a4.05 pr attn: document control desk technical
TRANSCRIPT
-- EEntergy Operations, Inc. 17265 River Road Killona, LA 70066 Tel 504 739 6660 Fax 504 739 6678
Charles M. Dugger Vice President, Operations Waterford 3
W3Fl-2000-0069 A4.05 PR
May 16, 2000
U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555
Subject: Waterford 3 SES Docket No. 50-382 License No. NPF-38 Technical Specification Change Request NPF-38-224 Response to Request for Additional Information Regarding Containment Cooling System
Gentlemen:
Entergy Operations, Inc. (EOI) submitted the subject change request by letter dated October 18, 1999. The NRC has requested additional information to support their review of this request in a letter dated May 12, 2000. Attached is the EOI response to those questions. This response is consistent with the discussion of these topics between EOI and the NRC reviewers conducted by telephone on May 9, 2000.
EOI is hereby requesting the attached information be considered in your review of the change request. The proposed TS change has been evaluated in accordance with 10CFR50.91(a)(1), using the criteria in 10CFR50.92(c), and it has been determined that this request involves no significant hazards consideration. The responses do not impact this conclusion.
EOI is requesting NRC Staff approval of the TS change prior to May 29, 2000 to allow the expeditious closure of Notice of Enforcement Discretion (NOED) 00-6-06. As noted in the NOED, the discretion period is in effect until either this license amendment is issued or until completion of an outage of sufficient duration to effect repairs.
Technical Specification Change Request NPF-38-224 Response to Request for Additional Information Regarding Containment Cooling System W3F 1-2000-0069 Page 2 May 16, 2000
This letter contains no new commitments. Should you have any questions or comments concerning this submittal, please contact Jerry Burford at (601) 368-5755.
Pursuant to 28 U.S.C.A. Section 1746, I declare under penalty of perjury that the foregoing is true and correct. Executed on May 16, 2000.
Very t rulyyus
C.M. Dugger Vice President, Operations Waterford 3
CMD/FGB/rtk
Attachment: Request for Additional Information for NPF-38-224
cc: E.W. Merschoff, NRC Region IV N. Kalyanam, NRC-NRR J. Smith N.S. Reynolds NRC Resident Inspectors Office
ATTACHMENT I To W3F1-2000-0069
NPF-38-224
Response to Request for Additional Information
Attachment 1 to W3F17-2000-0069
Page 1 of 7
RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING CONTAINMENT COOLING SYSTEM
The questions provided by the NRC in letter dated May 12, 2000 and the Waterford
3 response to each item are provided below.
Question I
Page 3 - Are any of the *many input and nodalizational* changes that were made to the Appendix K methodology to adapt it to containment (versus core) analysis new for these analyses?
Answer 1
The input and nodalization changes to the Appendix K methodology is a standard method for conservatively calculating LOCA mass and energy release data for containment pressure and temperature response calculations. This methodology is also in accordance with SRP 6.2.1.3. None of the changes are new for the Waterford 3 analysis.
In an Appendix K analysis, the goal is to contain core heat to maximize the fuel/cladding temperature. Thus, the model is biased to delay removal of energy from the fuel. In the Appendix K analysis, the core is typically represented by 5 axial nodes in each of the three radial zones, for a total of 15 core nodes.
Conversely, for a containment analysis, the goal is to maximize the heat removed from the core to maximize the severity of the mass and energy response. Thus, some of the inputs for the two analyses are required to be different. For the containment mass and energy release calculation, the core is typically represented by 5 axial nodes in one radial zone, for a total of 5 core nodes. Note that Section 6.2.1.3 of the SRP specifies that LOCA containment mass and energy release calculations should be done in general accordance with the Appendix K analysis, although additional conservatism should be included to maximize the release to containment. For example:
The Appendix K prediction of fuel clad swelling and rupture is not considered. This will maximize the energy available for release from the core.
Calculations of heat transfer from core to coolant assume nucleate boiling even though conditions may warrant departure from nucleate boiling. This will maximize the energy transfer to the exiting RCS coolant.
Attachment 1 to W3F1-2000-0069
Page 2 of 7
Question 2
Page 4- What basis justifies using the ANS 1979 + 2 sigma standard rather than the BTP ASB 9-2 guidance (1.2xANS 1971 standard)? Was this also done for MSLB analyses?
Answer 2
The decay heat curve used depends on the time period during the transient and the code used to analyze that part of the accident response. For the blowdown and reflood/post-reflood phase of the LOCA, the Klotz (1968 Nominal curve +20%) decay heat curve is used. This decay heat curve is hard wired into the CEFLASH4A and FLOOD3 codes, and there was no change to this decay heat for the Waterford 3 analyses. The Klotz decay heat curve was a conservative predecessor to the ANS 1971 standard. For the long-term cooldown phase of the LOCA, the ANS 1979 + 2 SIGMA curve was used. The 1979 + 2 SIGMA curve accounts for long term actinides and heavy metals that are not included in the BTP ASB 9-2 curve presented in the NRC's Standard Review Plan. The basis for the use of this curve was its approval for similar long term containment mass and energy response calculations for the Palo Verde 2% power uprate (Reference: Safety Evaluation related to Amendment No. 108 to Facility Operating License No. NPF-41, Amendment No. 100 to Facility Operating License No. NPF-51, and Amendment No. 80 to Facility Operating License No. NPF-74, Arizona Public Service Company, et al, Palo Verde Nuclear Generating Station, Units Nos. 1, 2, and 3, Docket Nos. STN 50-528, STN 50-529, and STN 50-530).
The 1971 ANS standard curve with 20% uncertainty was used in the SGNIII computer code for the MSLB mass and energy release analysis. This code is referenced in SRP Section 6.2.1.4.
Question 3
Page 9 - For MSLB analyses, was saturated steam assumed to leave the break. If not, what is the basis for the assumed entrainment?
Answer 3
No liquid entrainment was assumed for the MSLB analyses. Only pure (dry) steam was released from the break into the containment during the MSLB event. The methodology of this calculation followed the guidelines specified in SRP Section 6.2.1.4.
Attachment I to W3F17-2000-0069
Page 3 of 7
Question 4
Page 11 - The new MSLB analyses do not explicitly include the measurement uncertainties associated with parameters *specified above.*
(a) What is the justification for omitting the uncertainties?
Answer 4(a)
Instrument uncertainties have been implicitly considered in the containment performance analyses rather than explicitly included in the inputs listed in Table 5 of the submittal. This is consistent with the Entergy graded approach for treatment of instrument uncertainties based on the safety significance of the instrument function. This approach recognizes the large margin that exists in the containment safety function compared to the small impact that uncertainties have on the peak containment pressure and temperature. As a result, it is extremely unlikely that instrument uncertainties would cause containment to fail even if all the uncertainties were simultaneously at their worst case condition at the time of the accident. The Technical Specification Change Request included an estimate of the change in probability for containment failure due to overpressurization to be less than 1 E-1 1. Thus, instrument uncertainties have been considered and are accounted for in the margin available to failure.
The Entergy graded approach for instrument uncertainties was discussed with NRC management on December 2, 1999. Although it is realized that NRC did not formally endorse the Entergy approach at this meeting, the concept of considering the margin to failure of the safety function when treating uncertainties received favorable feedback. NRC concurred with Entergy that a graded approach to uncertainties was supported by the existing regulatory guidance.
In addition, during the week of February 28, 2000, NRC Region IV conducted an inspection at Waterford 3 on the graded approach treatment of instrument uncertainties. The inspection report (50-382/00-01, dated March 30, 2000) noted in several cases that adequate analysis margin existed to account for instrument uncertainty. The inspection report documented the NRC Staffs recognition that Entergy uses a graded approach to address instrument uncertainty for safety related systems and components at Waterford 3, and they closed out this issue with no violations or findings.
Therefore, instrument uncertainties have been accounted for implicitly in the margin available to failure of the containment safety function. Because of the large margin and small impact of the uncertainties, there is no need to explicitly include instrument uncertainties in the analysis input values.
Attachment 1 to W3F1-2000-0069
Page 4 of 7
(b) Was this also done for other parameters not specified?
Answer 4(b)
Yes. The four parameters listed in Table 5 of the submittal were the only parameters that were changed from the current Waterford 3 licensing basis containment analyses due to treatment of uncertainties.
(c) Were measurement uncertainties omitted for LOCA analyses also?
Answer 4(c)
Yes. The four parameters listed in Table 5 of the submittal describe the input used for both the MSLB and LOCA analyses.
(d) Do the analysis input values for parameters significantly affecting the results bound their as-operated plant values for both MSLB and LOCA?
Answer 4(d)
Analysis input values are typically selected to be worst case values that bound normal, as-operated plant values. For example, CCW temperature was assumed to be 11 50F throughout the accident although the initial as-operated CCW temperature is below 90°F. The CCW system temperature control setpoint with a Safety Injection Actuation Signal present is 115 0F. The initial containment temperature was assumed to be 120°F although the typical temperature inside containment at the containment fan cooler inlet is less than 1200F. However, instrument uncertainties were not included for these values. As discussed above, the large margin to failure of the containment safety function is sufficient to account for these small uncertainties.
(e) What is the reason for change in CCW design temperature from 120°F to
115 0F? Is it due to change in the Service Water System?
Answer 4(e)
The CCW temperature of 11 5°F was used to remove excess conservatism and be consistent with the CCW system post-accident temperature setpoint. This was not due to any plant or design change. The use of this value remains conservative because 11 50F is used from the beginning of the accident where the CCW temperature during normal operation is maintained below 90°F and is allowed to
Attachment 1 to W3F1-2000-0069
Page 5 of 7
increase to 11 50F post-accident. The peak pressure for both LOCA and MSLB occurs early enough during the transient (12.5 and 63 seconds, respectively) that the actual CCW temperature is expected to remain below 11 50F for most, if not all, of the time to the peak pressure.
Question 5
Please provide the comparison curves of containment pressure and temperature vs time for the benchmarking studies of the GOTHIC computer code with the current licensing basis computer code. Provide an explanation for any major differences in values and assumptions.
Answer 5
Graphical results for four LOCA (L-1 through L-4, see Figures 1 through 8) and two MSLB (S-1 and S-2, see Figures 9 through 12), GOTHIC and CONTEMPT, benchmark cases are attached. Input changes between the benchmark cases are given in the table below. On the figures, please note that LOCA BOGMAX is the same as case L-1, in the legend PR1 and TV1 refers to GOTHIC containment pressure and temperature results, respectively, and (RCB) stands for reactor containment building.
LOCA Cases
Case No. L-1(BOGMAX) L-2 L-3 L-4 Parameter #of CFC Fans 1 2 2 2 CCW Flow (gpm/fan) 1350 1100 1100 1350 CCW Temp. (°F) 115 115 120 112 SDCHX CCW Temp. (°F) 115 115 120 112 SDCHX U (Btu/hr-ft2-°F) 216 179 180.46 238.1 SDCHX CCW Flow (gpm) 3000 2550 2550 3000
Attachment I to W3F1-2000-0069
Page 6 of 7
MSLB Cases
Case No. S-1 S-2 Parameter # of CFC Fans 2 4 CCW Flow (gpm/fan) 1350 1100
Note: CFC = Containment Fan Cooler, SDCHX = Shutdown Cooling Heat Exchanger, U = Overall Heat Transfer Coefficient CCW = Component Cooling Water
The benchmark cases use the same plant input and assumptions for both GOTHIC and CONTEMPT analyses. However, some input differences are necessary due to the different modeling approach for the two codes. The benchmark cases show excellent agreement between the two codes with only minor differences in the calculated temperature and pressure for LOCA after blowdown. These minor differences in results are due to an improved modeling approach and more detailed calculation methods used in the more up to date GOTHIC code.
Some of the differences between GOTHIC and CONTEMPT models include:
0 CONTEMPT uses four predefined types of volumes to model the containment. GOTHIC, however, is capable of using any number of volumes to represent the containment. CONTEMPT models the Waterford containment as one volume, whereas in GOTHIC the containment is modeled using two connected volumes, one for the vapor region and one for the liquid region.
* GOTHIC and CONTEMPT also differ in calculating the heat and mass transfer to/from containment spray flow. CONTEMPT uses a simple equation and a built in spray efficiency to calculate the spray energy removal rate. In GOTHIC, however, the heat removal rate is calculated using the interface heat and mass transfer between the spray droplets and the vapor region.
* The condensation heat transfer coefficient used in GOTHIC during the reflood phase of the transient is higher than that used in CONTEMPT. The GOTHIC heat transfer coefficient more closely matches the Uchida heat transfer coefficient data given in BTP CSB 6-1, Table 3.
Attachment 1 to W3F1-2000-0069
Page 7 of 7
Question 6
Please provide the containment pressure and temperature vs. time curves for the new LOCA and MSLB analyses. Why are the LOCA and MSLB analyses performed at two different power levels?
Answer 6
The containment pressure and temperature results for the most limiting LOCA (peak pressure and temperature and pressure at 24 hours, Figures 13 through 16) and MSLB (peak pressure and peak temperature, Figures 17 and 18) cases are attached. The limiting peak temperature (MSLB) as reported in the submittal is 397.40 F. Including the uncertainties discussed in the submittal will increase the peak temperature by 5.2 0F.
The LOCA analyses were performed at a higher power level to bound future power uprate conditions. For the LOCA analysis, a higher power clearly yields more limiting results. The MSLB analyses were performed at the current power level (102% of the licensed power level). Other input changes for power uprate conditions (such as lower steam generator secondary side mass) caused it to be less clear that power uprate was more limiting for the MSLB.
FIGURE 1
RCB Pressures - LOCA BOGMAX 60
55 - _ -. ---
45
, 40 ..
35 ..
30 ..
-PRI - GOTHIC Volume I 25
x 90CONTEMPT
20 .
15
0.1 1 10 100 1000 10000 100000
Time (sec)
FIGURE 2
RCB Temperatures - LOCA BOGMAX280
260
240
220
200
180
160
140
120
0.1
S..
1 10 100 1000
Time (sec)
10000 100000
FIGURE 3
GOTHIC Case L-2 Pressure Comparisons
1 10 100 1000 10000
Time (see)
60
55
50
45
40
35
30
25
20
15
I-
0.1 100000
FIGURE 4
GOTHIC Case L-2 RCB Temperature Comparisons290
270
250
230
210
190
170
150
130
110
90
0.1
S
2
1 10 100 1000 10000
Time (sec)
100000
FIGURE 5
GOTHIC Case L-3 RCB Pressure Comparisons
1 10 100 1000
Time (sec)
60
55
50
45
40
35
30
25
20
15
I.q
I.e
0.1 10000 100000
FIGURE 6
GOTHIC Case L-3 RCB Temperature Comparisons 290
270
250
230
S210
I'v
S170
130 ....... ..- TV I - GOTHIC Volume I Vapor
1 1 0 . . .. .... ...... CONTEMPT-RCB
9 0 F . . .. . . . . .. ...
0.1 1 10 100 1000 10000 100000
Time (sec)
FIGURE 7
GOTHIC Case L-4 RCB Pressure Comparisons
1 10 100 1000
Time (sec)
60
55
50
45
40
35
30
25
20
15
a..
U
a.
0.1 10000 100000
FIGURE 8
GOTHIC Case L-4 RCB Temperature Comparisons
0.1 1 10 100 1000 10000 100000
Time (sec)
Oi
290
270
250
230
210
190
170
150
130
110
90
FIGURE 9
GOTHIC Case S-1 RCB Pressure Comparisons
20 40 60 80 100 120 140 160
Time (sec)
W3007OSL-new.xls, Page I
70
60
50
40
rA
30
20
10
0 180 200
FIGURE 10
GOTHIC Case S-1 RCB Temperature Comparisons 460
410 - - - ------ - - -
• 310
S260
210 -T- -OHI
0 CONTEMPT
160
110
0 20 40 60 80 100 120 140 160 180 200
Time (sec)
W3007OSLnew.xls, Page 1
FIGURE 11
GOTHIC Case S-2 RCB Pressure Comparisons 70
60 -- - - ~ - ~ 60 ... ...~~~~~ ~ ~~..... . ..... ... [...... .. ..
50 _
S40__
30
PRI - GOTHIC
--- CONTEMPT
2 0 -- .. .. . .. -----.. .. ....
10
0 20 40 60 80 100 120 140 160 180 200
Time (sec)
FIGURE 12
GOTHIC Case S-2 RCB Temperature Comparisons460 T 4 10 . . . . . ...
360 / .....
p310 -.
260 .....
210
-ITV I - GOTHIC
160 ------ CONTEMPT
110
0 20 40 60 80 100 120 140 160 180 200
Time (sec)
FIGURE 13
Figure 1: Hot Leg Break Cont. Pressure PRI. LocAC Feerk Pre-s. Cikse.
I 16 loo l6e6 le +004
Ti ue 5.0(QA)-c 01-/28/99 68:25:51
ReA k R- 'yq.1 -=3 .- 5. Z
U,
i
~i
I'S
S
S.
S 5.
"1
N
GOTH IC
- -------------- -------------- --------- ------ ............... ................
---------------- ----------- .... ....... .
- ------------- ------ -------- ------------ -------------- ---------------
. .............. -------------- ---------------- ---------- ---------------
flsia
FIGURE 14
S
N
'a N N
N 94
In 94
N 94
Hot Leg Break Cont. Vapor Temperature TUI
1 1i 1La 16s le +004
Ti mIe GOTHIC 5.@(QA)-c 01/28/99 08:25:51
2 5-4/S F (Pe-, jk /7" -
0
a'
............... ........ ..... ................ ............... ................
I --------------- --------- ---------------
---------- ............... -------------------------------- ---------------
------------------------------ L ............... ................
.Te-c- .
FIGURE 15
DEDLSB Min. SI - Cont. Pressure
PRI Z d> cA PeAk 2 -q PI-S. Ce.A0 inr
ci�
all
In. I8 I 1l
Ti me
GOTHIC 5.0(QA)-c 04/14/99 15:52:55
?Pe•k P= 4/ 7.q 7 PsA- -- 33.2~-7
1? & 2~cf ~- ~ v~z ?- -~ 5-.sP 1
P&
4
S.
S.
........... ....
.... .... . . . . . . . . . ......... ....... ..... .. .. ............. ......
........... ,--------i [ ' ' ' ii i ..'i i ii i i i i i l
Ps� 7
I
ISO 1888 le +004 le +0105
FIGURE 16
DEDLSB Min. SI - Cont. Temperature TUI
1 1l 188 1000 le +084 le +005
Ti neGOTHIC 5.0(QA)-c 84/14/99 15:52:55
0F7-1=e 25 1
a
U I..'
cI
C,,=
...o .... ........ ... . . ....... ..... ......
*----------..- ... .......... L ............ -- - - - - - --- -- -- -- --.... .--.. .. ... ..---..
...... I ------- ------- -----------
11 1 . 1 1 11 f l . I I 1 1 1 ,1 i l
P-ee, k
FIGURE 17
MSLB Peak Pres. Case-Cont. Pres. PRI
ID-.
c,
S
II S.
40 8B 128 1IS
Tine (sec) GOTHIC 5.@(QA)-c 09/20/99 16:34:15
5
............. ............... - - ........... .... ...............
.... ~ ~ ~ ~ -. .. . .L.. . . . . . ................ L .............. ---------------.
. ............ L ..p .... .. .... .. .............•. .. ............... I ...............
200
I%wlI
pe4 )c 'P = 5:ý - -s S. P si ck, == 112. jý 8, P-s 1- 5-
FIGURE 18
IISLB Peak Temp Case-Cont. Temp. TUI
24 48 72 96
Time (see) GOTHIC 5.6(QA)-c 01/28/99 08:41:05
P-?-1k T- 3 9 7-.
6
0v
aJ
II
S I-
%-4
S..................
128
II