ML091940277
| ML091940277 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 07/09/2009 |
| From: | Luminant Generation Co |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| GL-04-002 | |
| Download: ML091940277 (45) | |
Text
a Luminant NRC -Luminant Power Public Meeting Comanche Peak Nuclear Power Plant Emergency Sump Performance July 9 th 2009 Overview* This presentation provides clarifications and additional information to support the supplemental response to GL 2004-02" Focused on the NRC draft request for additional information
[RAI #s noted]" Conservatisms which offset uncertainties are noted in bold" Supplemental response will be revised 1 GSI-191 Challenge Analyst 3 GSI-191 Sumpology 101 SPTF Downstream Effects m 4 2 Modifications" Flat sump screens replaced with large complex strainers.Sump switchover setpoints lowered* Motor operated isolation valves replaced , Debris screens and strainers provided for drains in the refueling cavity* Drain holes added to the reactor vessel head stand shield wall" Modified features to minimize water holdup* Debris interceptor (curb) provided around strainers* Water control feature added to optimize sump performance
- ECCS and CSS pump suction pressure monitoring instrumentation upgraded to meet Regulatory Guide 1.97, Revision 2.5 Schedule -Specify, design, build and install new strainers before completion of analyses 0 -" 6!3 Flow Control Core Tube RAI 22 and 23 8 4 Strainer and I .Layout The sump pit is self venting through the strainers.
There are no vents to containment above the top of the strainers.
[RAI 21]-,o, o , o ..,:! 12 inch tall interceptor
.. ."[RAI 17]LFVATION VIFW '7-7'Revised calculation for ECCS Recirc for SBLOCA -810.56 ft. [RAI 19/22/23]* ,a toll 2 /w~m~i £I-~YVf. P-4f.2-4* ', "
- f' " 12 inch tall ,_ -- -interceptor I .(F WW11TiA[RAI 17]eIARI PLNM VIEi 5 New Sump Strainer with Debris Interceptor and Trash Rack RAI 17 New Sump Strainer L1 121 RAI 17 12 6 Computational Fluid Dynamics Model R~g- I I qpAr 4t Drin Sopig Dr~nsgp R~lC.cn Nb~ ' \\ ' dd.4S*mWr MAMAW..d 1. 1.11 e ...4! Mpdw6d g .k MVý. Modw4 d R.M o VWfltFidng olu" F1r anSm Took, B RAI 11 1 Capped Equipment Drain near sumps RAI 11 14 RAI 11 14 7 Normal sump drain near strainer RAI 11 1ii5: RAI 11 15 Flashing Mod RAI 11 1,6 8 Streamline for Loop 4 LOCA (2-train)Equipment hatch ng cavity rains , Inactive sump Refueli dr Z-- trainers 17 Solid plate on sump outboard ends 18 9 Refueling Cavity Drains Drain Strainer RAI 32 20 10 Upender Area Fuel Handling Bridge Crane32 RAI 32 a 21 Refueling Cavity Drain Strainer RAI 32 22 11 Upender Area Drainage* Maximum spray flow into the upender area (1.7% of upper containment) is 185 gpm.Strainer is 1.8% size of sump strainer.
Debris load on the drain strainer would be bounded by the debris load on the sump strainer.* The flooding calculation assumed 2 ft of holdup (< 400 ft 3) The CSHL at 250 gpm is 0.042 ft.10.8" submergence
-the 30 day debris laden head loss for the sump strainers is approximately 7.2 inches* No credit for the drain strainers were taken to reduce sump debris loads I RAI 32 I Strainer Prototype Testing" Informative testing based on BWR protocol" Demonstrated LBLOCA ECCS switchover with partial submergence" Demonstrated relationship of LOCA versus Secondary Line Breaks* Demonstrated settling is prototypical and would occur for most debris RAI 13 12 Prototype (small flume) Testing* Ultra-Conservative test temperature
-< 50 F* Fiberglass fibers settle in 20 to 60 minutes in 50 F water versus 20 to 30 seconds in 120 F water[NUREG/CR-2982, "...water temperature has a paramount effect on buoyancy..."]
- Tests conducted at 128 F and 169 F confirmed the effect on settling of fiberglass
[Test Report No. ITR-92-03N]* Fibrous debris was consistent with NUREG/CR-6808." Demonstrated settling with low approach velocity is prototypical and would occur for fines RAI 13 25 Prototype Fiber Testing 13 Free Floating Fines and Smalls Settled Fibers 28 14 Settled Debris During Head Loss Test Recirculation with partial submergence
- The strainer was designed for a transient water level during switchover
-LBLOCA 3.12 ft. At 4900 gpm-SBLOCA 2.56 ft. at 400 to 1200 gpm* The core tube would have > 9 inches of submergence at initiation
- Water level would be rising at over 1 inch per minute for all break sizes" The strainer would be fully submerged in less than 15 minutes (One train)30 15 Flood Up Transient* Water level at ECCS switchover.
All debris introduced at the prototype prior to pump start. Debris mixed at start.* Initial flow at maximum ECCS (39.5% of design flow)* Flood up at minimum flood rate was 25 min. (< 1 pool turnover).
- 100% flow at CSS switchover flood level.RAI 13 31 Transient Flood Up Testing RAI 13 32 16 Transient Flood Up Testing Transient Flood Up Testing RAI 13 34 PAl 13 34 17 Secondary Line Breaks* Secondary lines break have flow rate of 60.5% of design versus 12,420 gpm)a maximum (7520 gpm* The maximum mission time is less than 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> versus 30 days for LOCA.* Debris Head loss was 0.009 for LOCA ft. vs 0.482 RAI 24 35 Main Steam Line Break Testing -Drain Down 36 18 Reflective Metal Insulation" Debris Generation and Transport based on Diamond Power Reflective Metal Insulation." 11,269 ft 2 small/2,072 ft 2 large Sump A (2 train)* Unit 1 SG replacement with Transco RMI bounded" No credit for the scavenging of fibers by RMI was taken in analysis or testing.[Low Density Fiberglass Insulation
- For debris generation and transport, Nukon low density fiberglass
[2.4 lbm/ft3] was assumed to calculate the volume* For strainer design debris load, the maximum density of 4.9 Ibm/ft3 was assumed* 100% of the LDFG in each loop room was assumed to be within the ZOI* No credit for shadowing by grating and robust structures was taken 19 Small Fiberglass Holdup in Upper Containment" Based on DDTS testing, 10% of the small fiberglass debris in upper containment was credited for holdup on grating (7% debris on grating and 3% washing off operating deck through grating).No credit taken for holdup of fiberglass fines, or retention of small pieces on concrete floors, refueling canal, stairs, or any other structures besides RCS loop room grating..Removing a small amount of this conservatism would reduce the overall washdown transport fraction even if no credit is taken for holdup on grating.RAI 7 39% V~1RAI 7 LOOPS 1 & 4 LOOPS 2 & 3 40 20 V Low Density Fiberglass Insulation ZOI Transported*
Case LDFG 42.42 ft 3 33.87 ft 3 (Loop 4)Fines 7.16 ft 3 6.66 ft 3 Loop 4, 1 train 32.6 Ibm Smalls 29.01 ft 3 22.63 ft 3 Loop 4, 1 train 110.9 Ibm Large 3.03 ft 3 2.28 ft 3 Loop 4, 2 train 11.2 Ibm Jacketed 3.22 ft 3 2.30 ft 3 Loop 4, 2 train 1_ 11.3 Ibm* (166 Ibm) = 69 ft 3 NUKON for testing I 41 LDFG Debris Preparation" NUKON processed into fines representative of either eroded or latent fibrous debris and 'fines/smalls' by recognized mechanical process devices [i.e., chipper (smalis) & Munson shredder machine (fines)]" Sample of latent, fines/smalls, and larges' were provided to the Staff before any Large Flume Testing was initiated and were found to be representative of what the NRC Staff had expected, as long as they are adequately diluted* Fibrous debris has been processed, prepared, and introduced in accordance with the PCI white paper Sure-Flow Suction Strainer -Testing Debris Preparation
& Surrogates, the PCI/AREVA/Alden Large Flume Test Protocol which have been provided to and discussed with the NRC Staff* Observations and comments by the NRC Staff and lessons learned by PCI/AREVA/Alden during the initial Large Flume Test for Wolf Creek/Callaway were incorporated into all subsequent tests RAI 25 42 21 Fines diluted per the March 2008 Staff guidance LUI-U t-ines RAI 25 and 26 43 Latent Debris" 200 Ibm (70% particulate/30%
fiber) conservatively assumed" No settling in the recirculation pool credited* Pool fill transport excluded upper containment.
-inactive sump conservatively limited to 15%-active sumps evaluated to be 9%" Testing used bounding case of 80% transport to one sump" Latent fibers introduced on surface of test flume 5 minutes prior to start of flow. Head loss prior to first batch was comparable to fiber only test with fines RAI 9, 12, and 15 44 22 Robust design -Weld are continuous.
The end profile is low.Maximum specified Min-K thickness
0.5 inches
assumed.RAI 1 45 Seamless -no open gaps RAI 1 46 23 NUREG-6808
-Air Jet Impact Testing* Transco RMI with sheaths half the thickness of 0.050 was tested. No RMI sheath failed during testing. (App. B p. 5)" The tested Transco RMI had rivets and spot welds (3" max) in lieu of continuous welds" The RMI failures occurred due to separation of the outer sheath from the ends. RMI testing targeted the seams and joints" Secondary effects were considered insignificant" The ZOI for Transco RMI is conservative for Comanche Peak RAIl 4 Min-K Insulation Min-K ZOI Transported Case Fines 0.56 ft3 0.52 ft3 Surge Line, 1 8.2 Ibm train 100% fines assumed within the ZOI Pulverized Min-K used for testing RAI 3 24 Lead Shielding Blankets" The lead fibers were previously shown to not be transportable in testing and were not tested. No credit for fiber scavenging was taken." The Comanche Peak permanent lead blankets are the same blankets tested at Wyle." Wyle testing bounds the Comanche Peak configuration.
RAI 2 49 Lead Shielding Blankets -CPNPP RAI 2 50 RAI 2 50 25 Lead Shielding Blankets -Wyle Test 2 RAI 2 51 RAI 2 51 Lead Shielding Blankets -Wyle Test 2 RAI 2 52 RAI 2 52 26 Lead Shielding Blankets -Wyle Test 2 Lead Shielding Blankets -Wyle Test 2 RAI 2 54 27 Lead Blanket Cover Fines Size consistent with LDFG smalls RAI 2 and 27 55 Lead Shielding Blankets" All layers within the ZOI were included." The largest break, Loop 1 Cold Leg, was used for testing.* Fibers from the blanket cover were observed to reach the strainer during fiber bypass testing.* The fiberglass content was included in the chemical effects analysis.RAI 2 56 28 Coatings" All containment coatings were applied and maintained under either the Comanche Peak 10 CFR 50, Appendix B QA program or the Comanche Peak Non-Appendix B QA program." The reevaluation of all coatings inside containment was based on ASTM D 5144-00, the EPRI Guideline on Nuclear Safety Related Coatings using plant records." Where records were insufficient, sampling and testing were performed (e.g. material traceability).
[RAI 30 Coatings within the ZOI" ZOI of 4D used for acceptable concrete epoxy coatings based on WCAP-1 6568-P and JOGAR testing." ZOI of 1 OD used for acceptable steel epoxy coatings in lieu of 4D for conservatism." Pulverized acrylic coatings used for testing surrogate for epoxy" Tin Powder used for IOZ surrogate p 29 i~s ~Unqualified Coatings" Indeterminate coatings classified as unqualified." Size distribution for analysis and head loss testing based on test data.* Testing proved that paint chips >/= 1/64 inches cannot transport to and block the strainer.[Chemical Precipitates
-Analysis* Analysis of chemical effects were completed in accordance with WCAP-16530-NP,"Evaluation of Post-Accident Chemical Effects in Containment Sump Fluids to Support GSI-191," Revision 0, February 2006.* No WCAP-1 6785 refinements were used.* The current Sodium Hydroxide
[NaOH] buffer concentrations were used. There are no current plans to implement the buffer reduction which was previously proposed as a contingency.
RAI 34 and 36 30 V r Chemical Precipitates
-Testing* Although Sodium Aluminum Silicate (NaAlSi308) makes up 83% of the precipitate, Aluminum Oxyhydroxide (AIOOH) was used as the surrogate for testing." For head loss testing, WCAP-1 6530 was applied to generate the AIOOH precipitates.
Acceptance criteria for AIOOH batches generated were based on the settling characteristics of chemical precipitates generated at 2.2 g/l.* AIOOH precipitates generated were introduced within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of their generation
/ acceptance for use.RAI 34 Chemical Precipitates
-Testing The chemical concentration in the test flume was controlled over time so as to prevent the potentially accelerated settling of chemical precipitates from an over concentrated flow stream. This was in accord with the protocol reviewed with the NRC staff prior to testing.* WCAP-16785 was not used.* Calculations of head loss including chemical precipitates were performed at both 200 F and 120 F RAI 34 31 Supplementary Strainer Testing Loop 4 LOCA Train B I RAI 16 Supplementary Strainer Testing Loop 4 LOCA Train A unst_iectors_ovef_combined bmp 2007-10-25 CPSES Case 4-EF RAI 16 64 32 Test Mpdule -approach velocity 0.0073 fps/'109.5 ft 2 , net scaling factor 2.9225%RAI 16 65 Debris interceptor
-12 inches tall, 6 inch top Af aiq 0.18 fps test vs 0.12 fps avg RAI 16 66¸RAI 16 66 33 Transport flume -264 inches (22 feet)Max flow -0.62 fps RAI 18 67 Test Protocol* Test debris types were introduced separately with the most transportable first.* Although classified as fines, the lead blanket covers were characteristic of smalls but less transportable.
Therefore, they were added after LDFG smalls.* Particulate fines were introduced before fibrous debris for conservatism.
RAI 27 68 34 Head loss test -Debris head loss stable at approximately 0.6 ft Teot 4 Design Basis Tet Hlead Lom 0 7 -. .-.. -.. -----..............
06.Flow sweep 2 .............
... shows direct relationship 0 .between flow and dp 000 W 0000 20o00 30000 4W000 50000 60000 70000 TlA 29ln)RAI 29 6 Test 4 Head Loss Extrapolation Curve Exponential
-0.7497 ft.Head Loss (H) Vs. Tims (T): CP Flume 2--- ----OC-5 O"i-pj; II T T , )'A S :..D 4~-444 10(2' 15lI IWO. 20 .1 I 401.5 400 U01.1 Tim. 4sec.1 RAl 28 70 35 Test 4 Head Loss Extrapolation Curve Linear -4.2552 ft.*Head Loss (41) Vs. Time [T[ : CP Flume 2 Oe5I.z I. ~+/--13 5-.-,0itrt-n.-,, Fl It),1*49 XIt !' k2 C .)Ol Tne ae.RAI 28* Not used in analysis 71 Head Loss Margins* Exponential
-Net Positive Suction Head margin 5 Ft. at 30 days* Linear Days Extrapolation NPSH...._ _Margin Non- 14 2.331 ft 3.63 ft Harsh Ambient 25 3.654 ft 2.306 ft 72 36 Headloss test -drain down 73 Headloss test -drain down I est If rleauolusS 74 37 Light debris load/open holes -top half 75 Heavier debris load on bottom half Test 4 Headloss 76 38 Settled fines on top of interceptor 77 Miscellaneous Debris" Miscellaneous debris was tested for transport at both Alion and Alden test facilities to determine the margin in the 200 ft 2 sacrificial area." Electromark Series 1000 and 3000 labels were shown to be qualified (would not detach) or acceptable (would not transport or block the strainer).
- Other acceptable materials
-Lamacoid RAI 5 78 39 Miscellaneous Debris Testing* Separate flumes were used for miscellaneous debris testing because the transport over the debris interceptor was 0.18 fps as opposed to 0.12 fps in the plant. The average transport velocity in front of the debris interceptor is 0.08 fps; therefore, a minimum of 0.1 fps was used for testing.Some label testing was done in the main flume.RAI 14 79 Series 1000 Electromark labels RAI 5 840 40 Series 1000 Electromark labels, 6"x6" clear vinyl Test 3 Labels RAI 5 81 Series 1000 Electromark la els, 6"x6" white vinyl Test 3 Labels RAI 5 82 41 Series 1000 Electromark labels, 6"x16" clear vinyl RAI 5 83 RAI 5 83 Floating Debris* Although most miscellaneous debris sank readily in the cold flume water, duct tape, bumper sticker tape and radiation tape floated.* Based on previous testing of the same duct tape at Alion, it was concluded that the larger pieces of tape could entrain air during boiling.* The floating debris was considered unacceptable and included in the sacrificial area penalty.RAI 19 84 42 Vinyl labels and tags settled at 0.1 fps RAI 5 85 RAI 5 85 Unacceptable labels, tags, and tape" The impact on the sacrificial area margin was calculated to be the area equivalent to 75% of the total of the original single sided surface area of the unacceptable labels, tags, and tape (per SER) plus 20% for uncertainty." The sacrificial area penalty is 31.9 ft 2 for Unit 1 and 34.6 ft 2 for Unit 2" Paper tags are less than 22% of this total and represent an insignificant amount of fiber.RAI 4 and 5 86 43 NMI INIIIIIII In Vessel Downstream Effects Debris WCAP-16793-NP, R1 CPNPP Fiber 0.33 lb 0.02 lb Particulate 29 lb 98.78 lb Chemical 13 lb 1.26 lb Calcium silicate 6 lb N/A Microporous 3.2 lb 0.03 Insulation Comparison to the WCAP test data concluded Comanche Peak demonstrates reasonable-assurance-of-Long-Term Core.CoolIing.
RA133 Holistic Case" Conservatisms have been inherent in each phase of the analysis and testing to account for uncertainties" Debris generation and transport conservatisms assure that only debris that cannot get to the sump is excluded" Testing conservatisms assure that only debris that DG and DT show cannot get to the sump is excluded from testing 44 RAI Discussions
- RAI 9, Debris transport analysis, Pool Fill -Alion" RAI 6, Fiber erosion testing -Alion" RAI 10 and 11, Test Flume, velocity and turbulence questions
-Alden" RAI 15, Test protocol, latent fiber issue -PCI* RAI 8, Test protocol, erosion issue -PCI* RAI 20, Test Flume, fiberglass larges question -AREVA* RAI 22 and 23, SBLOCA, vortex, air ingestion, void fraction, and flashing issues -PCI/AREVA" RAI 24 and 37, Secondary line breaks, debris load and testing -Luminant Summary" Comanche Peak has been a leading participant in the pursuit to resolve GSI-191 since 2000* Comanche Peak has collaborated with industry groups seeking knowledge and data required to address technical issues." Extensive analysis and testing are complete that demonstrate with reasonable assurance that the ECCS and CSS recirculation functions under debris loading conditions are in compliance with the regulatory requirements delineated in Generic Letter 2004-02.[45