ML20086P119
| ML20086P119 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 12/17/1991 |
| From: | Creel G BALTIMORE GAS & ELECTRIC CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| TAC-M68525, NUDOCS 9112260208 | |
| Download: ML20086P119 (7) | |
Text
"'
]
,3
'3 u
- BALTIMORE ~
GAS AND' ELECTRIC
=i
- e 1650 CALVERT CUFFS PARKWAY a LUSBY, MARYLAND 20657 4702 Gcomoc C. CntEL
- vics PRESIDENT
- NUCLE AR cNERGY
-teso)aeo 44ss December 17,1991 N
5 U. S. Nuclear Regulatory Commission Washington, DC. 20555 l
ATTENTION:
Document Control Desk
SUBJECT:
( alvert Cliffs Nuclear Power Plant
- Unit Nos.1 & 2; Docket Nos. 50-317 & 50-318 Station Blackout Modifications Update (TAC Nos. M68525 & M68526)
REFERENCE:
L (a).
Letter from Mr. G. C. Creel (BG&E) to NRC Document Control Desk, - dated. March 20, 1990, ' Supplemental. Station Blackout Submittal G:mtlemen:
In response to the station blackout (SBO) rule (10 CFR 50.63), we committed to perform several
. modifications which would enable the plant to better respond to an SBO event. These modifications
'were briefly described in Reference (a). _We also described these modifications during a meeting with the NRC Stafl on November 7,1991. His letter is to further confirm our ii..entions regarding -
the modifications which support SBO rule compliance.
- Fourigroups of modifications are needed (other than _the addition of two safety related aicsel generators) to bring Calvert Cliffs into full compliance with the SBO rule. They are :nodificatica of the Control Room drop ceiling, providing DC-powered indication for some isolation valves, adding temperature indicating switches _ to the battery rooms, and providing ~AC-power (through the--
- inverters) to :the-Reactor Vessel Ixvel Monitoring System. These modifications are further-described in-Attachment (1). - Our schedule for performing these modifications has not changed,.
~
- however the refueling' outage dates have slipped due to our extended outages. These modifications
- are still scheduled to be performed before the end of the next Unit 1 refueling _ outage (tenth z
refueling outage) and the next Unit 2 refueling outage (ninth refueling outage).
- One other modification that was originally proposed in Reference (a) will not be performed. We will not_ supply DC power to the power operated relief valves (PORVs), A recent SBO analysis shows-that the pnmary system pressure is not high enough during an SBO event to challenge the PORVs
. and, therefore, this modification is not necessary. A description.of the analysis is provided in A
8h 7/ff h.
- Attachment (2).
g 19112260200 911217
- PDR-ADOCK'05000317-i g
P PDR
%_o m
p-p p.
e-
,p p.
'twmw
Document Control Desk December 17,1991 Page 2 As described at the November 7 meeting, our existing SBO emergency operating procedure (EOP 7) is being updated to ensure that all of the appropriate guidance is incorporated. Additionally, this
- procedure will continue to be updated, as needed, when modifications are made which affect the ability of the Units to respond to an SBO event. This includes the modifications described in this.
letter as well as the addition of the two safety-related dicsci generators.
We had previously described implementation of the SBO augmented quality (AQ) program which meets the requirements of Regulatory Guide 1.155, Appendix B.
Our previous schedule had indicated that the AQ program would be fully implemented by the end of 1991. However, due to the substantial changes needed to our Quality Assurance Program to implement the AO concept, full implementation has been delayed until March 16,1992. We do not feel that this has an adverse impact on our ability to respond to an SBO event since most of the response equipment is already safety-related and the remainder of the equipment is considered to be reliable.
Should you have any questions regarding this matter, we will be pleased to discuss them with you.
Very truly yours, _
f-
/
Hff b 12' GCC/ PSF / psf / dim Attachments (1)-
Description of Station Blackout Modifications (2)
' Reactor Coolant System Pressure Response Analysis cc:
D. A. Brune, Esquire J. E. Silberg, Esquire R. A. Capra, NRC D. G. Mcdonald, Jr., NRC T. T. Martin, NRC L E. Nicholson, NRC R. L McLean, DNR J. H. Walter, PSC 4
y--
V u-J p
y,5i re 3
Y
.8
-i-y
4.
g
~A'ITACIIMENT (I)-
DESCRIPTION OF STATION BIACKOUT MODIFICATIONS J/s Modification 1 Control Hoom Celline The Control Room acoustic dron ceiling layout will be modified by replacing 30% of acoustic tiles 4
with egg-crate diffusers.1 The current Control Room drop ceiling is comprised of two types of grid
. systems: acoustic tiles (solid grid panels) and luminous egg-crate type diffusers. The acoustic grids are at Elevations 58' 21/2" and 60'-11", respectively, and are suspended via adjustable hangers from an overhead concrete slab at elevation 69'-0".
The luminous grid is at Elevation 57'-5"
- (approximatel ) and is suspended from the acoustic grid (Elevation 58'-21/2"). The Control Room florescent stri liner lights are located between elevations 58'-21/2" and 57'-5" During a -Station Blackout (SBO), all Control Room heating / ventilation and air conditioning (HVAC) will be lost due to the complete loss of alternating current electrical power to the essential and nonessential switchgear buses and the faihire of the onsite emergency AC power system. This will result in a rise in the Control Room temperature.
Station Blackout analyses revealed that the removal of approximately 30% of the acoustical Control Room drop ceiling will help alleviate the effects of rising temperatures during SBO conditions.
Replacement of the acoustic tiles with egg-crate diffusers will allow free movement of air between the Control Room and the ceiling space. Heat will dissipate continuously from the Control Room and be absorbed in the heat sink surface areas around the ceiling space.
To preserve the stability of the ceiling structure, acoustic tiles removed will be replaced with -
- luminous egg-crate type diffusers. Panels designated for removal will be strategically h)cated in the Control Room in order to maximize heat dissipation and minimize noise levels while maintaining the aesthetics of the room. : Panels located directly below the HVAC system or in front of the electrical
_ control consoles w:11 not be removed because the reflective glare created on the board facings and/or excessive noise levels which could inhibit the operators in performing their duties.
Current electrical configurations and operational characteristics, i.e., emergency lighting, general
' lighting, electrical distribution system, etc., will not be changed at this time. The effects of acoustical
- panel removal o's Control Room iighting levels has not been determined. Lighting levels will be determined after panel replacement has been completed. At that time, it will be determined whether
. additional work to the lighting layout will be required.
}fodification 2 - Valve l'osition Indication The SBO analysis hr.(determined that position indication will not be available for the following valves:
C Containment Sump Discharge Valves (1[2]-MOV-4144 and 4145)
+.
Containment Sump to Miscellaneous Waste Receiving Tank (1[2]-MOV-5462 and 5463)
- e Main Steam Isolation Valve 11 Bypass (1[2]-MOV-4045)
+-
- Main Steam Isolation Valve 12 Bypass (1[2)-MOV 4052)
A modification will remedy this' situation for the valves listed above. Position indication lights at Control Room pancis for these valves will be changed from AC power to DC power. Red and Green
- position indication will be DC powered, while White lights will remain AC powered to indicate AC
~
power availability to the MOVs. ' Die indicating lights are part of the handswitches which will also be replaced, Computer identification' points for vahes 1[2]-MOV-4144,4145,5462, and 5463 will still be avadable; however, she existing 120 VAC input cards for this application will be removed from 1
.. ~.
g.
3.
~.
_-s 2(e e
'4 Y>
A'ITACIIMENT (1)
DESCRIPTION OF STATION BIACKOUT MODIFICATIONS service and the existing spare 125 VDC input cards will be utilized. This modification W nsure that the correct valve position is identified during loss of AC power.
This modification will not change the operational requirements of these valves except to provide position indication of the valves during SBO. He valves can be manually operated if AC power to the MOVs is not available and their position observed at the individuai motor control centers.
Modi 0cullos. 3 - ilattery Room Temperature Indicaflon This modification adds temperature indicating switches to the Unit 1 battery rooms, Unit 2 battery rooms, and the reserve battery room. This modification will allow Control Room operators to identify when a low or high temperature condition exists for the battery rooms and will alert the
-operators of abnormal conditions resulting from the possible failure of the hattery room HVAC system o' incorrect position of winteusummer ventilation selector switch. The alarm will be common for all five batery rooms.
Temperature indicating switches sense the' temperature in the Lattery rooms. Each of the temperature indicating switches has a low temperature setpoint of 710F and a high temperature setpoint of 950F, Each temperature indicating switch setpoint contact provides an input to the 1K02
- F logie cabinet. Annunciator windows at Control Cabinet IC34 alarm the low or high temperature condition.
Modification 4 - Reactor Vessel Level Monitoring The SBO analysis has determined that'neither of the two (per Unit) channels of the Reactor Vessel i
Level Monitoring System (RVLMS) will be available during an SBO event. The power source for
. Channel "b" of each Unit will be changed from a motor control center to a vital AC source (powered from the inverters).
No operational requirements of system will change except availability during an SBO event.
l 2
u-<
' - T ef e
*m' 3'
T meM.
T 3'
%-m w-g -- - '
-w-+
W g
voyu Wg'
-t
=P'ig-Tr=t is--g v r e us e
--9y----
me9 g--'v+*
h 9w--r-
-1M
4 ATTACHMENT (2)
REACTOR COOLANT SYSTE!W PRESSURE RESPONSE ANALYSIS An important aspect of our ability to cope with a station blackout event is the thermal. hydraulic behavior of the primary system during ti e event. %rce cases were analyzed to determine the maximum Reactot Coonant System (RCS) pressure followSg the initiation of a station blackout event. The results of this study predicted a peak RCS pressure of less than 2362 psia for a station blackout event initiated from full power. This peak pressure is below the 2370 psia Power Operated Relief Valve (PORV) setpoint (2385 psia less 15 psia uncertainty). %us, the PORVs are not required to mitigato this event and it is not necessary to provide the PORVs or their block valves with DC power.
AnalvticM Model The CENTS (Combustion Engineering Nuclear Transient Simulation) computer code was used to model the Calvert Cliffs station blackout event. CENTS is an interactive computer code for the simulation of NSSSs and related systems. The primary system models are based on those used in the CEFLASII-4AS computer code. The secondary system models are taken from the long Term Cooling (LTC) code. The database was taken from the BO&E simulator, hence the analysis is representative of the actual plant.
The analyses performed anume that a loss of offsite AC power (LOOP) occurs at time zero, which generates a turbine legic trip. The turbine trip causes the turbine stop valves to close in 0.2 seconds and the n. actor trip breakers to open 0.1 second or 0.2 seconds after the turbine trip. The control rods fall to the bottom of the core in 2.28 seconds, (realistic based on plant experience) or 3.1 seconds (safety analysis value). Operator control of auxiliary feedwater (AFW) was not assumed until five minutes into the transient (i.e., no operator action prior to five minutes).
l Assumptions General Assumptions a.
Loss of offsite power with nc. other independent failure l
b.
The RCS leak rate was minimized to maximize primary system pressure. The RCS fcak rate was assumed to be 20 gallons per minute (gpm). This is conservative since letdown flow will likely be greater than 20 gpm. The letdown valves (battery backed) will not fail shut until after instrument air header pressure is lost, which will occur after peak RCS pressure is reached. In addition, the charging pumps were assumed to stop at time zero. The minimum net RCS leak rate should therefore be at least 20 gpm.
l c.
The RCS leak rate was allowed to change during the event as co,,ditions change, d.
All cases were initiated with the plant operating normally.
l c.
He event began with a turbine trip at t = 0.0 seconds, f.
The PORV setpoint is 2385 pia +/- 15 psi l
4 A"ITACIIMENT (2)
REACTOR COOLANT SYSTEM PRESSURE RESPONSE ANALYSIS Available Systems a.
Control ruom instruments which are battery backed b.
Operator manual actions after five minutes c.
Primary (pressurizer) Safety Valves d.
Manual control of the steam-driven AFW pumps only e.
Turb:nc Stop Valves (normally closed upon turbine trip) f.
Main Steam Safety Valves for control of secondary pressure g.
Four sets of ': 3 safety relief valves each (per SG) with the following nominal setpoint:
1000 psia, lulu psia,1030 psia,1050 ps:a. Accumulation is 3% and blowdown 5%
h.
Tu'bine/ Reactor Protection i.
Safety Injection Tanks Unavailable Systems a.
Power Operated Relief Valves b.
Atmospheric Dump Valves d.
Reactor Coolant Pumps c.
High Pressure Safety Injection / Low Pressure Safety Injection Pumps f.
Charging Pumps g.
Pressurizer Heaters h.
Pressurizer main and auxiliary spray i.
Letdown assumed to be isolated after 30 minutes j.
Main Feedwater k.
AC power for four hours, except foi battery-backed systems 1
Operator actions before five minutes m.
Plant computer after five minutes (based on operator cecuring the plant computer due to loss of Heating, Ventilation and Air Conditioning [HVAC) systems) 2 b
c a
A'ITACIIMENT (?d kEACTOR COOIANT SYSTEM PRESSURE RESPONSE ANALYSIS Analysis Three cases were run to determine the maximum RCS pressure following the initiation of an SBO event. These cases were based on 102% of full power with 20 gpm total RCS initialleakage in order to maximize the increase in RCS p essure following the start of the blackout. Letdown was assumed to be isolated at the start of the trarient. Case 1 used time delap of 0.1 seconds from turbine trip to reactor trip and 2.28 seconds for Control Element Assembly drop time. The peak RCS presst.re from this analysis was 2322.6 psia at 4.4 seconds. Case 2 was identical to Case 1 except that a 0.2 seconds time delay from turbine trip to reactor trip was used which bounds the plant response during the 1987 LOOP at Calvert Cliffs. The peak RCS pressure from this analysis was 2329.8 psia at 4.5 seconds. Case 3 represents the most conservative analysis. A 0.2 seconds time delay from turbine tria to reactor trip was used t.nd a 3.1 seconds CEA drop time was used which corresponds to the m ue used in the safety analyses and Technical Specifications. The peak RCS pressure from this anal..s was 2361.3 psia at 5.0 seconds.
Conclusion The result of this study was that a peak RCS pressure ofless than 2362 r % was predicted for an SBO initiated at full power. This peak pressure is below the 2370 psia m: am m PORV setpoint (2385 psia less 15 psia uncertainty). Thus, the PORVs are not required to mitigate this event. Therefore,it is not necessary to provide the PORVs or their associated block valves with DC power.
Additional information provides support for this conclusion. Peak RCS pressure during the loss of offsite power event experienced by both Units 1 and 2 during 1987 did not challenge the PORVs on either Unit. Peak RCS pressure during a loss of offsite power should not be significantly different than during a station blackout since the peak pressure is reached in five seconds (bef,rc the diesel generators provide any power). Although the Loss of Non-Emergency Power event analyzed in Chapter 14.10 of the UFSAR showed a peak RCS pressure of 2493 psia, this does not detract from the acceptability of the conclusion. The event in Chapter 14.10 contains several assumptions which were considered too conservative for the station blackout analysis. They are: all feedwater is lost at time zero, all steam demand is lost at time zero, letdown is isolated at time zero, and a one second delay exists between the turbine and reactor trips. These assumptions do not reflect realistic plant responses, therefore, they are considered too conservative for the SBO analysis.
3 t