ML20210S213
| ML20210S213 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 05/13/1986 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20210S199 | List: |
| References | |
| NUDOCS 8605200464 | |
| Download: ML20210S213 (6) | |
Text
,. _
'+
/
UNITED STATES y
g NUCLEAR REGULATORY COMMISSION j
j WASHINGTON, D. C.,20556 g...../
i SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO AMENDMENT NO. 57TO FACILITY OPERATING LICENSE NPF-9
.AND AMENDMENT NO.38 TO FACILITY OPERATING LICENSE NPF-17 DUKE POWER COMPANY McGUIRE NUCLEAR STATION, UNITS 1 AND 2 INTRODUCTION By letter dated May 9,1985 and revised or supplemented October 2 and 14, December 17 and 23,1985, January 14, March 17 and April 8,1986, Duke Power Company (the licensee) requested changes to the Technical Specifications to provide for operation up to full power with the Upper Head Injection (UHI) system in bne of three conditions:
(1) fully operable in accordance with existing Technical Specifications., (2) physically installed but functionally the closing of both isolation valves in each of the two injection disabled by(3) deleted by physical removal of major portions of the system
~
lines, and such as injection piping, valves, support / restraints and instrumentation.
Other changes associated with UHI system isolation or removal were also re-quested. These include deletion of technical specifications requiring UHI system maintenance, surveillance, and leakage verification and modifications of Technical Specifications to reflect deletion of UHI related containment penetrations and associated conductor overcurrent protection devices, con-tainment isolation valves, and system piping snubbers.
The licensee's letter also requested Technical Specification changes for the ECCS cold leg injection accumulator which would be implemented prior to and
~
during the startup with the UHI :.ystem isolated or removed. These changes increase the operable range of the nitrogen gas cover-pressure.and decrease the operable range of the water volume from 8022 and 8256 gallons to 6870 and.
7342 gallons.
(The changes to the ECCS cold leg injection accumulator will also be accompanied by appropriate modifications to instrumentation, alarm functions and procedures, and by resizing the flow restricting orifices in the discharge piping; however, these accompanying changes do not involve a j
change to the Technical Specifications).
EVALUATION The present plant design for McGuire, Units 1 and 2 incorporates an ice condenser containment and Upper Head Injection System. The ice condenser containment was introduced as a less costly alternative to the large dry containment..The ice beds were designed to efficiently condense the steam from the postulated design basis large break loss-of-coolant accident (LOCA), limiting the pressure excursion in the containment. The low containment pressure contributed to increased steam binding in the primary system, thereby delaying reflood of the core and leading to higher calculated peak clad temperatures.
h kOO 64 860513 P
K 05000369 PDR
=-*.a-ce--3.----,,--.f-w.
,5------wm-.-,-.m,,w--,,,--..,eew.-,--,.---,,,,eh,es.--e.,.wev,c--g
.,--.-r,,w y w -r-- g r resr e ar-=
-g-+,,ww w c.m -
- t
-2, Upper Head Injection (UHI) was introduced in some Westinghouse plants to enhance core cooling during blowdown and to generally compensate for core cooling problems associated with low containment pressure. Five plants containing UHI have been licensed and three more are nearing operation.
The UNI system may be visualized as a pressurized accumulator tank contain-4 ing at least 1850 cubic feet of borated water which is connected through a system of shut-off and check valves to the upper head of the reactor vessel.
A nitrogen gas system keeps the UHI system pressurized to over 1200 psi, i
When the primary system is above 1900 psi, the UHI is isolated only by check i
valves. UHI plant upper head internals are designed to assure good mixing of the injected flow with the steam from the hot legs during a cold leg i
LOCA.
t l
l Experience has demonstrated that the UHI system adds to the complexity of l
plant operation, requires additional maintenance and generally reduces i
plant availability. Licensee event reports have identified numerous pro-j blems with.the UHI systems. Duke Power Company began discussions with the l
NRC early in 1985 while evaluating the possibility of eliminating the UHI j
systems from the McGuire Station, Units 1 and 2.
After a series of interchanges with the staff and the ACRS, it was generally agreed that there were adequate incentives to consider removal.
jl As identified above, the licensee submitted a number of documents to support the proposed removal of the UHI system during successive refueling outages.
This SER presents the results of the staff review of the licensee proposal to remove the UHI system and modify the associated Technical Specifications.
The :taff review included the revised Chapter 15 FSAR analyses and supporting discussion of plant characteristics as they would be affected by the removal 1
of the UHI system.
i The analysis of a large break LOCA transient is divided into three phases:
(1) blowdown, (2) refill, and (3) reflood. There are three distinct tran-sients analyzed in each phase:
(1) the themal-hydraulic transient in the reactor coolant system (RCS), (2) the pressure and temperature transient within the containment, (3) and the fuel and cladding temperature transient of the hottest fuel rod in the core. Based on these considerations, a system i
of interrelated computer codes has been developed for the analysis of the LOCA.
The SATAN-IV computer code analyzes the thermal-hydraulic transient in the RCS during blowdown and refill. The WREFLOOD and BART computer codes are used to calculate the thermal-hydraulic transient during the reflood phase of the accident. The BART computer code is used to calculate the fluid and heat transfer conditions in the core during reflood. The LOTIC computer code is used to calculate the containment pressure transient during all three phases of the LOCA analysis. Similarly, the LOCTA-IV computer code is used to compute the themal transient of the hottest fuel rod-during the three phases. References 1, 2, 3 and 4 describe the major phenomena i
modeled, the interfaces among the computer codes, and the features of the codes which assure compliance with NRC Acceptance Criteria. The staff has confirmed that acceptable evaluation models were used for the large break l
LOCA analysis, i
1 i
~
.g o
-3, j
I I
4 charge)ge break LOCA (double ended cold leg break at reactor coolant pump dis-The lar ar.alysis results are listed in Table 1 along with a column describing the 10 CFR 50.46 limits. The break spectrum was developed by varying the break discharge coefficient (C ) assuming the worst single failure. The limiting D
case (Cn = 0.6) resulted in a peak clad temperature (PCT) of 1896'F. When the j
limiting break analysis was repeated assuming full safety injection (SI) capability, including full containment spray, a PCT of 2132'F resulted. The assumption of full safety injection resulted in a lower containment pressure, i
leading to steam binding. This delayed reflood and resulted in the higher PCT.
l The full containment spray plus the large quantity of safety injection flow issuing from the vessel side of the break contributed to the condensation of 4
additional steam and the lower calculated containment pressure. The full SI i
case with a CD = 0.6 thus became the limiting case for the non-UNI configuration l
The small break LOCA reanalysis submitted by the licensee was perfomed with tne NOTRUMP code as described in Reference 5 and approved by the NRC in August 1985. Using the WFLASH code, the previous (1984 update of the McGuire FSAR) small break LOCA analysis identified the limiting small break as a break I
equivalent to a 6 inch diameter hole and reported a PCT of 1499*F. The re-analysis using the more realistic NOTRUMP code, identified the limiting small break as a break equivalent to a 3 inch diameter hole and PCT of 1488'F.
The differences between these analyses are attributed to modeling changes and 3
the use of non-equilibrium thermodynamics in NOTRUMP.
An acceptable version of the LOFTRAN code (reference 6) was used for transient analysis. Only two plant transients were reanalyzed because these were the only transients which were predicted to depressurize the primary system suffi-ciently to initiate the upper head injection. One was the " Inadvertent Opening of a Steam Generator Relief or Safety Valve" and the other was " Steam System l
Piping Failure". The THINC code was used to detennine if departure from nucleate boiling (DN8) occurred. The steam system rupture was the more I
i limiting transient. The DN8 analysis confirmed that DNB ratios were always greater than the fuel design limit value of 1.17.
{
In surunary, we have confinned that the safety analyses were done with approved codes. Plant and core characteristics without UHI were demonstrated to be within regulatory limits.
The NRC staff was aided by consultants from the Sandia National Laboratory (SNL) in the review of the proposed removal of UHI from the McGuire Station.
SNL was asked to use the TRAC-PFI/ MODI code to study the characteristics of a j
generic Westinghouse 4-loop reactor with ice condenser, with and without UHI, The UHI internals were modeled for both cases. The water in the upper head j
provides some cooling benefit whether or not it is augmen ed by flow from the l
UHI system. For the non-UHI case the cold leg accumulat6rs were modeled similarly to other four loop Westinghouse plants (accumulator pressure up, water volume decreased slightly, and accumulator line resistance decreased) so that cold. leg injection would be more effective.
I
= -
p i
o
-4, t
In our two cases the first part of blowdown was identical. The blowdown PCT occurred at about 21 seconds. As the UHI didn't come on until around 3 seconds, UNI had no effect on the blowdown PCT which was slightly less than 1400*F. At the end of blowdown +.he cladding temperature was about 100*F cooler with UHI than without. This is of interest primarily to show when some cooling benefit from UNI occurs.
In our UHI case there is no reheat of cladding after blowdown.
It stays at roughly 1000*F until the hot rod is quenched.
In our non-UHI case, only the hottest rods reheated to around 1400*F before they turned around.
In our TRAC analysis, the cladding temperatures reached during blowdown and during reheat were well below the 2200*F limit of 10 CFR 50.46.
The TRAC analyses have been characterized as less conservative than an Appendix K analysis but somewhat more conservative than a "Best Estimate" or "most likely case".
It has been pointed out by the ACRS and others that if a more realistic Fn (lower than 2.17) had been selected for the TRAC input there likely would have been little or no clad reheat after blowdown in the non-UHI 3
case. We believe that the analysis as done confims that even if a few of the hottest rods do experience some reheat, there is ample margin relative to the 10 CFR 50.46 limits.
SNL was also requested to investigate the effects of nitrogen cover gas enter-ing the primary system if the UHI isolation valves did not shut off at the proper tank low level rM. SNL concluded that a postulated failure to shut off the UNI at the low tank level was indicated to be slightly beneficial to core cooling. Continued UHI water flow into the core (nearly twice as much) provided additional cooling before the nitrogen cover gas would enter the upper head. When the gas did enter the upper head, there was significant venting to the cold legs and out the break and therefore very little nitrogen actually passed through the core.
We have reviewed the licensee's request to remove the UHI sytem and change the associated Technical Specifications. As stated previously, we have con-cluded that the McGuire Nuclear Station, Units 1 and 2, meets the requirements of 10 CFR 50.46 for large and small break LOCAs without UHI.
In addition, i
potentially affected transients were evaluated without UHI and found not to exceed fuel design limits. Therefore, the licensee's request to eliminate the UHI system for the McGuire Nuclear Station, Units 1 and 2 is accepta.ble.
Similarly, licensee's request to operate with the UHI functionally disabled by isolation is also acceptable.
ENVIRONMENTAL CONSIDERATION Pursuant to 10 CFR 51.32, the Consnission has detemined that issuance of the amendments will have no significant impact on the environment (51 FR 13574).
J CONCLUSION l
Notices of opportunity for a prior hearing were published in the Federal Register on July 26,1985 (50 FR 30548) and February 4,1986(51FR4449). No & quests
y
-5.
for a hearing were received. The licensee's letters of March 17 and April 8,1986, do not alter the requested changes and are addressed in the Commission's Environmental Assessment and Finding of No Significant Impact (51 FR 13574). We have concluded, based on the considerations discussed above, that:
(1) there is reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, and (2) such activities will be conducted in compliance with the Commission's regulations and the issuance of these amendments will not be inimical to the common defense and security or to the health and safety of the public.
REFERENCES 1.
Bordelon, F. M., Massie, H. W. and Borden, T. A., " Westinghouse ECCS Evaluation Model-Suomary", WCAP-8339, (Non-Proprietary), July 1974.
2.
Westinghouse ECCS Evaluation Model, 1981 Version", WCAP-9220-P-A, Rev. 1 (Proprtetary), WCAP-9221-A, Rev.1 (Non-Proprietary), February,1982.
~ 3.
Young, M., et al., "BART-1A: A Computer Code for the Best Estimate Analyzed Reflood Transients", WCAP-9561-P-A, 1984 (Westinghouse Proprietary).
4.
Chiou, J. S., et al., "Models for PWR Reflood Calculations Using the BART Code", WCAP-10062.
5.
Lee, H., Tauche, W. D., Schwarz, W.
R., " Westinghouse Small Break ECCS Evaluation Model Using the NOTRUMP Code", WCAP-10081-A, August 1985.
6.
Burnett, T. W. T., et al., "LOFTRAN Code Description", WCAP-7907-P-A, April 1984.
Principal Contributors: Darl S. Hood, PWRf4 J. Watt Dated: May 13, 1986 9
w --
-_Ar.
-. ~,.-
d d
O
?
TABLE 1 i
LARGE BREAK LOCA RESULTS j
l WITHOUT UHI C
0.4 C
0.6 LIMITS 0.8 C
0.6 C
- E =CLG D =CLG 10 CFR 50.46 D =CLG D =CLG l
I (MAX SI)
FUEL CLADDING DATA Peak Clad Temperature (*F) 1865 1895 1863 2132 2200 Peak Clad Temperature Location (ft) 6.75 6.75 6.75 6.50 I
local Zr/H 0 Reaction (max), (1) 2.53 2.12 2.16 5.05 17.00 p
Local Zr/H O Location (ft) 5.50 6.00 5.50 6.50 2
)
Total Zr/H O Reaction (%)
0.3 0.3 0.3 0.3 1.0 2
Hot Rod Burst Time, (sec) 61.4
'82.2 88.8 63.0 Coolable Hot Rod Burst Location. (ft) 5.50 6.00 5.50 6.00 f'
i I
I i
i e
4
+
l i