ML20072C014
| ML20072C014 | |
| Person / Time | |
|---|---|
| Site: | 05200004 |
| Issue date: | 07/31/1994 |
| From: | Dowlati R, Ishii M, Revankar S PURDUE UNIV., WEST LAFAYETTE, IN |
| To: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| Shared Package | |
| ML20072C010 | List: |
| References | |
| CON-NRC-04-93-049, CON-NRC-4-93-49 PU-NE-94-1, NUDOCS 9408170024 | |
| Download: ML20072C014 (257) | |
Text
.ua NUREG/CR
- . . PU-NE 94/1
= = = = = = = = = = = - = = = = = = = = = = = = =
Scientific Design of Purdue University Multi-dimensionalIntegral Test Assembly (PUMA) for GE SBWR
__=========================__
Manuscript completed: February .1994 Revised Manuscript Completed: July 1994 Prepared by:
M. Ishii, S.T. Revankar, R. Dowlati, M.L. Bertodano, I. Babelli, I W. Wang, H. Pokharna. V. H. Ransom and R. Viskanta School of Nuclear Engineering Purdue University West Lafayette, IN 47907 i
Prepared For l Division of system Research Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Contract No. NRC-04-93-049 1
\
9408370024 940809 PDR ADOCK 05200004 B PDR
l l
ABSTRACT l
The scientific design of the Purdue University Muld-dimensional Integral Test Assembly (PUMA) has been carried out under the " Confirmatory Integral System Testing for the GF SBWR Design" contract sponsored by the U.S. Nuclear Regulatory Commission. The design is based on a three-level scaling method developed for this task. The first level of scaling, the integral scaling, is based on a well-established approach obtained from the application of integral response function to the thermal-hydraulic system. This level ensures that the steady- l state as well as dynamic characteristics of the thermal-hydraulic loops will be similar between 1 SBWR and PUMA. The second scaling level addresses the boundary flow of mass and energy between components, insuring flow and inventory similarity. The third scaling level focuses on the similarity of key local phenomena governed by constitudve relations. The PUMA facility has 1/4 height and 1/100 area ratio scaling. This corresponds to the volume scale of 1/400. The ;
PUMA power scaling based on the integral scaling is 1/200. The present scaling method j predicts that PUMA dme scale will be one-half that of SBWR. The system pressure for PUMA 1 is full scale, therefore, a prototypic pressure is maintained. However, the facility is designed to I simulate relatively low pressure thermal hydraulic phenomena' which occur after the initial !
blowdown depn:ssmizadon phase. PUMA is designed to operate at and below 1.03 MPa (150 psi), which allows it to simulate the prototypic SBWR accident conditions below 1.03 MPa (150 psi). The facility includes models for all the major components of SBWR safety and non-safety systems that are important to the transient response to postulated LOCA and other transients.
The PUMA component designs and detailed instrumentations are presented in this report.
l I
l i
1 l
l
This page is Blank I
.- l 1
1 4
1 1
- l 1
l i
I i
i l
l 1
1 i
1 I
l l
l 1
1 I
\
l l
4
TABLE OF CONTENTS -
ABSTRACT...................................................................................................................i LIST OF FI G URES . .... .... .. .. .... ..... . .. .. . . .. .. .. . . .... . . .. .. ..'.'. .. . .... . .. . . .. .. . . . . .. .. .
LIST OF TAB LES .. .. .. .... .... .... .. .. .. . .. . . .. .. . . . .. . . . .. .. .. . . . . . .. . . . . . .. .. .. .. . . . . . . . . . .. . . . . . . .. . . . .. . . . . .. . . . . . .. x N OM ENC LATURE . . . .... .. .. ... .. .. .. .. . . .. . . .. .. .. . . . . . . .. .. . .. .. . . . . .. . . .. .. . . . . . . .. . . .. .. . . .. .. . .. .. . .. ... . . .. ..xi v
- 1. INTROD UCTI ON ...... . . .. .. .... . .... .... . .. ... ... . ..... . .. . . .. . .. . . . . .. .. ......................I-1
- 2. O B JECTI VES .. .. . . .. . . . .. .. .. .. .. .. .. .. . . . . . . . . .. ... .. .. . .. . . . . .. .. . . . . .. . . . . . . . . . . . .. . . . . . . . .. . . .. . . .. . . . . 2 - 1
- 3. PROGRAM REQUIREMENTS... . . .. ....... ... ... ..... . ... . . . .... . ... . . . . . . . . . . . . . . .. .. .... 3 - 1
- 4. DESIG N B ASIS FOR INTEGRAL FACILITY ........... ..... .. ../... .... . . . . ......... .. ...... ..... 4- 1 4.1 Systems Characteristics ( Safety) ... . .. . .. ... . .. ..... ... ..... . . . .... ............. .. . ... . .......... 4- 1 4.1.1 Reactor Pressure Vessel (RPV).. ........... . ... .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .... 4- 1 4.1.2 C on tain m e n t S yste m .... ...... .. . . .... . .... .. .. . ...... .. ...... .. . .. . ...... . . . . . . . . . .. ...............:4-2 4.1.3 M ai n S te am Li nes ....... . .. .. .. .. . .... . .. .. .......... . .... ........ .. .. .. .. .. .. . .. .. . . . . . . .... . . . ... .. ...... 4-3 4.1.4 Automatic Depressurization System, (ADS).. .. . .... . ....... .. ..... ........ ...... . ....... .. . 4-4 4.1.5 Gravity Driven Core Cooling System (GDCS)................. . ...........................4-4 4.1.6 Passive Containment Cooling System (PCCS) ................ .. .. ..................... ....... ...... . 4-6 4.1.7 Isolation Condenser System (ICS) . .... ... ........... . ..... ...... ....... . . . . . . . . . . . . ... 4-7 4.1.8 Control Rod Drive (CRD) System and Reactor Water Cleanup / Shutdown Cooling (RWCU/SDC) System .._..... ......... . .. .......... . . . . ... .... . ... . . . . . .. .. . . 4- 8 4.1.9 Spectrum of Postulated Breaks....... ...... . . .. .... ... ..... . . . . . . . . ........49 4.2 Governing Processes and Phenomena. .. .... .... .. . . . . .... . . .... . . .... . . .. . .. . . . 4 -9 4.2.1 Processes Following a LOCA or Transient... .... . . . . . . . . . . . . . . .. . ... . . . 4-9 i 4.2.2 Phenomena Accompanying LOCA or Transients. . . . . .. . . . . . . .. ..4 10 4.3 Identification of Important Phenomena.... .. ... ... ... .. ... . . . . ... .... 4- 13
- 5. SC ALING APPRO ACH...... .. . . ... ....... . .... ...... . .. . . . . . . . . . . . . . . . . . . . 5- 1
5.1 General Considera tion ...... .. .. . ............ ............. ................ ........ .... . .. .. . ... .. .. . . . . . .. ..... 5- 1 5.2 Scalin g App roac h . ........ ... ...... .. ... ... .... .. . .. . ..... .... . .. . ... .. .... .. .. . . . . . . . . . . . . . ..5-2 5.3 G l ob al S c alin g .. .. .. . . .. .. .. .. . .. . .. . . . . . . .. .. . . . .. . . . . . . . . . . . .. .. .. . . .. .. .. . . . .. . .. . . . . . . . . . . . .. .. . . .. . . . . . . . . .. . . 5- 3 5.3.1 Integral System Scaling (1st Level) .. ........ . .. . .. ...... ....... ....... . . . . . .... . ... .. 5-3 5.3.2 Mass and Energy Inventory and Boundary Flow Scaling (2nd Level)......... .. . . .5-12
- 5. 3.3 Pressure Scalin g . .. ...... ... .. .... .. .. ...... ... .. . .... .. ....... .. . .... ...... .. .. . . .. .. . ....... . .. .... .. .... 5- 15 5.3.4 Basis for Reduced Height Scaling... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... .. .5-18 5.4 Local Phenomena Scaling . . ....... ........... . . . . . . . . . . . . . .... . .. . ....5-20 5.4.1 Reactor Vessel Flow Dynamics and Instability Scaling . ... .. .. . . .. .. . . ... .... . 500 5.4.2 Choked Flow Case... ... . ... ..... .. ..... . . . . . . . . . . . . . . . . . . .. .... .. .. .. .. .....5-21 5.4.3 Un-choked Flow Case .... . . ........ . . ... ... . . . . . . . . . . . . . . . . . . . . . . . . ................5-24 5.4.4 Relative Velocity and Flow Regime . ... ........ .. ................. ... ..... . . . .. .. . ...5-27 5.4.5 Critical Heat Flux Scaling (CHF)....... .... ..... ... . . . . . . . . . . . . .. . . .......5-29 5.4.6 Flashing in the Chimney.... .. ...... ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . 5-31 5.4.7 Condensation in Suppression Pool ......... .... ......... .. .. .... . .. .. .. . . . . . . . . . . ....5-32 5.4.8 Vent Phenomena in Suppression Pool...... .. . . . . . . . . . . . . . . . . . . . . . . . . . . ..5-36 5.4.9 Mixing in Strotified Fluid Volumes. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-40 5.4.10 Natural Circulation .. .... .. .... .... .. ...... ... ... . . . . . . . . . . . . . . . . . . . . .. ...5-44 5.4.11 Heat Source and Sink .. ........ ................. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. . . . 5 -4 6 5.4.12 PCCS Venting into Suppression Pool .... .... . ... ....... . .. .. . .. .. . .. .. . ...5-47 5.4.13 Condensation in PCCS Condensers..... ....... .......... ... . ..... ..... .. . ... . . ... . .. 5-51 5.4.14 Stratification in the Drywell . . . . . . . . . . . . . . . . . . . . . . ..... . . .5-54 5.4.15 Stratification in the Suppression Pool.. . . ... . . . . . . . .. .. . ..... ..5-55
- 6. SCALE OF THE PUM A.. . .. . . . . . . . . . . . . . . . . . . . . . . 6- 1
- 7. DESIGN OF PUMA-SBWR FACILITY AND SCALING BASIS. . . . . . . 7- 1 7.1 Reactor Pressure Vessel Design. .. . . .. .. ... . . . .. . . . . 7- 1
. 7.1.1 Pressure vessel . .... ... ..... .. ... . .. .. ...... ..... ... ... .... . . . . . . . . . . . . . . . .......7-2 7.1.2 Lower plenum.. . ....... . ......... . . . . .. . . . . . . . . . .. ... .. . 7-2 7.1.3 Core plate.. ... ..... . .. . ..... ... . . . . . . . . . . . . . . . . . . . . .. . . . . . . . ., . 7-3 7.1.4 C ore .. .. . .. ...... . . .. ... . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . . .7-3
5 7.1.5 S hro u d desig n ... .... ....... .... ....... ...... .. .. .... .... .. ............ ..... . .. . . .. . . .. ...... ... . .... . . .. . . .. 7-7 7.1.6 Chim ney . .. .... .. .. . . .. .. .. . .. . . .. .. . . .. . .. . . . .. . . .. .. . . .. .. . . .. . .. . . . .. . . . . . .. . . .. . . . .. . . . . . . 7 - 8 7.1.7 D o wn c o m e r . . .. .. .. . . .. .... . .. .. .. .. .. .. .. .. .. .. . . .. .. . . .. . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...7-10 7.1.8 Separator / Dryer Assembly ........ .. ....... .. ... .. .. . . . ... . . . . . . . . .. . ..7-11 7.2 Drywell Design . . ... ...... . ........ .. .. .. . .. .. .. . ... .. . .. .. . ... .. .. . . . .. . .. . . . . . . . . . ... ..7-14 7.3 Suppression Pool Design .. ..... ....... .. ... .... . ..... ... .. .. . .. . . . . . . . . . . . . . . ..7-20 7.4 Gravity Driven Core Cooling System (GDCS) Design... ..... . . . . . . . . . . . . ..7-23 7.5 Passive Containment Cooling Systen (PCCS) and Isolation Condenser System (ICS) Design ... . . .. . ............ ... . .. .. . . . . .. . . . . . .. . ...7-28 ;
4 7.6 Automatic Depressurization System (ADS) Design . .. .. ... . . . .. . . . ..7-35 7.7 Feed Water Line and Auxiliary Reactor Coolant System............ . . . .... .. . . . ..... ...... 7-37 7.7.1 Feed Water Line (FWL).. .. . .. .... . . . . . ...... . . . . . . . . . .. . .. . ..7-38 7.7.2 Control Rod Drive (CRD) System . ... ......... . ........ . . . . . . . . . . . . . ...7-38 7.7.3 Reactor Water Clean-Up/ Shut-Down Cooling (RWCU/SDC) Systems. ... .. .7-39 7.8 Stored Heat. Heat Loss and Insulation Design..... ... . .... .... . . . . . . . . . . . . . .7-40 1
7.8.1 Ston:d Heat in RPV wall .. . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .7-40 7.8.2 Containment Heat Sink Desig n .. . ..... ............. ... ..... ... ........... . ....... . . . . .....7-42 7.8.3 Design of Insulation for Heat Loss.. ... . . . . . . . . . . . . . . . . . . . . . . . . . . 7-45 7.9 B reak Design . . . . . .... . .... .. . . .. . . . . . . . . . . .7-47
- 8. INSTRUMENTATION SYSTEM . ... . . .. .. . . . . .. . . . . . . . 8- 1 8.1 Overview of PUM A Instrumentation . . ... . . .. ... .. . . . . . .. .. . . 8- 1 8.2 Reactor Pressure Vessel Instrumentation.. .. .... ... . . . . . . . . . ..He2 8.2.1 Temperatures . .. ... . . . . . . . . . . . . . . . . . . . . . . . . ... . . .. .. .8-3 8.2.2 Pressures ... ...... .. . .. . . . . . . . . . .... .. .. .. . . . . . 8-3 8.2.3 Conductivity Probes .. . ... . . . . . . . . .. . . . . . . . ... 8-4 8.2.4 Additional Measurements.. ... ... ... . . . . . .. . ..8-4
8.2.5 B ou ndary M eas ure me n ts .. .. .. .. .. ...... ..... ............ .. .. .... ........ .. ... . .... .. ..... .. .. .... . ...... ...... 8-5 8.3 Containment Instrumentation..... ...... .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . .... 8-6 8.3.1 Tem peratures ...... ...... . .... ... .. .. . ... .. .. ...... l.. .... . .......... ... . .. . . . . .. ... . . . . . . . . . . .. . .. 8 6 8.3.2 Press ure................ . .... .... . . .. .. . . . . . . . . . . . . . . .................................8-6 8.3.3 Oxygen Concentrations .. ...... .. .. ..... . .. . . .. .. .. . ..... ... . . . .. . . .... . . . .. .. .. 8 6 8.3.4 Additional measurements......... . . . . . . . . . . . . . . . . .......................................8-7 8.3.5 Containment Boundary Measurements ..... ....... . ...... ...... .......................8-7 8.4 Data Acquisition System. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 8-7
- 9. PUM A S YSTEM DESIG N ..... ... ..... .... .. ... .. .... . .. . . ...... . .. . . . . . . . 9-1 9.1 Scaling of PUMA Facility .......... ... . .... ... . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 9- 1 9.2 Prelim i nary Desi gn . .. ... ... .. ... .. .... .... . .. . ... .. .. ... ....... .. .. .... .. .. , . . .. .......................9-2 9.3 Overa!! Instrumentation... . . ..... . . . .. ... . .. . . . . . . . . . . . . . . ... ... . .9-3 9.4 Initialization of Test ...... . .. .. ... .. ... ....... . . .... ... . . . . . . . . . . . . . . . . . . . . . . 9-4 9.5 Preliminary Test Matrix .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... . . ... 9-5 9.6 Scaling Distortion and Potential Impact on Integral Test. . . . . . . ............96 9.6.1 RPV Core.... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9-6 9.6.2 GDCS Lines.. .... ... .. ....... . .. . . . . .. .. . . . .. .. 9-6 9.6.3 PCCS Condensation ..... . . . . . . . . . . . . . . . . . . . . .. . 9-7 9.6.4 Drywell. . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . .. . 9-7 9.6.5 Suppression Pool .. .. . . . . . .. .. . . . . .. ... 9- 8 9.6.6 Helium Gas Distribution... .. . . . . . . . . .9-8
- 10.
SUMMARY
AND CONCLIJSIONS. . . . . . . .. . . ... . . 10-1
1 l
LIST OF FIGURES Figum 4.1 SBWR containment boundary [Ref. 4.1, p. 6.2-113].. .. . .. ... . . . . . . . . . . . ..4-16 Figure 4.2 Cross-sectional view of RPV [Ref. 4.1, j$. 5.3-22] . . .. . . . . . . .. . . ....4-17 Figure 4.3 Cross-sectional view of RPV with dimensions [4.6] . . . . .. . .. ... .4-18 l 1
I Figure 4.4 Horizontal vent module between the drywell and suppression pool
[Ref. 4.1, 6.2- 1 17] . . ..... ........ ................ . . . . . . . . . . . . . . . . . . . . .4-19 Figun: 5.1 'Ile static upstream stagnation pressure ratio as a function of position and the exit static pressure ratio for a smooth converging-diverging node.. . . . . . . . . . . . . . . . . . . . . . ...5-57 Figure 5.2 The nondimensional mass flow through the conveying-diverging nozzle as a function of nozzle pressure ratio .. .. . ... . ........ ........ .. ... .. .. ..... ...5-58 Figure 5.3 The static to upstream stagnation pressure ratio as a function of position and the exit static pressure ratio for an abrupt expansion nozzle . ........... . .. . . .. . . .. .. ... . . . . . . . . .. . . . . . . ..5-59 1
Figure 5.4 The nondimensional mass flow through the abrupt expansion nozzle as a function of nozzle pressure ratio. . . . . . . . . . . . . . . . . .. ..5-60 Figure 5.5 Drywell and suppression pool pressure transient schematic.. ... ... . ... . . ...5-61 Figure 5.6 Time period At to fill PCCS volume as a function of air concentration. . .5-62 Figure 7.1 Reactor pressure vessel (RPV) and internal components . .7-51 Figure 7.2 Flanges, view ports and special instrument parts on PUM A RPV . . .7-52 Figure 7.3 PUMA core shroud and mounting design. . .7-53 Figure 7.4 PUMA core inlet plate design showing heater and bypass llow area. .7-54 l
1 Figure 7.5 Decay heat curve fitting for SBWR . . . . ... .7-55 l l
l l
Figure 7.6 Cross-sectional view of PUM A core section ..... ........... . . .. ..... ......... . . . .. .. .. .. 7- 5 6 Figure 7.7 Typical heater design for PUM A RPV......................... . . . . . . .. . . . . . . . . . .. . . . . 7 - 5 7 Figure 7.8 Top of chim: Aroud and lower separate plate in PUM A RPV. .... ...... ... . .. 7-58 Figure 7.9 PUMA chimney shroud and mounting design. .. . .. .. ... . . ...... ... ... .7-59 Figure 7.10 PUMA cross-sectional view of chimney partition. .... . .. . . . . . ......7-60 Figure 7.11 Separator tube layout on lower separator plate in PUMA RPV . . ... 7-61 Figure 7.12 Orifice holes on upper separator plate and support structure in PUMA RPV . .7-62 Figure 7.13 Mounting of separator section on:o PUMA RPV, separator tube. and separator dryer shroud design... .. .... . . . . . . . . . . . . . . . . . .. . ...7-63 Figure 7.14 Design of drywell, suppression pool, connecting vent line and vacuum breaker lines in PUM A . . . . . . . . . .. . .... ... . . ... ..7-64 Figure 7.15 Jet deflector plate ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...7-65 Figure 7.16 Jet flow regime transition as a function of Froude number . .. ... . . . ..7-66 Figure 7.17 Gravity-driven cooling system (GDCS) design in PUM A . .. . . .7-67 Figure 7-18 Design of PUM A GDCS pool .. . . . . . . . . . . . .. .. .7-68 Figure 7.19 Design of PUMA ICS/PCCS condenser modules.. . .. . . . . .7-69 Figure 7.20 Lower header penetrations of the ICS/PCCS condenser modules in PUM A. . . 7-70 Figure 7.21 Design of PUMA passive containment cooling system (PCCS). . . . 7-71 Figure 7.22 Design of PUMA isolation condenser system (ICS) . . .7-72 Figure 7.23 Design PUMA ICS/PCCS pools . .. . . . . . . . .'.7 73
Figure 7.24 Plan view of automatic depressurization system (ADS) and main steam lines (MS L) in PU M A..................... .. ......... .. ....... . . . .. .. .. .... .. . ... . . 7-74 Figure 7.25 Schematic of ADS and MSL in PUM A,..... ... .... . ........ .. .... . .... .. ... ....7-75 Figum 7.26 Typical instrumented ADS line'in PUMA.... .... . .. .. . . .. . . ... . .. .. ...... 7-76 Figure 7.27 Reactor water clean-up/ shutdown cooling (RWCU/SDC) system and control rod drive (CRD) system design in PUMA. ..... .... . . . .. ...... . . .. .... . 7-77 Figure 7.28 Feed water line and sparger in PUMA... .. . . . . . . . . . . . . . . . . . . . . . . . . ...7-78 Figure 7.29 Estimated RPV wall temperature profile with time elapsed from blowd own for S BW R . . . ... . . ..... . . ..... .. ....... . . . . . . . . . . . . ..7-79 Figure 7.30 Estimated RPV wall heat loss with blowdown for SBWR.. . . . .. ...... . ..7-80 Figun: 7.31 Cross-sectional view of insulation on typical vessel wall . .. . . . . . .. ...... ..7-81 Figure 7.32 Typical break design in PUMA to simulate double-ended pipe break. . ... ... . 7-82 i
Figure 8.1 PUMA RPV temperature instrumentation ...... . .. . . . . . . . ... . .. . . . .. . . 8 22 Figure 8.2 PUMA RPV pressure and void fraction instrumentation.. .... . . . . . . . .8-23 Figure 8.3a Schematic of the conductivity probe. ......... .. ..... .. .... . . . . . . . . . . . . ... .8-24 Figure 8.3b Output signal of conductivity probe . . . .... .. . . . . .. . .8-24 Figure 8.4 PUMA RPV boundary measurements .. . ... . . . . . . . .. . ... ..8-25 Figure 8.5 Typical instrumented ADS line . . . . . . . .... .. . .8-26 !
l Figure 8.6 Capacitance probe schematic. . . . . . . . . .. ...8-27 Figure 8.7 PUMA containment temperatures and pressures... . . . .. .8-2F l 1
Figun: 8.8 PUMA containment boundary flow measurement.. . .. ... ....8-29 )
1 Figure 8.9 Water level ranges in the SBWR ................. .. .................. .... .... ....... .. .... . ...... 8-30 Figure 9.1 Overall schematic of PUM A facility ... .. .......... ............ . ... ..... . ... .... ... .... 9 24 Figum 9.2 Plan view of main vessels in PUMA ... l. .............. ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......9-25 1
1 1
1 l
1 1
1 1
l l
v
I LIST OF TABLES Table 4.1 Dimensions of Reactor Pressun: Vessel [Ref. 4.1, p.1.3-5,5.1-5,1.3-4) ......... . ..... 4-1 Table 4.2 Thermal-bydraulic Parameters for RPV [Nef. 4.1, p.4.4-6.1.3-2]... ........ ... ....... ..... 4-2 Table 4.3 Containment Parameters [Ref. 4.1, p. 6.2-20, 6.2-61) .... ... ..... ...... .. . . . .... ....... 4-3 Table 4.4 SRV and DPV Parameters, Water Levels, and ADS Actuation Timings ;
[Ref. 4.1, p. 21-39,6.2-120,6.2-121 Table 21.5-1-1, 6.3-28)................................................................ . . . .. . . . . .. 4- 5 Table 4.5 GDCS Parameters and Equalization Line Parameter '
[Ref. 4.1, p. 6.2-60. 6.3 5, 6.3-6] .... ...... . ... .... . ..... ... .. . .. ...............4-6 ;
Table 4.6 PCCS Parameters [Ref. 4.2, Appendix 2] . .. . .. ... .... . . ... . .. . . . . . . .. ... .. .. . 4-7 Table 4.7 IC pool and ICS Parameters [Ref. 4.2 Appendix 2].. . .. . . . . . . . . . . . . . .... 4-8 Table 4.8 Postulated Breaks [Ref. 4.1, p. 6.3-20, 6.3-21] .. .. .... .... . ..... .. . .. .. .. . . . . . .. . . . .. 4 - 9 I
Table 6.1 Comparisons for Major Components and Dimensions of Different Heigh t Scaling ......... .. .............. ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 6-4 Table 6.2 Comparison of Various Parameters and Non dimension'al Numbers Between Prototype and PUMA (at 1.03 MPa or 150 psi operating pressure).... . ......... . ...... . 6-5 Table 7.1 Reactor Pressure Vessel Dimensions..... . .. . . . . . . . . . . . . . . . . .. . . .. 7-2 Table 7.2 Coefficients for Decay Heat Polynomial Fit. . . . . . . . . . . . . . . . .. .......7-4 Table 7.3 Decay Power in SBWR and PUMA Facilities. . . . . . . . . .. . .. 7-4 Table 7.4 Core Power Levels at 1.03 MPa (150 psi) Following Blowdown.. . . .. . . . 7-5 ,
Table 7.5 Core Heater Rod Arrangement.... ........ . ..... . . .. . . ....................7-6 Table 7.6 Core Parameters . . . .... ...... . . . . . . . . . . . . . .. .. . ........7-6
Table 7.7 Core S hroud Design ........ . ..... ............. ................ .... . . . .... . . ... .. . . . . ... .. 7- 8 Table 7.8 Chimney Design ............. .. .. . ....... . . ... .. .. ..... ... . . . ... .... .... .7-10 Table 7.9 Required Separator Tube Numbers and T' heir Diameters.. .... .. . . . . . . . . . . . ...7-12 Table 7.10 PUMA Drywell Design.. . ... ... . .. . . . . . . .. . . . . . . . . ...7-16 Table 7.11 PUMA Suppression Pool (SP) Design ... ... .. ...... . ... .. . . . . . . .7-23 Table 7.12 PUM A GDCS Design.... .. ... . ..... .. . . . . . . . . . . .. .7-28 Table 7.13 PUMA PCCS/ICS Condenser Design Parameters.. . . .. ...7-33 Table 7.14 PUMA PCCS/ICS Pool Design Parameters. . . . . . . . . . .7-35 Table 7.15 SRV and DPV Capacities and Sequence of Action for.SBWR . . . . . . ..7-36 Table 7.16 Automatic Depressurization System (ADS) for PUMA Facility. . . ..7-37 Table 7.17 PUMA Feed Water Line: 2 Each . .. .... .. ... .. . . . . .. .. .. ..7-38 Table 7.18 PUMA Component Heat Loss Estimation .. . .. . . . . . .. .. .. .7-48 Table 7.19 Containment Heat Loss in PUMA with Addition of Fibrous-based Acoustic Tile to Simulate Concrete Heat Sink ... .. . ....... .. ..... . . . . . . . . ... . . . .7-50 Table 8.1 Mass and Energy Measurement. .. . . . . .8-2 i
Table 8.2 Vessel measurement. . . . . . . .. . . . . . .. 8 9 1
1 Table 8.3 Drywell Measurement.... .. . .. . . . . . .8-13 l Table 8.4 Suppression Pool measurement.. . . . . .. . . .8-14 Table 8.5 PCCS and ICS Measurement .. . . . . .8-15 2
Table 8.6 GDCS Measurement.. . . .. .8-18
- _ _ . _. . _. . . .. .. . - _ ___ . _ . _ _ ~. . _ _ _
. 8 Table 8.7 Boundary Flow Measurement .............. ....................... ............ ................. ...........g.19 Table 8. 8 Ins trum en ts ......... ... .. .... .......... . . .. .... .. ...... .......... .... .... .... .... .. .. .. . .. . . .. . . . . . . . . . . , , g . 2 1 Table 9.1 Comparison of Components Dimensions Netween Prototype and PUMA ................ 9 9 Table 9.2 Controls and Provisions on Various Components of PUMA...... .. ....... ... ........ ... 9 2()
Table 9.3 Phase 1 of th: Preliminary Test Matrix-Base Case and GE Counterpart Tests................................................................................ . . . . . . . . ... . . . . . 9-21 Table 9.4 Phase 2 of the Preliminary Test Matrix-Sensitivity Study and '
Be yon d D B A Tests .. . .. .... .. ... ... .... . . ..... . ... ..... .... . . . .. . . , .. .,,, . . . .. , , , , , ,, 9 22 Table 9.5 Geometric Distortion in R PV ... .... .... . ..... ..... ... ...... . ..... .. .. . ....... . .... .. . . . 9-23 e
i i
1 4
1 1
1 l
l 1
Y r- ,
NOMENCLATURE A Flow area scale 2
a Cross-sectional area [m )
B; Biot number (Eq. 5.19)
Co Distribution parameter (Eq. 5.109) co Concentration of non-condensables -
cp Specific heat [J/kg-C]
D,d Diameter [m]
D,t Bubble diameter [m]
E Energy [J)
E, Vapor latent heat flux [J/m2 )
F Total pressure loss coefficient f Friction factor, friction f Frequency [s-1]
G Mass velocity [kg-m/s2 )
Gr Grashof number (Eq. 5.193) 2 g Gravitational acceleration [m/s ) j j Superficial velocity [m/s]
Ja Jakob number (Eq. 5.189) 1 H Height [m]
h Enthalpy [J/kg]
2 h Heat transfer coefficient [W/m -C]
k Conductivity [W/m-C]
K Minor loss coefficient L Axial length scale I Length [m]
m,M Mass [kg]
th Mass flow rate [kg/s]
n.N Number Na Drift flux number (Eqs. 5.32 and 5.102)
Np, Froude number (Eq. 5.31)
Nr Friction numher (Eq. 5.35) l Nriash Flashing phase change number (Eq. 5.112)
Nc n Natural circulation number (Eq. 5.66)
No Orifice number (Eq. 5.36)
N pc3 Phase change numbe-(= Zuber number) (Eq. 5.29)
Ny CHF number (Eq. 5.110)
N,ui, Subcooling number (Eq. 5.30)
Na Thermal inertia ratio (Eq. 5.34)
Nu Nusselt number (Eq. 5.169) ,
Nzu Zuber number (Eq. 5.29) p,P Pressure [Pa]
Pr Prandtl number (Eq. 5.194) q Power [W]
2 q" Heat flux [W/m )
q'" Volumetric heat generation [W/m3 ]
Q Volumetic gas flow rate [m3/s]
Q Transfer function Qs Heat source number (Eq. 5.20) e Condensation power R Richardson number (Eq. 5.1)
R Universal gas constant [kJ/kgmol-K]
Ra Rayleigh number (Gr Pr) ,
Re Reynolds number (Eq. 5.190) 2 S Surface area [m )
St Modified Stanton number (Eq. 5.17) t Time [s]
T Temperature [ C]
T* Time ratio numher (Eq. 5.33) u Velocity [m/s]
ur Intemal energy ofliquid 2
U Overall heat transfer coefficient [W/m -C]
v,V Volume [m ]
3 Ve Drift velocity [m/s] (Eq. 5.l(X))
w Work [J]
W Width [m]
Wa Non-condensible mass fraction x Quality x Distance [m]
X Concentration of steam z.Z Distance [m]
i Greek Symbols Volumetric thermal expansion coefficient [K- ]
6 Conduction depth [m] '
S Perturbation A Difference Ahf3 Latent heat of vaporization [J/kg] ;
a Void fraction
- a. Thermal diffusivity [m 2f3) p Density [kg/m3 ]
t Time Constant [s]
p Dynamic viscosity [kg/m-s]
11 Heated perimeter [m]
v Kinematic viscosity [m 2f3) i x Time constant ratio O Dimensionless temperature a Surface tension [N/m] ,
I Summation ,
4 Wetted perimeter [m]
T Parameter
'll Subscripts a Ambient b Bulk b Bubble c Core, critical, containment cond Condensation DW Drywell e Exit i eq Equivalent f Fluid g Gas i ith component in Inlet jJ Jet .
voe
, . _ , _ , , , , -_m.
9 m Model m Maximum m Mean, average o Reference point / component out Outlet p poolside p Prototype pc PCCS R Ratio s Surface, solid sup,sp Suppression pool T Pool water t Throat th Thermal v Vapor v Vent, vessel w,W Wall ,
Wa Non condensable mass fraction 1
l Superscripts Dimensionless quantity I
I
1-1
- 1. INTRODUCTION The General Electric Nuclear Energy (GE) has developed a new boiling water mactor called the Simplified Boiling Water Reactor (SBWR) [1.1]. Major diffemnces between the current Boiling Water Reactors (BWRs) and the SBWP are in the simplification of the coolant circula-l tion system and the implementation of passive emergency cooling systems. There are no recir-culation pumps to drive the coolant in the vessel of the SBWR. The emergency com cooling and containment cooling systems do not have active pump-injected flows.
There are several engineered safety systems and safety-grade systems in the SBWR which are directly related to the relevant issues and objectives of the present program: 1) the Automatic Depressurization System (ADS),2) the Gravity-Driven Cooling System (GDCS),3) the Passive Containment Cooling System (PCCS),4) the Isolation Condenser Systems (ICS),
and 5) the Pressure Suppression Pool (SP). The GDCS and PCCS are new designs unique to the SBWR and do not exist in operating BWR's. The ICS is similar to those in some operating BWRs. Both the GDCS and PCCS are designed for low-pressum operation (less than 1.03 MPa or 150 psia), but the ICS is capable of high pressure operation as well (up to 7.58 MPa or 11(X) psia). ,
The ADS system becomes active at a prescribed vessel condition and depressurizes the reac- .
tor vessel so that the gravity driven cooling systems can be activated. The goal of these safety systems is to maintain sufficient core cooling by preventing core uncovery and dryout of the fuel pins.
The performance of these safety systems under a loss of coolant accident (LOCA) and other important transients is a major concern. Since the emergency cooling systems are driven by the gravitational head, interaction between the ADS, GDCS, PCCS and other auxiliary systems are important. The emergency cooling systems depend not only on the gravitational head but also on the relative static pressure differences between the vessel, drywell and wetwell (suppression pool). The safety systems and various natural circulation phenomena encountered after the ini-tial vessel depressurization in the SBWR are somewhat different from the systems and phenomena studied by the nuclear community in existing commercial nuclear reactors.
General Electric has performed tests to assess the GDCS performance in a low pmssure full-height GIST facility with a volume scale of 1/508 [1.2]. Results of this study have demonstrated the feasibility of the GDCS concept. The GIST facility was scaled from an older SBWR design in which the GDCS pools were combined with the SP. The PCCS was absent in the GIST facil-ity, hence parallel operation of the GDCS and PCCS was not observed in the GIST experiments.
GE has also performed tests to assess the PCCS performance in a low-pressure, full height Toshiba GIRAFFE facility in Japan with a volume scale of 1/4(X) [1.3]. The GIRAFFE tests pro-vided data to help model the prototypic SBWR PCCS units, and demonstrated the feasibility of the non-condensible venting concept. However, the GIRAFFE facility was scaled from an older
1-2 SBWR decign, and it did not investigate GDCS injection in the vessel.
A new PANDA facility in Switzerland is a low-pressure, full-height facility with a volume scale of 1/25 [1.4]. 'Ihe main focus of the PANDA facility is on PCCS performance and con-tainment phenor. na in a relatively large-scale facility so that three-dimensional effects can be assessed. Like GIRAFFE, however, the PANDA facility is not designed for assessing GDCS injection into the vessel. Although GE has performed experimental and analytical studies for the PCCS and GDCS systems and associated phenomena, the U.S. Nuclear Regulatory Commission (NRC) has identified a need to develop additional independent confirmatory data from a well-scaled integral test facility built to reproduce major thermal-hydraulic phenomena at relatively low pressure (< l.03 MPa or 150 psia) [1.5]. Purdue University was awarded a research con-tract, " Confirmatory Integral System Testing for GE SBWR Design," to design, construct and operate PUMA (Purdue University Multi-dimensional Integral Test Assembly) to obtain integral test data.
The objectives of this program are to build a scaled integral test facility and obtain confirmatory data for the NRC to assess the RELAP5/CONTAIN code. The general guidelines for assessing the code scalability and uncertainty associated with accident predictions have been developed at NRC [1.6]. This report summarizes the details of the scaling method and sciendfic design of the integral facility, including its instrumentation.
l i
l l
l t
Note: Thmughout this report, "pmtotype" refers to the GE-SBWR and "model" refers to the present PUMA test facility.
~
1-3 References 1.1 GE Nuclear Energy, "SBWR Standard Safety Analysis Report 25A5113 Rev. A, August (1992).
1.2 Billig, P.F., " Simplified Boiling Water Reactor (SBWR) Progmm Gravity-Driven Cooling System (GDCS) Integral Systems Test-Final Report," GEFR-(X)850, October (1989).
1.3 Tsunoyama, S., Yokobori, S., Arai,.K., " Development of Passive Containment Cooling System," Proc. Intemational Topical Meeting on Advanced Reactor Safety, Hyatt Regency, Pittsburgh, April 17-21 (1994).
1.4 Yadigaroglu, G., " Scaling of the SBWR Related Test," Report NEDC -32258, November (1993).
1.5 Han, J.T., Bessett, D.E., Shotkin, L.M., "NRC Confirmatory Testing Program for SBWR,"
Proceedings of the Twenty-First Water Reactor Safety Information Meeting, Bethesda, Maryland, October 25-27 (1993).
1.6 NRC Draft Report, " Compendium of ECCS Research for Realistic LOCA Analysis - Draft l l
Report for Comment", NUREG-1290 (1987).
l l
i
2-1
- 2. OBJECTIVES The major objectives of the PUMA program are to:
I. provide integral data to NRC for the assessment of the RELAP5/CONTAIN code for SBWR applications,
- 2. assess the integral performance of GDCS and PCCS, and
- 3. assess SBWR phenomena important to LOCAs and other transients The focus of the PUMA integral test program is the reproduction of the important phenomena expected in the SBWR for use in the assessment of the RELAPS/CONTAIN code.
The objective of the scaling method is to provide a facility design that will reproduce those phenomena which occur during both the later stages of depressurization of the SBWR pressure vessel and during the functioning of the gravity-driven safety systems. A corollary objective of the scaling will be to preserve, to the extent necessary and possible, the sequence and interrela-tion method of the key phenomena. In this way, comprehenrive data which can be related to prototypical condition will be provided for assessing the code models.
The particular focus of the integral experiments will be to obtain data on the nerformance and interaction of the GDCS and PCCS, particularly as related to the maintenance of the coolant level in the RPV, containment integrity, maintenance of natural circulation, possible occurrence of two-phase natural circulation instab.lities, the effect of non-condensible on PCCS perfor-mance and potential impact on coolability of the core. Data will also be obtained regarding sys- !
tem interaction between the GDCS, PCCS and regarding auxiliary cooling systems, and possible water hammer occurrence during GDCS injection, feedwater injection, and ICS condensate draining into the vessel.
The data collected will provide qualitative as well as quantitative tests of the code models l
and overall predictive capability of RELAP5/CONTAIN. In this way, the uncertainty associated ;
with the calculation of the safety margins predicted to exist for design-basis accidents can be ;
comprehensively assessed by the NRC using the code scaling and uncertainty analysis methodol- l ogy established several years ago.
l
_-.______i
4 31
- 3. PROGRAM REQUIREMENTS The PUMA tests are primarily concerned with: developing a well scaled integral test data base for NRC to assess the ability of RELAPS!CONTAIN code to simulate the effectiveness of the GDCS and PCCS, and assessing the interaction.among safety and non-safety systems. It has been found in previous studies that the largest uncertainties in predicting safety system perfor-mance are found in the later stages of accident events in which the system pressun: is reduced through the automatic depressurization systems or break flow. The stated objective of the Nuclear Regulatory Commission is to obtain confirmatory data from a scaled integral test facility for major thermal-hydraulic phenomena at low pressure after vessel depressurization in the SBWR In view of this, the test facility should be designed to reproduce the phenomena at LO3 MPa (150 psia) or below in the plant. This maximum pressure is one of the most important fac-tors affecting the design and cost of the proposed integral test facility.
The data collected will serve as pan of the basis in the assessment of the model applicability and code uncertainties associated with the use of the RELAP5/CONTAIN safety analysis code for SBWR applications. The data should confirm the performance of various engineered safety features in simulated accident conditions. ,
The integral test facility design should be based on a rational scaling method which embo-dies conservation principles. The scaled-down system design requires that separate considera-tions be used for the length scale and the flow area scale. The product of these two scales give the overall volume scale. For natural circulation-driven flow, both the driving force and flow resistance simulations are very important, since the natural circulation flow is essentially deter-mined by the balance between these two forces.
In the past, a full-height and reduced-area facility was often justified on the basis of preserv-ing the total driving force. However, this is often not a valid argument because such a scaling approach may lead to a significant distortion of frictional resistance. The magnitude of the fric-
~
tional resistance is proportional to the length-to-diameter ratio //d. For a full-height but reduced-area facility, the natural circulation rate may be significantly smaller as a result of much larger values for I/d in the pipe sections. One way to reduce this impact is to enlarge the diame-ter of piping sections. However, this will lead to significant distortions of the scaled mass and energy inventories, which usually must be conserved. Since the inventory balance beovcen the vessel, the containment, and the suppression pool largely determines the course of events in SBWR accidents and transients, inventory distortion is highly undesirable in the test facility.
The integral test facility scaling method should also provide a rational basis for scaling the
- integral test facility results up to the prototype conditions. Therefore, it is necessary to.have scaling criteria for time, velocity, pressure, void, mass inventory and energy inventory in addi-tion to the geometric scaling criteria. The integral test results should not only qualitatively iden-tify the key thermal-hydraulic phenomena and sequence of events, but also quantify the system
3-2 response, phenomena and sensitivity, and the effects of interactions of components and phenomena. However, because of some scaling distortions, data from test facilities, including PUMA, GIRAFFE, and PANDA are not expected to reproduce quantitatively the system response and sensitivity exactly as in SBWR.
The integral tests are primarily concerned with the phenomena encountered after the reactor vessel is depressurized to the level when the GDCS is activated. Therefore, the facility should represent the major vessel intemal structures, decay heat in the core, the depressurization sys-tems, GDCS, PCCS, ICS, suppression pool, and non-safety systems. In addition to these, a sys-tem for water injection to the feedwater lines is necessary in order to address the interactions of GDCS with the non-safety systems such as the control rod drive system (CRD), Reactor Water Cleanup / Shutdown Cooling System (RWCU/SDCS) and Fuel and Auxiliary Pools Cooling Sys-tems (FAPCS). The latter requires a spray system in the dry well.
The integral test facility should have instrumentation to provide phenomena comprehension, quantification, and evaluation, and also be usable by the NRC in model development and assess-ment of the RELAP5/CONTAIN code. The instrumentation should measure at least
. pressure at various locations
. A P in vessel for void measurements
. mixture level in the vessel
. in-vessel void fraction
. flow in various connecting lines 1
. non-condensible concentration
. temperatures I l
. natural circulation rate in the vessel l
. water level in the pools
. power input
. heater surface temperature
. heat loss The scaling method must address the following phenomena and issues; l) in-vessel natural circulation and two-phase flow instability,
- 2) flashing in the chimney,
- 3) inflow or outflow from various components and intercomponent flow,
- 4) initial and boundary conditions,
- 5) important containment phenomena,
3-3
- 6) single phase and two-phase natural circulation
- 7) condensation phenomena in the presence of non-condensible gases, and
- 8) system stored energy and decay heat.
Dus, tasks to be performed under the pmsent p ogram am:
. Perform phenomena identification for SBWR LOCAs and transients, to be used for the design of the integral experiments and development of a test plan.
. Develop a well-balanced and justifiable scaling approach for the design of the SBWR integral facility, PUM A.
. Design a well-scaled integral test facility having proper and sufficient instrumentation.
. Construct the scaled integral SBWR test facility.
. Develop boundary and initial conditions for the integral tests based on the scaling method and computer code simulation using RELAP5/CONTAIN.
. Perform the integral tests under strict quality assurance over the experiments, as well as the reporting procedures.
. Report the results in a NUREG/CR document.
1 l
l l
. _ - .m . . _ .__ _ _
4-1 i
- 4. DESIGN BASIS FOR THE PUMA INTEGRAL FACILITY 4.1 Systems Characteristics (Safety)
In this section, tie system characteristics of the SBWR that are relevant to the safety of the reactor are discussed. Some figures and data for the tables presented in the discussion were obtained from the GE Standard Safety Analysis Report (SSAR) [4.1] and the captions for these figures and tablec include references to the SSAR page number for immediate reference. All-dimensions presented ire for nominal size.
In Figure 4.1, the SBWR containment boundary is shown. Within the containment boundary there are the reactor pmssure vessel (RPV), drywell, suppression pool (SP), gravity driven-core cooling system (GDCS) pools, isolation condenser system (ICS) piping, passive containment cooling system (PCCS) piping and the automatic depressurization system (ADS). The con-densers and pools for the ICS and pools for the PCCS are located outside the containment boun-dary.
4.1.1 Reactor Prer.sure Vessel (RPV) ,
Figure 4.2 shows the cross-sectional view of the SBWR vessel. The dimensions of the vesse' are given in Table 4.1. The thermal-hydraulic parameters of the RPV at the normal full-power operation are given in Table 4.2.
Tal .e 4.1. Dimensions of Reactor Pressure Vessel (Ref. 4.1, p.1.3-5,5.1-5,1.3-4)
Inside Height 24.612 m ID 6m Wall Thickness 157.175 mm 3
Coolant Volume 607.3 m 3
Total Volume 669 m Active Fuel length 2.743 m In Figure 4.3, the levels of various parts and intemals of the RPV are shown. The overall height of the RPV is about 25 m. This permits natural circulation driving forces to produce the required core coolant flow. An increased thermal d.iving head is provided by a long " chimney" in the space which extends from the top of the. core to the entrance of the steam separator assem-bly. The chimney region has vertical panels for flow partition. The RPV volume provides a large reserve of water above the core. This volume assures a long period of time before core uncovery in the case of feedwater flow interruption or loss of coolant. The large RPV volume also reduces the reactor pressurization rates that develop when the reactor is suddenly isolated
- v- - , , e
4-2 from the normal heat sink.
Table 4.2. Thermal-hydraulic Parameters for RPV (Ref. 4.1, p. 4.4-6,1.3-2)
Core Power (l(X)%) 2000 MWth Core Inlet Flow 27.2 x 106 kg/h Feedwater Inlet Flow 3.88 x 106 kg/h Steam Dome Pressure 7.17 MPa Core Inlet Pressure 7.28 MPa Core Outlet Pressure 7.23 MPa Average Core Power Density 41.5 kW/ liter Average Heat Flux 430.58 kW/m2 Maximum Heat Flux 1225.23 kW/m 2 Core Average Quality 14.3 Feedwater Temperature 215.6 C Core Inlet Temperatum 278.5 C Core Outlet Temperature 280.3 C '
l The reactor internals consist of com suppon stmetures and other equipment. The core sup-pon stmetures locate and support the fuel assemblies, form partitions within the reactor vessel to sustain pressure differentials across the partitions, and direct the flow of coolant water. The major intemal structures consist of a shroud, shroud support, core plate, and integral fuel support and control rod guide tubes (CRGT).
'he other reactor internals are the control rods, feedwater spargers, in-core guide tubes, chimney, chimney partitions, chimney head, steam separator assembly, and the steam dryer assembly. The shroud stmport, shroud and chimney make up a cylindrical stainless steel assem-bly that partitions the upward flow of coolant through the core from the downward recirculation flow in the downcomer.
4.1.2 Containment System The SBWR has a low-leakage containment which is divided into the drywell and pressure suppression chamber. The containment is a cylindrical, steel-lined, reinforced concrete stmeture integrated with the reactor building. The drywell design conditions are 379 kPa (gauge) and 171 C. The suppression chamber design conditions are 379 kPa (gauge) and 121 C. The
~drywell is divided by the vessel support skirt into a lower drywell (below the skirt) and an upper drywell (above the skirt). There is an open flow area between the lower and upper drywells to allow for pressure equalization. The upper drywell houses the main steam lines and feedwater !
piping, the SRVs, GDCS pools, main steam drain piping and upper drywell coolers. The !
l l
4-3 suppression pool is higher in elevation than the top of the core. This provides a gravita"onal driving head for injecting suppression pool water into the vessel when the vessel is depressurized and the equalizing lines (total of three) between the suppression pool and the vessel are opened.
The gas space above the suppression pool serves as the LOCA blowdown gas reservoir for the upper and lower drywell nitrogen and other noncondensible gases, which pass through the eight drywell-to-suppression chamber venical vent pipes. Each vent pipe has three horizontal vents located below the suppression pool surface. There are 24 horizontal vents between the drywell and suppression pool. In Figure 4.4, a detailed view of the horizontal vent module is shown. To prevent suppression pool water from flowing into the drywell via the horizontal vents, there is a vacuum breaker system between the suppression chamber and drywell. The vacuum breakers consist of check valves which open when the suppression chamber pressure exceeds the drywell pressure at a preset pressure difference. In Table 4.3 relevant containment parameters are given.
Table 4.3. Containment Parameters (Ref. 4.1, p. 6.2-60, 6.2-61)
Drywell volume above skirt 4598 m 3 Drywell volume below skirt 892 m 3
Suppression pool gas volume 3819 m 3 Suppression pool water volume 3255 m3 SP vertical vents area 9m 2 SP surface area 588 m 2 Vertical vent pipe inside diameter 1.2 m Vertical vent pipe height 12.7 m Horizontal vent diameter 0.7 m Heights of horizontal vents, from pool floor Top vent 3.5 m Middle vent 2.13 m Bottom vent 0.76 m 4.1.3 Main Steam Lines Two main steam lines (MSLs) of 711 mm diameter carry steam from the RPV to the turbine main steam systems. Each of the two MSLs has a flow restricting nozzle built into the RPV exits. In the event of a MSL break accident, the restrictor limits the coolant blowdown rate from the reactor vessel to a choked flow rate equal to or less than 200% of rated steam flow at 7.07
4-4 MPa. Each MSL has two steam isolation valves, one inside and one outside the containment.
On each MSL, there are four safety relief valves (SRVs) and one depressurization valve (DPV).
These SRVs and DPVs are discussed below.
4.1.4 Automatic Depressurization System (ADS) #
The function of the automatic depressurization system (ADS) is to systematically depressur-ize the RPV in the event of LOCA or transient, to allow the GDCS water injection to the vessel, preventing the core uncovery and maintaining the core temperature below design limits. The ADS also keeps the reactor depressurized for continued operation of the GDCS after a system accident response, without power. The ADS consists of the eight SRVs, six DPVs and their associated instrumentation and controls. There are four SRVs and one squib-type DPV on each main steam line. Four DPVs are flange-mounted on hodzontal stub lines connected to the RPV at about the elevation of the MSLs. The SRVs discharge into the suppression pool through spargers. The DPVs discharge into the upper drywell.
The SR Vs and DPVs are actuated in several groups at staggered times as the reactor under-goes a controlled depressurization. This minimizes reactor ndxture level swell during the depressurization phase. When the low water level (Level 1) signal persists for at least 10 seconds, then the ADS is activated. First, four SRVs (two from each MSL) open and discharge steam to the SP, The remaining four SRVs open after an additional 10-second time delay, At 55 seconds after ADS actuation, the first group of two DPVs (on MSLs) start to discharge to drywell. Likewise, the second group of two DPVs open after 1(X) seconds and the third group of two DPVs open after 145 seconds of ADS actuation. In Table 4.4, SRV and DPV parameters, ADS actuation and water level definitions are given.
4.1.5 Gravity Driven Core Cooling System (GDCS)
The emergency core cooling systems (ECCS) of the SBWR are the GDCS and ADS. The ADS described above provides depressurization through the use of safety relief valves (SRVs) and depressurization valves (DPVs). Once the reactor is depressudzed, the GDCS provides gravity-oriven flow from three separate water pools located within the drywell at an elevation above the active core region. The GDCS initiation signal is related to the confirmed Level I sig-nal. Three squib valves are activated,150 seconds after confirmed Level i signal, one in each of j the injection lines connecting the GDCS pools to the RPV. The additional water flow from the SP can be injected into the RPV through three equalizing lines to meet long-term post-LOCA core cooling requirements. After 30 minutes and when the RPV coolant level decreases to 1 m above the top of the active fuel (TAF), squib valves are opened in each of three equalization lines. The 30 minute delay and the above criterion ensure that the GDCS pools have had time to drain into the RPV and, that as a result of the blowdown, the initial RPV level collapse does not open the equalizing lines.
4-5 Table 4.4. SRV and DPV Parameters, Water Levels, and ADS Actuation Timings (Ref. 4.1, p. 21-39,6.2-120,6.2-121, Table 21.5-1-1,6.3-28)
'SRV Inlet Line Diameter 203.2 mm SRV Outlet Line Diameter 254 mm SRV Flow Area 67 cm 2 DPV (MSL) Inlet Line Diameter 304.8 mm DPV (RPV) Inlet Line Diameter 457.2 thm DPV Flow Area 248 cm 2 Levels w.r.t. TAF Control Functions Normal water Level (NWL) 11767 mm Level 9 - (L9) 12870 mm Level 8 - (L8) 12200 mm Level 3 - (L3) 10840 mm Level 2 - (L2) 7930 mm Level 1 - (LI) 3930 mm Level 0.5 - (LO.5) 1(XX) mm TAF w.r.t. Inside Bottom of RPV 6493 mm BAF w.r.t. Inside Bottom of RPV 3750 mm Value Actuation Sequence af ter Level 1* Signal Confirmed 4 SRVs 0.0 s 4 SRVs 10 s
)
2 DPVs 55 s ;
2 DPVs 1(X) s 2 DPVs 145 s 1
- Maximum Allowable Time Delay to Confirm 1.evel 1 Signal 10 s RPV Water Level Control Function )
l L9 Initiate trip feedwater pumps runback, L8 Tdps CRD high pressure make up, scrams reactor, closes main turbine stop valves, and initiates feedwater pumps runhack L3 Runback RWCV pump, Trips leak detection and isolation system (LD & IS), and scrams reactor L2 Initiates the CRD high pressure makeup mode, Initiates alternate rod intention ( ARI),
Initiate IC's, and Closes MSIV's, containment isolation valves i except IC's L1 Initiate ADS, GDCS and Tdps LD & IS LO.5 Initiate GDCS Equalizing line
4-6 In Table 4.5, the G DCS parameters are given. As shown in Table 4.5, the minimum equaliz-ing line driving head 7f I m is determined by the elevation differential between the top of the first SP horizontal vent and the centerline of the equalizing line RPV nozzle. The 203.2 mm (8 inch) lines from each GDCS pool branch out into the 152.4 mm (6 inch) lines just before they enter R1 V.
Table 4.5. GDCS and Equalization Line Parameter (Ref. 4.1, p. 6.2-60, 6.3-5, 6.3-6)
GDCS Pool Numbers 3 Each GDCS Pool Minimum 329 m 3 Drainable Inventory Minimum Surface Elevation of GDCS 13.3 m Pool above the RPV Nozzle 3
SP Inventory 1 Meter above TAF 1475 m Minimum Equalizing Line Head 1m GDCS Line Size from GDCS Pool 203.2 mm (three total)
GDCS Line Size at RPV 152.4 mm (six total)
RPV-Injection Line Nozzle Size 76.2 mm (six total)
Equalizing Line Nozzle Size 50.8 mm (three total) 4.1.6 Passive Containment Cooling System (PCCS)
The passive containment cooling system (PCCS)is an engineered safety feature and there-fore it is a safety-related system. The PCCS removes the core decay heat energy, rejected to the containment after a LOCA, to outside of the containment. It provides containment cooling for a minimum of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> after a LOCA. The PCCS consists of three PCCS condensers. The con-denser is sized to maintain the containment within the design pressure limits of 379 kPa (gauge)
(55 psig) for design basis accidents (DBAs). The PCCS is designed as a passive system without power actuated valves or other components that must actively during the accident. ;
1 Each PCCS condenser i~s composed of two identical modules. One PCCS condenser assem- l bly is designed for 10 MWt capacity under conditions of saturated steam in tubes at 308 kPa (45 l psia) and 134 C and pool water at atmospheric pressure and 101 C. The noncondensible gas
47 purging system is driven by the pressure difference between the drywell and the suppression chamber.
The PCCS parameters are given in Table 4.6. Each PCCS condenser has two identical bun-dies of vertical heat-transfer tubes connected to a. steam drum above as an inlet plenum and a similar drum below as an outlet plenum. The vent and drain lines from each lower header are routed to the drywell through a single containment penetration. The condensate drains into an annular duct around the vent pipe and then flows into a line which connects to a large common drain line which also receives flow from the other header. The drain line discharges condensate into a GDCS pool. The non-condensibles from the PCCS condenser an: vented through the vent pipe into the suppression pool.
4.1.7 Isolation Condenser System (ICS)
The isolation condenser system removes core decay heat from the reactor by natural circula-tion. It can function with minimum or no loss of coolant inventory from the reactor when the normal heat removal system is unavailable. For example, it can be activated for the following events: 1) sudden reactor isolation from power operating conditions, 2) reactor hot standby mode, and 3) safe shutdown condition.
Table 4.6. PCCS Parameters (Ref. 4.2, Appendix 2)
Number of Units 3 Modules per Unit 2 Tubes per Module 248 Total Heat Transfer Area inside/Outside 400/430 m 2 Total Flow Area 2.6 m 2 Condenser Tube
- Length 1.8 m
-OD 50.8 mm
-ID 47.5 mm
- Material Stainless Steel Headers
- 1xngth 2.4 m
-OD 750 mm Condenser Tube Bundle Volume 3.2 m 2 1
1 l
l i
4-8 l
The ICS consists of three independent high-pressure loops, each of which contains a steam )
isolation condenser (IC). The steam is condensed on the tube side and transfers the heat to a large ICS/PCCS pool by evaporating water to the atmosphere. Each IC is designed for 30 MWt capacity and is made of two identical modules. ,
Each IC is located in a subcompartment of the IC/PCCS pool, and all pool subcompartments communicate at their lower ends for full utilization of the collective water inventory. Steam condenses inside vertical tubes and collects in the lower header. Two pipes from each lower header take the condensate to the common drain line leading to the RPV. Noncondensible gases are purged through vent lines into the SP.
During LOCA transients, the ICS is activated when the reactor water level falls below Level
- 2. In Table 4.7, the IC/PCCS pool and ICS parameters are given. Note that the 254 mm hot steam inlet line shares a steam line with one of the DPVs.
Table 4.7. IC/PCCS Pool and ICS Parameters (Ref. 4.2, Appendix 2)
IC Pool <
- Depth 4.4 m 3
- Air space 1.4 m 3
- Volume above top of tubes 1250 m IC Inlet Line Size (from DPV Stub / Tube) 304.8 mm IC Condensate Return Line Size 152.4 mm IC Vent Line (to SP) Size 19.05 mm Number of Units 3 Modules per Unit 2 Condenser Tube
- Length 1.8 m
-OD 50.8 mm
-ID 46.6 mm
- Number of Tubes. 120 per module 4.1.8 Control Rod Drive (CRD) System and Reactor Water Cleanup / Shutdown Cooling (RWCU/SDC) System The CRD and RWCU/SDC systems are pump-driven non-safety systems. They can also pro-vide high-pressure water injection into the vessel if the AC power is available. The injection of high-pressure makeup water to the reactor is initiated when the normal makeup supply system
l l
4-9 )
i l
(feedwater) is unable to prevent the reactor water level from falling below reactor water level 2.
This makeup water is supplied to the reactor via a bypass line which connects to the feedwater inlet piping via the RWCU/SDC retaining piping.
The SBWR does not have the BWR's residual heat removal system. For a normal shutdown and cooldown operation, the residual and decay heat are removed by the RWCU/SDC system and its condensers. The RWCU/SDC system provides two basic functions: reactor water cleanup and shutdown cooling. The RWCU/SDC system provides the core cooling one half-hour after control rod insertion.-
4.1.9 Spectrum of Postulated Breaks For the SBWR, breaks can be classified as large breaks, intermediate breaks or small breaks.
Table 4.8 gives a list of specific break and the flow area associated with each of them.
Table 4.8 Postulated Breaks (Ref. 4.1, p. 6.3-20, 6.3-21) ,
Equivalent Nozzle 2
Break Area (cm ) Size (cm)
Large Break
- DPV stub tube break 1056 36.61
- MSL break 977 35.22
- FWL break 390 22.77
- RWCU/SDC suction line break 295 19.36 Intermediate Break
- IC return line break 168 14.62 Small break
- GDCS injection line break 45.6 7.38
- Bottom head drain line break 20.3 5.08 4
4.2 Governing Processes and Phenomena 4.2.1 Processes Following a LOCA or Transient During a LOCA or transient (e.g. loss of feedwater) _the control rod drive system (CRDS) ,
shuts down the reactor. The primary function of the GDCS is to remove the core decay heat in
4-10 order to protect the com from uncovering and melting. The GDCS is designed to inject water into the vessel without relying on any active systems by using the gravitational head diffenmce between the GDCS tanks and the vessel. The reactor is operated at high pressure (1040 psia) which needs to drop to a value closer to the contain, ment pressure in order to gain a driving head difference between the GDCS tanks and the vessel. Therefore, before activating GDCS injec-tions, the vesselis depressurized using SRVs and DPVs.
The steam vented through the SRVs is sent to the suppression pool where it is condensed, but the steam vented through DPVs is added to the drywell, ..'hich raises the drywell pressure. To lower containment pressure, the PCCS condenses steam in condensers immersed in the ICS/PCCS pools. A mixture of nitrogen and steam is drawn from the containment atmosphere and is passed through condensers. The condensate is drained into the GDCS pools, and the non-condensible gases are purged into the suppression pool.
The SBWR containment is filled with nitrogen in order to prevent the combustion of hydro-gen that can form during core uncovery by the high-temperature interactions of steam and zir-conium in the core. Therefore, the non-condensibles expected to be present in the containment are nitrogen and possibly hydrogen. Purging non-condensibles into the suppression pool serves a dual purpose. First, the suppression pool water removes mdioactive contaminants in gases.
Second, the PCCS condenser performance deteriorates rapidly with the accumulation of non-condensibles in the tubes, thus purging will restore it to near pure steam environment.
The driving 4rce for the steam and gas mixture through the PCCS is provided mainly by the ,
condensation-induced pressure gradient. There is a pressure gradient between the drywell and suppression pool due to water level difference.
Long term core decay heat is removed in three steps. First, the GDCS injects water into the vessel, thus it removes core energy by boiling and venting into the drywell through DPVs, which ,
remain open once activated. Heat is removed from the core by natural circulation flow within the vessel. Second, the PCCS transfers energy from the drywell;o the PCCS/ICS pools by con- ,
densing steam frcm the drywell in the PCCS condensers. Third, the PCCS/ICS pools transfer j their energy to the atmosphere outside the containment by vaporizing pool water and venting it. l The PCCS also supports the heat removal process by feeding condensate to GDCS pools and by )
passing noncondensibles to the suppression pools which enhances condensation in the PCCS. l 4.2.2 Phenomena Accompanying LOCA or Transients ;
Inside the RPV, the steady state natural circulation mode can be significantly altered by a break flow which leads to inventory loss and a large pressure gradient within the vessel. At the same time, as the vessel depressurization occurs, the liquid is superheated and llashing can occur in the chimney section as well as in the downcomer.
l
1 1
4-11 l
During this early stage, in-vessel flow instabilities, such as manometric, geysering and den-sity wave instabilities, can occur due to the increased void fraction. The basic density wave ins-tability analyses, carried out in the 'early seventies, indicate that the two phase flow is much more unstable at lower pressure and higher void fraction. Hence, significant transient flow insta-bilities, including the parallel channel flow instabilities, may be encountered.
Outside the reactor vessel, several phenomena an: important. These are the steam mixing l and condensation in the containment atmosphere, and the condensation of steam in the suppres- !
sion pool. One important non-equilibrium phase change phenomena is the condensation of l steam by the subcooled GDCS water in the reactor vessel. This may lead to the condensation- )
induced water hammer phenomena or condensation flow instabilities.
Counter current flow limitation (CCFL) phenomena may occur within the reactor vessel.
This can lead to the voiding of some subassemblies while others are flooded. Since the natural circulation mode is relatively unstable due to the coupling of the flow and heat transfer (or void-ing), the occurrence of the instabilities such as density wave instability and CCFL are possible.
The other potential instabilities are manometer-type oscillations and geysering or flashing. The natural circulation instability studies by Ishii et al. [4.3-4.5) indicate that these instabilities can lead to very large amplitude oscillations or cyclic phenomena. Hence, their effects on the GDCS performance can be significant. l The following important goveming processes and phenomena should be considered in the model facility:
l Time-Dependent Vessel Water level: The vessel water level in the downcomer dropping below l the set limits actuates the ECCS (ADS and GDCS). During the operation of the ECCS, the core l natural circulation heat removal depends on the collapsed liquid level difference between the vessel and the downcomer.
Flow Rates of GDCS Water Draining and PCCS Condensate Draining: The difference between the draining rates of GDCS water into the vessel and PCCS condensate into the GDCS pools gives the net rate of emptying GDCS pools. This difference determines the the effectiveness of the GDCS at the later stage.
Thermodynamic State of Non-Condensibles: Presence of non-condensibles in the drywell degrades the performance of PCCS condensers. Therefore, the PCCS vents are designed to remove non-condensibles from the drywell into the suppression pool. The measurements of the pressure, temperature, and concentration of non-condensibles are useful to evaluate the con-denser performance. The concentration of non-condensibles may vary depending on the location in the drywell. -
l 4-12 !
RPVInventory with GDCS Flow: Since the GDCS flow will not be initiated until the RPV pres-sure is low, the steam flow is reduced significantly by the time GDCS injection begins. How-ever, it is possible that the GDCS water can be flashed into steam by contact with hot metal in the injection region. This may adversely affect the GDCS injection flow rate at the initial stage.
The GDCS injection flow, break flow and the DPV and SRV flow are the critical parameters in determining the vesselinventory.
GDCS Equilization Flow to the Vessel: The water inventory of the GDCS pools is slowly replenished with condensate draining from the PCCS. However, an effective long-term opera-tion of the GDCS is not possible because only condensed steam from PCCS can replenish GDCS inventory. Due to core boiling, the water level in the vessel will decrease. When the vessel water level reaches I m above TAF, the GDCS equilization line will open and water injection from SP to the RPV will begin. Because the driving head difference is small (0.91 m), it is important to know the core cooling c,:pability with the GDCS equalization flow.
Containment Integrity during a LOCA or Transient: During a LOCA or transient. SBWR safety systems operate continuously to remove decay heat from the core to the SP and outside of the containment through the PCCS. Steam is released from the vessel into the containment. The PCCS condenses steam and mitigates the containment over pressurization. The non-condensibles from PCCS condensers are purged into the suppression pool. The effectiveness of PCCS strongly depends on the proper functioning of this purging mechanism.
Thermodynamic and Thermal-Hydrmdic Conditions: Pressure and temperature of the drywell atmosphere are the main factors in determining containment integrity. The performance of PCCS condensers depends on the concentration of non-condensibles in the drywell. Water level and temperature in the suppression pool are important in determining the effectiveness of the i GDCS equalization flow for core cooling in the event of GDCS pool water depletion that could I occur in a long-term operation. l
)
Impact of Non-Condensibles on the Performance of PCCS and ICS: The presence of non-condensible gases is considered in the scaling because it can adversely affect the performance of i devices dependent upon condensation heat transfer. When vapor condenses on a heat transfer l surface, the concentration of non-condensibles increases. This layer of non-condensibles may insulate the surface from the vapor. In PCCS or ICS more global accumulation of non-condensibles occur. Therefore, it is necessary to continuously remove the non-condensibles and allow vapor to contact the heat transfer surface. PCCS venting into the SP is designed to serve this purpose. Situations which inhibit the removal of non-condensibles by interfering with the i PCCS vent flow to the suppression pool will also be studied. This includes the failure of the vacuum breakers to close.
l l
1 4-13 1
l l
Impact of Drywell Spray: Vacuum breakers will open when drywell pressure drops below the i SP gas space pressure. Their flow and opening characteristics are considered in the scaling. I Drywell pressure can be reduced due to an operation of the drywell spray which condenses a significant amount of steam from the drywell atmo," sphere. Thus, the drywell spray may lead to opening of the vacuum breakers.
A number of potential system interactions wen: identified by the NRC staff in their review of the SBWR testing program proposed by General Electric Nuclear Energy. These interactions include the connection of the DPVs and ICS pipes to a common stub tube, a situation which could allow some blowdown flow through the IC. The test facility will include this feature and will also be designed to allow the study of the combined injection from the GDCS and suppres-sion pools.
4.3 Identification ofImportant Phenomena For a scaled integral test facility, it is necessary to reproduce major themial-hydraulic phenomena ofinterest related to engineered safety features. This will ensure that the data can be used to assess the performance of the engineered safety features and to assess the model and code applicabilities for the SBWR safety analysis. For the present facility, the focus is on the phenomena expected to occur in the SBWR at a pressure of 150 psia or less after various accident initiations.
In identifying the key phenomena which should be reproduced in the test facility, two factors should be considered simultaneously. These are the importance of a phenomenon to an accident and the level of understanding of that particular phenomenon. Phenomena identification and ranking should address these factors. The scaling should focus on the highly-ranked phenomena that are the least understood. Such phenomena may not be quantitatively scaled because of the lack of understanding. Hon aer, it is necessary that the test facility be able to reproduce and retain the qualitative aspects ' these phenomena.
Since not all phenomena can be simulated in the test facility. the data from this test facility can first be used to assess the models in a code at test conditions. When the model can be vali-dated against data under integral test conditions, the scaling effect of that particular phenomena should be re evaluated in view of the reliable data and validated model. Thus, careful model i evaluations and the safety analysis code fills the gap between the integral test dati and the proto-type conditions.
In this section, a brief summary of the major phenomena of importance to SBWR safety is given. A preliminary BNL PIRT analysis performed for NRC as well as other PlRT analyses performed for GE, have been used together with in-house Purdue assessment of the phenomena.
Flow Instability in RPV: Due to depressurization, significant void fraction is generated in the
. . l l
4-14 I l
RPV. With increased void fraction, flow instability may occur. The natural circulation and break flow can be significantly affected by flow instabiliiy. The flow instabilities that are observed in such two-phase natural circulation flow systems are manometric oscillations, density wave oscillations, and flashing induced cyclic flow phenomena [4.7]. At low pnessure and low flow conditions, phenomena such as flooding, chan'nel-to-channel oscillation and geysering will also contribute to the oscillatory phenomena. The factors which can affect such oscillatory phenomena are density distribution in the downcomer, void distribution in the core, rate of remo-val of steam from the vessel and recirculating flow patterns in the core / bypass and chimney.
Simulation of void distribution and void propagation is the key to investigating these phenomena.
Blowdown Process: The blowdown process involves critical and suberitical flow of steam, water-steam mixture or water from the RPV. The critical and suberitical flow phenomena are well-known both for single-phase and two-phase. However, the fluid and energy inventory in the RPV depends on the blowdown process. The position of the water levelin the RPV triggers the activation of emergency core cooling systems (the GDCS and ADS). The blowdown process thus determines the boundary flow in various components of the SBWR and has the most significant impact on the mass and energy balance in the system. It also governs the pressure transient in the RPV, drywell and wetwell of the SBWR.
GDCS Flow: %e GDCS flow into the RPV begins when the RPV pressure is reduced enough so that there is a net higher head in the GDCS water. The draining of the GDCS water into the RPV at later stages of operation when the GDCS pools are replenished with condensate draining from the PCCS is also important. If the RPV level is reduced to the level LO.5, then the GDCS equalization line will open and injection of water from the suppression pool occurs. As the driv-ing head difference is small, the flow from the suppression pool is susceptible to manometric j oscillations.
PCCS Condensation: The PCCS is the ultimate heat sink for the containment, and in the long )
run PCCS performance determines the containment pressure. PCCS condensation efficiency is i dependent on the drywell steam non-condensible concentration. The type of non-condensible gases present is also important. If hydrogen is present in the drywell steam then its distribution in drywell, PCCS accumulation and purging process may be different compared to that of nitro- l gen.
Suppression Pool: The suppression poolis one of the two major heat sinks in the SBWR system. I It also works as the retaining tank for the non-condensibles gases. The PCCS performance depends on the effective purging of the non-condensible gases into the SP. The heat and mass transfer from the steam and non-condensible gases determine the SP pressure. Condensation of
4-15 i
I l
I I
steam with non-condensible in the form of jet flow or bubbly flow in the SP is an important phenc;aenon in determining the containment pressure.
9 i
a i
4 4-16 PASSIVE ISOLATION UPPER CONDENSER SHIELD CONTAINMENT DRYWELL COOLERS (TOTAL THREE) WALL (TOTAL THREE) . ,
i GDCS POOLS mim -
3,,m ---r---. _ (TOTAL THREE) 3' _
(.: I b ll /, Jj . .g f ! '
s DIAPHRAGM mum
/.,
m% '
, _ /4A m / FLOOR ll'/
- mm % /-_
. _.._, i
'm d L% ;
J /. '_ . ./_ ./[
== A .
GAS SPACE U' '
/~ .-
/.
e .
mm _
,.om - == i -
== .
d m
sum n
F1
% e = , .. = -stlalim ,
e ,. . . .,
- x =
a i ,,
- .s
'em _. - . . . . . _ .
,6i 8 g _.._ ._. _
~
nAnt inn i FVPt
/ 4 i .I _.N ELEV.10000 mso I l' . ,', -
v ~? .
k -
{N SUPPRESSION l
=
- l_ _ hl ,
b_ ,--, \ _7~
1 POOL
.iem h?? A l i i g
HORIZONTAL-VENT lL
, SYSTEM m ,
- / -
4' , a l
1XEC / .
l g i_ + sam j-j !f i
6 CONTAINMENT BOUNDARY LOWER ! ,mo l !
DRYWELL l
l Figure 4.1 SBWR containment boundary [Ref. 4.1, p. 6.2-113]
i t
4-17 ppy we l Steam eyer >
l a**g d l oupport om l Ed"If l
i j
[! Headvont l
/ .jf W-DPvac oumet "
T
'f a si l 7 e m now reemei.
- ~w
(~
. E4- d '-
Steam separators m fj "j siam - l N jn99n9L'
% ie i i , i i i ll l l l l l l ,
l
~+- *"*'* *a-o=
., L i i i i i i i L l i l l 8 Foodwater inlet Shh coonng _
yll ! [l ,
sunet _
- l. l I4l ~' SLC Inlet I CNmney restra.nt e
l CNmney IC retum l 4 i l ,
GDCS inlet Top guide
( l
) : ODCS equenzing e
hW l Support sw ,,,,,M , , , , , ; &,
n .
^
Fuel & contd rods - .
f Integraf fuel support , nn ,.
D -1l l l ! ll h:
~-
I l b l
< i ,/M CRD housing l
- '"8 :If l
4 i l .
) ! ! CRD restraint beam l
D'""' !dp '
i .:s J
i! l l
Figure 4.2 Cmss-sectional view of RPV [Ref. 4.1, p. 5.3-22] l
4-18 I
< 2d8'8 21267 1 i fif I) ld/
- ! \th l
q i ;
ic . 25- :_ 203:0 19935 19364
( ep._ _
s!
' bc ll l l l l l V ll l l l l l l !
< i i i i i i i --&-
I ! I I I ! l 17071 1M55 15996 *t _
15756 w -
i 4955 '
2 -
l 13501 12065 4s l, I <
t I
9493 @ I 1
7493 6934 /
[_
m (TAF)6493 lih 6100 5150j .
l (BAF) 3750 .
3523
- blll '
llk I 2146 j i 0 '
. Dimensions in millimeters
. i l 1, I l I li :i- 1:8
. )
i! I Figure 43 Cross-sectional view of RPV with dimensions
4-19 SHIELD VERTICAL UPPER WALL FLOW DRYWELL
\, CHANNELS (TOTAL 8) i 1200 .- /
/ IAPHRAGM FLOOR y
- E. /
N /
/ ONTAINMENT
' RPV f ALL
' i 1750 i m - -
i
, e
, /
i i NWL i
9 . . . u . 7_. . . , _ .
ELEV.10100 i
' ..1350
- F .
o
'
- 2
.: l 1370 I 'C l\.4 & ' TAF ELEV. 6493 ljll l l, i( s l[1370 760 ELEV. 4650
' \. ___
BAF ELEV. 3750 i N x
d EN S RPV 0.000 (TOTAL 24 =8x3)
VENT DIAMETER 700 s
hw l l
Note: Dimensions are in millimeters. Elevations relative to RPV invert EL 0.000 Figure 4.4 Horizontal vent module between the drywell and suppn:ssion pool [Ref. 4.1, 6.2-117]
4 20 References 4.1 GE Nuclear Energy,"SBWR Standard Safety Analysis Report." 25A5113 Rev. A August (1992). .-
4.2 U.S. NRC, " Request for Proposal", Requisition No. RS-RES-93-049, U.S. Nuclear Regu-latory Commission, Div. of Contracts and Property Mgmt. (1992).
4.3 Lee, S.Y. and Ishii, M., " Thermally Induced Flow Oscillation in Vertical Two-phase Natural Circulation Loop", Nucl. Engr. & Design, vol. 122, 119-132 (1990).
4.4 Lee, S.Y. and Ishii, M., " Characteristics of Two phase Natural Circulation in Freon-113 Boiling Loop", Nucl. Eng. and Design, vol. 121,69-89 (1990).
4.5 Lee, S.Y. and Ishii, M., " Simulation Experiments on Two-phase Natural Circulation in a Freon-113 Flow Visualization Loop", NUREG/CR-5082. ANL-88 (1988).
4.6 GE Nuclear Energy " Response to NRC RAI for SBWR Design", MFN No. 077-94, Docket No. STN 52-004, May (1994). ,
4.7 J.-T. Hsu and M. Ishii, " Experimental Study on Two-Phase Natural Circulation and Flow Termination in a Loop", NUREG/CR-4682, ANL-86-32, (1986).
I l
l
l 5-1
- 5. SCALING APPROACH 5.1 General Considerations An important part of the design of the SBWR. integral test facility is the development of a well-balanced and justifiable scaling approach. For this task, a three-level scaling approach is used: integral scaling, boundary flow scaling, and local phenomena scaling. The integral scal-ing is derived from the integral response functions for major variables in single and two-phase flow. This scaling ensures that both the steady-state and dynamic conditions are simulated. It also determines the geometric requirements and time scale. The integral scaling results in the simulation of all the major thennal-hydraulic parameters. The boundary flow scaling simulates the mass and energy inventory of each component and flow of these between components. The third level scaling ensures that key local phenomena can be simulated reasonably well. This is done by examining the constitutive relations, flow regimes and applicable correlations.
Once the test facility scaling is determined from the integral and boundary flow scaling, then the scaling for the local phenomena is considered. It is possible to encounter some distonions in local phenomena while simuhaneously satisfying the first two levels of scaling. Since the first two levels of scaling must be satisfied for a correctly scaled facility, the hical phenomena should be scaled as accurately as possible while remaining within the constraints imposed by the first two scaling levels. The local phenomena relevant to the SBWR facility are: flashing, choking (blowdown), bypass flow in the reactor core, circulation pattems (forced or natural), slip and phase distribution (flow regime), critical heat flux, condensation, mixmp, stratification stored energy and heat loss.
Once the integral system scaling is complete, the scientific dcign of the system can be per-formed. In this stage, various practical considerations, includiag the instrumentation, should be considered. The scientific design at the optimum conditions should then be translated into an engineering design which meets requirements such as state and local licensing codes, manufac-turability, operation and servicing of the test facility.
A well-scaled integral test facility will produce valuable integral experimental data that reproduces all the major phenomena of interest. However, neither the scientific design nor the engineering design can completely satisfy all the scaling requirements. Thus, some scale distor-l tions are inevitable, particularly in the third level of scaling. Distortions are encountered for two l major reasons: difficulty in matching the local scaling criteria and lack of understanding of a l local phenomenon itself. Therefore, the direct extrapolation of the experimental data to the pro- i totypic conditions is often quite difficult or impossible.
In analyzing data to obtain the corresponding prototypic conditions, the use of a computer )
code such as RELAP5/CONTAIN can be of great importance. The method that has been dev-ised for using the transient simulation code as a part of the scaling process is a useful approach,
5-2 and that will produce greater insight into the RELAP5/CONTAIN transient simulation capabili-ties and limitations. The method is based on the use of three separate analytical models: the pro-totype SBWR system model, the " ideal" scaled SBWR model, and the actual PUMA system model. The prototype SBWR model is based on the guidelines developed for applying RELAPS/ CON'IAIN in order to obtain the best estimate simulation. The " ideal" scaled model is based on the application of all the scaling' factors developed for the well-scaled SBWR experi-ment, again using the RELAP5/CONTAIN application guidelines. The third model is the appli-cation of RELAPS/CONTAIN to the actual experimental system and includes all the atypical conditions necessary or unique to the experiment. Some of the atypicalities that can be accounted for between the two scaled models will be: 1) the lack of thermal interaction between components caused by the use of a separated vessel rather than a concentric vessel integrated design for drywell,2) the difference in the core design necessitated by the minimum heater rod diameter and the cost of providing access to individual heater rods,3) the differences in the heat sinks of the vessel and piping walls, and 4) any difference caused by initiating the experiment at 1.03 MPa (150 psia) rather than starting from 7.16 MPa (1040 psia). Other minor differences can also be accounted for, such as any difference in boundary conditions, control systems, or differences caused by the instrumentation systems. ,
The advantage of this three-model system is that differences related to scaling issues can be separated from differences resulting from experimental compromises. These comparisons estab-lish that the experiments capture the qualitative behavior of the SBWR, and give some measure ;
of how closely the phenomena are simulated. The comparisons also assure that the experimental compromises do not qualitatively change the nature of the system response, and that the data can be used to assess the uncertainty in the ability of RELAP5/CONTAIN to predict the actual response of the SBWR system.
The final data analysis compares the PUMA data to the RELAP5/CONTAIN simulations of the PUMA experiments. Such an analysis should be supported by the basic modeling study addressing the true physical mechanisms at the local level. From this, the scalability of the local l phenomena in the code can be established. This establishes the ability of RELAP5/CONTAIN to simulate physical phenomena of this type. This ability may need to be increased or decreased based on the differences in response predicted to exist between the PUMA, the " ideal" scaled facility and the actual experiment, and the differences between the full-scale SBWR model and the " ideal" scaled model. As the code scaling and uncertainty analysis methodology [1.6] indi-cates, the code scalability critically depends on the two-phase flow models and scalability of constitutive models in the absence of near full scale experiments.
1 5.2 Scaling Approach The scaling criteria for a natural circulation loop under single-phase and two-phase flow con-ditions were developed by Ishii, et al. [5.1, 5.2, 5.3]. The criteria include the effects of fluid
5-3 properties, so one can also use them for redaced-pressure system scaling.
For a single-phase flow, continuity, integral momentum and the energy equations in one-dimensional area averaged forms are used. First, relevant scales for the basic parameters are determined, then the similanty groups are obtained from the conservation equations and boun-dary conditions. The hen transfer between the fluid and structure can be included in the analysis by using the energy equation for the structure. From these considerations, the geometrical simi-larity groups, fnction number, Richardson number, characteristic time constant ratio, Biot number and heat source number are obtained. It should be noted that the simulation of a long, large pipe section by a small scale model may encounter some difficulties if the prototype sys-tem does not have a reasonably large loss coefficient in addition to the wall frictional loss.
For a two-phase natural circulation system, similarity groups have been developed fmm a perturbation analysis based on the one-dimensional drift flux model. The set of mass, momen-tum and energy equations are integrated along the loop, and the transfer functions between the inlet perturbation and various variables are obtained. The scaling parameters are developed from the integral transfer functions, represent the whole-system similarity conditions, and are applicable to transient thermal-hydraulic phenomena. ,
The scaling approach that has been used for the design of many existing NRC thermal-hydraulic research facilities is summarized in an NRC NUREG Report prepared by Condie, et al.
[5.4]. The so-called " full pressure full-height method" was used for most of these facilities. The scaling approach recommended by the NRC, based on the experience accumulated from exten- !
sive LOCA studies in scaled integral test facilities, is summarized in a comprehensive paper by Boucher, et al. [5.5]. The present scaling method is an extension of the previously developed scaling approach by Ishii, et al. [5.1,5.2,5.3] and consists of three levels of scaling detail. First, i integral scaling methods are applied to the system circulation paths. Second, component boun-dary flow scaling considerations are applied in order to preserve integral mass and energy inven-tory. Third, scaling enteria are developed that preserve the similarity of local phenomena such ;
as choking, condensation and bubble rise time. These levels of scaling detail are described in the following section.
5.3 Global Scaling 5.3.1 Integral System Scaling (1st Level) !
It is imperative to have the single-phase flow similarity requirements as a ready reference, as they are needed to simulate the single-phase to two-phase flow transition. The system consists of a thermal energy source, energy sink and connecting piping system between components. For a natural circulation loop under single-phase flow condition the similarity parameters,are obtained from the integral effects of the local balance equations (continuity momentum and
1 l
5-4 l
l 4
energy) along the entire loop.
The fluid continuity, integral momentum, and energy equations in one-dimensional, area-averaged forms are used along with the appropriate boundary conditions and the solid energy equation. From the non-dimensional form s of these equations,important dimensionless groups characterizing geometric, kinematic, dynam,ic and energetic similarity parameters are derived.
They are given as follows: <
g ATolo Richardson Number, Rs (5.1) u o2 Friction Number, Fi n -+Kil d (5.2) 4hl' o Modified Stanton Number, Sti e (5.3)
, pfcpru od ,,
'lo/u o Time Ratio Number, T,i s 2 (5.4)
. b IA . i Heat Source Number, Q,i e (5.5)
, pxCps u o At, i ,3 where subscripts i, f and s identify the ith component of the loop, fluid and solid, and uo, ATo and t oand, temperature difference and equivalent length, respectively (for PUMA,lo is heated length and AT is temperature rise across core). The symbols appearing in the above set of equa-tions conform to standard nomenclature.
In addition to the physical similarity groups defined above, several geometric similarity groups are obtained. These are:
Axial Length Scale: L, a l /lio (5.6)
1 l
5-5 l Flow Area Scale: A ie ai a/o (5.7)
It is noted here that the hydraulic diameter of the ith section, d,. and the conduction depth,6,,
are defined by .
di e 4 ai /4; (5.8) and 6, s a,i %; / (5.9) where ai, a,i and (i are the flow cross sectional area, solid structure cross sectional area and wet-ted perimeter of the ith section. Hence, di and Si are related by di = 4(a/a.); S, (5.10) 9 The reference velocity, u o, and temperature difference, ATo are obtained from the steady-state solution. If the heated section is taken as the repn:sentative section, these characteristic parameters are expressed as follows:
m 40 lo a so 4 pg th u= PPf. "" . (5.I1) o ;
I Fi /A 2 and go lo a,o ATo= (5.12)
, Pr cu a
eo, r ,o, where the subscript o here denotes the heated section. ,
56 Eqs. (5.1) through (5.5) repn:sent relationships between the dimensionless parameters and the generalized variables characterizing the system under consideration. The similarity criteria between different systems can be obtained through detailed consideration of the similarity groups listed above, together with the necessary, closun: conditions. If similarity is to be achieved between processes observed in the protoiype and in a model, it is necessary to satisfy the following requirements:
A in = (ai/o a )g = 1 (5.13)
L i g = (l i/lo)g = 1 (5.14)
~ -
'I F /A,2 - t '.
i =
fd+K
/ (ai/o a )2 =1 (5.15) s 8
.R i i
. R Rg = (DATo lo/u 2)R = 1 (5.16)
Sti g = (hlo/ptcpru od i)g = 1 (5.17)
T[g = [(lo/uo)/(32/a s)i]g = 1 (5.18)
S i n = (hS/k,),a = 1 (5.19)
Q,ig = (q[lo/p,c ps u nAT,,)in (5.20) where subscript i designates a particular component and R denotes the ratio of the value of a model to that of the prototype, i.e.,
% v for model yn = yp ~- y for prototype (5.21)
5-7 As discussed in detail in [5.1-5.3], the frictional similarity requirement, Eq. (5.15), can be j satisfied independently of the remaining scaling requirements. Hence, from the remaining scal-ing mquimments, it can be shown that the following conditions should be satisfied for a com-plete simulation: ,
.,,w Pqo l o (uo )g = (5.22)
, P.Cp , j a qO O (ATo )g = (5.23)
, ExCpou sR 1/2 "a,lo (Si )g = (8)g = (5.24)
, " . R o
1/2 p,cp a,lo (di )g = (d)g = (5.25)
, Pr c e,g r ,
uo
,R "uo 1/2 (hi )g = (h)g = (k,)g (3.26)
,oIa s,g where the parameters withaut the component subsecipt, i, denote universal values that must be satisfied in all components. In addition to the above, the geometric similarity requirement, dic-tate that li a,
=1 and -
=1 (5.27)
,">RI
, .R must also be met. ,
With these conditions, Eqs. (5.22-5.27) and Eq. (5.15), the effects of each term in the conser-vation equ:.tians are preserved in the model and prototype without any distortions. If some of these requirements are not satisfied, then the effects of some of the processes observed in the
5-8 model and prototype will be distorted.
At this point, a few comments am appropriate regarding the practical implications of the similarity requirements:
- 1. The friction similarity may be difficult to satisfy, except in components having a subchan-nel geometry. Often, friction similarity imposes the most significant limit on the size of a scaled-down model [5.1-5.3].
- 2. The conduction depth ratio and hydraulic diameter ratio should satisfy certain criteria.
However, satisfying those criteria over the entire loop may be difficult. It is considemd that they are important naainly at the major heat transfer components where these condi-tions can be easily satisfied. However, the distortions in these criteria over a loop may lead to an overall scale-distortion in terms of structural heat losses. This should be cme-fully evaluated and compensated.
- 3. In contrast to the design parameters, the heat transfer coefficient cannot be independent of the flow field. Therefore, there may be some difficulties in meeting the constraint imposed by Eq. (5.26). Satisfying this condition depends on the flow regime. However, relaxation of this similarity requirement influences only the boundary layer temperature drop simula-tion. When the heat transfer mechanism is not completely simulated, the system will adjust to a different temperature drop in the boundary layer. The overall flow and energy distribution will not be strongly affected by the slow transients typical of a natural circula-tion system.
- 4. It is important to note that the above set of requirements does not put constraints on the power density ratio, gog. However, they do put a restriction on the time scale as follows:
lo lg o tg s - = , , ,
(5.28) u
,n,g [gq, /3)g/(p,c p,)gju3 The small perturbation technique and integral response function have been used by Ishii and Kataoka [5.1] to develop similarity criteria for two-phase flow systems. The important dimensionless groups that charceterize the kinematic, dynamic and energetic fields are given as follows:
4qo,,, Si n Ap Phase Cnange No. gNn= = Nzu (5.29) duoprAhrg) , ,
Ps j
l 5-9 i
This phase change number has been renamed as the Zuber number, Nzu n:cently in recog-nition of Zuber's significant contribution to the field.
.- Ahsub Ap
- Subcooling No. N ,uh n (5.30)
A hrg , pg u,o ' pr Froude No. Np, e (5.31)
Folco, ,
AP, d
Drift-Flux No. Nai e (or Void-Quality Relation) (5.32) un .
u a n T,i =
Ilo/u o ,
Time Ratio No. (5.33)
S2/a, ,,
' p,cpsS '
ThermalInertia Ratio Nth; e (5.34) -
, r pfped ,,
2 fl . I + X(AP/ Ps) . ao t Friction No. Nn m (5.35) d (1 + xAp/pg)"25 a, 2
3 a"
Orifice No. Noi e Ki [l + x /2 (Ap/pg )] --
(5.36) u
. n.
where Vg, Ahr ,s Ah uh, i and x are the drift velocity of the vapor phase, heat of evapora-tion, subcooling and quality, respectively In addition to the above-defined physical simi-larity groups, several geometnc similarity groups such as (li //o) and (ai /o a ) are obtained. ,
The Froude, friction and orifice numbers, together with the time ratio and thennal iner-tia groups, have their standard significance. Subcooling, Zuber and drift-flux numbers are associated with the two-phase flow systems. Their physical significance is discussed in
E D 5-10 detail elsewhere [5.1-5.3]. .
Eqs. (5.29) through (5.36) represent relationships between the dimensionless groups and the generalized variables of a two-phase flow system. The dimensionless groups must be equal in the prototype and model if the similarity requirements are to be satisfied.
Hence, the following conditions result:
(Nzo)g = 1, (N,os)g = 1, (Np,)g = 1, (Na;)g = 1 ,
(T*)g i = 1, (Nai)g = 1(Nn)g = 1, and (No)g = 1 (5.37)
It can be shown from the steady-state energy balance over the heated section that Nzo and N,os are related by :
Ap Nzo - N,os = x, - (5.38) ,
. Ps .
where x, is the quality at the exit of the heated section. Therefore, the similarity of the ,
Zuber and subcooling numbers yields -
t AP (x.)g =1 (5.39)
, Ps , ,
t This indicates that the vapor quality should be scaled by the density ratio. When com-bined with Eqs. (5.35) and (5.36), Eq. (5.39) shows that the friction similarity in terms of Na and Nc; can be approximated by droping the terms related to the two-phase friction multiplier. Furthermore, by definition it can be shown that Na. =
(Ap/pg x)[pg/Apa- 1]- 1. Therefore, similarity of the drift-flux number requires void fraction similarity Ap :
(a.)g =1 or (a )a= 1 (5.40)
Pr
, s R Excluding the friction, orifice and drift-flux number similarities from the set of similar-ity requirements, Eq. (5.37), and solving the remaining equations, one obtains the
l
. l
. l 5-11 i following similarity requirements:
(uo)g = (lo)[U (5.41)
AhrgP 8 (Ahsub)R = (5.42)
AP ,R (q )g = ( )R(lo'd' (5.43)
OP ,R Sa = (lo)[" (cz,,)[G (5A 0
"' (lo)[" (cz,)[G da = (5.45) e
, Pr pt , g The velocity scale shows that, in contrast to the case of single-phase flow scaling, the time scale for a two-phase flow is not an independent parameter. From Eq. (5.41), the time scale in two-phase flow is uniquely established. Thus, ta = lu = (lo)[U (5.46)
.".R This implies that if the axial length is reduce'd in the model, then the time scale is shifted in the two-phase flow natural circulation loops. In such a case, tSe time events are accelerated (or shortened) in the scaled-down model by a factor of (In g)3U over the proto-type.
It is important to note that when the two-phase f!ow velocity scale, Eq. (5.41), is used in the single-phase flow geometric scale requirements, the geometric similarity require-ments in both cases become the same. Hence, the same geometric scale can be used for single-phase and two-phase flows. However, using the time scale indicated by the two-phase flow scaling, namely Tg = Ql no , the single-phase time events are also scaled by the same criterion. This leads to the very important conclusion that for systems involving
5-12 both single and two-phase flow in a reduced length model, real-time scaling is not appropriate. l 5.3.2 Mass and Energy Inventory and Boundary Flow Scaling (2nd Level)
The scaled mass and energy inventory histories must be preserved for integral similar- l ity to be achieved. The integral system response scaling methods assure this senilarity when friction-dominated loop flow is considered. However, when vessel or system discharges occur that are dominated by non-frictional momentum effects, such as at a point of choked flow or any nozzle flow in which the pressure drop-flow relation is dom-inated by kinetic loss or by cavitation effects, then additional constraints apply. At such discharge points the fluid velocity depends upon the local pressure ratio across the device, which is preserved in a full-pressure scaled system such as the PUM A facility. In nonfrie-tional momentum-dominated flows, the fluid velocity is the same in the model as in the prototype. Therefore, the flow area at such discharge points must be scaled to preserve mass and energy inventory rather than loop kinematics. The purpose of this section is to develop the appropriate scaling relations to be applied at such points. This is particularly important for the SBWR experiment, since choked or mom'entum-dominated flows occur at the reactor vessel steamline discharge and at the safety and depressurization valve open.
ings. It is particularly important that similarity of these phenomena is preserved, since the timing of the SBWR safety systems is a strong function of the system pressure, and the depressurization system is activated based on the preset levels of the vessel coolant inven-tory. Furthermore, the ECCS tlaws begin only after near-equalization of the pressures in the reactor vessel and the containment. Thus, an overall criterion for similar behavior between the prototype and the model is that the depressurization histories he the same when compared in the respective (scaled) time frames, i.e.,
pJi m ) = p (tp p) $ 47)
This integral condition will be satisfied if the differential pressure change is the same at corresponding times, i.e.,
- P
=
(5.48) dt m dt p The scaling criteria for simi*arity of the fnetion-dominated natural :.irculation flows yields the result that the time scale c,f the model, or laboratory time, is related to the prototype time, by
5-13 tm = (1/2)tp = tgtp (5.49) and the depressurization rates of the model and the prototype are related by dpm dpp dpp (5.50) dtm = (1/ta) dtp = (2) dtp This condition will be satisfied if the corresponding component vessel inventories are similar, i.e.,
P 9 P 4 Mm Mp
=
V V (5.51)
. m. i. . p.9 where Mpand M arem the prototype and model vessel inveptory masses, and Vp and Vm are the respective prototype and model vessel volumes. This relation must hold for each component as well as for the overall system if complete similarity is to be ensured.
Mass inventory and Mass Flow Scaling For integral experiments, accurate simulation of the mass and energy inventory is essential. This requires a separate scaling criteria for the system boundary flows such as the break flow and various ECCS in.iection flows. The scaling criteria, stated in Eq. (5.51),
are obtained fro: the overall control volume balance equations.
i For the cools mass inventory, the total mass for a particular component is given by d
-M = I mm - Im ooi (5.52) dt By denoting the total volume by V and the mean density by
, the balance equation can be written in a non-dimensional form that applies to both the model and the prototype sys- ! tem as l d . .< > = Im,*n - I m *,oi (5.53) di 5-14 where t* a t/(Io/uo) (5.54) and . min to pi, ain uin mm s (5.55) pV p , a o, u o, where to = (lo/uo) for either the prototype or the model. The definition for m',u can be given similarly. For equal model and prototype pressure simulation, (p *,oi)g = (pour /p)g is simply unity. Hence, the simulation of the boundary flow requires ai oo uout =1 ' (5.56) ao U , .R This is a similarity condition for the flow area and velocity combined. Therefore, it is not necessary at discharge points to satisfy the independent conditions for area and flow given by Eqs. (5.27) and (5.41), which must be satisned by the other components of the loop. The form of the discharge scaling criterion given by Eq. (5.56) is very convenient from the standpoint of practical implementation. For example, the break flow velocity, uooi, can not be independently controlled if choking occurs. In the case of choking, Mach number similarity is maintained. Thus, for a equal-pressure system the break flow is prototypic in the sense that (unoi)g = 1, whereas the basic scaling (uo )g = (l o)jj2 and the criterion given in Eq. (5.50) predict that the break flow area should be scaled according to ) a l = (/og )v2 (5.57) l ^" .R I which would result in a reduction of the break flow area beyond the geometrical scale used for the loop flows. , i For the case of ECCS injection flows, the breaktlow scaling criterion is also very use- , ful. If the injection lines are scaled according to the geometrical scaling condition, Eq. (5.27), the line diameters become very small and the frictional resistance can be very l l 1 l 5-15 i large. This will result in mismatched ECCS injection flow, which is unacceptable. For-tunately, the boundary flow scaling criterion, Eq. (5.56), permits an enlarged flow area to obtain the correct volumetric or mass flow rate. Energy inventory and Energy Flow Scaling ' For the energy inventory, E, the control volume balance is given by dE - = q - w + Iminhin -I minhout (5.58) - dt d By non-dimensionalizing the above equation, it can be shown that the scaling criteria obtained for the natural circulation satisfies the similarity requirement for the heat input, q. The non-dimensional fann of the above equation is given by , = q' - w + Im[n h[n -I m *,ui h*,ui (5.59) dt where i m[n h[o = min h,n pVh,, -( p'")(b)( a,, ")( u,, h) (5.60) , in view of Eq. (5.56), for a full pressum simulation, i.e. (h,,)g = 1, it is necessary to require (h n)g = 1 (5.61) This physically implies that the inflow or outflow should have a prototypic enthalpy. The above non-dimensional energy equation also shows that the initial energy inventory should be scaled by the volume ratio. 5.3.3 Pressure Scaling
- The work scope and program objectives of the PUM A are focused on the low-pressure region of operation following the initial depressurization of the vessel. This implies that the prototype pressure maximum is about 150 psi (or 1 MPA). In considering the pressure
5-16 scaling of the integral test facility, two effects should be evaluated separately. These are:
- 1. System pressure level. .which affects all the thermal-hydraulic properties of the liquid, vapor and phase changes.
- 2. Individual component or inter-comporient pressure distributions.
Considering the pressure scaling in these two separate effects is somewhat analogous to the well-known Boussinesq assumption. As indicated in the report on pn:ssure scaling by Kocamustafaogullari and Ishii [5.3], the reduced-system pressure scaling or fluid-to-fluid scaling introduces large uncertaintis and scaling distortions due to the difficulties in matching all the non-dimensional groups under changing system pressure conditions. For transients involving system pressure changes, the adoption of the prototypic system pressure for a scaled model is the best and simplest approach. This guarantees that all the thermal-hydraulic properties are essen-tially prototypic, so that the system scaling criteria can be significantly simplified. Since the experimental focus is on the various low pressure phenomena associated with the emergency cooling systems and vessel depn:ssurization systems, the adoption of the prototypic pressure for the model integral test facility significantly increases the confidence level of the scaling approach and design of the scaled integral facility, as well as the usefulness of integral test data. Furthermore, the maximum pressure of 1.03 MPa (150 psia) is low enough that its impact on the overall cost of the project is small com-pared to a reduced-pressure model. The detailed scaling study carried out by Ishii, et al. [5.1-5.3] indicates that the reduced-pressure simulation is possible for a narrow window of pressure transients. However,it is very difficult to simulate all transients. In view of the above, the prototypic pressure is taken as the system pressure scaling base. Hence, the system pressure and all other fluid properties are considered to be proto-typic. This will greatly simplify the scaling procedures. Thus, we have the global pres-sure scaling given by pg = 1 (5.62) Under the above prototypic system pressure scaling, the thermodynamic and transport properties at every component are considered prototypic. However, the pressure distribu-tion in each component may not be prototypic. It should be noted that the pressure distri-bution within a component or between components can be the controlling factor in deter-mining the flow by forced convection or natural circulation. This aspect of the pre'ssure 5-17 effect in a reduced-height system should be considered separately. At the initial blow-down phase of a LOCA or other transient, the major intercomponent flow occurs due to the initial pressure difference between the reactor pressun: vessel and the containment. For this initial phase, the pressure difference between these two components should be prototypic at the same elevation. Thus, (A Pg)g = 1 at Zg = Ig (5.63) where the notation i and j stand for the reactor vessel and containment, respectively. However, in the case of natural circulation-dominated flow, such as the reactor vessel internal circulation, GDCS injection or PCCS venting, the hydrostatic head is the essential driving force. For this case, the differential pressure is scaled by the reduced height scal-ing. Hence, (A P)g = la (at AZa = Ig), (5.64) For all components where the flow is governed by the pressure difference due to hydrostatic head, the latter pressure scaling criterion should be used. This criterion is con-sistent with the hydrostatic pressure distribution within each component, and guarantees the proper intercomponent flows driven by the gravitational head. A significant deviation or distonion from this differential pressure scaling at any component m. lead to incon-sistent flow among components and may destroy overall scaling of integral phenomena of interest. For the PUMA, the initial differential pressure scaling is set by the initialization pro-cess with isolated components. At the later stages of accident simulation, most of the significant liquid flows between components are driven by the hydrostatic head. These flows are accurately simulated by using proper height scaling of all major elements and components based on AZg = / g (5.65) which implies the complete axial geometrical similarity. This condition, together with the void distribution simulation based on the integral scaling, insures that the differential pres-sure is scaled by the reduced height scaling. 4 5-18 5.3.4 Basis for Reduced Height Scaling Under the prototypic pressure simulation, the system geometry can be determined from the integral system scaling and the boundary flow scaling discussed above.. The dynamic scaling requirements for a two-phwe flow system are given by Eqs. (5.29-5.36). In general it is difficult to match all these similarity criteria for a scaled down system, so a careful evaluation of each of these requirements should be made. Based on the original scaling study [5.1-5.3] it is evident that the Froude number and Fdetion number scale the gravitational driving force and frictional resistance against the inertia term. The Zuber (phase change) number and subcooling number scale the energy transfer for a boiling process. It is essential that these latter numbers are satisfied for the energy and kinematic similadties. As indicated by Eqs. (5.38-5.40), these two similanty criteria give the simulation of the void fraction and the steam quality under the prototypic pressure simulation. In considering the dynamics of the system, two conditions should be considered separately. The first is on the quasi-steady flow simulation and the second is the dynamic response of the system, including the inertia effect. It is crear that the Froude number and fdction number scale the dynamic response. When the inertia forces are not important, only the balance between the frictional resistance and gravitational force should be con-sidered. This can be achieved by taking the product of these two numbers. Thu.., natural circulation number is defined as friction inertia N,m = Nr NF , = (5.66) interia gravity head This equation can be extended to include the minor loss coefficient as N,m = (Nr + No)Np, (5.67) In general, the requirement of i N,m)g = 1 or < (Nr + No )Nrr - =1 15.68) . .R l is less restrictive than (Nr)g = (No)g = (Np,)g = 1. 5-19 However, the energy and kinematic similarities require that the velocity be scaled by Eq. (5.41) and the void fraction by Eq. (5.40). Under these conditions, it can be shown that (Np,)g - - 1 (5.69) . Si ll bP,a Hence Eq. (5.68) can be reduced to (Nr + No)g = 1 (5.70) Combining the above equation with Froude number similarity, is seen that these two also constitute an approximate dynamic similarity between the inertia term, gravitational term and flow resistance. The advantage of Eq. (5.70) relative to the two independent requirements of (Nr)R = 1 and (No)g = 1 is significant. Under a homogeneous flow assumption, the requirement given by Eq. (5.70) can be approximated by ,2 fl no (Nr + No)g = -+K - =1 (5.71) d a . .R 8 .R By using the geometrical similarity criteria, il -+K =1 (5.72) d . .R A careful analysis of Eq. (5.72) clearly indicates the great advantage of using the i reduced-height system for a given volume scale in satisfying the dynamic similarity cri-teria. By reducing the flow area, the hydraulic diameter is reduced by da = (, except at bundle sections such as the core. For most small integral test facilities, it is necessary to have la > da in order to maintain a reasonably large axial height so that the naturally existing two-phase level fluctuations do not adversely affect various transient phenomena. l In general, the ratio of the first friction term itself is always larger than unity. However, by reducing the height of a facility, this ratio can be made closer to unity by increasing da for a fixed value of vg. The second significant point is that the minor loss coefficient is an 9 5-20 1 I 1 easy parameter to adjust through small design modifications in such a way that Kg < 1 to compensate for increased friction. Henc: by properly modifying the K value, Eq. (5.72) can be achieved. In view of the above and the cost consideration, the volume scale of 1/400 and the height scale of 1/4 appear to be most desirable for the Purdue integral test facility. This implies the general area ratio of 1/100. A more detailed discussion of the system geometry and other considerations is given in Chapter 6 and 7. However, for the subse-quent local scaling phenomena analysis, these geometrical scales are used as reference conditions. 5.4 Local Phenomena Scaling (3rd Level) 5.4.1 Reactor Vessel Flow Dynamics and Instability Scaling The dynamic behavior and stability of a boiling flow system can be analyzed by using a one-dimensional drift flux model [5.6] and a small penurbation method [5.7-5.9]. A per-turbation of inlet flow is introduced, given by , Sv(t) = Ee*' (5.73) where s = a + Jco. Thus, s is a complex number; the real pan gives the amplitication coefficient and the imaginary part represents the angular frequency,(o. By formally integrating the four differential balance equations in the one-dimensional drift flux model, various transfer functions between major variables. such as the velocity, void fraction, density, enthalpy, pressure drop and inlet velocity, can be obtained. These can be expressed symbolically by Sr (s z, t) = Q(s, z) Sv (s, t) (5.74) l 1 where Q represents the transfer function and Sf is the perturbed part of the variable, f, at location z. It has been shown that both the dynamic and transient response of the system , are governed by the transfer function between the internal pressure drop over the system and the inlet flow [5.7-5.9]. Thus, the most important relation is given by , I SAP (s, t) = Q (s) Sv (s, t) (5.75) l J 5-21 In a real physical system, the perturbation comes from the boundary condition on the pressure which induces flow change. Thus, 1' Sv = - (5.76) Q(s) SAP The dynamic response depends on the form of the transfer function,1/Q(s), whereas the linear stability of the system depends on the root of the characteristic equation given by Q(s) = 0 (5.77) The characteristic function, Q(s) can be nondimensionalized by introducing proper scales for various variables [5.7-5.9]. Thus Q*(s*) = Q*(s** Ni , N2. Nm) (5.78) where Ni , .... N mare the same non-dimensional groups listed in Section 5.3. This indi-cates that the dynamics of the system can be simulated if the scaling parameters, N i .. Nm, an: identical between two systems. The disappearance of z in Q is due to the fonnal integration of the momentum equation over the entire system length. Since our general scaling is based on these non-dimensional groups, as discussed in Section 5.3, it is assured that the dynamic behavior and instability characteristics can be well-simulated in the reactor pressure vessel of the integral test facility if the flow can be considered close to one-dimensional. The SBWR reactor pressure vessel has a relatively ; large I/d, which causes it to be dominated by the axial flow. Furthermore, the present reduced height system has the aspect ratio reasonably close to the prototype at (l/d)g = 2.5. This number, which is a good scale of the three dimensionality of the facil-ity, is better matched than most of the full-height integral test facilities for similar pur- q poses. 5.4.2 Choked Flow Case The RPV is depressurized by the discharge of steam or water from a break and by the l SRV and DPV flows when these systems are activated. In the early phase of the depres-surization, the upstream pressure is sufficient to cause sonic velocity at the minimum
- area section of the steam line venturi, or at the throat of the SRV or DPV, and at the break loca-tion. A bottom drain line break (BDLB) also results in choked tiow when cavitation occurs at the throat or minimum area in the break line. For these cases, the velocity at the
5-22 i l 1 l l break or the throat will be the same in the model as in the prototype since the pressun:s i and thermodynamic conditions are the same. The flow in short nozzles and valve contrac-tion sections can be considered to be nearly frictionless and adiabatic so that an isentropic model of the flow process is a good approximation. The prototype-to-model ratio of velo-city multiplied by the area should equal 200 ih order to satisfy the conditions for similarity of the mass and energy inventories as previously discussed. In the choked flow case, no additional restrictions on the geometry or loss mechanisms are required. However, as we shall see, additional restrictions will result from the need to preserve the pressure or pres-sure ratio at which the transition from choked flow to unchoked flow will occur. Two quite different nozzle geometries are found in the SBWR system. Grst, the steamline contains a converging-diverging de Laval type nozzle that is designed to limit the steam discharge rate in the case of a main streamline break, yet result in little pressure loss under normal operation. The low pressure loss under normal operation is achieved by the use of a low-angle conical diffuser downstn:am of the nozzle throat so that flow separation is avoided and good pressure recovery is achieved. This type of nozzle will become choked only when a modest drop in the discharge pressure occurs as a result of a break or other decrease in the downstream flow resistance.'The flow will remain choked until the upstream pressure drops to near the downstream value. This process is illustrated in Figures 5.1 and 5.2 for the case of isentropic flow through a converging-diverging noz-zle. Figure 5.1 is a surface plot of the static pressure ratio (ratio of static pressure to isen-tropic stagnation pressure) throughout the nozzle for a range of the exit pressure ratio. Notice that the throat pressure ratio drops as the downstream pressure ratio is lowered, until sonic flow is reached at the throat. From that point on. the upstream pressure distri-bution is constant independent of the downstream pressure (choked). The pressures in the downstream, or diverging section of the nozzle have a rather complex behavior due to the possibility of a normal shock (the discontinuous rise in pressure shown in Figure 5.1). The flow velocity changes from supersonic to subsonic across the normal shock. Note that subsonic diffusion downstream of the normal shock results in an increase in the static pressure as the flow velocity decreases. At a sufficiently low discharge pressure, the nor-mal shock will occur downstream of the nozzle exit. Thus, we see that it is possible for I such a nozzle to be choked under a wide range of pressure ratios. This is more clearly illustrated in Figure 5.2, where the nondimensional mass flow parameter is plotted as a function of the ratio of the nozzle exit pressure to the upstream stagnation pressure. Note that the mass flow rate is constant up to a pressure ratio of approximately 0.98. The SRV and DPV systems have a different nozzle geometry that consists of a smooth contraction down to the throat followed by an abrupt increase in the flow area. This type of nozzle has quite a different flow characteristic. The abrupt increase in flow area down-stream of the throat results in large pressure losses due to the irreversibility associated I 1 5-23 with turbulence downstream of the abrupt increase in area. The static pressure ratios through such a nozzle are illustrated in Figure 5.3 as a function of the downstream pres- ; sure ratio. Compare Figure 5.3 with 5.1 and r.ote the considerable drop in the discharge pressure that occurs prior to reaching the ch,oked condition, indicated by the constancy on the upstream pmssures. In this case, the downstream pressure losses have been modeled using the Bourda-Camot loss model for an abrupt area change. The dramatic difference in the flow characteristic between this nozzle and the converging-diverging nozzle previously discussed can be seen by comparing the non-dimensional mass flow as a function of pres-sure ratio (compam Figure 5.4 with Figure 5.2). In this case, choked flow is maintained only up to a pressure ratio of 0.56, which is characteristic of an orifice or any discharge point having an abrupt area change. In order to maintain similarity in the scaled mass and energy inventories it is necessary to scale the throat area between the model and the prototype by the ratio 1/2(X). However, in order to assure correct transition to subsonic or unchoked flow, it will also be necessary to preserve the diffusion characteristics of the downstream section of the nozzle. This requires that geometric similarity be maintained, and to a lesser degree that Reynolds similarity be maintained. Thus, the nozzle contour and especially the diverging section cone angle must be geometrically similar in the model and the prototype. For a break of flow area, ou a i, with break flow velocity, unoi, the boundary flow scaling requin: ment is given by (anoi uoui)R (Do Uo )R * (UdR (u o)R = 1/200 t (5.79) For critical flow, the ratio of velocity at the throat is given by (ui)g = 1. From boun-dary flow scaling, (a, ui)g = 1/2(X). Since the model has prototypic pressure, the density ratio pg = 1. Thus, the area ratio is (at)g = 1/2(X) (5.80) This shows that the throat area where chocking occurs should be scaled differently from loop sections in which (ao )g = 1/l(X). This can be accomplished by using a nozzle of throat area ratio (ai)g = 1/200. One additional case needs to be considered, and that is the case of a cavitating venturi such as would occur at a bottom drain line break (BDLB). This case is more complex *than the ideal gas case just discussed. However, the considerations are very similar and the resulting conditions for similarity are the same. The reason for this is that even though the choking phenomena is due to the onset of vaponzation caused by the lowering of the static 5-24 pressure to less than the saturation pressure, the pressure behavior is primarily governed by the same inviscid, or in this case Bernoulli, flow phenomena. The pressure recovery / loss mechanism that occurs downstream of an abrupt area change is more com-plex than the single-phase flow case, but it too is mainly governed by the geometry and is thus is approximately simulated by preservin'g geometric similarity. 5.4.3 Unchoked Flow Case For non-critical flow, the size of the connection line is determined by the loss coefficients and the total pressure diffen:nce between the ends of the line. The actual requirement can be obtained by the total loss coefficient of the line. Thus, 2 pum Apm = Km (5.81) 2 and a 2 pup App = K p (5.82) From pressure scaling Apg = 1, or the inter-component pressure difference is prototypic.
- Thus,
,2 K" = u" 1 = -- = [200(a.,)g ]2 (5.83) k,r u ug ,m, With a nozzle of throat area, ai, the total loss coefficient, K,is given by ,2 Bout K = K ,ut + K i . (5.84) an l Therefore, l l \ 5-25 ,2 atcu Kooi+ K i a, > g* , , ~" =7 (5.85) K p* , 2 Doui Kooi+ K, ai " . .p In general, (Kooi)p < 1 and (K )p = 2. For the model (Koot)m = 10, the typical diameter i ratio is 1/10, and (Ki ) = 2. This requires that ,2 -
- 2 Dout 3out
>> (5.86) I I . .m . .p , r Since (ai)g = 1/2(X), the above requin: ment can be satisfied if ou (a i)g = 1/l(X). In fact, this is in agreement with the geometrical scaling. In conclusion, for the break flow or for the flow in ADS lines, the pipe size area is scaled as (aoui)g = 1/l(X) (5.87) and the nozzle where choking occurs is scaled with throat area ratio (ai)g = 1/2(X). The flow in the GDCS injection lines is essentially single phase liquid. A simple momentum integral equation for these lines is given by e 2 du ff p 7 + pu2 1 ~+K + (E" ~ E') = pgH (5.88) where I, H and d are the line length, driving head and hydraulic diameter, respectively. Eq. (5.88) shows that the liquid hydrostatic head controls the GD.CS tiow against the liquid inertia, friction and pressure difference between the vessel and containment. For the fully depressurized case under quasi-steady state condition, the flow is deter-mined by the balance between friction and gravity. Thus, the velocity scale ratio is given by 1 l 5-26 1/2 H ua= ft/d + K (5.89) . s R Furthennore, the mass inventory scaling reg'uires that as u =1 (5.90) ' ,oa u o,, where ag is the line flow area. By combining these two requirements, the line size can be obtained: a oa l$ aya = m (5.91) H f//d + K , , which clearly indicates that the GDCS line flow area scale ratio, a gg, should be increased to compensate for the increased friction number. This increase in the flow area beyond the geometrical scale, aog, guarantees that the mass flow rate is scaled to preserve the mass inventory. A similar criterion applies to the PCCS draining lines. Similar scaling calculations can be made for PCCS venting by considening the hydros-tatic driving head at the suppression pool (the liquid level difference between the vent exit and the vertical flow channels). If this height, h,is scaled according to the basic geometri-cal scale, then Hg = log. Thus, the single phase gas flow scaling requires that il an=ag v n -+K (5.92) . .R This relation shows that the venting line flow area should be increased to compensate for the increased friction for a scaled model. The drywell/wetwell venting depends on the pressure difference between the suppres-sion pool and containment, as well as the water height above the vents, H. Hence, . l \ I 5-27 - Pc - P up 2 H p g (5.93) For a reduced height system there are several approaches for setting this pressure differ-ence. However, the best way to scale this Ap is according to the height scale as discussed in Section 5.3.3, thus (pe - Puup)R = lor = 1/2 (5.94) because for most components such as the vessel, suppression pool and GDCS, the pressure distribution is close to the hydrostatic. By scaling the pressure difference using log, all the component pressure balances in the containment can be easily preserved. It follows that lin = lon (5.95) . The size of the opening can be determined by a scaling' criterion similar to that of the PCCS venting. 5.4.4 Relative Velocity and Flow Regime The present scaling method, based on the integral response function. implies that the parameter given by the following expressions are correctly scaled: Apx/p, = Nzu-N,un (5.96) and APX/P; =1 (5.97) . R ; \ i For the full pressure scaling, this leads to the similarity of vapor quality given by lx)g = 1 (5.98) The two-phase llow regimes depend on geometry, vapor quality and relative velocity. The effect of the relative velocity is scaled by the drift flux number given by I i 5-28 I Na = (5.99) uo l i In the drift flux fannulation, the drift velocity is given in the fami [5.6]. l l l Vg= Co-1 j + <Vg > - (5. l(X)) where the total volumetric flux is l l l - . j=uo 1 + Apx/pg (5.101) Hence, the drift flux number becomes Na = (Co-1)(1 + Apx/pg)+ av gin (5.102) u o' The right-hand-side terms represent the slip due to void and velocity profile and the hical slip between phases, respectively. Since the distribution parameter, Co,is a weak function of the density ratio [5.6], the first term is completely scaled. The local drift velocity depends on the two-phase flow . regimes and the hydraulic diameter [5.10]. The major effect of the local slip is mostly-limited to the chimney section of the RPV. For a relatively large-diameter channel, the local drift velocity [5.10] is approximately given by <Vgi> = 0.54h for d < 30 (5.103) GAP , and i 'oghp'1/4 a <V g, > = 3 2 for d > 30 pg gap The former applies to slug flow and the later to cap bubbly flow. This shows that for a channel having d smaller than 30V(c/ gap), the bubbly flow becomes slug flo* at a . higher gas flux, whereas for a larger channel the bubbly flow becomes cap bubbly flow. 4 l l l 5-29 l Therefore, for a chimney section of an integml facility it is necessary to have i 1 (d)cnim > 30, (5.104) CPA in order to simulate the two-phase flow regimes. A hydraulic diameter smaller than above leads to a formation of slug flow accompanied by a cyclic flow behavior which is a charac-teristic of small channels. However, the simulation of the two-phase flow regime in the chimney section leads to some distortion of the drift flux number. This is due to the fact that the reference inlet velocity is scaled by uon =l$, but the local relative velocity itself remains prototypic. This will lead to some distortion in the void fraction also, because it can be expressed as 8 (5.105) a = Co j + <Vgi:> < where j, = u (Apx/pg). o However, it can be seen that the effect of <<V gj >> is small if j is large. Comparison of Eqs. (5.89) and (5.92) indicates that the impact of distortion in <<V gj >> / uois much smaller for the void fraction than for the drift flux number. S.4.5 Critical Heat Flux Scaling (CHF) The CHF condition at low flow has been reviewed by Leung [5.11], Katto [5.12], and Mishima and Ishii [5.13] among others. The modified Zuber correlation [5.10] for low flow is given by l/4 096P qs." = 0.14(1-a) pgAhrs 2 (5.106) .P G . Based on the limited data on blowdown experiments, the above correlation is recom-2 mended for the mass velocity range of -24 (down flow) to 10 g/cm s {5.13]. It is evident that this correlation is based on a pool boiling CHF mechanism and has been developed for LOCA studies. Katto's correlation [5.12] for low flow is given by 5-30 ) 1 do op ' O.043 Ah,os 9e" = 7 Ahrg G + G 2f, Ahrg which implies that the critical quality is xc = (op/G2 7,)o". Here G and Ah,os are the mass velocity and inlet subcooling. The typical value of xyis 0.5 to 0.8, so the underlying mechanism should be the annular flow film dryout. This correlation can be applied to most slow transient situations at low flow, except very low flow near flow stagnation. However, there is a possibility that the CHF may occur at much lower exit quality than given above, due to a change in two-phase flow regimes [5.14]. In a natural circulation system with very small flow fluctuations, the occurrences of CHF have been observed at the transition between the chum-turbulent to annular flows. Beyond this transition, the lack oflarge disturbance waves eliminated the pre-existing rewetting of dry patches. This leads to the formation of permanent dry patches and CHF. The criteria developed by Mishima and Ishii [5.13] for this case is given by d 1 4c" = * {-1 . Ahr, ypgAppd,, + GAh,os - (5.108) k 4 Here, Co is the distribution parameter for the drift-ilux model, given by C., = 1.2 - 0.2 yp,/p (5.109) These CHF criteria should be used to develop a similarity criterion for the fluid solid boundary instead of the heat transfer coefficient. This ensures that the CHF occurs under similar conditions in a simulated system. The CHF number is given by qc Ny= ,, (5.110) q<, For proper scaling of the CHF it is necessary to have . . ~ _ _ . _ _ . 5-31 9cR (Ny )g = 1 or =1 (5.111) QR The substitution of a particular CHF correlation yields the actual value of the CHF i number ratio. Thus, by using the scaling parameters obtained from the integral response - I function approach, . (Ny )g = 1 = l[/2 (Zuber Modified) 9R VR dR (Ny )g = = da (Katto) 9R IR dg/2 (Ny )g = /2 - da (Mishima) The above expressions show that it is difficult to satisfy the CHF number scaling criteria-for a model which does not have a prototypic bundle geometry (i.e.. /g = 1 and dx = 1). However, for most of the transients and accidents of importance in SBWR safety, the reactor power is scrammed and it stays at the decay heat level. The focus of the present integral tests is on the latter part of the RPV depressurization and subsequent natural circu-lation stages for various emergency cooling systems. At this level, the dryout in the proto- - type is mainly caused by the uncovery of the core due to the drop of the mixture level to the top of the con:. This type of CHF may be approximately predicted by the modified Zuber correlation. The CHF number for this correlation indicates the dependence on the length ratio,Ig. However, in reality it is dominated by the change in the liquid concentra-tion (1-a) between the liquid-continuous to vapor-continuous regimes (see Eq. (5.100)). Hence, this type of CHF can be correctly scaled if the void fraction is properly scaled. The Katto and Mishima correlations show that by using a larger hydraulic diameter in the core section, the premature CHF by other mechanisms can be avoided in a model. - 5.4.6 Flashing in the Chimney The above scaling criteria for a two-phase natural circulation loop were obtained by considering the phase change (or boiling) due to the heat addition in the heater section. This effect is represen'ted by the standard phase change number (or Zuber number) given , by Eq. (5.29). However, in the SBWR geometry, significant flashing or vaporization due to depressurization of two-phase mixture in the chimney section is possible because of the 1 l l 5-32 decreasing hydrostatic head along the axial elevation. Hence, it is necessary to introduce the second phase change number based on the depressurization superheat. By denoting the pressures at the bottom and top of the chimney section as pi and p2, the corresponding saturation temperatures.are given by iT = Toi(pi) and T 2= Tot (P2), where T = Toi(p) is the saturation relation. The flashing phase change number is given by cr >f (Il - T2 ) Ap (l-cz) Naa = (5.112) Ahts Pg This number represents the non-dimensional scale for the magnitude of the vapor volume generation due to flashing. The superheating (T -T 2) is directly related to the hydrostatic pressure decrease (pi - p2) through the Clausius-Clapeyron equation. For a reduced-height scale model, the length of the chimney section is smaller than the proto-type system. Hence, the magnitude of the flashing is also smaller in the reduced-height model at full-pressure simulation. It is noted, however, that the impact of the flashing can be completely simulated if an appropriate reduced press 6re level is used in the experi-ments, due to the density ratio factor in the above Naash expression. By using a reduced-pressure, the ratio Ap/pg can be made larger in a model than in a prototype to compensate for the reduction in (T -T 2) or (pi - p2). 3 The above option is considered as a part of the test matrix. The preliminary calcula-tions show that about a fifty percent reduction in the absolute pressure causes the flashing phase change number to be rnatched. 5.4.7 Condensation in Suppression Pool The scaling of the condensation phenomena in the suppression pool can be divided into two sections: one related to the injection from the SRV spargers and the other related to the injection from the vertical vent pipes. The essential parameters important for the condensation phenomena scaling are the condensation power and the vapor latent heat energy flux given by e = ai hmna(Tut-Tg,i ) (5.113) l and I 1 .5-33 - E, = p,Q, Ahr, (5.114) i where ai, hcond and Qg are the total interfacial area, condensation heat transfer coefficient and total vapor volumetric flow rate, respectively. For most cases, h na, (Tu -TM) and Ahr can s be considered approximately proto-typic. Hence, for the proper scaling of the condensation phenomena, it is required that ; Qe DiR -1 (5.115) E, Q,R , ,R where Q,g = (ainuin)g. The product of the area and velocity ratios is determined by the integral scaling method discussed in the previous sections. Thus, Q gR = (3snuin)R hor lor (5.116) This reqbires that ; DRi 3oR lop, (5.117)' . i For the sparger in the suppression pool, the total bubble interfacial area can he scaled by a,=nnND3 (5.118) where n. N and D3are the total number of sparger holes, the number of bubbles which can exist in the vertical line above each hole and the bubble diameter. By using the frequency of the bubble generation at a hole, f, /7 f N= (5.119) us -v. , 5-34 , where ir is the pool depth and un is the bubble rise velocity. The frequency is given by Q, 6Q, f= = (5.120) nV 3 . xnD$ Here V3is the volume of a bubble and Q,/n is the volumetric flow rate at each hole. Sub- . stituting Eqs. (5.113) and (5.114) into Eq. (5.118) yields, 6 Ir Q, ni = uD (5.121) n 3 Thus, en I TR = < (5.122) E,a (ung(Dng For the evaluation of condensation phenomena scaling, u n and Dn should be given. The relation between the Dn and the volumetric flow rate is given by a simple model, l/5 '2 Q8 1-Dn = 1.5 - - (5.123) - n p , where a cap bubble rise velocity and the maximum frequency has been used. The rise velocity of a bubble is given by ; un = 0.71 QgDn (5.124) where Dn is the volume equivalent or drag diameter of a cap bubble. Thus, cR ITR = i =1 (5.125) E,a De,$ I y r-m+p - - 5-35 By substituting the expression for Dna from Eq. (5.123);nto Eq. (5.125), ng = Q,n 17 (5.126) The above correlation gives the correct number of sparger holes for a model. For the present model, Qga = 1/200 and I rR = 1/4. Thus, nm ng - - I (5.127) np 20 This relatively large number of holes compensates for the reduced height of the suppres-sion pool. Little information is availabk in the literature on condensation in pools in the presence of noncondensibles. What information is available, is considered proprietary. Recently, results from an experimental study on condensation in the presence of noncondensibles under chugging condi' ions have become available [5.15]. Based on a transient conduction-diffusion nmdel, a transition criteria for the onset of chugging with the effects of noncondensible gases has been presented as follows:
- os kw Ja 6, ' D** (5.128) 0.06 3+ Ref Pr"5 Ja-' < l .0 kj 2 6, aw x" where Du is the mass diffusion coefficient of steam through air.
i Nearly all the parameters in the above criterion are property-dependent and can be considered to be prototypic in the model. Only the steam Reynolds number contains scal-l able velocity and length parameters. The velocity is based on the steam velocity at the l exit of a vent opening, and the length scale is simply the vent opening diameter. Hence, the model-to-prototype ratio of the above criteria is given by '05 o3 . , o3 ; Rey = U, D, = - . .R . R 20 + The above result indicates that the transition criteria for the onset of chugging in the I 5-36 presence of noncondensibles may not be properly scaled. Liang and Griffith [5.15] have also pn:sented a criteria to detennine the critical air content in steam mixture condensation required for the diffusion resistance mode of con-densation, which would begin to dominate.outside conduction resistance due the heated thermal layer: kj 2 6a (If xn2 k, Ja S, Du It should be emphasized that the transition criteria given above are the only ones available in the literature, and the validity of transition criteria Eq. (5.129) under various test condi-tions has yet to be established. 5.4.8 Vent Phenomena in Suppression Pool In the initial blowdown stage, the total gas flow throug6 the vertical flow channels and horizontal vents into the suppression pool is expected to be very large. Under this condi-tion, the vent flow forms a horizontal jet at the exit of the horizontal vents. The axial extent of this jet depends on the condensation rate, concentration of non-condensible gas and jet instability. For the scaling of the condensation phenomena for this horizontal jet, the first criterion to be satisfied is the one given by Eq. (5.115), which scales the global interfacial area for condensation. By denoting the width and height of the horizontal vents by W in and l in, respectively, the total exit flow area is given by I a in = n,n(Winl in) (5.130) From the boundary flow scaling. the gas volumetric flow rate should satisfy QpR (U inuin)R aRo l oR (5.131) l Hence, for PUMA one has Q,g = 1/200. For the prototype system having circular vents of a diameter DJr , the interfacial area is .- - , - . . . . = . . - - - .- - . . . _ . _ . . i 5-37 s a;p = (nnDjls)p (5.132l For the model system, interfacial an:a is given by , i aun = (2n(W in + l i n)lj }, (5.133) However, scaling the inlet velocity by (uin)g = yl og , < (nWinl n)mi = 1 aon = (5.134) - 100 (nrDf)p , Thus, it is required that f e aig . 2 (Win + I,)m ljk , -na =1 (5.138) E, Q,a n Dop Q,g ~ ~ . .R-The height scaling of the vents implies I lni " 1 .(5.136) Dop ""4 The axial extent of the horizontal .iet lj is scaled by the pool size. Thus, ln = Qan i o = (5.137) Using Eqs. (5.122), (5.130) and (5.131), the condensation scaling criterion reduces to ! -na 41+W, 20 . in . ( = 1 (5.138) n Dop . 6 ww w rmr W *pr a+- my 5-38 This can be approximated by nas E (5.139) 5 This implies that ng is between 1/2 and 1. Since the prototype number of the vertical flow channels in the prototype is eight, four to eight sets of the horizontal vents are required. With the prototypic n for the model, the relative condensation rate is about twice that of the prototype. For the intermediate gas flow rates in the vertical channel, it is possible to form a verti-cal gas jet at the exit of the horizontal vents. There is great uncenainty in estimating the diameter of this gas jet in a pool of water. However, a good scaling reference for this vert-ical gas jet is the width of the horizontal vents. Using a similar formulation as the one explained above, the relative condensation rate is scaled by he - (ai)g - ng ITR DJR (5.140) E, Qga Q,a . .R where I T si the depth of water and Dj is the jet diameter. The boundary flow scaling with (u,n)R = 1/2 implies ' (a,n)g = (nWin lin)g = 1/100 (5.141) l l Since (li n)g = log = 1/4, one obtains 1 1 (n Win)g = 1/25 (5.142) ) l Thus, by using the approximation D; = W n i (5.143) one obtains 5-39 e 1 = (5.144) { g/TR (nWm)x = 2 . .R This implies that the relative condensation rate for this case is about twice as high in the model as in the prototype. It is noted that the result is independent of the vent width or the number of the vents. The above results for horizontal and vertical jets show that the condensation rates an-somewhat higher in the model than in the prototype system. However, their orders of magnitude are similar. These results apply before the jet disintegrates into smaller bub-bles. The disintegration is expected when the condensation rate is insulliciently high or there is a significant amount of non-condensible gas in the gas mixture. When the vertical or horizontal jet disintegrates, the most probable fow regime in the pool is the maximum size cap bubble regime. In this case, a model similar to the one used for the sparger should be used to evaluate the relative condensation rate. In the later stages of an accident after the initial blowdown, the gas flow through the vertical channel is considerably reduced. In this case, the formation of muhiple bubbles at the exit of the horizontal vents is expected. As explained in the sparger scaling, there are two regimes in terms of the bubble size: the flow-rate dependent bubble size regime and the maximum size cap bubble regime. Considering very large diameter of the horizontal vents, the prototype system should be in the maximum size cap bubble regime. For the initial scaling estimate, the model vent size is about 2 cm x 17 cm, which is also a large oritice. Thus, the bubble size in the model is the maximum or slightly below it. Both the jet disintegration case and the bubbling case may be modeled as the max-imum size cap bubbly flow. The relative condensation rate scaling is given by Eq. (5.116). Thus. Dy l rR (5.145) E, (us)a(Dng , s R For the maximum size cap bubble regime, 1 5-40 I i I l (un )g = 1 and (D3 )g = 1 (5.146) Then P 4 Qc I (5.147) Es = TR= 7 , .R This shows that in the bubbly flow regime, the relative condensation rate is conside: bly smaller in the model than in the prototype. However, this is a conservative estimate because the bubble size in the model can be much smaller than the maximum cap bubble size given by Ds = 30 , (5.148) gbP In the flow-rate dependent bubble size regime, the bubble size is given by Eq. (5.123). This equation indicates that the bubble sim decreases with decreasing gas flow rate and increasing number of horizontal vents. In the preliminary design of the horizontal vents, the relative condensation rate is increased in the early stage and decreased in the later stages of accident simulation in the model. In the intermediate region where both the jets and bubbles exit, the scaling should have a compensating effect between these two flow regimes so that the total condensation can be well simulated. Furthemiore, if the conden-sation rate is high enough such that either the bubbles or the entire jet can condense before i they reach the top of the water pool, the rate itself has little impact on the overall simula-l tion. For SBWR or PUMA this is expected in the bubbly llow due to the large subcooling and considerable pool depth. 5.4.9 Mixing in Stratified Fluid Volumes l Mixing and stratification are multidimensional effects. Scaling these phenomena l requires considerations beyond the one-dimensional integral scaling criteria. The addi-tional criteria were obtained from Peterson and his co-workers [5.16,5.17). The model that has been developed [5.16] is a buoyant jet pointed upwards in a square cavity with a stratified fluid. The jet is assumed to be cylindrical, although a conical jet would produce similar results. The objective is to model the amount of mixing that occurs as the jet ascends into the stratified layer at the top. This mixing process then determines the level 1
- 5-41 of stratification.
In the model for inertia-dominated jets (i.e., the inertia of the jet is much greater than the buoyancy), four time constants are identified:
- 1) The transport time of the jet, njfdfH
= (5.149) t) Qo where nj is the number of jets, dj is the diameter of the jets, H is the height of the cavity and Qo is the volumetric flow at the inlet.
- 2) The entrainment time constant, which is the time it takes for the jets 'n entrain a volume of surrounding fluid equal to their own volume, njfdfH tj, = (5.150) nj u n d, H where u is the average velocity of the surrounding fluid into the jet. The entrainment volumetric flow into the jet is given by n, u nd, H = a 4E " Q,, (5.151)
J where at is Taylor's jet entrainment constant so that n nj - dj, t,= (5.152) 4EaQ,, 3)The transport time of the surrounding fluid, l 1 I l 5-42 V t,i = (5.153) Qo where V is the volume of the cavity. ,.
- 4) Entrainment time of the surrounding fluid, V
T, = (5.154) nj u xdj H - However, this last time constant can be obtained from the previous three: T xt T>e T,,, = (5.155) Tji So, in effect, there are only three independent time constants. Two independent non-dimensional groups can be formed out of these three by taking ratios. Peterson. et al. [5.16] chose the following two: = 4E a "- (5.156) xj = Tje dj and nj nd 2 t,, 7,H ; x, = = (5.157) T si V These non-dimensional groups could have been obtained from the Buckingham-n theorem as well. The second non-dimensional number may be expressed as 2 d' x , = nj 2 (5.158) i I . , . ~ . - - - . ._. -. , _ _ . . . . - _ _ . . _ _ _ - . . _ _ _ _ _ - _ _ - - - _ .. .- - . . - . -. .-. - = . . - . - 4 5-43 / The ratio x,opn ni x,poioiyp = 1 means that njR djR = D[ = an (5.159) 2 which is already satisfied by area ratio scaling. Dividing the above equation by H results in 2 '2 d3 D nja - . (5.160) H .R H s . . But, preservation of x) imposes d, -1 (5.161) . H .R , which means that I (5.162) nja = (H/D)[ To obtain similar mixing in the experiment, it is necessary that the number of jets be reduced by a factor that is inversely proportional to the ratio of the vessel aspect ratios squared. However, this imposes a very restrictive limitation on the number of jets; since the ratio of the aspect ratios is 2.5 this means that the number of jets would be reduced by a factor of 6 in the PUMA facility. . Peterson, et al. [5.16] then developed another model for buoyant plumes where buoy-ancy is dominant over inertia. The entrainment process is governed according to List's model, to 5 gp h Q,, nj u xd; H = 7 ku ZH (5.163) > 4 where Z is the vertical position and ko is an entrainment constant. The jet scaling criterion 8 E 5-44 becomes 2n - 5/3 x 5 nj = 4 ko.Ri]^ - d, (5.164) 3 Not only is it necessary to preserve the aspect ratio of the jets but also each jet's Richard-son number Rij . However, the Richardson number is aln:ady preserved by the integral scaling criteria, so this additional constraint is automatically satisfied. In the case of the discharge through the suppression pool and PCCS vents that occurs during long-term decay heat removal, the rising plumes are formed by condensed steam and noncondensible bubbles. The buoyancy in the plume is caused primarily by the non-condensible bubbles. The Richardson number for two-phase thw is given by gHa x2 df H R1, = 2 = - gn - (5.165) u 2 16 x29,2 dj where a is the void fraction in the plume and x is the molar concentration of noncondensi-bles in the nitrogen-steam mixture coming through the vent. Assuming this is the same for model and prototype, [Ri2 ,]g = an (5.166) It is difficult to predict the value of the void fraction. However, since this is the only difference, Ri2 , should be similar in both cases. The condition for similarity is given by Eq. (5.155) for single-phase buoyant jets and plumes, assuming Eqs. (5.151) and (5.156) are valid for two-phase bubbly jets and plumes. 5.4.10 Natural Circulation For natural circulation turbulent single-phase convective heat transfer in the walls of the containment, the correlation by Tokuhiro[5.18-5.19] may be used: I l - Ra l'3 (5.167) (1 + Nu s)23 where Ra = Gr Pr is the Rayleigh number and 5-45 l Tiop - T *"
- 3_ (5.168)
T.-T is the stratification number. For small values of s, 8 Nu = h H - Ra -H (5.169) k so that h is not a function of H. However, at a high level'of stratification, the heat transfer coefficient increases as the height is reduced. A more significant effect on the containment heat transfer is the increased surface-to-volume ratio in the reduced volume system. This ratio is scaled by S 1 - = (5.170) V ,g Da > which indicates that the reduction in the facility's cross-sectional area has a major impact on the available heat transfer area. As the cross-sectional area decreases, the surface-to-volume ratio increases. Hence, the taller the facility, the more distortion in the heat transfer area. This is one of the important factors to be considered in determining the height and area ratios,la and ag. In general, a tall and thin system has overwhelming heat transfer to the structure, which has been the major shortcoming of most of the integral test facilities built in the U.S. and other countries. For the proposed facility, the height is reduced to 1/4 to accommodate a relatively large cross-sectional area. As pointed out, the proposed facility has the smallest aspect ratio among existing and under-construction of test facilities for the SBWR. This is one of the significant advantages of the proposed design. For the effect of mixing at the wall boundary layers produced by natural circulation, Peterson, et al., [5.16] recommended the following scaling criterion: PliaH['" n n ig = (5.171) tg QVg where Pli compares the actual perimeter to the perimeter of a cylindrical vessel with ' equal 5-46 volume and height. This scaling criterion does not account for conden3 1 or other more complicated phenomena. 5.4.11 Heat Source and Sink It is important to account various, heat so rees and sinks in the integral facility follow-ing a transient, so that energy balance is satisfied. The thermal transients after the initial depressurization are relatively slow processes. Hence, the thermal energy stored is more imponant than the instantaneous heat transfer to or from the structure. For the structure, the thermal penetration depth is given by di s = a]n%t (5.172) From this, the characteristic time constant for conduction heat transfer in the wall can be calculated. For the reactor pressure vessel wall, the characteristic time is about 5(X) s. For the SBWR blowdown process, the time required is about 5(XL7(X)s. It is necessary to esti-mate the energy stored in the vessel wall. This can be done by solving the one-dimensional transient conduction heat transfer equation to get the temperature pmfile. Integrating the temperature profile over the period of time, the stored energy can be calcu-lated. During and following the blowdown process, the stored energy is released into the vessel liquid. Since the vessel in the present model is designed for 1.03 MPa (150 psig) rather than 7.16 MPa (1040 psi) as in the prototype, the wall thickness is thinner. Hence, an additional heat source which is 1/2(X), should be provided for the model vessel, to simu-late the heat energy released from the prototype vessel wall. 1 In the containment, the thick concrete wall has a large conduction time constant. ! However, in this case the containment wall is at a lover temperature, and the discharged I steam acts as a energy source to the wall. Since the c':aracteristic time constant of the transient for concrete is much longer than the blowdown period, a lumped parameter , model is appropriate for the containment wall, which acts as heat .%k following transient. Hence, ; 1 l dT a, p, cp 7 = nm q", - go, q",,ui (5.173) where a, is the cross-sectional area of the structure. The heat flux, q" and heated perime-ter, a couple this equation to the fluid energy equation. The important scaling parameter for the structural heat transfer is the thermal inertia ratio given by 4 4 5-47 Nth = (5.174) apcp The similarity between the model and protolype is satisfied by (Ni n)g = 1. For very thick materials, the cross-sectional area is given by a, = Thn du, (5,175) where d ih is the maximum penetration depth calculated from the end of the transient. Thus by substitution Eq. (5.175)into Eq. (5.174) the scaling requirement reduces to (k, p,Cp )j[2 = R = l (5.176) i T}g [ SE For concrete, the thermal diffusivity is small. 7his indicates that for most transients, only the region near the inner wall participates in the energy transfer. For example, about 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> are required, for 15 cm (6 inch) thermal penetration. The thennal inertia of this thickness can be simulated by 1.5 cm (0.6 in.) thick equivalent insulation material. 5.4.12 PCCS Venting into Suppression Pool Here, we consider the mechanism of PCCS venting into the suppression pool. Most of the steam flow rate into the drywell from the RPV is proportional to the decay heat of the core. The steam is condensed by the PCCS condensers at later stages of the blowdown following a LOCA. Since the drywell steam is mixed with air, this steam-air mixture enters the PCCS condensers and condenses, leaving air to accumulate in the lower header of the PCCS condenser. After sufficient accumulation of air in the PCCS condenser, the condensation becomes ineffective. This leads to an increase in the drywell pressure. There is a water head pressure diffe'rence between drywell and the suppression pool. When the drywell pressure exceeds this head, the gas from the PCCS is vented into the suppression pool. This venting process decreases the non condensible concentration in the PCCS condenser, and the PCCS condenser begins condensing steam again, which will decrease or stabilize the drywell pressure. The schematic of the PCCS venting process is shown in Figure 5.5. At time t=to, the PCCS has just vented gas into the suppression pcol. After t=to, condensation in the PCCS is very efficient and the drywell pressure decreases and will reach a point where steam flow rate from the RPV is equal to the condensation rate. As the condensation proceeds, i 5-48 the non-condensible gas accumulates in the PCCS enclosure. This will decrease PCCS efficiency. At t=t , ithe condenser is filled with non-condensible gas. Then, the pressure in the drywell rises due to steam flow from the RPV. At t=t 2, the pressun: in the drywell is such that Pow - P,p 2 Pna, and the PCCS vents into the suppression pool. After this, the cycle continues. For each cycle, the minimuili pressure difference APmin decreases as the SP pressure increases due to the non-condensible gas vented into it. Eventually, suppres-sion pool pressure exceeds the drywell pressure after a number of vent cycles. When the suppression pool pressure is higher than that of the drywell by 3.45 kPa (0.5 psi), gas passes through the vacuum breakers. When the vacuum breakers are opened, the drywell and suppression pool pressure equalize. The time period required for each venting is At i + At 2. The time period Att is almost constant and independent of Pow. However, At2 is dependent on APmin, If the APmin = 0, that is Paw = P,p, then At2 becomes longest. For the scaling of the venting phenomena, the time periods Ati and At2 need to be evaluated. First, we derive the time period At i. At the end of this period the PCCS condenser is filled with non-condensibles and condensation is totally stopped. If yV is the volume of the condenser tubes, then V y is the maximum volume of air required to fill the condenser tubes so that condensation is totally stopped. In the PCCS condenser, the flow near the tube inlet is like forced convection and near the tube end the flow is stagnant. The air accumulates at the bottom of the condenser tube. For simplicity and conservatism, it is assumed that the air begins to accumulate from the bottom of the condenser tube, and the condensation stops when the air has occupied the whole tube length. For each fraction of the air filled, the condensation area is decreased by that amount. For the first filling of the condenser with steam containing c,, faction of air, the volume , occupied by the air is Vg co. This leaves Vy (1-c o) of volume for the next tilling. After l the second filling the remaining volume in the PCCS condenser is Vy( 1-c,,)2 Sim larly, after the third filling, the volume left is Vg ( 1 -c,,)3 Thus, after n fillings, the volume left ! is Vg(1-co )" I Thus, the accumulated air after n fillings is given by Vg[1-(1-co )"] (5.177) Lettci be the time for single filling of the condenser, then for n fillings the required time is approximately given by t en =tci(1-c o)"-3 (5.178) 5-49 Thus the total time required to fill the condenser by non-condensible is At = I ten = tci I (1-co )"I = 41 c (1,179) n=1 n=1- o The characteristic time for a sihgle filling of PCCS condenser volume by steam / air mixture is given by tci =t= (5.180) m Here, p is the density of the vapor mixture entering the PCCS condenser. The rate of steam produced in the drywell, di, is given by rii = 9 (5.181) Abig Where q is the decay heat rate of the RPV core. From Eqs. (5.180) and (5.181) the time constant for the prototype is tp = 0.3168 s, and for the model tm = 0.158 s. Thus, the time constant for the model is half that of the prototype. This is consistent with the scaling of time for the model chosen. In Figure 5.6, the time period At required to fill the PCCS condenser with non-condensible as the function of non-condensible concentration is shown. For 10% of air concentration in steam, the time period At, for the prototype ispt x 10 = 3.16 s. For 1% of air concentration, this time period is 31.6 s. For the PUMA, Atim = 1.58 s for c , = 0.1 and at co= 0.01 (1%), Atim = 15.8 s. ; After the condensation has totally stopped, the pressure in the drywell increases due to the addidon of steam. Now we evaluate At 2, the time period at which the drywell pressure exceeds the suppression pool pressure by an amount greater than the water head, Pnua, that exists between SP and drywell. If SP and DW are at the same pressure, then from the ideal gas law we have T 4 pHud = R2 2 y9w A hrg 5-50 Thus, ; pilew Vow At2= T R (q/Ahr ) (5.183) 2 s where T is the absolute temperature of steam, R2 is the gas constant for steam and Vow is the drywell volume. Now we take the prototype-to-model time period ratio: PitegVow (At2 )R = (5.184) TR 2(q/hrg) ,g When the fluids used and the pressure condition are same for the prototype and model, then (TR2 )g = 1. For 1/4 height,1/4(X) volume and 1/2(X) mean flow rate scaled model, one obtains 1 2 (5.185) (At )R = 8 This indicates that the time period for the model is shorter than for the prototype. Typical time period for the prototype at 206 kPa (30 psi) drywell pussure can be calculated from Eq. (5.183). At a decay power level of 40 MW,it can be shown that At2p = 12.63 s. The corresponding time period for the model is At2m = 1.6s. The above time period was estimated assuming both the DW and SP were initially at the same pressure level. However, as shown in Figure 5.5, the pressure difference between the DW and SP could be anywhere from Apu,a to 0. The At2p given in Eq. ; (5.176) gives the maximum time required for the vent to open when the PCCS condenser l stops condensation due to accumulated non-condensibles. It can be assumed that the time period required to vent the non-condensible gas from the PCCS is negligible. Hence, for each venting cycle, the time period is At i + At 2 . The time period (At i) is higher (- 30 s) for PCCS performance when the air concentration in ; l the drywell is reduced. For most cases, At2 is smaller than 12.63 s for the reason men-tioned above. Hence, the vent cycle period is largely determined by the (Ati ) time period, which is well-scaled in the model. However, during the initial phase of the decay heat cooling, when the air concentration in the drywell is higher, the time period (Att + At2 ) for i I 5-51 the model will be shorter than in the prototype, so frequent venting is expected in the model. However, as the drywell air concentration is reduced, the time period required for PCCS venting in the model will be close to the correct by scaled time period of the proto-type. It is expected that for the long time transients, the model PCCS venting will perform as required applying the current scaling metliodology. 5.4.13 Condensation in PCCS Condensers The PCCS condensers provide decay heat removal by condensing steam from the drywell and supplying condensate water to the RPV through the GDCS tanks. The scaling of the heat transfer rate through the condenser is given by hpec, = N iows Nuniu U 3 iA (T, - Tp ) (5.186) where N i uw, is the number of PCCS condenser tubes. N as is the number of PCCS units. Un is the overall heat transfer coefficient. A; is the inner surface area of a condenser tube, and Tgand T are p the steam and PCCS pool temperatures. The overall heat transfer coefficient is given by ~ 1 in(Do/Di ) 1 Un = + + (5.187) hc A; 2nk,L hpAn In the RHS of Eq. (5.187), the first term corresponds to the tube side condensation heat transfer coefficient, the second term corresponds to the tube wall conduction heat transfer coefficient, and the third term corresponds to the outside tube pool heat transfer coefficient. l The condensation heat transfer coefficient is for the condensation of steam and air or nitrogen mixture in a vertical tube. Siddique, et al. [5.20] have studied the condensation heat transfer coefficient for steam-air mixture in the tube. The condensation heat transfer Nusselt number given by them is i Nuc - h' D' o - 6.213 Re .223 ' W" -W* a .Ja .2n (5.188) k, - W o. 3 where 5-52 p 3 Ja = e (T -T.) (Jacob number) (5.189) Ahrg V Di Re = Pm m (Reynolds number) (5.190) Pm a Non-condensable Wa= mass fraction (5.191) g, , g , The pool side heat transfer coefficient, h p, depends on whether pool side transfer is due to boiling or natural convection. For natural convection, the heat transfer Nusselt number is given by P '" Nu,= r = 0.021(GrPr)"4 (5.192) v - where 3 Gr = g (T - T p)Liu w (Grashaf number) (5.193) 2 V Pr = 1 (Prandtl number) (5.194) ct For the correct scaling of heat removal by the PCCS condensers we should have Qg,/sohrs" ', =1 (5.195) Qy,/sAhr, I where m is the inlet steam mass flow rate to the PCCS condenser. From Eys. (5.186) and (5.191), the scaling requirement for PCCS condenser heat removal rate is given as 5-53 Nt ow Nunii, Un A;(T3 -T )hhAhrg p =1 (5.196) If the prototype and model have the same operating pressure condition and use the same operating fluid (water), then the temperature difference can be preserved. From Eqs. (5.195) and (5.196) we have y,/mAhrg y,/rnAhrg ,p 'N m 'N m 'U m 'A m [(T3 - Tp ), rii n ' Ahryp N p,, N p, Up Ap ,, (5.197) (T3 - Tp)r , mm , Ahrgm , "U m - - ' First, we evaluate or U 3 U p,, , sR. From Siddique, et al. [5.20], the average Reynolds number for the prototype PCCS condenser tube is approximately given by R7p= (13(XX) + 1(XX))/2 = 7(XX). This is for a prototype condenser tube length of 18 m. For the model, with the present choice of 1/4 length scale and 1/4(X) volume scale, the condenser tube length is 0.45 m. If the model tube diameter is the same as that of the prototype, then the results of the Siddique, et al. [5.20] can be directly applied. For the model, the average Reynolds number is jiem= (13(XX) + 8(XX))/2 = 105(X). The corresponding condensation heat transfer coefficients for the prototype and model are hem = 4(XX) W/m 2 K and hyp = 2500 W/m 2 "K. The pool 2 side heat transfer coefficient is available from SSAR [5.21] as ph = 45(X) W/m K. For the prototype tube material, stainless steel with OD = 50.8 mm and ID = 47.5 mm. the ratio (Un)g is calculated from Eq. (5.187) as (U3 )g = 1.442 (5.198) In the present scaling, we have heat transfer area scaling of 1/2(X) and boundary mass flow rate scaling of 1/2(X). Hence, 5-54 'Nm 'Nm 'Am = I (5.199) NP . tube N P iJ uniu A . 2(X) , . . P.i ~ and rh P 1 = (5.2(X)) mm 2(X) Since temperature differences are preserved in the model we have from Eq. (5.197) through (5.2(X)), Qy,hhhhty ', ' , " = 1.442 (5.201) Qg,hhAhrg ' This indicates that the pmsent model condenser removes about 449< more heat than the correct scaled model. Apparently, this is a distortion of the heat removal rate of the PCCS condensers. It is proposed that cenain portions of the PCCS condenser tube's outer sur-face be covered with insulating material like Teflon
- sheet, so that smaller areas of the PCCS condenser tube is exposed to the pool side. This will effectively reduce the heat transfer rate of the condenser. The insulator will be snuggly wrapped around the con-denser tube to cover 309, of the tube's outer surface. The reduction in the condensibles l
total surface will result in satisfying the correct scaling criteria for the PCCS condensers 1 given by Fq. (5.195). I i j 5.4.14 Stratification in the Drywell The stratification in the drywell will affect the operation of the PCCS. If the steam stratifies above the nitrogen, the amount of non-condensible gas entering the PCCS will be reduced. During the blowdown stage, the drywellis at a higher pressure than the suppres-sion pool, and the steam coming out of the reactor vessel will mix with the steam and nitrogen in the drywell and vent into the suppression pool chamber. A simple calculation shows that at the end of blowdown, the amount of nitrogen left in the drywell is almost negligibly small. l l I 5 55 The steam jets coming out of the drywell during blowdown carry a lot of momentum. Therefore, mixing is expected to be good. The Richardson number for these jets is approximately 1 x 10". The amount of steam produced during the vessel depressuriza-tion is given by , M,i,2m uf initiai - ut finai + Q/Miniciai (5.202) Mmitial h, - ut tinal where hg is the me.m enthalpy of the steam coming out of the vessel, Mmitia is the initial mass of coolant in the vessel, ut is the internal energy of the coolant and Q is the amount of decay heat during depressurization. The steam mass produced is approximately 1(X) 3 tons, which corresponds to a volume of 60,(XX) m at 3 atm. This is twelve times the volume of the drywell. Assuming perfect mixing, the concentration of steam at the end of blowdown is Xa am = 1 - e-12 = g,999994 (5.196) so the amount of non-condensibles left in the drywell accounts only for 0.(W15% of the mixture mass. After blowdown, the vacuum breakers discharge the nitrogen from the suppression pool back into the drywell. In this case, the Richardson number is 0.(X12~, so again, these jets are inertia dominated and will reach the top of the drywell and deflect along the containment walls, resulting in good mixing. The steam coming out of the reactor vessel during decay heat removal caries relatively little momentum. The Richardson number for these plumes is 0.2. The possibility of stratification in this case is greater. Because the Richardson number is part of the integral scaling criteria, these phenomena should be similar in the SBWR and in PUMA. However. because the number ) of jets in PUMA is greater than Eq. (5.162) specifies, the amount of mixing will be i greater. This implies that in case of the steam plumes during decay heat removal, the amount of stratification will be reduced. 5.4.15 Stratification in the Suppression Pool l The stratification of the pressure suppression pool affects the pressure in the contain-ment system. The pressure in the pressure suppression pool chamber is the sum of the pressure of the non-condensible gases and the saturation pressure of the steam at the tem-perature of the pressure suppression pool surface. I 5-56 During vessel blowdown, the jets coming out of the drywell vents and the SRV ! spargers are well mixed. The question that remains is the effect of the PCCS vents discharge during non-condensible venting on the pool stratification. The PCCS vents are very shallow,2(X).mm in the PUMA design and 8(X) mm in the SBWR. While a full fluid dynamic evaluation of stratification may be very complicated in this case, a simple argument shows that PCCS venting effect is negligible. Assuming that the period of non-condensible discharge from the PCCS is 15 seconds and that each discharge corresponds to the whole volume of the PCCS, it may be shown that this is equivalent to a heat source of 11(X) Watts in PUMA. With this heat input, a layer of 2(X) mm thick on top of the suppression pool would heat up at a rate of approximately 1 C per hour. The reason for this layer will heat up uniformily is that, the bubbles will entrain water from the vicinity of the discharge so there is an induced liquid flow that is of the same order as the bubble volumetric flow. Therefore, the liquid layer above the discharge is constantly getting mixed. The worst case to assume is that there is no mixing with the rest of the pool and to neglect evaporation at the surface, conduction to the walls and con-duction to the rest of the pool. Then the liquid layer above the discharge heats up adiabat-ically at the rate previously mentioned. _Ab inlet pressure a= 2 MM9 .? M 'h ll, ,'lf,'f. "J.-~. ^ ' w% M - _ ' ' '_- '---~--N mm= $f ~BwlS _ I flIIII;f ((I ggj r 1 M5~:w Q w (Illill(IlllTT 111U1111ifI III?lilli! ( 'I1111111111111ll IIIIIIII II) Iy] i ~lIIIllfl111 , IIIIIIIIIII l { "lllllUllH IIIIIIf f ITT1 '//////////# outlet pressure th roat '!lllllll,h' 1T+'lllllR ?h'h'h'h'? / i normal shock M' m . Figure 5.1 The Static to Upstream Stagnation Pressure Ratio as a Function of Position and the Exit Static Pressure Ratio for a Converging-Diverging Nozzle ? M 0.6 i i i i i i i e i i nondimensional mass flow rate 1 0.3 - - I I I I ' ' ' I I l O 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 ratio of exit to stagnation pressure . Figure 5.2 The Nondlemensional Massflow Through the Converging-Diverging Nozzle as a Function of Nozzle Pressure Ratio u
.-----n-_.---- _a -_- - - _ - - - _ - - - - - -- n - -m - -
1 inlet pressure ;
outlet pressure -
th roat w gg y ,
k ~
1 ,
i OllE3 l* - -
1 a
0$ S0 $ ! -
jhP ML"*s$=0 !$5 Oli emsEE ;gg C00.0 Figure 5.3 The Static to Upstream Stagnation Pressure Ratio as a Function of Position and the Exit Static Pressure Ratio for an Abrupt Expansion Nozzle 4
t 0.6 , , , ,
nondimensioal mass flow rate 0.3 - -
I ' ' I 0
O 0.2 0.4 0.6 0.8 1 ratio of exit to stagnation pressure Figure 5.4 The Nondimensional Massflow Trough the Abrupt Expansion Nozzle as a Function of Nozzle Pressure Ratio y.
8
a _
ieod P ~-
Prnin '
P [ +
Nj - -
d2 ~
1
2 -*
1 2.
t --
Figure 5.5 Drywell and suppression pool pressure transient schematic E
l 5-62 1 03 .......i .
. . . ....i . . . . . . . . .
.g 1 02 _
\ ,
i \ SBWR (prototype) !
N .
<i
,6
\ -
8 PUMA N a
e 10 -
\ '
e : \ :
.5 H
N .
\ .
N 1 00 -
N ,
\
\\! N .
s 10-1 10-3 10-2 10-1 1 00 Concentration, c o ,
I 1
1 Figure 5.6 Time period At to fill PCCS volume as a function of air concentration
5-63 l
)
References 5.1 Ishii, M., Kataoka, I.. " Similarity Analysis and Scaling Criteria for LWRs under Single Phase and Two-Phase Natural Circulation", NUREG/CR-3267. ANL-83-32 (1983). .-
5.2 Kocamustafaogullari, G., Ishii, M., " Scaling Criteria for Two Phase Flow Natural and Forced Convection Loop and their Application to Conceptual 2x4 Simulation Loop Design", ANL-83-61, NUREG/CR-3420 (1983).
5.3 Kocamustafaogullari, G., Ishii, M., " Reduced Pressure and Fluid to Fluid Scaling Laws for Two-Faase Flow Loop", NUREG/CR-4584, ANL-86-19 (1986).
5.4 Condie, K.G., Larson, T.K., Davis, C.B., McCreery, G.E., " Evaluation of Integral Continuing Experimental Capability (CEC) Concepts for Light Water Reactor Research-PWR Scaling Concepts", NUREG/CR-4824, EG&G 2494 (1987).
5.5 Boucher, T.J., DiMarzo, M., Sho:
. L.M., " Scaling Issues for a Thermal-Hydraulic Integral Test Facility". NPC Paper.
5.6 Ishii, M., "One Dimensional Drift-Flux Model and Constitutive Equations for Rela-tive Motion Between Phases in Various Flow Regimes", Argonne National Labora-tory Report ANL 77-47 (1977).
j 5.7
' Ishii, M. and Zuber, N., " Thermally Induced Flow Instabilities in Two-Phase Mix-I tures," 4th International Heat Transfer Conference, Paris, Paper No. B5.11 (1970).
5.8 Ishii, M., " Study on Flow Instabilities in Two-Phase Mixtures," Argonne National Laboratory Report, ANL-76 23 (1976).
! 5.9 Ishii, M., " Thermally Induced Flow Instabilities in Two-Phase Mixtures in Thermal Equilibrium," Ph.D. Thesis, Georgia Institute of Technology (1971).
5.10 Kataoka, I. and Ishii, M., " Drift Flux Model for Large Diameter Pipe and New Correlation for Pool Void Fraction",Infl. J. Hear Maxx Tranx/er, vol. 30,1927-1939 (1987).
I 5.11 Leung, J.C.M., " Occurrence of Critical Heat Flux during Blowdown with Flow Reversal". Argonne National Laboratory Report, ANL-77-4 (1977).
5.12 Katto, Y., "A Generalized Conelation of Critical Heat Flux for Forced Convection Boiling in Venical Uniformly Herted Round Tube", Intl. J. Hent Mass Transfer, vol. 21, p.1527 (1978).
I 5.13 Mishima, K. and Ishii, M., " Critical Heat Flux Experiments under Low Flow Condi-tions in a Vertical Annulus", NUREG/CR-2647, ANL-82-6 (N2).
- . - - . ~ -- . .- - -- - .- - .. - . . . --.. - --
l 5-64 l 5.14 Griffith, P., Pearson, J.F., and I.epkowski, R.J., " Critical Heat Flux During a Loss-of-Coolant Accident", Nuclear Safery,18(3), p. 298 (1977).
5.15 Liang, K.S. and Griffith, r.. " Experimental and Analytical Study of Direct Contact Condensation of Steam in Water," Nuc/. Engr. andDesign,147 pp. 475-435 (1994).
5.16 Peterson, P.F., Schrock, V.E. and Greif, R., " Scaling for Integral Simulation of Mix-ing in Large Stratified Volumes", NURETH-6, October 5-8, 1993, Grenoble, France.
5.17 Peterson, P.F., " Scaling and Analysis of Mixing in Lrrge Stratified Volumes," Intl.
J. Hear Mass Transfer, 37,97-106 (1994).
5.18 Tokuhiro, A.T., Lykoudis, P.S., " Natural Convection Heat Transfer from a Vertical Plate - 1. Enchancement with Gas Injection", Intl.1. Hear Maxx Transfer,37(6),
997-1003 (1994).
5.19 Tokuhiro, A.T., Lykoudis, P.S., " Natural Convection Heat Transfer from a Vertical Plate - 2. With Gas injection and Transverse Magnetic Field", Intl.1. Hear Maxx Transfer,37(6) 1(X)5-1012 (1994). '
5.20 Siddique, M .Golay, M.W., Kazimi, M.S., " Local Heat Transfer Coefficient for Forced Convection Condensation of Steam in a Vertical Tube in the Presence of Air," Nuclear Technalagy, Vol.106, No. 2, pp. 202-215 (1994).
5.21 G.E. Nuclear Energy, "SBWR Standard Safety Analysis Repon," 25A5113. Rev. A ' .,
August (1992).
l
i l
1 6-1 l l
1 I
1 l
- 6. SCALE OF THE PUMA I l
i At the prototypic pressure simulation, the following relations are obtained from the integral system scaling and the boundary flow scaling results given in Chapter 5:
pg = 1 (6.1) ai Ag=
i ,
-1 (6.2)
, ,R li Lg=
i =1 (6.3) 7" . R uR I (6.4) l 1
G gg = 1//g (6 4)
At the major heat transfer sections, such as at the core and heat exchanges, it may be possible to make dx = 1 (6.6) i which implies that the heater or heat exchanger sections should have a prototypic hydraulic diameter.
According to Eqs. (6.2 and 6.3), the major geometric configurations of the scaled model are determinated by A;g = 1 and L in = 1. This leaves some freedom in choosing on the height ratio, i Ig, and area ratio, an. The overall cost of the facility is reflected in the volume ratio, j Vg = la ga and height ratio, Ig. In determining the overall size of the proposed facility, it is )
necessary to consider four essential factors: !
1
- 1) the need to consider scale relations to the existing facility,
- 2) the need to compensate for the shortcomings of existing facilities or complement the overall data base, -
3). the necd forjustifiable rationale for the chosen values of an andIg,and 1
6-2 I
1 l
- 4) the overall impact on the total cost. I For the PUMA, the above factors have been examined in detail. Based on these considerations, a 1/4 height and 1/400 volume scale have been chosen as the opdmum design.
The existing or under-construction integral facilities for the SBWR are all full height. The GE's GIST facility is a low pressure, full-height facility with la = 1 and aa = 1/508. The GIPAFFE facility in Japan has la = ! and ag = 1/400. The planned PANDA facility has lg = 1 and ag = 1/25. The aspect ratio Ig/da, for these facilities are 22.5,20 and 5, respectively. In view of the overall cost, the volume scale of these facilities, friction and structural heat similari-ties, a new facility at the volume scale of about 1/400 appears to be optimum. This will match the mass and inventory of the GIST and GIRAFFE facilities also.
Since the existing facilities are all full height, the impact of the actual total height on various phenomena can be evaluated sufficiently. However, the existing facilities fall into the category of thin and tall systems, which have some major shoncomings. In Table 6.1, the dimensions of various components of the SBWR are compared between prototype and full-height,1/4-height and 1/8-height scaled model for a 1/400 volume scaled facility. As seen in Table 6.1 for a full-height scaled model, the flow area is reduced substantially. The reduction in flow area increases the frictional resistance significantly. This effect can be quantified by the friction number, or more specifically by a factor fl/d, where f is the friction factor. For most nuclear reactor sys-tems, the major flow resistances are located at the bundle sections such as the core or steam gen-erators. These sections can be made quite prototypic by taking the similar bundle cross-sectional geometry, so that the reduction of the flow area, up to a certain level, does not r.trongly affect the overall frictional resistance. I!owever, a flow area ratio ofless than 1/200 can have a significant impact on the total frictional resistance. Since the friction number has a factor fl/d, a reductian in I can significantly help to scale flow resistances. At the same time, da can be increased by decreasing la for a fixed volume scale. Hence, the reduction in height can effectively eliminate the significant problem in the simulation of the friction number.
The complete scaling analysis presented in Chapter 5 indicates that the reduction in the height does not necessarily distort the natural circulation phenomena because the circulation rate is determined by the balance between the driving head and the loop frictional resistance. In order to preserve the kinematic and energy similarity, the system velocity and time scales should be reduced by the specified factors in a reduced hight system. Under these conditions the driv-ing head and loop frictional resistance can be matched. In contrast, a thin and tall facility may significantly distort the natural circulation by increasing one of these two balancing forces.
The second advantage of using the reduced-heigh ~t system is that the aspect ratio becomes much closer to the prototype system. Thus, it is a much better system to simulate the two- or three-dimensional phenomena expected in the SBWR core, chimney, downcomer and
6-3 l
l containment. As shown in Table 6.1, the 1/8-height scaled model is close to a linearly scaled I model. It has very fat vessels, especially the upper drywell and suppression pool. For 1/8 height scale, the requimd core power is also large. The 1/4 height and an = 1/100 scaled facility has moderate power requirement and makes the aspect scaling ratio factor to be only 1/2.5, which is very close to the prototype system.
The present quarter-height system with a volume scale of 1/400 has the advantage of well-matched gravitational and frictional forces. Funhermore, due to relatively large cross-sectional areas, the important phenomena of two- or three-dimensional voiding patterns and flow regimes in the core and chimney can be simulated well. This is particularly important for assessing the effects of various instabilities such as the manometer oscillation, density wave instability, geysering and flashing-induced cyclic phenomena in the natural circulation cooling and stability of the GDCS. Even in comparison with the PANDA facility, the present system has a smaller aspect ratio, and therefore a significant advantage for simulating certain phenomena.
The present facility, which simulates the 2-D and 3-D phenomena very well because of its smaller aspect ratio, is called the Purdue University Multi-dimensional Integral Test Assembly (PUMA). In Table 6.2, the important non-dimensional numbers, derived from scaling considera-tions, are compared between the prototype and PUMA. These numbers are calculated for 1.03 MPa (150 psi) operating pressure condition. From this table, the numbers for PUMA match the prototype dimensionless numbers, except N,ius. The Na, in PUMA is also smaller than in the prototype due to a large contribution oflocal slip to the total slip. The chimney Nu in PUMA is smaller than in the prototype. However, the frictional pressure drop in chimney is not important, as it is very small compared to the gravitational pressure drop.
One of the possible shortcomings of the reduced height system is the reduction in the flash-ing phenomena in the chimney. The superheating due to the reduction in the hydrostatic head as the two-phase mixture rises in the chimney section is essentially related to the height of the chimney. Hence, by reducing the height, the flashing is somewhat reduced.
l Since flashing can be one of the important local phenomena of interest, it is necessary to l focus some of the experiments on this flashing phenomena distortion. The vapor generation due to flashing can be simulated by reducing the system pressure by about 459 relative to the proto-type system. By reducing the pressure, the vapor density can be made significantly smaller.
l This will lead to similar vapor volume generation and void fraction as in the prototype. This type of experiment, focused essentially on the impact of the flashing phenomena, will comprise a small portion of the preliminary test matrix.
l l
l
l 6-4 TABLE 6.1 COMPARISONS FOR MAJOR COMPONENTS-AND DIMENSIONS OF DIFFERENT HEIGHT SCALING
- l COMPONENT PROTOTYPE fli 1M ja HEIGHT HEIGHT HEIGHT (PUMA) (LINEAR)_
REACTOR PRESSURE
- VESSEL
- Total height (mm) 24600 24600 6150 3075 1.D. (mm) 6000 300 600 848.7 Total volume (m3) 669 1.67 1.67 1.67 .
CORE Rod material Zr clad S.S alloy S.S alloy S.S alloy Active length (mm) 2743 2743 685 342.9 Total power 45 mW 112.5W 225 kW 318.2 kW Core shroud I.D (mm) 5150 257.5 515 728.4 CHIMNEYSECTION Total height (mm) 9000 9000 2250 1125 Partition height 6500 6500 1625 812.5
- of divided areas 25 9 , 9 9 1.D. of shroud (mm) 4955 247.75 495.5 700.8 CONTAINMENT Wall material Concrete / steel S.S S.S S.S Upper head volume (m3) 3770 9.4 9.4 9.4 '
Upper head height (mm) 6100 6100 1525 762.5 Upper head dia.(mm) 28050 1402.5 2800 3967.5 i
Lower head volume (m3) 1696.5 4.24 4.24 4.24 Lower head height (mm) 27200 27200 6800 3400 ,
Lower head dia. (mm) 8911 445.55 891.1 1260.4 SUPPRESSION POOL Initial water volume (m3) 3255 8.13 8.13 8.13 .
Initial gas space (m3) 3819 9.55 9.55 9.55 Height (mm) 11950 11950 2987 1493 Diameter (mm) 27450 1372.5. 2800 3882.6 GDCS POOL (1 OF 31 Diameter (mm) Not circular 450 900 1273 Height (mm) 6100 6100 1525 762.5 Volume (m3) 348 0.87 0.87 0.87
- Note: The volume scaling ratio is kept as 1/400 for all different height scaling.
l l
l
( .
65 TABLE 6.2 COMPARISON OF VARIOUS PARAMETERS &
NON DIMENSIONAL NUMBERS BETWEEN PROTOTYPE & PUMA l (at 1.03 MPa or 150 psi operating pressure)
I QUANTITY / NUMBERS PROTOTYPE PUMA CORE EXIT QUAUTY d.3x104 4.166x 104 CORE EXIT VOID FRACTION 0.4 0.41 Np(core) .72 .694 Nn h(chimney) 0.298 0.074 Nd 0.85 0.46 Nr
-core support plate and core 4.4 4.2
-chimney 3.4x 104 1x104
-separator 1.19 0.91 Losses for major lines in both PROTOTYPE and PUMA Component PROTOTYPE PUMA K ffD K ffD MSL 6.6 0.4 34.8 2.1 SRV 5.8 1.4 56.6 13.7 DPV (MSL) 101.8 0.14 526.6 0.5 DPV (RPV) 14.7 0.1 113.4 0.8 IC supply line 6.5 1.0 131.6 20 IC condensate line 3.4 1.4 26.2 10.8 IC vent line 10.2 32.9 20148.2 64987.8 PCCS supply line 4.6 0.5 23.3 2.5 PCCS vent line 3.3 1.3 16.7 6.1 PCCS condensate 5.5 1.2 42.4 9.3 GDCS equalization line 40.5 1.1 19.5 0.5 GDCS line 12.4 1.2 121.1 11.7 FWL 6.2 1.1 8.2 1.5
- SRV losses were estimated since Purdue did not receive its isometric drawing from GE.
- Losses were estimated using the Crane Paper.
1 i
1 I
i 7-1 7.0 DESIGN OF PUMA-SBWR FACILITY AND SCALING BASIS 7.1 Reactor Pressure Vessel Design The Reactor Pressure Vessel (RPV) is the most vital component in the scaling, design and experimentation of the Simplified Boiling Water Reactor (SBWR) system. This observation is based on the fact that nearly all SBWR safety systems are designed to prevent core uncovery, and that the RPV is the main source of steam / energy inventory in the integral facility. Of the components that make up the RPV, the most significant are:
. Pressure vessel
. Lower plenum
. Core plate
. Core
. Chimney
. Steam separators
. Steam dryers
. Downcomer.
The issues of scaling and design of these RPV components are discussed in later subsections of this chapter.
Proper scaling of the RPV requires careful scaling of the vessel geometry. The core power is properly scaled next, followed by the pressure drop across each component. Even though the geometrical scaling is straightforward, it may be restricted in more than one component by other considerations, such as core power and core heat flux.
The core power scaling is derived from the heat source number, Eq. (5.5), where the volumetric heat generation is given by
. uo i q =-= (7.1) lo
{
From the integral and boundary flow scaling of the reactor system, the area ratio, an, and length ratio,Ig, are given by 1 1 a and (7.2) a = 100 Jt = 4
1 I
7-2 which results in the following value for the core power scale:
1 QR = (7.3) 7.1.1 Pressum Vessel The pressure vessel scaling is purely geometrical. The choices of vessel material and struc-tural design are dictated by factors such as vessel integrity and operating conditions, which are independent of scaling ise"es. The PUMA pressure vessel dimensions are given in Table 7.1.
In Figure 7.1, a schematic of the pressure vessel is shown with the major internal com-ponents and their corresponding heights. As shown in Figure 7.2, the vessel has three parts separated by flange connections, which allow access to the internals during assembly and maintenance.
Table 7.1 Reactor Pressure Vessel Dime nsions.
Component SBWR PUMA Scale Total Height (mm) 24505 6126 1/4 ID (mm) 6040 604 1/10 2
Area (m ) 28.65 0.2865 1/100 3
Volume (m ) 669 1.755 1/4(X)
Wall Material Carbon Steel Stainless Steel l 7.1.2 Lower Plenum i
The SBWR lower plenum houses the Control Rod Drive (CRD) mechanisms and other inter-nal structures, whereas the PUMA lower plenum houses the non-heated length of the fuel-simulated electrical heater rods. The SBWR intemal structures in the lower plenum are not con-sidered in this scaling analysis. This leads to distortion in the flow area and volume scaling in the lower plenum approximately 26% greater than the volume scaling criteria of 1/400. Such a distortion will affect pressure drop and vessel inventory. In order to minimize this distortion, a volume filler piece is added to the lower plenum, as shown in Figure 7.3, to reduce the volume distortion to 16.4%. The tentative design of the filler pieces calls for a sealed block made of thin ,
stainless steel sheets, which will have a negligble effect on stored heat. l
7-3 7.1.3 Core Plate In order to obtain a properly-scaled nominal flow rate through the core, the core plate in PUMA is designed considering the total pressure drop across the core. The distortion in the over-sized PUMA core flow area leads to a distortion.in the com flow rate and core pressure drop.
Hence, the core plate in PUMA is designed not only to match the scaled pressure drop across the core plate, but also to account for the area distortion in the core. As shown in Figure 7.4, the PUMA core plate is designed to house two orfice groups:
- i. Bypass orifice group 1 x 10 mm I.D. at core plate center 12 x 10 mm I.D. equally spaced at a radius of 179 mm from core plate center ii. Core heaters orifice group 38 x 34 mm I.D. giving an annular geometry for heater orifices 7.I.4 Core
- a. Core Heater Rods Adopting the scaling approach of Kocamustafaogullari and Ishii [7.1], the appropriate scal-ing of the core fuel rods results in use of 439 heater rods,12.27 mm O.D., capable of providing 455.6 W of power per rod. The cross-sectional area of the 439 heater rods, in addition to the flow ama around each rod, consumes the total core area. This is clearly unacceptable as the com area must also include the bypass area and several instrumentation penetrations into the core. An alternative configuration, therefore, would be a reduced number of enlarged heaters.
Other important considerations in the heater design are the need for a variable power profile and the need to over-design the heater power. These provisions are necessary to compensate for the unexpected power failure of individual heaters and to accommodate over-power tests where 10% over power is used.
SBWR realistic decay heat data [7.2] were fitted with a tenth order polynomial. The decay power is given by 10 y = 1 b nx" (7.4) n=0
7-4 Hei:, y = log p/po, p is decay power at time t, pc = 2000 MWt, x = logt, and the coefficients bn are given in Table 7.2.
Table 7.2 Coefficients for Debay Heat Polynomial Fit bo 0.45865 b -1.20325 b2 -0.5128 b3 1.140821 04 0.35704 b5 -0.91525 b6 0.126003 by 0.289916 b8 -0.16738 b9 0.035941 bo i 0.0028 In Figure 7.5, the polynomial 6t is shown with decay heat data. In Table 7.3, SBWR time and decay power are compared with PUMA time and power scaled values 6(X) seconds following scram.
Table 7.3 Decay Power in SBWR and PUMA Facilities l SBWR Time SBWR Power PUMA Time PUMA Power l (kW) (kW) 4 600 s 4.50x10 3(x) s 225 1 4
1h 2.65x10 0.5h 4
2h 2.15x10 1h 108 3h 1.91x10 4 1.5 h 4
4h 1.75x10 2h 88 4
8h 1.43x10 4h 71 ,
I The experiments on PUMA are designed to stan at a blowdown condition of 1034 kPa (150 psi).
Table 7.4 lists the power levels in both the SBWR and PUMA facilities at 1034 kPa (150 psi) for various postulated break transients with one DPV failure. Table 7.4 is derived from the informa-tion given in Figures 6.3-10 and 6.3-80 in [7.3]. PUMA core power levels are based on the core
7-5 power scaling criteria of 1/2(X).
Table 7.4 Core Power Levels at 1034 kPa (150 psi) Following Blowdown Break Location 1 GDCS Valve 1DPV SBWR PUMA Failure (s) Failure (s) Power (MW) Power (kW)
STB 290 51.8 259 Inside SLB 270 52.49 262.45 FWLB 230 53.87 269.35 RWCU/SDC Break 370 49.53 247.65 IC Retum LB 600 44.1 220.5 GDCS Injection 310 51.12 255.6 BDLB 500 46.06 230.3 STB - Stub Tube Break BDLB - Bottom Drain Line Break From Table 7.4, it can be seen that the highest scaled decay power level (269.35 kW) is for the feedwater line break (FWLB) at 1034 kPa (150 psi) following blowdown. Hence, a total of approximately 300 kW is needed for the core power of the PUMA facility to accomodate all pos-sible blowdown conditions.
A number of design iterations were performed to determine the PUMA core heater rod lay-out each taking into consideration PUMA total power requirement, accommodation of bypass flow area, provision for variable power profile, commercial availability, and practical design and assembly considerations. The resultant design includes thirty-eight heater rods, each one 25.4 mm (1 inch) in diameter, placed in three concentric rings as shown in Figure 7.6. The layout of the heaters allows sufficient space for instrument probe penetrations into the core through the l core plate. The three-ring design allows for a non-uniform power profile by separately control-ling the power to each ring.
Typical radial power profiles from the SSAR, Chapter 4.1 [7.3), were studied. Relative to l the inner ring of heaters, the PUMA middle ring power level per heater rod is designed to be greater by a factor of 1.2, whereas the outer ring power level per heater rod is lesser by a factor of 0.8. Table 7.5 lists the arrangement of the heater rods for a maximum core Pcwer of 400 kW.
I
7-6 Table 7.5 Core Heater Rod Arrangement Circle No. of Max. Operating Maximum radius rods rod power power heat flux (mm) (kW) (kW) (kW/m2 )
Inner ring 70 6 ' 11.0 5.5 201.0 Middle ring 145 12 13.5 6.6 246.7 Outer ring 240 20 9.0 4.4 164.5 A typical heater rod is shown in Figure 7.7. The heated length of the heater rod is based on the scaled length of the SBWR active fuel length. For assembly and access purposes, the heater rods penetrate from the bottom of the RPV and extend up to the inlet of the chimney section. Heater rods are manufactured from commercially available Inconel or stainless steel with commercially prepared surfaces. ,
- b. Bypass Constructing the PUMA core as three rings of variable-power heater rods suggests using the central area of the core as one bypass area and the annular space between the second and third heater rods rings as the second bypass area.
The core parameters for PUM A are compared with prototype in Table 7.6.
Table 7.6 Core Parameters Component SBWR PUMA Scale l Number of Rods 43920 38 Rod Diameter (mm) 12.27 25.4 Rod Material Zr Clad SS Alloy Clad ,
l Rod Active Length (mm) 2743 686 1/4 2
Flow Area-Fuel (m ) 7.4 0.1255 distorted Flow Area-Bypass (m2 ) 5.6 0.02491 distorted Mass Flow Rate (kg/s) 7555.6 37.8 1/200 l Pressure Drop (Pa) 48200 12050 1/4 l
7-7 i
Based on the pressure drop across the core and inlet plate and the flow area for bypass and fuel channels, the following design was chosen for PUMA:
. Central bypass pipe with I.D. of 54.0 mm
. Annular area with a smaller diameter of 340 m'rn and a larger diameter of 380 mm.
Figure 7.4 shows the bypass regions and the inlet holes at the core plate.
From the design we see that the available area for the flow in the core is 0.15041 m2 , which j
is larger than the flow area of the appropriately scaled PUMA core by approximately 15.7%.
i This enlargement will result in a distortion in the scaled velocity in the PUMA core of approxi-mately 15.7% as well. Therefore, the velocity ratio in the core for the enlarged area is (V.)g = 1/3.6.
The pressure drop is similarly distorted. To compensate for the decrease in pressure drop across the core due to the enlargement of the flow area, the core plate orifice area is substantially decreased, as discussed above.
7.1.5 Shroud design The core shroud acts as a thermal shield, preventing heat from directly reaching the vessel wall. It also separates the core from the downcomer section of the RPV. The core shroud can be modeled as a cylindrical wall placed around the core.
Dimensions of the core shroud are summarized in Table 7.7. The core shroud extends from the top of the fuel to a certain distance below the core plate. The height and diameter of the core shroud are scaled using the geometrical scaling criteria oflength and area.
The PUMA core shroud consists of a single cylinder surrounding the core assembly extend-ing from above the heaters to the bottom of the RPV, as shown in Figure 7.3. The lower portion of the core shroud is perforated with large holes to simulate the open flow area in the lower ple-num. These perforations begin at the corTect scaled height location in the core shroud (i.e.,
536.5 mm from the bottom of the RPV)in order to preserve the flow pattem in the downcomer and the region below the core plate. The perforations extend to the bottom of the RPV.
The shroud is welded to the bottom flange of the RPV. This single-shroud design provides the rigid structure needed for precise positioning of the electrical heaters. Removal of the core assembly for maintenance is also made simpler with the single-shroud design.
To prevent bypass leakage between the inside of the core shroud and the outer perimeter of the core inlet plate, an 0-ring is placed on the core inlet plate, as shown in Figures 7.3 and 7.4.
7-8 The core inlet plate is supported by bolting the core shroud to the plate with angle supports welded onto the outer rim of the core inlet plate, as shown in Figures 7.3 and 7.4.
Table 7.7 Core Shroud Design Component SBWR PUMA Scale Shroud Height (mm) 4788 1197 1/4 Core Shroud OD (mm) 5250 525 1/10 Core Hydraulic Diam. (mm) 19.7 62 distorted 2
Core Flow Area (m ) 7.4 0.1255 distorted 7.1.6 Chimney The chimney extends from the top of the core to the bottom of the steam separator assembly.
The chimney shroud creates an annulus with the vessel wall, providing the downcomer tiow area. Following an interruption in feedwater line or a LOCA, the large reserve of water in the shroud above the core allows for an extended period of time in which core uncovery can be prevented. The height of the chimney which contains a two-phase mixture provides the natural buoyancy driving force that circulates water to the core through the downcomer.
Vertical partitions placed in the chimney shroud extend from the bottom of the chimney to about 75% of the chimney height. The partitions act as a deterrent to any lateral flow distur-bances due to non uniform void fraction profiles, and prevent flow redistribution. Such distur-bances may cause two-phase flow dynamic instability in the chimney section. The instabilities, if no partitions are present, may then lead to local circulation of the two-phase mixture in the chimney region, thus decreasing the overall natural circulation flow in the RPV. Therefore, it is important that partitions also be simulated in PUMA, not only to prevent any dynamic instabil-ity, but also to preserve the inherent two-dimensionality of the two-phase flow mixture.
A mixing plenum is present at the exit of the chimney partition and extends to the top of the chimney shroud,immediately preceding the entrance to the steam separator. The mixing plenum is an open area which allows the two-phase mixture exiting the partitions to mix and form a more homogeneous mixture, avoiding channeling effects in the separator. Table 7.8 summarizes the geometrical dimensions of the chimney. The height and diameter of the chimney are scaled using the geometrical scaling criteria of length and area. Figure 7.1 shows the location of the PUMA chimney in the RPV.
7-9 1
Figures 7.8 and 7.9 show a general schematic of the chimney shroud mounting and assembly. l The bottom of the chimney shroud is connected to the top of the core shroud by a double slip joint (Figure 7.9). The slip joint serves a number of purposes. During initial assembly, the slip joint allows for proper centering / positioning of the chimney shroud. During normal operation, the slip joint prevents any flow leakage between tlie downcomer and core region at the location where the core shroud ends and the chimney shroud begins. Finally, during RPV maintenance, the slip joint allows the bottom ponion of the vessel housing the heater rods to be readily disen-gaged. This is accomplished by simply unbolting the lower flange connection, and removing the bottom portion of the vessel with a minimum lowering clearance. The weight of the chimney shroud is supponed by bolting the top of the chimney shroud wall to a cylindrical ring welded to the bottom of the lower steam separator plate (Figure 7.8). This design also allows for minimiz-ing the number of devices needed to suppon the shroud, which would have otherwise created more obstacles in the downcomer flow area.
As mentioned earlier, the SBWR chimney shroud contains 25 square partitions. Each square partition covers approximately the cross-sectional flow area of 9 fuel cells (3 X 3). In PUMA design, it is not necessary to geometrically scale the panitions or to use panitions of similar configurations, since the SBWR chimney partition configuration'is based mainly on the square layout of the fuel assemblies.
A number of criteria were set in order to determine the PUMA chimney partition configuration. The first was the need to preserve two radial flow areas that coincide with the ring-like layout of the PUMA heaters. The second criteria called for approximately equal cross-sectional flow area through each of the partitions. Also, the hydraulic diameter of each partition would need to be larger than the expected slug size, which is discussed below.
After a number ofiterations, the PUM A chimney partition was designed to have live cylindr-ical tubes,165 mm I.D. each, with one tube placed in the center of the shroud and the remaining four symmetrically positioned around the center in each quadrant. This design results in nine partitioned flow areas and is unique in that each panition has the same cross-sectional flow area.
A cross-sectional schematic of the PUM A chimney putitions is shown in Figure 7.10.
The partions need to preserve the two-dimensionality of the two-phase flow mixture through the chimney. The duct size should be larger than the expected maximum slug bubble size, D, which is given by the following correlation [7.4):
Dn = 40 (7.5)
Sph
7-10 For steam-water mixture, the maximum slug size, which is equal to 108.4 mm, is calculated from this correlation. It is clear that the PUMA partition duct size is larger than the minimum set by the size of slug bubble formation.
7.1.7 Downcomer The RPV downcomer area extends from the top of the chimney to the lower plenum. In the SBWR, the width of the downcomer in the chimney section is .slightly larger than that in the core region, due to the difference in the outer diameters of the chimney and core shrouds. The width of the downcomer in each n:gion of the RPV is scaled in PUMA using the geometric scaling cri-teria of 1/10. This results m downcomer widths of 47.3 and 37.5 mm for the PUMA chimney and core regions, respectively. A schematic of the PUMA RPV indicating the downcomer width is shown in Figure 7.3.
In the previous section, the choice of the partitior. dimension in the chimney region was based on the requirement to support the maximum bubble slug size. In the PUMA downcomer region, similar consideration is made. It is clear that the PUMA downcomer width is smaller than the largest bubble slug size predicted by Eq. (7.5). However, the annular geometry of the downcomer region has two length scales: namely, the width of the annulus and its perimeter.
The maximum bubble size depends on.these two length scales. Since the perimeter is much larger than the maximum slug bubble size calculated from Eq. (7.5), it is expected that the bub-ble shape will be distorted, having a thickness of approximately the gap size and a width given by Eq. (7.5). Such a bubble is expected to rise with a velocity corresponding to the spherical cap bubble having a diameter equal to that of a critical size bubble. Therfore, the relative velocity in the PUMA downcomer section is well-simulated.
Table 7.8 Chimney Design Component SBWR PUMA Sb Chimney Height (mm) 9060 2266 1/4 Nominal ID (mm) 4955 496 1/10 Partition Height (mm) 6570 1642 1/4 No. of Partitions 25 9 --
Average Coolant Flow 18.55 0.185 1/100 2
l Area (m )
Nnass (i.554 0.128 1
l l
7-11 7.1.8 Separator / Dryer Assembly The steam separator assembly located directly above the chimney shroud is designed to efficiently remove entrained water from the steam-water mixture entering the separators. This provides moist steam to the dryer assembly, which then undergoes additional separation to pro-vide dry steam for the turbine generators. Information provided by GE on the steam separator performance under normal operating conditions allows for approximate pressure drop scaling of the sepamtor for use in PUMA. Scaling the pressure drop across the separator would simulate the effective flow resistance in PUMA during vessel operation.
The proprietary SBWR separator consists of a large array of standpipes welded to the top of the separator plate. Each standpipe contains internal vanes to create a centrifugal force on the steam-water mixture, pushing the water against the inner walls of the standpipe. At different elevations, pick-off rings are located in the standpipes. The pick-off rings remove the water flowing along the inner wall and divert it outside of the standpipe through openings in the pipe wall adjacent to the rings. Water separated in the standpipe flows back into the downcomer region.
The elevation of the pick-off rings (and their adjacent openings in the standpipe wall) are designed to be above the normal operating water level. During a LOCA, however, there may be sufficient swell in the two-phase mixture level to reach the openings in the standpipe wall. This can potentially create a problem as water can flow back down into the chimney mgion by enter-ing the standpipes through the openings.
Total pressure drop across the SBWR separator under normal. operating conditions is given as 26.3 kPa [7.5, Attachment B, p. 9). The gravitational pressure drop needs to be subtracted from the total pressure drop in order to obtain the frictional component, which is then scaled by 1/4 to obtain a proper scaling of the pressure drop. Based on typical operating conditions of 3
14% flow quality at 7.23 MPa (i.e. pm = 231 kg/m ), and using the SBWR separator height [7.6],
the gravitational pressure drop is calculated to be approximately 7.65 kPa. Therefore, the proto-type frictional pressure drop under these operating conditions is estimated to be 18.65 kPa. Scal-ing this value by the pressure drop scaling of 1/4, PUMA frictional pressure drop is given as 4.66 kPa.
In order for the PUM A separator to provide this pressure drop, the mass flow rate through the separator must also be scaled to obtain the correct flow area and determine the number of separa-tor tubes. Typical SBWR mass flow rate in the core is estimated to be 7,690 kg/s {7.3, p.1.3-2).
When scaled by the flow scaling criteria of 1/200, the PUMA mass flow rate in the core is found to be 38.5 kg/s. Setting the minor loss coefficient, IGvalue, across an opened tube to 1.5, which cormsponds to one sudden expansion and one sudden contraction in a short open-ended pipe, the
. 1 7-12 l
l i
velocity of the mixture for a pressure drop of 4.66 kPa is calculated to be 5.18 m/s from the fol- :
Icwing relation: .
I KpV 2 AP = n (7.6) 2 Assuming n separator tubes, atypical flow condition of 14% flow quality at 7.23 MPa (i.e.
pm=231 kg/m 3) and a mass flow rate of 38.5 kg/s, the flow area can be calculated from 4 l
p a V = '38.5 (7.7) i I
Solving for the flow area, a, the following condition is obtained:
a=
(7.8) n A list of tube diameters and the equivalent number of tubes satisfying the above condition is given in Table 7.9.
Table 7.9 Required Number of Separator Tubes Number and Diameters n Diameter (mm) 10 63.8 13 56.0 16 50.8 17 49.0 63 25.4 From this table, commercially available tube sizes and a reasonable number of tubes, a value of sixteen for @ italic n@ was chosen, with a pipe size of 50.8 mm (2 inch) diameter. The length of these tubes and the mounting mechanism of the lower steam separator plate will be discussed later in this section. The proposed layout of the tubes in the separator plate is given in Figure 7.11.
In order to provide an additional mechanism to separate entrained liqu'id from the steam, an upper steam separator plate containing a similar number and size of holes as the lower plate (described above) will be placed above the separator tubes. This upper steam separator plate
I 7-13 l l
1 will contain short open-ended tubes placed directly above the holes in the plate. The tube lengths ,
will be discussed later in this section. The openings are positioned so they are not directly above !
the separator pipes, causing a change in the steam-water mixture flow path. Any entrained liquid not separated in the first stage will exit the separator tubes with the steam, impact on the top separator plate and fall back down without flowing directly into the top separator tubes. The proposed layout of the tubes in the upper separator plate is given in Figure 7.12. Elevation of the -
upper steam separator plate and its mounting mechanism will be discussed later in this section.
Since the PUMA steam separator design is not based on geometrical similarity with the SBWR separator assembly, the choice of separator tube length and upper steam separator plate is somewhat arbitrary. The only device in the SBWR steam separator which needs to be properly scaled in PUMA using the length scale criteria is the elevation of the first level of pick-off ring i openings in the SBWR stand pipes. As mentioned above, in the case of sufficient water inven-tory swell in the downcomer region these openings will be the first access points for water flow back into the standpipes and down into the chimney region. Hence, four symmetrically posi-tioned holes an made in the PUM A separator tube walls, as shown in Figure 7.13. The elevation of these holes is based on 1/4 length scale of the elevation given at the lowest level of pick-off ring openings in the SBWR standpipes (7.5].
Allowing a sufficient clearance from the side holes in the separator tubes, the determination of the length of the open-ended tube, skwn in Fgure 7.13, is complete. A sufficient gap is allowed between the top of the separator tuas and the upper steam separator plate to allow the steam-water mixture to separate as the mixtare impinges on the upper plate. Hence, the steam separator tubes extend from the lower steam separator plate to 60 mm below the upper steam separator plate, as shown in Figure 7.13.
The SBWR dryer consists of vertical perforated plates placed in a parallel configuration.
Several flow path changes through the vertical channels cause moisture from the wet steam to collect on the plates and fall into horizontal collecting trays located at the bottom of the dryer plates. The trays then feed into a vertical duct which acts as a skimmer and retums the water to the downcomer region beneath the normal water level. This prevents the collected water from being entrained back into the upflowing steam mixture.
In order to simulate the skimmer duct, a single vertical tube is connected beneath the upper steam separator plate, as shown in Figure 7.13. The length of the tube is nased on the 1/4 length scale to properly position scale its bottom end in respect to the normal water level.
In order for water collected on the upper steam sep.arator plate to drain back down through the skimmer tube, only short tubes (76.2 mm) are needed above the separator plate, as shown in Figure 7.13.
1
7-14 As shown in Figure 7.13, the lower steam separator plate, the separator tubes and chimney shroud are all supported by a ring which is, in turn, anchor bolted to the inside of the vessel at several locations. This support ring does not exist in the SBWR, so it is perforated to allow for communication across the ring wall.
The support ring extends approximately 50.8 mm below the lower steam separator plate.
This Icwer portion of the support ring is not perforated and serves several purposes. First, it allows the chimney shroud to be properly positioned during installation procedures by guiding the shroud n fit closely against the support ring. Also, the weight of the chimney shroud and partitions are supported by bolting the top of the shroud to the inside of the ring, as shown in Figums 7.8 ano 7.13 Finally, the close fit between the chimney shroud and supporting ring minimizes the pcssibility of steam rising from the chimney and escaping into the downcomer region.
In the SBWR, the outer periphery of the steam dryers forms a wall-like barrier, preventing the separated Jry steam from directly entering the downcomer an:a. As shown in Figure 7.2, the steam rises toward the top of the vessel, then reverses direction as it travels into the downcomer region. Tnis steam flow redirection is simulated in PUMA usipp a cylindrical ring called the dryer dirt. The height of this skirt is scaled by 1/4 height scaling. The bottom of the skirt extends down into the separator region similar to the SBWR barrier around the separator region.
The bottom elevation corresponds to the point where the skimmer tubes n: turn water to the downcomer region. This elevation is dso scaled by 1/4 height scaling. The main purposes of this shroud-like barrier are to direct steam leaving the separator standpipes into the dryer region and to prevent steam from escaping into the downcomer region.
The weight of the upper steam separator plate, short tubes and dryer skirt is supported by anchor bolting the bottom of the skirt to the inside the vessel. A vessel flange connection is placed at a specified location in the steam separator assembly to allow access to the support structures, as shown in Figure 7.2.
7.2 Drywell Design The SBWR containment, as shown in Figure 4.1, is a reinforced concrete cylindrical struc-ture which houses the RPV, GDCS, SP, ADS and their related components. The containment wall is designed to provide a leak-tight containment boundary. The containment is divided into a drywell (DW) region and a suppression pool (SP) chamber region (or wetwell) which are inter-connected through vent lines.
The drywell region is designed to provide a leak-tight gas space and boundary against the release of radioactive fission products, steam and/or water released during a LOCA. It is also l
i
7-15 designed to withstand the pressure and temperature loading associated with a bmak in any of the main system pipes inside the drywell and to withstand the negative differential pressures associ-ated with a depressurization event. The drywell structure is designed for maximum temperature and pressure conditions of 171 C and 483 kPa (70 psia). The drywell directs nearly all the steam released during a LOCA into the suppression pool through the DW/SP vent system.
The prototype drywell consists of an upper drywell volume and a lower drywell volume, which an: divided by the vessel suppon skirt and connected by an open flow area to allow for pressure equaliz.ation. The upper drywell surrounds the upper portion of the RPV and houses the ADS, PCCS/ICS piping, GDCS and other related systems. The lower drywell houses the portion of the RPV below the vessel support skin, the control rod drive system and other vessel-bottom piping. It also acts as a sump, to collecting any water that enters the containment.
In integral systems, the mass flow from one component to another is an important parameter in measuring the system's accurate mass conservation and in explicitly studying flow behavior.
In order to allow for such measurements in the drywell and the RPV, the drywell is designed as separate from the RPV, though they are connected by piping. This simplification in design facili-tates construction of the vessels, instrumentation and flow visualization.
From the geometrical scaling criteria, the volume of the drywell is scaled by 1/400 and the height is scaled by 1/4. The resultant model will have an appropriately-scaled cross-sectional area. In Table 7.10, the dimensions of the PUMA drywell are given along with information on the piping connected to the drywell.
In the present model, the annular geometry of the drywell around the RPV is modeled as a single cylinder. The reasons for this are two-fold. First, if the annular geometry of the drywell were to be scaled as an annulus, the width of the annulus would be scaled by 1/10 due to geometric scaling consid: rations. This would substantially increase frictional losses due to the reduced hydraulic diameter. Second, construction of a single cylinder would be simpler compar-ised to an annulus. Figure 7.14 shows a schematic of the PUMA drywell design.
The top portion of the upper drywell is shown in Figure 7.14. The scaled volume of the upper dome in the prototype drywell has been added to the upper head volume in PUM A. This is due to the fact that the dome volume contributes only 5% of the total drywell volume. and its presence does not significantly affect the mixing or flow distribution in the upper drywell. This approach also simplifies construction of the PUM A drywell.
l
7-16 Table 7.10 PUMA Drywell Design
- Number of Units: 1 e Upper Drywell Height: 1525 mm o Upper Drywell Diameter: 2750 mm -
e Annular Section Diameter: 850 mm o Annular Section Height: 43(K) mm e Lower Drywell Height: 68(X) mm e Lower Drywell Diameter: 890 mm o Connecting Lines:
Line Function Line From Line To Pipe I.D. (mm/ inch)
MSL RPV Drywell 76.2/3 MSL (To DPV) RPV Drywell 76.2/3 DPV Line RPV Drywell 76.2/3 FWL Break RPV Drywell 38.1/1.5 GDCS Line Break GDCS Line Drywell 50.8/2 PCCS Supply Line Drywell PCCS Tank 38.1/1.5 Drywell Vent Drywell SP 407/16 Vacuum Breaker SP Drywell 50.8/2 Vacuum Breaker Leak Vac, Break Line Drywell 6.35/0.25 RWCU/SDS Bnzak RWCU/SDS Line Drywell 6.35/0.25 A brief analysis is given here to compare important safety phenomena between the SBWR and PUMA in the containment following LOCA behavior. An important phenomena expected to occur in the drywell region of the containment is mixing and stratification of steam and non-condensible gases. The degree of mixing / stratification in the drywell affects a number of other processes in the SBWR. For example, the performance of the PCCS, whose inlet is drawn from the upper region of the drywell, will be affected by the level of mixing / stratification of steam and '
non-condensible gases in the drywell. As the main driving force for the flow of steam and non-condensibles into the PCC condensers is based on the condensation-induced pressure difference between the drywell and suppression pool, pressure in the drywell will also be affected by mixing / stratification. In turn, the mixing / stratification phenomena will also be affected by the rate of steam condensation on the walls of the drywell.
I As the PUMA containment is initially filled with air, the main source of steam into the con-tainment will come from DPVs or breaks in the steam lines. Steam flow out of the RPV may also include some hydrogen from radiolysi'. of wata within the core. The behavior of the 4
. =. _
7-17 steam / hydrogen flow through DPVs or breaks into the drywell will initially be characterized as turbulent buoyant jets. As the density and velocity of the jet diminish, gravitational forces begin to affect the jet's motion. The jet will enter into an intermediate region, and eventually buoyancy foxes will dominate the jet's motion [7.7]. One of the important dimensionless parameters used when the inertia and buoyancy forces are coupled,' which can also be used for scaling purposes, -
is the modified Froude number defined as u'2 (7.9)
Frj = g d j(p, - pj)/pj where uj, dj and rj are the jet exit velocity, jet diameter at the source and jet density at exitcand
- p. is defmed as the containment mixture density.
Based on a large experimental database for round jets discharging into an infinite volume of fluid of uniform density, Rodi [7.7) provided empirical relationships for predicting the axial decay of the' maximum (centerline) jet velocity away from the jet source and the excess density for three different flow regimes. The empirical correlations fo'r each flow regime are given below. ,
(a) Non-buoyant region:
t/2 ' '
-l um ' pj ' x
= 6.2 (7.10) u, p, dj in i Pa - Pm Pi x
= 5.0 (7.1I)
, Pa - Pi , , Pa , dj 1
(b) Intermediate region:
'900 " '
-4/5
- l ' *
= 7.26 Fri '3" (7.12) u, p, d)
I l
l l
i
l 7-18
-7/16 ' '
-5/4
~
= 0.44 Fr8 A 1 dj (7.13)
, Pa - P; , , Pa , ,
(c) Plume region:
10 '
lO f1
= 3.50 Fr-lo A (7.14) u)
2 p, ,
dj 10 ' '
-5B
' ~ *
= 9.35 Fr3 38 A 1 (7.15)
, Pa - P3 , Pa , d, where the subscript m refers to the jet centerline, and x is the coordinate in the normal direction.
A dimensionless group incorporating the modified Froude number is used to set the criteria for transition from one flow regime to the next.
-u4 < 0.5 non-buoyant l/2 PJ X Fri 0.5 s and s 5 intermediate (7.16)
- E' s J
> 5 plume Consider the steam flow from a DPV stub valve after the ADS has been activated. Although the orientation of the DPV opening provides for a horizontally-directed jet flow, the SBWR design includes a deflector plate located directly in front of the flow path. The main function of this plate is to prevent internal damage to the containment due to direct impact of the steam.
The tentative SBWR design of this deflector resembles a semi-circular pipe, vertically oriented.
Acting as thrust-reversers, the deflector plates can be considered to divide the mass flow into two approximately equal vertical streams as shown in Figure 7.15. Therefore, the effective jet diam-eter may be taken as dj = E d y, (7.17) l l
1
_______________l
7-19 The empirical correlations of Rodi [7.7] have been applied to a single-source steam jet for the SBWR and PUMA drywells at the same velocities. Figure 7.16 shows the results of the cal-culations for a range of Froude numbers. The range of Froude numbers covered here corresponds to the steam velocities at the jet source from initial blowdown to long-term decay heat vaporiza-tion from the core. The value of x/dj indicated for each drywell corresponds to the top of the drywell, where dj is based on the effective jet diameter. The top of the drywell is the most important location to consider when determining the dominant flow regime, since it is at this point where the jet impacts the containment wall and determines the degree of mixing / stratification and condensation.
The results indicate that all three flow regimes can be expected in the PUMA drywell. :,;milar to the SBWR. As the Froude number (i.e., jet velocity) reaches low values, buoyancy effects become dominant and the jet behaves as a buoyant plume. Hence, possible flow stratification in the drywell may be expected in the long-term behavior of the drywell after blowdown. The Froude number at which transition occurs from one flow regime to another, however, is slightly shifted to higher Froude numbers in PUMA. This implies that for the same jet velocity, the tran-sition from non-buoyant to intermediate to plume flow regimes may occur sooner in the PUMA drywell than in the SBWR drywell. '
It should be emphasized, however, that the Rodi corn:lations are strictly valid for round, vertical jets in an infinite volume of fluid of uniform density. They do not include any interac-tions such condensation, the effects of internal structures or walls, or mixing due to shear, all of which can affect the degree of mixing / stratification and condensation. There is currently no data-base for the dynamic behavior of jets in finite volumes of fluids, enclosed by boundaries. Hence, the above analysis pmvides for approximate results and comparison between drywell behavior in SBWR and PUMA.
It is also interesting to compare the PUMA-to-SBWR dimensionless numbers considered here. Assuming similar fluid properties, the modified Froude number ratio for both critical and suberitical flow velocities is given as
~
u 2i p
.R
$ 14 critical flow (7.10 gd)Ap/p> .R
. 8 .R The PUMA-to-SBWR ratio of the dimensionless group used by Rodi [7.7], Ey. (7.16), for deter-mining flow regime transition is also determined here for the critical and suberitical flow velo-city conditions,
7-20
~ ~
1/4 Fr-t/2 .i P x _
[ 0.94 critical flow M *d b (7'19)
- p. , dj _j[,g u
. .R The previous discussion centered on the' comparison of the jet flow regime in the drywells of the SBWR and PUMA. We now apply the scaling criteria discussed in Section 5.4.7, where additional scaling issues introduced by Peterson, et al. [7.8] regarding mixing and stratification were presented. The final analysis of that study resulted in a dimensionless number representing the number ofjets in a given volume, 1
njR = .
(7.20)
(H/D)',a where H/D is the aspect ratio. The PUMA to SBWR ratio of H/D is given by (H/D)g = 2.5 (7.21)
Therefore, from Eq. (7.20) the ratio of the number ofjets is approximately 1/6.
Based on the Peterson, et al. argument, preserving this ratio would result in a proper scaling of the degree of mixing and stratification. However, this ratio is clearly not preserved in PUMA, as the PUMA to-SBWR ratio of potential jet sources in the drywell is 0.66 (4 DPVs in PUMA,6 DPVs in SBWR). It can be expected that, for given jet conditions, a higher level of mixing is expected in PUMA.
7.3 Suppression Pool Design As discussed in Section 7.2, the containment is divided into a drywell region and a suppres-sion chamber region (or wetwell), which are connected by vent lines. The suppression chamber consists of the suppression pool and the pas space located above the pool. The suppression pool is a large reservoir of water capable of absorbing a large amount of energy by condensing the steam discharged from the safety relief valves (SRV) in case of pipe break accidents a LOCA.
The water serves as an additional source of reactor water makeup and as a reactor heat sink through the GDCS equalization line that connects the suppression pool to the RPV.
7-21 The gas space above the suppression pool is designed to be leak-tight, as it serves to collect ,
and retain any non-condensible drywell gas following a LOCA blowdown without exceeding the I design pressum of the containment. The design tempemture and pressure of the suppression chamber are given as 121 C and 483 kPa (70 psia), respectively.
The suppression pool is connected to the drywell through the drywell/ suppression pool (DW/SP) vent system, which is comprised of eight vertical / horizontal vent modules. Each module consists of a vertical flow channel extending into the suppression pool water with three horizontal vent pipes opening into the pool, as shown in Figure 4.1. In the event of a LOCA or pipe break within the drywell, the incmased pressure inside the drywell forces a mixture of steam, water and non-condensible gases thmugh the DW/SP vent system. The steam quickly condenses in the pool, and the noncondensible gases rise and collect in the gas space volume of the suppression chamber. Hence, the suppression pool also acts as a means of preventing over-pressurization of the drywell.
The SRVs also discharge steam into the suppression pool through discharge piping con-nected to spargers near the bottom of the pool. Similarly, the noncondensible gas vent lines from the PCCS and ICS are routed through the suppression pool and vented mio the pool water. The gas rises to the top of the water level and collects in the pas space volume.
In the event that the pressure in the gas space exceeds the pressure in the drywell, vacuum breakers located on top of the suppression chamber open to the drywell region. Each of the three vacuum breakers are equipped with check valves designed to begin opening when the pressure in the suppression chamber rises to 3.45 kPa (0.5 psi) above the pressure in the drywell. The vacuum breakers become fully open when the pressure in the suppression chamber rises to 13.8 kPa (2 psi) above the pressure in the drywell. The vacuum breakers prevent the overpressuriza-tion of the suppression chamber, which can only operate properly when its pressure is below that of the drywell under these conditions, the DW/SP vent system and PCCS remain operational.
Based on the previously discussed geometrical scaling criteria for the PUMA model, the volume of the SBWR suppression pool is scaled by 1/400, and the height is scaled by 1/4. This scaling approach ensures that the cross-sectional area is also properly scaled. The configuration of the SBWR suppression pool resembles an annular pool nearly surrounding the RPV. How-ever, a simplification is made in the PUM A model and the pool is designed as a single cylindri-cal vessel, as shown in Figure 7.14.
To model the eight vertical vent modules connecting the drywell to the suppression pool, a single vent line, scaled by the total prototype vent flow area using the boundary flow scaling cri-teria, is used, as shown in Figure 7.14. The single vent line is centered in the suppression pool with multiple openings on the perimeter of the submerged line, simulating eight prototype vents
7-22 opening into the suppression pool. Table 7.11 lists the suppression pool design parameters.
A number ofissues need to be considemd when scaling the horizontal portion of the vent lines opening directly into the suppression pool. As discussed in Section 5.4.7, different scaling criteria apply to the vent openings depending on w.hether the steam /non-condensible gas mixture exiting the suppression pool is in the jet or bubble flow regime. Applying the general boundary flow scaling to determine the PUMA vent size may lead to over-condensation in the horizontal and vertical jet flow regimes. The condensation may be under-rated in the bubbly flow regime.
Themfore, different vent sized openings may lead to different rates of condensation in the suppression pool.
Scaling the height of the vents also needs to be considemd. This approach is more straight-forward. As 1/4 height scaling has been implemented throughout the PUMA facility, we main-tain the 1/4 height scaling for consistency. Hence, the location of the three rows of vent open-ings in the single vertical vent pipe is determined using 1/4 height scaling.
Because of the difficulty in scaling the vent sizes to m tintain similar condensation rates under all possible flow regimes, the driving factor in the PUMA design is flexibility of use. This requims designing the vent size in such a fashion which the openings can be adjusted depending on the local conditions, so as to maintain similar rates of condensation. This approach will also allow separate tests to be performed on the condensation phenomena in the suppression pool.
The PUMA design calls for three rectangular-shaped slots at each of the thme raw elevations on the vertical vent line, as shown in Figure 7.14. The size of each opening is oversized, due to the general boundary flow are scaling criteria, to measure 2(X) mm in width and 175 mm in height.
Bolt holes are are placed around the periphery of each opening. The bolt holes will allow plates of various sizes and configurations to be connected to the openings in order to adjust the number and size of vent openings.
The three vacuum breakers located at the top of the SBWR suppression chamber, opening into the drywell, are also scaled in PUMA. Each of the three lines is scaled by the boundary flow scaling criteria. Therefore, the flow area of the PUMA vacuum breakers is scaled by 1/l(H) of that in the SBWR. Each vacuum breaker line in PUMA is equipped with a check valve which is designed to open when the pressure in the suppression chamber rises to 862 Pa (0.125 psi) above the pressure in the drywell. This value is obtained by applying the 1/4 differential head scaling criteria to the prototype value of 3.45 kPa (0.5 psi).
Once the pressure in the suppression chamber equals that in the drywell, the vacuum break-ers are expected to shut to a leak-tight position. This is essential for the suppression
7-23 pool / chamber and PCCS to perform properly. However, in the case of a malfunction where a leak path remains in the vacuum breakers, this will have a negative affect on the long-term safety system of the SBWR. In order to simulate this scenario, a separate, smaller line is con-nected between the suppression chamber and the drywell, as shown in Figure 7.14. This line is to be equipped with an operator-actuated solenoid valve. During a test where the vacuum breaker leak condition is to be included, tlie solenoid will open to simulate a leak in the vacuum breakers.
Table 7.11 PUMA Suppression Pool (SP) Design e Number of Units: 1 e Tank Height: 2838 mm e Tank Diameter: 2817 mm
. Connecting Lines:
Line Function Line From Line To Pipe ID (mm/ inch)
SRV Line MSL SP 50.8/2 GDCS EQ Line SP RPV 12.7/0.5 PCCS Non-Cond Vent PCCS Tank SP 38.1/1.5 ICS Non-Cond Vent ICS Tank SP 12.7/0.5 SP Vent Drywell SP 356/14 SP Feed / Drain SP Drain 25.4/1 Vacuum Breaker Drywell SP 50.8/2 l
7.4 Gravity Driven Core Cooling System (GDCS) Design i
As part of the Emergency Core Cooling System (ECCS), the Gravity Driven Core Coolmg j System (GDCS) plays a major role in the SBWR safety mechanisms. The GDCS can be con-sidered as two separate systems: a short-temi safety system and long term safety system.
The former is designed to provide short-term water makeup to the reactor vessel for main-taining fuel cladding temperatures below safety limits. Three separate water pools, located within.the upper drywell at an ekvation above the active core region, provide gravity-driven water makeup to the rector vessel. A 203 mm (8 inch) drain pipe from each of the GDCS pools passes through a loop seal, then branches into two 152 mm (6 inch) lines which feed into the dowricamer annulus region of the reactor vessel through flow-restricting nozzles. Squib valves located on the GDCS drain lines are actuated 150 seconds after Level 1 is reached in the reactor vessel. ,
The long-term GDCS safety system is designed to provide long-term vessel cooling by keep-ing the core region covered with water, again through gravity-driven flow. This is accomplished
7-24 through three GDCS equalization lines connecting the suppression pool to the reactor vessel.
Each line is independent and designed to open when the water level in the reactor vessel reaches 1 m above the top of the active fuel (TAF).
For PUMA, the volume and elevation of each of the GDCS pools is scaled using volumetric (1/400) and height (1/4) scaling. Due to the reduced diameter and height of the PUMA reactor vessel, it is also necessary to reduce the number of GDCS penetrations into the vessel in order to maintain some similarity to the SBWR in the degree of mixing occurring in the downcomer region by each GDCS penetration's plume. A reduction in the number of penetrations into the PUMA reactor vessel is also consistent with the overall PUMA aspect ratio of 2.5. Hence, the two 152 mm (6 inch) lines branching from the main 203 mm (8 inch) drain line are combined and scaled into one line for PUMA. As shown in Figure 7.17, this two-line-into-one combina-tion is performed for 2 of the 3 PUMA GDCS pools.
In order to simulate s ,alfunction or break in one of the 6 inch SBWR penetration lines, it is also necessary for PUM .o maintain a scaled, branched line for the third GDCS pool. There-fore, the total number of JDCS drain penetrations into the PUMA reactor vessel is reduced from six to four. Loop seals in each GDCS drain line are also scaled in PUMA using height scaling criteria.
In the SBWR, the GDCS drain line from each GDCS pool is elevated from the bottom of the pool, leaving a small volume of water at the bottom of the pool once the pools are drained. The elevation of this drain line penetration is designed to prevent water flowing out of the reactor vessel through the DPV lines, once all GDCS water is drained into the vessel. The elevation of this penetration into the GDCS pool is also simulated in PUMA using 1/4 height scaling, as shown in Figure 7.17.
Each GDCS drain line in PUMA is equipped with a fast-acting, electric-actuated, full-port ball-valve. Analog signals are used to activate the valves. The logic timing governing the ana-log signals and the opening of the valves is similar to that used for the SBWR GDCS drain line activation.
Each of the three GDCS equalization lines used for long-term cooling of the reactor core is simulated in PUMA, as shown in Figure 7.17. The penetration elevation of each line between !
the suppression pool and reactor vessel is scaled using 1/4 height scaling, and the pipe size is chosen using boundary flow scaling criteria.
Each PUMA GDCS equalization line is equipped with a fast-acting electric actuator mounted on a full pon ball valve. The actuator will receive an analog signal from the main relay control board to determine the appropriate time for the valve to open. As described earlier in this
7-25 section, the GDCS equalization line is designed to open when the water level in the RPV drops to a pre-specified level. Since the water level m the RPV will be continuously measured during
)
the tests, a set point representing the specified level in the RPV can be programmed into the data l acquisition software that will send a signal to the relay control board to activate the eleitric actuator to open the ball valve. An electronic flow switch mounted on the electric actuator will send a signal back to the a: ,
indicating that the valve has fully opened.
Boundary flow scaling entena are used to properly scale the flow area for gravity-driven flow lines such as the GDCS drain lines and GDCS equalization lines. This approach is common to all piping in the PUMA facility. The methodolo,cy for applying the boundary flow scaling cri-l teria to piping is given in detail here.
In boundary flow scaling, the approact- pply the continuity equation and derive the area and velocity scaling relation as discussed previously in Chapter 5. From the boundary flow scal-ing criteria we have r -
au
=1 (7.'2) o o Since,[a]g = 1/l(X) and [u]g = 1/2, the bc undary mass flow rate requirement becomes tha = 1/200 (7.23)
Using the momentum equation, a relationship between the area ratio and the loss coefficient ratio can be obtained from the momentum equation Ap = pu K = h 2
,' K (7.24)
- pa Here, K is the total loss coefficient, including the friction and the minor losses. a is the cross-sectional flow area of the pipe and Ap is the pressure drop across the pipe for the mass flow rate, !
rh. Hence, the ratio of op for prototype to model becomes l
l 1
l l
1
. I 7-26 l
.2 mR Ap =
2 ER (7.25)
.R ag 1
i . 1 Substituting tha = in the above equation, we obtain ag 1
= (7.26)
QKg 2W)iApg From pressure scaling, it is known that if the flow is mainly gravity-driven, then the pressure drop ratio becomes equivalent to the height ratio, i.e.,1/4. Water flow in pipelines such as the GDCS drain lines, GDCS equalization lines, PCCS/ICS condensate lines and PCCS supply lines can all be considered to be gravity-driven, and the scaling criteria of (AP)g = 1/4 applies to these lines. However, flow in lines such as the steam lines, DPV lines and SRV lines are mainly deter-mined by the pressure difference between two components or vessels. In these cases, the pres-sure scaling given by (AP)g = 1 is applied. From Eq. (7.26), since the prototype loss coefficients and the diameters of the pipes are known, the PUMA pipe size can be determined by an iterative process. For practical purposes, it is necessary to choose commercially available pipe sizes and adjust the pipe loss coefficient of PUM A by placing an orifice in over-sized pipes.
As described earlier, each SBWR GDCS pool has one 203 mm (8 inch) pipe that branches i into two 152 mm (6 inch) lines before connecting to the RPV. In the PUMA facility, the two I branch lines are replaced by a single drain line for two of the three GDCS pools. From the ;
mcmentum equation applied to the prototype drain pipe we have !
l Ap = pu2 Ki + h pu2K 2 (7.27) where subscript I refers to the 203 mm (8 inch) line and subscript 2 refers to the 152 mm (6 inch)line. From the continuity equation we have l
aiui = .!a:u2 -
(7.28)
The factor 2 on the right hand side of Eq. (7.28) is due to the two 152 mm (6 inch) lines
7-27 branching from the 203 mm (8 inch) lines. From Eqs. (7.27) and (7.28) we have Ap = pK,qu3 (7.29) where K,y is defined as the equivalent loss coefficient given by 4a3 K,y = 2 K+K2 i (7.30) ai As the flow is driven by gravity for the GDCS drain lines, the pressure drop scaling is given by (Ap)g = 1/4. Hence, Eq. (7.26) can be written as 2
d,2n/d 3
= (7.31) l(K QKm/K p The PUMA drain pipe replaces two 152 mm (6 inch) Enes with a single line. Therefore, an equivalent diameter for this single line can be calculated by combining the flow area for each 152 mm (6 inch) line, which is given by d p.,y = 215.7 mm (8.49 inch) Using the loss coefficients for 203 mm (8 inch) and 152 mm (6 inch) lines from infonnation given in [7.3], K,y for the prototype can be calculated from equation (7.30) as Kp ,,y = 13.59. From Eq. (7.31), the scaled PUMA pipe diameter can be detennined if Km is known. However, Kmcannot be deter-mined until the pipe size is known. A pipe size has to be assumed for which equation (7.31) can be satisfied. For 25.4 mm (1 inch) pipe, the total Kmwas calculated as 33.69 for the PUMA facility. Substituting these values K p. y = 13.59, Km = 3. dp ,y = 215.7 mm(8.49 inch). dm = 25.4 mm(1 inch) into Eq. (7.31) we have (dm/dp)2 3
= (7.32) l ! 3'4 QKm/Kp This is slightly less than the scaled requirement of 1/100 as given in Eq. (7.31). Next,
7-28 considering a 38.1 mm (1.5 inch) pipe, the corresponding loss coefficient is 131.75. Then the left-hand side of Eq. (7.31) is calculated to be 1/99.75, which is nearly the same as the scaled requirement. Hence,38.1 mm (1.5 inch) pipe is the appropriate choice for the GDCS drain line where 2 branched lines are combined into one. ,
In order to conserve space and simplif: istruction procedures, the three PUMA GDCS pools are designed to fit into a single cylindr As shown in Figure 7.18, the tank is verti-cally partitioned into three independent poc Sting each of the three GDCS pools. The tank is sized so that the liquid volume in eac: nartitioned pools conesponds to the 1/400 volume scaling of the SBWR GDCS pool. Th. of the partition is designed to be above the normal water level, though it does not extend m the top cover of the tank. This allows for a common gas space above the pools. The GDCS pools in the SBWR are open to the containment at the top, allowing for pressure equalization. In PUMA, a single line connects the common gas space at the top of the GDCS tank to the upper drywell. Table 7.12 summarizes the PUMA GDCS design parameters.
Table 7.12 PUMA GDCS Design e Number of Tanks: 1 e Tank Height: 1525 mm e Tank Diameter: 1540 mm o No. of Partitions 3 e Connecting Lines:
Line Function Line From Line To Pipe 1.D. (mm/ inch) )
GDCS Drain GDCS Tank RPV 38.1/l.5 GDCS Line Break GDCS Line Drywell 50.8/2 i GDCS Air Supply Drywell GDCS Tank 50.8/2 )
7.5 Passive Containment Cooling System (PCCS) and Isolation Condenser System (ICS)
Design.
The Passive Containment Cooling System (PCCS) is designed to remove the core decay heat that has been rejected to the containment after a LOCA, whereas the Isolation Condenser System (ICS) has been designed to remove decay heat from the-RPV after any reactor isolation follow-ing interruption in normal reactor operation.
i l
7-29 In the SBWR, a total of three PCCS condensers and three ICS condensers are submerged in a large, interconnected ICS pool of water, which is located outside and above the containment.
Each PCCS and ICS condenser is designed to dissipate a maximum of 10 MWt and 30 MWt energy, respectively. The volume of water in the,ICS pool is sufficient to provide decay heat removal for 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> following a LOCA without replenishment. Boil-off from the pool is vented to the ambient.
Each PCCS condenser consists of two identical modules with 248 tubes per module [7.6].
The tubes are 1800 mm in length, with 50.8 and 47.5 mm O.D. and I.D. respectively. Each PCCS condenser is connected to a 254 mm (10 inch) line that vents the non-condensibles to the suppression pool, and a 152 mm (6 inch) line that returns the condensed water to the GDCS pool. The inlet supply to the PCCS condenser is a 254 mm (10 inch) line. This inlet is always open to the drywell to allow free flow of steam / gas from the drywell to the PCCS condenser tubes. The vertical condenser tubes of the PCCS modules are connected between two drums act-ing as the inlet and outlet plenum. The driving head of the PCCS is provided by the pressure difference between the drywell and the suppression pool. There are no valves, pumps or fans in the PCCS, which makes it a passive system by design. The PCCS is a unique design of the SBWR, which does not exist in any operating BWRs. On the other hand, the non-condensible vent from the ICS is not a fully passive system like the PCCS. The vent lines from the ICS are equipped with manualy-operated solenoid valves.
Each ICS condenser also consists of two identical modules, but with 120 tubes per module
[7.6]. The tubes are 1800 mm in length, with 50.8 and 46.6 mm O.D. and I.D., respectively.
Each ICS condenser is connected to a 254 mm (10 inch) inlet line that receives steam from the RPV and a 152 mm (6 inch)line that returns the condensate to the vessel. Any non-condensibles in the ICS condenser are vented to the suppression pool via a 19.05 mm (0.75 inch) line. The 254 mm (10 inch) steam supply line shares the same stub line with a squib-type depressurization valve (DPV) which is part of the ADS system.
In scaling the PCCS and ICS for PUMA, first the PCCS condenser tubes will be considered.
This will be followed by the ICS condenser tubes, then the PCCS and ICS headers, and finally the pool sizes. ;
Since the basic function of the PCCS is the removal of decay heat, the condenser tubes are- ;
l scaled using power scaling. For a 1/4 height scaled system, the power density is scaled by
=2 (7.33) gi = di~
I l
7-30 I l
l l
For the mactor core, the total power is scaled using the power ratio 1
[ power]g = q[ Va = (7.34)
~
Themfore, the condensation capacity of the PCCS is scaled by a factor of 1/200. Since the con-densation capacity is proportional to the tube inner surface area, the tube surface area is also scaled by 1/200. The underlying assumption is that negligible condensation is occurring in the PCCS headers. The surface area ratio is then given by Sg = [ power]g = ng(xd)g Hg (7.35) where n is the number of tubes, d is the tube diameter, and H is the tube height.
Maintaining a tube diameter similar to that in the SBWR, and applying the 1/4 height scaling criteria to the tube length, the PCCS tube number scaling criteria is given by i
ng = 50 (7.36)
As noted earlier, each of the three SBWR PCCS units consists of two modules, with 248 con-denser tubes per module. The approach taken in PUMA is to combine two modules into one.
Applying the criteria given by Eq. (7.36), this results in approximately 10 condenser tubes for each of the three PUMA PCCS condenser units.
Volume scaling of the PCCS condenser tubes is not considered here, since the tubes do not store an initial water inventory as the ICS condenser tubes do, and the volume scaling in the PCCS tubes is considered to be a secondary effect in comparison to the heat transfer area scaling preserved as described above. The volume of steam in the PCCS condenser tubes (assuming tubes full of steam) represent a relatively negligible volume of water in comparison to the RPV inventory.
In scaling the ICS condenser tubes, a similar method to that described for heat transfer scal-ing in the PCCS is also applied. However,in addition to heat transfer scaling, volume scaling in I the ICS condenser tubes must also be considered. Since the steam supply from the RPV to the ICS is always open, and the condensate drain line returning to the RPV is closed until activated, condensate is expected to fill the condenser tubes and headers in the ICS. By applying the volume scaling to the ICS condenser tubes, this initial volume of condensate will also be i
7-31 properly scaled. Volume scaling of the headers will be discusred later in this section.
The volume in the ICS condenser tubes is scaled by the volume ratio given by Vg = ng(nd[) Fig = 1/4(X) (7.37)
In applying Eqs. (7.35) and (7.37) to the ICS condenser tubes, it is clear that by maintaining 1/4 height scaling in the tube lengths and prototypical tube diameter, both equations cannot be satisfied with a single tube number ratio. Hence, we need to allow the PUMA tube diameter to differ from that of the prototype, while maintaining 1/4 height scaling on the tube lengths.
Dividing Eq. (7.37) by Eq. (7.35), we obtain a scaling criteria for the ICS condenser tube diameter:
Vg i Sg = dg = 2 (7.38)
Substituting Eq. (7.38) in either of Eqs. (7.35) or (7.37), the ICS tube number scaling is given by I
ng = g (7.39)
As noted earlier, each of the three SBWR ICS units consists of two modules with 120 con-denser tubes per module. Similar to the PUMA PCCS modeling, the approach taken here is to combine two ICS modules into one. Therefore, based on the criteria given by Eq. (7.39), the PUMA ICS is designed with 10 condenser tubes for each of the three condenser units. As well.
application of Eq. (7.38) determines the PUMA ICS condenser tube diameters as being 1/2 the prototype diameter. Hence, the PUM A ICS tubes are chosen to be 25.4 mm (l inch).
As noted above, the reduction of the SBWR ICS tube condenser diameter from 50.8 mm (2 inch) to 25.4 mm (1 inch) in the PUMA ICS is primarily based on the need to maintain volumetric scaling in the ICS at 1/4(X). Since the ICS tubes are initially filled with water, any distortion in the volume scaling will affect the inventory of the system as the water drains into the RPV during the isolation stage. The reduction in condenser tube diameter is not expected to cause any appreciable distortion to phenomena occurring in PUMA. First, the role of the ICS in the SBWR safety system is appreciably reduced after the ADS system has been activated, which is the time scale with which the PUMA facility is primarily concerned. It can be shown from
7-32 Dukler's analysis of condensation in vertical tubes that a reduction in tube diameter from 50.8 mm (2 inch) to 25.4 (1 inch) will not appn:ciably affect the overall rate of condensation. The analysis of Dukler is outlined in [7.9]. Working from the definition of eddy viscosity and using the Deissler equation for its variation near a solid boundary, Dukler found that the velocity dis-tribution curve in the liquid film was dependent on the interfacial shear and film thickness.
Assuming that the ratio of the eddy thermal diffusivity to the eddy viscosity was unity and that the viscosity was unity and that the physical properties of fluid do not change in the direction of heat transfer, equivalent temperature profiles were constructed. Integrating the velocity and tem-perature profiles, the liquid film thickness and point heat transfer coefficients were computed.
The results are displayed in [7.9] as a function of the Reynolds and Prandtl numbers with the interfacial shear as a parameter.
For the present analysis, we assume that the interfacial shear is proportional to the steam /non-condensible gas velocity, which is,in tum, related to the flow area.
Compared to the general flow scaling of the condenser tubes, where ng=l/50 and the tube diameter remains prototypic, the adjusted scaling approach of the ICS condenser tubes (ng=l/25 and dg=l/2) results in a doubling of the steam /non-condensible gas velocity through the tubes for a fixed mass flow rate. Based on the above assumption, the interfacial shear can also be expected to increase by approximately a factor of 2. Using Dukler's analysis for the appropriate Prandtl number, an incremental increase of the interfacial shear of this magnitude results in only an incremental increase in the condensation rate. Hence, a reduction of the ICS condenser tube from the prototypic value of 50.8 mm (2 inch) to 25.4 (1 inch) is not expected to add any appre-ciable distortion to the condensation rate in the tubes.
Next, we consider scaling the PCCS and ICS headers. In each case, the SBWR headers are designed as horizontal cylinders. In scaling the headers for PUMA, we impose two sealing cri-teria, namely 1/4 height and 1/400 volume scaling.
Each SBWR PCCS header has a volume of 0.779 3m with height of 600 mm. Applying the height and volume sealing criteria to these dimensions and noting that in PUMA two module headers are combined into one, the PUMA PCCS header dimensions are given as 3.895 x 10-3 m3 in volume and 165 mm in height. Considering only vertical cylinders, this results in a cylinder diameter of approximately 174 mm.
Based on a cylinder of this diameter,it was found that it is impossible to accommodate all of ten PCCS condenser tubes, with 59.8 mm (2 inch) tube diameter each. After several iterations, still it was decided that in order to best accommodate the condenser tubes and maintain the header volume and height scaling, the PUMA PCCS headers would be designed by connecting two cylinders of different cliameters, as shown in Figure 7.19.
,y
7-33 The diameter of the cylinder was chosen in order to accommodate the condenser tubes.
Once a practical height was chosen for this portion of the header, the dimensions of the smaller cylinder were chosen in order to satisfy the volume and height scaling. Hence, even though the PUMA PCCS header is designed as two sections, the combined dimensions are chosen is such as way as to maintain proper volume and height sca' ling. As shown in Figure 7.19 the inlet and outlet PCCS headers are designed to be similar, as is the case in the SBWR.
The PUMA ICS headers are also scaled using the 1/4 height and 1/400 volume scaling, simi-lar to the PCCS headers. The dimensions of the ICS headers are given in [7.10]. Applying the height and volume scaling criteria to these dimensions and noting that in PUMA two module headers are combined into one, the PUMA ICS header dimensions are given as 4.978 x 10-3 m 3 in volume and 170 mm in height. Again, considering a vertical cylindrical geometry, this results in a cylinder diameter of approximately 193 mm.
Unlike the difficulty that was encountered with accomodating the PCCS condenser tubes with the scaled dimension of the cylinder base, the ten PUMA ICS condenser tubes can be readily positioned in a cylinder base of 193 mm in diameter. This is primarily due to the fact that the PUMA ICS tube diameters are half of those in the PCCS. This implies that a single cylindri-cal header can be designed for each PUMA ICS unit.
However, in order to maintain consistent geometries between the PUMA ICS and PCCS headers, the ICS headers are also designed as two combined cylinders with different diameters, as shown in Figure 7.19. The diameter of the larger cylinder is chosen to accomodate the con-denser tubes. Once a practical height is chosen for tnis portion of the header, the dimensions of the smaller cylindrical section are determined in order to satisfy the volume and height scaling using the combined dimensions of both sections. Again, the inlet and outlet header in the PUMA ICS are designed to be similar, as is the case in the SBWR. The PUMA PCCS/ICS con-denser design parameters are given Table 7.13.
Table 7.13 PUMA PCCS/ICS Condenser Design Parameters e No. of IC Condenser Units: 1 e No. of IC Condenser Tubes: 10 e IC Condenser Tube Diameter: 23.3 mm
.No. of PCCS Condenser Units: 1
. No. of PCCS Condenser Tubes: 10 e PCCS Condenser Tube Diameter: 47.5 mm
.IC/PCCS Condenser Tube Length: 450 mm
1 l
7-34 1
A simple design is adopted for both the PUMA PCCS and ICS headers in order to allow the efficient separation of non-condensibles from condensate in the outlet headers. As shown in Fig-ure 7.20, a single condensate drain pipe is positioned at the lowest elevation in each of the outlet headers. Condensate collecting on the bottom cylinder head will readily drain through the drain pipe. The diameter of the PUMA PCCS and ICS condensate lines is determined based on boun-dary flow scaling.
Also shown in Figure 7.20, the non-condensible gas is removed from the PCCS and ICS headers at a higher elevation in the lower headers than the condensate drain. Penetration of the non-condensible vent line into the larger cylindrical portion of the outlet header prevents any condensate collecting in thc bottom header from entering the vent line. Also,in order to prevent any condensate from flowing directly into the vent line from the condenser tubes above the vent opening, a hood-like device is positioned on top of the vent opening. The diameter of the PUMA PCCS and ICS vent lines is determined based on boundary flow scaling.
In the SBWR, each ICS/PCCS condenser is located in a subcompartment of the ICS pool.
The subcompartments are formed by venical baffles. All pool subcompartments communicate through openings beneath the baffles to the larger ICS reservoir, enabling full use of the water inventory independent of the operational status of any given ICS/PCCS loop.
Pool water can heat up to about 101 C, forming steam. This produces a slight positive pres-sure relative to ambient forcing the steam from the space above each ICS/PCCS condenser to be released to the atmosphere through discharge vents.
Applying the volume scaling criteria (1/400) to the SBWR ICS/PCCS inital total water inventory [7.11, Doc.25A5044, Sh.30] results in a relatively large volume, which from a cost and design point of view would be difficult to accomodate in PUMA, as the ICS/PCCS are located at the highest elevation in the facility. The imponant issue to consider in the scaling of the ICS/PCCS pool sizes in PUMA is maintaining 1/4 height scaling of the water, thereby main-taining the submerged level of the condenser units.
To maintain this scaling criteria, and avoid the difficulty of dealing with large water inven-tory, the width (or diameter) of the PUMA pools are chosen in such a way as to allow a practical and sufficient volume of water to be stored at a high elevation. To this effect, the ICS and PCCS pools in PUMA are separated into two smaller pools. The diameter of each pool is designed so that it is large enough to accommodate all three ICS condensers or all three PCCS condensers.
For simplicity, the dimension of the PUMA ICS and PCCS pools are kept similar. Each pool is partitioned into three pie-shaped sections (similar to the PUM A GDCS pool), with an opening at the bottom of the partitions to allow for communication.
7-35 The height of the pools is determined by the need to accomodate the 1/4 scale water height.
The water level is controlled by a system which can supply water from a storage tank located at ;
the ground level. The water is fed independently into the bottom of the pool. The dimension of the PUMA PCCS and ICS pools is given in Table 7.14. For simplicity, the dimension of the l PUMA ICS and PCCS pools are kept the same.
A schematic of the PUMA PCCS system is shown in Figure 7.21. The inlet for each PCCS condenser is connected to the upper drywell. In order to simulate the effect of the steam /non-condensate distribution in upper drywell on the PCCS condenser performance, three sets of steam feedlines are provided with manual valves. These steam feedlines can selectively supply to the PCCS condenser from the center, middle or wall locations of the upper drywell. The diameter of the PCCS steam supply lines is determined by boundary flow scaling.
A schematic of the PUMA ICS system is shown in Figure 7.22. The inlet for each ICS is connected to the RPV via the DPV lines. Since there are two DpV lines in PUMA, one line from the DPV is branched into two lines to allow for a total of three ICS supply lines. The diameter of the ICS steam supply lines is determined by boundary flow scaling. A summary of the PUMA ICS and PCCS line sizes are given in Table 7.14.
Table 7.14 PUMA PCCS/ICS Pool Design Parameters e Number of Pools: 2 (3 condensers each) e Tank Height: 1450 mm e Tank Diameter: 1225 mm
. Connecting Lines:
Line Function Line From Line To Pipe ID (mm/ inch)
PCCS Supply Drywell PCCS Tank 38.1/l.5 PCCS Non-Cr :d Vent PCCS Tank SP 38.1/l.5 PCCS Con &nsate Drain PCCS Tank GDCS Tank 25.4/1 ICS Supply DPV Line ICS Tank 38.1/l.5 ICS Condensate Drain ICS Tank RPV 25.4/l ICS Non-Cond Vent ICS Tank SP 12.7/0.5 7.6 Automatic Depressurization System (ADS) Design The ADS is part of the SBWR emergency core cooling system (ECCS). Its function is to depressurize the reactor so that the gravity-driven GDCS water can be injected into the reactor. ;
7-36 The ADS consists of eight safety relief valves (SRVs) and six depressurization valves (DPVs) and their associated instrumentation and control.
The SRVs are mounted vertically on top of the main steam lines (MSLs) in the drywell.
Each SRV discharges steam through a discharge line to a point below the minimum water level in the suppression pool. Four DPVs are horizontally mounted on horizontal stub tubes con-nected to tne RPV at about the same elevation as the MSLs. The two remaining DPVs are hor-izontally mounted on lines branching from each MSL. The SRVs are spring-loaded valves and can be operated in safety (steam pressure) mode and relief (power-actuated) mode. Each SRV is equipped with a pneumatic accumulator and a check valve for safety mode and manual opening I
functions. The DPVs are straight-through, squib-actuated, non reclosing valves with a metal diaphragm seals. The DPV is closed with a cap covering the inlet chamber. The cap will readily shear off when pushed by a valve plunger actuated by an explosive initiator-booster. The DPV size provides about twice the depressurization capacity of an SRV.
The ADS automatically actuates on a water level (Level 1) signal that persists for at least 10 seconds. For the PUMA facility, Level 1 is defined as WK; mm above the Top Active Fuel (TAF). The SRVs and DPVs are actuated in groups of two or four valves at staggered times as the reactor undergoes relatively slow depressurization. In Table 7.15, the SRV and DPV capa-city and their opening sequence are given.
Table 7.15 SRV and DPV Capacities and Sequence of Action for SBWR 6
Total Max. DPV Flow Capacity at Vessel Pressure 6.35 x 10 kg/h.
7.481 MPa Total Max. SRV Flow Capacity at Vessel 4.084 x 106 kg/h.
Initiating Signal (Level 1) 36(X) mm (above TAF)
Max. Allowable Time Delay to 10 s Confirm Level 1 Signal Valve Actuation Sequence after Level 1 Signal Confirmed:
4 SRVs 0.0 s 4 SRVs 10 s 2 DPVs 55 s 2 DPVs 100 s 2 DPVs 145 s
l l
1 7-37 For the PUMA facility, the four SRVs on each MSL are replaced by one SRV line. Emm boundary flow scaling, the size of the SRV line is scaled by 1/10. Thus four 203 mm (8 inch) l SRV lines are replaced by a 50.8 inm (2 inch) SRV line. For the discharge nozzle, the nozzle size is scaled by 1/(10xE), according to break flow scaling criteria. In PUMA, the four DPVs on the RPV are replaced by two DPVs, and each DPV on a MSL is replaced by one DPV each.
In Table 7.16, the sizes of DPVs and SRYs in the PUMA facility are compared to the prototype SBWR system.
The plan view of the ADS is shown in Figure 7.24. On one of the MSLs, the DPV is treated
.as a MSL break for MSL break simulation. The front view of the ADS is shown in Figure 7.25.
This figure indicates the DPV and SRV lines leading to the upper drywell and suppression pool, c
Full-port ball valves with electrical actuators are used for the opening and closing of the SRVs
- and DPVs. These ball valves have response times of 2 to 5 seconds. Each of the SRVs and DPVs are instrumented with flow, temperature and pressure measurement devices. A typical ADS line is shown in Figure 7.26. The ball valve is followed by pressure and temperature sen-sors, then a capacitance meter and a magnetic flow meter. At the end of the line a flow nozzle of appropriate size is used for measuring flow with DP transducers. The opening and closing of the ball valve is controlled by a logic circuit which is monitored from the computer control network.
Table 7.16 Automatic Depressurization System (ADS) for PUMA Facility Component SBWR (mm/ inch) PUMA (mm/ inch) Area Scale SRV line 203/8 (4 each) 50.8/2 (1 each) 1/100 SRV Nozzle 203/8 (4 each) 29/1.13 (1 each) 1/200 DPV (MSL)line 305/12 76.2/3 1/100 DPV (MSL) Nozzle 305/12 21.6/0.85 1/200 DPV (RPV)line 457/18 (4 each) 76.2/3 (2 each) 1/100 DPV (RPV) Nozzle 457/18 (4 each) 30.5/1.2 (2 each) 1/200 t
7,7 Feed Water Line (FWL) and Auxiliary Reactor Coolant System Main components of the Feed Water Line (FWL) and auxiliary reactor cooling system are shown in Table 7.17.
7-38 Table 7.17 PUMA Feed Water Line: 2 Each Auxiliary Systems e Components: 1) CRD, 2) RWCU/SDCS e Connecting Lines Line Function Line From Line To Pipe I.D. (mm/ inch)
RWCU/SDC Line RPV Bottom Aux. Tank 12.7/0.5 RWCU/SDC Line Aux. Tank RPV Side 25.4/1 CRD Line Aux. Tank RPV Bottom 25.4/1 RWCU/SDC & CRD Break RPV Bottom Drywell 50.8/2 7.7.1 Feed Water Line (FWL)
The SBWR FWL is designed to supply water to the RPV over the full range of reactor power operation. The two FWLs are vented from the turbine building and carry condensate the feed water lines (FWLs) consist of two 355.6 mm (14 inch) diameter lines connected to RPV nozzles.
Each line branches into two lines which then connect to the RPV at an elevation near the top of the chimney section. The use of two lines minimizes the number of containment penetrations while providing two separate flow paths. Each of the four penetrations into the RPV connect to spargers located on the inside wall of the vessel. This design allows for proper feed water flow distribution in the downcomer region. The Control Rod Drive (CRD) system provides makeup water via the Reactor Water Clean-Up/ Shut-Down Cooling (RWCU/SDC) system piping to the core any time the feed water flow is not available. The CRD and RWCU/SDC system are described in the following sections.
In the PUMA facility, the FWL is scaled by boundary flow scaling and two lines are com-bined into one. The PUMA FWL then branches into two lines symmetrically penetrate the RPV, as shown in Figure 7.27. Each FWL penetration then connects to a sparger as shown in Figure 7.28. The sparger flow area is scaled by the total sparger flow area in the SBWR. The choked area is sealed from the FWL nozzle area.
7.7.2 Control Rod Drive (CRD) System The CRD system is designed to operate only when on site AC power is available. As such,it is considered a non-safety system. The system is composed of three major components:
Electro-hydraulic Fine Motion Control Rod Drive (FMCRD) mechanisms, Hydraulic Control
7-39 Units (HCU), and the Control Rod Drive Hydraulic Subsystem (CRDHS). The FMCRDs pro-vide electric-motor-driven positioning for normal insertion and withdrawal of the control rods and hydraulic powered rapid insertion (scram) of control rods during abnormal operation condi-tions. The hydraulic power required for scram is provided by high pressure water stored in the HCUs. Each HCU contains a nitrogen-water accumulator charged to high pressure and the necessary valves and components necessary to scram two FMCRDs. The CRDHS supplies dem-incralized water to provide charging of the HCU scram accumulators and purge water flow to the FMCRDs during normal operation. The CRDGS is the source of pressurized water for purging the RWCU/SDC system pumps. The primary task of the CRDHS is to provide high pressure make-up water to the reactor during events, such as LOCA, in which the feed water syster.. is unable to maintain reactor water level. This makeup water is supplied to the reactor via a bypass line off the CRD pump discharge header which connects to the feed water inlet piping via the RWCU/SDC return piping.
In the PUMA facility, the following function of the CRD system is important, as it occurs during a reactor LOCA event: The CRD system supplies high pressure makeup water to the reactor when the normal makeup supply system (feed water) is unable to prevent reactor water level from falling below reactor water Level 2. In the PUMA facility the CRD system is simu-lated for the above function, as shown in Figure 7.28. The water is taken from the auxiliary tank and pumped to the CRD line. Part of it is bypassed to the feed water line through RWCU/SD system return lines. Here a break on the CRD line is also shown. The pipe sizes of the CRD lines are scaled by the boundary flow scaling method. The CRD lines are only on the "B" feed water line.
7.7.3 Reactor Water Clean-Up/ Shut-Down Cooling (RWCU/SDC) Systems The RWCU/SDC system is designed to operate only when on-site AC power is available.
Hence, similar to the CRD system, it is also considered as a non-safety system. The RWCU/SDC system performs two basic functions: reactor cleanup and shutdown cooling fune-tions. The important functions of RWCU/SDC that apply to PUMA facility are the control of reactor water level during shutdown and shutdown checking. The RWCU/SDC system is com-posed of two independent pump-and-purification equipment systems. During nonnal plant operation, the system continuously recirculates water taken from the mid-RPV level and from the reactor bottom and returns it via the feed water line to the RPV.
In the case of the loss of preferred off-site AC power, the RWCU/SDC system brings the plant to cold shutdown in 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> in conjunction with the ICS Following a transient, the j RWCU/SDC system has the capacity of removing the core decay heat, plus compensate the CRD purge flow, one-half hour following scram. One of the possible LOCA events is the break i
1
7-40 of the RWCU/SDC line near reactos bottom. This break has the potential of draining reactor coolant coming from both RWCU/SDC line.s, one near mid section of reactor and the other at the bottom, through the break. In Figure 7.28, the RWCU/SDC system for PUMA facility is shown with the break location. ,
7.8 Stored Heat, Heat Loss and Insulation Design 7.8.1 Stored heat in RPV wall The relatively thick wall (154 mm) of the reactor pressure vessel (RPV) has a very large heat capacity ar.d will act as an imponant source of stored energy during the reactor blowdown pro-cess. In addition to the core decay heat, stored enugy from the vessel wall will be released into the reactor, adding to the fluid enthalpy. This additional source of heat needs to be properly scaled in order to balance the total energy of the system.
The stored energy in the prototype vessel wall cannot be readily physically scaled in PUMA, due to its relatively thin walls. An alternative would be to install electric heaters on the outer perimeter of the PUMA RPV in order to match the scaled stored energy. This procedure is also difficult to implement, since the electric heaters need to be distributed in such a way as to pro-vide uniform heat to the wall in order to avoid localized heating. In addition, adequate insulation needs to be added to prevent heat loss from the electric heaters to the surrounding ambient environment.
The approach taken here is to quantify the amount of stored energy in the SBWR vessel wall, and then scale this energy to the PUMA model. The scaled energy can then be compared to the core decay energy in order to gauge its magnitude to the overall energy. This would determine how the additional energy can be added to the system through the core heaters. The stored heat from the vessel wall can also be calculated as a function of time to observe the importance of the heat release as it decreases with time.
The one-dimensional, transient heat conduction equation for a wall can be written as 1
3T - I BT 2
(7.40) 8x2 a at I
The initial and boundary conditions for the RPV wall are l 1
7-41 T(x,0) = Ti T(0 t) = Ts = f(t)
BT
-8xl x=t. - 0 (74l)
The information needed to evaluate f(t)is obtained from SBWR vessel blowdown curves, where TRACG has been used to predict vessel pressure as a function of time. The saturation tempera-ture is then correlated in terms of pn:ssure in order to obtain the inside vessel wall temperature as a function of time.
Solving Eq. (7.40) for a constant surface temperature, using the initial and boundary condi-tions listed above, the approximate solution given a semi-infinite slab assumption Tx - Ti x
= 1 - erf (7.42)
Tx - Ti 2Mt In order to calculate T i and obtain the wall temperature profile, the following iteration scheme was implemented:
T i li=j i = T i(x = L)li=) (7.43)
Figure 7.29 shows the temperature profile for several time steps beginning from 600 seconds, which is the time at which the RPV is at 1.034 MPa (150 psi) following a loss of feedwater break. The calculations show that within one hour of the initial blowdown, a nearly uniform tem-perature profile exists across the RPV wall. Hence, shortly after one hour from the initial blow-down, the RPV wall no longer acts a m. in.portant heat source and no additional measures need to be taken into account in PUMA for scahag purposes.
The total amount of heat released to the inside of the vessel from the RPV wall can then be estimated by 4 = pep VATm (7.44)
7-42 where the mean wall temperature difference, ATm, from 6(X) seconds to 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> after blowdown, can be estimated from Figure 7.29 to be approximately 84 C. The total amount of heat released to the inside of the vessel within this time period is then calculated to be approximsely 0.6948 MJ. Scaling this value by the power scaling criteria of 1/2(X), results in the PUMA heat loss equivalent of 3.474 kJ. This amount of additional heat needed to properly scale the RPV wall heat loss can be readily compensated by increasing the PUMA core electrical heaters. Hence, there is no need to install additional electrical heaters on the outer surface of the PUM A RPV to simulate the heat loss.
Finally, Figure 7.30 illustrates the heat transfer calculated for the SBWR RPV wall as a func-tion of time, following blowdown and using the calculation procedums described above. The curve shows an initial sudden rise in heat transfer, which is due to the fact that after the initial blowdown there is a sudden depressurization taking place in the vessel, resulting in lower satura-tion temperatures inside the vessel. This creates a steep temperature gradient between the inside of the vessel and the vessel wall, which still maintains a large of amount of stored heat. The rate of heat transfer reaches a peak at about 8(X) seconds and then aegins to decrease as the thermal penetration depth extends further into the vessel wall, thus reducing the thermal gradient.
The heat transfer profile of the vessel wall can also be simulated in PUMA when the power to the electrical heaters is increased to compensate for the vessel wall. This ma., he accom-plished by inputting a time function to the designated power controllers to follow a similar profile as that in the prototype.
7,8.2 Containment heat sink design Due to the massive concrete wall structure of the SBWR containment, the concrete has the potential to serve as an important heat sink during the reactor blowdown process. As steam comes into contact with the containment wall, condensation will occur due to the cold surface of the wall. This rate of heat removal needs to be properly scaled in order to match the boundary flow of energy.
l To properly scale the containment heat sink in PUMA, it is necessary to evaluate the tem-perature penetration and the thermal inertia of the concrete wall structure to obtain appropriate scaling parameters.
Treating the wall as a semi infinite slab, the one-dimensional transient heat conduction equa-l tion is given by Eq. (7.40). By using the thermal diffusivity of the containment wall ots and the following initial and boundary conditions l
7-43 T(x, o) = T; T(o, t) = To T(*, t) = T i the solution of the temperature profile for the above equation can be expressed in terms of the heat flux at the surface (x=0):
k, (To - Ti > ,,
q,, x__o = q x=o = k, BT x=o (7.45)
Qna,t 0*
Equation (7.45) shows that the thermal penetration depth, dih , can be given by dm = ina,t (7.46)
Proper scaling of the penetration depth would result in the correct temperature profile in the con-tainment wall structure. Since the time scaling in PUMA is half that of the SBWR, the thermal penetration ratio can be written as (cz,)g (dm)g = (7.47)
In order to obtain an appropriate scaling parameter for the thermal inertia of the wall strue-ture, an energy balance needs to be considered. The lumped parameter energy equation for the wall structure is dT a, p, cps - = (4 q") - (( q"),,, (7.48)
By non-dimensionalizing the above equation and performing several manipulations, an appropn-ate scaling parameter, the thermal inertia ratio, Nth, can be defined as
7-44 a, p, cp, Nm = (7.49) af pr Cp r , g In order to properly scale the heat sink, as well as the heat transfer, the value of Nm should be unity. By defining the effective solid cross-sectional area, a,, as the product of the inner cir-cumference and the penetration depth, one obtains a, = x d dth (7.50)
Substituting Eqs. (7.46) and (7.50) into Eq. (7.49) and noting that (a r)R I/IIX). dR = I/IO.and similar fluid properties, the thermal inertia ratio reduces to N c es)R (7.51) th = E (Is Ps Substituting the properties of conen:te into Eq. (7.51), the PUMA requirement necessary to satisfy proper scaling of the heat sink is obtained:
(k sp, cps)m = 20900 (7.52)
This product has units of J2 /s-md -K2 It is noted that this requirement only depends on the solid material properties. Based on commercially available material, with sufficient flexibility to be installed around cylindrical vessels it was found that fibrous-based acoustic tiles, 3
(p = 290 kg/m ,k = 0.058 w/m-K,c p= 1340 J/kg-K) which are commonly used in office build-ings would best match the above criteria:
(k, p, e ps
)m,u c m,. = 23(XX) (7.53)
Hence, the concrete surrounding the containment in the prototype can be scaled very closely by using the fibrous-based acoustic tile as a source of heat sink in PUM A.
In order to evaluate the necessary thickness of the PUMA heat sink, the penetration depth in the SBWR needs to be estimated. Considering a time of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> after the initial blowdown, Eq.
l
(
l l
7-45 !
l (7.46) can be used to calculate a penetration depth of 23.6 mm in the SBWR. Scaling the depth by a length scale of 1/10, the PUMA heat sink thickness is estimated to be 23.6 mm.
7.8.3 Design ofinsulation for heat loss It is important to estimate the rate of heat loss from the PUMA components in order to quan-tify the amount of heat that needs to be added to the system to compensate for the losses. The estimated rate of heat loss in the system can be compensated for by increasing the power input to the system and preventing heat loss from distorting the thermodynamic equilibrium in the sys-tem.
The rate of heat loss from each of the components can be estimated by a simple one-dimensional heat conduction equation:
q"=UAT (7.54) where AT is the overall temperature difference. Since all major components in PUMA have a cylindrical geometry, the heat conduction equation can be expressed in cylindrical coordinates.
The overall heat transfer coefficient, U, expressed in terms of the inside surface area for a cylinder with one layer ofinsulation and of unit length (Figure 7.31)is given by
_i ri r ri r ri 1 U= i In 2 +-In-,+ (7.55) ka ri kg r2 r3 h o .
The outside heat transfer coefficient, ho, can be determined from empirical free convection correlations. For vertical cylinders of height, L, satisfying the criterion given by, d '15
-2 (7.56)
/ Grl" the following correlation [7.12] is recommended to calculate the average heat transfer coefficient for both constant heat flux and constant surface temperature conditions over the entire range of Rayleigh number, Raf, under both laminar and turbulent coriditions:
i
. ~_
i 7-46 l l
[ '2 0.387 Ralt6 h=I, 0.825 + w27 -
(7 57)
/ + (0.492/Pr)9/16 where Gri and Raf are defined as, 2
Gri = g ATL 2 (7.58) v 3
Rai = g ATL (7.59) va In order to calculate the heat transfer coefficient from pipes, the following correlation [7.13]
recommended for long horizontal cylinders over a wide Rayleigh number range can be used:
-2 0.387 Raj'6 h=1, 0.60 + .
w27 -
(7.60)
/ 1 + (0.559/Pr)9'l6 In order to calculate the Rayleigh number, Ru. in the above equations, a wall temperature (T,)
of 35 C and ambient temperature (T ) of 24 C are assumed as being typical of normal operating 9
conditions. For the major components, the calculated value of Rn ranges from 10 to 10", with an average fme convection heat transfer coefficient calculated to be approximately 4 W/m2K.
For pipe diameters ranging from 25.4 to 76.2 mm, which are typical of the pipe sizes in PUMA, Rayleigh number calculations range from 10' to 10', with an average free convection coefficient calculated to be approximately 5 W/m2 g, Table 7.18 lists the estimated heat loss calculations for the PUMA RPV and the suppression pool, including three different pipe diameter sizes. Various insulation thickness are shown in order to choose an appropriate thickness for the insulation material. Each table specifies an assumed inside wall temperature for the specified component, which is based on the maximum temperature expected during the initial stage of each experiment. As this temperature decreases with time, the heat loss will also decrease. Calculations shown in these tables are maximum values and will decrease as the blowdown process continues. Considering 76.2 mm (3 inch)
i i
7-47 i l
l insulation for the components listed, the overall heat loss is seen to be relatively small.
Similar calculations are also performed to estimate the heat loss from the PUMA contain-ment. However, since a large portion of the containment surface area is covered by heat sink material (see Section 7.8.2 concerning scaling heat sink material), this additional layer will also act as insulation and needs to be accounted for in heat loss calculations. Table 7.19 lists the estimated heat loss calculations for the PUMA containment with 31.75 mm (1.25 inch) thick heat sink material (fibrous acoustic tile) for various insulation thickness. Again, considering 76.2 mm (3 inch) insulation, the overall heat loss is seen to be relatively small.
The sum of all heat loss from the PUMA components and piping is estimated to be unA:r 5 kW over the duration of the experiments. This amount of heat can be readily compensated by increasing the core electrical power.
7.9 Break Design The double-ended pipe breaks on lines such as the GDCS drain. GDCS equalization, ICS condensate drain, RPV bottom drain, RWCU/SDC, and FWL are simulated in PUMA, as shown in Figure 7.31. The break sizes for the prototype are listed in Table 4.8. As seen in this table, the largest break size in this category is that of the FWL break (area 390 cm2 ). In PUMA, the break area is scaled by 1/2(X). The equivalent nozzle cross sectional area for the FWL break in PUM A 2
is 1.95 cm . The break receiver tee shown in Figure 7.31 is 50.8 mm (2 inch) in diameter (cross 2
sectional area 20.26 cm ). Thus, the flow area of the receiving pipe is about 10 times larger than the FWL break area. For choked flow through the break, the downstream conditions do not effect the flow through the break. For non-choked flow, the receiving tee is large enough such that there is no significant pressure loss in the drain side of this tee section. Also, as the receiving side of the tee has a large flow area compared to the break, the flow from the break and the top side of the tee are not affected, since there is no pressure build-up in the Lee section. This break simulation requires a single instrument set to measure the total break flow coming out of the double-ended pipe break.
1 i
7-48 1
l Table 7.18 PUMA component heat loss estimation RPV insulation heat heat thickness (mm/ inch) flux (W/m2) loss (W) 0 500.20 6219.02 12.7/0.5 230.98 2871.79 25.4/1 152.21 1892.38 50.8/2 92.52 1150.29 76.2/3 67.75 842.35 101.6/4 54.17 673.49 127/5 45.58 566.68 SUPPRESSION POOL insulation heat heat thickness (mm/ inch) flux (W/m2) loss (W) 0 323.35 12337.83 12.7/0.5 143.62 5480.08 25.4/1 92.60 3533.29 50.8/2 54.43 2076.77 76.2/3 38.73 1477.86 101.6/4 30.17 1151.35 127/5 24.79 945.81 GDCS insulation heat heat thickness (mm/ inch) flux (W/m2) loss (W) 0 527.16 9603.88 12.7/0.5 173.32 3157.54 25.4/1 104.37 1901.46 50.8/2 58.73 1069.93 76.2/3 41.24 751.26 101.6/4 31.99 582.75 127/5 26.26 478.47 i
1
7-49 l
1 Table 7.18 continued...
3 INCH PIPELINE insulation heat heat thickness imm/ inch) flux (W/m2) loss (W/m)
C 710.98 174.06 l 12.7/0.5 305.45 74.78 25.4/1 207.66 50.84 50.8/2 137.61 33.69 76213 108.94 26.67 101.6/4 93.06 22.78 127/5 82.83 20.28 2 INCH PIPELINE insulation heat heat thickness (mm/ inch) flux (W/m2) loss (W/m) 0 716.36 118.16 12.7/0.5 360.42 59.45 25.4/1 258.16 42.58 50.8/2 179.99 29.69 76.2/3 146.56 24.17 101.6/4 127.57 21.04 127/5 115.13 18.99 1 INCH PIPELINE insulation heat heat thickness (mm/ inch) flux (W/m2) loss (W/m) 0 976.87 81.77 12.7/0.5 488.88 40.92 25.4/1 357.22 29.90 50.8/2 258.51 21.64 76.2/3 216.18 18.10 101.6/4 191.88 16.06 127/5 175.76 14.71
7-50 Table 7.19 Containment heat loss in PUMA with addition of fibrous-based acoustic tile to simu-late concrete heat sink l
TOP OF UPPER DRYWELL insulation heat heat l
thickness imm/ inch) flux (W/m2) loss (W) 0 117.54 3015.73 0.5 81.14 2081.75 1 62.08 1592.80 ,
2 42.42 1088.25 j 3 32.34 829.87 i 4 26.22 672.81 5 22.11 567.23 I
ANNULAR SECTION insulation heat heat thickness (mm/ inch) flux (W/m2) loss (W) 0 118.88 1385.40 12.7/0.5 91.44 1065.81 25.4/1 74.65 869.09 50.8/2 55.15 642.66 76.2/3 44.15 514.52 101.6/4 37.09 432.21 127/5 32.16 374.81 1
BOTTOM OF LOWER DRYWELL insulation heat heat thickness (mm/ inch) flux (W/m2) loss (W) 0 115.67 880.74 12.7/0.5 88.39 672.97 25.4/1 71.72 546.10 50.8/2 52.41 399.01 76.2/3 41.54 316.29 101.6/4 34.57 263.24 127/5 _29.72 226.33
,_ -. . ~ . - . . .- - . . . . . , _ - . - - - - - _ . - - _ - - - .. . . .
7-51 .
544.7 ALL DIMENSIONS IN mm 740.8 DRYER SKIRT nn nnnnn ~
UPPER STEAM
, , , SEPARATOR PLATE 576.3 SEPARATOR TUBES STEAM SEPARATOR SUPPORT 265.7
!,7 SKIRT (PERFORATED)
I^
LOWER STEAM SEPARATOR PLATE 623.7 =
CHIMNEY MIXING PLENUM
=
CHIMNEY PARTITION 1641.8 CHIMNEY SHROUD ,
3 e ELECTRIC HEATER
' =
(ACTIVE AREA) t :
ess.o ; i CORE INLET PLATE
!~ CORE SHROUD 343.5 l l l
'= ELECTRIC HEATER (INACTIVE AREA)
I !
- 1. PERFORATED PORTION I : l l l OF CORE SHROUD 536.5 g ; , ; g i !
I
. s ,
l Figure 7.1 Reactor pressure vessel (RPV) and internal components i'
l
'~
.7-52 ,
?
i I
11 11 f 11 tl if I_' I! #
7 i g '
f- FLANGE 50.8 mm FLANGE -- E -- - - - - - --1= SPECIAL ,.
(CAMERA PORT) INSTRUMENT PORT 4
- SPECML :
INSTRUMENT PORT j 50.8 mm FLANGE -P" ----------
SPECIAL (CAMERA PORT)
~
] INSTRUMENT PORT :
50.8 mm FLANGE E ---------- SPECIAL l (CAMERA PORT) - INSTRUMENT PORT o 9 O l Y e
, f a ! -
I i e
i c
al- ma j e t
t ,
I i
- - 1 1
' ' I 50.8 mm FLANGE J'~.'
i (CAMERA PORT) l l 7; M i . . . i rs/ArirArrrrirA ,
f Figure 7.2 Flanges, vierv ports and special instrument parts on PUMA RPV 4
r- -, . , . y
7-53 DOUBLE SUP JOINT CONNECTING CORE $HROUD
{ ,47.3 mm (INSIDE CAP)
TO CHlWNEY SHROUD l-CHlWNEY SHROUD TOP Or CORE HEATERS 5R rl b W OE
~L ____ ,_
r t / : 2
/ CORE SHROUD l [
(t E
l d
l%j,g .
POSTTIONER (3 AZIMUTHAL r: LOCATIONS)
' I NG
$0EB0 Ou 0F RPV UME
/
h- 0-RING (9 3 T CKNESS) BOLTED e
/ ANGLE _o is- 37.6 mm SUPPORT (INSIDE CAP)
/
' ' RPV WALL 300 mm
/ Y 8 -
CORE SHROUD y HEATER (3.2 mm THICKNESS)
/
CORE
/ ~
a DOWNCOMER
/ , FLOW AREA j
i --Q
/ 536.5 mm FRou
/ a ,
80TTOu 0F RPV t
/
l a I, /
l PERrORATED PORT)DN Or CORE SHROUD l
- ,L /
i /
I -
I
/
"~
T k%
/
/ LOWER PLENUM l VOLUME FILLER i /
/ 290 mm INSIDE 80TTOW CORE SHROUD BOLTED Or RPy TO 80s rwcE i 1 ,
u I///A W////////A = ~"
Figure 7.3 PUMA core shroud and mounting design
RPV WALL DOWNCOMER FLOW AREA O O o -
() -
O CORE SHROUD O O O CENTER BreASS RING 54 mm DIAM., 853 mm HEIGHT O O oo O O
\
FuEt CootiNo WATER CHANNEL, 32 mm 1. D.
/-
Q o O O O O O. o Q -
- 37.5 m m ci" SIDE c^e)
QQ 38 HEATER RODS, 25.4 mm DIAM.
OUTER BYPASS RING -
O 0 0 O o 380 mm Di^u. 853 mm HEicHT INNER BYPASS RING O 340 mm LIAM., 853 mm HEIGHT O O (' CENTER BYPASS ORIFICE HotE. lo m m i.o.
12 OUTER BYPAS' ORIFICE l HOLES, 10 mm 1.D.
POSITIONED 179 mm FROM
! CORE VOLUME FILLER CENTER OF RPV l
l l
Figure 7.4 PUMA core inlet plate design showing heater and hypass llow area
i l
j 7-55 l
l l
. i .g ........g ........y . . . . ....g ........g . . . . . . . .
1.00 -. .
e DECAY HEAT [Ref. 7.2] :
- POLYNOMIAL FIT
Q.
s Q_
0.10 -. .
' ' ""l 0.01 ""l ' ' ' " >l ' ' ' " "l " "l 0.1 1.0 10.0 100.0 1000.0 10000.0 time (sec) !
l Figure 7.5 Decay heat curve fitting for SBWR l
l l
4 7-56 ,
E E E E
n E E
E ,
Eg e .g d
an O -
E s
O a 'b b i
? Po g\ 8 A 6 w ar -
- , - - .- - - a
+
' 7-57 t733.5 mm FROW ,
RPV WSIDE 80TTOW --*
a 15213 mm FROW RW NSIDE BOTTOW ~
i l'
CROSS-SECTIONAL VIEW OF HEATER AND COOLING WATER CHANNEL ACTNE AREA FOR ELECTRICAL HEATER '
HEATERS '
CHANNEL WALL (32 mm LD.)
HEATER POSITIONER (3 PER CHANNEL)
WATER CHANNf.L f HEATER POSmONER 837 O ** NN -
RW Bom CORE WLET PLATE (0.S3 mm THICKNES5)
(800 mm FROM RPV SOTTOW) r-MATER CHANNEt watt ruCL C00Lmc !
(40 mm HEIGHI ABOVE WATER CPMNNEL CORE INLif PLATE) ')
.j
- - 25.4 mm i
l 1
l 1
RPV INS 4DE BOTTOW-Figum 7.7 Typical heater design for RPV
7-58 I.
1 l N -
- . RPV . WALL l, '
STEAM SEPARATOR ,_ _ l l SUPPORT RING _
(PERFORATED) ,
l -
STEAM SEPARATOR - '
~
TUBE, 50.8 mm 1.D. -
g l
l.
[
l
' LOWER STEAM '
~
\' SEPARATOR PLATE l
I7 ul uuuu (6.35 l uu u mm THICKNESS)
U _
SPARGER
,,__t u ______T- - @.- d- . ---
FWL (3999 mm FROM
,' RPV INSIDE BOTTOM)
.; li j IN #
[ SUPPORT BOLT CHIMNEY
) SHROUD TO SHROUD Dl . SHROUD-0 '
POSITIONER RING - .
t (4 AZIMUTHAL LOCATIONS) 50.8 mm- POSITIONER RING --e= de-- 47.25 mm
!'= 300 mm O '
l'
=
l CHIMNEY
'- s DOWNCOMER l - FLOW AREA TOP OF CHIMNEY SHROUD CHIMNEY SHROUD (3999 mm FROM RPV H (3.2 mm T. lCKNESS)
INSIDE BOTTOM)
Figun: 7.8 Top of chimney shroud and lower separate plate in PUM A RPV 0
5
-v v. - - . _ . - - , .,
l '
7-59 s
' )
= 300 mm = LOWER STEAM i SEPARATOR PLATE
,. l
/
I I I I I I I I i l I i I iI I I i 1 /
I 1 _1 _
l TOP OF CHIMNEY
- SHROUD, 3999 mm
, FROM RPV BOTTOM
- l: :::
i I '
= -
CHIMNEY MIXING PLENUM l -
CHIMNEY
~
_ 00WNCOMER i , < FLOW AREA h ,
$= RPV WALL i
i d=-47.25 mm INSIDE GAP
' E ,
E
' ,i = ,
CHIMNEY SHROUD
/ *
~ '
DOUBLE SLIP JOINT I $ l CONNECTING CORE SHROUD CHIMNEY : -
~
l~
~
- - ~~ TO CHIMNEY SHROUD PARTITION -
l 10 mm-r ,
i e
/ FLANGE l
l h :-:-: ,
)
- ~::-
i
/ -:: l - ,
TOP OF HEATERS / EhEEE_ ! '
1733 mm FROM j : :--
BOT 10M OF RPV gEg g = CORE SHROUD
/,
i Figure 7.9 Chimney shroud and mounting design
CHIMNEY PARTITION 5 VERTICAL PIPES 165 mm DIAMETER
' CHIMNEY SHROUD
)
) 47.3 mm
/ (INSIDE GAP)
./
)M RPV WALL
+
EQUAL FLOW AREA l
l DOWNCOMER FLOW AREA 1
Figure 7.10 PUMA cross-sectional view of chimney partition a
STEAM SEPARATOR ANGLE SUPPORT.
o r-O ' N STEAM SEPARATOR SUPPORT .
RING (PERFORATED)
/ \
s
/
=
RPV WALL
[
I
\ f N LOWER STEAM SEPARATOR
\
\ -
PLATE .
\- .
> 50.8 mm
\ s
/ (SEPARATOR TUBE)
.\ , 1- / (TOTAL OF 16)
~ .-
)
400 mm DIAM. J L- 200 mm DIAM.
l Figure 7.11 Separator tube layout on lower separator plate L
i
. e- SKIMMER TUBE (25.4 mm DIAM.)
DRYER SKlRT
-- UPPER STEAM
' SEPARATOR PLATE s
43.3 mm
' =
RPV WALL S
i
- }
L , g =
ANGLE SUPPORT PLATE s UPPER STLAM SEPARATOR TUBE (50.8 mm DIAM.)
x 76.2 mm HEIGHT)
TOTAL OF 15 TUBES)
BOLT SKIRT TO PLATE q
DRYER SKIRT POSITIONER 400 mm DIAM.
240 mm DIAM- 120 mm DIAM.
Figure 7.12 Orifice holes on upper separator plate and support structure on RPV y b
7-63 4
I TOP OF DRTTR SKurf
$$81 mm FROW RW BOTTOW ORvtM p i M l.* samT P0srrowcR L
DmTR ,
SKIRT --
i l i i i
-== =- 413 mm g 300 mm (50.8 mm DmW. 76.2 mm FEIGHT)
,p l[ (TOTAL OF 15 TUBES) uPPtR sTtm
$EPWtATOR PLATE
- /'/ g l i EOLT SKIRT (6.35 mm THICKNES$)
l,7 TO PLATE
} j simTOR TWES (ma mm DuW.) ' '
(TOTAL OF 16 TUSCS) f PV 80 OW MS$",%
F - 0 0 4nsa == rROM RPV BOTTOW (12.7 mm on4W )
4640.5 *am FROW RPV BOTTOW i i SloWWCR TUBE RPV WALL l
(25 4 mm)
- DRYTR SKMT
, ANCLE SUPPORT PLATE
,[
(ANCHOR BOLT) l '
i BOTTOW Or DRYTR i F@O[ 4264 S mm FROW SKRT (4264.5 mm RPy 30TTOW FROW PV 90TTOW) l' i MW///
_' STEAW SEPARATOR SUPPORT ARATOR SU* PORT RING
(,L ANGLI SUPPOR1 PLATE l F. ,
(ANCHOR 80LT) J 9
LDwf R STEAW 4 SipaRATOR PLAft i o .,
p- O ; FWL (3999 mm FROW NSIDE RPV BOTTOM) )
!{
N. BOLT CHiWNfv SHROUD TO SHROUD POSITIONER RsNC SHROUD P05ffl0NER RINC CMWNtv becuD l
Figure 7.13 Mounting of separator secdon onto PUMA RPV, separator tube, and separator dryer shroud desigh 1
1 l
i r AN-HOLEu l
TMY f
' M
- n A UPPER DRYWELL DRYWELL TO SUPPRESSION POOL VENT 2750 mm r-
'"** 355.6 mm 1525 mm [ ,,, _
SPRAY 4300 mm U- '
L] urn-H0tt n A ,
/g V ANNutAR
--@ /
sECTION
/ \
2775 mm / O #
VACUUM BREAKER 2837.5 mm LEAK (6.35 mm) 3 ROWS OF vacuuu g 3 RADIAL SLOTS ER 7 BRNS U 175"mm HEIGHT
- ~850 mm (Sa8 mm) p 200 mm WIDTH a N ,, D' l
1162.5 mm ORtrtCE PLATE l- 2817 mm --- !
1525 mm * *
- SUPPRESSION POOL
<t LOWER 0 mm - DRYWELL - --
-300 mm -
926 4 mm y
M 1524 mm b a
Figure 7.14 Design of drywell vessel, suppmssion pool vessel, connecting vent line k. >
and vacuum breaker lines
_. -_-.____:_-________-m__u___ _ . -___ -___m-_ m . "- T -- ,,
7-65 DEFLECTED JET m.
J h
JET DEFLECTOR PLATE ([]~
(e - ~.,
~~~,
\ $
a l
l gs l
/ >
/
/ ,
E a
INCOMING JET , l m
Jo
=2m.
J
~_s q]
\l l
Y l
DEFLECTED JET m; Figure 7.15 Jet deflector plate
l 7-66 103 -
i i i , i i i , i i i
- N :
N PUMA (x/d=30) -
1 02 r \ --- SBWR (x/d=10) 5 E \ E
- N :
\ -
1 01 - Plume N _
x ~J' : N i x : N :
E , -
N -
, a Intermediate a a 10o r \ -
t s : x :
? : N :
f 10~' r N
\
Non-buoyant ;
\ :
- N :
g .
102 7 x _
i i i i i i i i i i i 10-3 104 102 10o 1 02 1 04 1 08 1 08 Froude No.
Figure 7.16 Jet flow regime transition as a function of Froude number 1
l
GDCS COVERGAS UNE FRov DRYWELL (50 8 mm) CONDENSATE FROM
-W ,PCCS (25 4 mm)
TOP V!EW OF GDCS TANK:
TO M A - THREE POOLS OF EOUAL 'LLUVE TOP GDCS m m
- POOL "& HAS NO PCCS ORNN UNE POOL ' TO POOL C
- ALL 3 POOLS COMMUNICATE AT TOP
- PARTITION]
WALL 4 g
h r---- -
POOL 8 r CDCS BOTTOU DRAIN I{
' OPENINGS *
@5 rrrn ? t D. 38.1 mm l 1525 mm 1287.5 mm t g l1 POOL C POOL A ,
1706.2 mm 4300 mm -
TANK FEED &
SUPPRESSION POOL 75 mm +---1540 mm t.D.-
UNES (HEIGHT OF SEAL = 348 25 mm) (38.1 mm)
\
2373 mm \ GDCS DRAIN UNE CHECK I. . _ _ _ .._ A f VALVES D / DETACHABLE
- ^
EOUAll2ATION UNES (12.7 mm) 1854 mm - ,jM s @ < 460 mm
\ GDCS EQUAUZATION g -
UNE CHECK VALVES c.a If EOUAUZATION UNE If BREAK (50.8 mm)
@ TO DRYWELL 0 mm - @ TO DRmELL GDCS DRAIN UNE BREAK (50.8 mm) .
. Figure 7.17 Gravity-driven cooling system (GDCS) design in PUMA y
w
l 385 mm RADIUS PCCS CONDENSATE 1 RETURN LINE VESSEL GDCS COVERGAS LINE B PENETRATIONS USING FROM DRYWELL (50.8 mm) CONDENSATE FROM FLANGE FITTINGS TO
-W PCCS (25.4 mm)
MAKE THE UNES REMOVABLE.
TO POOL A I PENETRATION INTO TOP GDCS " # "
GDCS POOL *B* ONLY t POOL ' TO POOL C
' (TWO LINES) #
' C TO CONTAIN A FLANGED PARTITION WALL
\ 9 e_ _ .
g jA PORT AND CAPPED OFF FOR POSSIBLE FUTURE h
_____I_____ USE.
I-J_
l 6% mm 1525 mm 1287.5 mm i 4 TOP VIEW I-~h I
7 106.2 5 mm o I y o TOP VIEW OF GDCS TANK:
FRONT VIEW
- THREE POOLS OF EOUAL VOLUME
- POOL "B" HAS NO PCCS DRAIN LINE
- ALL 3 POOLS COMMUNICATE AT TOP POOL B GDCS BOTTOM DRA!N OPENINGS
? 1.D. 38.1 mm W CT POOL C POOL A e-1540 mm 1.D.- ,
BOTTOM VIEW Figure 7-18 Design of GDCS pool j,
PCCS CONDENSER 900JLE 'CS CONDENSER MODULE pcts coNoENSER HEADER VOLUVE 3894 cm-3 HEADER VOLUVE 4977.8 cma3 uGOULE HEADER INSIDE HEIGHT 165 mm HEADER INS!DE HEIGHT 165 mm io ruaEs PER voout.t i
O OO O TUBE PARAMETERS:
7 OO - LENGTH 450 mm
- O.D. 50.8 mm
- i.D. 47.5 mm k [ B B OO SECTION A-A CONDENSER 10 TUBES PER uootAE 50 mm 75 mm Q Q 270 mm ^ !
' 270 mm f .
{ (
g Q Q . LENGTH 450 mm
- O.D. 25.4 mm 1.Q6 mm -
98 mm _ _
i O O -
'.D. 2 3.3 m m
~
O O 115 mm I 95 mm }
(NOTE: ALL D!MENSIONS ARE VEASURED INTERNALLY)
Figure 7.19 Design of ICS/PCCS condenser modules
?
PCCS CONDENSER uGDULE ICS CONDENSER VCDULE l
! ]
NON-CONDENSABLE VENT ttNE HOOD NON-CONDENSABLE VENT LINE HOOD
/ PCCS NON-CONDENSABLE /
/
ICS NON-CONDENSABLE VENT UNE (38.1 mm) 4 VENT LINE (12.7 mm)
TO PENETRATE TOP SECT 10N TO PENETRATE TOP SECTION OF LOWER HEADER BY 25.4 mm OF LOWER HEADER BY 25.4 mm PCCS CONDENSATE ICS CONDENSATE DRAIN LINE (25 4 mm) DRAIN LINE (25.4 mm)
O O '
o o BOTTOM VIEW BOTTOM VIEW
?
E Figure 7.20 Lower header penetrations of the ICS/PCCS Condenser modules in PUMA .
- _ - _ . . _ _ _ . _ _ _ - - _ _ _ _ _ - i_. _ __ _-_A _-__._ _ _ _ -
PCCS STEAv SUPPLY -
UNES FROM UPPER DRY WELL (38.1 mm) 1225 mm 8
N -
r--
=
VENT OPEN TO ATuGSPHERE '
2A1 PCCS STEAM SUPPLY 1450 mm LINE (38.1 mm) l y , , ,I NON-CONDENSA8LE
' y y y VENT LINE (38.1 mm)
I TANK FEED & I DRAIN UNE If If If 6325 mm 1r 17 37 rROM RPV BOTTOu CONDENSATE DRAIN UPPER LINE TO GDCS DRY WELL POOL (25.4 mm)
SUPPRESSION _
j_
POOL DETACHABLE c o- u N__J g
__g g 1175 mm
'\ NOTE FEED, DRAIN, & OVERFLOW WATER LINES CONNECTED TO PCCS POOL ARE NOT SHOWN IN DRAWING TOP VIEW OF CONTAINMENT STEAM FEED LINE TO PCCS y
Figum 7.21 Design of PUMA passive containment cooling system (PCCS) d
ICS STEAu VENT TO SUPPLY LINE (38.1 mm) 1225 mm __
r.-jl ATUCSPHERE T
b 3 w g 1450 mm ;
, , I, ,
NON-CONDENSABLE Il ll ji LJ L/ t/ VENT UNES (12.7 mm)
- c s
6325 mm !
LINE SIZE Q TANK FEED &
U 50.8 mm DRAIN UNE U b DPV LINE opV UNE h @
SUPPRESSION
@ POOL ICS CONDENSATE " N DRAIN L!NE RPV (25.4 mm) ___
3016 25 mm __._ L -. ~- -~ T N '
, N FI NCS IC CONDE SATE DRAIN LINE HEtGHT = 162.5 mm U
TO DRYWELL
! 1175 mm 212.5 mm tCS DRA!N LINE BRrJK (50.8 mm) L
/
DETACHABLE SPARGERS NOTE FEED. DRAIN. & ~ OVERFLOW WATER UNES CONNECTED TO ICS O mm - POOL ARE NOT SHOWN IN DRAWING l
l Figure 7.22 Design of PUMA isolation Condenser system (ICS)
T l TOP VIEW OF .lCS ' CONDENSERS TANK TOP view OF PCCS CONDENSERS TANK 0 0 N l
l
- N -
l -306.25 mm (RADrus) 306.25 mm (RADIUS)
- CONDENSER WODULES
- r CONDENSER WODULES
( CENTER UNES DSTANCE CENTER UNES DSTANCE
. g FROM TANK CENTER ,
h FROM TANK CENTER WATER 64 EATER WATER HEATER PCwER = 30 kw POWER = 30 kW (50.8 mm FRou 80TTOW Or TANK) (50.8 mm FRCu 90TTou or TANM)
P ON WALL 1225 mm PARTITCN WALL HEIGHT = 1162.5 mm j e 1225 mm =l HEIGHT = 1162.5 mm lc rl h OVERFLOW UNE OVEQFLOW UNE (50.Q mm) (5').8 mm) g I
IL 1 h 1[
[
l" "i l 3 i--.1 i---i 1450 mm 1450 mm 1100 mm 1100 mm -
Il OPEMNC AT BOTTOM OPENING AT BOTTOM
' PARTITION BETWEEN THREE 73.75 mm - PARTITION BETWEEN THREE 133.75 mm- POOLS (130 mm RADtOS)
POOLS (130 mm' RADIUS)
(130 mm HEICHT) (130 mm HEtCHT)
- ATER FEED & DRAIN WATER FEED & DRAIN UNE (25.4 mm) UNE (25.4 mm)
I FRONT- VIEW OF ICS CONDENSERS TANK FRONT VIEW OF PCCS CONDENSERS TANK w
- Figure 7.23 Design ICS/PCCS pools U Y
e
$ *n 4' y m-!= r r e.-e n. b w w .e
n
?
L O
O P
N
. O I
S S
E R
P P
U S
L
, S p M d
= n a
= S
\
\ / D
= / A f
=N o w
%_ / i e
v
'= - n a
~ lP 4
x 2
' 7 m
- I I u
g i
F
% = T p "'"' % N J
g " E E
" M
>g N 0
5 g
1 I A
o g T N
2 m
,6 m O 8
% C B) 0
- 5
) ,
8
( >,
V R
S
'*)
e
- s S
V P
R
o Cs SUPpty (A) 38.tmai c S stsRT (C) 1 381 awa -
@ Cs SLIPPLT (B) 381 -
UPPER DRY WELL Wr R,
)
ust (A) 7s r - _ _ _ _ _ ___
'~~~
mes s -
'1
' 8 opvA)(E 7s.: a-
~~~~ ::: ----- - se tar see -
SUPPRESSION POOL
, c,3 ,, , _
,s ust (B) 7e - [_
~
y 8 e ::: <,,r r.
8 y i o." -
r-. ./ tut,.s o
"#8 "
RPV sw unts (see awa) r jo ol jo LOWER DRYWELL pg
/
" /- .
ORinCE PLATE 'y OVENCHER h$g nr $$,"g[ L*E DETACWGLt 0 m m _._
-300 mm] o aesstt on== tut rtto l , vtssa rtto A DRAtN Lmt PIT T l
l l
r Figure 7.25 Automatic depressurization system (ADS) and main steam lines (MSL) schematic T'
a l
l I
i
l P,T DP
/ N g
m .
RPV NLL-PORT CAPACITANCE MAGNETIC FLOW WALL BALL VALVE WETER FLOW WETER NOZZLE W/ ELECTRIC ACTUATOR 4
Figure 7.26 Typicalinstrun ated ADS line w
?
t
- . _ _ - _ . - _ , - - . _ _ _ _ - - . - - - - - - _ _ . . - - . _ - . _ - - - _ _ - _ _ - _ _ . - - ~ _ _ - - - - _ - - _ - w - e e -
7-77 1
l RPV FEEDWATER LINE (A & B) 4114 mm 3939 mm (38.1 4 mm) @
n g g 38.1 mm II FEEDWATER LINE BREAK
@ N (50.8 mm) h AUXILIARY mT TANK IIOE N 1 CUBIC PUMP METER VOL.
8%
$5 i @
l I
4 l II l l l l
0 mm - g l CRD LINE 123 m l I l (25.4 mm)
L. ______,_,,,,_j
(_. _
L------g--- __. __ _ _ J RWCU/SDS CRD BREAK y if BREAK LINE (50.8 mm) , (50.8 mm)
Figure 7.27 Reactor water cir2n-up/ shutdown cooling (RWCU/SDC) system and control rod drive (CRD): , car design in PUMA .
TOP V!EW OF FEED WATER LINE SPARGER 12.7 mm 1.D.
25.4 mm f.D. ;
38.1 m m 1.D. F
\ /
M.
/ \ RADIAL INJECTION FEED WATER HOLES
/
,7 ,7 O'
- f
/
UNION ORIFICE PLATE 14.63 mm 1.0.
FRONT VIEW OF FEED WATER LINE SPARGER BOTH ENDS
/ ARE CAPPED u...................................nf l
35 HOLES OF 4.7625 mm (3/16")
Figure 7.28 Feed water line and sparger in PUMA m
Do
7 79 250 g 200 -
o a.
E .
S 150 -
-- - ~ - ~ - . ~ - . ~ = , . - . = . - . _ . _ .
100 - ' . , ' f O ,
0.05 0.1 0 15 mside RPV wall (m) wall 600 s --750 s -
1 hr -- 5 hr Figure 7.29 RPV wall temperature profile with ti:ne elapsed from blowdown i
7-80 1 03 . . .
....... . .' ......i . . . . . . . ..
1 02 .
A 10' - A A b
x i
. 5 v . .
o y) -
.9 U) d .
a
. 8 100 -
A -
10-' - :
i E
A i 10- ' ' ' " ' ' ' ' ' ' ' ' ' ' ' ' '
1 02 3os 1 04 1 05 time (s)
- Figure 7.30 Estimated RPV wall heat loss with blowdown for SBWR
7-81 k
8 +
++
k +++
A +++
++
+ +
+ +
r1 + + AIR
+ +
r2 ,+ + + h n
1 ; + +
+ + +
(W/m2K)
+ + T=24 C
+ + +
+ +
+ +
+
+v
++
STAINLESS STEEL +++
+++
++
+
INSULATION l
l I
Figure 7.31 Cross sectional view ofinsulation on typical vessel wall
i i
7-82 l
WATER DRAIN LINE INTO RPV (e.g. GDCS WATER DRAIN LINE)
UNION f, REDUCER RPV WALL I/
I =
TEE
\ T UNION BREAK LINE (50.8 mm, 2")
(t n (X) v l Figure 7.32 Typical break design in PUMA to simulate double-ended pipe break 1
i 1
7-83 Reference 7.1 Kocamustafaogullari, G., Ishii, M., " Scaling Criteria for Two-Phase Flow Natural and Forced Convection Loop and Their Application to Conceotual 2x4 Simulation Loop Design", ANL-83-61, NUREC/CR-3420 (1983).
7.2 G.E. Nuclear Energy, MFN. No.170-93 October (1993).
7.3 G.E. Nuclear Energy, "SBWR Standard Safety Analysis Report", 25A5113 Rev. A.
August (1992).
7.4 Kataoka, I. and Ishii, M., " Drift Flux Model for Large Diameter Pipe and New Correlation for Pool Void Fraction", Intl. J. Heat Afass Transfer, vol. 30,1927-1939 (1987).
7.5 G.E. Nuclear Energy, MFN. No. 077-94, May (1994).
7.6 U.S. Nuclear Regulatory Commission Request for Proposal, No.RS-RES- 93-049, November (1992).
7.7 Rodi, W. " Turbulent Buoyant Jets and Plumes", HMT Volume 6 Pergammon Pre ;s, Lon-don (1982).
7.8 Peterson, P.F., Schrock, V.E. and Greif, R., " Scaling for Integral Simulation of Mixing in large, Stratified Volumes", NURETH-6, October 508, Grenoble, France (1993).
7.9 Collier, J. G., Canvective Balling and Candensatian,2nd ed., McGraw-Hill, New York, (1981).
7.10 Private Communication with Brookhaven National Laboratory (BNL) Staff, April (1994).
7.11 G.E. Nuclear Energy MFN. No.167-93 October (1993). !
l 7.12 Churchill, S.W. and H.H.S. Chu, " Correlating Equations for Laminar and Turbulent Free I Convection from a Vertical Plate," Int /.1. Heat Afass Trun.sfer, 18,1323 (1975). !
7.13 Churchill, S.W. and H.H.S. Chu, " Correlating Equations for Laminar and Turbulent Free l
Convection from a Horizontal Cylinder," Intl. J. Heat Afass Transfer, 18,1049 (1975).
l
8-1
- 8. INSTRUMENTATION SYSTEM 8.1 Overview of PUMA Instrumentah The PUMA instrumentation consisti of over 340 measurements that will provide a detailed picture of the mass and energy content in each vessel as well as pressures and liquid levels.
The steam and air mass inventory for component j in terms of the measurements is given by mg j = { p (Tj,i p,j, xij) Avij (8.1) where i is the subscript for measurement i, p is density, T and p are temperature and pressure, x is the concentration of steam and Av is the volume corresponding to measurement i. Therefore, temperature, pressure and concentration are key measurements.
For the case of a liquid or two-phase mixture, the liquid mass inventory is simple to measure if the hydrostatic assumption is made and if p, < pr. Then the mass is given by Apii Aji mij = Ii (8.2) 8 where Apij si the pressure across volume i and A ij si the cross-sectional area. Therefore, Ap measurements are essential for the liquid inventory. Furthermore, these op measurements can be used to determine the liquid levels. In the case of the two-phase level in the reactor chimney, which is one of the most important variables in these tests, a number of Ap measurements will be performed in series. Assuming that the void fraction within the chimney is uniform, the two phase level is given by li Ap,
= (8.3)
Azi-1 Ap,-i ,
where l i is the level above pressure tap i Azgi is the distance between pressure taps i and i-1, l Api si the pressure drop between taps i and i + 1 and Ap,_ is the pressure drop between taps i !
and i-1.
The measurement of the energy inventory for component j is given by 1
I
8-2 m) hj = I p (Tj.i pij xij) h (Tj.i pij. xij) Avij (8.4) so the measurements required are basically the same as those for the mass inventory.
Table 8.1 is a list of the instrumentation associated with the mass and energy inventories for each component. This table is a subset of Tables 8.2 to 8.7, which list the complete measure-ments. Table 8.8 is a list of the instruments. A more detailed discussion of the specific measure-ments follows.
Table 8.1 Mass and energy measurements Steam W ater Component Pressure Concentration Temperature Level Reactor pl - p2, cne 11 T1 - T66 p3 - pl7 vessel p 72 Drywell p58 - p59 cnc5 - cnc9 T136 - %141 p60 - p61 T157 - T161 Pressure p66 - p67 ene10 T146 - T153 p63 - p64 suppression T162 - T164 pool GDCS p58 - p59 ene12 T82 - T90 p36 - p41 PCCS + -- --
Tl 12 - T123 p62 - p63 ICS pools Boundary flows between the vessels will also be measured. In addition, high frequency con-ductivity probes will be used for the first time in an integral test facility to detennine the local value of the void fraction and the flow regime.
8.2 Reactor Pressure Vessel Instrumentation The reactor vessel is the component that contains the majority of the instruments. Table 8.2 !
gives a summary of the technical specifications. The instrumentation layout for the reactor vessel is shown in Figures 8.1,8.2 and 8.4.
The most important measurement is the two-phase level over the reactor. This measurement j will be performed with sets of dp-cells and conductivity probes approximately 0.60 m apart in l the axial direction. The conductivity probes will provide a check on the local density. This, I 1
I
8-3 combined with the average density obtained by the dp-cells, will allow an accurate estimate of the two-phase mixture level, assuming that the void fraction in the chimney is uniform. Since the experimental transients are expected to be slow, this assumption should be valid except at the initial instants of the blowdown. ,
b Temperatures Temperatures are measured with K-type thermocouples (Omega Engineering, Inc., Stanford, CT). The range of measurements is 20 C to 180 C. The estimated uncertainty is 2 C.
The most important temperature measurement is at the steam dome. Therefore, a redundant measumment is performed (i.e., two thermocouples). A large number of themlocouples are used throughout the vessel to obtain a multidimensional picture of the temperature distribution. The downcomer has twenty-four thermocouples at six elevations and four azimuthal locations. Com-bined with the pressure measurements, they provide a detailed picture of the overall energy inventory. Furthermore, coolant bypass in case of a GDCS line break may be detected with these thermocouples. The temperature of the GDCS water is 20 C whereas the reactor vessel temperature will be greater than 100 C at the time of GDCS injection.
The lower plenum has twelve themmcouples at four elevations and three radial positions.
The reactor core has eight thermocouples to measure flow temperatures and twelve thermocou-ples to measure heater surface temperatures. These are arranged at four elevations: 25%,50%,
75% and 100% of the core height. The chimney has eight thennocouples at four elevations and two radial positions: the central duct and one peripheral duct.
Figure 8.1 also shows three wall heat flux sensors for measuring the heat loss from the reae-tor vessel. These are performed with ITI-TFM thermopile heat flux sensors (ITI Co., Delmar, CA). The uncertainty of these sensors is 1% of full scale. However, the uncertainty due to the intrusiveness of these probes needs to be determined.
8.2.2 Pressures Figure 8.2.2 shows the locations of the pressure taps. Pressures will be measured with STD924, STD930 and STD974 pressure sensors (Honeywell, Inc., Phoenix, AZ). The uncer-tainty in these sensors is 0.1% of full scale, in case of the STD924 this is 10 mm of water. For practical considerations, a minimum resolution of 25 mm water head has been selected to account for various sources of experimental error such as noise, condensation in the pressure lines, etc. 'Iherefore, the minimum uncertainty in the measurement of the liquid level will be approximately 25 mm.
Aside from the dome pressure and the bottom of core pressure, all other pressures are dif-ferential measurements. There are twelve differential pressure transmitters in the core-chimney
l i
1 8-4 I
section to obtain a detailed picture of the two-phase levels. One of these dp-cells measures the pressum drop across the separators standpipes and the top separator plate. There are five dp cells in the downcomer. These have been arranged in the same way as in the SBWR design (see Figure 8.9). ,
8.2.3 Conductivity Probes A dozen conductivity probes will be installed in the reactor and the chimney. A schematic probe is shown in Figure 8.3. These probes have been developed at Purdue over the past ten years and they provide reliable measurements of the indicator function [8.2]. They are sampled at 10 kHz and this signal is then integrated to obtain the void fraction. PUMA is the first integral test where this fundamental measurement will be performed. The accuracy of this measurement for bubbly and slug flows is approximately 10% of measurement. However, this is only a local measurement so the uncertainty in terms of the volume-averaged void fraction must be greater.
The conductivity probes provide additional information on the local flow regime which may be valuable in case of two phase flow instabilities. Moreover, they provide a check on the two phase level measun:ments.
These twelve probes are located at six elevations and two radial locations to obtain a multidi-mensional set of data. Half of the probes are located in the core and the other half are located in the chimney.
The probes located in the chimney are insened from the side of the vessel through special instrumentation ports that also hold pressure taps and thermocouples. The probes located in the reactor core together with the thermocouples are inserted from the bottom plate. These probes have the electrodes facing downward so they only detect bubbles that are moving up.
8.2.4 Additional Measurements There are four view ports with two video cameras. One in the lower plenum and three in the chimney. The view ports in the chimney will consist of double windows to view the flow within the chimney.
Figure 8.4 shows the vessel and boundary flow measurements. Three flow measurements are performed at the same elevation and three azimuthal locations in the downcomer: two pitot tubes and an EGG-TSA-12-C-D turbine meter (EGG Flow Technology, Phoenix AZ). This last probe can measure flow velocities with an uncenainty of 0.1 m/s. It has 2 R.F. pickups so it can sense the flow direction as well.
I l
I
1 i
8-5 8.2.5 Boundary Measurements The objective of these measurements is to obtain the mass and energy flow rates flowing out of the reactor vessel under single-phase and two-phase conditions. The accuracy of these meas-urements in the two phase regime has not yet been determined.
i All the boundary lines shown in Figure 8.4 are instrumented equally except for one: the feed- i water flow. He mason is that the SRVs, the DPVs, the steam line break, the feedwater break and the small break all represent blowdown outflows. The schematic of the instrumentation in one of ;
these lines is shown in Figum 8.5.
There are six measurements in each line. The first two are pressure and temperature to deter-mine the single phase thermodynamic state. For two-phase flow, there is a capacitance probe l that measures void fraction. The capacitance void meter was chosen over the gamma densitome-ter because of the cost and lengthy permit procedures regt< red with gamma ray sources. Capaci- ;
tance void meters have already been used by other researchers in two-phase flow [8.3-8.5]. We j intend to calibrate these sensors in an air-water loop. j The capacitance probe presently under development is shown in Figure 8.6. It is driven by a j 1 MHz signal. At this frequency the impedance is capacitive. The flow is expected to be homo-i geneous during the blowdown stage of the accident. This is imponant because the output of the capacitance probe is strongly dependent on the flow regime.
l The boundary measurements also include three flow probes: a magnetic flow meter, a flow nozzle and a vortex flow meter. The magnetic flow meters are Honeywell MagneW-3(XX) j models. Magnetic flow meters may be used also to measure the liquid velocity for two-phase
! flows if the void fraction is lower than 40% [8.1].
The flow nozzle provides a measurement of the kinetic energy of the flow. For homogeneous two-phase flow, this and the void fraction measurements obtained by the capacitance meter may be combined to obtain the flow rate. However, for separated flows with different phase veloci-ties, the number of measurements is not sufficient to uniquely determine the flow. Fortunately, during blowdown the flow is homogeneous and after blowdown the flow is expected to be single-phase vapor or single-phase liquid.
For low steam flow rates during decay heat removal, J-TEC vortex flow meters have been chosen (J-Tec Associates, Inc., Cedar Rapids, IA). They are composed of an ultrasonic vortex sensor that can detect very low steam flows. These flow meters include a flow computer that accounts for temperature and pressure variations. With the aid of this computer the manufac-turer claims an uncertainty of 1% of measurement across the full range.
\
l l
8-6 l
8.3 Containment Instrumentation The schematic of the containment measuremants in the drywell, pressure suppression pool and other tanks is given in Figure 8.7. The boundary flows are shown in Figure 8.8.
8.3.1 Temperatures Temperatures will be measured at four elevations in each partition of the PCCS tank as well as of the ICS tank. On one ICS condenser tube and on one PCCS condenser tube, detailed tem-perature measurements will be made: three center line temperatures and five surface tempera-tures of the tube. Three temperatures will be measured in each of the GDCS tanks. Six tempera-tures at six elevations will be measured within the drywell and four at the wall. Six temperatures at three elevations will be measured within the pressure suppression pool and two more in the gas volume.
8.3.2 Pressures The PCCS/ICS tanks will have two differential pressure transducers for level, as will each GDCS tank, the drywell and the pressure suppression pool. The drywell and the pressure suppression chamber will also have two pressure probes each to measure absolute pressure.
Additionally, there will be a differential pressure cell in the dryell vent to the pressure suppres-sion pool.
8.3.3 Oxygen Concentrations I
A Siemens Oxymat Se oxygen sensor (Siemens Industrial Automation Inc., Alpharetta, GA) will be used to measure all the oxygen concentrations. This instmment can produce fast and reli-able concentration measurements. The error in these measurements is 1/2% of full scale.10 points will be sampled: five in the drywell, one in the pressure suppression chamber and four in I
the PCCS/ICS lines. These measurements will provide data on the distribution of non-condensibles. In order to sample the locations systematically, an automatic array of solenoid valves will be cycled periodically by the data acquisition system. The tubing will be heated to j prevent condensation.
l The bleed time for each concentration measurement will be approximately 10 s, and the measurement time will be 5 s so each measurement will take approximately 15 s. Concentra-l tions that vary over periods of time shorter than that cannot be measured accurately. This is not expected to be a problem during decay heat removal. Each measurement will bleed 300 c.c.
from the system. This corresponds to 1 liter / min approximately and represents a very small per-turbation of the containment systems.
1
8-7 8.3.4 Additional Measurements The only additional measurement in the containment is a viewpon to be installed in the pres-sure suppression pool.
8.3.5 Containment Boundary Measurements Figure 8.8 is a schematic of the boundary measurements. Containment boundary measure-ments are simpler than the reactor vessel measurements because there are no two-phase flows.
However, there are non-condensibles.
For the PCCS and ICS extraction lines, the flows are a mixture of steam and air. Concentra-tions and temperatures will be measured for thennodynamic state. The flows will be measured with vortex flow meters. Additionally, one venturi will be located in one PCCS line and another in one ICS line.
For the PCCS and ICS condensate lines, as well as the GDCS lines and the equalization line, the flows are single-phase water only. Temperatures will be measured for thermodynamic state.
Liquid flows will be measured with magnetic flow meters.
The PCCS and ICS non-condensible bleed lines represent a more difficult problem since the flows are so low that they are hard to measure. Therefore, these flows will not be measured.
Instead, differential pressure transducers will be used to determine the level wi;hin the drains into the suppression pool. This will indicate the times at which noncondensibles are being discharged. Thermodynamic state will be obtained by measming temperature and concentration in these lines.
Finally, differential pressure transducers located across the valves will be used as vacuum breaker flow indicators.
8.4 Data Acquisition System The data acquisition and control system will be based on a network of 5 PCs. To increase the capability of the DAS in the future, all that is needed is to add more PCs to the network.
Since a sampling rate of I Hz is sufficient for this experiment, the total data collection rate is only 340 Hz which is substantially below the transfer rate of an Ethernet network. The server PC is a 66 MHz 486 machine. The other PCs are 486 machines of various clock speeds. The PC server will send a signal to all the DAS software across the network to launch the programs simultaneously. After that, no further synchronization is necessary, given the accuracy of the PC clocks.
A maximum data collection time of eight hours and a sampling rate of 1 Hz imply a total number of 9,792,200 data samples or approximately 20 MBytes (1 datum equals 2 Bytes). This
8-8 is well below the hard disk stowge capability. All data will be stored in the server and each PC will store its own data for redundancy. 'Ihe sampling rate of I Hz is sufficiently short with respect to all transients. For example, the blowdown stage of the main steam line break transient lasts approximately 4(X) s. ,
Keithley-Metabyte A/D and D/A converter boards will be used throughout the data acquisi-tion system. The software for data acquisition and control will be Lahtech Notebook. The software for communications will be Microsoft Windows for Workgroups. The bottleneck in the DAS is Labtech Notebook for Windows. At present it can only sample 3(X) channels per second on a 66 MHz PC, which nevertheless satisfies our requirements.
The central PC will collect all the data and send the necessary information to the control PC.
It will communicate with the four PCs through the Ethernet local area network.
PC #1 will be fully dedicated to acquire the 156 temperatures and nine heat fluxes. Two DAS-801 boards and eleven EXP-16 terminal boards will be used.
PC #2 will be used to acquire the seventy-six pressures, twenty magnetic flow signals, eight capacitance probe signals and data from the concentration probe. Hardware is the same as PC
- 1.
PC #3 will be fully dedicated to the twelve conductivity probes. These are the only probes that need to be s.,mpled at a rate higher than 1 Hz: 10 kHz. two DAS-58 data acquisition boards and two STA-U terminal boards will be used for this task.
PC #4 will be dedicated to monitor and control the valves and the heaters. Two PIO-24 parallel Digital I/O boards will be connected to two SSIO-24 relay boards with forty-eight SM-OAC5A relays and two PIO-HV boards will monitor the signals from the valves.
l l
l
Table 8.2: Vessel Measurement MEASUREMENT TYPE LABEL MAX RANGE MIN RANGE CATALOG #
STEAM DOME PRESSURE P P2 0-1 MPa 0-0.4 MPa STD 974 LOWER PLENUM PRESSURE P P72 0-1 MPa 0-0.4 MPa STD 974
- STEAM DOME TEMP T T1 20-180 C N/A K-TYPE STEAM DOME TEMP T T2 20-180'C N/A K-TYPE CORE & CHIMNEY TEMP T ._ _ _ __
T3 20-je'rC .
N/A __
K-TYPE CORE & CHIMNEY TEMP T T4 20-180'C N/A K-TYPE CORE & CH!MNEY TEMP T T5 20-180*C N/A K-TYPE CORE & CHIM_NEY TEMP T I6 20180'C N/A . .K-TYPE CORE _& CHIMNE.Y__ TEMP ..T T7 20_- 180'C _
N/A .
K-TYPE CORE & CHIMNEY TEMP T T8 20-180'C N/A K-TYPE CORE & CHIMNEY TEMP T T9 20-180'C N/A K-TYPE
- CORE & CHJMNEY TEMP .
T__ .
T10 20_-1801Q N/A K-TYPE CORE & CHIMNEY TEMP T T11 20-180'C N/A K-TYPE CORE & CHIMNEY TEMP T T12 20-180*C N/A K-TYPE CORE & CHIMNEY TEMP T _.
T13 20-180 C_ N/A K-TYPE CORE & CHIMNEY TEMP T T14 20-180'C N/A K-TYPE CORE & CHIMNEY TEMP T T15 20-180'C N/A K-TYPE CORE & CHIMNEY TEMP T T16 20-180'C N/A K-TYPE CORE & CHIMNEY TEMP T T17 20-180'C N/A K-TYPE CORE & CHIMNEY TEMP T T18 20-180'C N/A K-TYPE .
CORE VOID FRACTION CND CND1 0-1 N/A PURDUE CORE VOlD FRACTION CND CND2 01 N/A PURDUE CORE VOID FRACTION CND CND3 0-1 N/A PURDUE CORE VOID FRACTION CND CND4 0-1 N/A PURDUE CORE VOID FRACTION . QND CNDS 0-1 N/A PURDUE CORE VOID FRACTION QND_ QND6 0-1 N/A PURQ.UE CHIMNEY VOID FRACTION CND CND7 0-1 N/A PURDUE CHIMNEY VOID FRACTION CND CND8 0-1 N/A P.URDUE CHIMNEY VOID FRACTION CND CND9 0-1 N/A PURDUE CHIMNEY VOID FRACTION CND CND10 0-1 N/A PURDUE CHIMNEY VOID FRACTION CND CND11 0-1 N/A PURDUE CHIMNEY VOID FRACTION CND CND12 0-1 N/A .PURDUE .
CORE _& CHIMNEY LEVEL DP P3 0-30 KPa N/A .
STD 924 CORE & CHIMNEY LEVEL DP P4 0-30 KPa N/A STD 924 CORE & CHIMNEY LEVEL DP .P5 0-30 KPa N/A STD 924 CORE & CHIMNEY LEVEL DP P6 0-30.K_Pa N/A STD 924 CORE & CHIMNEY LEVEL DP P7 0-30 KPa N/A STD 924 CORE & CHIMNEY LEVEL DP P8 0-30 KPa N/A STD 924 oo b
Table 8.2: Vessel Measurement MEASUREMENT TYPE LABEL MAX RANGE MIN RANGE CATALOG #
SEPARATOR DP DP P9 0-30 KPa N/A STD 924 DOWNCOMER LEVEL DP P12 0-30 KPa N/A STD 924 DOWNCOMER LEVEL DP P13 0-30 KPa N/A STD 924 DOWNCOMER LEVEL DP P14 0-30 KPa N/A STD 924 DOWNCOMER LEVEL DP P15 0-30 KPa N/A STD 924 DOWNCOMER LEVEL DP P16 0-30 KPa N/A STD 924
._DOWNCOMER LEVEL DP P17 0-30 KPa N/A .
STD 924 DOWNCOMER FLOW DP P10 0-5 KPa N/A STD 924 DOWNCOMER FLOW D_P P11 0-5 KPa N/A STD 924 DOWNCOMER FLOW TURBINE TU1 . 0-3_m/s N/A . EGG-TSA-12-C-D VIDEO PORTS YP
.yIDEO PORTS yP vlDEO PORTS VP . __
VIDEO PORTS __
VIDEO CAMERA CCD VIDEO CAMERA CCD VIDEO CAMERA CCD INSTRUMENT PORT .
QOUBLE-PORT INSTRUMENT PORT DOUBLE-PORT INSTRUMENT PORT DOUBLE-PORT
.IN_STRU_ MEN _T PORT QQUBLE-PO_RT HEATER TEMP .T T19 20-180'C N/A K-TYPE -
. HEAT ER TE_MP T T2_0 20-180'C N/A K-TYPE HEATER TEMP T _ _ _ _ I21_ 20-180'C ..N/A .K-TYPE HEATER TEMP T T22 20-180'C N/A K-TYPE
'UEAT5R TEMP T__ T23 20-180'C N/A K-TYPE U5ATER TEMP T T24 20-180'C N/A .
K-TYPE SEATER T5fviP T .T25 20-180'C _N/A K-TYPE Ti5AiER TEMP T T26 20-180'C N/A K-TYPE HEATER TEMP T .T27 20-180'C N/A K-TYPE HEATER TEMP _
T _
T28 20-180'C N/A K-TYPE T T29 20-180'C N/A K-TYPE
_ HEATER TEMP HEATER TEMP T .T30 20-180'C N/A K-TYPE TEMP CONTROLLER T-CONTROLLER
. TEMP CONTROLLER T-CONTRQLLER TEMP CONTROLLER T-CONTROL _LER POWER CONTROLLER PWR-CONTROLLER 0-60 KW N/A POWER CONTROLLER PWR-CONTROLLER 0-60 KW N/A POWER CONTROLLER PWR-QONTRQLL_ER__ 0-60 liW N/A S
Table 8 2: Vessel Measurement MEASUREMENT TYPE LABEL MAX RANGE MIN RANGE CATALOG #
POWER CONTROLLER PWR CONTROLLER 0-60 KW N/A POWER CONTROLLER , P_WR-CONTROLLF.R_ _ 0-60_ KW Ng POWER CONTROLLER PWR CONTROLLER 0-60 KW N/A POWER CONTROLLER PWR. CONTROLLER 0-60 KW N/A POWER CONTROLLER PWR-CONTROLLER 0-60 KW N/A POWER METER PWR-METER 0-300 A N/A POWER METER PWR _ METER 0-300_A N/A POWER METER PWR-METER POWER METER PWR-METER POWER METER PWR-METER POWER METER PWR-METER ~
POWER METER PWR-METER POWER METER PWR-METER
~~ ~ -
DOVINCOMER TEMP I T31 20-180'C N/A K-TYPE DOWNCOMER TEMP T T32 20-180'C N/A K-TYPE DOWNCOMER TEMP T T33 20-180'C N/A K- TYPE QOWNCOMER TEMP T T34 20-180'C N/A K-TYPE DOWNQOMER TEMP .
T T35 20-180*C N/A K-TYPE DOWNCOMER TEMP T T36 20-180'C N/A K-TYPE
.QOWNCOMER TEMP T T37 20-180'C N/A _K-TYPE DOWNCOMER TEMP T T38 20-180'C N/A K-TYPE DOWNCOMER TEMP T T39 20-180'C N/A K-TYPE DOWNCOMER TEMP T T40 20-180'C N/A K-TYPE DOWN_QO_MER TEMP .
T T41 20-180'C N/A K-TYPE DOWNCOMER TEMP T T42 20-180'C N/A K-TYPE DOWNCOMER TEMP T T43 20-180*C N/A K-TYPE DOWNCOMER TEMP T T44 20-180*C N/A K-TYPE DOWNCOMER TEMP T T45 20-180'C N/A K-TYPE DOWNCOMER TEMP T T46 20-180'C N/A K-TYPE
.DOWNCOMER TEMP T _
T47 20-180'C N/A K-TYPE
.DOWNCOMER TEMP T T48 20-180'C N/A _K-TYPE .
DOWNCOMER TEMP .
T T49 20-180'C N/A K-TYPE DOWNCOMER TEMP T T50 20-180'C N/A K-TYPE DOWNCOMER TEMP T T51 20-180'C N/A K-TYPE DOWNCOMER TEMP T __
,T52 20-180*C N/A K-TYPE DOWNCOMER TEMP . T_ T53 20-180'C N/A K-TYPE DOWNCOMER TEMP T T54 20-180'C N/A K-TYPE LOWER PLENUM TEMP T T55 20-180*C N/A K-TYPE T56 20-180'C N/A K-TYPE oo LOWER PLENUM TEMP T
Table 8.2: VesselMeasurement
. MEASUREMENT TYPE _ .
LABEL .M_AX RANGE MIN RANGE CATALOG #
LOWER PLENUM TEMP T T57 20-180'C N/A K-TYPE
- LOWER PLENUM TEMP T TSB 20-180'C N/A K-TYPE LOWER PLENUM TEMP .T ,_ .T_59 20100'C N/A K-TYPE LOWER PLENUM TEMP T T60 20-180'C N/A K-TYPE LOWER PLENUM TEMP .
T T61 20-1gQ:C N/A K-TYPE LOWER PLENUM TEMP .T T62 20180'C 2 N/A K-TYPE LOWER PLENUM TEMP T T63 20-180'C N/A K-TYPE LOWER PLENUM TEMP T T64 20-180'O N/A K-TYPE LOWER PLENUM TEMP T T65 20-180'C N/A K-TYPE LOWER PLENUM TEMP T T66 20-180'C N/A K-TYPE VESSEL OUTSIDE TEMP T T70 2_0-180'C N/A .
K-TYPE VESSEL OUTSIDE TEMP T T71 20-180'C ..N/A K-TY.PE VESSEL OUTSIDE TEMP T_ T72 211801C N/A K-TYPE
_WALL HEAT FLUX O Q1 0-100W/m^2 N/A ITI-TFM WALL HEAT FLUX .Q Q2 0-100W/ma 2 N/A ITI-TFM WALL HEAT FLUX .O O3 0-100W/m^2 N/A .
ITI-TFM INLET CONCENTRATION CNC CNC11 0-20 % N/A OXYMAT SE I
E O
G
Table 8 3: Drywell Measurement i
MEASUREMENT TYPE LABEL MAX RANGE MIN RANGE CATALOG #
PRESSURE P P58 0-0.5 MPa 0-0.1 MPa STD 930 ,
PRESSURE P P59 0-0.5 MPa 0-0.1 MPa STD 930 TEMPERATURE T T136 20-180'C K-TYPE TEMPERATURE T T137 20-180'C K-TYPE TEMPERATURE T T138 .20-180'C K-TYPE TEMPERATURE T T139 20-180'C K-TYPE TEMPERATURE T T140 20-180'C K-TYPE TEMPERATURE T T141 20-180'C K-TYPE TEMPERATURE T T157 20-180'C K-TYPE <
TEMPERATURE T T_158 20-180'C K-TYPE
, TEMPERATURE T T159 20-160'C K-TYPE TEMPERATURE T T160 20-180'C K-TYPE TEMPERATURE T T161 20-180'C K-TYPE SUMP LEVEL DP P60 0-30 KPa N/A STD 924 SUMP LEVEL DP P61 0-30 KPa N/A STD 924 CONCENTRATION CNC CNCS 0-20 % N/A OXYMATSE
, CONGENTRATION .QNQ QNQ6 .
0-20 % N/A OXYM AT SE CONCENTRATION CNC CNC7 0-20 % N/A OXYMAT 5E '
CONCENTRATION CNC CNC8 0-20 % N/A OXYMAT SE i .QONCENTRATION QNC CNC9 0-20 % N/A OXYMATSE WALL TEMP T T142 20-180'C N/A K-TYPE '
_W _A_LL_ TEM P .
T T143 20-180*C N/A - .K_-TYPE
_WALL TEMP T T144 20-180'C N/A K-TYPE WALL TEMP T 1145 20-180'C N/A K-TYPE HEAT FLUX O 04 0-100W/m^2 N/A ITI-TFM HEAT FLUX O 05 0-100W/m*2 N/A ITI-TFM ,
HEATFLUX O .Q6 0-100W/m^2 N/A ITI-TFM HEAT FLUX O 07 0-100W/m^2 N/A ITI-TFM RPV-DRYWELL DP P P62 0-10 KPa STD 924 1
C
Table 8.4: Suppression Pool Measurement MEASUREMENT TYPE LABEL MAX RANGE MIN RANGE CATALOG #
WATER TEMP T T146 20-180'C N/A K-TYPE WATER TEMP T T_147 20-180'C N/A K-TYPE
.WATER TEMP T T148 20-180'C N/A K-TYPE WATER TEMP T T149 20-180'C N/A K-TYPE WATER TEMP T T150 .20-180'C NJA K-TYPE WATER TEMP T T151 20-180'C N/A K-TYPE TEMPERATURE T T162 20-180lO N/A K-TYPE TEMPERATURE T T_163 .20-180'C N/A _
K-TYPE TEMPERATURE T T164 _ 20 .180'C .N/A K-TYPE LEVEL .DP _
P63 0-14 KPa 0-10 KPa STD 924 LEVEL DP P64 0-14 KPa 0-10 KPa STD 924
! VENT LEVEL DP . .P65 .
0-14 K_Pa_ 0-10 KPa STD 924 l- PRESSURE P P66 0-0.5 MPa 0-0.1 MPa STD 930 PRESSURE P P67 0 0.5 MPa 0-0.1 MPa STD 930
. GAS CONCENTRATION _ QNQ CNQ10 .
0-20 % N/A OXYMAT SE GAS TEMPERATURE T- T152 20-180'C N/A K-TYPE GAS TEMPERATURE T T153 20-180'C N/A K-TYPE VIDEO PORT VP VIDEO PORT VP VIDEO PORT VP HEAT FLUX Q 08 0-100W/ma2 N/A ITI-TFM HEA_T FLUX O 09 . 0-100W/m^2 . N/A ITI-TFM l CAMERA CCD I
_ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ = _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . . __
Table 8 5: PCCS and ICS Measurement MEASUREMENT LABEL TYPE MAX RANGE MIN RANGE CATALOG #
INLET CONCENTRATION CNC01 CNC 0-21 % 0-21 % OXYMATSE INLET CONCENTRATION CNCO2 CNC 0-21 % 0-21 % OXYMATSE NONCONDENSABLE VENT CONC CNCO3 CNC 0-21 % 0-21 % OXYMAT SE NONCONDENSABLE VENT CONC CNC04 CNC 0-21 % 0-21 % OXYMAT SE CONDENSER PRESSURE DROP P44 DP STD 924 CONDENSER PRESSURE DROP P45 DP STD 924 CONDENSER PRESSURE DROP P46 DP STD 924 CONDENSER PRESSURE DROP P47 DP STD 924 CONDENSER PRESSURE DROP .P48 . DP. STD 924 CONDENSER PRESSURE DROP . P49 DP STD 924 INLET FLOW P42 _
DP 0-3 KPa 0-3 KPa STD 924 INLET FLOW P43 . .DP 0-3 KPa 0-3 KPa STD 924 NONCONDENSABLE VENT PRESS..P50 DP -1KPa-5KPa -1 KPa-5KPa STD 924 NONCONDENSABLE VENT PRESS P51 DP -1KPa-5KPa -1 KPa-5KPa STD 924 NONCONDENSABLE VENT PRESS PS2 DP -1KPa-5KPa -1 KPa-5KPa NONCONDENSABLE VENT PRESS PS3 DP -1KPa-5KPa -1KPa-5KPa NONCONDENSABLE VENT PRESS P54 DP -1KPa-5KPa -1KPa-5KPa NONCONDENSABLE VENT PRESS P55 DP -1KPa-5KPa -1KPa-5KPa POOL LEVEL P56 DP 110 cm 110 cm STD 924 POOL LEVEL _P57 DP 110 cm 110 cm STD 924 ,
CC 9DENSATE FLOW MAG 12 MAG N/A 0-0.1 m/s MagneW 3000 t CC 4DENSATE FLOW MAG 13 MAG N/A 0-0.1 m/s MagneW 3000 l CONDENSATE FLOW MAG 14 MAG N/A 0-0.1 m/s MagneW 3000 CONDENSATE FLOW MAG 15 _ MAG N/A 0-0.1 m/s MagneW 3000 l CONDENSATE FLOW MAG 16 MAG N/A 0-0.1 m/s MagneW 3000
- CONDENSATE FLOW MAG 17 MAG N/A 0-0.1 m/s MagneW 3000 INLET FLOW N10 VENTURI N/A 4.7-14.1 m/s INLET FLOW N11 VENTURI ..N/A 4.7-14.1 m/s CONDENSATE TEMP T100 T 20-180'C 20-140'C K-TYPE l CONDENSATE TEMP T101 T 20-180'C 20-140'C K-TYPE T102 T 20-180'C 20-140'C CONDENSATE TEMP K-TYPE CONDENSATE TEMP TiO3 T 20-180'C 20-140'C K-TYPE CONDENSATE TEMP T
_104 T 20-180'C 20-140'C K-TYPE CONDENSATE TEMP ,T105_ .T 20-180*C 20-140'C K-TYPE CONDENSER TUBE TEMP T124 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP .
T125 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP _ _. T126 _ T 20-180*C 20-140'C K-TYPE CONDENSER TUBE TEMP T127 _ T_ 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP , _ T128_ _ _ T 20-180'C 20-140'C K-TYPE -
?'
G
Tabte G 5. PCCS and ICS Measurement MEASUREMENT LA_ BEL _ TYPE MAX RANGE MIN RANGE CATALOG #
CONDENSER TUBE TEMP _ _ _ _ _. .T129 T .20-180'C 20-140'C K-TYPE CUND5NSER TU_BE T5@F _ ~ TR ~ f] T
~~ 20-180'C M 80'C 20-140'C K-TYPE CONDENSER TUBE TEMP __ _ T131 _ _ 20-140'C K-TYPE CONDENSER TUBE TEMP T132 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP T133 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP T134 T 20-180*C 20-140'C K-TYPE CONDENSER TUBE TEMP _ .
T135 .
T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP T165 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP T166 T 20-180'C 20-140*C K-TYPE CONDENSER TUBE TEMP .
T167 T 20-180'C 20-140'C K-TYPE ~-
CONDENSER TUBE TEMP T168 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP .
T169 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP T170 T 20-180*C 20-140'C K-TYPE CONDENSER TUBE TEMP T171 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP T172 T 20-180'C 20-140'C K-TYPE CONDE_N SER TUBE TEMP T173_ _ T 2
_ 0-180'C .20-140'C K-TYPE CONDENSER TUBE TEMP T174 T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP .
T175 .
T 20-180'C 20-140'C K-TYPE CONDENSER TUBE TEMP T176 T 20-180'C 20-140'C K-TYPE INLET TEMP _
T94 ,T 20-180'C 20-140'C K-TYPE .
INLET TEMP 195 T 20-180'C 20-140'C K-TYPE INLET TEMP _
T96 T 20-180'C 20-140'C K-TYPE INLET TEMP T97 T 20-180'C 20-140*C K-TYPE INLET TEMP T98 T 20-180'C 20-140'C K-TYPE INLET TEMP T99 T 20-180'C 20-140'C K-TYPE NONCONDENSABLE VENT TEMP T106 T 20-180'C 20-140'C NONCONDENSABLE VENT TEMP T107 T 20-180'C 20-140'C NONCONDENSABLE VENT TEMP T108 T 20-180'C 20-140'C K-TYPE -
NONCONDENSABLE VENT TEMP T109 T 20-180'C 20-140'C K-TYPE NONCONDENSABLE VENT TEMP T110 T 20-180'C 20-140'C K-TYPE NONCONDENSABLE VENT TEMP . T111 T 20-180'C 20-140*C K-TYPE POOL TEMPERAT_URE T112 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T113 ..T 20-180'C 20-140*C K-TYPE POOL TEMPERATURE T114 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE _ _ _ _ T115_ _ T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T116 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE _. T117____ T_ .
20-180*C 20-140'C K-TYPE POOL TEMPERATURE _ _ _ _ _ T118____ T 20-180'C 20-140'C K-TYPE POOLTEMPERATURE T119 T 20-180'C 20-140'C K-TYPE
?'
E
Table 0.5: PCCS and ICS Measurement MEASUREMENT LABEL T(PE MAX RANGE MIN RANGE CATALOG #
POOL TEMPERATURE , _ _ _ _ T120._._ T __ _ __ 20-180'C 20-140*C K-TYPE POOL TEMPERATURE . T121 __ T .20-180"C 20-140'C K-TYPE POOL TEMPERATURE T122 _, T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T123 _ _ T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T177___ T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE _
T178 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T179 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T180 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T181 _
T 20-180'C 20-140'C K-TYPE POOLTEMPERATURE T182 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T183 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T184 T 20-180'C 20-140'C K-TYPE -
POOL TEMPERATURE -T185 T 20-180'C 20-140*C K-TYPE POOL TEMPERATURE T186 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE _
T187 T 20-180'C 20-140'C K-TYPE POOL TEMPERATURE T188 _ _ _ . . T 20-180'C 20-140'C K-TYPE FLOW RATE V07 VORTEX 0.0267kg/s 0.0089 kg/s VP715TS FLOW RATE V08 VORTEX 0.0267kg/s 0 0089 kg/s VP71STS FLOW RATE V09 VORTEX 0.0267kg/s 0.0089 kg/s VP71STS FLOW RATE V10 VORTEX 0.0267kg/s 0.0089 kg/s VP71STS .
FLOW RATE V11 VORTEX 0.0267kg/s 0.0089 kg/s VP71STS FLOW RATE V12 VORTEX 0.0267kg/s 0.0089 kg/s VP715TS r
1 Y
G
Table 8.6: GDCS Measurement MEASUREMENT TYPE LABEL MAX RANGE MIN RANGE CATALOG #
LEVEL DP _.
P36 0-14 KPa N/A STD 924
. LEVEL DP P37 Qd4 KPa Nf_A STD 924
. LEVEL DP P38 0-14 KPa N/A STD 924
. LEVEL .
DP P39 0-14 KPa N/A STD 924 LEVEL DP P40 0-14 KPa N/A STD 924 I LEVEL DP P41 0-14 KPa N/A STD 924 TEMPERATURE T T82 20-180'C .N/A K-WPE TEMPERATURE T T83 20-180'C N/A K-TYPE TEMPERATURE . T_ T84 20-180'C N/A K-TYPE TEMPERATURE T T85 20-180'C N/A K-TYPE TEMPERATURE T T86 20-180'C N/A K-WPE TEMPERATURE T T87 20-180 C N/A K-TYPE TEMPERATURE' T T88 20-180'C N/A K-TYPE TEMPERATURE T T89 20-180*C N/A K-TYPE TEMPERATURE T T90 20-180'C N/A K-TYPE OUTLET FLOW MAG _ __ MA.G9 Od_m/s N/A MagneW 3000 OUTLET FLOW MAG MAG 10 0-1 m/s N/A MagneW 3000 OUTLET FLOW M_AG MAG 11 Q-1_m/s N/A MagneW 3000 OUTLET TEMP T T91 20-180'C N/A K-TYPE ,
.OUT_LET TEMP T- T92 20-180'C N/A K-TYPE
.OU_TLET TEMP T T93 20-180'C N/A K-TYPE COVER GAS CONCENTRATION. CNC CNC12 0-20 % N/A OXYMATSE
'oo
f Table 8.7: Boundary Flow Measurement COMPONENT MEASUREMENT TYPE LABEL MAX RANGE MIN RANGE CATALOG #
FLOW RATE DP PIB 0-8 KPa STD 924 STEAM LINE BREAK FLOW RATE VORTEX VI 0.Q22 m/s 0.66 m/s VP702TS STEAM LINE BREAK FLOW RATE NOZZLE N1 Q.0_22 m/s 0.66 m/s STEAM LINE BREAK STEAM LINE BREAK ,FL_QW RAT.E (LIQUID)__ _.
MAG __ MAG 1 Q-5 m/s _N/A MagneW 3000 VOID FRACTION CAP CAP 1 0-1 N/A PURDUE STEAM LINE BREAK PRESSURE P ..P19 0-1 MPa 0-0.01 MPa STD 974 STEAM LINE BREAK TEMPERATURE T T73 20-180'C N/A K-TYPE STEAM LINE BREAK FLOW RATE DP P20 0-0.5 MPa STD 930 VESSEL _ DPV FLOW RATE DP P22 0-0.5 MPa STD 930 VESSEL DPV FLOW RATE VORTEX V2 0.0082 m/s 1.11 m/s VP715TS VESSEL DPV V3 ..0.0082 m/s 1.11 m/s VP71STS VESSEL DPV ..FLQW RATE_
VQRT_EX .
N2 0.0082 m/s 1 ;1 m/s YESSEL_DPy ELOW RATE _ NQZZLE _ _
NO_Z ZLE N3 0.Q0_82 m/s 1.11 m/s VESSEL DP_V FLOW RATE ___._
MAG MAG 2__ 0-7 m_/s N/A MagneW 3000 VESSEL DPV FLQW RATE (LIQUID) _ _ _
MagneW 3000 MAG M_AG3 0-7 m/s N/A VESSEL DPV FLOW RATE (LIQUID) _
VOID FRACTION CAP CAP 2 0-1 N/A PURDUE VESSEL DPV VOID FRACTION QA_P QAP3 0-1 N/A PURDUE VESSEL DPV .
STD 974 PRESSURE P P21 0-1 MPa 0-0.04 MPa VESSEL DPV STD 974 0-1 MPa 0-0.04 MPa PRESSURE P P23 VESSEL DPV N/A K-TYPE TEMPERATURE T T74 20-180'C VESSEL DPV N/A K-TYPE T T75 20-180'C YESSEL_DPV .T EM P.ER AT_URE_ .
FLOW RATE DP P24 0-0.5 MPa Q-8 kPa STD 930 MSL DPV 0.19 m/s y_QRTEX V4 0.Q012_m/s VP7QOTS MSL__DPV ..FLQW RATE FLOW RATE NOZZLE N.4 0.0M2_m/s OM_m/s MSL DPV ___
0-1.2 m/s N/A MagneW 3000 MAG _ MAG 4 MSL DPV FLOW RATE (LIOUID) .
VOID FRACTION . QA_P QAP4 0-1 .N/A PURDUE M5L DPV P P25 0-1 MPa 0-0.01 MPa STD 974 MSL DPV PRESSURE T T76 20-180'C N/A is-TYPE MSL DPV TEMP _ERATURE K-TYPE TEMPERATURE T . T76-1 20-180'C N/A MSL DPV 0-0.5 MPa 0-8 KPa STD 930 DP P26 SRV .. FLOW RATE STD 930 DP P28 0-0.5 MPa 0-8 KPa
~5RV ELOW RATE FLOW RATE NOZZLE N5
.SRV NOZZLE N6 SRV FLOW RATE MAG . . M AG5 0-2.9 m/s N/A MagneW 3000
'5RV FLOW RATE ILIQUID) MagneW 3000 MAG MAG 6 0-2.9 m/s N/A
'5RV FLOW RATE _(LIQUID)
CAP 5 0-1 N/A PURDUE SRV VOID FRACTION . CAP CAP CAP 6 0-1 N/A PURDUE SRV VOID FRACTIO _N P P27 0-1 MPa 0-1 MPa STD 974 x PRESSURE SRV PRESSURE P P29 0-1 MPa 0-1 MPa STD 974 g SRV
l Table BJ: Boundary Flow Measurement l
l COMPONENT MEASUREMENT TYPE .
LABEL MAX RANGE MIN RANGE CATALOG #
SRV TEMPERATURE T T77 20-180*C N/A K-TYPE '
, . TEMPE_R_A_TURE . _T78 N/_A l' FEED WATER BREAK FLOW RATE DP .. P_30 0-1 MPa STD 930 l FEED WATER BREAK FLOW RATE .
VORTEX V5 0.0051 ko/s 0.0153 kg/s VP701TS FEED WATER BREAK FLOW RATE NOZZLE ..N7 0.0051 kg/s 0.80 ko/s FEED WATER BREAK FLOW RATE (LI.OUlD) MAG MAG 7 0-1 m/s N/A MagneW 3000 FEED WATER BREAK VOID FRACTION CAP CAP 7 0-1 N/A PURDUE
, FEED WATER BREAK PRESSURE P P31 0-1 MPa 0-0.4 MPa STD 974 FEED WATER BREAK TEMPERATURE T T80 20-180*C N/A K-TYPE SMALL BREAKS FLOW RATE .
DP .
P32 STD 924 SMALL BREAKS FLOW RATE VORTEX V6 VP700TS SMALL BREAKS FLOW RATE NOZZLE N8 SMALL BREAKS FLOW RATE (LIQUID) M A.G .
MAG 8 MagneW 3000 ,
SMALL BREAKS VOID FRACTION CAP CAP 8 0-1 N/A PURDUE SMALL BREAKS PRESSURE P P33 0-1 MPa 0-0.4 MPa STD 974 SMALL BREAKS TEMPERATURE T T81 20-180*C N/A K-TYPE FEED WATER FLOW RATE DP P34 -
FEED WATER FLOW RATE ORIF N9 FEED WATER . PRESSURE P P35 0-1 MPa 0-0.01 MPa STD 974 FEED WATER TEMPERATURE T T79 20-180'C N/A K TYPE FEED WATER FLOW CONTROLLER FLW - CONTROL VACUUM BREAKER V_AQUUM BREA_KER_ FLOW DP_ P68 0-20 KPa N/A .STD 924 VACUUM BREAKER VACUUM BREAKER FLOW DP P69 0-20 KPa N/A STD 924 VACUUM BREAKER VACUUM BREAKER FLOW DP P70 0-20 KPa N/A STD 924 VACUUM BREAKER VACUUM BREAKER LEAK DP P71 0-20 KPa - N/A _STD 924 EQUAllZATION LINE FLOW MAG _ MAG 18 0-0.5 m/s N/A MagneW 3000 EQUALIZATION LINE FLOW MAG MAG 19 0-0.5 m/s N/A MagneW 3000 EQUAllZATION LINE FLOW MA.G ..M_AG20 0-0.5 m/s N/A MagneW 3000 TEMP T T154 20-180*C N/A K-TYPE
.EOUAllZATION LINE EOUAllZATION LINE TEMP T T155 20_-180*C N/A K-TYPE EQUALIZATION LINE TEMP T T156 20-180'C N/A K-TYPE 9
E
l-I Table 8.8: Instruments
. CATALOG # TYPE _ MIN RANGE MAX RANGE SENSITIVITY VP702TS .
VORTEX 0.0056-0.181 kg/s 1% reading ,
VP715TS VORTEX 0.0028-0.113 kg/s 1% readina VP701TS VORTEX 0.0014-0.0396 kg/s 1% reading VP700TS VORTEX 0.00048-0.0112 kg/s_ 1% readirm MAGNEW 3000 MAGNETIC 0.1-1 m/s 1-10 m/s 2% reading at 10%.1/2% reading at 40%
STD 924 PRESSURE 0-0.0062 MPa 0-0.1 MPa 250 Pa or 1/4% rance STD 930 PRESSURE 0-0.035 MPa 0-0.7 MPa 250 Pa or 1/4% range
.STD 974 PRESSURE _
0-0.7 MPa 0-21 MPa 1/4% RANGE OMEGA K-TYPE TEMPERATURE -200*-1250'C 22C -
PURDUE QO_NDUQTIVITY___ 0-100 % 10% reading _ _ _ _ , . _ _
EGG-TSA-12-CA-HG-43 el TURBINE 0.07-0.76 m/s 0.07-6.1 m/s 2% readire FLOW SYSTEMS VENTURl+ NOZZLE 1.25% reading ITI-TFM HEAT FLUX - 0-200 w/m 2w/m SIEMENS OXYMAT SE CONCENTRATION 0.1-20 % 1/2% readina PURDUE CAPACITANGE 0-100 %
N
, , -.-a- .n.. - . , - - - . . _ . _ . _ - . . - _ - . - - - _ . . - - -
.7-e 6- '
I I UPPER CHIMNEY = uasa s'e'e's 1 @@
6.tsa h LOVER CHIMNEY = t625m GGGG , 88 6666 , es CORE N @@
SECTION =uesa g gg s es QQQQf w---O S 9 LOVER w eee PLENUM = la h. ,
\\*dhh
\\' bbh HBOOO SOOO Figure 8.1 PUMA RPV temperature instrumentation -
8 23 e .
~
e 6.15m flh e -
pp e ee hb e _- e 06m Figure 8.2 PUMA R.PV pressure and void fraction instrumentation
8-24 l l
GLASS M
STAINLESS STEEL T
Figure 8.3.a Schernatic of the conductivity probe Vout N, "*"
PROBE DO o threshold V, ____ ____ ___ ___
Vw _ __, J ___ w ___ w _ _
- t1- -
- t3_ _
Time
! T !
1 a = - E t, T'
Figure 8.3.b The output signal of the conductivity probe
8-25 ;
1 l e e VESSEL VESSEL DPV
- X , _ W ---C>C _ g X
FEED- v FEED-VATER VATER i -' i .
^ BREAK l
t-6 e I SMALL X - e n v
- BREAKS Figure 8.4 PUMA RPV boundary measurements
- -- - _ _ - - - - i
4 8-26 F
P,T DP
_] n same.
s __-
l l e
meer
- a -
Q nu-Pos CAPM:RMCE MOCDC N N BALL VALVE NETER FLDur WETER N0ZZLE m WETER W/ DELTuc Acnavon Figure 8.5 Typical instrumented ADS line
)
l
'8-27 i l
t-STAINLESS STEEL PIPE
' STAINLESS STEEL TEFLON TUBE RING ELECTRODES
/
/ \ \ / f (f\ [T'\ [l'\ . { [) . >
1 11 11 l ll ll
( CN//AE////)f
--IMPEDANCE' CIRCUIT
- TO COMPUTER -
Figure 8.6 Capacitance probe schematic i
l
8-28 N @ " @-
QSi i n/
G@@*d-uuu.,b_ @_-
., g:-
i i + -@g @n f
b nn
++ e~p _4 ii -64
%e@, a e s b ,___,
ee G 8 m o
\ e e -
e g_ , ___,
I h 8 __
e \ e a
C
! \ e :
e 0 i
8 1
8 Figum 8.7 PUMA containment temperatures and pressure l
f
, _ 1 8-29 ;
l l
t PCCS Extraction ICS Extraction x =
CNC1 794 W CNC sh4 !3 ;" [*@
W 4' PCCS ICS 71.
Noncondensable Bleed Lines p i
'PCCS Condensate O "d' "
DCS h) @" *@ gny N ni CDCS 3
E E Line ri i . . . . .
Vacuu n Brooker 3
" Vacuum Brec kor_ Lagh .,',
v Ecuollrotion Qne_ _ _ am 1154 wie .3 -iw ,3
-30 I
Figum 8.8 PUMA containment boundary flow measurement-
8-30 I
I u l l l l 8 J/ f i g
s
) p g ;L -
?lj c:
_esG::r a '[ n b ,!
Level 9 '
< 3l l i<ino#m4" va l l Level 8 j'
8 l
NWL l 11 i i i i i I
- ll g ,
l ! '., Level 3
- Y l i i '. '! ; ii l s q FW
$ ! i l ! i i Ill U'y
. U l F ej j
Level 2
{
c: !
I 8l E i El j-I
- l
, Level.1 l t d'l
.. l Es l l
i
? GDCS i
H g TAF l ini i I
a l
BAF ll f
_ ghill l llQ D '
C
(, ug I !
I '. : : Instrument l Tap Range I
l' o = .2 g ___ _. _ y Instrument N J Readout Range Il l Figure 8.9 Water level ranges in the SBWR
8-30 References 8.1 Knoll, K. E., " Investigation of an Electromagnetic Flow Meter for Gas-Liquid Two-Phase Flow Measurement," ANS Winter Meeting, San Francisco, CA, November 11,1991.
8.2 Revankar, S.T., Ishii, M., " Local Interfacial' Area Measurement in Bubbly Flow," Intl. J.
Hear Mass Transfer, Vol. 35,913-925 (1992).
8.3 Andreussi, P., D.I. Donfrancesco A. and Messia, M., "An Impedance Method for the Measurement of Liquid Hold-up in Two-phase Flow," Intl. J. Multiphase Flaw, Vol 16, No. 6,777-785 (198P).
8.4 Merilo, M., Dechene, R.L., and Ciconlas, W.M., " Void Fraction Measurement with a Rotating Electric Field Conductance Gauge,"1. Heat Transfer, Vol. 99,330-332 (1977).
8.5 Cimorelli, L. Evangelisti, R., "The Application of the Capacitance Method for Void Frac-tion Measurement in Bulk Boiling Conditions," Intl.1. Heat & Mass Transfer, Vol.10, 277-288 (1967).
l l
I
1 i
l 9-1 !
- 9. PUMA SYSTEM DESIGN 9.1. Scaling of PUMA Facility The scaling approach for PUMA is based on thr.ee levels of scaling: integral system scaling; mass and energy inventory and boundary flow scaling; and local phenomena scaling. In the integral system scaling, the fluid continuity, integral momentum and energy equations in one-dimensional, area-averaged forms are used with the appropriate boundary conditions and the solid energy equation. From the non-dimensional form of the integral response functions under small perturbations, important dimensionless groups characterizing geometric, kinematic, dynamic, and energetic similarity parameters an: derived. These are the geometrical non-dimensional groups, friction number, Richardson number, characteristic time constant ratio, Biot number and heat source number. For a two-phase system, the non-dimensional numbers are the phase change number, subcooling number, Froude wmber, drift flux number, thermal inertia ratio number, two-phase friction number and orifice number.
In the mass and energy inventory and boundary flow scaling, the overall control volume bal-ance equations for mass and energy are used. The scaling criteria give the similarity for flow area, velocity of fluid and enthalpy flow in the channel.
In the local phenomena scaling, the dominant phenomena are examined individually and appropriate similarity parameters are derived. The local phenomena considered are flow insta-bilities, choked flow, un-choked flow, flow regime and relative velocity, critical heat flux, natural circulation, flashing, condensation, heat source and sink, and mixing and stratification.
The scaling factors chosen for the confirmatory integral test facility (PUM A) are, of course, a compromise between several factors. For example, there is the requirement of keeping the facil-ity within manageable size and cost, while at the same time constructing as large a facility as possible in order to provide a meaningful basis for extrapolation to the full-scale nuclear reactor.
The plan to construct a facility scaled by 1/4 in height and 1/4(X) in volume was such a compromise.
The overall cost of the facility is mainly determined by the facility volume scale, and the number of instrumentations, and in lesser degree is affected by the total height of the facility.
The proposed 1/4(X) volume scale is determined by the overall cost constraint, scale relation to the existing facility, and simulation capability of key phenomena.
The advantages of the present scale model are summarized below.
- 1. Well-balanced simulation of the frictional resistance and the hydrostatic driving head.
This balance is essential in preserving the thermal-hydraulic characteristics of the loop.
9-2 l 2. Small aspect ratio similar to the prototype system. The value of la/da is only 2.5 relative I
to the linear scale of 1. This value of 2.5 is closer to unity than any facility which has been built or is under construction. Hence, the 2-D and 3-D phenomena are best simu-lated. .
- 3. Relatively small heat loss to the structure by using the reduced-heght model. For a fixed-volume scale model, the heat loss distortion is smaller for a shorter facility. The reduction of height by 1/4 gives considerable advantage over a full-height facility.
- 4. Accurate simulation of flow rates in interconnecting lines driven by hydrostatic head. By using the scaling method presented here, the flow rates in PCCS, GDCS and other con-necting lines are well-scaled to simulate the prototype conditions. The scaling study indi-cates that it is not necessary to have the full height. This is contrary to the common belief that full-height simulation is necessary for the best simulation.
- 5. Preservation of flashing phenomena due to hydrostatic head decrease as the liquid rises in the RPV. A detailed study of the flashing phenomena indicates that reduction of the height beyond 1/4 scale may start to decrease the liquid flashing phenomena significantly.
Since the mixture level is one of the most important parameters in terms of safety, the choice of the 1/4-height appears to be optimum.
9.2 Preliminary Design Based on the above system scale ofla = 1/4 and ag = 1/lm a preliminary design of the pro-totypic pressure, reduced-height integral test facility has been developed at Purdue. The facility is designed to simulate the low-pressure phenomena of SBWR' The basic dimensions and .
parameters are given in Table 9.1. The overall configuration of the facility is shown in Figures 9.1 and 9.2. These figures show the relative locations and components.
It is clear from these drawings that the reactor vessel, containment and suppression pool are installed as independent pressure vessels with interconnecting piping systems. This option has been chosen over the integrated component system for increased accessibility to each com-ponent, simplified instrumentation systems, including flow visualization capability, and flexibil-ity of the system. Any later modifications by GE to the SBWR prototype design can he easily accommodated by this separate component design.
The reactor vessel itself is geometrically scaled and the vessel internals are simulated. This insures that important 2-D and 3-D voiding phenomena and natural circulation instability phenomena can be well-preserved. It has an annulus downcomer with cylindrical core and chimney sections. Since the reactor vessel is outside of the containment, the vessel and all the connecting lines for ADS, GDCS and suppression pool are accessible from the outside.
9-3 9.3 OverallInstrumentation The instrumentation consists of approximately 340 measurement points. The main objective is to measure the mass and energy content in each vessel as well as the two-phase level in the .
reac:or vessel. High-frequency conductivity probes will be used for the first time in an integral f i
test facility to determine the local value of the void fraction and the flow regime in the reactor vessel.
Approximately 2(X) measurements will be made in the reactor vessel. The most important quantity is the two-phase level in the core-chimney section. For this purpose, pressure taps are located approximately every (dX) mm axially and at two radial locations corresponding to the central and peripheral sections of the chimney. These, combined with the conductivity probe measurements at the same locations, will provide the position of the two-phase level within the resolution of the pressun: sensors (i.e.,25 mm of water) assuming a uniform void fraction above I
the measurement point closest to the surface. Another set of differential pressure transducers will be used to measure the downcomer level and a redundant measurement at the steam dome provides the vessel pressure. Seventy-two thermocouples will be used for detailed temperature distributions in the downcomer, lower plenum, reactor and chimney.
Measurements in the containment include single-phase levels, pressures, temperatures and non-condensible concentrations. Temperatures at six elevations will be measured in the suppres-sion pool to determine the level of stratification. For drywell stratification, ten temperatures and five concentrations will be measured.
Boundary flows between the vessel and the containment will be measured as well. The measurement in the blowdown lines represent a very challenging task because the flow may be either single-phase liquid or vapor or two-phase flow. Moreover, these measurements must be non-intrusive since the vessel and containment are strongly coupled during decay heat removal.
) Capacitance probes will be used to determine the density. The flow will be measured by a variety of instruments.
Magnetic flow meters will be used to measure the velocity of single-phase liquid and bubbly two-phase flows. These measurements are effective up to 409 void fraction.
Nozzles will be used to measure the inertia of two-phase flows and high-speed vapor flows during blowdown. If it is assumed that these flows are homogeneous, the flow can be calculated from the flow nozzle pressure drop and the capacitance meter measurements combined.
Vortex flow meters and RTDs will be used to measure low-speed vapor flows during decay heat removal. These can measure very low steam velocities because of a unique ultrasonic sens-ing technique.
9-4 9.4 Initialization of Test The design of the PUMA facility includes the necessary provisions to establish initial condi-tions for the proposed set of experiments. The PUMA facility is designed to study transient phenomena at and below 1.03 MPa (150 psia). Hence, the initial conditions for each test should correspond to the conditions at 1.03 MPa (150 psia), it is proposed to use results of RELAP5/CONTAIN computer code calculations to determine initial conditions. A team from Brookhaven National Laboratory (BNL) will conduct analysis using RELAP5/CONTAIN code, for each break simulation. The computer code calculations are performed for the prototype SBWR from the initiation of an accident (Break-LOCA) and progress through blowdown to the long term cooling scenario. The results of the code calculations are the basis for the initial test conditions on the PUMA facility. Purdue will determine the initial condition for PUMA testing at 1.03 MPa (150 psia) from the initial blowdown phase predicted by the code. It should be noted that the code predictions for the blowdown phase have been validated against experimen-tal data.
The RELAP5/CONTAIN code involves models for RPV, core, vessel wall heat structures, PCCS/ICS condensers, containment, suppression pool, connecting piping, etc. The transient cal-culations provide the following parameters in all components of the facility modeled:
e Pressure e Liquid levels in RPV, SP, DW, etc.
- Reactor decay power e Mass inventory in each component e Temperature of both phases e Heat structure temperature (walls of RPV, DW, condenser, etc.)
e Heat flux in the heat transfer components e Non-condensible concentration e Void fraction e Mass flow rate e Liquid and gas velocity Thus, all necessary initial conditions are obtained from RELAP5/CONTAIN code calculations at 1.03 MPa (150 psia) pressure. In order to establish these initial conditions in the experiments, the RPV, DW, SP and other vessels and piping have adequate provisions. All component.s in the facility have fill and drain lines to establish the mass inventory. Components with thermalhy-l draulic phenomena, such as RPV, SP and PCCS/ICS pools, have independent electrical heaters. !
Using these heaters, the required temperature in these units are established. To set the steam / air
]
l l
1 l
9-5 l
concentration in DW and SP, an independent steam generator is used. The RPV, SP and DW- ,
l have independent purge and pressure relief lines. These purge and relief lines each contain two sets oflines, one for coarse setting and the other for fine tuning. Using these lines, the required void fraction in the RPV can be established. Pressure conditions in all vessels can be established using the purge and relief lines. The reactor power level is controlled electronically through the computer.
A number of shakedown tests will be conducted to study the characteristics of heating and cooling, steam and liquid flow from component to component, fluid motion within each com-ponent, pressure response and stability of the thermodynamic condition. These characterization tests guide the nature of establishment of initial conditions required for each test. Once the loop is characterized for both adiabatic and heated cases, the actual test will be conducted. Table 9.2 summarizes the provisions and controls associated with each component for establishing the ini-tial conditions for a test.
9.5 Preliminary Test Matrix The preliminary test matrix consists of a total of approximately 50 tests which cover a broad spectrum of LOCAs and transients [9.6]. The preliminary test matrix is divided into two phases.
Phase 1 of the preliminary test matrix consists of 17 tests - 5 base case tests (Tests 1,4,7,10, and 12),8 GE counterpart tests (Tests,2,5,8. II, and 14-17), and 4 complementary tests (Tests, 3,6,9, and 13). In Table 9.3, Phase 1 of the preliminary test matrix is listed. Four types of LOCAs and transient tests will be investigated in this phase: bottom drain line break (BDLB),
main steam line break (MSLB), GDCS line break (GDLB), feed water line break (FWLB), and loss of feed water (LOFW). Note that Tests 12 and 13 for FWLB have no counterpart tests in GIST, GIRAFFE, or PANDA. The numbers of operational components for Tests 14-17 are left in blank due to current lack ofinformation. Few repeatability tests will be carried out as a part of the Phase I tests.
In Table 9.4, Phase 2 of the preliminary test matrix is listed. These tests include sensitivity study tests and beyond design-basis accident (DB A) tests. In Table 9.4, Tests 25-50 are not listed as they have yet to be selected and are reserved to accomodate future NRC needs. Tests 18-22 are for BDLB concurrent with a single component failure. Test 23 has multiple component failure - all three vacuum breakers failed in open position. Test 24 is a station blackout test (namely, loss of all AC power) coeurrent with a PCCS unit not available for operation.
9-6 9.6 Scaling Distortion anti Potential Impact on Integral Test The present scaled model cannot represent all detailed geometrical features of the SBWR.
The model design was based on number of considerations. First, the requirements of global scal-ing were met. Then, important phenomena were identified and scaled for local scaling. In some cases, both global and local scaling cannot be satisfied simultaneously. In such cases, the requirement on global scaling was kept intact and certain compromises on local scaling were made. In addition to scaling considerations, hardware components were evaluated for ease of construction, operation and cost. Certain geometrical features in the model for some com-ponents were thus distorted. The most relevant distortion of components and their impact on the phenomena are discus. sed here.
9.6.1 RPV Core The geometry of the RPV core is not modeled exactly as in the prototype. In Table 9.5, the major geometric distortions in the core and lower plenum are shown. The impact of these distor-tions is that the flow distribution in the heated section and in the by-pass flow section will be dif-ferent from the prototype. The inventory of the fluid is also distorted in the core. This will dis-tort the flow regime and void fraction in the core, which will impact the RPV thermal hydraulics in cases where the RPV water level is near or below the top of the core. The core behaves as the void generating section. Total height of the core is about 30% that of chimney length. In the chimney, the fluid flashes as it rises. The chimney provides the largest buoyancy head to the natural circulation flow in the RPV. Hence, the distortion of the model core geometry on the natural circulation flow is negligible. Distortion of liquid inventory in the core amounts to only about 37c of the total RPV liquid inventory. Moreover, during transients, the core has at least 30% void fraction. Hence, the impact of the distorted liquid inventory in the core on RPV ther-mal hydraulics is insignificant.
In the present facility (PUMA). the four SRVs on each MSL are replaced by the one SRV equivalent to the scaled flow area of four SRVs. Also, four stub tube DPVs from RPV are replaced by two DPV lines. However, for one of the ADS stages, four SRVs open simultane-ously and for a later stage, two stub tube DPVs open simultaneously. Hence. combining these _;
lines does not affect the blowdown phenomena.
I 9.6.2 GDCS Lines The six GDCS lines are combined into three main injection lines in the present model facil-ity. One of the main injection lines is branched into two lines at the RPV, so that a single,line i
j
9-7 break can be simulated. The important phenamena to be considered are the plume spread and mixing in the downcomer. The distortion introduced by reducing the number of injection lines on the plume spread and mixing in the downcomer needs to be investigated. The present facility has a #d ratio of 2.5 compared to that of the prototype. This means that the effect of plume spread is almost 2.5 times larger in the model facility. Hence, reducing the number of injection lines by half compensates for the excess p'lume spread effect. Thus in the present model facil-ity, the reduction of six GDCS injection lines to three main injection lines has little distortion on the mixing phenomena in the downcomer region.
9.6.3 PCCS Condensation In the present 1/4-height scale facility, the PCCS condenser heat transfer rates are higher.
This distortion in PCCS condenser heat transfer will have an impact on a number of phenomena.
The PCCS condensate supply to the GDCS tank increases which,in turn, provides cooling waar to the RPV. This will accelerate the RPV coolant supply. The increased condensation rate will increase PCCS venting frequency into the suppression pool. Higher condensation in PCCS decreases the drywell pressure fa3 ter. The opening of the vacuum breaker is also accelerated.
Hence. the transient may divergt from the ideal case. This distonion can be eliminated by reducing the condensation area of tm PCCS condenser. This is implemented in the present facil-ity by wrapping the required outside sarface area of the PCCS condenser tubes (tentatively 30%
of the tubes - to be confirmed by separite effect tests) with insulating sheets of material such as Teflon , after the initial facility characterization and shakedown tests.
9.6.4 Drywell The drywell design is simplified in the present model facility. The geometric distortion in the lower and upper drywell will have an impact on phenomena such as mixing, concentration and thermal stratification. natural circulation and heat transfer to the wall. Distortion of the wall material and area of the drywell affects the heat transfer characteristic. This can be corrected by using a material determined by thermal penetration scaling methodology. The drywell is
- covered by this material to simulate the concrete. The excess surface area of the drywell can be thermally insulated to avoid excess heat loss. The mixing and stratification phenomena in the drywell was found to depend on the jet Froude number, as presented in Chapters 5 and 7. How-ever, the boundary flow of mar.s and energy is correctly scaled in the drywell. Thus, the global phenomena are preserved. The distonion of local phenomena, mixing and stratification, is not expected to significantly affect the overall integral response of the facility.
9-8 9.6.5 Suppression Pool In the suppression pool, a single vent line is drawn from the drywell. This vent line has eight openings corresponding to the eight vent lines of the prototype. This distortion in design does not impact the discharge phenomena from the horizontal vents during the initial blowdown.
However, at later stages of blowdown when the discharge is in the form of bubbles, the bubble vent phenomena may be distorted, as discussed in the scaling analysis of Chapter 5. Discharge phenomena from the sparger of the SRV have similar distortions. Also, due to the 'l/4-height scale, the bubble residence time is smaller. Thk may have some impact on the condensation of vapor containing non-condmsible gases. H . recent study of condensation of steam bub-bles containing non-condensible gases (9.7: . tes that the bubbles condense within a short distance of the discharge nozzle. Thus, the c!!ect of 1/4-height scale is not expected to be significant. The thermal stratification of the pool water may introduce significant scaling distor-tions. Rising bubbles (non-condensible) purged from the PCCS condensers induce plume, which will raise the hot water to the pool surface. The heat from the pool surface may not quickly dif- ,
fuse to the pool bottom. Hence, thermal stratification may occur. Due to 1/4-height scaling and 1/10 diameter scaling, plume dimensions are different in the present model. This will impact the degree of thermal stratification, which will have an impact on the suppression pool pressure.
Scaling considerations on this phenomena are presented in the scaling analysis (Chapter 5).
9.6.6 Helium Gas Distribution The simulation of hydrogen gas release into the drywell is carried out by injecting helium gas into the upper drywell through the bottom wall of the upper drywell. This is a distortion of initial hydrogen source location. However, present tests do not make any local measurement of helium concentration. The impact of the total helium injection on the PCCS condensation is stu-died by observing the response of the integral loop.
I
}
9-9 j
. Table 9.1 Comparison of compon%ents dimensions between SBWR and PUMA COMPONENT SBWR SOURCE OF PUMA ORIGIN I 1
BEV_
Total height (mm) 6126.3 1.D. (mm) 6000 Ref. 9.1, p.1.3-5 600 Total volume (m)) 669 Ref. 9.4 1.732- ,
Free volume (m3) 607.3 Ref. 9.1, p. 5.1-5 1.5706 Wall material Iow alloy steel Ref. 9.1, p.1.3-5 S.S Top of separator tubes 19364 Ref. 9.4 4841 Bottom of dryer skirt (mm) 4267.8 Top ofdryer skirt (mm) 5316.8 Elevation of penetrations (mm)
-GDCS equalization line 7493 Ref. 9.4 1873 3
- GDCS dmin line 9493 Ref. 9.4 2373.3-
-IC drain line 12065 Ref. 9.4 3016.3
- SDCS line 3938.8
- FW line 16455 Ref. 9.4 3999
- DPV line- 19935 Ref. 9.4 4983.8
- MSL line 20380 Ref. 9.4 5095 CORE No. of rods 43920 Ref. 9.1, p.1.3-3 38 Rod dia. (mm) 12.27 Ref. 9.1, p.1.3 25.4 Rod clad material Zr clad Ref. 9.1, p.1.3-3 S.S alloy clad Hydraulic dia. (mm) 19.73 Calculated 63.2 Flow area (m2) 7.4 Ref. 9.1, p. 4.4-6 0.153 (inside channel) 5.6 Ref. 9.4 (outside channel)
Core pressure drop (MPa) 0.048 Ref. 9.1, p. 4.4-6 0.012 Core height (mm) 3830 Ref. 9.1 p. 4.4-8 957.5 Active fuellength (mm) 2743 Table 1.3-1 685.8 Total power (at 2% power 40 Ref. 9.1, p.1.3-2 0.2 level)(MW)
Power per rod (at 2% power 0.91 Ref. 9.1, p.1.3-2,3 5.26 level)(kW) t
9-10 Table 9.1 Continued ...
COMPONENT SE_EE SOURCE OF PUMA ORIGIN Power density (at 2% power 8.61 Ref. 9.1, p.1.3-2 96.13 level)(kW/m2) '
Core shroud I.D (mm) 5150 Ref. 9.2, p. 2.1-9 520 TAF elevation (mm) 6493 Table 21.5-5-5 Sheet 1623.3 7
BAF elevation (mm) 3750 Calculated 937.5 (6493-2743)
Total fuel length + coatrol rod 6934 Table 4.4-2 1733.5 (mm)
Top of core plate elevation 880.8 (mm)
Bottom of core shroud 2146 Table 4.4-2 536.5 elevation (mm)
Core shroud O.D. (mm) 5250 Table 2.1.1-2 (Tierl) 525 Core Shroud wall thickness 50 Table 2.1.1- 2 2.5 (mm) (Tier 1)
Core downcomer width (mm) 375 Calculated 37.5 (6000-5250y2 Core downcomer annular area 6.627 Calculated 0.0 63 (m2) n/4(6.02-5.252)
Core downcomer hydraulic 750 Calculated 75 diameter (mm)
CHIMNEYSECTION l
\
Chimney shroud I.D. (mm) 4955 Ref. 9.4 502 Chimney shroud O.D. (mm) 5055 Ref. 9.4, p. 3 of 7 508 Top of chimney partition 3375.3 ;
(mm) ;
Chimney partition height 6500 Ref. 9.4 1625 (mm)
- Top of chimney (mm) 15996 Ref. 9.4 3999 Gap between TAF & bottom 110.75 ofchimney(mm)
I
9-11 Table 9.1 Continued ...
COMPONENT SBWR SOURCE OF PUMA ORIGIN Clumney wall thickness (mm) 50 Ref. 9.4 3 Chimney downcomer width 472.5 ,
Calculated 46 (mm)
Chimney downcomer area 8.205 Calculated 0.0801 2
(m )
Chimney downcomer 945 Calculated 92 hydraulic diameter (mm)
- ofdivided areas 25 Ref. 9.4 9 Dimension ofdivided areas Ref. 9.4 five 165 (mm) dia. pipes Waterlevels inside RPV measured from inside bottom (mm)
- Normal water level 18260 Ref. 9.4, Table 21.5- 4565 (NWL) 1-1, SH. 7
- Level 9 19355.5 Ref. 9.1, Table 15.0- 4838.9 1
- Level 8 18713 Ref. 9.4, Table 21.5- 4678.3 1-1, SH. 7
- Level 3 17333 Ref. 9.4, Table 21.5- 4333.3 1-1, SH. 7
- Level 2 14423 Ref. 9.4, Table 21.5- 3605.8 1-1, SH. 7
- Level 1 10423 Ref. 9.4, Table 21.5- 2605.8 1-1, SH. 7
- Level 0.5 7493 Ref. 9.4, Table 21.5- 1873.3 1-1, SH. 7 Waterlevel APlines:
- Fuel zone range: bottom 2600 Ref. 9.4. Table 21.5- 650 sense pon (mm) 1-1, SH 7
- Fuel zone range: top 20600 Ref 9.4. Table 21.5- 5100 sense pon (mm) 1-1, SH 7
- Wide range: bottom 9600 Ref. 9.4. Table 21.5- 2400 sense port level (mm) 1-1, SH 7
9-12 Table 9.1 Continued ...
COMPONENT SBWR SOURCE OF PUMA ORIGIN
- Wide range: top 20600 Ref. 9.4. Table 21.5- 5100 sense port level (mm) 1-1, SH 7
- Shutdown range: bottom 17060 Ref. 9.4. Table 21.5- 4265 sense line level (mm) l-1, SH 7
- Shutdown range: top 24612 Ref. 9.4. Table 21.5- 6150 sense line level (mm) 1-1, SH 7
- Narrow range: bottom 17060 Ref. 9.4. Table 21.5- 4265 sense port level (mm) 1-1, SH 7
- Narrow range: top 20600 Ref. 9.4. Table 21.5- 5100 sense port level (mm) 1-1, SH 7 CONI AINMENT Wall material Reinforced concrete Ref. 9.1, p.1.3-9 S.S with steelliner Upper dry well free volume 10.99 (above RPV skirt level)(m3) lower dry well free volume 2.56 3
(below RPV skirt level)(m )
Total volume (m3) 13.55 Orifice plate elevation (RPV 1525 skirt)(mm)
Top clevation (central section) 5825 (mm)
Bottom elevation (central -2250 section)(mm)
Elevation of penetrations MSL break line (mm) 20380 Ref. 9.4 5095 DPV break line(mm) 19935 Ref. 9.4 4983.8 SRVs (mm) 20380 Ref. 9.4 5095 Suppression pool vent (mm) 17200 Ref. 9.1, p. 6.2-115 4300 Vacuum breakers (mm) 17200 Ref. 9.1, p. 21-3 4300 PCCS supplyline (mm) 23300 Ref. 9.1, p. 6.2-113 5825 GDCS opening (mm) 23300 Ref. 9.1, p. 6.2-113 5825 FWL break (mm) 15996 Ref. 9.4 3699
. 1
. 9-13 Table 9.1 Continued ...
COMPONENT SBWR SOURCE OF PUMA I OF,1 GIN !
GDCS drain line break (mm) 9493 Ref. 9.4 2073.25 GDCS equalization line break 7493 Ref. 9.4 1573.25 I (mm)
ICS condensate drain line 12065 Ref. 9.4 2716.25 break (mm)
CRD line break (mm) 0 Ref. 9.4 0 RWCU/SDS line break (mm) 0 Ref. 9.4 0 Vacuum line leak (mm) 17200 Ref. 9.1, p. 21-3 4300 SUPPRESSIONPOOL Initial water volume (m3) 3255 Ref. 9.1, p. 6.2-60 8.492 Initial gas space volume (m3) 3819 Ref. 9.1, p. 6.2-60 9.193 Top elevation (mm) 16000 Ref. 9.1, 6.2-113 4000 Height (mm) 11350 Ref. 9.;, p. 6.2-113 2837.5 Pool surface area (m2) 588 Ref. 9.1, p. 6.2-60 6.133 Pool equivalent diameter (mm) 27450 Ref. 9.1, p. 6.2-60 2817 Waterlevel height (mm) 5450 Calculated 1362.5 Vertical vent area (m2) 9 - Ref. 9.1, p. 6.2-60 0.0993-Vertical channel diameter 1200 (total 8) Ref. 9.1, p. 6.2-61 356 (one (mm) channel only) .
- of horizontal vents 24 Ref. 9.1, p. 6.2-61 12 Diameter of horizontal vents 700 Ref. 9.1, p. 6.2-61 175x200 (mm)
- of vacuum breakers 3 Ref. 9.1, p. 21-9 3 Diameter ofvacuum breaker 508 Ref. 9.1, p. 6.2-15 50.8 (mm)
Elevation of penetrations w.r.t. RPV inside bottom Dry well vent (mm) 16000 Ref. 9.1, p. 6.2-117 4000 <
Vacuum breakers (mm) 16000 Ref. 9.1, p. 21-3 4000'
9-14 Table 9.1 Continued ...
COMPONENT SBWR SOURCE OF ORIGIN PUMA 16000 Ref. 9.1, p. 21-3 3562.8 SRVs (mm)
GDCS equalization lines (mm) 7493 Ref. 9.4 1872.8 Non-condensable retum line 16000 ,
Ref. 9.4 4000 Non-condensable sparger 860 Ref. 9.4 212.5 height (mm)
ICS/PCCS Pools Initial water height (mm)
- 1100 Pool height (mm)
- 1450 Bottom elevation (mm) 25300 Ref. 9.1, p. 6.2-113 6325
- 7775 Top elevation (mm)
- 2.158 Initial total water volume (m3 )
Initial gas volume (m3 )
- 0.825 Total volume ofIC/PCC
- 3.42 pools (m3)
Boiloff height above PCC
- 377.5 tubes (mm)
IC (or PCC) pool diameter Not circular 1225 (mm)
ISOL4 TION CONDENSER (10.F 3)
Number of units 3 Ref. 9.1, p. 21-5 3 Modules per unit 2 Ref. 9.1, p. 21-5 1 Tubes per Module
- 10 Total heat transfer area (m2 )
- Inside
- 0.988
- Outside
- 1.077
- 0.0128 Total flow area (m2)
Condenser tubes
- length (mm)
- 450
- 25.4
- OD (mm)
- 23.3
- ID (mm)
- Material
^
l l
l 9-15 l l
Table 9.1 Continued ... l i
COMPONENT SBWR SOURCE OF ORIGIN PUMA Headers per module: -l
-I.D. (height ofliquid
- 170 l space)(mm) ;
Volume (m )
3
- 0.0024889 i Vent nozzle size (mm) 4.949 (critical flow) 7 (non-critical flow)
- 230 centerline elevation oflower '
header from bottom ofIC pool (mm)
PCCS CONDENSER (10F3)
Number of units 3 Ref. 9.1, p. 5.4-9 3 Modules per unit 2 Ref. 9.1, p. 5.4-9 1 .
- 10 Tubes per Module 2
Total heat transfer area (m )
- 2.012
- Inside
- 2.155
- Outside Total flow area (m2) 0.0532 Condenser tubes -,
-length (mm)
- 450 l
- 50.8
- OD (mm)
- ID (mm)
- 47.5 -
- Material Headers -
- 165
- I.D. (height ofliquid '
space (mm)
- 0.001947 Volume (m2)
- 167.5 Centerline elevation oflower header form bottom ofIC Pool (mm)
.GDCS POOL Water volume (m3) 2.608 l
9-16 Table 9.1 Continued ...
COMPONENT SBWR SOURCE OF ORIGIN PUMA Normal water height (mm) 1400 Gas space volume (m3 ) 0.233 Gas space height (mm)
- 125 l Total volume (m3) 2.841 Non-drainable water volume 0.14 3
(m )
Non-drainable water height 75 (mm)
Total height (mm) 6100 Ref. 9.1, p. 6.2-115 1525 Diameter (mm) Not circular Ref. 9.1, p. 6.2-115 1540 ADS & MSL MSLlength (mm) 22865 Ref. 9.4 4431 MSL diameter (mm) 711.2 Ref. 9.1, p. 6.2-118 76.2 MSL nonje (mm) 353.822 Ref. 9.1, p. 6.2-118 25.019 MSL break size (mm) 711.2 Ref. 9.1, p. 6.3-20,21 50.289 Minor loss 6.90 Calculated 36.37 SRV line length (mm) 20320 Calculated 8000 SRV line (mm) 203.2 ( 4 each) Ref. 9.1, p. 21-39 50.8 SRV nozzle (mm) 92.202 ( 4 each) Ref. 9.4 13.056 Minor loss 7.17 Calculated 70.04 DPV (MSL)line length (mm) 4712 Ref. 9.4 4431 DPV (MSL)line diameter 304.8 Ref. 9.1, p. 21-39 76.2 (connected (mm) to MSL)
DPV (MSL) nozzle (mm) 177.876 Ref. 9.4 12.573 Minor loss 101.95 Calculated 537.50 DPV (RPV)line length (mm) 2300 Ref. 9.4 3686 DPV (RPV)line diameter 457.2 (4 each) Ref. 9.1, p. 21-39 76.2 (2 each)
(mm)
DPV nozzle (mm) 304.8 (4 each) Ref. 9.4 30.48 (2 each)
Minor loss 14.71 Calculated 113.43
9-17 Table 9.1 Continued ..
COMPONENT SBWR SOURCE OF ORIGIN PUMA IC SUPPLYLINE (1 OF3)
Length (mm) 18118 Ref. 9.4 11000 Diameter (mm) 254 Ref. 9.1, p. 21-45 38.1 Elevation change (mm) 9765 Ref. 9.4 2441.25 Minor loss 7.51 Calculated 152.08 IC CONDENSATE LINE (1OF3)
Iength (mm) 27560 Ref. 9.4 11000 Diameter (mm) 152.4 Ref. 9.1, p. 21-45 25.4 Elevation change (mm) 17635 Ref. 9.4 4408.75 Break size (mm) 146.2 Ref. 9.1, p. 6.3-20,21 50.8 Minor loss 4.8 Calculated 37.04 IC VENTLINE (1 OF 3)
Iength (mm) 4000 Diameter (mm) 12.7 Elevation change (mm) 5055 Minor loss 43.08 Calculated 85096.3 PCCS SUPPLY LINE (1 OF 3) 1.ength (mm) 7960 Ref. 9.4 7000 Diameter (mm) 254 & 203.2 Ref. 9.1, p. 21-53 38.1 Elevation change (mm) 6400 Ref. 9.4 1600 Minor loss 5.08 Calculated 25.72 PCCS VENTLINE (1 OF3)
Length (mm) 24039 Ref. 9.4 4000 Diameter (mm) 203.2 & 254 Ref. 9.1, p. 21-53 38.1 l Elevation change (mm) 16620 Ref. 9.4 4155 Minor loss 4.6 Calculated 23.29
9 18 Table 9.1 Continued ...
C.OMPONENT SBWR SOURCE OF ORIGIN PUMA PCCS CONDENSATE LINE !
1 (1OF3)
Length (mm) 8661 Ref. 9.4 10000 Diameter (mm) 101.6 & 152.4 Ref. 9.1, p. 21-53 25.4 Elevation change (mm) 490 Ref. 9.4 122.5 Minorloss 6.68 Calculated 51.54 GDCS EOUALIZATIONLINE (1OF3) ;
Iength (mm) 11432 Ref. 9.4 5500.
Diameter (mm) 152.4 Ref. 9.1, p. 21-54 12.7 Elevation change (mm) 1840 Ref. 9.4 460 Nozzle size (mm) 50.8. Ref. 9.1, p. 6.3-6 3.59 Line break size (mm) 152.4 Ref. 9.1, p.6.3-20,21 50.8 Minorloss 41.67 Calculated 20.1
. GDCS DRAINLINE (1 OF3)
Iength (mm) 25766 Ref. 9.4 6000 Diameter (mm) 203.2 & 152.4 Ref. 9.1, p. 21-54 38.1 *
^
Elevation change (mm) 8107 Ref. 9.4 2026.75 Line nozzle size (mm) 76.2 Ref. 9.1, p. 6.3-6 5.385
.Line break size (mm) 152.4 Ref. 9.1, p. 6.3-20,21 50.8 Minorloss 13.59 Calculated 132.72 FEEDWATER LINE (1OF2) length (mm)
- 4000 Diameter (mm) 355.6 Ref. 9.1, p. 21-41 38.1 Line break size (mm) 227.7 Ref. 9.1, p. 6.3-20,21 50.8
. sparger reducing tee (mm) 127 Ref 9.5, p. 3 12.7 Sparger nozzles I.D. (mm) 7.37 Ref. 9.5, p. 3 4.763 Minorloss 7.37 Calculated 9.71 AUXILIARYLINES CRD line(mm) 31.75 Ref. 9.1, p. 21-35 25.4 Bottom break line (mm) 50.8 Ref. 9.1, p. 21-34 50.8
9-19 .
Table 9.1 Continued ...
COMPONENT SBWR SOURCE OF ORIGIN PUMA RWCU/SDCline (nun) 76.2 Ref. 9.1, p. 21-46 12.7 RWCU/SDCline (side of 203.2 Ref. 9.1, p. 21-46 25.4 RPV)(nun)
Line break size (mm) 193.6 .
Ref. 9.1, p.6.3-20,21 50.8
- Infonnation available from GE Nuclear Energy 1
1 i
i 1
l l
l
.1
9-20 Table 9.2 Controls and Provisions on Various Components of PUMA Control / Provision Component Liquid Injection and Drain Line RPV, Drywell, SP, GDCS, PCCS
, and ICS pools Steam Injection and Relief Valve Systems RPV, Drywell, SP Liquid Level Measumment RPV, Drywell, SP, GDCS, PCCS/ICS pools Temperature and Pressure Measurement RPV, Drywell, SP, at Different Height GDCS, PCCS/ICS pools Non-Condensable Concentration Measurement Drywell, SP, ICS/PCCS Condensers Void Fraction Measurement RPV Heater Power RPV Core 400 kW, SP 50 kW, ICS/PCCS pools - 30 kW each Timer Controlled by Computer ADS (SRVs and DPVs)
9-21 Table 9.3. Phase 1 of the Preliminary Test Matrix - Base Case and GE Counterpart Tests Operational Components Test Event PCCS ICS GDCS . DPV VB EOLN DWS WWS Lines 1 (Base) MSLB 3 3 6 6 3 3 0 0 2* MSLB 0 0 ,
4 6 3 3 0 0 (GIST B01) 3 MSLB 3 3 4 6 3 3 0 0 4 (Base) BDLB 3 3 6 6 3 3 0 0 5* BDLB 0 0 4 6 3 3 0 0 (GIST A07) 6 BDLB 3 3 4- 6 3 3 0 0 7(Base) GDLB 3 3 5 3 3 0 0 0 8* GDLB 0 0 4 6 3 3 0 0~
(GIST C01A) 9 GDLB 3 3 4 6 3 3 0 0' 10 (Base) LOFW 3 3 6 6 3 3 0 0 )
1 11* .LOFW 0 0 4 6 3 3 0 0 l (GIST D03A) )
12 (Base) FWLB 3 3 6 6 3 3 0 0 13* FWLB 0 0 4 6 3 3 0 0 l
14 !
MSLB (GIRAFFE / PANDA) 15 BDLB (GIRAFFE) 16 GDLB (GIRAFFE / PANDA) 17 ICRLB l
- Test will be terminated when a temperature or pressure set point is reached to prevent damage, i VB = vacuum breaker between drywell and wetwell, EQLN = equalization line connecting suppression pool to the vessel (e.g.,3 means all three equalization lines will be open if actuated automatically or manually), DWS = drywell spray, WWS = wetwel spray, MSLB = main steam line break, BDLB = bottom drain line break, GDLB = GDCS line break, LOFW = loss of feed water, FWLB = feedwater line break, ICRLB = isolation condenser condensate retum line break.
No information is currently available regarding the number of operational components in the GIRAFFE and PANDA tests
+
9 22 Table 9.4 Phase 2 of the Preliminary Test Matrix - Sensitivity Study and Beyond DBA Tests Operational Components Test Event PCCS ICS GDCS DPV VB EOLN DWS WWS Lines 18 BDLB 3 3 6 5 3 3 0 0 19 BDLB 3 3 5 6 3 3 0 0 20 BDLB 3 3 6 6 3 2 0 0
{
21 BDLB 3 3 6 6 2 3 0 0 (1 VB failed in open posithn) 22 BDLB 3 3 6 6 2 3 0 0 (1 VB failed in open position) 23 BDLB 3 3 6 6 0 3 0 0 (all 3 VBs failed in open position) ;
24 Blackout 2 3 4 6 3 3 0 0 l In addition to the above tests, the following tests will be selected-
- 1. Several additional tests with multiple component failure.
- 2. A few tests to assess the impact of non-safety systems upon GDCS and PCCS (e.g., control rod drive flow or RWCU/SDCS flow on GDCS performance, drywell spray on PCCS and ,
GDCS performance, wetwell spray on suppression pool flow to the vessel via equalization !
line).
- 3. A few tests at different break sizes (e.g.,50% of BDLB). ;
- 4. A few tests to assess natural circulation flow characterization by measuring core flow as a !
function of power, downcomer water level, and vessel pressure (including the determination l of any flow oscillation or instability).
- 5. Several sensitivity tests by varying core power or other PIRT-identified important parameters.
- 6. A few helium tests to investighte the presence of hydrogen on PCCS performance.
- 7. A few repeatability tests.
i l
l
9-23 Table 9.5 Geometric Distortion in RPV Ideal Scale Item SBWR Model PUMA Distortion Lower Plenum 71.4 .
0.1785 0.2077 +16.4%
3 Free Volume (m )
Lower Plenum 16.51 0.1651 0.1972 +19.5%
Average Flow Area (m2 )
3 Core Free Volume (m ) 39.7 0.0992 0.1209 +21.8%
2 Com Flow Area (m ) 13 0.13 0.1584 +21.8%
Chimney Free Volume (m3 ) 171.46 0.4367 0.4370 0%
3 Downcomer Volume (m ) 73.85 0.1846 0.1846 0%
Core and Chimney 211.16 0.5279 0.5579 +5.7%
3 Free Volume (m )
Downcomer and Lower 145.25 0.3631 0.3923 +7.4%
3 Plenum Free Volume (m )
9 i
R PCCS/tCS POOLS (TOTAL OF 2)
_ VENT 1 PCCS/ICS CONDENSER $
_h6 ' '
7 (3 EACH)
PCCS/ICS MAIN STEAM LtNE PCCS STEAM CDCS POOLS PCCS DRAIN NE ES --[] [}/% g ,[ -CONDENSABLE 3 SU{3 )
REACTOR
$ $ $ 8 PRESSURE VESSEL
~ r' ICS STEAM O - - b: s UPPER DRY WELL f3EACH) DRYWELL/
WETWELL DPV UNE f- .
' ! VENT (2 EACH) -
JN rs 4 @ [
l es g
{
\
~~~~
ICS DRAIN UNE 8 FEEDWATER VACUUM S LRE 9 f (3 EACH) BREAKER o x UNES - 2 9
GDCS DRAIN UNE (3 EACH) $
b (3 EACH)
SRV UNE (2 EACH) M ROCU/SDCS l- &'
UNE 3 p~ "
3 GDCS EQUAUZATION LINE r-(3 EACH) 1-ANNULAR W cROUND LEVEL 3 ROWS OF
-@ CRD UNE 8 RADIAL SLOTS i
(175 X 22 mm)
LOWER [ '
!(((((((((( [ [ CONTAINMENT pit h
t/a/a////A Figure 9.1 'Overall schematic of PUMA facility e
~
- e l
/
N O
I S /
U S
L O
SL /
PP / A M
U P
i n
l s
/ e s
s e
v
/ i n
a m
f o
w e
L L
/ i v
n E a W l P
Y R
D R
E
/ 2 W 9 R O E L e r
P u P g U i F
/
/
/
S C
/
D G
V P
R
/
.l.l. I*l.I'I Isl.l.l.
/
9 26 References 9.1 SBWR Standard Safety Analysis Report (SSAR), GE Nuclear Energy,25A5113 Rev. A.
August 1992.
l 9.2 SBWR Tier 1 Design Certification Document, GE Nuclear Energy, 25A5354 Rev. A, February 1993.
9.3 Private Communication with Brookhaven National Laboratory (BNL) Staff, January 1994.
i 9.4 U.S. Nuclear Regulatory Commission Request for Proposal No. RS-RES-93-049, .
November 25,1992.
9.5 GE Nuclear Energy Report: MFN No.170-93, Attachment B.
9.6 Han, J. T., Bessett, D. E., Shotkin, L. M., "NRC Confirmatory Testing Program for SBWR," Proceedings of the Twenty-first Water Reactor Safety Information Meeting.
October 25-27, Bethesda, Maryland (1993).
9.7 Liang, K-S," Experimental and Analytical Study of Direct Contact Condensation of Steam in Water," Ph.D. Thesis, MIT (1994).
l l
1 e
l l
1 1
10-1
- 10.
SUMMARY
AND CONCLUSIONS The U.S. NRC requires independent confirmatory data for the General Electric Nuclear Energy (GE) Simplified Boiling Water Reactor (SBWR) which can be used to assess the uncer- -
tainities associated with the use of the RELAPS/CONTAIN code for simulation of SBWR safety system performance under accident and operational transient modes. In order to satisfy this requirement, the NRC has awarded a contract to the Purdue University School of Nuclear Engineering to obtain this data through the design, construction, and operation of a scaled -
integral test facility named the Purdue University Multi-dimensional Integral Test Assembly (PUMA).
The major objectives of the confirmatory research program are to:
. provide integral data to NRC for the assessment of the RELAP5/CONTAIN code for SBWR -
application
. assess the integral perfonnance of the GDCS and PCCS
. assess SBWR phenomena important to LOCAs and other transients These objectives are being accomplished by the following sequence of tasks: ,
. develop a well balanced and justifiable scaling approach for use in design of the PUMA facility.
. identify important SBWR phenomena associated with LOCAs and applicable operational transients and, based on the identified phenomena, establish a test matiix
. design the scaled integral test facility, including necessary and sufficient instrumentation
. constmet the scaled integral facility
. develop boundary and initial conditions and procedures for conducting the integral tests based on the scaling methods and by using computer code calculations (RELAP5/CONTAIN)
. perform the required integral system tests under strict quality control procedures and provide associated documentation The scaling methods that have been developed for the purposes of the PUMA facility design and the results of the scientific design analyses are documented in this reporf.
The necessary scaling relationships for design of a full pressure, but less than full neight scaled facility have been developed. The criteria that have been used for scaling are to maintain similarity of: integral system response (loops), interaction between components, and preserva-tion of local phenomena. The results of this development show that all of the objectives of the -
, , . ~ - :- <
10-2 program can be achieved by use of a full pressure but reduced-height scaled facility.
Parametric pmliminary design studies show that a 1/4 height scaled facility could best meet all the program objectives without compromising either geomenic similarity or preservation of the important two-phase phenomena. The 1/4 height scaled facility is easily constructable and satisfies the geometrical requirements in order to preserve important thennal-hydraulic phenomena.
The details of the PUMA scientific design, and the supporting analysis are documented herein. The basic design featun:s 1/4 height and 1/400 volume scales which result in a 1/2 time scale. The facility has a reactor pmssure vessel height of 6 m and 0.6 m diameter. The system contains all the important safety and non-safety systems of the SBWR. The facility is designed to simulate the phenomena at and below 1.03 MPa (150 psi) fol' awing SBWR scram. The pres-sure is scaled 1:1 and the power is scaled by 1/200 of prototype. The facility is designed with 340 instrumentation devices to measure pressure, temperature, flow rate, local void fraction, two-phase flow void fraction and non-condensible gas concen7ation. The site for the facility has been identified and the preliminary layout of the facility is completed. Purdue has contracted the mechanical design and construction of the PUMA facility to an engineering firm (Phoenix Solu-tions Company, Minneapolis, MN).
In conclusion, all program objectives can be achieved using a 1/4 height scaled facility. The resulting design fits nicely within the range of scaled national and intemational SBWR experi-mental facilities that either exist or are in the construction phase. This will provide the ability to cross check the system behavior for important thermal hydraulic phenomena. These characteris-tics will help increase the accuracy of uncertainity assessments that will be based on data from this facility.
l 1
1