ML20196J765
| ML20196J765 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 06/29/1999 |
| From: | Langenbach J GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 1920-99-20342, NUDOCS 9907080009 | |
| Download: ML20196J765 (22) | |
Text
_
g
{
GPU Nuclear,Inc.
(
Route 441 South NUCLEAR Post Office Sox 480 June 29,1999 Middletown, PA 17057-0480 Tel 717-944-7621 1920-99-20342 U.S. Nuclear Regulatory Commission Attention: Document. Control Desk Washington, DC 20555 Ladies and Gentlemen:
Subject:
Three Mile Island Nuclear Station, Unit 1 (TMI-1)
Operating License No. DPR-50 Docket No. 50-289 Additional Information - Technical Specification l
Change Request No. 279 - Core Protection Safety Limit This letter piosides the updated additional infonnation discussed on June 16,1999. GPU Nuclear's calculations supporting the Loss of Feedwater and Loss of Electric Power accident analyses have been updated to reflect more accurate or more conservative assumptions in several areas. These changes are associated with the ongoing TMI-1 UFSAR upgrade project. The attachment provides an updated discussion of the Loss of Feedwater and Loss of Electric Power accident analyses supporting the subject Technical Specification Change Request (TSCR). The event acceptance criteria as identified in the existing TMI-l UFSAR Chapter 14, and as reanalyzed for TSCR No. 279 previously submitted in GPU Nuclear's letter to the NRC dated March 26,1999, are maintained.
If any additional information is needed, please contact Mr. David J. Distel, Nuclear Licensing and Regulatory Affairs at (973) 316-7955.
Sincerel
???
-Oi 0d7 James W. Langenbach Vice President and Director, TMI
/DJD Attachment
\\
\\
cc: Administrator, Region I e
TMI-l Senior Project Manager i
TMI-l Senior Resident Inspector File No. 98195 9907090009 990629
PDR ADOCK 05000209 P
PDft;
1920-99-20342 METROPOLITAN EDISON COMPANY JERSEY CENTRAL POWER AND LIGEff COMPANY PENNSYLVANIA ELECTRIC COMPANY d/b/a GPU ENERGY GPU NUCLEAR, Inc.
Three Mile Island Nuclear Station, Unit 1 Operating License No. DPR-50 Docket No. 50-289 Technical Specification Change Request No. 279 AdditionalInformation I
COMMONWEALTH OF PENNSYLVANIA
)
) SS:
COUNTY OF DAUPHIN
)
This additionalinformation on Technical Specification Change Request 279 is submitted in support of Licensee's request to change Appendix A to Operating License No. DPR-50 for Three Mile Island Nuclear Station, Unit 1. All statements contained in this submittal have been reviewed, and all such statements made and matters set forth therein are true and correct to the best of my knowledge.
7//
/4 BY:
Vice President and Director, TMI Sworn and subscribed before me this yih d D M,1999.
day of f.J
/
Notary Public Notadas Seat unde L futer, Notary Put*c MyCommisalon uhlree 26 2002 Member, Pennsylvania Associatonof Notaries
1920-99-20342 Page1of20 ATTACHMENT I
LOSS OF FEEDWATER AND LOSS OF ELECTRIC POWER RE-ANALYSIS The Loss of Feedwater and Loss of Electric Power descriptions and results transmitted in GPU Nuclear's letter to the NRC, dated March 26,1999, have been updated to reflect the following analysis changes.
The reason for each change is provided below. The updated assumptions remain conservative.
Change: Use the ANS79 standard decay heat option with 3 isotopes with two sigma uncertainty adjustments for each isotope decay heat curve.
Reason: The previous calculation used 1979 ANS 5.1 Decay Heat Stundard with a 1.05 multiplier.
This was based on the assumption of decay heat from U235 only, together with the conservative assu nption of an infinite operating period. This was considered sufficient to account for other fissioning isotopes and for two sigma uncertainty since a significant fraction of fissions occurs in Pu239, which generates less decay heat than U235 and more than offsets the increased decay heat (relative to U235) from the much smaller fraction of fissions i
occurring in U238. In response to NRC concerns regarding adequacy of the two sigma uncertainty used, the re-enalysis we per onned by specifying the fraction of decay heat from r
each fissile isotope, U235, U238, and Pu239, with the total decay heat being the sum of these three contributions. The decay heat was assumed to be at equilibrium conditions and the fissile isotope split was obtained by using plant specific values for a typical two-year cycle. In addition, a two sigma uncertainty of 4%,20%, and 10% was applied to each decay heat curve for U235, U238, and Pu239 respectively. This uncertainty was applied to each isotope j
contribution before they were summed.
Change: Increase the trip and reset setpoints for the MSSV's to account for 3% setpoint drift.
Reason: TMI-l Surveillance Procedure 1303-11.3 specifies a i 3% as-found and i 1% as-left requirement for the MSSV's. This procedure satisfies the requirements of TMI-l Technical Specification Table 4.1-2 for the MSSV setpoint testing. These tolerances are in accordance with.ASME Section XI 1989 requirements. For conservatism the analysis was revised to account for a + 3% setpoint drift.
Change: increase the initial pump heat to 22.4MW (5.6MW per pump).
Reason: The previous calculation used a pump heat based upon the Crystal River pump curve. Since Crystal River and TMI-l are sister plants, it was assumed thct Crystal River and TMI had the same pumps. However, it was discovered that the TMI pumps were slightly less eflicient than the Crystal River pumps, resulting in a higher pump heat input.
~
g n
, 19,20-99-20342-
" Attachment.
iPage 2 of 20
' Change: Increase the EFW temperature to 135F.
Reason: The primary EFW water source is the Condensate Storage Tank (CST) which is normally less than 90F. The previous analyses assumed an EFW temperature of 120F, which was -
conservative with respect to the overall CST temperature. The CST level is periodically raised by transfer of water from the Hotwell. During this periodic transfer, the piping between
- the Hotwell and the CST would be filled with water which could be hotter than 120F, but less than 135F. Since this is the same piping to which the EFW is connected, the updated analyses conservatively used an EFW temperature of 135F, in the event that a LOFW or loss of electric power would occur during the time of tiansfer from the Hotwell to the CST.
Change: Model the pressure dependence for EFW flowrate to both steam generators linearly between 500gpm to both SG's at 1090 psia and 550gpm at 1065 psia (Loss of Feedwater only).
' Reason: During the transient, and prior to the EFW heat removal being equal to the RCS decay and pump heat, the steam generator secondary pressure is controlled by some cycling of the small main steam safety valves. The small MSSV's open setpoint including a +3% setpoint drift would be 1087 psia and closure at 4% blowdown (+3% drift) would be 1043 psia, for an average steam generator pressure of about 1065 psia. EFW flow from any two EFW pumps was previously modeled as a constant flow of 550 gpm (at 1065 psia), and is now revised to more accurately model flow as a function of steam generator pressure. The flow rates of 500 gpm at 1090 psia and 550 gpm at 1065 psia are established from two (2) points of the f
EFW pump performance curves. A linear dependence is conservative with respect to the actual pump head curve.
Change: Increase the FW valve closure time to 8.9 seconds (Loss of Feedwater only).
Reason: The closure time of the FW control valves was changed from 7 seconds to 8.9 seconds. The 7 seconds was based on the original valve specification sheet. As per Surveillance Procedures (1300-3R and 1302-5.35), the allowable full stroke closure time range is 16.2 to 21.8 seconds.
The actual tested valve closure times are greater than the minimum acceptable closure time.
Since the valve is at 55% open during normal operation, and the valve stroke is at a constant rate, the fastest allowable closure time would be 8.9 seconds.
Change: c Increase the capacity of the PORV to 113,000 lb/hr at 2450psig (Loss ofFeedwater only).
Reason: The previously used PORV capacity of 100,000 lb/hr at 2450 psig was documented in FSAR Table 4.2-8. This was in conflict with the nameplate rated capacity of 106,451 lbm/hr at 2300 psig. Calculations performed in accordance with Section III of the ASME B&PV Code resulted in a PORV rated capacity at 2450 psig of 113,350 lbm/hr. The LOFW calculation was revised to use 113,000 lbm/hr, and the FSAR will be changed to reflect the correct value.
~
m
)
1920-99-20342'
- Attachme'nt -
' Page 3 of 20 E
Change: ' Account fc,r the heat capacity of the metal mass of the primary coolant piping, the reactor vessel and the SG upper and lower heads (Loss of Electric Power only).
. Reason: Heat conductors for the metal mass of the primary coolant piping, the reactor vessel and the l
SG upper and lower heads were previously included in the Loss of Feedwater Accident
-)
l Analysis. For corsistency of models, they were also included in the Loss of Electric Power L
analysis.
Change: Terminate MFW in 2.0 seconds due to condensate and condensate booster pump trip (Loss 1
of Electric Power only).
[
l Reason: ' This change terminated main feedwater flow to the steam generators in 2 seconds instead of
]
10 seconds. The 10 seconds was consistent with the analysis of this event in the TMI-l FS AR.
j However, data retrieved from the June 23,1997 main breaker failure event at TMI-l (resulting in a q
LOOP and Reactor Trip) showed that flow would have stopped entering the OTSG afterjust j
under 3 seconds. The re-analysis of this event conservatively used 2 seconds as the time in which the main feedwater system stops delivering flow to the OTSG.
Change: Assume MS-V13 A is available and TDP flow delivery will begin at 43 seconds after the initiating signal (Loss ofElectric Powet only).
Reason: Since for the complete loss of all on-site and off-site power no additional single failures are required beyond the postulation of the event, TDP flow delivery will begin at 43 seconds after the initiating signal. The previous analysis conservatively assumed the additional failure of MS-V13A, resulting in TDP flow with a 103 second delay.
' Change: Model pressure dependence for EFW flowrate: 330gpm to both SG at 1090 psia and 350gpm at 1065 psia (Loss of Electric Power only).
Reason: During the transient, and prior to the EFW heat removal being equal to the RCS decay and pump heat, the steam generator secondary pressure will be controlled by some cycling of tha small main steam safety valves. The small MSSV's open setpoint including a +3% setpoint drift would be 1087 psia and closure at 4% blowdown (+3% drift) would be 1043 psia, for an average steam generator pressure of about 1065 psia. EFW flow from one TDP was previously modeled as a constant flow of 350 gpm (at 1065 psia), and is now revised to more accurately model flow as a function of steam generator pressure. The flow rates of 350 gpm at 1090 psia and 300 gpm at 1065 psia are established from two (2) points of the TD EFW pump performance curve. A linear dependence is conservative with respect to the actual pump head curve.
This updated discussion of the Loss of Feedwater and Loss of Electric Power Accidents supercedes the l~
description provided in GPU Nuclear letter 1920-99-20088 dated March 26,1999.
l
l" c1920-99-20342.
? At'tachme'nt Page 4 of 20:
L 1 DISCUSSION OF LOSS OF FEEDWATER ACCIDENT L
LA loss of feedwater may result from abnormal closure of the feedwater isolation valves, abnormal
~
control valve failure or main feedwater pump failure. A loss of normal flow through the secondary.
- system will result in a reduction in secondary heat removal, causing the reactor coolant system (RCF) temperature to increase. Due to this temperature increase, the reactor coolant begins to expand causing the RCS pressure to increase.
L
' Increasing RCS temperature and pressure could result in the RCS filling solid, a failure of the RCS, or
[
a fuel cladding failure. Reactor protection for these events is provided by the high RCS pressure trip function of the reactor protection system (RPS). When the RCS pressure reaches the high RCS pressure setpoint, the reactor is tripped. ' Shortly after reactor trip, only decay and pump heats are added to the reactor coolant. Initially, the emergency feedwater (EFW) flow rate is not able to keep l
up with decay and pump heat. Therefore, the reactor coolant will continue to expand until the EFW l
heat removal matches decay and pump heat. Subsequent to this point in time, RCS pressure will decrease and the reactor coolant will contract.
l ANALYSIS ASSUMPTIONS
. The LOFW accident with increased tube plugging was analyzed in' GPU Nuclear Calculation C-1101-224-E610-070, Rev. 3, using the RETRAN-02 Mod 5.2 code. The initial power level is j
,2619.34 MWt which is 102% of the rated power level. A heat balance error is required by Reg.
1 Guide 1.49, which states that the accident analysis calculations must be performed at a power level 2% greater than rated power to account for uncertainties in the determination of power level through the heat balance calculation. Starting the event at a power level higher than nominal is conservative.
l t
L The analyses are performed using beginning-of-cycle (BOC) kinetics since these conditions are the
-most limiting for this event. A smaller (less negative) Doppler coefficient and a larger (more positive) moderator temperature coefficient (MTC) are conservative for a loss of feedwater analysis. Since this accident is analyzed from rated power, a 0.0 pcm/'F MTC was used. TMI-l Technical Specifications require that the MTC shall not be positive at power levels above 95% rated power.
. Reactor trip is modeled to occur when the neutron power reaches 112 percent of rated power of 2568 MWt or when the primary system pressure reaches 2402 psia (2355 psig + 32 psi error + 15 psi j
to absolute) at the hot leg pressure tap. The high-pressure trip setpoint includes 32 psi string error.
l
- A trip delay of 0.4 seconds is used for the high flux trip. The high RCS pressure trip delay is modeled as 0.6 seconds. These values represent the delay from the time the trip condition is reached to the time the control rods are free to fall and bound the actual delays for TMI-1.
1920-99-20342-Attachment Page 5 of 20 The percent of reactivity insertion versus time curve is for 2/3 insertion at 1.4 seconds after reactor j
trip. A minimum tripped rod worth (MTRW) of 2.36% Ak/k is used. The MTRW is comprised of a
_ power deficit of 1.2%, a maximum allowable inserted rod worth of 0.16%, and a shutdown margin of 1.0%. This MTRW corresponds to a required minimum shutdown margin of 1% Ak/k and provides for the maximum worth stuck rod.
The initial pressurizer liquid level is set to 232 temperature-compensated inches, which is the typical hot full power (HFP) preburizer level of 220" plus 12" error allowance. The initial cold leg
{
temperature was 555.7 F and the initial RCS pressure was 2170 psia in the hot leg, which are the approximate operating values with 20% tube plugging.
The analysis was run with an assumption of 20% average tube plugging in the once-through steam generators (OTSGs). Both loops were adjusted to account for a reduction in the coolant flow area in the primary and a 20% reduction in the heat transfer area between the primary and secondary. Since the Loss-of-Feedwater is a symmetric transient (i.e. feedwater is lost to both OTSGs) and the reactor coolant pumps remain operational for this event, an asymmetric tube plugging of 25%/15% (10%
i asymmetry) will show the same results as the 20% average tube plugging. The smaller heat removal from the steam generator with the higher tube plugging will be compensated by the larger heat removal through the steam generator with less plugging, resulting in heat removal equivalent to 20%
average tube plugging.
The RCS flows used in the FSAR Chapter 14 Accident Analysis are conservatively based on 106.5%
of design flow. The average measured RCS flow rates for TMI-l are 110 % of design flow. In order to re-analyze these events for 20% tube plugging, analyses were performed to determine RCS loop flow rates and pump coastdown flow rates. As discussed in the TSCR No. 279 submittal, a minimum flow of 102% of design flow or 133.8 mlbm/hr was used.
t The initial steam generator inventory provides a measure of the heat removal capability of the l
secondary system. For a LOFW, a smaller initial secondary system inventory in the steam generators will lead to a smaller integrated heat removal. The smaller the heat removal, the higher the resultant reactor coolant pressure. The inventory predicted for a steam generator with a level at 50% of the operating range has been calculated to be approximately 39,000 pounds per steam generator. TMI-l normally operates at 60% of the operate range. In addition, the mass of feedwater between the isolation valves and the affected steam generator (approximately 35,500 lbm) was conservatively not modeled and not available to cool the affected steam generator.
l The EFW system is initiated by a low OTSG level signal. The OTSG low level initiation signal of 10 inches is measured by the startup range instruments with an assumed instrument error of 10 inches.
The EFW system was conservatively assumed to deliver flow to the steam generators starting at 43 seconds after the initiation signal with any combination of EFW pumps (2 MDPs or ITDP and IMDP). One EFW pump is assumed to have failed which is the worst single failure. The EFW flow is modeled as a function of OTSG pressure, having a value of 500 gpm at 1090 psia and 550 gpm at 1065 psia.
)
l l
l j
1920-99-2'0342 Attachme' t n
Page 6 of 20 The flow rate through the pressurizer power operated relief valve (PORV) is 113,000 lbm/hr/ valve at 2450 psig. The flow rate through the pressurizer safety valves (PSVs) is 297,846 lbm/hr/ valve at 2575 psig (the setpoint includes a 3% lift tolerance). The pressurizer spray set point is 2205 psig in the hot leg. The spray capacity is 190 gpm. Tr.ble 4 summarizes the analysis input values to be used.
These values bound the TMI-l specific values.
4 Two separate transients were analyzed. First, with no credit taken for the PORV or pressurizer sprays to determine peak RCS pressure. The second, with PORV and spray active to determine a worst case pressurizer level. The pressurizer power operated relief valve (PORV) is a non-safety grade component. Therefore it is not usually modeled in safety analysis. However, in the case of a j
LOFW, actuation of the PORV to control system pressure would aggravate the liquid insurge to the pressurizer. Consequently, the PORV was included in this analysis to provide a conservative prediction of pressurizer liquid level. Similarly, pressurizer spray is a non-safety grade pressure control system. However, actuation of pressurizer spray flow could worsen the pressurizer liquid level response during the event by condensing the pressurizer steam bubble. Consequently, the LOFW accident was analyzed with pressurizer spray to provide a conservative prediction of pressurizer liquid level.
1 ANALYSIS RESULTS The loss of feedwater is conservatively assumed to be initiated by a closure of the I
feedwater control valves. The feedwater control valves are conservatively assumed to close in 8.9 seconds. The rapid loss of flow to the OTSGs causes an immediate reduction in OTSG secondary mass as shown on Figure 7 and a consequent reduction in heat
)
transfer. Reduction of primary-to-secondary heat transfer causes Tha and Tem in both loops to increase because OTSG heat transfer is inadequate to remove primary system heat (Figures 8 and 9).
j Increasing RCS temperatures increase RCS liquid volume and therefore pressurizer level (Figure 10). Increasing level compresses the pressurizer steam bubble, increasing RCS l
pressure (Figure 12) until the reactor trips on high pressure at 18 seconds (Figure 11). The j
trip of the reactor and the resultant, momentary drop in average RCS temperature causes a shrink of RCS inventory and a decrease in RCS pressure. Reactor trip also initiates a turbine trip, so that steam flow out of the OTSGs is reduced, as is heat removal. Reactor l
power stays at the initial value of 102% of 2568 MWt until reactor trip, and so thermal power will not exceed this value. EFW is initiated on low SG level signal at 50 seconds and conservatively assumed tr> start delivering flow after a 43 seconds delay. Since the OTSG pressure at the time of transient termination is below 1065 psia, and there is no level in the OTSG, the EFW flow is about 550 gpm (Figure 12a).
Initially, the EFW flow is not suflicient to remove decay and pump heat, and the system stays pressurized with pressure relief through the PORV and/or pressurizer safety valves as shown on Figure 12. By about 250 seconds, the OTSG heat transfer exceeds the heat generated and the RCS pressure and temperatures start to decrease. However, the pressurizer level continues to increase due to the spray flow.
-1920-99-2'0342-l At'tachment Page 7 of 20 For the case without pressurizer spray or PORV, the PSV cycles several times in the first -
200 seconds, and the maximum RCS pressure is predicted to be 2666 psia in the lower plenum. In the case allowing for pressurizer spray and PORV, the spray trips on at 11.4 seconds and continues to flow untillow pressure signal at 410 seconds. The PSV setpoint is reached once after 20 seconds and the PORV cycles several times in the first 150 seconds. The pressurizer level reaches a maximum of 39.69ft which is below the inlet of the PORV/PSV (39.72 ft), and the pressurizer does not become water solid. As can be seen from Figure 12, the RCS pressure at the time of peak pressurizer level is decreasing and well below the open setpoint of the PORV and PSV. Thus, there is no potential for any water relief through the PORV or PSV The sequence.of events for these cases is shown on Tables 5 and 6.
Based on these results, it is concluded that the reactor is protected against a loss of feedwater accident for a power of 2568 Mwt with 20% average OTSG tube plugging since.
the existing UFSAR acceptance criteria are met.
1920-99-2'0342
- Attachment
- Page 8 of 20 TABLE 4
SUMMARY
OF ANALYSIS INPUT VALUES PARAMETER ANALYSIS VALUE HFP BOC Moderator Temperature 0.0 Coefficient, pcmf'F HFP BOC Doppler Coefficient, pcmf'F
-1.17 HFP Delayed Neutron Fraction, pa 0.007 Spray Capacity, gpm 190 PORV Capacity, Ibm /hr/ valve 113,000 PSV Capacity, Ibm /hr/ valve 297,846 PSV Setpoint Drift, %
3 Initial Core Power, MWt 2619.36 (102% of 2568)
Decay Heat model(2a uncertainty)
ANS 1979 High Flux Trip, percent full power 112 High Flux Trip Delay Time, sec 0.4
)
High Pressure Trip, psia 2402 High Pressure Trip Delay Time, sec 0.6 RCS Inlet Temperature, F 555.7 Initial RCS Pressure, psia 2170 RCS Flow Rate, Mlbm/hr 133.8 EFW Full Flow Rate, gpm 550 EFW Temperature, "F 135
1920-99-20342 Attachme'nt Page 9 of 20 Table 5 Sequence of Events LOFW with PORV and Spray Event Time, seconds Main feedwater control valve closure initiated 0.0 Main feedwater flow reaches zero 8.9 Pressurizer spray on 11.36 RCS high pressure trip setpoint reached 17.81 j
1 Turbine trip 18.31 PORV lift (first) 18.47 Peak RCS pressure reached (2647.7 psia) 21.0 PSV lin 21.0 OSTG low level setpoint reached 50 33 EFW flow initiated 93.33 PORV lift (last) 146.87 Peak RCS temperature reached (613.7 "F) 256.0 Pressurizer spray off 409.71 Peak pressurizer level reached (39.69 ft) 410.0 End of transient 800.0
1920-99-20342 At'achmeht t
Page 10 of 20 Table 6 Sequence of Events LOFW without PORY or Spray Event Time, seconds Main feedwater control valve closure initiated 0.0 Main feedwater flow reaches zero 8.9 RCS high pressure trip setpoint reached 17.75 Turbine trip 18.25 PSV lift (first) 20.31 Peak RCS pressure (2666.4 psia) 21.0 OSTG low level setpoint reached 50.30 EFW flow initiated 93.30 PSV lift (last) 205.41 Peak RCS temperature (613.98 *F) 246.0 Peak pressurizer level (36.40 ft) 323.0 End of transient 800.0 l
1 i
I.
1920-99-20342-
- At'tachme' t n
Page 11 of 20 a00s
.m00s E
b *ms i
1000tLO 0.0 2004 4004 000.0 000.0 ine (a)
Figure 7: OTSG A Mass For LOFW At 2568MWt u0a L
J
~
=0 E
ei l
j 0004 I
I S40.0 "b
ima
=0 me 800.0 Tns (s) 1 Figure 8: Hot Log RCS Temperature for LOFW At 2568MWt
1920-99-20342 At'tachment Page 12 of 20 -
as
~~
w maa E
i,-
E
).
saa
" no ma an as
.oo.o Tne (s)
Figtve 9: Cold Leg RCS Temperature for LOFW At 2568MWt 4
boob sen saw ens %
g an m nw - w
}
J i
' 300.0 "c.o m
an as an Tm M Figure 10: Pressurizer Level for LOFW At 2568MWt i
F, 1920 99-20342' Attachme'nt Page 13 of 20 1J 1D Eu I
ju t
- l. 0.4 oJ CAg ga 400.0 600.0 20 Time (e)
Figure 11: Percent of Rated Power for LOFW At 2568MWt was am0 i mu F
e I
g uma )
=004 1800.0 g3 400.0 400.0 Ed Time (s)
Figure 12: Hot Leg RCS Pressure for LOFW At 2568MWt
1920-99-20342 At'tachment Page 14 of 20 40.0
.10.0 k8 g 20.0 e
10.0 0.0 200J 400.0 600.0 800.0 1
Tm (a)
Figure 12a: OTSG A EFW Flow for LOFW At 2568MWt
1920-99-20342 Attachment Page 15 of 20 DISCUSSION OF LOSS OF ELECTRIC POWER ACCIDENT The loss of all ac power (Station Blackout) transient is the hypothetical case where all unit power except the unit batteries is lost. A loss of power results in gravity insertion of the control rods and trip of the turbine valves. The reactor coolant pumps, main feedwater and condensate booster pumps will also trip. After the turbine stop valves trip, excessive temperatures and pressures in the RCS are prevented by natural circulation with excess steam relief through the main steam line safety valves and the atmospheric dump valves (turbine bypass valve steam reliefis lost due to loss of power to the condenser cooling water circulating pumps). Excecs steam is relieved until the RCS temperature is below the pressure corresponding to the set point of the atmospheric dump valves. Thereafter, the atmospheric dump valves are used to remove decay heat. The turbine-driven emergency feedpump (TDP) provides feedwater for decay heat removal. The TDP takes suction from the condensate storage tanks and is driven by steam from either or both steam generators. Decay heat removal after coastdown of the reactor coolant pumps is provided by the natural circulation characteristics of the system.
ANALYSIS ASSUMPTIONS The initiating event for a station blackout transient is the loss of all ac power, and results in a trip of
)
the reactor, the reactor coolant pumps and main feedwater pumps. The analysis assumptions are the j
same as for a Loss of Feedwater Accident and are sununarized in Table 7. Table 7 also provides a sequence of events for SBO. The natural circulation cooldown with increased tube plugging was j
analyzed for the SBO analysis in GPU Nuclear Calculation C-1101-224-E610-069, Rev.1, using the RETRAN-02 Mod 5.2 code. This code is approved by the NRC foi use in TMI-l licensing applications (NRC SER dated February 10,1997). The analysis was performed from an initial power level of 2620 MWt, which is 102% of rated pov r. Conservative beginning-of-cycle kinetics parameters were used. The model uses a conservatively low initial OTSG mass of 39,000 lb.
For the station blackout (SBO) transient, the emergency feedwater (EFW) system would be initiated by a loss of reactor coolant pumps signal. With a loss of ac power the motor driven EFW pumps (MDP) are unavailable and only the t,urbine driven EFW pumps (TDP) are assumed to provide flow.
Subsequent to the EFW initiation signal, the steam admission valve (MS-V-13 A/B) to the turbine driven pump (TDP) receives an immediate open signal and is fully open in 24 seconds. Turbine testing shows the TDPs are at full speed in 11 seconds after the steam admission valves are full open.
An additional 8 seconds for flow coastup is modeled resulting in TDP flow delivery at 43 seconds, when MS-V13 A is available. MS-V13 A/B provide motive power to the turbine driven EFW pump from the ' A' and 'B' once-through steam generators (OTSGs) respectively. The EFW system is designed so that MS-V13 A receives an immediate open signal as described above, while MS-V-13B has a delay timer that is set between 40 and 60 seconds to avoid lifting of the EFW steam line relief valve (MS-V-22).
l
1920-99-20342 At'tachme'nt Page 16 of 20 Since for the complete loss of all on-site and off-site power no additional single failures are required beyond the postulation of the event, TDP flow delivery will begin at 43 seconds after the initiating signal. EFW is assumed to deliver flow to both steam generators 43 seconds after initiation signal.
The EFW flow is modeled as a function of OTSG pressure, having a value of 330 gpm at 1090 psia and 350 gpm at 1065 psia.
No credit is taken for the pressurizer power operated relief valve (PORV) or pressurizer sprays and the turbine bypass and atmospheric dump valves are conservatively assumed to be unavailabla.
ANALYSIS RESULTS A loss of power results in gravity insertion of the control rods and trip of the turbine stop valves.
Since the eveN is initiated by a reactor trip, thermal power never exceeds the initial value. The reactor coolant pumps, main feedwater and condensate booster pumps will also trip. Consequently, the transient proceeds in a mostly symmetric manner and the results for one loop are presented in Figures 13 through 18 and may be considered as representative of the other loop.
]
As a result of the condensate and condensate booster pump trip, feedwater is conservatively terminated in 2.0 seconds as shown in Figure 13. An almost instantaneous EFW signal is initiated on l
loss of RCPs, and EFW will try and control level at 50% of the operating range. After reactor trip and turbine stop valve closure, the main steam safety valves (MSSVs) open to relieve excess steam.
The MSSVs cycle several times during the first 100 seconds of the transient, following which the steam generator pressure is controlled by the small safety valve which cycles about its open and close setpoints.
Following loss of power to the reactor coolant pumps, the reactor coolant system flow decays (Figure 15). Decay heat removal after coastdown of the reactor coolant pumps is provided by the natural circulation characteristics of the system. Only the TDP provides emergency feedwater for
. decay heat removal. This pump is assumed to statt delivering flow to both SGs at 43 seconds after
- reactor coolant pump trip as shown on Figure 14. The EFW flow is a function of the OTSG pressure and Figure 14 shows the EFW flow varies from 330 to 350 gpm as the small safety valve cycles.
The RCS system pressure decreases to post-trip values as shown on Figure 16, then there is a rise in pressure when the secondary side heat removal decreases in response to the low SG inventory. The RCS pressure rises until the pressurizer safety valve lifts at 910 seconds. The RCS pressure starts to rise again then levels off as the secondary side heat removal begins to balance the decay heat. The pressurizer level shows a gradualincrease following its post-trip value as seen on Figure 17. Figure 18 shows a temperature difference of greater than 20*F exists between the hot and cold legs at the end of the transient. The cold leg temperature corresponds to the saturation temperature at the lowest MSSV setpoint. As decay heat decreases, the energy removed by the EFW will exceed the core decay heat and RCS temperatures, pressure and level will be reduced. Similar results were achieved with asymmetric steam generator tube plugging of 25%/15%.
i
1920-99-20342 Attachment Page 17 of 20 TABLE 7
SUMMARY
OF ANALYSIS INPUT VALUES PARAMETER ANALYSIS VALUE HFP BOC Moderator Temperature 0.0 Coefficient, pcmfF HFP BOC Doppler Coeflicient pcm/ F
-1.17 HFP Delayed Neutron Fraction, par 0.007 PSV Capacity Ibm /hr/ valve 297,846 PSV Setpoint Drift, %
3 Initial Core Power, MWt 2620(102% of 2568)
High Flux Trip, percent full power 112 High Flux Trip Delay Time, sec 0.4 j
I High Pressure Trip, psia 2402 High Pressure Trip Delay Time, sec 0.6 RCS Inlet Temperature, F 555.7 Initial RCS Pressure, psia 2170 RCS Flow Rate, Mlbm/hr 133.8 Sequence of Events for Station Blackout Event Time, seconds Reactor and RCS pump trip 0.001 Turbine trip 0.50 Main feedwater flow reaches zero 2.0 Small Safety setpoint reached (first) 2.74 MSSV Banks 1& 2 setpoint reached (first) 2.81 MSSV Bank 3 setpoint reached 4.36 EFW TDP flow initiated 43.0 MSSV Banks 1& 2 setpoint reached (last) 95.59 Pressurizer Safety valve lift 910.84 End of transient 1500.0
l
'1920-99-20342 Attachme'nt -
Page 18 of 20 2000.0 1900.0 7
E g _0 o
500.0 0.0 S00.0 1000.0 1500.0 Ime (a) i Figure 13: Feedwater Flow to OTSG A For SBO At 2568MWt With 20% Average SGTP 30.0 20s
?
E E
10.0 00 600A 1000.0 1500.0 1me (s)
Figure 14: TDP EFW to OTSG A For SBO At 2568MWt With 20% Average SGTP
r :i
.?
1920-99-20342-Attachme'nt Page 19 of 20 10000,0 8000.0 A
- j
.0 0
1 E0
==0 0.0 800.0 1000.0 1500.0 1me (a)
Figure 15: Loop A RCS Flow For SBO At 2568MWt With 20% Average SGTP 2600.0
--a I
E i
1
- g nme 200lLD 1me (a)
Figure 16: RCS Hot Leg Pressure For SBO At 2568MWt With 20% Average SGTP
p
' 1920-99e20342 -
- Attachme'nt Page 20 of 20 14.0
/~
21.0 g.-
}-
s m0 1'
)
18.0 I
0.0 SOES 1000.0 1500.0 Eme (e)
Figure 17: Pressurizer Level For SBO At 2568MWt With 20% Average SGTP no.O
$10.0
- d W E
300 i
y im0 I
3 =0 Smo 550.0 0.0 500.0 1000.0 1500.0 I
Eme (s)
Figure 18: RCS Temperatures For SBO At 2568MWt With 20% Average SGTP j
d