ML20196J527
| ML20196J527 | |
| Person / Time | |
|---|---|
| Site: | Waterford  |
| Issue date: | 04/07/1999 |
| From: | Hannon J NRC (Affiliation Not Assigned) |
| To: | Richards S NRC (Affiliation Not Assigned) |
| Shared Package | |
| ML20196J523 | List: |
| References | |
| TAC-MA3778, NUDOCS 9907070327 | |
| Download: ML20196J527 (4) | |
Text
s f "*6Q\\
UNITED STATES
[-
{
NUCLEAR REGULATORY COMMIS810N I
WAsHmeTON, D.C. SeteMesi Apr11 7, 1999 q,,
MEMORANDUM TO: StWa*Illitheses;Sisector Project Directorate IV and Decommissioning DMsion of Licensing Project Management Ottice of Nuclear Reactor R tion FROM:
John Hannon, Chiet
. Plant Systems Branch Division of Systems fety and Analysis Office of Nuclear Reactor Regulation l
SUBJECT:
WATERFORD UNIT 3 UCENCING AND DESIGN BASIS FOR TORNADO MISSILE PROTECTION OF TURBINE DRIVEN EMERGENCY FEEDWATER PUMP VENT STACK AND STEAM SUPPLY PlPING (TAC NO. MA3778)
The Plant Systems Branch (SPLB) has reviewed the memorandum dated August 28,1998, from Mr. Donald Norkin, Operating Reactor inspection Support Branch Chief, to Mr. John Hannon, Project Directorate IV 1 Director requesting assistance regarding the tomado missile protection for the turbine driven emergency feedwater (EFW) pump vent stack and steam supply piping.
I During a design inspection at Waterford Unit 3, which concluded on June 19,1998, the inspection team identified a tomado missile protection issue. The turbine-driven EFW pump vent stack and steam supply piping that are located on the top of the reactor auxiliary building
}
are not protected f rom tomado missiles. During the inspection, neither the licensee nor the Inspection team could identify any document that exempted these components from being j
protected from tomado missiles.
q The SPLB staff was requested to review three items: (1) determine whether the turbine driven EFW pump vent stack and the steam supply piping are protected adequately from tornado missiles consistent with the design and licensing basts; (2) determine whether the licensee's probabilistic risk assessment methods for not protecting safety-related equipment for safe shutdown from tomado missiles are consistent with the design and licensing basis; and _
(3) determine whether the licenses should have submitted a license amendment in accordance w)th 10 CFR 50.90.
The staff has reviewed the following documents related to the issue: letter dated October 21, 1996, from J. J. Fisicaro, Entergy to NRC; letter dated June 4,1997, from E. C. Ewing, Entergy CONTACT:
Diane Jackson, SPLB/DSSA/NRR 301 415 8548 ATTAQlMENT 9907070327 990512 DR ADOCK 05000382 PDR 1
.~
Stuart Richards 2
1 to NRC; the description of the inspection followup item (IFl 50 382) from inspection Report i
No. 98-20109; a memorandum dated May 6,1990, response to the task interface agreement for Beaver Valley Power Station; and NUREG/CR 2219, *Waterford Steam Electric Station Unit 3 Auxiliary Feedwater System Reliability Study Evaluation," dated September 1981. For the design basis, the staff reviewed Updated Final Safety Analysis Report (UFSAR)
Section 3.5,
- Missile Protection," Section 10A9, " Emergency Feedwater System,' Section 0.2.5.3.3, " Site Related Phenomena (for the ultimate heat sink],' and Appendix 10.4.9, "EmerDency Rellability Assessment," end the associated sections from NUREG-0787, ' Safety Evaluation Report (SER) for Waterford Unit 3," dated July 1981. Additionally, the staff reviewed Standard Review Plan (SRP) Section 2.2.3," Evaluation of Potential Accidents," SRP Section 3.5.1.4, " Missiles Generated by Natural Phenomena," and negulatory Guide 1.117,
'Tomado Design Clacsification."
With regards to whether the turbine-driven EFW pump vent stack and the steam supply piping I
are protected adequately from tomado missiles consistent with the design and licensing basis, the licenses states the followlng in the UFSAR:
a) The turbine-driven pump or both motor-driven pumps together are required to provido the minimum flow rate of 575 gallons per minute (gpm) to the steam generators upon loss of feedwater flow in order to remove decay heat and to reduce the reactor coolant system (RCS) temperature and pressure to the shutdown cooling entry conditions.
b) The EFW system is designed to perform its intended functions in the event of a single I
failure in the system.
c) The EFW system is able to perform its design functions following a tomado, including tomado missiles.
Therefore, by the license basis Waterford Unit 3 should be able to withstand a singte active f ailure and a tomado (including tornado m!ssiles) and have either the turbine driven pump or both motor-driven pumps operable. Since the }urbine-driven pump could be damaged by a tomado and one motor-driven pump could be 16st as a single f ailure,it is possible to have only one motor-driven pump available to perform its safety function. One motor driven pump is insufficient to supply the minimum flow as stated in the licencing basis. The staff also reviewed the SER ahd NUREGICR-2219 and did not find any exceptions that would approve these portions of the EFW system to be unprotected. Therefore, the staff finds that the existing configuration does not meet the current licensing basis.
b With regards to whether the licensee's probabilistic risk assessment methods for protecting safety related equipment for safe shutdown from tomado missiles are consistent with the design and licensing basis, 'UFSAR description SER, and the licensee's letter dated June 4, 1997, were reviewed. In the UFSAR, the licensee demonstrated adequate missile protection for the ultimate heat sink through the use of the concept of probability and demonstrating that the ultimate heat sink could perform its Intended function with the unprotected portions of the wet and dry coolinD towers assumed to be unusable from tomado missile damage. For the licensee to use the same methodology for tne EFW system, it should address the concept of probability of a tomado missile damaging the unprotected portions of the EFW system and
l-1 Stuart Richards a
dernonstrate that the turbine-driven EFW pump could perform its intended function assuming the unprotected portions were unusable. This method would be within the licensing basis.
A 10 CFR 50.59 review would need to be completed to ensure that prior approval from the staff was not needed. However, in the information reviewed, the staff finds that the licensee used the concept of probability but did not demonstrate that the turbine-driven EFW pump would perform its intended f unction with the unprotected portions unusable from tomado missile damage ' Because the licensee did not use the complete methodology to demonstrate adequdte tomado missile protection, it is not consistent with its design and licensing basis, in the letter dated June 4,1997, the licen'see states that the TORMIS computer code was used
)
to demonstrate the concept of probability of tomado missile strikes. The NRC genedcally reviewed and approved the Electric Power Research Institute (EPRI) Report NP-2005 for the use of the TORMIS computer code in a memorandum dated October 26,1983. For the licensee j
to use the TORMIS computer code alone to demonstrate adequate tomado missile protection, the 16consee would need to submit a licensing change for review and approval by the staff. As part of the submittal, the licensee should address the conditions for use desenbed in the staff's safety evaluation for the computer code. These conditions were not mentioned in the licensee's discussion of the TORMIS code.
Since the licensee did not use the complete methodology approved in the licensing basis by demonstrating that the turbine driven EFW pump could still perform its intended function, the staff assumes it cannot withstand the tomado missile damage without adversely affecting the capability of the pump to perform its function. The staff finds that a 10 CFR 50.59 review would result in an unreviewed safety question. Since it was destgrmd to be fully protected from tomado missiles, the change increases the probability of a malfunction of equipment important-to safety and is a different malfunction than previously evaluated in the safety analysis report.
Therefore, the licensee should have subniin3d a license amendment in accordance with 10 CFR 50.90.
in the letter dated June 4,1997, the licenses sta?cs that 4 follows SRP Section 2.2.3. However,
,l It also states that Waterford Unit 3 is not committeo ;. Regulatory Guide (RG) 1.11'r and disconnects the regulatory guide from SRP Section 3.5.2. SRP Section 2.2.3 is the review guide for the evaluation for several potential accidents. S.RP Section 3.5.2 is the review guide for extemally generated missiles, if probability is used in the evaluation of tomado missiles, the probabilistic guidelines in SRP Section 2.2.3 are used in conjunction with the probabilistic guidelines in RG 1.117 and the deterministic guidelines in SRP Section 3.5.2. A probabilistic methodology could not be approved by the use of only SFIP Bection 3.5'.2 because this section t
does not contain any probabilistic criteria to. grant approvst if the licenses chooses to use a probabilistic methodology for the tomado missile protection, it should demonstrate that the probability of a tomado missile striking any target it) question. ls less than or equal to a median g
value of 1.0 E.7,or a mean value of 1.0 E-6 and that the cumulative probability of all tomado i
p missile strikes to all of the targets is below 1.0 E 6 for the entire plant.
'l i
4 i
The August 28, t 998, memorandum referenced the task interf ace agreement (TIA) response j
for the Beaver Valley Power Station (BVPS) dated May 6,1996. To understand the statement 1
in the TIA response that BVPS could riot use probability to demonstrate missile protection, it must be recognized that the determination was based on the initial and existing licensing basis j
for BVPS. At the time of the review, the use of probability was not part of the BV.PS current licensing basis. Because probability was not used as a means of demonstrating tomado i
)
Stuart Richards 4
missi!e protection in the current licensing basis, the use of probabilistic risk assessment (PRA) was not an acceptable allemative to consider in a 10 CFR 50.59 review. However, the staff did not prohibit the licensee from changing its licensing basis to use a probabillstic methodology.
An amendment could be submitted for staff review and approval for the use of probability as a means to demonstrate adequate missile protection.
As referenced in the memorandum dated August 28,1990, the original branch technical position AAB 3-2.did not allow the use of probability to show adequate protection from tomado missiles. In later versions, which became Regulatory Guide 1.117, the staff revised this positipn to allow the use of probability. Current!y, several plants are approved to use probability as a means to demonstrate that positivo misslie protection is not necessary due to the probability of a tomado missile strike. Any licensee that uses probability for the tomado missile strike must ensure that the probability of a tomado missile striking the target in question is less than or equal to a median value of 1.0 E-7 or a mean value of 1.0 E 6 and that the cumulative 1
probability of all tomado missile strikes to all of the targets is below 1.0 E 6 for.the entire plant.
Attached for your information is the staff guidance memorandum dated November 7,1983, for the probabilistic criteria applied to torredo missile protection.
Based on our review, we have determined that the turbine driven EFW pump Vent stack and the steam supply piping are not adequately protected from tomado missiles consistent with the I
design and licensing basis. Further, we have determined the licensee's use of the concept of I
probability without the associated analysit for not protecting safety-related equipment for safe shutdown from tomado missiles is not cons 4 tent with the design and licensing basis that was demonstrated for the ultimate heat sink. Finally, the licensee allowed a change in the plant that was not consistent with its design and licensing basis. The staff finds that a 10 CFR 50.59 mvlew would result in an unreviewed safety question. Since the equipment was designed to be fully protected from tomado missiles and the licensee did not demonstrate that it was adequately protected with the method in its licensing basis, the change increases the probability of a malfunction of equipment important to safety and is a different malfunction than previously evaluated in the safety analysis report. Therefore, the licensee should have submitted a license amendment in accordance with 10 CFR 50.90.
l
Attachment:
As stated I
l l
i li 1
L 4 1
Ib