ULNRC-03747, Forwards Responses to Request for Addl Info Re MSSV Setpoint Tolerance Change.W/One Oversize Drawing

Forwards Responses to Request for Addl Info Re MSSV Setpoint Tolerance Change.W/One Oversize Drawing

ML20217P225
Person / Time
Site:	Callaway
Issue date:	03/04/1998
From:	Passwater A UNION ELECTRIC CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
TAC-M99419, ULNRC-03747, ULNRC-3747, NUDOCS 9803100128
Download: ML20217P225 (10)

Text

___

Union Elscrric One Ameren Plaza

}

1901 Chouteau Avenue i

PO Box 66149 St. Louis, MO 63166-6149 314.621.3222 March 4, i.998 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Mail Station P1-137 Washington, D.C. 20555-0001 Gentlemen:

ULNRC-03747 TAC No. M99419 hl.

7//ggg DOCKET NUMBER 50-483 CALLAWAY PLANT UNION ELECTRIC COMPANY MSSV SETPOINT TOLERANCE REVISION

References:

1) ULNRC-03627 dated August 8,1997

2) ULNRC-036% dated December 16,1997

3) ULNRC-03717 dated January 20,1998

4) SLNRC 84-50 dated March 23,1984

5) B. C. Westreich letter to G. L. Randolph dated February 27,1998 References 1,2 and 3 transmitted information to support a change to the MSSV setpoint tolerance value contained in Callaway Technical Specification Table 3.7-2. Reference 5 requested additionalinformation relating to our amendment request.

Attached are responses to the request for information. Ifyou have any questions concerning this information, please contact us.

Sincerely, ses Alan C. Passwater Manager, Licensing and Fuels

[?[

WEK/pir Attachment kkl,kkkk!!! !

0083 PDR a substdiary of Amoren Corportlico

.M

I i

STATE OF MISSOURI

)

)

SS COUNTY OF CALLAWAY

)

j Alan C.

Passwater, of lawful age, being first duly sworn l

upon oath says that he is Manager, Licensing and Fuels

{

(Nuclear) for Union Electric Company; that he has read the

)

foregoing document and knows the content thereof; that he has i

executed the same for and on behalf of said company with full

{

power and authority to do so; and that the facts therein stated are true and correct to the best of his knowledge, information and belief.

N By Alan C'.

Passwater Manager, Licensing and Fuels f

Nu near 1

1 J

,rd JUBSCRIBED and sworn to before me this day of

/}']/fff/-/

1998.

f

)

1

[

J A CLARK NOTARY PUP,LICSTATE OFMISSOURI*

COLE COUNTY MYCOMMlsS10N EXP NOV 4,2000 f

i

r L

cc:

M. H. Fletcher Professional Nuclear Consulting, Inc.

19041 Raines Drive Derwood, MD 20855-2432 l

Regional Administrator U.S. Nuclear Regulatory-Commission l

Region IV 611 Ryan Plaza Drive Suite 400 Arlington, TX 76011-8064 Senior Resident Inspector

'Callaway. Resident Office U.S. Nuclear Regulatory Commission 8201 NRC Road Steedman, MO 65077 Barry C. Westreich (2)

Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission 1 White Flint, North, Mail Stop 13E16 11555 Rockville Pike Rockville, MD 20852-2738 Manager, Electric Department Missouri Public Service Commission P.O. Box 360 Jefferson City, MO 6510?

Ron Kucera Department of Natural Resources P.O. Box 176-Jefferson City, MO 65102 Denny Buschbaum TU Electric P.O. Box 1002 Glen Rose, TX 76043 Pat Nugent Pacific Gas & Electric Regulatory Services P.O. Box 56 Avila Beach, CA 93424

ULNRC-03747 Attachment Page 1 of 7 UNION ELECTRIC RESPONSE TO THE REQUEST FOR ADDITIONALINFORMATION REGARDING MSSV SETPOINT TOLERANCE CHANGE Ouestion 1 Discuss how the 1-second response time ofpressurizer pressure-high trip is ensured and what licensing basis document this limit is controlled by.

Resnonse l

Under the procedure that defines the Response Time Testing Program at Callaway the response l

times of the sensors and analog circuitry of various instrument loops are compiled from their j

time response testing procedures after every refueling outage.

The maximum response time of the sensors is added to the maximum response time of the analog l

circuitry. The response time of the Solid State Protection System (SSPS) logic is conservatively i

t.ssumed to be 10 maec. This, along with the nu.asured response time of the reactor trip breakers L

is added to the total to determine the total loop response time.

An example of a recent compilation of maximum response times for the Pressurizer Pressure -

High Reactor Trip loops follows:

Sensor: 0.453 sec.

Analog Circuits: 0.050 sec.

SSPS: 0.010 sec.

ResaorTrip Breakers 0.043 sec.

Total: 0.556 sec.

l l

FSAR Table 16.3-1 item 10 currently requires a response time of52.0 seconds, although l

Callaway is currently administratively controlling the response time to $ 1.0 sec. FSAR Table i

16.3-1 item 10 will be revised to reflect the change to 51.0 sec after approval of this amendment.

Ouestion 2 I

Address the Westinghouse NSAL-94-001 and NRC IN 94-60 issues related to neutron flux high setpoint and its relation to the number ofMSSVs inoperable.

Resnonne Westinghouse Nuclear Advisory Letter (NSAL)94-001 and Information Notice 94-60 " Potential Overpressurization ofMain Steam System" notified utilities of a calculation error in determining the maximum power level for inoperable main steam safety valw. Using the NSAL l

methodology results in lower allowed power levels with inoperable MSSVs. We evaluated these documents and imposed administrative controls to limit the maximum number ofinoperable I

V.

L* -

ULNRC-03747 Attachment Page 2 of 7 MSSVs based on the maxi um allowable power range neutron flux high setpoint. These controls exist as an mterpre.ation to our Technical Specifications.

1 L

Union Electric originally proposed revising Technical Specification Table 3.7-1 with the NSAL methodology as part of an amendment application dated September 12,1994 (ULNRC-3%9).

l The NRC Staff rejected this amendment for reasons unrelated to the changes proposed in Table 3.7-1, Technical Specification Table 3.7-1 was not revised per this amendment application because The Callaway Improved Technical Specification (ITS) amendment revised Table 3.7-1 by utilizing the methodology as described in the NSAL. The revised values are contained therein. We will-maintain the administrative controls t otil such time as the ITS is approved and implemented.

Ouestion 3 l

With regard to safety valve acumulation, it is the staff's position that either this must be accounted for as has been done in the past (3% accumulation) or a justification based on test data presented for an alternative method. An alternative method was reviewed for another plant (see Palo Verde safety evaluation dated May 16,1994) which accounts for a pressure increase l

resulting from appropriate delays in valve opening time. Please provide furtherjustification of your proposed method with regard to valve accumulation.

l l

Response

Per ASME Section III subparagraph NC-7512.1, the steam generator safety valves (commonly referred to as main steam safety valves (MSSVs) must attain lift for the stamped capacity at a

- pressure which does not _ exceed the set pressure by more than 3%. Similarly, per ASME Section l

III subparagraph NB-7512.1, the pressurizer safety valves must attain lift for the stamped capacity at a pressure which does not exceed the set pressure by more than 3%. Safety analyses have historically assumed that spring-loaded safety valves actually open over a pressure range of 3% above the set pressure. But safety valves do not behave that way. Historical performance l

shows that once the valve begins to open, it essentially pops open in milliseconds. Vendor design information and actual observed valve behavior confirm that these spring-loaded safety valves open almost instantaneously on steam once the system pressure reaches the valve set pressure.

Extensit a industry testing of pressurizer spring-loaded safety valves was conducted by EPRI in the early 1980's. The test data, behavior and perfonnance are documented in References A and B.

l-Reference A summarizes and addresses the performance of the safety valves tested by EPRI. In l

all cases the steam tests clearly demonstrate that the valves pop open almost instantaneously j

when the set pressure is reached. For loop seal tests the valve pops open on steam

(

instantaneously once the loop seal water purges through the valve. Reference B concludes that the spring loaded safety valve opening time for Crosby HB valves is 10 msec. Given this, if the maximum set pressure range is assumed to be +3%, the valve is expected to pop open following the attainment of the 3% upper set pressure va!ne Westinghouse has determined that a 5 psi i

ULNRC-03747 Attachment Page 3 of 7 i

overpressure perturbation above the 3% set pressure is a bounding value for all anticipated transients and coraervatively bounds system behavior during the very short valve opening delay.

The valve opening time and the 5 psi overpressure assumption can be used to calculate a maximum allowable pressurization rate. For the PSVs, the maximum pressurization ate is 500 psi /sec (5 psi divided by 10 maec). The actual pressurization rate for each transient must be less than the maximum allowable rate to ensure conservatism of the 5 psi assumption. For all RCS transients, the maximum pressurization rates are much less than 500 psi /second Thus, the selection of 5 psi for overpressure is conservative and justified for the pressurizer safety valve applications identified below.

The revised Loss of Load / Turbine Trip and Locked Rotor analyses recently performed for Callaway also model the effects of pressurizer safety valve loop seals as described in Reference C. For these events, no pressurizer pressure reliefli, credited until the loop seal has been fully purged which occurs at pressures well in excess of 3% above the PSV set pressure Crosby design information originally projected that their HB spring loaded pressurizer sefety valves, such as those installed at Callaway, would achieve lift in 60 msee maximum.

Reference B has shown that the Crosby design information is conservative with respect to opening time of the pressurizer safety valves (HB). Crosby also has design data that predicts the opening time for the type HA Main Steam Safety Valves (installed at Callaway) is 80 mnec maximum. Since both the HA and HB valves are spring loaded pop-open steam safety valves, the HA MSSVs will open in the same manner u the HB PSVs did during the EPRI steam tests.

However, there is no test data supporting a faster opening time for the MSSVs, as there is for the PSVs. Thus, since the maximum vendor design time of 80 msec is conservative, it will be used to determine a maximum allowable pressurization rate. Dividing the 5 psi overpressure by 80 msec results in a pressurization rate of 62.5 psi /sec, which greatly exceeds any pressurization rate expected for secondary side transients. Thus, the selection of 5 psi for overpressure is conservative for the MSSV application identified above, and justifiable based on vendor design information, and the comparison to the EPRI industry testing on PSVs.

The discussion above is consistent with previous evaluations performed by Westinghouse for overpressure protection analysis. A 5 psi overpressure assamption has been used for applications where the set pressure tolerance of PSVs or MSSVs has been increased to +3%. For these applications the analyses assume that the valve does not open until the system pressure is at least 3% above the nominal set pressure. Since the rated capacity of these valves is based on 3%

ovwpiessure, credit can be taken for the full-rated flow capacity once the valve has attained rated lift, which will occur within an additional 5 psi above the opening pressure, as discussed above.

Also, as discussed earlier, the Callaway analyses apply this method of modeling safety valves on PSVs with a water loop seal. In these cases, a set pressure shift is explicitly accounted for as 9

described in Reference C. Additionally, the purge time of the inlet water loop seal is accounted for such that the valve does not provide immediate pressure reliefuntil the system pressure is well in excess of 3% above the PSV set pressure.

Based on the above, it is concluded that the modeling technique used by Westinghouse in the Callaway analyses of the Loss of Load /furbine Trip and Locked Rotor events to model both the pressurizer and steam generator safety valves is sufficiently conservative. Also note that the

ULNRC-03747 Attachment Page 4 of 7 method employed in the Callaway analyses is similar to that used to model pressurizer safety valves in the Palo Verde safety evaluation dated May 16,1994. Furthermore,the modeling of the pressurizer safety valves in the revised Callaway analyses is conservative in terms of l

maximizing system pressure since it also incorporates the eff' cts of PSV loop seals using e

methodology described in Reference C, previously approved by the NRC.

1

References:

(A) Burns, E. M., et al., " Review ofPressurizer Safety Valve Performance as Observed in the EPRI Safety and Relief Valve Test Program," WCAP-10105, June,1982.

(B) "EPRI/C-E PWR Safety Valve Test Report, Volume 6: Test Results for Crosby Safety Valve," EPRI NP-2770-LD, March 1983.

(C) Barrett, G. O., et al., " Pressurizer Safety Valve Set Pressure Shift, Westinghouse Owners l

Group Project MUHP2351/2352," WCAP-12910 Rev I-A, May 1993.

Ouestion 4 l

Your position to not account for test instrument uncertainties of +2/-l% is inconsistent with approved methods by which instrument uncertainties are calculated or accounted for. Please account for this uncertainty or provide furtherjustification for its exclusion.

Resnonse Callaway's position is that the test equipment accuracy of +2/-l% required by the OM-1,1987 fall of 1988 Code is accounted for by, the Code allowable stresses. The required test equipment l

accuracy insures the system pressure is controlled within a band covered by the Code's built in safety factor of 4. The Code's recognition of this conservatism is evident in the recent approval of a factor reducSon from 4 to 3.5 in Section VIII of the Code. Therefore the test results obtained i

using equipment which meets the Code required accuracy may be treated as nominal values. The l

current accident analysis then bounds the requested allowable variance in this nominal value.

l This methodology is discussed in the Section XI forward. " Se code does notfully adeess l

tolerances. When dimensions, sizes, or otherparameters are not specipedwith tolerances, the l

walues ofthe.se parameters are considered nominal andallowable tolerances or local wariances l

w be considered acceptable when based on EngineeringJudgement and standardpractice as l

determined by the engineer. " It should be noted that the actual Callaway test equipment accuracy is +/-1% as documented by our testing vendor. Therefore, there is no need to further account for l

test equipment accuracy.

Ouestion 5 In the FSAR, the initial assumptions are discussed with respect to RCS peak pressure. How are they applied for the MSS peak pressure analyses and how does this ensure conservative analyses with regard to the MSS 7

ULNRC-03747 Attachment Page 5 0f 7

)

Redponse The current methodology used by Westinghouse in the analysis of the Loss oiLoad / Turbine Trip event for Callsway considers two cases: inaddresses DNB concerns, while the other evaluates the eff'ects of the transient on RCS pressure. The first case assumes that the pressurizer pressure control system (spray and PORVs) is operable, which yields limiting DNBR results; the second case assumes that these control functions are inoperable, only crediting actuation of the pressurizer safety valves. Uncertainties are not directly applied to the initial conditions in the DNB case, as these are accounted for in the statistical DNBR limit value (Reference D). In the analysis of the peak RCS pressure case, the initial condition uncertainties are directly accounted for in the most conservative direction. Of the two cases analyzed, the case that assumes automatic pressurizer pressure control (DNB case) is typically the most limiting with respect to secondary side pressme. This is because the reactor trip occurs later in this case, which results in additional heat transfer to the secondary side.

I It is recognized that not all assumptions have been specifically biased in such a way as to yield the absolute worst secondary side pressure transient during this Condition II event. However, it is judged that there is sufficient conservatism in the cases analyzed to ensure that the peak steam generator pressure would not change significantly if all parameters were biased in the most conservative direction. Steam generator overpressure protection is conservatively demonstrated

)

by verifying that the total rated flow capacity of the valves is not exceeded. If the accident analysis demonstrates that the full-open pressure of the highest set MSSV is not exceeded, this I

confirms that the relief capacity is sufficient to prevent overpressurization. A review of the j

Callaway FS AR Loss of Load /furbine Trip analysis results indicate that only 4 of the 5 valves are actuated during this event. The additional available relief capacity of the fifth MSSV would be more than sufficient to preclude secondary side overpressurization, even if all analysis assumptions were biased to maximize steam generator pressure.

Reference:

(D) Chelemer, H., et al., " Improved Thermal Design Procedure," WCAP-8568-A (Non-Proprietary), February 1989.

Question 6 What effect will crediting pressurizer pressure control have on the MSS peak pressure analyses?

RAER9aat As shown in the response to Question # 11 in ULNRC-36% (Reference 2), the modeling of automatic pressurizer pressure control during a Loss ofLoad / Turbine Trip event has the effect ofincreasing peak secondary side pressure by 14 psi (from 1281 psia to 1295 psia). Note that the peak steam generator pressure value presented corresponds to the '.ase anaiy:-d to address DNB concerns, as discussed in the response to Question #5 above.

ULNRC-03747 Attachment Page 6 of 7 Ouestion 7 The staff's RAI, dated December 9,1997, asked the licensee to provide various data to allow the staff to understand certain details of the change requested by the licensee. The response to the RAI did not allow this understanding.

The staff requests that the licensee provide:

(a)

Reference 4, the Westinghouse methodology; (b)

A copy of the previous setpoint calculation; (c)

. A copy of the new setpoint calculation; and (d)

An explanation of any differences between (b) and (c).

ReSDonSe (a)

Reference 4 (SLNRC 84-50 dated 3/23/84) has previously been docketed as proprietary information. A copy was recently provided to the Callaway Licensing Project Manager.

(b)

Reference 4 provided the enclosed Table 3-16 and requested that it be withheld from public disclosure, since it contains information held as proprietary by Westinghouse.

Table 3-16 should pntinue to be treated as such.

j (c)&(d) The new setpoint calculation contains the following changes:

]

On the vertical axis, line item 16 corresponds to the his pressurizer pressure reactor trip function discussed in ULNRC-3627 dated 8/8/97 (Reference 1) and in ULNRC-3696 dated l

12/16/97 (Reference 2). As discussed in the previous amendment request letters, the changes to setpoint input values for the subject trip function are:

1)

Sensor drift (column 6) has been changed from 0% span (as denoted by the dasa) to 1.0%

span. This value is then added to the sensor calibration accuracy and M&TE terms in column 3 and that sum then squared in the CSA equation.

2)

The Barton drift allowance bias (column 7) has been eliminated since those transmitters are no longer installed. This term had been an algebraic bias, handied outside the radical in the CSA equation.

3)

The Safety Analysis Limit (column 12) has been decreased from 2445 psig to 2410 psig.

i This directly affects the calculated value for Total Allowance (TA) in column 15, since TA = S AL - Nominal Trip Setpoint (the latter has not changed).

With the above setpoint calculation input value changes, calculated values based on those inputs have also changed, i.e., Allowable Value (column 13), Total Allowance (column 15), CSA

ULNRC-03747 Attachment Page 7 of 7 (column 16), and Margin (column 17, Margin = TA - CSA). Since the Z and S values in ULNRC-3627 are also calculated values they are likewise changed.

OVERSIZE DOCUMENT

PAGE(S) PULLED SEE APERTURE CARD HLES '

APERTURE CARD / PAPER COPY AVAILABLE THROUGH NRC FILE CENTER kUMBER OF OVERSIZE PAGES FILMED ON APERTURE CARD (S)

ACCESSION NUMBERS OF OVERSIZE PAGES:

WO900/.27-o/

f l