ML033240164
| ML033240164 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley, Millstone, Calvert Cliffs, Mcguire, Palisades, Palo Verde, Indian Point, Kewaunee, Catawba, Harris, Wolf Creek, Saint Lucie, Point Beach, Watts Bar, Sequoyah, Byron, Arkansas Nuclear, Braidwood, Summer, Prairie Island, Seabrook, Surry, North Anna, Turkey Point, Ginna, Diablo Canyon, Callaway, Vogtle, Farley, Robinson, South Texas, Cook, Comanche Peak, Fort Calhoun, McGuire, 05000360  |
| Issue date: | 11/18/2003 |
| From: | Schiffley F Westinghouse Owners Group |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| WCAP-15872-NP, Rev 0, WOG-03-608 | |
| Download: ML033240164 (56) | |
Text
111 63TAU1.L5HED7 A Domestic Members AmerenUE Caflaway America n Electric Power Co.
D.C. Cookl &2 Arizona Public Service Co.
Palo Verde 1. 2 & 3 Constellatlon Energy Group Calvert Cliffs 1 & 2 Dominion Nuclear Connecticut Millstone 2 & 3 Dominion Virginia Power North Anna I & 2 Surry I & 2 Duke Energy Catawba 1 & 2 McGuire I & 2 Entergy Nuclear Northeast Indian Point 2 & 3 Entergy Nuclear South ANO 2 Waterford 3 Exelon Generation Company LLC Braidwood I & 2 Byron I & 2 FirstEnergy Nuclear Operating Co.
Beaver Valley I & 2 FPL Group St. Luie I & 2 Seabrook Turkey Pont 3 & 4 Nuclear Management Co.
Kewaunee Palisades Paint Beach 1 & 2 Prairie Island Omaha Public Power District Fort Calhoun Pacific Gas & Electric Co.
Diablo Canyon I & 2 Progress Energy H. B. Robinson 2 Shearon Harris PSEG - Nuclear Salem I £ 2 Rochester Gas & Electric Co.
R. E. Ginna South Carolina Electric & Gas Co.
V. C. Sumrner Southern California Edison SONGS 2 & 3 STP Nuclear Operating Co.
South Texas Project I & 2 Southern Nuclear Operating Co.
J. M. Farley &2 A. W. Vogtle 1 & 2 Tennessee Valley Authority Sequoyah1 & 2 Watts Bar I TXU Electric Comrnanche Peak 1 & 2 Wolf Creek Nuclear Operating Corp.
Wolf Creek International Members Electrabel Doel 1 2.4 Tihange 1 & 3 Electricit6 de France Kansal Electric Power Co.
Mihama I Takaharna 1 Ohi 1 &2 Korea Hydro & Nuclear Power Co.
Kon 1-4 ULchin3&4 Yonggwangl -5 British Energy pIc Sizewell B NEK Krtko Spanish Utilities Asco 1 & 2 Vandellos 2 AlnUraz I & 2 Ringhals AB Ringhals 2 - 4 Taiwan Power Co.
Maanshan I & 2 November 18, 2003 WOG-03-608 WCAP-15872-NP Project Number 694 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, DC 20555-0001 Attention: Chief, Information Management Branch Division of Program Management
Subject:
Response to Request for Additional Information - WCAP-15872-NP, Rev. 0, "Use of Alternate Decay Heat Removal in Mode 6 Refueling,"
References:
- 1.
NRC Letter, D. Holland (NRC) to G. Bischoff (Westinghouse), "Request for Additional Information - WCAP-15872-NP, Revision 0, Use of Alternate Decay Heat Removal In Mode 6 Refueling," (TAC No.
MB9020), October 2, 2003.
- 2.
WOG Letter, R. H. Bryan to US NRC Document Control Desk, "Transmittal of Report WCAP-15872, Rev 00 (Non-Proprietary), Use of Alternate Decay Heat Removal in Mode 6 Refueling dated January 2003,"
WOG-03-254, May 12,2003.
By letter dated October 2, 2003, the Nuclear Regulatory Commission (NRC) issued a Request for Additional Information (RAI) for WCAP-15872-NP, "Use of Alternate Decay Heat Removal in Mode 6 Refueling," (Ref. 1). Westinghouse Electric Company LLC (Westinghouse) on behalf of the Westinghouse Owners Group (WOG) submitted WCAP-15872-NP for approval in May 2003, (Ref 2).
The purpose of this letter is to transmit responses to the staff RAIs (Enclosure 1). In addition, changes were required to portions of WCAP-15872 to be consistent with these RAI responses. These changed pages are provided in Enclosure 2. All updated pages will be integrated into the final approved version of WCAP-15872-NP.
U. S. Nuclear Regulatory Commission November 18, 2003 WOG-03-608 Page 2 of 2 Information transmitted by this letter is non-proprietary and may be released to the public. If you require further information, please contact Mr. Jim Molkenthin in the Owners Group Program Management Office at (860) 731-6727.
Sincerely yours, Frederick P. "Ted" Schiffley, II Chairman Westinghouse Owners Group
Enclosures:
(2) cc:
S. Dembek, NRC, Westinghouse D. G. Holland, NRC (via Federal Express)
Management Committee Steering Committee Analysis Subcommittee Project Management Office C. B. Brinkman, Westinghouse J. S. Galembush, Westinghouse V. A. Paggen, Westinghouse
U. S. Nuclear Regulatory Commission WOG-03-608 November 18, 2003 WCAP-15872, "Use of Alternate Decay Heat Removal in Mode 6 Refueling" RAI Responses
WCAP-1 5872 RAI Responses Page 1 WCAP-15872-NP, ROO "Use of Alternate DecayHeat Removal in Mode 6 Refueling" Request for Additional Information dated October 2, 2003 Main Report RAI 1.
What is a shutdown cooling "train?" Describe the physical setting of the two "trains" mentioned in Sec 2.2 of the text when they are inoperable at the time of the initiation of the alternate heat removal alignment, and when they are supplementing the shutdown cooling system.
Response
A shutdown cooling (SDC) train is a dedicated flow path consisting 'of piping, valves, a low.'
pressure safety injection pump and a SDC heat exchanger that provides cooling of the reactor core during shutdown conditions in Modes 4, 5 & 6. Two such shutdown cooling trains constitute the shutdown cooling system installed at licensed plants. A brief description of the shutdown cooling system and the alternate cooling alignment'for removing decay heat from the refueling pool during Mode 6 operation is given in Sections 2.1 and 2.2, respectively,'of WCAP-15872.
Standard Technical Specifications, e.g., NUREG-1 432, LCO 3.9.4, require that one of the two SDC system' trains be operable and in operation during Mode 6 conditions with the refueling pool fully flooded. The alternate'heat removal (AHR) alignment will furiction as a corplete substitute for the SDC system, thereby permitting the shutdown cooling system to be taken but of service once decay heat removal using the alternate cooling alignment is placed in service.
Thereby, AHR promotes outage schedule flexibility when maintaining plant equipment during Mode 6 operations.,
The reference to supplementing the SDC system refers to the opportunity for a utility to ensure decay heat removal by having AHR capability available to support normal SDC, either in combination with an operable SDC train, or as stand-by'should normal SDC become inoperable.
RAI 2.
Is your methodology predicated on the use of the spent fuel pool cooling system as the alternate heat removal system?
Response
The alternate heat removal system is predicated on use of any appropriate and available cooling system that has adequate heat removal capability, can be aligned to rem'ove heat from the refueling pool, and is judged to be sufficiently reliable. In WCAP-15872, the alternate heat removal alignment is modeled after that of Calvert Cliffs, where a spent fuel pool cooling train can be used as the alternate system to receive decay heat in Mode 6 with the refueling pool fully flooded.
WCAP-15872 RAI Responses Page 2 Appendix A: Algorithm for Natural Convection between Core and Refueling Pool For the one-dimensionai model of the core and refueling pool:
RAI A1.
Superimpose the nodalization that your methodology assumes on Fig. A-1.
Demonstrate that it is robust.
Response
The analysis is based on division of the refueling pool and reactor vessel internals into a series of control volumes. The state points for these one-dimensional control volumes are shown in Figure A-1 and identified as follows:
1 =
Reactor vessel inlet at the level of the vessel flange.
2 =
Core inlet at the level of the fuel alignment plate.
3 =
Reactor vessel lower plenum at the bottom of the core.
4 =
Core exit at the level of the fuel alignment plate.
5 =
Reactor vessel exit at the level of the vessel flange.
6 =
Bulk refueling pool.
7 = Alternate cooling inlet to pool.
8 = Alternate cooling exit from pool.
9 =
Shutdown cooling inlet.
10 = Shutdown cooling exit.
These state points represent natural boundaries between the control volumes and are consistent with the set of assumptions used to reduce the refueling pool coupled circulation problem to tractable form. The robustness of this model is demonstrated by its close agreement with the test data obtained at Calvert Cliffs.
RAI A2.
What are the assumed mass, momentum and energy equations for the related control volumes?
Response
The one-dimensional model is based on the following general control volume formulations for conservation of mass, momentum and energy:
Conservation of mass,
-JpdV+fpi-dA=O Conservation of energy, QCV - JCV - TSHEAR +l q dV = aepdV
+ f e+ P.
CV Cs.
P)
WCAP-1 5872 RAI Responses Page 3 Conservation of momentum, ECVF=F5+Fs =+tJiVPdV+Jilpi.dA CV Cs.
where Wcv is mechanical work, Wshear is work done by shear and Qv is the heat generation within the control volume.
These equations, based on the following assumptions and expressed in finite difference form, are solved using the algorithm shown in Figure A-2.
Assumptions involving flow through the core:
Upper guide structure and fuel alignment plate have been removed.
One-dimensional, steady-state flow with no horizontal cross-flow for vertical flow paths.
Neglect changes in kinetic and potential energies of the water flowing through the core.
Neglect any ambient heat loss, Qoss = 0.
Heat generation is constant and uniformly distributed throughout the core control volume, f qdV=Qcv cv Work associated with rotating shafts and moving boundaries is zero, WV 5 = 0.
Work due to shear stress is negligible, and shear stress on the surface of the control volume is uniformly distributed, T
- t(z).
Temperature increases with depth for down flow path, T2 < T3 so that P2 > 3 Density varies linearly with elevation, p = P2 -P where AP = P2 and L = Z2 - Z3-No heat storage in the fuel.
The upflow and down flow areas are identical, A2 =A3 -
'°"r_ A 1=2co.
2 Heat generation in the core control volume results in an increase in temperature, so that fa fepdV*O..
- cv Refueling Pool: assumptions:
One-dimensional, steady-state flow along a streamline.
Change of momentum within CV, a. [l J(pdV)] = O.
Frictionless flow, i.e., no viscous losses.
4 WCAP-15872 RAI Responses Page 4 Heat transfer from the pool surface due to natural convection and evaporation, Qcv = -[eL Asu< (T6 - Tmb) + fileaphfg].
Neglect kinetic and potential energy changes of the water flowing through the pool.
Neglect work due to shear.
A fraction of the pool water, e., mixes with the core flow.
For one-dimensional flow through the core, shown as flow path 3 - 4 on Figure A-1:
Conservation of mass:
h13 = 'h4 = p3A3V3 = p4A4V4 Conservation of energy:
p 3 p 3 +V 3 2 /2+gZ3 = p 4 p 4 +V4 2 /2+gz 4 +K 3 4 v 2 /2 Conservation of momentum:
-p 4 A 4 +P3A 3 -T34 Asl 34 -g 2
+
4 A coec=m4v4 -i 3v3 For one-dimensional flow through the pool:
Conservation of mass:
m5 = th6 = p5A5V5 = p6A6v6 Conservation of energy:
EULrMpooIX P dT6 + °surf = n.Cp 5 -T. )-
Cp (-
T.7)
Conservation of momentum:
A V2 +2 p5 + V5 + gZ 5 = A + 2+ gZ 6 p5 2
p6 2
The fraction of the alternate heat removal cooling flow that does not mix with the thermal plume is expressed by the bypass coefficient, eypass. Thus, the refueling pool exit temperature, T8, can be expressed in terms of the bypass coefficient, the pool average temperature, T6, and the alternate heat removal inlet temperature, T7, as:
T8 = (1-ebypas ) T6 + ebypass (TZ)
When the bypass coefficient is zero, all alternate heat removal cooling flow mixes with the thermal plume, or T8 equals T6. If none of the alternate cooling flow mixes with the thermal plume, then eypass equals one and the pool exit temperature T8 equals T7.
RAM A3.
What is meant by "The'effective mass is determined by engineering judgment?" How is the numerical value for use in the one-dimensional model computed?
Response
The effective mass, defined as emiX times the pool mass, identifies the quantity of fluid in the refueling pool that mixes with the natural convection flow from the core. This mass is
WCAP-15872 RAI Responses Page 5 determined through CFD analysis when solving for the mixing coefficient. Engineering judgment refers to the review to ensure that predicted results are verified by test data.
RAI A4.
What results show that the mixing coefficient 6x is about 0.90? What are the
- parameters to which the value of 9,v, is most sensitive? What is the sensitivity of dmJx to these parameters?
Response
The mixing coefficient is described in terms of the initial pool temperature and the pool average temperatures from one-dimensional and CFD computations. Since the mixing coefficient influences the rate of temperature change in the one-dimensional model, it was necessary to use a transient CFD case to evaluate emix. For a refueling water pool cooling configuration typical of CCNPP but having no alternate cooling flow, the mixing coefficient was evaluated based on the time required for the average pool temperature to reach saturation as determined by the CFD model. Table D-3 illustrates the time required to reach the boiling point for three different pool elevations and the associated mixing coefficient as predicted by the CFD model.
Based on this data, a mixing coefficient of 0.9 was selected as the best representative value for use in one-dimensional analyses.
The principal parameters affecting the mixing coefficient are the refueling pool cooling configuration and the mass flow rate driven by natural circulation between the core and the refueling pool. No alternate heat removal cooling flow was assumed when computing the mixing coefficients given above, which ensures conservative results for all alternate heat removal cooling configurations. In addition, parametric evaluations using the one-dimensional model based on arbitrary variations of the mixing coefficient did not produce significant variations in pool temperature or core flow rate.'
With regard to the sensitivity of these parameters, based on the alternate heat removal conditions at Calvert Cliffs, an arbitrary reduction in core flow rate of 20% resulted in about a 10% reduction in the mixing coefficient. Also, for the same core flow rate, the mixing coefficient was found to vary approximately +/- 5% when based on average temperatures at specific locations rather the entire refueling pool.
A typographical error was found in Table D-3. The temperatures shown in the column labeled
'Bottom" should read 8740F, 212'F and 215.50F, respectively. The CFD value for Jx should be 1.03, while the one-dimensional value for emlX is 1.0. Table D-3 has been revised to show these corrected values.,
RAI A5.
How is the value of the by-pass fraction Ebypass computed?: What results show" that Ebypass is close to 1.0? How close?, What is the sensitivity of.ebypass to key parameters?
Response
The by-pass coefficient is defined in terms of mass flow rates and is computed using the expression for eypass shown in Section D-2. Mass flow rates, in turn, are determined from pool temperatures predicted by the CFD model. For the Calvert Cliffs configuration modeled in this analysis and represented by Configuration A in Table D-2, results demonstrate that the value of
I I
I WCAP-15872 RAI Responses Page 6 the bypass coefficient is approximately zero for alternate heat removal cooling flow rates varied from 200 to 2000 gpm.
Table D-2 also shows that the value of the bypass flow coefficient depends strongly on the refueling pool configuration, specifically the relative locations of the inlet and outlet for the alternative cooling flow. Comparing configurations A and B, it is seen that a factor of ten difference in alternate cooling flow rate has a minor impact on the bypass coefficient when the coolant flow interacts with the natural convection plume from the reactor core, whereas configurations with the inlet and outlet on the same side of the pool have significant differences in the bypass coefficient.. A similar result is seen when comparing configurations C and D, although computations indicate substantial entrainment of the pool water by the alternate cooling flow occurs for large flow rates in configuration C.
RAI A6.
Are ebypass (in the equations) and f (Table A-I) the same coefficient?
Response
The terms eyp,,, B3, and 3/4ypa5s as used in WCAP-1 5872 Rev 00 are the same coefficient. For consistency, the term "qypS," is used to define the bypass coefficient in these RAI responses and in any revisions made to WCAP-15872.
RA A7.
Please show the derivation of the values of,,,, and bypass used in the results shown in Figs. A-3 and A-4 for Case 2 and Case 3.
Response
The mixing and bypass coefficients are defined in Appendix A and derived as shown in Appendix D. However, for the results shown in Figure A-3 and Figure A-4, these coefficients were assumed well mixed, i.e., ems =1.0 and all alternate heat removal flow fully mixed with the natural convection flow from the core, ebypass = 0.0. In Appendix A, Case 2 represents full SDC flow plus alternate cooling flow; Case 3 represents only alternate cooling flow. (Note that sample Cases 1 - 4 in Appendix A are not the same as test Cases 1 - 4 listed in Appendices B, C and D.)
Appendix B: Comparison of Predictions with Test Data RAI B1.
Fig. B-1 is confusing. Under the alternate cooling alignment do you have a separate spent fuel pool (SFP) pump and heat exchanger for both the refueling pool and the SFP, or do these represent separate alignments? Please indicate the complete flow paths of fluid associated both with the refueling pool and core, and the SFP. In your figure, how and when do you get flow "from the refueling pool to the spent fuel pool?"
Response
Figure B-1 illustrates the specific alternate heat removal alignment at CCNPP. The figure describes the capability to align a "spare" spent fuel pool cooling train to cool the refueling pool while a second train remains aligned to the site's spent fuel pool.
The complete alternate heat removal process fluid flow path at Calvert Cliffs is where heat from the core exchanges with the refueling pool through natural convection, then forced flow from the
WCAP-1 5872 iRAI Responses Page 7 pool through a train of the spent fuel pool cooling system (pump, heat exchanger and piping).
The discharge from this alternate cooling alignment flow path is then returned to the refueling pool.,,.
The statement in Section B.1, uThe suction from the refueling pool to the spent fuel pool cooling line is through a drain in the bottom of the refueling pool, at the side of the pool opposite the inlet point," refers to the alternate heat removal alignment at Calvert Cliffs. In this alignment, major components (pump, heat exchanger, piping) from one train of the spent fuel pool cooling system are cross-connected to suction and discharge fittings in the Calvert Cliffs refueling pool.
A direct exchange of coolant between the spent fuel pool and the refueling pool is not relied upon to support the alternate heat removal process.
The actual configuration of the alternate cooling alignment implemented at other plants may vary depending upon the available plant equipment capabilities. Refer also to Figure 1 of WCAP-1 5872 which illustrates a generic shutdown cooling decay heat removal system, and to Figure 2 which illustrates the decay heat removal flow path when using the Alternate Heat Removal process. A different alternate heat removal alignment may be selected by other plants, depending on the heat removal loops available to cool the refueling pool. The alternate heat removal process does not envision altering the traditional method of cooling the spent fuel pool.
RAI B2.
In Table B-I, what is 'SW?"
Response
The term USW" refers to Service Water. This term is included in an updated acronym list for WCAP-1 5872.
RA B3.
You report average temperatures. These are averaged over what?
Response
Temperatures given in Table B-1 are averaged over times recorded for the tests.
RAI B4.
Table B-2, B-3 and B-4 report time in days, hours and minutes respectively. Also, the
-figures use two different time scales. 'Pleaste resubmit for review all tables and figures based on one time scale. (If there is a specific reason, such as clarifying a relationship, state so.)
Response
Time scales in Tables B-2, B-3 and B-4 are' expressed in terms of clock time, total elapsed time and time in 'days' after shutdown in order to expediently illustrate'a particular result. For
example, an everit having a duration' of min'utes is'not easily illustrated if expressed using a time-scale of days. Total elapsed timeis used t6'compare measured 'and predicted values,'
while days after' shutdown is the importantfparamrter for tracking the point at which changes such'as initiation and securing of shutdown'cooling,-head removal, initiation and securing of alternate cooling, and return to shutdown cooling occur.
WCAP-1 5872 RAI Responses Page 8 RAI B5.
Please give a table describing the physical conditions associated with each of the five cases. That is, for each of the five cases, give the initial and final time and the corresponding initial, final and average shutdown cooling and SFP temperatures (computed and measured), flows and core decay powers. For average values, give the explicit method by which they were computed.
Response
The physical conditions, time, and temperatures associated with the test cases listed in Table B-3 are given below. The reactor is in Mode 6 with the refueling pool fully flooded for Cases 2 - 5.
Case 1: SDC flow reduced while the reactor vessel head is removed; Case 2: SDC flow restored to value prior to head removal.
Case 3: AHR flow initiated,- SDC flow continued.
Case 4: SDC flow secured, AHR cooling only.
Case 5: SDC flow restored, AHR flow secured.
DAS AnalysisTime Terperature(°F)
Case Event Date and DS
..-.. L_______L.............
C E ent D(Days Start-End-SDC-in SDC-AHR-in.
AHR-RFP Time hr hr.out out 1
03/23/01, 04:30 5.75 0
11 73.58 102.90 NA NA NR 2
0323/01, 15:30 6.21 11 285 92.01 102.73 NA NA NR 3
04/03/01, 22:00 17.62 285 298 99.00 103.30 92.03 96.78 101.30 4
04/04/01,13:00 18.21 298 348 NA NA 78.16 92.95 99.26 5
04/07/01, 13:40 20.49 348 375 96.72 102.89 NA NA NR The purpose of Table B-3 is to document measured temperatures with their corresponding times.
Table B-4 lists the analysis times used for predictions corresponding to Cases 1 - 4 in Table B-3.
Time histories of the data for each of these cases are documented in Figures B-3 (SDC flow and temperatures), B-4 (AHR temperatures and flow rate) and Figure B-5 (RFP temperatures).
Predictions for Cases 2, 3 and 4 are shown in Figure B-6.
Appendix C: Comparison of CCNPP Unit 2 Test Data with Computational Fluid Dynamics (CFD) Predictions RAI C1.
For these calculations, please show the natural circulation flow path in the core region. Is that how is the core cooled?
Response
Decay heat is transferred from the core to the refueling pool through natural circulation. While-this heat removal is not dependent on the direction of the circulatory pattern through the core, good agreement between fluid temperatures based on the CFD analysis and the Calvert Cliffs test data at the reactor vessel flange elevation was predicted assuming a natural circulation path with down-flow in the center of the core and up-flow at the core periphery. This flow pattern was found to best represent the post-refueled conditions, where fresh fuel occupies a checkerboard arrangement in the core center, which existed during the alternate heat removal test phase at Calvert Cliffs.
WCAP-15872 RAI Responses Page 9 RAI C2.
The resultsfrorm the lumped parameter model (core, flow rate) are'dependent on d,,*
and ebypass, 'Thesetwo coefficients are determined via a CFD calculation. How does the' CFD 'calculation of 4igx and Ebypass differ from the CFD calculation in this' appendix?
Response
The CFD evaluations of Appendices C and D are based on parameters for the CCNPP refueling pool/reactor cavity geometry. Appendix C contains 'an evaluation of the specific flow and temperature'fields' associated with the' CCNPP flow 'alignment (simrilar to Configuration A of Appendix'D) at the initial and boundary conditions associated with the CCNPP Unit 2 test data.'
Appendix D contains the evaluation of the heat removal capabilities of permissible flow alignments and includes the evaluation of the mixing and bypasses coefficients for each alignment. As such, Appendix C represents a validation of the CFD computations'and the application of the mixing and bypass coefficients from Appendix D into the lumped 'parameter model which computes the core flow rate. Small changes in the initial and boundary conditions associated with the CCNPP2 test data, including a lower alternate cooling flow rate, do not' substantially alter the computed mixing'and bypass coefficients presented in Appendix D. Thus, the methods used to calculate the mixing and by"pass coefficients given in Appendix C are the same as those for the remainder of WCAP-15872.'
RAI C3. Is the CFD calculation in this appendix a steady-statd calculation?'
Response
The'CFD computations are steady state-based on the observation that the refueling pool is in a quasi-steady state condition for the purposes' of Appendix C.
RAI C4.
The'data appear to show no temperature gradient at the flange level, while the CFD
'calculation shows a distinct gradieht.' Your proffered 'explanation in paragraph eight is
'not clear.' Please provide a drawing indicating the flows and temperatures that support your argument.
Response
The application of a rectangular Cartesian grid to represent a cylindrical reactor vessel cavity" accentuates local temperature 'differences when comparing CFD temperature predictions with thermocouple 'data at the flange level. Pool temperature'data from CCNPP Unit 2 were taken in four strings starting just aboveihe reactor vessel flange;'these thermocouples-are radially near,'
but not necessarily in,' the rising thermal plume. 'The corner cells just above the reactor cavity and within the'computed thermal plume' are the closest representations in the CFD model to these thermocouple locations. As a consequence, the average temperature of the four' computational cells would be expected to be higher than the average of the test data. This rationale is confirmed in Table C-I where the average CFD temperature'exceeds the data by" only 3.60F at the 44-ft elevation.' The'average temperatures'are much-closer at the mid-pool and pool-surface elevations since the CFD model can better represent the global turbulent diffusion and convective diffusion.
The horizontal temperature gradients at the flange level are more pronounced as a consequence of the rectangular grid approximation to the circular reactor cavity opening at the flange. The rectangular grid causes a more'pronounced channeling of pool currents around the
WCAP-1 5872 RAI Responses Page 10 flange opening than might be expected from currents around a circular flange opening. As shown in Figure C-6, the channeling of current is evident as longer velocity vectors passing one side of the flange opening in the velocity distribution of the horizontal plane just above the flange. In turn, the enhanced channeling promotes a somewhat larger temperature difference between opposite sides of the flange, as evident in the temperature distribution in the horizontal plane just above the flange and seen in Figure C-3.
Both of these effects are localized at the reactor cavity opening. The turbulent thermal diffusion and convective diffusion of the thermal plume into the bulk refueling pool are otherwise well represented and indicated by the good agreement in temperatures at higher elevations.
RAI C5.
How is the difference in mixing, described in C4 above, taken into account in your estimate of dmb,?
Response
The pool mixing coefficient is defined in terms of pool average temperatures. The impact of localized currents is accurately represented in the global mixing although the localized temperature results may not precisely correlate with the CCNPP data in the flange area.
Appendix D: Evaluation of Alternative Heat Removal Alignments The key to your methodology is the estimation and validation of the mixing and bypass coefficients. Please define your terminology clearly; indicate the type of calculation and the results precisely so that the comparisons are clear.
RAI D1.
Please describe the simplified one-dimensional computational model and its relation to the two-dimensional computational fluid dynamics model. How does it differ from the one-dimensional model discussed in Appendix A?- When you say "computational fluid dynamics model" (without the adjective uone-dimensional' in D.3, what are you referring to - A 3D model? Figures D-3 through D-6 give 2D results. So, how are you treating the situation in Figure D-2?
Response
The mixing and bypass coefficients reflect three-dimensional effects into the one-dimensional analysis, shown in Appendix A, for natural circulation flow rates and refueling pool temperatures. The mixing coefficient is a measure of the uniformity of the refueling pool temperature, while the bypass coefficient, represented schematically in Figure D-2, is an indicator of the flow rate from the alternate cooling alignment that bypasses the natural circulation plume from the core.
Predictions of refueling pool temperatures using the three-dimensional CFD model, described in Appendix C, are then used to calculate both mixing and bypass coefficients. These values are then used in the one-dimensional model. Final values are selected based on agreement between the one-dimensional predictions, the CFD analysis results, and the data.
RA D2.
You say "The one-dimensional evaluations based on perfect mixing... are summarized in Table D-2, " yet you show bypass flows that are not one-dimensional.
WCAP-1 5872 RAI Responses Page 11 In Table D-3 what is your point? The table indicates that the mixing coefficient is spatially dependent (given at different locations). How can that be when it is defined on page D3 in terms of pool average temperatures?
Response
The statement referring to perfect mixing (ejX = 1.0) and all alternate cooling flow passing over, the core (cYpass = 0.0) are assumptions used in the one-dimensional scoping analysis shown in Appendix A.
The mixing coefficient is defined in Appendix D in terms of the initial pool temperature and the pool average temperatures from one-dimensional and CFD computations.' A number of CFD cases were run to evaluate the range of the mixing coefficient since the mixing coefficient influences the rate of pool temperature change in the one-dimensional model. Results for the.,
case selected to best represent the mixing coefficient are reported in Table D-3.- In that table, a one-dimensional model with the mixing coefficient set equal to 1.0 establishes a time, 886 minutes, when the pool average temperature reaches saturation. -By interpolation, the equivalent time predicted by the CFD model to achieve a pool average temperature of saturation is 851 minutes, which reasonably agrees with the one-dimensional prediction. Results of the CFD model at other times, which correspond to reaching the saturation temperature at an elevation representing the core exit, the free surface, and the bottom of the refueling pool are also shown in the table. For these locations, the mixing coefficient was found to be 0.88, 0.98, and 1.03, respectively, from which a representative value of 0.90 was selected for use in one-dimensional analyses.
Appendix E: CCNPP Specific Evaluation of Conditions for Alternate Decay Heat Removal
-in Mode 6 I
RAI El.
In section E. 1, your discussion of Figure E-3 is inconsistent with the text. The text indicates that the initial refueling pool temperature is 75 F, while the value in the figure at t = 0 is 90'F.
Response
The initial temperature of the refueling pool was taken as 900F in the analysis. Page E3 of Appendix E has been corrected to be consistent with Figure E-3.
RA E2.
Where are the data that reflect the last statement on page E3? What is the basis for the "expected" high and low limits?
Response
The statement concerning expected high and low limits is not needed and has been deleted.
WCAP-15872 RAI Responses Page 12 RAI E3.
What is the purpose of footnote I on page E4? Where and what is Reference 6.1?
Response
The footnote was meant to reference standard methods used to determine heat exchanger effectiveness and outlet temperatures. This footnote and reference are not needed and have been deleted.
RAI E4.
In the paragraph Limiting THS vs. TAS on page E4, Figure E-5 does not show a family of curves. - What do you mean by a 90OF heat sink temperature when the refueling pool inlet temperature in also 90OF?
ResDonse:
The statement has been corrected to refer to Figure E-4, not E-5. Figure E-5 is a cross-plot of the data shown on Figure E-4. The heat sink statement refers to the temperature of the heat sink for heat removal, which in this case is the inlet temperature to the spent fuel pool heat exchanger.
RAI E5.
The time scale of minutes on the x-axis of the figures is inappropriate for the phenomena described on the figure. Please submit a revised figure that uses a consistent time scale (see Appendix B, Question B4).
Response
The different time scales reflects differences in the information represented in the figures. For example' Figures E-1, E-3, E-5 and E-7 reflect the influence on the days after shutdown on the value of decay heat assumed in the subsequent analyses. Figures E-2, E-4 and E-6, reflect the time, the order of magnitude being minutes, for the refueling pool temperature to reach a new steady state value after the noted changes in operating conditions. Thus, the time scales selected are appropriate to the information represented and do not warrant changes to the report.
RAI E6.
What is Reference 6.4 which gives the CFD analysis that establishes the maximum fluid velocity for the computation of the force on the fuel assembly?
Response
The reference was for the CFD analysis and is not needed. This reference has been deleted.
RAI E7.
How do you get from a one-dimensional model the flow rate in the core for a lateral velocity of 0.22ft/sec in the refueling pool? The precision is astounding!
Response
The velocities were taken from the CFD analysis and are representative of the magnitude of lateral velocities that could be expected. The text has been revised to state that the velocity is approximately 0.2 ft/sec.
U. S. Nuclear Regulatory Commission WOG-03-608 November 18, 2003 WCAP-15872, "Use of Alternate Decay Heat Removal in Mode 6 Refueling" Changed Pages
WCAP-1 5872-NP Use of Alternate Decay Heat Removal in Mode 6 Refueling List of Changed Pages Report Body:
- Pg ii (List of Acronyms)
Appendix A
- Pgs A2 - A8
- PgA13
- Pg A14 Appendix B
- Pgs B2 - B5
- Pg B10
- Pg B11 Appendix C
- Pgs C2 - C4 Appendix D
- Pgs D2-D6
- Pg D8 Appendix E
- Pgs E2 - E6
- PgsEl4-E16
List of ACRONYMS AHR....
Alternate Heat Removal CCNPP....
Calvert Cliffs Nuclear Power Plant CCW......
Component Cooling Water CDF....
Core Damage Frequency CFD....
Computational Fluid Dynamics DAS....
Days after Shutdown DHR....
Emergency Operating Procedure FPCS....
Fuel Pool Cooling System gpm....
Gallons per Minute HPSI....
High Pressure Safety Injection HX....
Heat Exchanger LCO....
Limiting Condition for Operation LOCA....
Loss of Coolant Accident LPSI......
Low Pressure Safety Injection MEEL..
Minimum Essential Equipment List NPSH....
Net Positive Suction Head NRC......
Nuclear Regulatory Commission RCS....
Reactor Coolant System RFP....
Refueling Pool RV.....
Re.acto'rR Vessel SDC....
Shutdown',Coolirig Systemn SFP.
Spent Fuel Pool SW....
Service Water Tamb Containment 'ambient temperature TAS.....
Time after Shutdown THSHSHS H.....
eat Sink Temperature TRM..
Technical Requirements Manual TS..
Technical Specifications
................ ;..'......Upper Guide Structure E E...
.................. Ratio of mixed RFPmmass t6 total REP mass EbAya bypass......... Ratio of AHR'flow bypassing core to total AHR flow November 2003 WCAP-1 5872, Rev 01 Page ii
.,1 APPENDIX A ALGORITHM FOR NATURAL CONVECTION BETWEEN
. CORE AND REFUELING POOL i
4 S
9
.< 61 ' ' '
November 2003 WCAP-15872, R01 Page Al of A14
APPENDIX A ALGORITHM FOR NATURAL CONVECTION BETWEEN CORE AND REFUELING POOL A.1 MODEL In Modes 5 and 6, forced convection provided by the shutdown cooling system is used to transport decay heat from the reactor core to the ultimate heat sink. In the absence of shutdown cooling flow during Mode 6 refueling operations with the refueling pool flooded, the reactor core decay heat is transported by natural circulation into the refueling pool water. The buoyancy force causing this natural circulation is driven by the density difference between the, cooler, denser, fluid in the refueling pool and the hotter, less dense, flow through the core. Interaction between the natural circulation flow through the core with the circulating currents in the refueling pool results in a variation of fluid temperatures and velocities within the refueling pool. Properties controlling the natural convection from the reactor to the refueling pool as well as natural convection and evaporation from the free surface are primarily functions of temperature.
The model described in this Appendix has been developed to calculate the natural convection flow between the core and refueling pool that occurs during Mode 6 refueling conditions when the shutdown cooling system is not in operation. This model divides the reactor vessel and refueling pool into a series of control volumes that describe the upper guide structure, core and refueling pool, Figure A-1. Mass flow rates and inlet temperatures are prescribed for the alternate heat removal flow path. Conservation of mass, momentum and energy for these control volumes are solved to predict the mass flow rate between the reactor vessel and refueling pool. Temperatures are calculated for the refueling pool, the flow into and out of the pool, and the flow rate through the alternate heat removal alignment. The model also considers the heat lost at the pool surface due to natural convection and evaporation from the free surface. Dependent and independent variables are defined in Table A-1.
The flows into and out of the control volumes are assumed one-dimensional. However, the natural convection flow being driven by the temperature difference between the core and refueling pool is allowed to vary with time. This heat storage is accounted for in the mass of coolant in the pool as well as the coolant and structural masses for the upper guide structure and the core. Without active heat removal provided by the alternate heat removal alignment, the temperature of the refueling pool would continue to increase until the boiling point is reached. With active heat removal, steady state temperatures are eventually reached for core, pool and outlet flow.
The geometry of the pool results in regions where the cooler fluid near the bottom of the pool does not fully mix with the core flow. This is modeled by defining a mixing coefficient, E,44 which is defined as the ratio of the effective mass of coolant in the refueling pool that mixes with the reactor vessel flow to the total mass of coolant in the refueling pool. Therefore, the mixing coefficient is the effective fraction of the pool water that participates in the core-to-pool flow process.
Emx "' MAlrefueiing pool The effective mass is determined by engineering judgment from the temperature and velocity distributions in the computational fluid dynamics model used to address the November 2003 WCAP-1 5872, ROI Page A2 of A14
refueling pool. The flows between the core and the fraction of the mass of fluid in the refueling pool, defined by the value £,,'which participates in the fluid transfer, are assumed to be fully mixed. Analysis shows the majority of the refueling pool inventory mixes with the natural convection flow from the core, resulting in a value for the mixing '
coefficient of about 0.90. '
In addition, not all the flow from the alternate-cooling path mixes with the natural convection driven flow from the core. This is accounted f6r byidefining a bypass fractiorh, defined as the ratio 'of the flow bypassing the core plume flow to the total pumped alternate heat removal flow, or:
Ebypass = mbypass flow / rAHR flow The value of the bypass coefficient is determined'from the computational fluid dynamics model., Analysis shows that essentially all of the alternate heat removal cooling flow injected into the refueling pobl mixes with the natural circulation plume above'the vessel, resulting ina bypass coefficient close to zero.
A.2 Algorithm The solution algorithm solves for the core exit temperature, T4, and the pool temperature, T6, for each time step, tW.1 = tn + At. The algorithm iterates on core exit temperature at each time step, with the following basic steps;
.. Select Qcore Assume T4 (_ Tout of core) > T6 (- Tpoo, Tinto core) = T Solve for p (T4)
Solve for ihcore Solve for new T4 (- Tnew out of core, new)
' Iterate'until Tcore new minus' Tcore oldis within the convergence criteria (0.1 0F)
Solve for new pool temperature, T6.new This algorithm, Figure A-2, is evaluated for each time step until a steady state or until the saturation temperature is reached, T4 = Tcore new = Tsat.
Values for the independent variables for CCNPP Units 1 or 2 are listed in Table A-2.
tanipie cases for four combinations of shutdown cooling and -a"teriate h'eatareroVal flow are listed in Table A-3. The upper guide structure has been removed in all cases.
Thus, values of structural mass and loss factors for the upper guide structure are taken as zero. Selection of values for the time step (15 seconds) and convergence criteria (0.100F) are based on a convergence study. Output parameters are defined in Table A-4. Sample results are shown in Table A-5.
Results for average refueling pool temperatures and natural circulation flow are shown in Figures A-3 and A-4. Case 1 represents normal alignment for active shutdown cooling.
In Ca'se 2, iboth'the alternate heat removal and 'shutdowncooli'g are active, iesulting in the lowestIv'alues of refueling pool'temrperature. Case 3 is for alternate heat removal alone.,The refueling pool temperatures reraalnbelow.sattion i l
e 4 w'ith both shutdown cooling and alternate heat removal flow secured, represents the condition for no' adtive' c'oing of the' refueling pool.
November 2003 WCAP-15872, ROl
..1 ;
I.~
Page A3 of A14
I I
With'shutdown. cooling flo6w in operation, the flow rate between the core and;refueling.
pool due6to natural circulatl6n is approximately 2000 gpm; with shutdown cooling flow, secured this natural circulation flow rate increases to approximately 4000 gpm'as shown on Figure A-4. These flow rates are driven by the temperature difference between the core and refueling pool. Cases 1, and 2, where shutdown cooling is active, have lower flows and lower temperature differences. Case 4, with no forced cooling flow, has the largest natural circulation flow through the core and the largest values of temperature difference.
November 2003 WCAP-15872, R01 Page A4 of A14
Table A-.
Definition of Variables ANALYSIS
.'DEFINITION
.QBASIC UNITS T1 UGS inlet temperature
- __TPne_
OF T2 Core inlet temperature
- p..w.
-p-'F T4 Core outlet temperature OF Ts UGS outlet temperature
'_,_'-T__e_,
°F
,.T 6.
Refueling pool temperature' Tp OF X.T7
,SFP flow inlet temperature Ton OF T
8 SFP flow outlet temperature OF T9 SDC flow inlet temperature '.Tsddn OF T10 SDC flow outlet temperature Tcnew' OF
, Mcore Core flow due to natural convection mcore Ibm/sec m7 SFP flow rate mpot Ibm/sec Msdc SDC flow rate msdc Ibm/sec Mass of water & metal in the core Mt,Mcm Ibm M16 Mass of water & metal in the UGS
- Mugsf Mugsm Ibm -
M6 Mass of water in the refueling pool Mp Ibm Pemb Containment pressure P-psia Tgmmb' Containmeni tempbrature.-t° Ow__l Decay heat C-btu/sec Qs.rf Heat loss at pool surface due to Qpnc+pevap btu/sec natural convection and evaporation At Time step'
.. A t.
sec Alternate heat removal cooling flow Eb Note I bypass coefficient Refueling pool mixing coefficient Note 2 Notes:
I No bypass =- 0 all bypassed = I 2
N-mixing = 0,-complete mixing I
I I 1, ;
November 2003 I
WCAP-15872, R01 Page A5 of Al 4
Table A-2 Input for CCNPP Units I & 2 COMPONENT PARAMETER SYMBOL VALUE UNITS NOTES Containment Pressure Pamb 14.7 psia 1
Ambient Temp Tamb 75' F
1 Refueling Pool Water mass Mf1 3084708 Ibmr Water depth LI 23 ft Free surface Asurf 1750 ft2 Wetted Perimeter P.et 190 ft Equiv Length Le 9.21 ft 2
SFP flow rate
__fP 9__
gpm Case dependent SFP inlet Temp T-fi OF Case dependent Mixing Coefficient 0 <' e 1
0.90 Natural Conv
>0
> 0 yes Evaporation
>0
>0 = yes Bypass7 O0< tass 0
Coefficient Initial Temp.
Tfpl T 2 Tamb OF Case Dependent UGS Metal Mass Mm2 0
Ibm 3
Water Mass Mf2 0
-Ibm 3
Flow Area A2 0.9565 ft2 3
Height L2 13.375 ft Loss Factor.
K2 21 73 ft 3, 5 Core Metal Mass Mm3 303800 Ibm Water Mass Mf3 46488 Ibm Flow Area A3 53.46 ft2 Height L3 12.917 ft Loss Factor K3 12.328 SDC flow rate Qsc O.3000 gpm Cadp et SDC inlet Temp Tscin 75 0F Case dependent Thermal Load Q/Q 020%
4, 8 Calculations Time Step At
<15 Seconds 6
Maximum Time tmax Minutes Temp error AT
< 0.5
°F 6
Print NPRT
> 0 print output Print per time
< Nmax 7
Plot NPLT
> 0 to txt file Plot per time
<Nmax 7
See NOTES next page.
November 2003 WCAP-15872, R01 Page, A6 of A14
Notes for Table A-2 Input for CCNPP Units 1 & 2 Table A-2 Notes 1
Pmib used in calculation of subcooled boiling temperature 2
Leq = Asxage / Wetted Perimeter 3
UGS removed; Loss factor & Area included for information only 4
- QO = 2754 x 106 watts-thermal = 9399 x 106 btu/hr 5
K= 6787 when based on core flow area of 53.46 ft2 6
Number of time steps = tajx
- 60 / A t = Nmax---
7 Recommended values based on convergence study (0.100F) 8 0.20% selected for test cases.
Table A-3 Sample Case Input Listing SDC SFP RFP Containment' Cases Qddc. gPml TedCmn, 'F Qsfp, gPm TSwn, IF Tsfp',° F Ta-b.
0F l
-Case 1 3000 75 0 ° l
NA l
75
-l-75 Case2 3000 75
--.1200 75 75 75 Case 3 0
.NA 1200 75 75
- 75
~ Case 4
~
-~~ 0 -
75 0
75 75 75 (sdc Tsddnl Qsfp Tsfpin Tambi Shutdown Cooling System flow Shutdown Cooling System inlet temperature Spent Fuel Pool flow Spent Fuel Pool inlet temperature Initial Refueling Pool temperature Containment ambient temperature November 2003 WCAP-15872, ROI Page A7 of A14
Table A-4 Output Parameters PARAMETERS VARA DEFINITION As functions of Time Tcore (OF)
T4 Core outlet Temperature Tpool ('F)
To Refueling Pool Temperature Core (gpm)
Natural Circulation Flow Rate Tpumpo (IF)
T8 Spent Fuel Pool Outlet Temperature Tavgc (OF)
T4+ Ts 0.50 x (Tcore-in + Tcore-out)
Error Q(-)
At the last time step Core Outlet Temperature (IF)
T4 Core Outlet Temperature Subcooled Boiling Temperature (OF)
T4sc Tsat = f (Pressure at top of core)
Pool Bulk Temperature (OF)
T6 Refueling Pool Temperature Surface Heat Loss[NC+Evap] (Btu)
Qsurf Surface Heat Loss Surface Natural Convection (Btu)
Qnc Heat Loss due to Natural Convection Evaporation (Ibm)
Mevap Amount of Surface Evaporation Surface Evaporation (Btu)
Qevap Heat Loss due to Evaporation Spent Fuel Pool Pump Heat Load (Btu)
Qsfp.
Total SFP Heat Removal SDC Heat Load (Btu)
Q1 Total SDC Heat Removal Core Convection Heat Load (Btu)
Q2 Convection Heat Transfer Core-RFP' Qcoretotal = Qcstored + Qsdctot + Qcnctot (Btu)
Q3 Total Heat Transfer from Core' Opooltotal = Qpstored + Qsfpumptot + Qnctotal(Btu)
Q4 Total Heat Transfer from the RFP' Decay Heat = Qd
- Time (Btu)
Q5 Total Heat Generation from Core' Heat Balance: (Qcore - Qdecay) / Qdecay (%)-
Change in Core Heat = Decay Heat Heat Balance: (Qpool - Qnacore) / Qnccore (%)-
Change in Heat to RFP = Core Convection Time Constant (minutes) fp Time Constant for RFP Heat Up2 Note 1: Following heat balances must be satisfied: SDC + Core Convection: Q1 + Q2 = Q5; Core convection = Decay Heat, Q2 = 04.
Note2: Time constant =
RrPmnaturai circuIaton A Variables in the analysis, see Table A-i.
November 2003 WCAP-15872, R01 Page A8 of A14
ft Figure A-3 Sample Cases: Average Refueling Pool Temperature
- I I-;,
I
..I 165 155 145 135 125 T(0F) 115 105 95 85 75 Case I
--Case 2
6 Case 3
--Case 4
0 100 200 300 400 500 600 700 Time (minutes) 800 November 2003 WCAP-15872, R01 Page A13 of A14
Figure A-4 Sample Cases: Natural Circulation Flow between Core and Refueling Pool 5000 4500 4000 3500 3000 f 2500 CY 2000 1500 1000 500 0
-4Case 1 Case 2 Case 3 Case 4 0
100 200 300 400 500 600 Time (minutes) 700 800 November 2003 WCAP-15872, R01 Page A14 of A14
APPENDIX B COMPARISON OF PREDICTIONS WITH TEST DATA
.. I I I
.. : r I.
November 2003 WCAP-15872, R01 Page Bi of B14
APPENDIX B COMPARISON OF PREDICTIONS WITH TEST DATA B.1 Test Data Validation of the model developed in Appendix A is based on a comparison with data recorded at CCNPP Unit 2 during the March 2001 refueling outage. Under limited conditions, CCNPP units are permitted to use an alternate refueling pool cooling system during Mode 6 with the refueling pool flooded and with shutdown cooling secured. In this alternate cooling alignment a train of the spent fuel pool cooling system is manually aligned so that the' spent fuel pool cooling pump takes suction from the refueling pool.
After passing through the spent fuel pool cooling heat exchanger, the flow is directed back into the refueling pool. This flow is directed into the refueling pool through piping near the bottom of the pool (Figure B-1). The suction from the refueling pool to the spent fuel pool cooling line is through a drain in the bottom of the refueling pool, at the side of the pool opposite the inlet point.
Test data were recorded for two days during which the alternate pool cooling alignment was in use. Fluid temperatures in the refueling pool where recorded by thermocouples located at the reactor flange level, at mid-level in the pool, and close to the pool surface.
Approximate locations of these thermocouples are noted in Figure B-2. Additional parameters recorded are listed in Table B-1.
The approximate time for initiation and securing of both shutdown cooling and refueling pool flows are listed in Table B-2. Measurements of flow rates and temperatures versus time, in days after shutdown (DAS) are shown in Figures B-3, B-4 and B-5.
Figure B-3 shows shutdown cooling flow rates, plus inlet (into the cold leg) and outlet (out of the hot leg) temperatures versus time. Note the reduction in shutdown cooling flow from 3000 gpm to 1500 gpm at about 6 days into the shutdown to facilitate flooding the refueling pool, and detensioning and removing the head. Once the head is removed, natural convection between the core and refueling pool starts. Thus predictions are only valid after the head is removed'.
Figure B-4 gives '
go4f~le l alternate heat removal cooling system flow rates and temperatures into the refueling pool and out of the refueling pool. These data were taken about 17 to 20 days into the outage. As shown in this figure, both the shutdown cooling system and the 0pent-fu4el-peo alt'erniateheat:
reimo Val cooling system are activated near the start and end of the time period. This is to assure that the switchover into and out of the alternate alignment is successful.
Figure B-5 shows the average refueling pool temperatures at each of the three elevations. As expected, the fluid temperatures are highest at the reactor flange and decrease toward the pool surface.
F~~~~~~~
1 Heat removal via the SDC indicates a decrease of about 17% after removal of the head. This reduction is due to natural circulation flow between the core and refueling pool.
November 2003 WCAP-15872, R01 Page B2 of B14
B.2 Comparison of Predictions with Test Data Switching from the conventional shutdown cooling decay heat removal, both before and after the head is removed, followed by switching to the alternate decay heat removal are represented for the following cases:
Case 1:
Reduce shutdown cooling flow for vessel head removal.
Case 2:
Restore full shutdown cooling flow.
Ca'se 3:' initiate alternate heat removal cooling flow, continue shutdown cooling flow.
Case 4:
Secoure shutdowii cooling flow," continue alternate heat ireioval coolng flow.
Case5: 'Secure alternate heat removal flow, restore shutdown cooling flow.
Temperatures and flow rates for these cases are list6d in'Tables 'B-3 and 1B-4.
Predictions for shutdown cooling and spent fuel pool (alternate heat removal) outlet temperatures versus time, Figures B-6 and B-7, compare well with outage data. Time-averaged values of the shutdown cooling, spent fuel pool cooling (alternate heat removal) and refueling pool temperatures are compared in Table B-5. With the exception of Case 1, the'predicted shutdown cooling anrd refueling pool temperatures
'are in reasonable agreement as 'shown
'n Figure B-8. The 10% difference in SDC predictions and data are related to uncertainties in decay heat values and initial refueling pool temperatures at the time the head is removed.
A comparison of predicted and measured average refueling pool temperatures is shown in Figure B-7. Experimental values are taken as the numerical average of the readings shown in Figure B-5. Agreement is good except for the initial portion where variations in the data are due to'operator controlled changes in the SDC'flow to reach a'aicceptable operating-point.
Table B-1 Measured & Calculated Parameters based on CCNPP2 Data MEASURED CALCULATED PARAM ETERDESCRIPTION HEAT BALANCES SFPin TSFPI T-into the RFP SFPout TSFPO T-out of the RFP SFPflow MRFP Flow into the RFP QRFP = MRFPCP (TRFPO - TRFPI)
SWin TSw1 T-into SW-HX SWout Tswo T-out of SW-HX SWflow
___SW Flow thru SW-HX Qsw = MswCp (Tswo - Tsw1)
SDCout.
TSDC1 T-out of RV hot leg SDCin TSDCO T-into RV cold leg SDCflow MSDC Flow in SDC QSDC = MSDCCP (TSDCO - TSDCI)
November 2003 WCAP-1 5872, ROI Page B3 of B14
Table B-2 Event Time Related to CNNP2 Outage EVENT DATE TIME' (hr:rmin)
JAS(-Da-.)
QDECAY (btulhr)
DECAY HEATb (%)
Refueling Pool Cooling Load MODE 5 03/16/2001 23:55 0.000 2.264E+08 2.409%.-
Full Core SDC start 03/19/2001-09:01 2.000 --- -
4.630E+07-0.493%
HEAD removed 03/23/2001 04:30-
- 5.750 3.089E+07 0.330%
RFP start 04/03/2001 22:00 17.625, 1.320E+07 0.140%-
125 Assy SDC secured 04/04/2001 13:00 18.208, 1.303E+07 0.139%
AHR 'steady-state 04/05/2001 00:00 18.715' 1.290E+07
'. 0.137%
SDC ritoredc !04/07/2001 13:40 20.486
' 1.248E+07
' 0.133%
AHR end data 04/07/2001 14:42 20.722 1.238E+07 0.132%
RFP secured 04/08/2001 05:00 21.358 1.223E+07 0.130%
- a. Approximate times
- b. QO = 9.399E+09 btuthr
- c. End of steady state period Table B-3
~
Average Values Based on Experimental Data CASE TIME (hours)
=
TEMPERATURE (°F)
FLOW (gpm)
QP'
(
9399E+09 l___ Range Tstart Tend SDC In SDC out SFP In SFP out RFP Qsdc Qsfp Qdecay
%decay 1
Reduce SDC flow . 0 11 73.58 102.90 NA NA -
NR-1521.87.
0 2.034E+07 0.216%
2 Full SDC flow..
11 285 92.01 102.73 NA,
- NA NR 3071.05 -
0 1i.572E+07 0.167%
3 SDC -+AHRflow 285 298 99.00 103.30 92.03 96.78 101.30 3088.98 1195.65 1.313E+07 0.140%
4 AHR flow only 298 348 NA
- *'NA 78.16 92.95 99.26 0
1194.23 1.276E+07 0.136%
5 SDC,-AHR=0 348 375 96.72 102.89 NA NA NR 0
0 1.231E+07 0.131%
NA = Not Applicable NR = Not Recorded November 2003 WCAP-15872, R01 Page B4 of B14
Table B-4 Input for Algorithm Cases CASE TIME (minutes)
TEMPERATURE (°F)
FLOW (gpm) l Decay Heat J
Range ATime:
Time SDC in SFP in RFPi Qsdc Qsfp
(%)
Reduce SDC~flow 660 660 73.58 NA 75 1521.87 0
0.216%
2 Full SDC flow 16440 17100 92.01 NA 92.3' 3071.05 0
0.167%
3 SDC -+ AHR flow 780 17880 99.00 92.03 102.03' 3088.98 1195.65 0.140%
4 AHR flow only 3000 20880 NA 78.16 100.25' 0
1194.23 0.136%
5 ISDC,;AHR-= 0 1680__ I NA NA NA.
. 99.52_
0 0
0.136%
- 1. RFP average temperature taken from prior Case.
- 2. Time when RFP temperature reaches 2120F.
3...SFP inQsfp-referto AHR flow.
Table B-5 Comparison between Predictions and CCNPP2 Data for Average Temperatures Tsdc-outlet (°F)
Tsfp-outlet (°F)
Trfp-average (°F)
CASE CALC DATA CALC l.
DATA CALC DATA 1
91.98 102.84 NA NA*
NA NR 2
102.14 102.74 NA NA NA NR 3
104.39 99.00 100.59 96.78.
100.59 101.07 4
1 NA No Data 99.60 92.95 99.60 99.23 Ref:
Figure A (next page)
Figure B Figure C November 2003 WCAP-15872, R01 Page B5 of B14
Figure B-4 CCNPP Unit 2 Outage Tests: Spent Fuel Pool Flow Rate and Temperature versus Time E
a.
a.aL 1500.00 -_
1450.00 -_
1400.00 -_
1350.00 -_
1300.00 -
1250.00 -_
1200.00 -
1150.00 17.0
-120.00 110.00
' 100.00
- 90.00 0
I--
80.00
- 70.00 60.00 21.5 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 DAS (days)
~~.
I
.I November 2003 WCAP-15872, ROI Page B10 of Bl4
Figure B-5 CCNPP Unit 2 Outage Tests: Average Refueling Pool Temperatures versus Time 110 108 106 104
- c. 102 r-100 98 96 el=44
- - - eI=53
-eI=62 94 4-17.000 17.500 18.000 18.500 19.000 19.500 20.000 20.500.
21.000 DAS (days)
November 2003 WCAP-15872, ROI Page B11 of Bl4
~APPENDIX C-COMPARISON OF CCNPP UNIT 2 TEST DATA
.WITH.
COMPUTATIONAL FLUID DYNAMICS PREDICTIONS
~.
C
.I I I..
November 2003 WCAP-15872, R01 Page Cl of C10
APPENDIX C COMPARISON OF DATA WITH COMPUTATIONAL FLUID DYNAMICS PREDICTIONS This Appendix provides a comparison of CCNPP Unit 2 test data with predictions based on a computational fluid dynamic model of the refueling pool.
The geometry of the CFD model (Figure C-1) for the refueling water pool preserves the volumes of the refueling pool. Core flow rate and heat generation rate, from the lumped parameter model, are applied as boundary conditions.
Computational fluid dynamics computations based on a decay heat generation rate of 0.0946% predict a temperature difference between the refueling pool outlet and inlet of 15.00F, approximately 1 % above an average of the measured temperature difference of 14.820F (Table C-1).
Refueling pool temperature data at different elevations above the reactor vessel flange
'indicates that the pool temperature decreases with elevation. This suggests that the hot plume from the core thermally mixes with the colder refueling pool water and cools as it rises to the top of the pool.
Computational fluid dynamics predictions of the refueling pool water temperatures at locations corresponding to the measurement points compare favorably with the measured temperatures, as shown in Table C-1. In general, computational fluid dynamics predictions are higher than measured values. The highest differences occur in the SE-NE quadrants (0° to 1800) due to a non-uniform distribution of the inlet (in the 1800 to 2700 quadrants) to outlet (in the 2700 to 3600 quadrants) over the reactor. (Refer to Figure B-2 for quadrant orientation.) Measurements being lower than predictions indicate a higher degree of mixing and a more uniform distribution of inlet flow than predicted by the computational fluid dynamics model.
Features of thermal hydraulic mixing in the refueling water pool are depicted in Figures C-2 and C-3, which show the temperature distribution of the thermal plume from the core in a vertical plane and through a series of horizontal planes. (Note: the temperature scale shown is in degrees Rankine; subtract 460 to obtain Fahrenheit). These temperature distributions illustrate the thermal plume rising above the core and then being transported downstream toward the drain. In Figure C-3, the bias of flow around the core to the SW and NW result in the lower temperatures predicted for those two locations.
Predicted values of fluid temperatures decrease with rising elevation above the vessel and are higher on the downstream side (angles of 450 and 1350) than the upstream side (angles of 2250 and 3150). These differences are due to spent fg pnei heating'of-thii alternatd cooling flow as it crosses the core and mixing not being as complete in the CFD model as in the refueling pool. Predicted values are, on the average, about 3%
higher than measurements.
In general, the thermal plume is predicted to rapidly mix in the vertical direction while the cavity of the pool that is associated with the incoming core flow remains cold. Some of this cold mass does short-circuit the core to the cavity on the drain side. Within the drain November 2003 WCAP-1 5872, ROI Page C2 of CIO
cavity, the pool temperature is warmer and reduces to the drain temperature at 940F. At the surface of the pool, the maximum temperature is 103OF and the volume weighted average temperature is 940F. As noted from Figure C-4, the test data shows temperatures are more uniform in the vertical direction than those predicted by'the computational fluid dynamics model.
Circulation due to the thermal plume results in the predicted values for fluid velocity in the vertical plane (Figure C-5) and horizontal plane (Figure 0-6) being the highest in the
-region above the reactor flange. These velocity profiles above the core are an indication of the strong mixing and recirculation occurring in that region. CFD results -show the higliest fluid velocity in the natural circulatior plume to be approximiately 0.2 feet/second.
Tho highoet fluidd velenities, of about 0. ft6eood, oc.curin the. o onfoabmo'6 tho G coro.
Since the thermal plumo is turbulont, there is alc6 an additional fRuctuating volocity '
compOnent of approximatoly 0.02 fno econd. The moan and tho fi9ctuti un".Welcitoc!
result in a maximum '.'locity of 0.22 fcocond. Tho'largect vertical velocitio, of about' 0:2 4ft'6oond, rccur niF the rofueling pool drain.
November 2003 WCAP-1 5872, RO I Page 03 of C10
Table C-1 Comparison of Thermocouple Data with Computational Fluid Dynamics Predictions Location TEMPERATURES (F Direction NE SE SW NW Average Altemate Cooling Flow Angle4 135° 225° 3150
.IN OUT Elevation DATA I CFD DATA l CFD DATA CFD DATA CFD DATA CFD DATA CFD DATA l.CFD 44-ft 101.42 108.71 101.31 107.68 101.93 100.78 100.74 102.95 101.35 105.03 53-ft 99.49 101.34 98.30 1101.70 97.50 93.73 9.1 95.0 5 9.0 7.96 78.82 78.57 93.39
-93.57 62-ft 98.77 100.69 98.57 101.75 97.72 97.55 9
8.77 98.51 98.46 99.63 Refer to Figure B-2 for quadrant orientation.
November 2003 WCAP-15872, R01 Page C4 of C10
APPENDIX D EVALUATION OF ALTERNATIVE HEAT REMOVAL ALIGNMENTS
- i. -
I
i, - "
I r
.I I :
iI I.-
. ), -
...1,
,... I i
i
': i,
I I I
. I.
I
I I
I 1 : ;
.
1 t
f I I
.. I ::.
I I.
November 2003 WCAP-15872, R01 Page Dl of D12
APPENDIX D EVALUATION OF ALTERNATIVE HEAT REMOVAL ALIGNMENTS The objective of this Appendix is to document predictions of fluid temperature at a value of 0.315% decay heat, seven days after reactor shutdown, considering four alternatives for location of the inlet and suction. In all cases the analysis is based on the parameters for the CCNPP refueling pool /reactor cavity geometry.
D.1 REACTOR CAVITY CONFIGURATIONS The four configurations to be analyzed are described in Table D-1, shown schematically in Figure D-1. The selection of configurations were chosen to represent a variety of possible conditions that may exist and that none of these configurations represent the exact configuration of the CCNPP units when they are aligned for alternate heat removal. The analyzed configurations are identified as follows:
Configuration A:
Alternate Piping: Suction across core.
Configuration B: Alternate Piping: Suction same side.
Configuration C: TransferTube: Suction across core.
Configuration D: TransferTube: Suction on same side.
The influence of the different flow paths on the one-dimensional model is manifested through the mixing and bypass coefficients. To evaluate these coefficients, computational fluid dynamics niodels are prepared for each of the configurations. The following are the assumptions used for these one-dimensional evaluations:
Containment temperature = 1 000F Inlet temperature = 850F Decay heat = 0.315% (seven days after shutdown)
SpeAt44~e~f~I 6terI..ate heat removal flows; 200 gpm and 2000 gpm D.2 ONE-DIMENSIONAL MODELING The one-dimensional computational fluid dynamics model uses mixing and bypass coefficients to incorporate the mixing of the core flow with the reactor cavity fluid and the alternate cooling flows. The mixing coefficient, ei,,,,, accounts for the portion of the reactor cavity fluid that does not mix (remains close to the initial pool temperature) with the core flow. The bypass coefficient, bbass accounts for thetpent fuei pool aitermate hfiat ?emn6v'al flow that does not mix (remains close to the inlet temperature) with the core exit flow. The bypass flow path is shown schematically for Configuration A in Figure D-2.
The one-dimensional model assumes values for the mixing and bypass coefficients.
This model is independent of locations of the inlet and drain for the alternate cooling paths. Thus, computational fluid dynamics models of the various arrangements must be used to re-evaluate these coefficients for use, in what is an iterative procedure, in the next one-dimensional model calculations.
The relationship between the definition of the mixing coefficient and temperatures in the computational fluid dynamics model isdrived4outsid6 this text shown below. The mixing November 2003 WCAP-15872, R01 Page D2 of D12
coefficient is expressed in terms of the pool average temperatures for the one-.
dimensional and computational fluid dynamics analyses as, Em'k-...A,;fmIM 6-=(TCFD;-T j/(T-where T. is the initial pool temperature.
The bypass coefficient rep'resents the fraction of the 'peit -fuolpool alteriate heat rem ovaIcooling flow that wdoes not mix with the flow out of the core. Conservation of energy for the mixed and unmixed flows then gives the outlet temperature for this flow' as,
- (1-
,)m.,cpT.+
ms fpcpT p
--ln fcTp 0
The bypass coefficient is solved for as,
.~~~~~~~~~~~~~~~~~R
.)IT where 7' is the 'pool average temperature for either the one'dimensional or'.
computational fluid dynamics models. For the computation fluid dynamics model, the alteriat6eheat remroval cooing fiow'tha't 'oes 'ot mix with the'.core pium'e flow'.is,
.'rby TpoC) /(Tc,; *TPi )(
where' po and pi refer to the refueling pool outlet (drain) and inlet temperatures.'
Tho'onodimonsi~ionai evaluatione'bae~doi 'n;p6e'rctmixirig,':that Ts,'those 'with'i mixirig coofficient of 1.0 and aiip a'coofficiont of Q.0,'aro summarized in'Tbl6 D'2.'AII rocults use tho assumeied valu-of 0.315 for the dociy het. :Caso 1 and.'Cse 5 use flo'..rato6 lof 2000i-nd 200 gpmfo'r'tho'&6pnet fdol pool flow.! ~Tho c'rio flovw rates' of 8563 gpm'and '10108:g'pm,-podsetiv'oly,'.for theso'c'3ss are'bounldar' con'ditione fr.tho comnputtational -fluid d,-i'na'mic's',eviluatio'n'.'
Re'sulte fo'r the'-~o'ri'dim'e~sonai 'e'vau tibnith 'cp rt'fu'oi'po'i'io 'ol~f '20 O0pfor'the influenco o'f 'm'ixin'g anb'r'diapassco'efficlonts' is cho'w In Table'D '3.-Tho".'ariation".'.ith mixing'.coeffici~en.t,.Ca'se6':1.-1,"is 'nogligigble, whero'ae, the '.ariitio~n w.ith b'ypa'ss c'o'effjcie'ntc, 'Cases-5 7 is signhifica'nt:t Caise 8 variee"'both the'emixing 'c'oefficioint (0:50) a'nd the bypass 'coefficient (0.'50) 'o'siulting in'po'ol tempe'raturos' cio60 t6"1500F-the
'!^-.!
~
V ve-4 v
setg !
resdlt~'f~rCases6.an C.
e6hvih0.0~xn cefiiit, ar td h
'o6n'e. iimen'siona calculation'.wvith' themixin.'coe'icient-equal to one'and.the" ba coeff icien t e ' Ual to: ze'ro Is use id to dete'rmine the core flow' rate tha't is 'appided "'s a
- botun'daryic'nditionto th'e 'computation'al fluid dy amic evaluation of the alternate cooling fio'w
-aignments.
iF1or an assumed 0.315°%-.decay heat level,'the predicted core-flow 'rates
.for alternative Wooing flow'rates of 200.ahnd 20'00'gphl iarei10408 a'nd 8563 gpm, respectively November 2003 WCAP-15872, ROI Page D3 of D12
D.3 COMPUTATIONAL FLUID DYNAMICS MODEL EVALUATION'*
The mixing coefficient is meant to represent the influence of a non-uniform distribution of fluid temperature on the transient behavior of the fluid in the reactor cavity. The bypass coefficient is intended to represent the alternate cooling flow that may not transport heat from the core. It is assumed that the mixing and bypass coefficients are independent.
Thus, the mixing coefficient may be determined based on no transport flow into or out of the cavity. However, evaluation of the bypass coefficient is dependent on the flow rate and the pool configuration. ResiIti'of this evaluation, based on the fol6loing core flow.
rates corresponding to the one-dirnensional flow rates for perfect mixing, ea,'= 1, and no bypass, ebbS, = 0, cases are shown in Tables D-2 and D-3.
AHR flow rate SF= = 200 gpm Qcore = 10408 gpm AHR flow rate S1-P = 2000 gpm Qcore = 8563 gpm Inlet flow to the refueling water pool from the spent kuel poo alternate heat removal flow path colihng' syster' may be introduced from either a pipe at the upper surface of the pool or from a low-level inlet'th tra'nsfr d'uct low in one of the pool cavities. Cooling flow may exit the pool through one drain which may be in either pool cavity. Since the CCNPP pool is nearly symmetric, four configurations bound the general possibilities for inlet and exit flow locations. For each inlet location, the drain location may be in the same cavity'or in the cavity on the opposite side of the reactor vessel. With the inlet and exit in the same cavity, the 4p t-fei p"'el alternate heat removal cooling flow may short circuit the reactor vessel. With the inlet and exit on opposite sides of the reactor vessel, the spentIt Ne'peoe aIteriate'6 h'eat re"mov'al cooling flow must at least pass by the open vessel. The slight non-symmetry of the refueling water pool, principally due to different depths of the cavities and the off-center location of inlets, should not be significant to these computations.. These configurations are defined in Table D-1 and shown schematically in Figure D-1.
Results of this analysis, in the form of temperature profiles for the four configurations at the 2000 gpm alternate hea r'oal it4`-l pet flow rate, are shown in Figures D-3 to D-6.
D.4 BYPASS AND MIXING COEFFICIENTS Results for the bypass coefficients are documented in Table D-2. For Configuration A, the flow that crosses the core and mixes with the flow from the core is reflected in a' value of the bypass coefficient of about ze'R for both high and low flo'x' piont hljpoI aernate heat removal'flw rates. In contrast, for Configuration B the majority of the alternate heat remova flow goes directly to the drain, which is reflected in values of the bypass coefficients is close to unity.
Configurations C and D represents the arrangement where the alternative cooling path entitrs tffe refueling oofr ow.
aiWteele°Vchia' througlj the fuel transfer tube. In Configuration C the flow is forced up and over the core. Computational fluid dynamics ana ysi in i' ctVti i-n >
dthf1 tha atcoss the core results in the inlet flow into the core being closer to the spent fuel pool co'ling "ystem flow.' temperature of 850F rather than the refueling, poolaverage temperature assumed in the one-dimensional analysis. Forjthis case' the resulting temperature of the flow out of the core is predicted to be lower than the average pool temperature. a'nd 'ro"lts in asl'o of'tlid bypass November 2003 WCAP-15872, R01 Page D4 of D12
coofficient groater than one' Th same isobsorved,'
t4o a loeoFr oxtont, at the In~'o cpnt fuelpolflow, ^.ho<< To to byp ass te ^fficit I+ cloe to one.
t I.
I In the alternate cooling mode, decay heat is transported by natural circulation from the core into the 'rfuelinig pool. A bypass coefficient having a value greater than zero denotes that a portion of the alternate cooling flow bypasses the natural circulation thermal plume above the core. For example, the alternate heat removal cooling inflow in Con'Configuration 'B enters near the 'pool surface with the drain at the bottom of the refueling pool on thfe ssan side as the inlet. The temperater ' distribution for this configuratiohn,
- shdwn in'Fiiguep D4,4 siggests tiat moist of the aitermiate coo ing infl ow'onymixes with refueling 6ool water the inlet side ofhe' th'ie exits the pooI without significant mixing with the-core theim...al plu'rbe.'- Thus,`a bypass'-coe'fficient greater'than'zero represents a reductio irn the alt e'ative-cooling flow that interacts to remove decay heat from the core thermal plume and results in a higherpool average temperature, Tcfd, as shown In Table D'2.
Tho'obypiasscoo'ffisiont rioprosents tho fSow that is' rnot bffoctivoly used 'to romo'.o'tho
-docayheat, transportod by n~atdral circuls'tion from'tho core, from th'ector ca'ity.
Vlsut to ee than onib zro denote a-ortio-n of
.h.
Inflo;' from hon altern te coolingl path, in this case tho&pont fuel 'pool coolinig s'stem, is not usedt t4 roo4 this hoat Fot oxamplo in Conf;^gurtior A tho^ sent fo;+l pb^l alter-nate hoe+ -ram^val 6doling.
iniflowv ontor at the free curfac" 'ith the drain at tho bottom on the oppocitb sido of the cor'o 'Horo the temprarrture distribultion, assonn F~igure D-3;,^ ch2ews.^ that a Porio ;^of the inflow forM a F6recirculation at the bottOm of the reactor civity en the earn side as the inflow. Tho, te
-_n _-,'6*;r~toof thbfow ha
^9 otTh it the^...+
niatural '6frcu^laion; fromghercre reduos h'f that can effectively remove the decay heat, resulting in :
hg-tlt
.et p
I -
In Configuration C the flow enters through 'a lo'-le've Ilobation wsuclh as thle&trarisfert'ub e and.
exits thugh a drain at the bottom on the opposite side of the core. The
-- -temperature 'distribution in Figure D-5 shows a portion of the flow entering from the low-le'el inleft 44Aiii-be remains near the bottom of the 'cavity; but most of the flow goes up and over the core. This cooler flow mixes directly with the natural circulation from the' core before being drawn to the outlet. The higher rate of coolerflow passing by the core
- inlet results in lower values of core outlet temperatures. This may be reflected in the one-dimensiohal model by a value'of the bypass coefficient les-s`
thsatn'zzero, which is
-equivalent to increasing the mass flow entrainment of Ejeit fuel-peosalte rnate"heat rempval flow in the one-dimensional model.
Results for Configuration D, where the drain is on the same side as the eFb low level inlet, are similar to Configuration B. In' both cases, the spent-fob poolternate hoeat'.rernov-l cooling'flow short-circuits directly to the reactor cavity drain. The thermal effects of this short-circuiting are manifested in low temperatures in the path between the.
spen~t fti3 pool cooling stoe'm alte'rnate heat removl inlet and outlet and relatively higher temperatures elsewhere '(Figures D-4 and D-6)..'Configu ratins B and D remove heat from the vicinity of the reactor core through the action of recirculation currents and turbulent diffusion in the active cavity of the refueling water pool that are produced by the natural circulation plume resulting from the core heat generation.
November 2003 WCAP-15872, R01 Page D5 of D12
Values of the mixing coefficients are all close to unity. Based on this data, a value of 0.90, close to the minimum value of 0.88, was selected is rbcommondod for use with the one-dimensional model.
Table D-1 Refueling Water Pool Cooling Configurations Configuration Inlet Location Drain Location A
Pipe flow directed downward in upper Drain in floor of cavity on opposite side corner of pool of reactor vessel B
Pipe flow directed downward in upper Drain in floor of cavity on same side of comer of pool (same as A).
reactor vessel C
Transfer tube 4.(low elevation in the Drain in floor of cavity on opposite side pool) of reactor vessel D
Transfer tube duA (low elevation in the Drain in floor of cavity on same side of pool) reactor vessel Refer to Figure D-1 for a schematic of these configurations.
Note that these configurations do not represent the specific configuration of the pool at the CCNPP Units.
Table D-2 CCNPR Unit 2 Bypass Coefficients Based on CFD Analysis Config A
B C
D Flow(gpm) 2000 200 2000 200 2000 200 2000 200 Ti (OF) 85 85 85 85 85 i 85 85 85 Tsfp (OF) 85 85 85 85 85 85 85 85 Tmax (F) 138.3 396.8*
170.1 444.6*
134.4 397.9*
193.8 421.8*
Tcfd (0F) 114.3 371.4*
149 418.9*
104.9 369.6*
156 393*
To (0F) 115 383.8*
110.3 377.3*
115 383*
108.9 382.7*
Tsurf (OF) 115.5 i 376.3*
149.3 423.9*
110.7 376.7*
166 400.8*
I;..-"0.024 R0.43 00 00:508 0.047.
0.663 0;033
- i.e., 200 GPM is insufficient to prevent boiling for the decay heat used.
Table D-3 CCNPiP~ Unit 2 Mixing Coefficient Based on CFD Analysis Analysis Computational Fluid Dynamics 1-D.
Location' l
Core Exit r
Surface lPo61 Bottom Uniform Time (min) 750 833 874 886 Tsaturation (OF)
- 215 214 212 214 Taverage (OF) 197 209.4 215.5 212
_MIX_0.88 0.98 1.03 1-.
November 2003 WCAP-1 5872, RO1 Page D6 of D12
Figure D-2 Flow Paths for Bypass Flow Bypass Flow N
N N
/NLE7 CORE 1
\\
DRAIN N \\ \\
\\
\\
\\
t\\
Alternate Heat Removal Flow i \\
\\
November 2003 WCAP-15872, Ro1 Page D8 of D12
4 APPENDIX E CCNPP SPECIFIC EVALUATION OF CONDITIONS FOR ALTERNATE DECAY HEAT REMOVAL IN MODE 6
.4
.. ~ ~~~
.I-I -i I,
. I Noeme 203WA-82 O
-November 2003 WCAP-15872, R01 Page El of El 8
APPENDIX E CCNPP SPECIFIC PARAMETRIC EVALUATION OF CONDITIONS FOR ALTERNATE DECAY HEAT REMOVAL IN MODE 6 This appendix presents the results of several evaluations testing the sensitivity of various parameters on performance of normal decay heat removal and the alternate heat removal alignment for the CCNPP Units. Limits on the use of the alternate alignment for the removal of decay heat, while removing one or both trains of shutdown cooling from service, and the possibility of moving fuel, all depend on the temperatures in the refueling pool. At CCNPP Units I and 2 the alternate heat removal alignment is accomplished with a train of the spent fuel pool cooling system (FPCS). Hard piped connections from the FPCS are available to establish dedicated coolant circulation with the refueling pool.
Per Section 4.0 of the body of this report, the limits on the use of the alternate alignment for the removal of decay heat, while removing one or both trains of shutdown cooling from service, and moving fuel, depend on the temperatures in the refueling pool. The refueling pool temperature, in turn, depends on the ability of the aligned cooling systems to reject heat to the ultimate heat sink. This heat rejection is a function of the performance of the heat exchangers used to reject the heat and the heat sink temperature (THs). Limits on refueling pool temperatures are discussed in Section 4.1.
Steps in determining refueling pool temperatures for values of heat sink temperatures are outlined in Sectiorr 4.2.
Removal of one or both trains of shutdown cooling from service will be limited by the fluid temperature reaching some value that represents the margin between the selected value and the core becoming uncovered. For the CCNPP Units the operating limit has been set at a value of 140 0F, coincident with the limiting temperature for the spent fuel pool.
Temperatures and time to reach specific temperature limits can be predicted based on the one-dimensional, lumped parameter algorithm developed to predict refueling pool and core outlet temperatures versus time as described in Section 2.2. The algorithm contains provisions for the usual Mode 6 shutdown cooling alignment as well as an alternate alignment utilizing spent fuel pool cooling.
Fuel assembly movement during refueling operations can depend on local fluid velocities due to the thermal convection between the core and refueling pool and subsequent mixing with the local pool fluid circulation. The limiting fluid velocity is such that it is below values at which the fuel assembly can become tilted and difficult to insert into the core.
November 2003 WCAP-15872, R01 Page E2 of E18
Changes in the Technical Specifications, discussed in Section 5.0, needed to support implementation of alternative heat removal and evaluation of limiting conditions for operation to meet these r6quirements, include:.
Conditions under which the alternate heat'removal alignment may be used.
Limiting conditions are a function of decay heat as a function of days after shutdown, refueling pool temperature as a function of heat sink temperature, flow rate and inlet temperature for the alternate heat removal alignment (Section E.1).
X Requirements for removing the shutdown cooling system from service.
Time limits for interrupting the alternate heat removal flow.
Limiting conditions for operation are based on time to reach a limiting value of.
refueling pool temperature (Section E.4).
Fuel movements allowed when using alternate heat removal alignment.
Limiting conditions for operation are based on fluid velocities 'induced by natural, convection, in the region above the core, and the influence of the'resulting fluid forces on alignment of the fuel assembly with its core location (Section E.5).
The following outlines the procedures and methodology for determining'the above conditions. Values presented are based on calculations for CCNPP Unit 2.
E.1.
RFP Temperatures vs. Inlet Temperature With the head off, at assumed times after shutdown, the' refueling pool (RFP) f'6fiperature is a function of the decay heat, shutdown cooling (fiterfnte heat rem oval) flow and inlet temperature, and refueling'pool initial temperature.
TRFP = f(Qdey, msDc, TSDCO, TRFPM)
-Decay Heat:' Based on assumed values of time after shutdown,-values 'of decay heat' are obtained from the decay heat curve, assumed for conservatism, for a full core,jfor example Figure E-1.
Conventional Decay Heat Removal: Values are calculated for refueling pool temperature versus time, at different values of days after shutdown, and constant values of shutdown cooling system flow (3000 gpm), inlet temperature (900F) and initial refueling pool temperature (900 F), for example in Figure E-2. -The values of steady state temperatures, in this case at a constant value of TSDCI of 900F, are shown in Figure E-3.
E.2.
RFP Temperatures vs,. Heat Sink Temperature Alternate Decay Heat Removal: Values of the spent fuel pool temperature, TSFPi, are a function of the performance characteristics of the heat exchanger(s) used to remove November 2003 WCAP-15872, R01 Page E3 of E18
heat from the'refueling pool and the final (ultimate) heat sink. Thus, upon switching to' the alternate cooling alignment, at assumed times after shutdown, the refueling pool temperatures are calculated as a function of the decay heat, spent fuel pool (alteinate heat removal) cooling system flow rate and inlet temperature and steady state temperature of the refueling pool at the time of the switch-over:
TRFP = f(Qdecay. msFp, THS, TRFPI)
Predicted values of refueling pool temperatures versus time, are shown in Figure E-4 and steady state values in Figure E-5. Both figures are based on a heat exchanger effectiveness and flow, multiplied by specific heat ratio, Cr, of one, so that TSDCi = THS.
As with conventional heat removal the calculation is repeated for values representing the expected high and lower limits of the heat sink temperature, THS.
Limiting THS vs. TAS: Repeated calculations for RFP temperatures result in a family of curves such as shown in Figuriie E-4. Refueling pool equilibrium temperatures will decrease with lower values of heat sink temperature and increase with higher values of heat sink temperatures. Selection of a limiting value of refueling pool temperature results in the time after shutdown that the alternate heat removal alignment can be aligned and not exceeds this limit. For a limiting value of 1400F, based on Figure E-5, the limiting condition of operation for entering alternate heat removal alignment with a 900F heat sink temperature is about 5 days.
E.3.
Time to Reach Limiting Temperatures Results in Figure E-5 show that, for CCNPP Unit 2, the alternate heat removal alignment is sufficient to keep the refueling pool temperatures below the values of both the selected limiting value of 140OF and saturation (212 0F) temperatures. However, the time to reach saturation decreases the higher the steady state values of the refueling pool temperatures. With loss of alternate heat removal alignment, refueling pool temperature versus time, for a constant value of heat sink temperatures, is a function of the decay heat and temperature of the pool at the time alternate heat removal cooling is lost; TRFP = f(QdeCay. TRFPI)
Refueling pool temperature as a function of time, at constant values of days after shutdown is shown in Figure E-6. Parametric relationships between the time, At, to reach, either the limit on SFP temperature of 140OF or a value of 2120F, are shown in Figure E-7.
At= f(DAS, Qdecay. mSFP, mSDCPTSFPI, TSDCI,TRFPi)
The outage schedule calls for initiation of alternate heat removal alignment from 15 - 25 days into the shutdown, for a duration of 5 days. Times to reach limits on temperature during this operating period are as follows:
November 2003 WCAP-1 5872, RO1 Page E4 of El8
Time into Time to Reach Temperature.
Shutdown Limits (hours)
(Days) 2-15 1.67
.13.3 25 6
16.7 E.4.
Fuel Movement Fuel movement depends on fluid velocities due to the thermal convection between the core and refueling pool and subsequent mixing with the pool circulation flow.. The fuel assembly can become tilted and difficult to insert into the core when these local fluid velocity values are below limits. 'The limiting condition can be determined as follows.
Tilt An-ile: With reference to Figure E-8, the horizontal component of drag force on a fuel assembly titled from vertical by an angle 0 is given by:
FD =~C pV 2 AP COS 6 where CD is the drag coefficient, p the fluid density in units (Ibm/ft3), V the average velocity over the length of the bundle, in units (ftlsec), Ap the projected surface area (bundle height times width) of the bundle, in units (ft2).
Upon equating'the drag force, the component of weight in the same direction as the drag component, CD
-CDPV22A cos = MAg -* sin O 2
where MFA is the mass,' in units (Ibm),' of the fuel assembly and g the acceleration'of gravity (32ft/sec2). The tilt angle is then given by, 0- tan 1
This relationship is shown in Figure E-9.
Evaluation: The maximum value of 2.4 for the drag coefficient, is based on the assumption of the fuel assembly being modeled as an infinite beam, with a square cross section rotated 450 to the flow. The density, based on a refueling pool temperature of 1 000F, is 62.4 Ibm/ft3. Tilt angle as a function of fluid velocity is shown in Figure E-1 0.
While the angles are small, the limiting value will depend on plant specific experience with insertion of fuel assemblies during refueling.
Fluid Velocity: Based on the CFD analysis h a
gA the maximum velocity occurs in the thermal plume region above the core. Furthermore, the velocities tend to November 2003 WCAP-15872, R01 Page E5 of El8
I be higher the closer to the top of the vessel. Based on the assumption that the velocities are proportional to the natural convection flow, QNC, from the vessel, the velocity is, Vmax QNcIAFLOW Based on the model in Figure E-1 1, the flow area corresponds to a circular flow area of about 6 feet in diameter, which corresponds to about half the flow area at the top of the vessel.
Predictions based on the one-dimensional model, of flow rate due to natural convection between the core and refueling pool, of 2900 gpm result in a velocity of about O.2 feet per second. Review of the CFD analysis indicated that the velocities in both the vertical and radial directions are about equal.
Limiting Conditions: Values of tilt angle as a function of time after shut down is calculated as follows.
The natural convection flow rates between the core and the refueling pool is a function of the decay heat, Figure E-1. Corresponding flow rates as a function of days-after-shutdown, DAS, are shown in Figure E-12.
Based on these flow rates, maximum velocity as a function of DAS is calculated from, Vm.
= QNc/AFLOW where AFLow is taken as 29 ft2.
Corresponding values of tilt angle can then be computed based on the following relationship.
0 = tan-'[cDpV2AP]
L 2MFA j
Limiting values of tilt angle will depend on plant specific experience with fuel assembly insertion. Values of velocities and corresponding tilt angles are shown in Figure E-13.
The allowable window for initiation of AHR should be based on temperature limits and then determine if the tilt angles are sufficiently small so as not to result in problems with insertion of fuel assemblies.
November 2003 WCAP-15872, R01 Page E6 of E18
II FDRAC MASS OF FUEL ASSEMBLY Figure E-8
..Limiting Conditions for. Moving Fuel 90 75 60 00 aI*45 a.2
<30 15 I
0 0.001 0.01 0.1 I
10 100 (Drag ForceIFA Mass)
Figure E-9 a Function of the Ratio of Drag Force to Fuel Assembly Mass Tilt Angle as November 2003 WCAP-15872, R01 Page E14 of E18 I
100.00 10.00 U,~
0 0) a
- 0
.6 CD 1.00 0.10 0.01 0.00 Z
0.01 0.1 Velocity (fps) 10 Figure E-10 Tilt Angle as a Function of Fluid Velocity November 2003 WCAP-15872, R01 Page El5 of El8
Figure E-11 Flow Areas for Natural Convection Flow November 2003 WCAP-15872, ROI Page El6 of El8
WCAP-1 5872-NP End of Change Pages