150 gate vahc (6 inch) dimensional data (Attachment 6) were assumed The body length used was 15 7/8 inches. The inside diameter assumed was 6 065 inches The wall thickness considered w as 0.28 inch (per ANSI Bl6 34 1977). Cast iron properties assumed included a thermal conductivity value of 31.8 Bruhr-ft2 F, a specific heat value of 0.1 Btu /lbm F, and a density of 474 lb/fD. The valve body and water initial temperatures were assumed to be 70of. Recent BRP control room data sheets were reviewed to assess this assumption Typically - aice water eemperatures range from 30 to 50*F for most of the year. The months of August and Septe nber ediibit the highest temperatures which are in the mid to high 60's. Consequently an initial yalue of 70af was used. This data was costaborsted by resiewing 1982 temperature data (Reference 7) Furthermore, the FIISR Chapter 3 design bases analyses assume 70aF. ANALYSIS CONSERVATISM'S Contamment Temper:1ws As shown in Attachments 4 and 5 VSW 3 is located outside the recire pump room in the sphere at the 604' elevation adjacent to the pipe tunnel. Temperatures in this region would be about 25'F to 50*F colder than the recire pump room for a 0.63 A2 steam line break. This conclusion is based on extensive analyses of the Big Rock Point containment using a node / junction model. The nodelfjunction model or Generalized Containment Model (GCM) was developed by the Department of Energy's Advanced Reactor Severe Accident Program,in cooperation with General Electric. This code has been used extensiwly to model the GE Simplified Boiling Water Reactor (SBWR), as well as modeling of the ABWR, and is the basis for the BRP severe accident evaluation (i c. the 1994 BRP Level 2 individual Plant Evaluation). Reference 6 provides further examples of the sphere relatively cold atmosphere temperatures for both LOCA's and steam line breaks from a node and-junction representation of containnent. IkaLTutufer Coefficients As noted above the Uchida data werr - i in the conduction analysis As noted in Reference 5 the local heat transfer coeflicient may be extremely v tor a few seconds after the break for a location (the recire pump room) near the break. Ilowever, for a non local au (sphere) adjacent to the break ine average heat transfer coefficient could be very low. These observations were corraborsted by experiments as vell as the unexpected rupture of a simulated containment vessel used by Kolflat [ Reference 5 page 2]. The following briefly describes the different blowdown regimes. In performingantainment analyses, twu separate time periods are generally considered following a loss of coolant accident in which di1Ierent correlation's for condensing steam heat transfer coefficients apply. The first period is characterized by high turbulence caused by the decompression of the primary coolant system. This period is referred to by various investigatcrs as the forced convection period, the turbulent portion, the blowdown period, or the transient state. The end point of this state is not defmed explicitly but is the erd of pressunzation of the containment resulting from the initial injection of primary coolant into the containment. The injection of pnmary coolant into the BRP containment would likely occur in the recire pump room where most of the primary system pipmg is located Later additions of mass and energy from the core and containment spray systems would not affect the end point of the forced convec. ion region For a 0 63 ft2 steam ime break, the end of this segment would occur in about 70 seconds, The second time period is referred to as the natural convection portion, the steady state portion, the post blowdown portion, or the stationary state. As the terms imply, this segment is characterized by lower turbulence following the decompression of the primary system. The Uchida correlation characterizes this penod. Since lower turbulence would be expected in the sphe e where VSW-3 is located, the Uchida correlation was applied as part of the boundary condition in tl e conductios analysis of VSW 3 both prior to and following the peak containment pressure. page 9-of-12 vsw3.med

~4-BRP 97-027 fmally the SAFE mass / energy data were used to calculate the forced convection filrn coefficient after the initial blowdown phase. His resulted in a more conservatively applied boundary condition whenever condensation ceawd. In summary, the heat transfer to VSW 3 was calculated using the Uchida correlation (u rmwncaded by NUREO 0588 Appendix B for electrical equipment). He original Reference 1 and 2 calculations used just the peak Tagami value in defining the valve boundary condition. His was an overly conservative and non-mechanistic assumption. 6AlallkaLRcm9nl describes the VSW 3 piping isometrics. As is evident from the drawings several feet of piping extend from VSW 3 to the senice water pumps to the post incident system. The availability oflong pipire ms in a relatively cold environment limits the potential heatup of VSW 3. For example, there is almt a 100 fed of piping from the downstream VSW 3 tee branch to the reactor cooling water heat exchangers. In a parallel connection from the same tee approximately 120 feet of piping (ignoring the underground seEment) coimects to the senice wa bay (recall that senice water temperatures nominally range from the 30's to 5(Ts for most of the year). Similarty several feet of piping extend to the P!S connection. Rese heat sink paths were not credited in the analyses. LinMaskt the following assesses the conservatism ofjust modeling the valve cylinder in the conduction analysis by comparing the thermal time constant ofjust the cylinder to the valve. t:= Defme valve radius. 2 Wt :190 Weight in Ibs, Refer to Attachment 7. I Where 490 lbs is the density of steel. Valve yo 490 Cy% vol '., x-((r + 0.28)* - r ) 15.88 2 Cylinder volume, td. For wall tinckness (0.28') refer to Attachment 7. Cylinder,,,, ': x (6M5 + 0.28) 15.88 I-Area of cylinder used in the conduction analysis, 144 n2, area : x r-(31 - r) 1 Approximate area of the stem, ft2 Steta 144 Valve

Cylinderarea + Stem Approximate area of the valve, ft2, nres area 6=

Thermal time constant. Where c p V (or "C')is the thermal capacitance and h-A s bA ( r 'R*)is the thermal resistance s page 10-of 12 vsw3.mcd

EA-AIR-BRP 97-027 j .Valvemi Valve e rstio : Ratio of valve thermal time constant to the cylinder modeled in the Cylindervoi conduction calculaticut Cylinder,,,, e rstio = 4.1 The valw has a slower time constant than the modeled cylinder by a factor of 4. nis is pictwed in Figwes 8 and 9 below. t ::.001,.002.. 5 Figwe 5 Valve Time Constant Figwe6 CylinderTuneConstant I I I I I i 4 4 g 0.5 05 I t I I 0 0 O O2 04 06 0 0.2 9.4 0.6 tune (seceeds) timme (second4 in summary, only the cylindrical portion of the velve body was modeled in the analysis. This is earmvative as the major portions of the valve body and stem were ignored. When compenng the thermal time constant of the entire valve (i c., the valve thermal resistance muhiplied by the valve thmnal capacitance) to the cylindrical representation of VSW 3 in the TLA code, the ratio is approxunately a factar d4. Tids suggests that if the entire valve were modeled in the conduction analyses (thus aediting the additanal cast iron thermal inertia), the heatup rate of the valve would be 4 times slower. Therefore the peak temperstwe would not reach 164*F as containment sprsys would how cooled the gas temperatwe before VSW 3 would bestup i page 11-of-12 vsw3.rncd i

EA-AIR-BRp-97-027 - Figure 1 - VS5N-3 Assuming Licensing - Basis Accident Conditions l VSW-3 250 i Peak Temperature of164 F occurs at 1216 sec g 200 2. 8. E150 W VSW-3 100 50 0 200 400 600 800 1000 1200 1400 1600 Time (sec) 1 page 12 of12 -}}