LIC-99-0002, Forwards Response to RAI Re GL 97-04, Assurance of Sufficient Net Positive Suction Head for ECC & Containment Heat Removal Pumps

Forwards Response to RAI Re GL 97-04, Assurance of Sufficient Net Positive Suction Head for ECC & Containment Heat Removal Pumps

ML20199D850
Person / Time
Site:	Fort Calhoun
Issue date:	01/05/1999
From:	Gambhir S OMAHA PUBLIC POWER DISTRICT
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
GL-97-04, GL-97-4, LIC-99-0002, LIC-99-2, TAC-M99992, NUDOCS 9901200229
Download: ML20199D850 (10)

Text

- _ _ - _ - - _ - - _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . . --

?.

0998 Omaha PWic Power Ostnct 444 South 16th Street Mat Onaha, Nebraska 681@2247 January 5,1999 LIC-99-0002 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Mail Station P1-137 Washington, DC 20555

References:

1. Docket No. 50-285

2. Letter to OPPD (S. K. Gambhir) from NRC (L. R. Wharton) dated December 7,1998

Subject:

Response to Request for Additional Information Related to GL 97-04,

" Assurance of Sufficient Net Positive Suction Head for Emergency Core Cooling and Containment Heat Removal Pumps" (TAC No. M99992)

The Reference 2 letter documented information requested by the NRC Staff sad discussed during several telephor,a calls conceming Omaha Public Power District (OPPD) submittals in response to GL 9. 04 Enclosed are the OPPD responses to the information requests.

Please contact me if you have any additional questions.

Si carely,

'M k

S. K. Gambhir Division Manager Nuclear Operations TCM/ tem '

l I

c: E.W. Merschoff, NRC Regional Administrator, Region IV ,

L.R. Wharton, NRC Project Manager W.C. Walker, NRC Senior Resident inspector Winston and Strawn 9901200229 990105 PDR ADOCK 05000285 p PDR I

psm tmooyment utn taa opportunn

~. ._. _ . _ . _ _ _

i ,

LIC-99-0002 ,

Enclosure Page 1 ~

Omaha Public Power District Responses to NRC Request for Additiona'l Information Generic Letter 97-04 ,

i.

REFERENCES:

(1) ABB IOC PSA-75-013, dated January 23,1975, Verification of the CONTRANS Digital Computer Code (2) ABB Topical Report Number CENPD-140-A, dated June 1976, Description of the CONTRANS Digital Computer Code for Containment Pressure and Temperature Transient Analysis (3)- USNRC letter from Olan D. Parr to A. E. Scherer, Topical Report Evaluation of CONTRANS code NRC REQUESTS:

1. In calculation O-MECH-CALC-021 Rev. 00, Appendix D, explain the difference between I the CONTRANS analyses for cases AVOC and OSNVBV97.

l i

Response:  ;

Case AVOC is a LOCA peak containrpent pressure case in which its inputs are biased to )

w . maximize the containment pressure / temperature response. Case OSNVBV97 is a maximum i safeguards case. A comparison of the inputs and assumptions for these two cases is included in Table 1.

2. Please confirm that only case OSNVBV97is being used to calculate the subcooling  ;

available and explain why that is acceptable. l Response: I Case OSNVBV97 is the case from which the subcooling credit is extracted. The following information is provided to clarify the GL 97-04 response. When calculation O-MECH-CALC-021 Rev. 00 was done, ABB-CE reviewed the large LOCA containment transient analyses available at the time. The results of that review were provided in ABB-CE memo OPS-92-0653, which is included in Appendix D to 0-MECH-CALC-021 Rev. 00.' The memo states, The CONTRANS analysis mns chosen from References (1) and (2) are either tha analysis of record, Case

1. AVOC (Reference 3), orprovide a scenario where the sump temperature would be maximized

' and the containment pressure minimized, Case #AGTL and Case #0SNVBV97 (Reference 1 and 2)." Reference (2)is ABB-CE calculation 002-NT90-C-012 Rev. 00, which addresses 26

~!

t

~

l l

LIC-99-0002 Enclosure l

Page 2 different large LOCA analytical cases (including AVOC and AGTL). Reference (1)is ABB-CE calculation 002-AS91-C-001 Rev. 00, which includes largo LOCA case OSNVBV97. The ABB- i CE memo indicates that of the 27 large LOCA containment transient analysis cases available for l review at the time, case OSNVBV97 had the lowest sump subcooling and at that time it was identified as a limiting case. With respect to LOCA, ABB-CE and OPPD only have containment analyses for the large break event for Fort Calhoun Station.

CONTRANS is a four-region model, with Region 11 being the containment sump. Due to the use of the containment sprays and shutdown cooling heat exchanger (s) during the recirculation phase of a LOCA, the sump is an integral part of the containment model, which must be valid to j support the calculation of the containment pressure and temperature response. The degree of l

conservatism in the CONTRANS model is a function of the inputs chosen for the particular l analysis. Thus, due to the flexibility of these various inputs, CONTRANS can be biased to suit l whatever containment application may be required. )

i

3. O-MECH-CALC-021 Rev. 00 assumes that the height of the waterin the sump is 6 feet l l based on Reference 3.11. However, page 4 of 25 of the current USAR Section 6.2 only l l assumes 2.8 feet of water. Explain why there is a difference in assumed height of water in the sump.

Response

The 996.8' recirculation phase water elevation used in USAR 6.2.1 is an assumption in calculation O-SEC-23 done by Combustion Engineering in 1968 (i.e., before the containment was built). That calculation formed the basis for the recirculation phase NPSH information in i section 6.2.1 of the 1971 Final Safety Analysis Report. Reference 3.11 of O-MECH-CALC-021 Rev. 00 is a more recent computation which indicates water depth would be six feet (elevation 1000') for a LOCA in which the SIRWT and Si Tanks' inventory is transferred to containment. In view of this, the existing USAR 6.2.1 value for recirculation phase sump water level is conservatively low for large break LOCA NPSH considerations.

l

4. Explain why the NPSH calculations are not time dependent or based on the temperature of the sump during your design basis accident.

Response

The AP between the containment atmospheric pressure and the sump water saturation pressure was determined as a function of time after a large LOCA. This is shown in the data tables in O-MECH-CALC-021 Rev. 00 Appendix D. For maximum conservatism, the time point at which this AP is minimized was found, and the subcooling credit taken as a fraction of that AP. Given this methodology, it is not necessary to calculate available NPSH as a function of time after the accident.

i l

l l

g l

LIC-99-0002 Enclosure Page 3 Psr the purposes of the NPSH assessment, a maximum spray pump flow value (based on system hydraulic calculations) and corresponding NPSHa value are used. (The maximum per- i pump flow requires less than three spray pumps to be operating. The per-pump flow and corresponding NPSHn will be lower if all three spray pumps are operating.)

The NPSH calculations are r based on the temperature of the sump during a DBA, because (1) the subcooling contribution is based on the difference between the containment atmospheric pressure and the sump water saturation pressure, and (2) sump water saturation pressure is a function of its temperature. Sump water temperature is not used by itself because the subcooling available depends on the containment atmospheric pressure present at the time.

For the large LOCA event of interest, containment pressure will be elevated above atmospheric.

5. The staff believes that you are referencing NRC SER dated April 11,1976, which reviewed CONTRANS Revision 1. Please confirm that the CONTRANS code usedin the 1992 analysis is the same as the code accepted in 1976.

6. Have there been any modifications to the CONTRANS code since its acceptance by the NRC? Provide the current version number of the code that was used for this analysis. If modifications have been made, provide the basis for the modifications to the code.

Response

Questions 5 and 6 will be answered together.

The version used for case OSNVBV97 is CONTRANS2-MOD 1. The CONTRANS code used in the 1992 analysis is a different revision of the same code accepted by NRC in 1976. The changes made in that period are of a nature which does not affect the results as calculated by the code. While more options have been added to the CONTRANS computer code, the basic vapor space, sump response, and long term boil-off model calculations have not been changed since its inception. As stated in the NRC SER (Ref. 3), ' Future applications which utilize this code and reference CENPD-140 must indicate the code revision number and if different from that reviewed as par 1 of this topical report, all modifications to the code must be identified."

Using this and ABB-CE's Quality Procedures as guidance, each analysis calculation clearly l indicates the current version of the code being used, together with the reference of the l Verification and Validation report which directly indicates or can be used to trace the differences l to the code since its inception. Additionally, as stated in Reference (3), "The staff does not intend to repeat the review of CENPD-140 when it appears as a reference in a particular licensing application." By ABB-CE's adherence to its approved Quality Procedures, the use and l intent of the CONTRANS code has not been compromised, and the version used has clearly l been indicated in the recorded calculations.

i Since the original version of the CONTRANS code was documented in Reference (1) to support

Reference (2), subsequent versions have been documented in accordance with the approved ABB-CE Quality Procedures in effect when the revisions were created. Each of these new I

I

+.

l l

l LIC-99-0002  !

Enclosure Page 4 versions has been benchmarked to the old with no or negligible difference from the previous version. The need for these updated versions of CONTRANS has ranged from changing to different models of the old mainframe computers, to different types of workstations. A )

conversion from FORTRAN IV to V has been documented, as well as options which make the code easier to use by providing additional output information and which give the user the flexibility to utilize NRC-initiated issues.

The chronology of CONTRANS versions up to the 1991/1992 time frame is as follows:

• CONTRANS Revision 0 (CENPD-140, PSA-75-013) l This was the original version submitted in the April,1974 topical transmittal.

• CONTRANS Revision 1 (CENPD-140-A, PSA-75-013, PSA-75-075)

This was the version which incorporated NRC comments and was the approved model

)

j documented in the SER for the June,1975 topical re-issue. l 1

. CONTRANS Revision 1D Identical to Rev.1 with the following additions:  ;

extended range of the convective heat transfer coefficient properties j extended fan capability in annulus building (so that the EBASCO designs could be l l

accommodated) e CONTRANS Rev. 2, Cyber 855 Mainframe (00000-OPS-002, Rev. 00)

This was an interim FORTRAN IV version on a different mainframe computer.

• CONTRANS Revision 2, CDC 7600 Mainframe (00000-OPS-002, Rev. 01)

This was different from the Rev.1D version because it included an option to model the first CCT(40) walls using the heat transfer coefficients specified in NUREG-0588 for equipment environmental qualification (EO). This option is only used when studying the  !

maximum temperature sensed by a piece of equipment for EO purposes. The verification shows that '.nis modification did not affect the containment peak pressure !

since the peak conta' ament pressures from the benchmark runs were identical. This l was also the first FORTRAN V version of CONTRANS.

• CONTRANS2 Version NPL, CDC 990 Mainframe An interim FORTRAN V version of the code.

. CONTRANS2 IV.OD1 (CTN2M1U/CTN2M1L), CDC 990 Mainframe (K-FS-C-042)

An updated version which had some output format changes, and the incorporation of the option to account for reevaporation of condensate from the heat sinks, per NUREG-0588 guidelines. This would only be used for EQ application during a MSLB.

l

.. , j LiC-99-0002 Enclosure Page 5 4

. CONTRANS2 MOD 1 (pooh / users /fisib/ appl /ctn2m1/exe/ctn2m1) Apollo Domain N10000, 400 Series (CODES-FS-C-003)

Minor changes included to make the code run on the workstation. Very good agreement j shown with previous test cases.  ;

7. For the CONTRANS analysis, provide the inputs, assumptions, and bias that were used to ensure that the containment pressure was minimized.

See responses to Questions 1 and 2. Case OSNVBV97 is a " maximum safeguards" case in which the overall containment pressure response is more benign than the AVQC peak pressure case.

8. According to calculation 0-MECH-CALC-021 Rev. 00, page D2, the pump manufacturer suggested that "any type of orifice or control valve is possible to alter the system curve to reduce the potential of run out flow. An orifice within the pump would also achieve similar results by a steeper head / capacity curve causing less run out flow." Explain why this option was not chosen to resolve the run out flowissue.

9. Since the run out flow condition occurs during the recirculation phase, are there any procedures in place that instruct the operators to throttle the containment spray pumps in order to prevent the run out flow condition? If not, explain why this option was not considered feasible.

Response

Questions 8 and 9 will be answered together.

If a fixed resistance (such as an orifice) was added to the containment spray system, the increased resistance would decrease system flow during the injection phase of a LOCA, which is when spray flow is most important for LOCA containment peak pressure mitigation. The existing containment spray pump hydraulic performance operability limit for LOCA peak pressure mitigation does not have margin which can accommodate a substantial increase in system hydraulic resistance. This is why a fixed resistance was not chosen as a resolution to this issue.

Floating Step F " Containment Spray Termination"in Emergency Operating Procedures EOP-03 and EOP-20 instructs the operators to terminate spray when containment atmospheric pressure falls below 3 psig. There are no steps in the EOPs instructing the operators to throttle containment spray flow. Crediting throttling capability in the long term of an accident would require a modification to the spray header valves. The OSNVBV97 analysis results indicate that sump subcooling will be available for the large LOCA event, and NPSH will be adequate for the pumps in the recirculation phase. It was therefore not considered justifiable to modify the spray header valves and impose an additional post-DBA procedural burden on the Operators to throttle spray flow, i.e., create an operator work-around and potential human performance issue.

. LIC-99-0002 TABLE 1 Enclosure COMPARISON OFINPUT DATA -

4 a

Page 6 (PEAK PRESSURE CASE AVQC VS. MAXIMUM SAFEGUARDS CASE OSNVBV97)

Card VariatWe AVQC Value OSNVBV97 Units Variable Description Remarks

1. Name Value 8 CCT(26) 105.0 -85.0 *F RWT Temperature for A higher temperature is worse for .

SIP flow containment P/T response CCT(27) 2900 1410 sec Time of RAS A function of SIP & spray flow used. Shorter time worse for long term temperature response.

CCT(29) 55.306293 147.95 lbm/ Total Si pump flow Case OSNVBV97 assumed full Si pump sec after RAS flow, therefore it is greater than one Si train

assumed in Case AVOC. This' reflects only HPSI flow.

, CCT(30) 442.45035 797.84 lbm/ Total Si pump flow Case OSNVBV97 assumed full Si pump

, sec before RAS flow, therefore it is greater than one Si train assumed in Case AVQC. This reflects HPSI/LPSI flow.

13 KCOOL 6 3 -

Number of values in Userinput l CFC table TINT (l) 120.0 90.0 *F Initial heat sink temp Higher temperature is more conservative for containment P/T 22 PAIR 16.85365 14.351 pria Partial pressure of air Higher pressure is more conservative for.

in containment containment P/T 24 PSTM 0.84635 0.349 psia Partial pressure of Higher pressure is more conservative for steam in containment containment P/T TSTM 120.0 90.0 *F Initial Temp of Higher temperature is more conservative for containment vapor containment P/T space i 25 PWTR 17.7 14.7 psia initial pressure of Higher pressure is more conservative for containment vapor containment P/T space 1

_ _ _ _ m__ _ . _ _ _ _ _ _ _ _ . _ _ . _ _ . _ __ _.__..m___.._.__._mm-__.__._.____.-.-.______________m _____._______..._.____.m.__.__ . .._____________.___m_______m_____ . _ . _ _ _ _ _ _ _ _ _ _ . _ _ . -

. ?i LIC-99-0002 TABLE 1 Enclosure COMPARISON OF INPUT DATA - 1 Page 7 (PEAK PRESSURE CASE AVQC VS. MAXIMUM SAFEGUARDS CASE OSNVBV97)  ;

Card Variable AVQC Value OSNVBV97 Units Variable Description Remarks

1. Name Value TWTR 120.0 90.0 F Initial Temp of sump Higher temperature is more conservative for ,

water containment P/T, however, only a small amount of initial sump inventory is used to initiate the code.

28 DELTBL(M) .002,.005, .002,.005, sec Time step size Only one point difference as shown

.005, S.,10., .005, 5.,15.,

20.,50. 20.,50.

38 COOLN 0.0 1.0 -

Number of CFCs Case AVQC did not assume the operation running before of any CFCs. This is conservative for STFAN1 containment P/T analyses.

STFAN1 100E+07 45.0 sec Time at which fan Case OSNVBV97 simulated a realistic coolers switch from switchover from one to two CFC units.

COOLN to COOLN1 COOLN1 0 2.0 -

Number of CFCs Case AVQC did not assume the operation running after STFAN1 of any CFCs. This is conservative for containment P/T analyses.

39, TSCOOL(M) 0.0,120., 90.,288., F Temperature for CFC Not used for Case AVQC. Best data 40 185.,244., 500. performance table available at that time for Case OSNVBV97.

288.,500 QHCOOL ,

QHCOOL(M) 0.0, 0.0, 0.0,58333., Btu / CFC performance for Not used for Case AVQC. Best data ,

12900., 58333. sec use with TSCOOL available at that time for Case OSNVBV97.

24600.,

33343.,33343 41 SPRAYO(l) 0.,0.,0.,2085 0., 0., 0., gpm Containment spray flow Less spray flow is more conservative for 5100 cont. P/T analysis. Full spray flow assumed for Case OSNVBV97.

t

__=-_._. _ . _ _ _ _ _

-a j '

LIC-99-0002 - TABLE 1 Enclosure -*

COMPARISON OF INPUT DATA Page 8 '(PEAK PRESSURE CASE AVQC VS. MAXIMUM SAFEGUARDS CASE OSNVBV97)

Card Variable . - AVQC Value OSNVBV97 Units Variable Description Remarks

1. Name Value TSPRAY(1,1) 0.,0.,0.,-5.0 0., 0., 0., -

For Case AVQC, the More realistic spray start time used for Case psig 55.0 sec absolute value of OSNVBV97.

TSPRAY (4,1) is the CSAS setpoint. For Case OSNVBV97, this -

is the time at which sprays are actuated.

TSPRAY(1,2) 0., 0., -2900., 0., 0., 0., sec For Case AVQC, the More realistic spray operation used for

-159.99 1410.0 absolute values of Case OSNVBV97.

TSPRAY(3.2) and (4.2) are the shutoff time and delay time after setpoint is reached.

For Case OSNVBV97, this is the time at which sprays end injection mode.

43 SDUMPQ(1) 2085.0 5100.0 gpm Cont. spray flow after Less spray flow is more conservative for recirc. cont. P/T analysis. Full spray flow assumed for Case OSNVBV97.

TSUMP(1,1) 2900.0 1410.0 sec Time at which recirc Time of RAS spray flow starts 44 TRFTNK 105. 85.0 F RWT temp for spray Higher temperature is more conservative for flow cont. P/T analysis.

46 AREH 5020.0 10040.0 ft2 Area of SDCHX One SDCHX for Case AVOC and two SDCHXs for Case OSNVBV97-49 QSEC 2407.0 4814.0 gpm CCW flow rate CCW flow rate to one vs. two SDCHXs TINSEC 120.0 90.0 *F CCW temp Higher temperature is more conservative for containment P/T analysis.

____.__._m.__.___ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ - _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ - _ - . _ . . _ _ _ . , _ _ . . , _ _ . . _ _ , , - , ~~vm.. .,- _ .._ _ 4, .,m. , ,. . . . _ - -~.-# .,_. .

.-__s - . - _ . .

~

. 9 - '

~~

., 3:

- LIC-99-0002 a= . TABLE 1

. Enclosure COMPARISON OF INPUT DATA Page 9 . .(PEAK PRESSURE CASE AVQC VS. MAXIMUM SAFEGUARDS CASE OSNVBV97)

Card Varistne AVQC Value OSNVBV97 Units Variatne Description Remarks

-#. Name . Value 81 TZERO 2900.0 1410.0 sec Time of RAS A function of Si and spray flow used. Early time worse for long term temperature .~

response.

TEND 442.45035 797.84 Lbm/ Si pump flow before Case OSNVBV97 assumed full Si pump  :

sec RAS flow, therefore it is greater than one Si train assumed in Case AVQC. This reflects  !

HPSI/LPSI flow.

TEND 1 55.306293 147.95 Lbm/ Si pump flow after RAS Case OSNVBV97 assumed full SI pump sec flow, therefore it is greater than one Si train assumed in Case AVQC. This reflects only -

HPSI flow.

82 PCT 1.2 1.0 -

Decay heat multiplier A larger decay heat multiplier is more conservative for long term release model since more energy will be released to . t containment.

Acronym Definitions:

RWT Refueling Water Tank i P/T Presr>ure/ Temperature SIP Safety injection Pump r RAS ~ Recirculation Actuation Signal CFC Containment Fan Coolers CSAS Containment Spray Actuation Signal e SDCHX Shutdown Cooling Heat Exchanger .

CCW Cornponent Cooling Water HPSI High Pressure Safety injecton LPSI Low Pressure Safety injection

. - . . _ . . . .-- - - . . . . . . - . - . . - _ . . . . - - . . - . .-