ML20216H647

Forwards Response to 970414 RAI Re License Amend Request for MNGP Power Rerate Program

ML20216H647
Person / Time
Site:	Monticello
Issue date:	09/05/1997
From:	Hill W NORTHERN STATES POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
TAC-M96238, NUDOCS 9709170009
Download: ML20216H647 (51)

Text

.

l Northem States Power Company Monticello Nuclear Generating Plant 2807 West Hwy 75 Montice'lo, Minnesota $5362 9637 September 5,1997 US Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555 MONTICELLO NUCLEAR GENERATING PLANT Docket No. 50-263 License No. DPR-22 Response to Request for Additional Information (RAl) on Monticello Power Rerate Program (TAC No. M96238]

By letter dated April 14,1997, the staff provided a request for ad6iSonal information (RAl) to facilitate its review of NSP's license amendment request for the Monticello Nuclear Generving Plant (MNGP) Power Rerate Program.

This letter provides NSP's response to the staffs request Please contact Joel Beres, Monticello Licensing, at (612) 295-1436 if additional information is required.

William J. Hill Plant Manager Monticello Nuclear Generating Plant c:

Regional Administrator - Ill, NRC NRR Project Manager, NRC Sr. Resident inspector, NRC State of Minnesota, Attn: Kris Sanda J Silberg, Esq.

Attachments:

(1) Affidavit to the US Nuclear Regulatory Commission (2) NSP RAI Rasponse (3) GE NEDC-32498, Rev.1, " Reactor Pressure Vessel Power Rerate Stress Report Reconciliation for Monticello Nuclear Generating Plant" (4) GE NEDC-32647,"Monticello Cobalt Transport and Shutdown Drywell Dose Rate Model Calculation Results" 0/'

(5) Figures 9,10, and 12 of GE-NE-B1100683-1 (6) Load Histogram for Core Spray Piping / Safe End (Duty Map)

(7) EQ Profiles j

9709170009 970905 PDR ADOCK 05000263 b

P pog l

$Ihl5$llIII}$l}l@l(([llll war xe cusomecumonwurew ooc UM

UNITED STATES NUCLEAR REGULATORY COMMISSION NORTHERN STATES POWER COMPANY MONTICELLO NUCLEAR GENERATl; G PLANT DOCKET NO,50-263 Response to Request for Additional information Regarding License Amendrnent Request dated July 26,1996 Northern States Power Company, a Minnesota corporation, by letter dated September 5,1997 provides its response for the Monticello Nuclear Generating Plant to a US Nuclear Regulatory Commission (NRC) letter dated April 14, 1997, with the subject "Monticello Nuclear Generating Plant - Request for Additional Information on License Amendment Request Dated July 26,1996 Entitled ' Supporting the Monticello Nuclear Generating Plant Power Rerate Program"(TAC No. M96238)'." This letter contains no restricted or other defense information.

NORTHERN STATES POWER COMPANY By

/

William J. Hill /

Plant Manager Monticello Nuclear Generating Plant On this f day of _.%M.rb

/997 before me a notary public in and 10r said County, personally appeared William J. Hill Pfant Manager, Monticello Nuclear Generating Plant, and beiry first duly sworn acknowledged that he is authorized to execute this document on behalf of Northern Sutes Power Company, and that to the best of his knowledge, information, and belief the statements made in it are true, 6_ ___Wy Comm Exp Jan STEPHF R. BLEGEN

/

NOTARY PUBLIC MINNESOTA j_

Ste/ phen R. Blegen /

~ ' ' - - - - - - " -

31.2000; Notary Public - Minnesota Sherbume County My Commission Expires January 31,2000

• L97 EB C elSOFFICEWINwoRnTDIPLATEiit ti fxr 2

Electrical Systems 1,

- Information provided in Exhibit A (page A 24) indicates no change for normal conditions for temperature, pressure, and humidity inside containment for the power rerate conditions while Exhibit E (section 10 2.1.1) indicates a slight increase. Provide clarification.

NSP Response Exhibit E is incorrect, and the phrase 'and norraal" will be deleted. Except for radiation levels, normal environmental conditions in the drywell will not change as noted in Exhibit A page A 24. Containment pressure is regulated by procedure and will not change under rerate conditions, Humidity levels will remain the same as there is no operational change being made to affect humidity in the inerted containment atmosphere. Although a small increase in ambient temperature is expected, this increase is well within the capability of the drywell cooling system. Present administrative controls and procedures will assure that the bulk average drywell air temperature will stay within the limits of 135*F at rerate conditions.

The Exhibit E wording will be changed in NSP's revised license amendment request.

2.

Information provided in Exhibit A (page A-24) indicates no change for accident (design-basis accident / loss-of-coolant accident) (DBA/LOCA) conditions for temperature, pressure, and humidity inside containment for the power rerate conditions while Exhibit E (Section 10.2.1.1) indicates a slight increase. Provide clantication.

NSP Response Exhibit A is incorrect. It appears from Question 1 that some unintended administrative errors occurred in the description of the rerate changes in regard to equipment qua!ification.

Equipment required to be qualified per 10CFR50.49 are qualified to the bounding environmental conditions.

The bounding accident temperature conditions in the drywell used for environmental qualification are based on the small steam :ine break accidents. The bounding accident pressure conditions in the drywell occur during the DBA LOCA. The humidity for accident scenarios is assumed to be 100%. The humidity assumption is not changed for rerate conditions.

The Monticello environmental qualification central file references General Electric Report AE-083-0983,

• Extended Drywell Temperature Analysis, as containing the drywell accident temperature profile.- The central file references General Electric P.eport NEDO-30485, "Monticello Design Basis Accident Containment Pressure and Temperature Response for FSAR Update," as containing the drywell accident pressure profile. General Electric Report NEDO-30477," Safety Analysis of the RHR Intertie Line Monticello Nuclear Generating Plant" analyzed for the peak short term (30 second) cor,tainment response. This analysis reported a peak drywell pressure of 42.3 psig at 1.2 seconds. The environmentally qualified equipment inside containment was venfied to be qualified to the peak drywell pressure of 42.3 psig.

Recent NSF correspondence with the staff addressed environmental qualification of equipment at 1880 MWt. Containment response curves for drywell temperature and pressure at an initial power level of 1880 MWt was provided by letter dated July 16,1997. See " Response to Request for Additional Information Regarding Revision 2 to MNGP License Amendment Dated January 23,1997.* Power rerate results in a slight increase in the peak drywell temperature and in an extension of the temperature profile for the bounding 1880 MWt DBA LOCA case. See response to question 3.b below.

A comparison of the peak drywell environmental conditions for current and rerate power is presented in the table below. A improved containment model was used for the rerate evaluations and for re-evaluating containment response for current power (See Response to Question 50). Because of modelimprovements, the results from the revised containment response for 1670 MWt are different than the results for the EQ containment response of record. For comparison purposes, both sets of containment responses for the 1670 MWt case are shown in Table 21 below.

MM JI B C StWFFiCEWINWORIPIDtPt.ATD111i DOC 3

~

~ ~

Table 2-1 Containment Response Parameters RATED DBA LOCA SHORT TERM SBA-LOCA POWER LEVEL

• DRYVf_LL PEAK LOCA DRYWELL 4

PEAK TEMP DW PRESSURE PEAK TEMP 1670 MWt 282 F (1.2) 42.3 psig (3) 335'F (4) a 1670 MWt 286.8*F (5) 40.6 psig (5) 330'F (6) 1880 MWt 285.5"F (5) 39.5 psig (5) 331'F (6)

Notes (1) NEDO 30485 Table 1: Code HXSIZ with May-Witt decay heat 4

]

(2) NEDO - 32418. Table 3-1; Code HXSIZ with ANS 5.1 decay heat (3) NEDO -30477, Table 3-3 ; Code M3CPT with May-Witt decay heat i

(4) AE-083-0983, Table 1; Code SHEX with May Witt decay heat 1

F) GE NET 2300731-1 Table 3; Code SHEX with nominal ANS 5.1 decay heat (b) GE NET 2300731 1. Table D-1; Code SHEX with nominal ANS 5.1 decay heat

• Evaluations were conducted assuming 102% of the rated power level (e g.1703 MWt was assumed for the 1670 MWt case).

The Exhibit A wording will be changed in the revised rerate license amendment request.

4 3.

Slight increases in the current vcident (DBA/LOCA) and normal condit:3ns for temperature, pressure, and humidity forpower rerate are considered insignificant as stated in Exhibit E (section 10.2.1.1),

a. Define the shght increases fortemperature, pressure, and humidity

b. Explain why the slight increases are considered insignificant.

1) Has each piece of equipment been evaluated to ensure it is still qualified?

2) Explain why equipment remains quahhed.

c, Section 10.1, Exhibit E, states that these increases are weII within the margins in the existing environmental qualification (EQ) envelopes

1) Do these increases cut into test margins or do they cut into the margin between quahtication levels and actualpredicted profiles?

2) Dehne how margins are being cut.

Z NSP Response See Question 1 above for a discussion of normal conditions. The responses below address accident conditions.

a.

Dehne the slight increases for temperature, pressure, and humidity.

NSP Response WW IEB C h. tWFICBWINWOR&TDtPLATBt til DOC 4

_.. _ _ ~.

____.m l

The first row in Table 21 above shows the peak containment temperature and pressure conditions used as the basis for equipment qualification for the current power level. The table also shows that the current peak containment temperature and pressure used for equipment qualification bounds the temperature and pressure results for the 1880 MWt power level. For the short term condition, the EQ temperature profile is derived from the tetr perature response associated with the SBA-LOCA. The slight increase identifie ? in Exhibit E is based on the DBA/LOCA containment istoonse curves for temperature and pressure and ref.ects changes in non-limiting pressures and tempratures for the DBA/LOCA at long term conditions.

Humidity is assumed to be 100% for all power ca:,es.

b.

Dplain why the slight increases are considered insignificant.

NSP Response Certain accident profiles change under rerate conditions. As a result, the integrated exposure to temperature increases slightly, This slight increase, howcver, does not significantly affect thermal degradation and does not preclude qualification of the affected equipment.

Since portions of the 1880 MWt temperature response were not contained within the current power EQ temperature profile, an evaluation was performed to demonstrate quahfication. An equivalent integrated temperature evaluation for EQ equipment in containment was calculated using the Arrhenius methodology. This methodology was previously approved by the Staff by its SER for Monticello dated January 4,1983 ( See section 4 therein).

J In order to evaluate the differences, a drynell DBA temperature envelope was developed. The j

DBA temperature enve' ope was constructed by choosing points that bounded the 1880 MWt containment DBA temperature profile. The results from MNGP Calculation CA 97176 show that l

the equivalent temperature exposure time for the EQ temperature evaluation profile exceeds the equivalent temperature exposure time for the DBA temperature profile. Therefore, the existing EQ temperature evaluation profile bounds the DBA temperature profile, Portions of the 1880 MWt pressure response were not contained within the current power EQ pressure profile. This is considered acceptable since the current power peak containment pressure used for environmental qualification bounds the peak containment pressure at forate conditions, and the failure mode mechanisms associated with the prassure pararneter do not include time dependent aging effects.

The evaLiation above supports containment equipment qualification with the current envelopes of pressure and temperature.

1)

Has each piece of equipment been evaluated to ensure it is still qvnIslied?

NSP Response 1

Equipment located in areas where temperature, pressure and humidity requires environmental qualification (EQ) was evaluated to determine the effect of changes (if any) to the environmental profiles. For equipment located in containment, equipment qualified at 1670 MW remains quahfied at the rerate conditions since the profile used for qualification is still bounding. For areas outside containment an evaluation was completed that concluded that each piece of environmental qualified equipment would remain qualified at the 1775 MWt power level.

Prior to implementa+ ion of power rerate at MNGP, environmental quahfication files will be revisrd to reflect the envirorimental profile changes required by the power rerate program.

M97 JEB C WiSOFFK DWINW0lu>iDIPLATOiiil DOC 5

4 i

2)

Explain why equipment remains qualified.

NSP Response Equipment located :n areas where temperature, pressure and humidity conditions requires equipment qualification (EQ) was evaluated to determine the effect of changes (if any) to the environmental profiles. For the in containment area, equipment qualified at 1670 MWt remains qualified at the rerate conditions since the profile used for qualification is still bounding. For areas outside containment the environmental conditions tested to exceeds the expected environmental conditions at 1775 MWt power level with the one

(

exception noted. No new environmentally harsh areas are created at power rerate conditions.

Section 10.1, Exhibit E, states that these increases are well within the margins in the existing c.

environmental quahhcation (EQ) envelopes.

1)

Do these increases cut into test margins or do they cut into the margin between quahncation levels and actual predicted profiles?

NSP Response These increases do not affect test margins. DBA/LOCA containment response curves for temperature and pressure, as discussed above, contain changes in non-limiting pressures and temperatures for Qe DBA/LOCA at long term conditions as compared to the current power EQ containment response curves.

The margin between the predicted profiles ( the analytical response) and the EQ evaluation profile (a tool for EQ evaluation purposes that is a conservative approximation of the analytical response) has been reduced. The margin between the EQ evaluation profile and the test profile has not been reduced.

2)

Define how margins are being cut.

3

{

NSP Response Please see the response to 3.b and 3.c.(1) above.

4.

Provide an EQ Package for one piece or type of electric equipment thatis within the scope of 10 CFR 50.49 which demonstrates (1) continued qualification for the verate environment and (2) the process for establishing quahncation for the increased temperature. pressure, humidity and radiation levels forpower rerate.

l See the attached EQ file.

5.

On page A-58 it is implied that minor modincations (required to assure the continued qualincation of electrical equipment outside the scope of 10 CFR 50.49) are not considered unreviewed safety questions and thus will be implemented under the provisions of 10CFR 50.59 during implementation of power rerate.

10 CFR 50.59 requires, in part, that a proposed change be deemed to involve an unreviewed safety question if the probability of malfunction of equipment important to safety previously evaluated in the safety

^

analysis report may be increased. The increase in system temperature, pressures, orpower requirements identified to exist on page A-24 Exhibit A (no matter how minor) can be interpreted to increase the probability of malfunction of heat sensitive electrical equipment important to safety. Thus. any modification.

to electrical equipment due to increased tett forature, pressures, or power requirements associated with MNGP power rerate could be considered an unreviewed safety question. Explain why (or how) these future mod Scations (which have yet to be identified) should (or will) not be considered unreviewed safety questions.

MM JEB CwtsOFFICEwNWOltDTDtPLATEllilIXsc 6

1 l

- - - - - -. - - _ _ - ~. -

l l

l Electrical equipment outside of the scope of 10CFR 50.49 has been evaluated to ensure that it remains within appropriate design hmits. This equipment will operate at the same flow rates and pressures as currently allowod. Temperatures for ambient conditions will be maintained within original design limits.

The modification program requires the screening of each modification for 10CFR50.59 applicabihty. Page l

A-58 states

• minor system modifications are to be performed to enhance the capacities and capabilties of l

installed plant systems." Exhibit D lists those minor system modifications. None of these listed modifications are to equipment that requires environmental quahfication or are in the scope of the EQ piogram. This list of hardware changes does not involve electrical equipment important to safety and therefore would not be an unreviewed safety question.

The equipment referred to in Exhibit A page A 24 will be quahfied for its application and location d j

accordance with 10 CFR 50.49. This includes all equipment within the scope of 10 CFR 50.49.

i i

)

Page A 24 identifies that equipment quahfication can be met in almost all cases, NSP's evaluation identified the following components that require additional work.

One cable type that meets quahfication requirements at 1775 MWt power level, but that additional testing, documentation or cable replacement with a different cable type would be required at 1880 MWt power level One type of conduit seal wou!d have a reduced quahfied hfe in some applications at 1775 MWt power level, and would require replacement, reanalysis or retesting at 1880 MW power levei.

6.

Section 6 of Exhibit E indicatos that a Northem States Power (NSP) gnd stability analysis has been performed at 1775 MM to venly no significant effects on grid stabihty and reliabihty. Explain why there are "no significant effects on grid stabihty and reliabihty.'

NSP Response it is important to note that Monticello is not licensed to the requirements of GDC 17 of Appendix A to 10 CFR 50 and is not heensed to the s; ability enteria of any lEEE standards However, in light of the staffs recent concems with electric grid rehabihty and in accordance with good engineering judgment, NSP determined that a calculation to address these concerns was prudent. NSP Calculation CA 97-144,

' Summary of the Effects of MNGP Power Rerate on Transmission System Reliabihty and Stability

• has been completed and shows that the transmission system remains staole and rehable with MNGP initially operating at 1775 MWt for the following grid contingencies identified in IEEE Standard 765, " Preferred Power Supply for Nuclear Power Generating Stations.*

(1) Loss of the nuclear power generating unit (2) Loss of NSP's largest generating unit (3) Loss of the largest transmission circuit or intertie (4) Loss oflargest system load The acceptance enteris for stabihty and reliability is based on the design standards for bulk transmission system performance as delineated by the Mid-Continent Area Power Pool (MAPP). These standards include, among other requirements, steady state pre-contingency voltage limits of 0.95 to 1.05 pu and post-contingency voltage hmits of 0.9 to 1.1 pu.

NSP requests that this calculation not be misconstrued as a commitment to change the plant'c licensing basis in the future.

7.

Provide results of analysis which demonstrates that sufficient power will remain available and connected to safety systems from the offsite system (transmission network) immediately following reactor trip caused by LOCA when operating MNGP at 1775 MM for all expected modes of operation of the transmission network.

I M97 EB C %i10FHCE\\WINWORD rEMPL.ArEillli DOC 7

-.- - -~.--

.~

1 NSP Response Please note that the response for Question 7 is contained in the response to Question 10.

l 8.

Provide results of analysis (or otherJushncation) which demonstrates that there has been no reduction in margin (due to power rerale) between trip setpoints forloss of voltage or degraded grid voltage protective schemes installed on safety buses and transient voltage on safety buses that are expected following 1

reactor trip dut to a LOCA NSP Response j

Please note that the response to Question 8 is contained in the response to Question 10.

9.

Technicalspecifications will allow plant operation with the IR and 2R transformers operable while the 1AR transformeris out of service. It is not clear that the 1R and 2R transformers esca have sufficient capacity I

and capability to supply safetynlated loads for this mode of operation. It is also not clearif operability requirements need to be established for the automatic load shedding feature on the 1R transformer (or for the administrative procedures forlimiting load on the 1R transionner) for this mode of operation. Provide technical specification changes that preclude this mode of operation orprovide a system description, the results of analysis that demonstrale compliance with design-basis requirements, andproposed limiting conditions for operation (if applicable) for this mode of operation.

e NSP Respon_se s

Please note that the response to Question 9 is contained in the response to Question 10.

10.

Technical spectScations will allow plant operation with the 1R and 1AR transformers operable while the 2R transformeris out of service. Provide technical specification changes that preclude this mode of operatiotr

- orprovide a system description, the results of analysis that demonstrate compliance with design-basis requirements, and proposed limitin.y conditions for operation (if applicable) for this mode of operation.

NSP Response Please note that the responses for Questions 7 through 10 inclusive are contained in the response to Question 10.

The central issue to questions 7 through 10 is the effect of rerate upon the operation of the 1R transformer.

The 2R transformer has significant loading margin and does not approach any loading limitations due to the increase in loads due to rerate. The 1 AR transformer scheme includes a load shed that leaves only the safety related busesc The 1 AR loads are not affected by power rerate. In addttion to the 1R X winding loading issue discussed in detail in the response to Cuestion 11 below, the following issues describe the effect of rerate for 1R transformer as currently configured.

t The rerate loading analysis has focused more attention on voltage capabilities of the local 115 KV System.

Currently, the plant has adn.inistrative limits on minimum 115 KV system voltage which are dependent upon whether No.10 transformer is in service. The " strong grid' (Iow grid impedance) case is when No.10 transformer is in service to provide a lower impedance connection between the 345 KV transmission system and the 1R primary winding. The " weak grid" case is the case when No.10 transformer is not in service. Under the weak grid cases, the minimum 115 KV system voltage is required to be higher than the strong grid case. A minimum voltage requirement corresponds to that voltage sufficient to recover uteady state voltage above the degraded voltage reset point when the plant is on 1R and No.10 transformer is out of service. Under rerate conditions, the minimum voltage requirement increases above present limits, and this voltage may not be available from the grid unless special provisions are made.

NSP has been considering several options to address the above issue as well as the 1R X steady state loading issue discussed in the response to Question 11 below. As discussed in a September 4,1997 phone conversation with the staff, tap changes to plant transformers have excellent potential for resolving the 1R loading issues.

DW RB C WtSOFHCEWWWOR!ATEMPLATE\\llla rMIC 8

Currently, the no load tap on 1R is positioned to the 115,000-4160 position. The engineering staff is analyzing the effects of changing the 1R tap setting to the 112.125-4160 position which would result in higher secondary voltages for a given primary voltage. In conjunction with this change, new lower 115 KV grid voltage limits and a lower No.10 transformer LTC voltage control band would be established.

Preliminary calculations indicate that the new 1R tap position, when coupled with the revised 115 KV allowable voltage band and the revised No.10 transformer LTC voltage control band, would significantly improve load margin for 1R and would also serve to eliminate the operational dependencies for 1R on the availability of the 10 transformer. This chnge would therefore obviate any technical specification changes which may have been necessary under th, present configuration.

Due to the significant improvement in offsite source reliaoitity and availability which appears to be achievabic via the change in 1R tap settings and the new 115 KV operating limits, NSP intends to pursue the above changes outside of the ferate program. In order to address rerate issues, however, the supporting calculations will support terate loading conditions. When calculations for the new configuration at terate conditions have been formalized, a separate submittal will be made to provide background information on the tap settings and to provide supporting analyses from the load study for staff review. This submittal will more fully address Questions 7 through 10 for the 1R transformer. The expectation is that this information will be submitted by September 30,1997.

11.

Page 2 of 3 of Updated Safety Analysis Report (USAR) 8.2 Revision 12 states that the 1R transformeris of adequate size to provide the plant's full auxiliary load requirements. Exhibits A and E of the rerate submittal indicate that the MNGP design has been modified (and will be further modified as part of relate) such that the capacity of the 1R transformeris something less than the 100 percent capacity required by the licensing basis for MNGP documented in the USAR. Provide the descdption and analysis for this modification.

NSP Response 9

With existing plant auxiliary system loading and substation voltage limits, the 1R transformer is of adequate size to provide the plant's full auxiliary load requirements. The 1R transformer is rated at 37.33 MVA and has 2 secondary windings, designated X and Y, each of which are rated for 18 67 MVA. The X winding supplies the Reactor Feed Pump and Reactor Recire MG set motors and is loaded to a higher level than the Y winding.1R has a relatively high impedance which results in a voltage drop across the X winding of the transformer. The plant's load study calculations indicate that with the USAR minimum source voltage of 117.5 KV present on the 115KV system which supplies the 1R primary winding, 3885 V would be available on the loaded X winding secondary under steady state non-accident conditions.

As transformer MVA ratings are based upon secondary winding current carrying limitations, it is Monticello's practice to proportionately derate the transformer MVA rating directly with the percent in which loaded secondary voltage is below transformer rated se.edary voltaae. For the 1R X winding, this results in a derate to 3885 / 4160

• 100 or approximately 93.4% (17.4 MVA). For existing plant full auxiliary load conditions, the X winding is loaded to 16.7 MVA or 89.5E As the degree of loading (89.5%) is less than the derated MVA limit (93.4%),1R is of adequate size to provide present full plant auxiliary load requirements.

Under rerate conditions, the higher loading on the feed pump motors results in a calculated X winding loading of 17.5 MVA (93.7%). This increased loading also results in an increased voltage drop across the 1R X winding such that approximately 3853 V will be present when the 115KV source voltage is at the USAR minimum value of 117.5 KV. Applying the reduced voltage derate as discussed above, the X winding MVA limit drops to 3853 / 4160

• 100 or approximately 92.6% Thus for the present 1R configuration and substation voltage limits, rerate loading levels would result in X winding loading (93.7%)

in excess of the rarate derated MVA limit (92.6%). The limits referred to in Exhibits A and E of the rerate submittal were intended to inform the staff of NSP's intent to establish controls such that the 1R X winding would not be loaded beyond its derated MVA limit.

NSP intends to obviate this 1R X winding loading concem via the change in the 1R primary tap setting and establishment of new 115KV system voltage limits as discussed in the response to Questions 7 through 10 above. Preliminary calculations indicate that the higher available voltages which will be available on the 1R M97 whee J ilCEN5E'JOEL JtERATE1ARillin rMM' 9

secondary windings willlower the amount of MVA derate required such that 1R X winding loading would be less than the derated MVA limit under rerate conditions.

t%97 whse J tlCENsE'JtMiLMRATEtAR.Ilit (MW 10

Materials Engineering 12.

Provide an assessment of how the propose <1 thermal uprate will affect the end of hie (EOL) upper shelf energy analysis for Vessel Plate No 1-15 (Heat No. C2220-2), and the equivalent margins analyses for Vessel Plates I-16,1 17, and I-14 (Heat Nos. A0946-1, C2193-1, and C2220-1) and the beltline vessel welds (no heats given). Include appropriate calculations, figures, or references demonstrating continued comphance with the requirements of 10 CFR Part 50, Appendix G, under the proposed increased power conditions and also updated values for the 1/4T fluence and the upper she!! energies for beltline materials of the MNGP reactorpressure vessel (RPV) at EOL.

NSP Response Table 121 lists the conservatively estimated EOL 1/4 T fluence value at a power uprate of 112.6% (1880 MWt) to be 3.9?E+18 n/cm2. Using Figure 2 from Reg. Guide 1.99 Rev. 2 and the fluence value of 3 99E+18 n/cm2, the percent decrease in shelf energy can be determined. Listed below are the results of the decrease in USE (Upper Shelf Energy) for the Monticello beltline materials based on uprate. Table 12-2 lists the estimated EOL 1/4 T fluence at 1670 MWt without taking rerate into consideration.

TABLE 12-1 (Effects on USE at Rerate Condition )

s Plate /

Heat INITIAL

%Cu E.O.L.

%Decr.

E.O.L Weld Weld Type TRANS. USE FLUENCE USE TRANS.

E+18 n/cm2 USE l 14 C2220-1 59.1 (1)

,17 3.99 20.9 46.7(2) 1-15 C2220-2 71 0 (3)

.17 3.99 20.9 56.2(4) 1-16 A0946-1 59.1 (1)

.14 3.99 18.5 48.2(2) 1-17 C2193-1 59.1 (1)

,17 3.99 20.9 46.7(2)

Wald 87.0(1)

.10 3.99 19.4 70.1(4)

Notes (1) initial Transverse USE values were obtained from NEDO-32205-A,(10CFR50 Appendix G Equivalent Margin Analysis for Low Upper Shelf Energy in BWR/2 Through BWR/6 Vessels)

(2) Acceptable value per NEDO-32205-A in 135 ft-Ibs.

(3) From response to GL 92-001 (Reactor Vessel Structural Integrity). This value is also identified in the Reactor Vessel Internals Data Base.

(4) Acceptable Value per Appendix G of 10CFR 50 is t 50 ft-lbs.

TABLE 12-2 (Effects on USE at Current Conditions).

Plate /

Heat INITIAL

%Cu E.O.L

%Dect.

E.O.L W eld Weld Type TRANS. USE FLUENCE USE TRANS.

E+18 n/cm2 USE l-14 C2220-1 59.1 (1)

.17 3.82 20.6 46.9(2) 1-15 C2220-2 71 0 (3)

.17 3.82 20.6 56.4(4) 1-16 A0946-1 59.1 (1)

.14 3.82 18.1 48 4(2) l-17 C2193-1 59.1 (1)

.17 3.82 20.6 46.9(2)

Weld 87.0(1)

.10 3.82 19 0 70.5(4)

M97 JEB C WISOFFICEiWIN%ORIPTEMPLATEM111 DOC 11

Notes (1) Initial Transverse USE values were obtained from NEDO-32205-A, (10CFR50 Appendix G Equivalent Margin Analysis for Low Upper Shelf Energy in BWR/2 Through BWR/6 Vessels)

(2) Acceptable value per NEDO 32205-A is $35 ft lbs.

(3) From response to GL 92-001 (Reactor Vessel Structural Integrity). Also vahe is identified in Reactor Vessel Internals Data Base.

(4) Acceptable Value per Appendix G of 10CFR part 50 is t 50 ft-lbs.

Given the above, the upper shelf energies for 1-14,1 15,1 16,1 17, and the limiting weld are all above the acceptable upper shelf acceptable energy levels for both rerate and current conditions. In addition, a comp 3rison of Table 12-1 to Table 12 2 shows that the difference in USE values are insignificant.

Question 12 References

1. NEDO 32205-A,(10CFR50 Appendix G Equivalent Margin Analysis for Low Upper Shelf Energy in BWR/2 Through BWR/6 Vessels.

2. Reg. Guide 1.99 Rev 2, (Radiation Embrittlement of Reactor Vessel Materials).

3. Response to GL 92-001, (Reactor Vessel Structural integrity) 4, Reactor VesselIntegrity Database Summary File for Upper Shelf Energy, t 3.

Provide an assessment of how the uprated conditions will affect the scope or schedule of the surveillance capsule wnthdrawalprogram (10 CFR Part 50, Appendix H Program) for the MNGP RPV.

NSP Response No changes in the Appendix H program met,4 red because there are no significant changes in criteria important for Appendix G. The relevant criteria are the system pressures and the vessel fluerr.:e at End-of.

Life. Or more precisely, the end of the current license. The system test pressure used for the current P.T curves is 1100 psi, which bounds both the current nominal operating pressure and the nominal operating pressure for rerated conditions. Any revisions to these P-T curves would similarly be expected to bound operating pressures for both current and rerated conditions. The neutron fluence calculated for the end-of-license time period with the addition of rerated power operation to 1775 MWt is very close to the same numerical value of neuan fluence that was calculated for a 40-year plant license operating at 1670 MWt.

This occurs because the operating capacity factor for Monticello to date is about 75% compared to the original assumption of 80%. Thus operation at 106.3% of 1670 MWt for the balance of the current license

- period results in a total integrated fluence level very close to that corresponding to an average 80%

capacity factor over a 40-year license operating at 1670 MWt the entire time period.

14.

Provide a more detailed evaluation of the effects caused by extended power uprate on reactorintemals (i.e., expand on yoursubmitted determination of the effects).

Power rerate has only a limited effect on the reactor intemals. Most of the operational parameters defining the operating environment for the reactor intemals are unchanged for power rerate operation. The maximum reactor operating ptessure is unchanged. The core flow operating range at rerated power is bounded by the core flow range at current power, The maximum recirculation drive flow is unchanged for power rerate operation because the recirculation system is currently at the maximum flow during increased core flow operation. The downcomer and core inlet enthalpy range at terated power is bounded by the enthalpy range at current power. The maximum steam generation in any single fuel bundle is unchanged for power rerate because the bundle thermallimits remain the same for power rerate.

The primary effect of power rerate operation is a slight increase in the reactor intemal pressure differences (RIPDs). The increase in RIPDs is due to the higher two-phase flow losses caused by the increased steam generation in the core. The reactor intemals have been evaluated for the higher RIPD loading at normal, upset, and faulted conditions due to power rerate in NEDC-32546P, Section 3.3.2.1.

Further information on the reactor intemals stress evaluation is presented below in response to Question

17. This response shows the expected changes in stress levels for the highest stressed reactor internals locations. See also NEDC-32498, Revision 1, provided in response to Question 19 for further information W HB C AW*FICEMIWORDTEMPLArDillt [XX' 12

on reactor internals stress evaluations done for Rerate. This is consistent with the approach defined in Section 1.3 of NEDC-32424P,

• Generic Guidelines for General Electric Boiling Water Reactor Extended Power Uprate"(ELTR1).

The steam separators see a higher inlet quality and the steam drye.s see a higher flow velocity as a result of the increased steam generation in the core. The steam dryer and separator performance evaluation is presented in NEDC-32546P, Section 3.4.

Intergranular stress corrosion cracking and erosion / corrosion have been addressed generically for the reactor internals in Section 3.6.1 of NEDC-32523P, " Generic Evaluations of General Electric Boiling Water Reactor Extended Power Uprate,"(ELTR2) for extended power uprates up to 20%.

Provide a more detailed evaluation of the effect caused by extended power uprate on components exposed to single-and two-phase fluid flow.

NSP Response As described in the response to Question 14, the steam separators and steam dryers are the only reactor internals components that experience a change in the fluid flow conditions for power rerate. The steam separators see a higher inlet quality and the steam dryers see a higher flow velocity as a result of the increased steam generation in the core. Intergranular stress corrosion cracking and erosion / corrosion have been addressed generically for the reactor internals in Section 3.6.1 of NEDC-32523P,

• Generic Evaluations of General Electric Boiling Water Reactor Extended Power Uprate," (ELTR2) for extended power uprates up to 205 697 JEB C StNOFFKEMINWORIATEMPL ATPilli NC 13 1

Mechanical Engineering 16.

In reference to Section 2 5.1 of Exhibit E, provide an evaluation of the controlrod4 rive mechanism with regard to the stress at,d fatigue usage as a result of the 6 3 percent power uprate. Also, provide the allowable code limits for the cnticalcomponents evaluated, and the code and code edibon used for the evaluation. Ildifferent from the code of record. jusbly and reconcile the differences.

NSP Response The control rod drives (CRDs) are in direct communication with the reactor pressure vessel and are exposed to reactor pressure and temperature. Since the reactor pressure and temperature remain unchanged for a power uprate of 6.3%, the original design conditions for the CRDs are applicable.

The CRDs have been designed for 1250 psig which is higher than the bottom head pressure of 1045 psig for normal and power uprate trA.ctor conditionc. The components of the CRD mechanism, which form part of the primary pressure boundary, have been designed in accordance with the applicable ASME B&PV Code, Section 111, The limiting component of the CRD mechanism is the indicator tube which has a ca!cula'ed primary membrane plus bending stresses of 20,790 psi. The allowable stress is conservatively specified as 26,060 psi (i.e.,1.5 Sm). The maximum stress on this component results from the maximum CRD internal hydraulic pressures of 1750 pt.ig caused by an abnormal operating condition.

The CRDs have been designed for temperatures of up to 575*F, which is higher than the bottom head temperature of 530*F for normal and power uprate,onditions. The analysis for cyclic operation of the CRD was coriservatively evaluated in accordance with applicable requirements specified in the ASME B&PV Code. Section lit. For example, when considering the loadings resulting from scram with a leaking scram discharge valve, scram with a failed buffer, and scram without CRD cooling water flow, the limiting component was found to be the CRO main flange. The fatigue usage factor is 0.15 which is less than the allowable limit of 1.0, All requirements are satisfied even when considering the increased power uprate vessel bottom head pressure, thereby satisfying the peak stress intensity limits governed by fatigue.

Since the reactor pressure and temperature remain unchanged for power rerate, the original evaluation of the CRD for stress and fatigue usage, described above, remains applicable.

17.

In regard to Section 3.3.2, provide the maximum calculated stress at the cnticallocations of the reactor intemal components, the allowable code limits, and the code and code edition used in the evaluation for the power uprate 11different from the code of record, providejustification.

N3? Response The evaluation of the reactor internals uses ASME Boiler and Pressure Vessel Code Section ll1 as a guide for congn 3cceptance criteria; no specific edition or addenda was specified. The specific applicable Code Edition for the reactor pressure vessel, including the shroud support, is the 1965 Edition with Addenda to and including the Summer 1966 Addenda.

The structuralintegrity of the component is demonstrated by comparison with applicable allowable stresses.

The stress results for three of the highest stress components are shown in Table 17-1.

I e na cmsOFFIC0WINWORurEMPLATDill) DOC 14

Table 17-1: Selected Results, Reactor intemals Stress Summary COMPONENT MAX.

UPSET FAULTED STRESS LOCATION Ma.:

Max.

Allowable Max. Value

Max, Allowable Value Value current Value Current Rerate psi psi uprate psi psi psi psi Shroud Shroud 15,575

<22,900*

34,t60 Not 22,900 69,900 Support Supt.

Determined Cylinder Core Beam 5,170 5,560 6.200 6,748 6,748 12,400 Plate Buckling Stress Jet Pump Riser Elbow 9,052 10,156 38,025 29,719 33,344 60,840 Assembly Stresses

• This value is less than allowable by inspection of the faulted value. No calculation was performed.

" The allowable values are based on the code of record.

18.

In Section 3.3.2 2, an assessment of flow-induced vibration of the reactorintemal components due to power uprate is performed to address the increase in steam product in the core, the increase in the core pressure drop, and the increase in the recirculation pump speed. In that assessment, the vibration levels were estimated by extrapolating the recorded vibration data at Monticello and by using the operating experience of similarplants. Provide a sample evaluation and the basis for using the operating experience of similar plants.

NSP Response During the start up flow induced vibration testing of the Monticello Nuclear Power Station, vibration data for the reactor intemal components were recorded during the cold flow pre-operational testing and hot flow power operation of 50%,75%, and 100% rod line testing. Operating conditions during these tests included steady-state balanced flow two pump operation, unbalanced flow, single loop operation, and transient flow conditions. The observed vibration responses were all well below the acceptance criteria limits under all tested conditions. See Section 3.6.3.1 of the Monticello USAR for additionalinformation on measurements, acceptance criteria and the basis for using operating experience from similar plants.

The acceptance criteria comprise a set of frequencies and corresponding allowable amplitudes derived from an analytical model. An acceptance criteria of 100% corresponds to a peak stress intensity of 10,000 i

psi due to vibration. At this stress level, sustained operation is allowed without incurring any fatigue usage.

When the stress level exceeds 100% of acceptance criteria, the component is subjected to fatigue usage.

Vibration data obtained from operating BWR plants have shown conclusively that, for a broed range of BWR sizes the conservatively chosen long te"m steady state vibration criteria are not violated for normal balanced flow conditions. Since all BWR jet pump plants are geometrically similar, it la not expected that there is any signi'icant difference in vibration response of plants in various limited size ranges. Therefore, a ccmplete series of vibration tests is not necessary for individual units. Monticello is one of the units that was instrumented and measured by General Electric to obtain vibration testing data.

Two sets of extrapolations beyond the bounds of the original test data were made to account for the potential effects of rerate operation. One extrapolation was made to determine the effect of increasing core flow with a nearly constant rod line and the other to determine the effect ofincreasing rod line with nearly constant core flow. Figures 9,10 and 12 of GE-NE-B1100683-1 (attached) are provided as an example. In MM JEB C at%OFFM'0WrNWORDr1MPL ArEiniII Dir 15

b l

these figures test data existed for a 50% rod line to 108% core flow, and therefore no extrapolation was needed for this case. Extrapolations that were made were shown with dashed hnes on the figures.

Since the start-up tests were not performed with operating conditions in the proposed power rerate region, the expected vibration responses were estimated from the existing start-up test data, Extrapolation from the start up test conditions to the power rerate operating condition prevides a reasonable estimate of the expected vibration level. Two sets of extrapolations were made: 1) a set to determir,e the Wet of increasing core flow with nearly constant rod line, and 2) a set to determine the effect of inert. sing rod line with nearly constant core flow.

The extrapolation procedure is an accepted engineering practice and has also been applied to the flow induced vibration evaluations for all other GE BWR power rerate programs. Power rerate operational testing has been conductad at one other GE BWR plant. Reactor intemal components with similar designs as Monticello have shown no significant increase Li flow induced vibration and have also shown that vibration levels are within the extrapolated values.

During the start up flow induced vit ration tes ;g of the Monticello Nuclear Power Station, vibration data for the reactor internal components were (ecorc,J during the cold flow pre-operational testing and hot flow power operation of 50%,75%, and 100% rod line testing Operating conditions during these tests included steady-state balanced flow two pump operation, unt:alanced flow, single loop operation, and transient flow conditions. The observed vibration responses were all well below the acceptance criteria hmits under t il tested conditions.

Since the start-up tests were not performed with operating conditions in the proposed power rerate region, the expected vibration responses were estimated from the existing start-up test data. Extrapolation from the start-up test conditions to the power rerate operating condition provides a reasonable estimate of the expected vibration level. Two sets of extrapolations were made: 1) a set to determine the effect of increasing core flow with nearly constant rod line, and 2) a set to determine the effect of increasing rod line with nearly constant core flow.

The extrapolation procedure is an accepted engineering practice and has also been apphed to the flow induced vibration evaluations for all other GE BWR power rerate programs. Power rerate operational testing has been conducted at one other GE BWR plant. Reactor internal components with similar designs as Monticello have shown no significant increase in flow induced vibration and have also shown that vibration levels are within the extrapolated values.

19.

In reference to Sections 3.3 and 3.3.2, provide the methodology, assumptions, and loading combinations used for evaluating the reactor vessel and intomal components with regard to the stresses and fatigue usage for the power uprate. Were the analytical computer codes used in the evaluation different from those used in the originallicensing-basis analysis? If so, identify the new codes used and providejushfication for using the new codes and state how the codes were quahlied for such applicabons.

NSP Response Reactor Vossel The methodology, assumptions, and loading combinations used for evaluating the reactor vessel components with regard to the stresses and fatigue usage for power uprate are provided in GE Report No. NEDC-32498 Revision 1 (attached). No new analytical computer codes were used in the evaluation.

ReactorIntemals This evaluation primarily considers the concem that operation at rerate power level may subject reactor intemal components to greater reactor it.ternal pressure d:fferentials (RIPDs) than previously considered. The operation 6 basis earthquake and maximum earthquake loadings on reactor intemal components are not affected by power rerate, but these loadings must be considered in appropriate combination with those loadings which are affected by rerate operation.

Load combinations and stress limits from the structural criteria for reactor intemals in Section 12.2.1.4 of the USAR were utilized for the power rerate evaluation. The stresses or loads for the W M B C WWHCP%iNWWU7 rtmp 1 ATollliIKY 16

{

0 major reactor intemal components were evaluated by either confirming that terate load combinations are bounded by previous analyses or scaling stresses using conservedve load ratios from these analyses, in sorae cases, previous analyses were repeated as required to demonstrate acceptance. No new computer codes were used in these evaluations.

20.

In reference to Section 3.6, provide the methodology and assumptions used for evaluating the reactor coolant piping systems for the power uprate. Also, provide the calculated maximum stress, critical locations, allowable stress limits, and the code and code edition used in the evaluation for the power uprate. Ildifferent from the code of record, justify and reconcile the differences.

NSP Response The reactor coolant pressure boundar) piping design is based on the reactor pressure vessel (RPV) design tamperature and pressure. There are no changes in the RPV design temperature and pressure due to power rerate as stated in Section 3.2 of the power rerate license amendment request. Therefore, the existing piping design and pipe stress analyses for the reactar coolant pressure boundary piping for the maa steam pipir.g, feedwater piping, CRD piping, RPV bottom head drain line, RPV head vent line, RCIC steam piping, Core Spray piping, RHR piping, HPCI steam piping, SRV discharge piping and RWCU piping bound the power rerate cond'tions.

Additional evaluation of the reactor recirculation system piping is provided below. Refer to the response to Questions 21 and 27 for a discussion of the main turbine stop valve closure loads impacting the main steam lines as well as the HPCI steam piping and SRV discharge lines which are attached to th6 main steam i

lines. For the evaluation of the main steam piping, feedwater piping RCIC piping, Core Spray piping, RHR piping and HPCI piping not within the reactor pressure boundary, refer to the response to Question 24. The SRV discharge piping evaluation for Mark l Containment hydrodynamic loads is contained in the response to Question 26.

The design of the recirculation system piping including the applicable codes and stress limits is described in Sections 12.2.1.9 and 12.2.2.12.7 of the USAR. Piping and supports were designed for pressure, temperature, seismic and thermal transients due to normal and upset conditions. The operating parameters for the recirculation system such as temperature, pressure and heatup/cooldown rate wul remain unchanged under rerated conditions. Therefore, power rerate has no impact on the design thermal transients, and the existing fatigue analyses of the recirculation piping bounds the power rerste conditions.

The existing recirculation piping stress analyses used the code of record as described in the USAR.

The recirculation system piping was designed to ensure flow induced vibration stresses under steady state and transient vibration do not exceed acceptable limits. Under power rerate conditions, the recirculation system operating conditions that influence system vibration such as flow and pump speed will not be increased beyond the flows and pump speeds that have been used in the past. Therefore, the existing recirculation system vibration analysis bounds power rerate conditions.

In conclusion, there are no changes to the reactor pressure boundary piping design parameters for power rerate, therefore, the existing piping evaluations bound power rerate conditions. The current calculated maximum stress, critical locations, allowable stress limits, tne code and code edition used for piping evaluation are not changed for power rerate.

21, Discuss the analytical methodc!ogy and assumptions used in eve'uating pipe supports, nozzles, penetrations, guides, valves, pumps, heat exchangers. and anchors at the power uprate conditions. Were the analytical computer codes used in the evaluation different from those used in the originallicensing-basis analysis? If so, identify the new codes and provide justification for using It's new coues and state how the codes were qualified for such cpplications.

NSP Response As stated in the response to Question 20, there are no enanges in the RPV design temperature, pressure, or severity of recirculation system thermal transients due to normal and upset conditions under power rerate conditions. Except for main steam stop valve loads, power rerate has no impact on piping within the

% W RB J 4tsof flCE'wlN%fMPTMtP1.ATEllit DOC 17

mactor coolant pressure boundary (RCPB) including pipe support loads, nozzles (e g. RPV nozzles),

penetrations (e g. containment penetrations), guides, valves, pumps and anchors. The existing temperature and pressure design values for the above equipment bound the power rerate conditions.

Consequently, no computer codes different than those used in the originallicensing. basis analysis were used in RCPB power rerate evaluations.

Methodology and assumptions used in the analysis of piping wr.s consistent with the descriptions provided in USAR Section 12 with piping issues predominantly described in US AR Sections 12.2.1.4,12 2.1.10 with all subparts, 12.2.2.10 and 12.2.2.12 with all subparts. No changes were made to methodology or assumptions previously communicated with the NRC. The predominant impact for some systems not associated with the RCPB was a small change in temperature used for design values. Further discussion of non-RCPB piping analysis is provided in the response to Question 24.

1here is an increase in main turbine stop valve closure loads due to rerate which impacts the main steam system piping back to the reactor vessel nozzles. Refer to the response to Question 27 for a discussion of the main turbine stop valve closure loads on the main steam, SRV discharge and HPCI steam piping, pipe supports and associated equipment. There are no heat exchangers within the reactor pressure boundary piping. The evaluation of the heat exchangers not within the reactor coolant pressure boundary is d:scussed in the response to Questions 24,28 and 29.

22.

The power uprate fatigue cumulative usage factors (CUFS) (shown on Table 3-4) for the reactor vessel are given in thrse locations: at the refueling bellows skirt, the closure region bolts, and the recirculation onlet nozzles. Provide CUFs for the limiting components of the reactor coolant piping systems. Discuss how the calculated CUFs for the reactor vessel and piping components compare to the CUFs resulting from the actualloading cycles based on the data recorded during plant operation.

NSP Response (Par,1) CUFs for the Limiting Components of the Reactor Coolant Piping Systems Monticello piping was originally designed and installed in accordance with USAS B31.1.0-1967 which did not require a fatigue analysis. The following reactor coolant pressure boundary piping systems were replaced, or added, since initial plant operation. The. modifications for the replaceme.,ts, or additions, considered the requirements of ASME Boiler and Pressure Vessel Code, Section 111, Class I and as such now have fatigue evaluations.

RHR intertie line (added in 1984)

RHR shutdown cooling cupply and retum lines (replaced in 1984)

Recirculation system lines (replaced in 1984; Core spray hnes (replaced in 1986)

Fatigue analyses was performed using a duty map that provided a bounding set of assumed thermal cycles that may occur over the hfe of the plant. Fatigue evaluations for the replacement piping, based on these postulated thermal cycles, resulted in calculated usage factors of approximately, but less than 1.0.

Conservative sets of assumed design cycles were used for the fatigue analyses. Aa example of the core spray duty mao is attached.

(Part 2) Comparison of Actual Loading Cycles and Calculated Fatigue:

For cycle counting, actual plant cycles were conservatively classified relative to design cycles. That is, the actual thermal transient is typically less severe than the assumptions shown on the duty map. The core spray duty nvp assumes a 100 *F/hr heatup rate followed by a cooldown rate derived from a core spray injection from ope.ating conditions. For countMg purposes at Monticello, a normal cooldown cycle of less than 100 'F/hr is conservatively considered to be equal to a core spray injection. For piping, this M97 Rit f $150f f K0%NWORfitEMPL.ATEllit t DtM' I8 I

conservative counting of experienced cycles showed a maximum CUF of 0.25 for the time period from 1986 to date for core spray piping. Extrapolating the Core Spray data, a life of over 40 years (110 25 x 11 years) is predicted before a usage factor of 1.0 is reached. This significantly exceeds the end of the current operating license, it should be pointed out that terate operation does not impact fatigue on plant piping systems that have fatigue evaluations since there is no change in design temperature, pressure or thermal i.

rate of change for the piping.

Similarly, the fatigue analyses for the vessel components in Table 3-4 are based on conservative sets of assumed design cycles. The current maximum actual usage has occurred for the refueling bellows skirt.

Based on corcervative countbg of experienc4J cycles, fatigue usage is approximately 0.50 after 26 years of plant operation. By extrapolating the experienced cycles, a life of 26 additional years is indicated until a fatigue usage factor of 1.0 is reached Due to the bias of increased startup/ shutdown cycles experienced during early plant operations, this extrapolation is conservative.

Each of these calculations and extrapolations is based on conservative analysis and conservative characterization of experienced cycles. Additional margin in calculated fatigue is avai'ebie if more detailed analysis and less conservativo c5racterization of experienced cycles are used.

23.

Discuss the operability of safe ly-related mechanical components (i e., valves and pumps) affected by the power uprate to ensure that the performance specifications and technical sp,scification requirements (e.g.,

flow rate, close and open times) will be met for the proposed power uprate. Confirm that safety related motor-operated valves (MOVS) will be capable of pehrming their intended functions) following the power uprate including such affected parameters as fluid how, temperature, pressure and differentialpressure, und ambient tempora;ure conditions. Idenbly mechanical components for which operability at the ut. rated powerlevel could not be confirmed.

NSP Response (Part 1) As stated in Section 4.3 uf Exhibit E of the LAR, the ' current ECCS system performance requirements were used in the power rerate analysis.' There were no changes in flow rates, pump differential pressures or valve operating times from what is used in the current 1670 MWt calculations.

HPCI and RCIC were evaluated for operation at a RPV pressure of 1142.3 psig. This is equivalent to the SRV opening setpoint of 1109 psig with a + 3% as-found tolerance as allowed by Code. The nominal SRV setpoint of 1109 psig will not be changed for rerate. The low-low set SRVs have logic that, following a scram, opens one SRV at 1052 psig, a second SRV at 1062 psig, and a third SRV at 1072 psig. A high reactor pressure scram occurs at 1056 psig, Since the low low set SRVs and the reactor high pressure scram logic are kth safety related and single failure proof, reactor operation at the SRV overpressure setpoint of 110S pig would not occur for design basis accidents and normal reactor transients. HPCI and RCIC technical specifications cased on an RPV pressure of 1120 psig are adequate and no changes are needed.

Peak primary containment pressures actually dactine slightly with rerate. While suppression pool -

temperatures do rise, the resultant temperatures are within the capabilities of pumps and valves that communicate with the suppression pool.

There were no mechanical components for which operability at the uprated power level could not be confirmed, (Part 2 ) The response to this question as it applies to MOVs is in progress and will be submitted at a later date.

24.

In reference to Section 3.13. list the balance-of-plant (BOP) piping systems that were evaluated for the power uprate. Discuss the methodolo,7y and assumphons used for evaluating BOP piping. components, and pipe supports, oczzles, penetrations, guides, valves, pumps, heat exchangers and arcchoc. Were the analytical computer codes used in the evaluation different imm those used in the original design-basis MH KB C SMOFFICIWINWORD IDIPL ATin til DOC 19

analysis ? If so, identify the new codes and provide justincation for using the new codes and state how the codes were qualined for such applications.

t NSP Response Balance-of-plant piping (BOP) was reviewed to determine the impact of power rerate on plant piping and equipment. The following BOP piping systems were determined to be affected by power rerate.

Main Steam Residual Heat Removal High Pressure Coolant injection Reactor Core isolation Cooling Feedwater Condensate Feedwater Heater Piping Emergency Service Water RHR Service Water Service Water Fuel Pool Cooling & Cleanup p=

Reactor Building Closed Cooling Water Miscellaneous Steam Piping including Extraction Steam and Turbine Moisture Separator Piping Circulating Water Evaluation of those portions of BOP piping attached to the torus, torus penetrations, and valves and pumps that may be affected by Mark I containment hydrodynamic loads as well as peak suppression pool temperature is discussed in the response to QueWon 26.

The increase in temperature, pressure, and flow due to power rerste was determined for the affected BOP piping. Piping and supports were evaluated in accordance with requirements of USAR Section 12, The results show that all piping and all but a few pipe supports are within the applicable code allowable limits under rerate conditions (See response to Question 25 for appicable ccdes). Minor modriications to supports are required as described in the response to Question 30.

All safety related piping and some non-safety related piping was analyzed using the licensing basis compuMr code. Piping evaluations were performed using the PISTAR program which was approved for piping applications by a Staff Safety Evaluation Report (SER) dated September 11,1985. Piping was analyzed for deadweight, pressure, thermal expansion and Safe Shutdown Earthquake (SSE) loads in accordance with USAR ;oction 12. Pipe supports were qualified per applicable codes and standards.

Examples of non-safety related piping systems analyzed in the above evaluation include If,3in steam, RCIC, feedwater, portions of condensate, fuel pool cooling & cleanup, and service water, Certain system piping is not affected by power rerate conditions. The impact of power rerate on RBCCW and circulating water systems was determined to be insignificant.

The remaining non-safety related piping was evaluated using the Algor PipePlus computer code or by using comparative evaluations. The Algor PipePius computer program was used to analyze portions of condensate piping, feedwater heater / cooler associated piping, and turbine extraction steam piping. Piping was analyzed for deadweight, pressure, and thermal expansion in accordance with USAR Section 12. The Algor computer code was procured from Algor inc, with the trade name of PipePlus as QA software for analysis of safety related piping systems. Algor Interactive Systems, Inc. has verified the adequacy of PipePlus by compuing results with the NUREG/CR-1677 benchmark, which provides the confidence to justify its use as a piping analysis tool for rerate.

The remaining condensate, feedwater heater / cooler associated piping, turbine extraction steam and turbine moisture separator piping wa: 3 valuated using comparative evaluations. This method of analyses was considered to be adequate since the increase in the above piping operating temperatures and pressures at rerated conditions is insignificant, and the existing pipe stress analyses show large margins exist in piping stresses and pipe support loads. The existing piping stresses were increased in proportion to the increase in piping temperatures and pressures at terated ronditions. The new calculated stresses were compared to 9497 JEB C \\M%Of F10b%IN% ort 1TEMPl.ATMitii DOC 20

the code allowables. Pipe supports were evaluated using a similar approach by comparing calculated support loads at power rerate and comparing them to the support design loads. Pipe stresses were found to be within the code limits. Support loads vere found to be within the original support design loads. The results show that significant margin exists in pipin0 and supports at rerated conditions.

Equipment nozzles were evaluated to ensure that the calculated piping reaction loads under rerate conditions are less than the piping reaction loads from the original design analysis or less than manufacturer specified allowables. In some cases where nozzle allowable loads were not available, certain non-safety related nozzles were qualified using standard analytical methods such as generally accepted formulae for cylindrical shells with attachments. Torus penetrations were qualified in accordance with the requirements of NUREG-0661 and the ASME Code, Section 111 as identified in Section 12 of the USAR. All equipment nozzles were found acceptable for power rerate conditions Design pressure and temperature ratings for pumps, valves and heat exchangers were found to envelope power terate conditions.

25.

Provide the calculated maximum stresses for the critical BOP piping systems, the allowable limits, the code of record, and code edition used for tlie power uprate conditions. If different from the code of record Justify and reconcile the differences.

NSP Response The maximum pipe stress increases for BOP piping are shown in Table 3-5 of the power rerate license amendment request. The maximum piping stresses are below the code allowable limits as stated in USAR y

Section 12. The construction code for the original plant piping systems was USA Standard Code for Pressure Piping, Power Piping, USAS B31.1.0,1967 edition. Plant piping that was affected by power rerate conditions was evaluated in accordance with ANSI B31,1,1977 edition with Addendt up to and including Winter 1978 which meets or exceeds the requirements of the original code of construction. This code was used for evaluation of plant piping systems in response to IE Bulletin 79-14. The allowaole stress limits are in accordance with ANSI B31.1 for sustained, thermal, and occasionalloads.

26.

Refemng to Sections 3.6 and 4.1.2, provide the evaluation ofpiping systems attached to the torus shell, vent penetrations, pumps, an't valves that may be affccted by the LOCA dynamic loads (pool swell, condensation oscillatio,t, and chugging) considered in the evaluation for the power uprate.

NSP Resr.anse Piping was analyzed in accordance with the requirements of NUREG-0661 as stated in USAR Section 12.

NUREG-Co61 established the ASME Codu, Section ill allowable stresses as acceptable stress limrts for torus attached piping. The Staff a Safety Evaluation (SER) dated September 11,1985 concluded that the origmal torus attached piping analyses,' Plant Unique Analysis Report * (PUAR) met the acceptance enteria contained in NUREG-0661. See NSP's letter to the staff dated August 12,1997 regarding code reconciliations for certain Mark I analyses.

As stated in Section 4.1.2 of the power terate license amendment request, Mark I containment hydrodynamic loads (e g. pool swell, condensation oscillation, chugging) are not affected by power rerate.

Thus the existing torus attached piping analyses for LOCA dynamic loads bound power rerate conditions.

However, torus attached piping was evaluated for peak suppression pool temperatures under power rerate conditions. The results show all piping and supports are in compliance with the requirements of NUREG-0661 and the ASME Code stress limits.

Since LOCA dynamic loads are unchanged for power rerate, valve becelerations due to dynamic loads are bounded by the existing design values. Therefore operability of valves is not arfected. Valve and pump pressures and temperatures are within valve and pump design values.

27.

In relemnce to Section 3.8.2, provide a detailed discussion of the effects of the steam flowincrease, identified in Table 1-2, on the design-basis analysis of the main steam piping due to main steam isolation valve (MSIV) closure and turbine stop valve (TSV) closum loads. Also, provide an evaluation of MSIV structuralintegrity and functionality due to the increase in the hydraulic pressure for the higher flow rate following the power uprate, as discussed in Section 4.7 of GE's Generic Evaluations of General Electric MW M3 C St%OFFICiiWIWORDTplPLArEitiiI(MK 21 Y

Boiling Water Reactor Extended Power Uprate (NEDC 32523P) (proprietary information not pubitcly available).

NSP Response (Part 1) Effects of Steam Flow increase The design-basis analysis of the main steam piping considered dynamic loads due to closure of the main turbine stop valves. The loads created by operation of the main turbine stop valves are limiting and bound MSIV closure loads. For power ferat4 the main steam temperature and pressure during normal operating conditions will remain unchanged. Thus the increase in the main stop valve closure loads is directly proportional to the increase in the main steam flow. For power rerate operation at 1775 MWt, tl's increase in main steam flow is 7.1% as shown in Table 12 of the reiate amendment request. For the analysis of main steam piping and pipe supports, an increase in main steam flow of 14.4% corresponding to a reactor thermal power of 1880 MWt was used. The original design basis forcing functions for main turbine stop valve closure loads were increased by 14,4%, The new forcing fut.ctions were applied to the main steam lines including main steam, main steam equalizing and main steam bypass lines, inside and outside containment. The piping models included the attached SRV discharge piping, the HPCI steam lines, and the RCIC steam lines. This analysis methodology was consistent with the original main steam piping analyses. The results show all piping and pipe supports are within the code stress limit; as identified in USAR Section 12.

e (Part 2) MSIV tmpact As stated in Section 3.8.2 of Exhibit E of the rerate license amendment request, the maximum flow and differential pressure that the MSIVs are required to close against during a steam line break will not change under power rerate. This is because the maximum steam flow and differential pressure are dependent on maximum reactor dome pressure and main steam line venturi design which are not being changed for power rerate. Therefore, power rerMe will not result in an increase in main steam maximum hydraulic pressure so MSIV structural integrity and functionality are unaffected.

28.

Discuss the potential for flow-induced vibration in the heat exchangers following the power uprate.

' NSP Response No safety related heat exchangers will experience increased flows under rerate conditions. No increased flow will be required by emergency core cooling systems or containment cooling systems. Thus no safety related heat exchangers will experience potential flow-irduced vibration.

The reactor building closed cooling water system, reactor water cleanup system and fuel pool cooling system flows are not required to increase under rerate operating conditions, Only heat exchangers on the condensate and feedwater systems will be subject to increased flow under rerate operation, These systems are not safety related. The impact of increased flow in the condensate and feedwater system is addressed in the response to question 2g, 2g.

In reference to Section 7.4, provide the evaluation of the feedwater heater for the power uprate with regard to vibration, stress, and fatigue usage.

NSP Response The feedwater heaters have been evaluated by the original equipment manufacturers (OEM) for increased stresses resulting from operation at higher temperature and pressure. The evaluations were performed to the original code of construction for each heater, The code year and addenda vary due to replacement of several heaters over the life of the plant but all are designed and constructed to ASME Code Section Vill Division 1 and Heat Exchanger Institute (HEI) standards. The results of the evaluations show all feedwater heat exchangers are within the applicable code allowable limits.

WM7 JEB C S60FHCOWINwORDitMPL ATEilli rKK 22

A3ME Section Vill, Division 2, paragraph AD-160, states that ?t in the users responsibility to determine whether a fatigue usage evaluation is lequired. Since all of the feedwater heaters are designed to Division 1, the fatigue usage evaluation does not apply, ASME Section Vill Division 1 provides appropriate design safety factors to justify not performing a fatigue evaluation.

An evaluation of the feedwater heaters for vibration effects has also been performed by the OEMs and operation at terate conditions is considered to be acceptable.

Given the above. power terate will not require any physical modifications to the feedwater heaters in addition, it is important to note that there is no safety impact associated with this equipment.

30.

In Exhibit O, the stats ment is made that modincation to piping or equipment supports for some plant systems due to loed t hanges involves approximately 12 pipe supports. Provide examples of pipe supports requiring modification and discuss the nature of these modshcations.

NSP Re;ponse

~

Piping support modifications due to load c,hanges are described below.

One spring hanger on a feedwater heater drain line will be replaced with a rigid support.

=

The RHR heat exchanger supports require minor modifications to base plates.

Non-safety related drain lines from each main steam line to the condenser were evaluated to ensure that a qualified path for MSIV leakage to the condenser exists during a seismic event. All drain lines and equipment within the scope of this evaluation were seismically verified. The evaluations show several new supports and modifications to the existing piping and equipment supports are required in order to limit piping displacements and to increase load carrying capacity of supports during a seismic event.

These modifications will be completed prior to startup at rerate conditions. The formal commitment will be documented in NSP's revised license amendment request.

1 MT M B C WMWICOW WWORf> fEMPLATLi t t i DOC 23 l

-________.-.___m_

Human Factors 31.

Discuss whether the power uprate wdi change the type and scope of plant emergency and abnormal operating procedures Will the power uprate change the type, scope, and nature of operator actions needed for accident mitigation and willit require any new operator actions?

NSP Response For power rerate, reliance will continue to be placed on symptom-based Emergency Operating Procedures.

As such, the type, scope, and nature of operator actions needed for accident mitigation remains unchanged. PRA analysis has determined that no new operator actions are required. Some operator actions require reduced response times in select scenarios. See NSP response to Question 32 below.

32.

Provide examples of operator actions potentially sensitive to power uprate and address whether the power uprate will have any effect on operator reliability orperformance. Identify operator actions that would necessitate reduced response times associated with a power uprate, Please specify the expected response times before the power uprate and the reduced response times. What have simulator observations shown relative to operator response times for operator actions that are potentially sensitive to power uprate. Please state why reduced operator response times are noeded. Please state whether reduced time available to the operntor due to the power uprate will significantly affect the operator's abiltty to complete manual actions in the times required.

NSP Response Section 10.5.3 of NEDC-32546P discusses many of the above topics in detall as it describes the results of PRA analyses performed for the requested power uprate. These PRA analyses were performed at a bounding 12% higher power level (1870 compared to 1670 MWt), and thus incorporate additional conservatism compared to expected results for the requested power uprate to 1775 MWt.

The underlying physical phenomena driving essentially all observed changes in required operator actions are the approximately 12% higher decay heat and the corresponding ATWS power levels. Any postulated e-sequence of events leading to core damage driven by these higher power levels and dependent upon intervening operator actions are impacted by the higher power levels. The referenced PRA calculations utilized screening methodology to determine for what events these power levels and the human actions to mitigate the effects of the changed power levels were important, then determined what the new required time for operator actions was to ensure equivalent prevention of core damage, then input these new '

operator action times into a human reliabihty analysis to determine the increased chance of failure M the required operation action (s), and then finally input tnese changes in operator action success probabilities to calculate changes in the overall Core Damage Frequency (CDF).

In regard to operator actions, no changes to assumed operator actions for design basis event mitigation is required. Operator actions affected by rerate are those required for severe accidents and for those events outside the plant's design basis but within the licensing basis. Examples of the operator actions most sensitive to power rerate laclude: (1) manually depressu:izing the reactor vessel, and (2) injecting boron with the Standby Boron L6 quid Control (SBLC) system. Several other operator actions show significantly less sensitivity to power rerate and include: (1) restoration of power during a Station Black Out, (2) recovery of emergency diesel generators, and (3) repair of failed plant equipment prior to exceeding allowed primary containment pressure limits or vessel water level dropping below 2/3 core height.

Section 10.5.4 of NEDC-32546P (Exhibit E of the license amendment request) discussed examples of the changes in required operator respcase times before and after the bounding 12% power rerate. About 2/3 of the increase in Core Damage Frequency due to the power rerate arises from high pressure core damage sequances characterized by high pressure injection system failure after a successful reactor scram and subsequent failure to depressurize the reactor to allow low pressure makeup. It was noted that the required time to initiate manual depressurization of the reactor vessel was changed from 26 minutes to 23 mir.utes.

Initiation of this manual action prevents water level tiom dropping below the top of active fuel. The shorter time allowed to complete the required action is driven by the 12% higher decay heat. Similarly, about 1/3 of WiM JES C %150FFICDWINWORDTEMPLArBillt DOC 24

..__m

the increase in Core Damage Frequency anses from failure to inject boron with the SBLC system during various ATWS scenanos. The ATWS scenario for which the time required to initiate SBLC changes most is an ATWS with feedwater continuing to operate (thereby sustaining a higher reactor power level) and no isolation from the main condenser. Energy is deposited to primary containment due to Monticello's relatively small(1415%) turbine bypass valve capacity. The time required to initiate SBLC changes from 21 minutes to about 13 minutes. Initiation of SBLC prevents the suppression pool temperature from exceeding 260 'F. The shorter time allowed to initiate SBLC is driven by ths.t higher ATWS power resultmg from the bounding 12% power level Although required times to accomplish manual operator actions are decreased as illustrated abo.*e, there is still adequate tirne to accomplish these actions, and an expectation that tne actions would indeed be accomplished. Operators are rigorously trained and evaluated on the symptom-based emergency operating procedures utilizing the Monticello simulator, input of the somewhat shorter response times and the resultant somewhat higher human error probabilities rasults in a minor effect on the overall Core Damage Frequency. Section 10.5.7 of NSP's previous rerate license amendment confirmed that the change in Core Damage Frequency was minor by comparison with appropriate quanthativa screening criteria.

33.

Discuss any changes the power uprate will have on control room instruments, alarms, and displays. Are zone markings on meters changed (e g., normal range, marginal range, and out-of-tolerance range)?

NSP Response in regard to human factors, the main control room panel instrumentation, as presently configured, can support power rerate operation without modification. No changes to zone markings have been identified. Certain changes to instrumentation, such as setpoint changes, will be necessary. See Exhibit D of the license amendment request amendment and NSP's responses to questions 34 and 37, 34.

Discuss any changes the powor uprate will have on the Safety Parameter Display System (SPOS).

NSP Response The impact of power rerate on SPDS, as well as other portions of the process computer and core monitoring software, is limited. Calculations and outputs that utilize power expressed absolutely in MWt are expected to continue to function with only minor adjustments in data validity checks to accommodate the higher expected MWt value when operating at 100% of terated power. Some calculations that calculate power as a percentage of rated power will require that new databank values be loaded for the rated power value of 1775 MWL that appears in the denominator of such calculations.

Validation calculations currently exist that check reactor ther.nal power by performing alternate calculations based on plant parameters that are independent of the primary feedwater flow nozzles. The input range of the validation calculations be extended to accommodate operation at the higher rerated power level.

Operation at a higher turbine-generator electrical output will require adjustments in data validity checks in this area. Thermal margin limits for core operation that are a function of reactor power and are monitorert by core monitonng software (e g. 3D-Monicore) will be reviewed as part of the normal core reload proct,..

to ensure that they are applicable to operation at 1775 MWt.

Prior to implementation of power rerate, all affected SPDS data points will be validated. The formal commitment will be oocumented in NSP's revised license amendment request.

35.

Describe any changes the power uprate wril have on the operator training program and the plant simulator..

Provide a copy of the post-modification test report (or test abstracts ) to document and support the effectiveness of simulator changes as required by ANSI /ANS 3.5-1985. Section 5 4. f. Specifically, please propose a license condition and/or commitments that address the following:

(a) Provide classrocm and simulator training on the power uprate modification.

wm na e momcewmomm.rummi poc 25 l

l

_ = _ _

(b) Complete simulator changes that are consistent with ANSI /ANS 3 5-1985. Simulator tidelity will be re.

validated in accordance with ANSI /ANS 3.5-1985, Section 5.41,

• Simulator Performance Testing?

Simulator revalidation willinclude comparison ofindividual simulated systems and components and simulated integrated plant steady state and transient performance with reference plant responses using similar startup test procedures.

(c) Complete control room and plant process computer system changes as a result of the power uprate.

(d) Modify training and plant simulator relative to issues and discrepancies identified dunng the stars'up testing program.

NSP Response a) Classroom and simulator training on new knowledge and abilities associated with the power rerate will be provided to all operations and licensed personnel in accordance with Monticello Training Center procedures. This training will be completed prior to implementation of power rerate, b) Simulator changes will be completed in accordance with ANSI /ANS 3 5 1985 section 5 4,1 simulator performance testing and Monticello simulator configuration control procedures. Initial simulator changes will be completed prior to porteer uprate and verified against actual plant startup data.

c) See Question 34 above.

d) Training and simulator will be modified in accordance with applicable Monticello Training Center procedures to reflect issues and discrepancies identified during startup testing.

The formal commitments to these conditions will be documented in NSP's revised license amendment request.

wkW A.B C Whof FICDWINWORDtTIMPLATBlitt DOC 26 l

l l

Instrumentation and Controls 36.

For power uprates, the GE setpoint methodology discussed in GE topical report NEDE-31366 has been used to determine instrument setpoints. Therefore, this methodology should be referenced in the basis section of the technicalspecifications.

NSP Response A statement si.nilar to the following will be added to the MNGP Technical Specification Bases related to section 3.1,

" Reactor Protection System,* and 3.2,

• Protective instrumentation.*

GE setpoint rnethodology provided in NEDC 31338,

• General Electric Setpoint Methodology

• ls used, as applicable, in establishing setpoints.

The formal commitment will be contained in the revised license amendment request.

37.

Tne submittal does not address the effect of power uprate on instrumentatior, range / span. Also, Section 5.2.1, Control Systems Evaluabon, states that,

• process control valves and instrumentation have been evaluated for range and adjustment capabihty for use at the expected terated cond: tion. Any required changes will be performed prior to operation at the terate..

• However, the submittal does not identify any such instrumentation an. ' mtrol valves. Provide this information for staff review.

NSP Response The capacity of the turbine control valve has been analyzed and will be venfied during rerate startup testing. Refer to rerate report GE NE-L120082901, *Rerate Test Program Recommendations for the Monticello Nuclear Generating Plant.*

(Proprietary, available for inspection onsite.) The instrumentation and control valves requiring adjustment and modification are identified in Exhibit D of the license amendment request.

Review of the individual systems affected by rerate included the affect ore instrumentation and controls. No instrument changeouts have been identified.

i 38.

Table 5-1 provides changes in the analytict

• limit for setpoints for the current and power uprate condition.

Thejustification for these changes is based on the assumption that they do not increase the probabihty and consequences of postulated accidents, or reduce significantly the margin of safety. In order for the staff to arrive at the same conclusion, information is needed on instrument setpoints and allowable values in addition to the analytical hmit for the instrumentation identified in Table 51 at both the currvnt and uprate power conditions.

The requested information is provided in the table below.

M97 IER C StSOFHCDWINWORDtTBtPLAfblillIME' 27

EXISTING NEW (1670 MWt)

(1775 MWt)

APRM Rod Block Setpoint 107.5 %

107.5 %

Allowable value 110 %

110%

Analytical Limit 112 %

112 %

APRM Scram Setpoint 119 5%

119 5%

Allowable value 122 %

122 %

Analytical Limit 125%

125%

Vessel High Pressure Setpoint 1051 psig 1051 psig Allowable value 1085 psig 1085 psig Analytical Limit 1090 psig i 1091.5 psig ATWS High Pressur3 ATWS Setpoint 1135 psig 1135 psig Allowable value 1155 psig 1155 psig Analytical Limit 1162 psig 1162 psig SRV Setpoint (maximum) 1120 psig 1120 psig Allowable value 1142.3 psig 1142.3 psig Analyti:allimit 2 1142.3 psig 1142.3 psig Turbine 1st Stage 3 Setpoint 27%

27%

Allowable value 30 %

30 %

Analyticallimit 45%

45%

Main Steam Line High Flow Setpoint 127.5 %

137.5 %

Allowable value 142 % -

142%

Analytical Limit 146 %

146 %

Condenser Low Vacuum 4 Setpoint 23.25' hg 22.25" hg Allowable value 22.5' bg 2165* bg Analyt: cal 22.5'hg 21.5" hg Notes 1.

Changed due to setpoint methodology improvement independent of terate.

2.

- A maximum reactor pressure of 1279 psia will result with the SRVs set at their analytical limit. The maximum allowable reactor pressure is 1350 psia.

3.

The final relation between turbine 1st stage pressure in psig and % power will be determin-O during startup testing. The setpoint in psig will be set at a value less than 30% power by an amount determined by the setpoint methodology (approx.

3%).

4.

Change supported by NSP's license amendment request.

• W JtB C WWFICIDWINWORDihtPLATDilll CKM' 28 1

Rad.jtyon PrNechon 39.

Sochon 8 $ 2 of Eshilst E slales t/mt W5P has established successful cobalt reduchon, zinc injection, and hydrogen water chemistry programs. These prograrns and other dose reduchon programs wtII adequately 1

compensate for the possiobe tacion:ms it; individual dosos due to power 1910102 Provi!e addrhonal informabon (;onceming theso dosa reduction programs, state when these programs w0re implemented at MontoccIlo, and descnbe what effect they have had on reducing overall dosos at MonbccIlo Compare the estomated annualreduchon in overall dosos resulhng from the amplementaban of these dose reduchon progrb,Js with the eshmated annual tracrease in doses at Monbcollo due to the proposed power relate NSP Response Part 1. Information Concerning Dose Reduction Programs A-Cobalt Reduction Program Description The cobalt reduction program is a formal effort, cons: stent with the ALARA philosophy, to eliminate sources of stable Co-59 in p' ant systems which communicate with ine pnmary system The goalis to minimize tha production of radioactive Co-60. To facilitate this effort, Co-59 sources have been identified and tanked according to the estimated Co 59 release rate.

Implementation Cobalt reduction started in aMut 1982, when the first non-stellite control blades were installed for testing Sirce then, the t'tajonty of originally installed high cobalt blades have been removed from the reactnr. Other ar tivities include replacement of numerous valves, drain lines, and high erosion surfaces on turbine blading and inner casings.

D.

Zinc injection Description Studies performed by General Electric bhow that small concentrations of zinc in the reactor water will result in a reduction in the amount of Co-60 incorporated into the oxide fiim established on stainier.s steel piping. This reduction in Co 60 incorporation provides substantial reductions in dose rates, particularly in primary containr.ient. When first introduced, zinc injection utilized natural zinc. Later, it was determined that zinc dcpleted of the isotope Zn-64 was a better choice because it eliminated the problems associated with Zn-65, which is produced by neutron activation of Zn-64.

Implementation Zinc injection was implemented in 1989. In 1993, injection of natural zine was terminated in favor of depleted zinc.

C.

Hydrogon Water Chemistry Description The presence of oxygen generated by radiolytic decomposition of water produces ar' environment favoring IGSCC of the components exposed to coolant. This mode of degradation can be controlled by suppressing the dissolved oxygen concentration with hydrogen injection and by maintaining high purity reactor coolant water. This process is called Hydrogen Water Chemistry (HWC).

Implementation 29 l

l l

Hydrogen Water Chemistry was implemented in 1989.

D.

Effects of Doss Reduction Programs it is difficult to separate the effects of cobalt reduction and zine injection because both programs have the same result,i e., reduction of Co-60 concentrations on out of core surfaces. The combined effect is evident from the average concentration of Co-60 in reactor coolant, which has been decreasing since 1991. This is significant because the opposite was true over the five 8uel cycles prior to that time. The decreased coolant concentrations have resulted in less deposition in recirculation piping, lower exposure rates in the drywell, and lower personnel doses Based on average drywell exposure rates since 1993 and recorded personnel doser, it is estimated that dose saved dunng outages over the last two fuel cycles is in excess of 150 person-rem.

The benefits of the Hydrogen Water Chemistry (HWC) program are less evident. An immediate effect of HWC is increased dose rates in certain areas of the plant. Injecting hydrogen into reactor water increases the fraction of volatile N 10, which is carried over in the steam. As a result, dose rates due to primary steam increase approximately 3 to 5 times normal. The long term benefit which must be considered, however,is avoidance of the personnel dose that would be required to repair or replace reactor internals.

In an effort to offset the effect of increased dose rates due to N 16, routine plant practices and policies were examined and changed where appropriate to keep personnel exposure ALARA.

Changes hcluded reduced frequencies for some inspections and a practice of reducing the rate of hydrogen injection periodically to accomplish work in steam-affected areas. The net ef ect has been an overall reduction in personnel doses during operating periods. Companson of average annual personnel dose outside of refueling outages, prior to and following HWC implementation, shows a decrease from 121 rem per year (1986 through 1988) to 81 rem per year (1989 through 1996). Not all of this decrease can be attributed to the actions taken to compensate for N 16 increases, but it is clear that the potential personnel dose increases were adequately offset.

Part H.

Estimated Annual Dose Reductions vs. Increase On an annual basis, the reduction in overall dose due to dose reduction programs is estimated to be 75 person-rem. This number is based on the following assumptions.

. Two year operating cydes Drywell dose is comparable to recent refueling outages (150 person-rem per outage) 1993 average drywell exposure rate (272 mR/hr)

- Current average drywell exposure rate (138 mR/hr which is about 50% of base rate)

Dose w/o dose reduction programs: 300 fem (150 sem per outage + 50% reduction)

Dose with dose reducten programs: 150 rem (assumed average ou.ag4 Annual overall dose rWuction:

75 rem (150 rem / outage + 2 years / outage)

The estimated increase in drywell dose rates at Monticello due to the proposed power rerate to 1880 MWt is 8%. (See NEDC-3264'F, Monticello Cobalt Transport and Shutdown Drywell Dose Rate Model Calculation Results, attached.) Since Co-00 and other activated corrosion products are the source of more than 90% of drywell dose and because Co-60 and other activation products account for the largest portion of doses outside of the drywell, it is safe to assume that annual doses due to the proposed power increase willincrease by approximately 10% overall. Since 1988, average annual dose for non outage years has been 90 rem. The average dose for outage years (conservatively based on last two outage years) has been 315 rem. Therefore, the annual dose increase due to rerate is about 20 person-rem per year.

Average non-outage year:

90 rem (1988 through 1996)

Average outage year:

315 rem (1994 and 1996)

Two-year total:

405 rem Annual average total:

202 rem 10% increase:

20 rem SW HJI C \\M%Of tlCf4WWORDTlMPLATUitiI tylc 30

Given the above, the expected dose increase from power rerate of 20 tem /yr is more than offset by dose reduction programs.

40.

Section 8.4 2 of Exhibit E states that the power terate may result in a net increase in the activated corrosion product production due to the increase in activation rate in the reactor region combined with the decrease in folter effsciency of the condensate domineralizers (due to the feodwater flowincrease). Descnbo the magnitude of the estimated increase in activated corrosion products in the reactor piping and descnbo how this will affect dose rates in the vicinity of this piping Descnbe any plans (such as increasing the amount of zinc it;lechon to the reactor coolant system) that you may have to reduce the increased amounts of activated corrusion products in the piping caused by the proposed power verate.

NSP Response Monticello analyzed the potential increase in activation products that may occur as a result of the reactor power rerate. This analysis is documented in GE Report NEDC 32647, Monticello Cobalt Transport and Shutdown Drywell Dose Rate Model Calculation Resu!ts. Final Report The cobalt transpori model developed by CC Lin (GE) has become a widely recognized and accepted algonthm for such modehng.

Power uprate affects the reactor system in the following ways, 1.

Increased feedwater flow rate resulting in increased mass transport of feedwater impurities.

2.

Increased core average neutron flux resulting in more activation events.

3.

Increased heat flux on fuel surfaces resulting in higher corrosion product deposition.

4.

Increased feedwater impurity levels due to increased feedwater flow conditions and reduced efficiency n condensato treatment systems.

At Monticello's direction, GE modeled rerate conditions for both 6 3 and 12.6% power terate conoitions.

The overwhelming contributor to dose at Monticello is the isotope cobalt 60. This being the case, the cobalt transport model is very appropriate to use in determining the effect power rerate may have on dose rate buildup of recirculation piping. The GE model was run for a number of cases. A base case was run to provide a benchmark from which tojudge the results of the other cases. The base case input parameters were chemistry parameters and conditions during calendar year 1996 The following parameters were used as model inputs: zine injection is considered, reactor power is 1670 MWt, reactor conductivity was assumed to be 0.114 uS/cm, feedwater iron was assumed to be 1.4 ppb, and feedwater cobalt concentrations were assumed to be 3 ppt. The results of the power rerate cases predict potentialincreases in Co60 relative to the baba case. Modeling indicates a power rerate of 6.3 % (1775 MWt) may result in an increase in Co60 activity and dose rate of 2-4% through EOC-22. For a power level of 12.6% (1880 MWt), an increase of 6-13% both in Co60 activity and dose rate is predicted. To pet these values in perspective, the 6 3% and 12.6% power increases would result in recirc piping dose rate increases of 3 mr/hr and 13 mr/hr respectively. From a radiation protection perspective these low level increases are quite manageable.

There are several factors thct willlikely reduce or negate the predicted dose rate buildup increases resulting from power rerate. Feedwater iron reduction efforts have been very successful to date. In the Cobalt Transport model study the base case assumed a feedwater iron concentration of 1.4 ppb. Modehng shows an overall reduction in recirc piping dose rates of 14% through EOC 22 for feedwater iron concentrations around 0.5 ppb. The average feedwater iron concentration for cycle 18 is about 0.8 ppb.

Also, a significant reduction in cobalt source term should be evident with the changeout of the high and low pressure turbines during the 1996 refuehng outage. The original turbine had stelhte faced blades at the latter stages of the LP turbine. The new turbine does not use stellite but rather employs a flame hardening technique to provide the needed hardness to protect against moisture impingement. It is estimated that reductions h feedwater cobalt may achieve a dose rate reduction of 8% through EOC-22.

The injection of derdeted zine oxide (DZO) at Monticello has been demonstrated to effectively reduce recire piping dose rate buildup. The DZO injection rate is varied depending primarity on the soluble Co60 concentration trend in the reactor coolant. Other parameters are considered as well, however the direct link between reactor coolant Co60 concentrations and recire pipe dose rates are widely accepted and well M91 n.B C iMWif KTWIN%OkkintPLATINiil tx4' 31

established. Should C000 concentrations increase as a result of power rerate implementation DZO injection ates will be aftered accordingly while considering all available parameters.

41.

Section II H.3 b of Exhibit A states thal

• Reference to containment spray is to be deleted from lhts discussion

• Provide your reasons for not taking credit for containment spray and state whether deletnng reference to the containment spray in this section constitutes a change in your accident dose analysis.

NSP Response There are no design bas'ts event sequences that rely on containment sprays for accident mitigation. In some severe accident scenarios, the use of containment spray is credited. Containment spray operation was postulated in the development of certain EQ probles. The use of containment sprays is a general enhancement of safety such that NSP has decided to retain this credit. Section ll H.3 b will be amended to reflect this change.

42.

Exhibit D (p D f) states that one of the hardware changes forpower terate will be to modofy the Control Room Emergency Filtration Train system *to reduce control room ventilation folter bypass leakage to establish consistency with control roono dose calculation inputs.* Discuss what you mean by estaolishing

' consistency with controlinom dose calc.sim; unalinputs.* In Table 9.4 of Exhibit E you state that the eshmated thyroid dose in the control room fo10 wing a LOCA at the terate power of f 360 MM would be f 3 rom. State what the estimated LOCA thyroid dose in the controlroom would be (at 1880 MM)if the control room ventilation fi!!er bypass leakage were not reduced. Provide both the current and the reducert control room bypass leakage figures in cubic feet per minute.

NSP Response Under the current dose calculations (NSP Calculation 94 009 Revision 1. dated 7/22/94), an inleakage of 500 cfr.:is assumed for the first 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of the accident. After 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, ingress and egress into the control room woulu be restored back to normallevels and positive pressure will be assured. The inleak6ge used in the calculation then orops to an assumed 250 cfm for the duration of the accident. This inleakage is due to 240 cfm ci leakage across the inlet isolation dampers on the operating CRV train and an additional 10 cfm from normalingress and egress from the control room. This calculation used 90% for the Standby Gas Treatment and the Control Room fitter e'fderes and 0;% plateout of iodine in the steam lines and in the main condenser.

Preliminary work performed for the forate effort showed that with an inleakage of 500 cfm for Tc8 hours and inleakage of 250 cfm for T>B hours, control room operator dose to the thyroid would be 38 rom. This work was done as a preliminary sensitivity study and was never finalized. It used a SBGT efficiency of 81% and an control room filtration efficiency of 95% 11 also used the new BWROG methodology for modeling lodine plateout in the steam line and the condenser (GE Report NEDC-31858P).

To limit the dose to less than 30 rem, inteakage from the inlet isolation dampers had to be reduced or eliminated and credited filter efficiencies increased. GE performed the dose calculation for the rerate enndition Under this analysis, control room inleakage was 250 cfm for Tc8 hours and 10 cfm for T*8 hcurs. Standby Gas Treatment and Control Room filter efficiencies were changed to 85% and 98%

respectively. This model used the BWROG method for modeling lodine plateout in the stream line and the condenser provided in NEDC-31858P.

To ensure no inleakage for the control room ventilation inlet dampers, blanking plates were installed on the system in August of 1996. Dedication of the main steam system and condenser to meet the requirements of NEDC-31858P is being done under other parts of the same modification.

43.

Exhibit A (p. A 2 f) states that, based on a radiological analysis for the proposed verate, you wtIIimprove the etnctencies of the control room emergency filtration system filter and the standby gas treatment syste.n falter. State the current and proposed tiller officiencies and discuss your timetable for making these changes.

MM ft B C WlMW KT UNWOitD itAtPL ATDiili (UC 32 l

NSP Response The current and proposed filter efficiencies are summanted below.

Control Room Filters Standby Gas Treatment Filters Current Proposed Current Proposed 90 %

98%

90%

B5%

As stated in paragraph 4, page A 21 of Exhibit A to the license amendment, the credited overall efficiency of the fdters in the Standby Gas Treatment system actually will be decreased. Reducing the efficiency credited in the SBGT system provides additional margin for fdter bypass.

A license amendment request entitled.

• Reactor Coolant Equivalent Radioiodine Concentration and Control Room Habitability,* was submitted on July 20,1990. Revision one to this 1.AR was submitted on April 11, 1997 and supersedes the original submittalin its entirety. This submittal supports the acceptance criteria for the control room filtration system efficiencies indicated above The procedure which governs control room filter testing has been changed to incorporate the enore testrictive acceptance criteria. Review of past testing results indicates that these criteria can be satisfied.

44.

The table on page A.20 of Exhibit A lists the calculated potential offsite doses at the esclusion area boundary (EAB) and low population tone (LPZ) from the following design-basis accidents: loss of coolant, refuehng, controlrod drop, and steamhne break. These doses were calculate by the AEC statt and are contained in the staff safety evaluation dated March 18,1970.

a For the same four accidents descnbed &bove, provide a hsting of the postulated doses (both whole body and thyroid) at the EAB, LPZ, and control room that were calculated by the hconsee during the initial licensing of the plant.

On page A.21 of Exhibit A, you state that the inputs and evaluation methods for the MNGP power terate dofler from those used in the current licensing basis evaluation contained in the USAR and in the AEC safety evaluation You state that you have estabhshed dose mulhphers that should be used to multiply the doses contained in your original hcensing basis evaluation to obtain the does calculated for the MNGP power verate,

b. Show how you applied these dose multipliers (listed in Table 14.7 22 of the USAR) to the doses calculated using your current licensing basis evaluation to arrive at the revised accident doses for the proposed power relate (listed in Table 9-4 (Appendix E) of the relate licensing amendment request).

NSP Response Part a.

Postulated accident doses calculeted during the initiallicensing of the Monticello plant are presented in Section 14.0 of the original Monticello Final Safety Analysis Report (FSAR). This is now Section 14,7 of the Monticello Updated Safety Analysis Report (USAR). These analyses were performed using the methods

- desenbed in General Electric topical report APED 5756,

• Analytical Methods for Evaluating the Radiological Aspects of the General Electric Boding Water Reactor l March,1969.

The FSAR dose calculations generally used more realistic assumptions than those used by the AEC Staff.

For example TID.14844 cource terms were not used in the loss of coolant accident radiological analysis, in addition, the methodology for atmospheric dispersion calculations differed significantly from methods used by the AEC and those currently in use.

Control room doses were not calculated in the FSAR. The originallicensing and design basis for the Monticello plant predated 10 CFR Part 50, Appendix A, GDC 19. A safety grade control room fittered air supply system was later added to the plant in response to the NRO Three Mile Island Action Plan, and dose calculations were completed in 1981 to demonstrate conformance to GDC 19 using the existing NRC guidance.

The original FSAR accident dose calculations are presented in the table below.

em na c monwmonunwwram tw 33

5 y

ACCIDENT ~

LOCAllON WHOLE BODY THYROIDDOSE DOSE (REM)

(REM)

LOSS OF COOLANT EAD (1/3 MIL E) 5 BE 4 5 2E 5 LPZ (1 MILE,)

3 GE-4 4 2E 5 CONTROL ROOM REFUELING EAD (1/3 MILE) 7.0E 3 3 OE 3 LPZ 4 4E 3 3 OE 3 CONTROL ROOM CONTROL ROD DROP EAB (1/3 MILE) 5 2E 3 3 4E 4 LPZ (1 MILE) 4 6E 3 2 BE 4 CONTROL ROOM CTEAM LINE BREAK EAB (1/3 MILE) 4 SE J 20 LP7 (1 MILE) 3 OE-3 10 CONTROL ROOM in Section 14.10 of the origina! Monticello FSAR (now Section 14.7.7 of the Monticello USAR),' Design Basis Accident Radiological Dose Multipliers' were presented which, when multiplied by the accident doses calculated in Section 14 6, would give results which more closely resemble doses calculated using AEC Division of Reactor Licensing (DRL) methodology. These ' Dose Multipliers

• are presented in Table 1410-8 of the original FSAR (Table 14.7 22 of the USAR). They are reproduced below ACCIDENT WHOLE BODY T HYROID 2-HOUR 30-DAY 2-HOUR 30-DAY (500 M)

(3218 M)

(500 M)

(3218 M)

LOSS OF COOLANT 1.08 E + 02 1.96E + 03 8 86E+05 000E+05 REFUELING 6.10E+01 4 51E+00 3 22E+02 2 50E+02 CONTROL ROD DROP 2.89E+00 3 03E+00 8 94E+04 4 56E+04 STEAM LINE BREAK 1.00E + 01 571E+01 The original dose calculations and the ' Dose Multipliers' presented in the FSAR and USAR are not consistent with current updated NRC guidance and were not used for power rcrate analyses Part b.

The " Dose Multipliers

• desenbed in response (a) above were used only in the onginal accident radiological analyses performed for the Monticello plant. They were not used to arrive at the revised accident doses for the proposed power rerate.

Plant specific accident dose analyses for Monticello power rerate were calculated by Genera; Electnc in accordance with current NRC guidance using a reactor thermal power of 1880 MWt These calculations are desenbed in Sec* ion 9 2 of Appendix E of the Monticello terate license amendment request.

Dose analyses were also performed using the same current methods at the existing licensed power of 1670 MWt. Table 9-4 of Appendix E of the Monticello terate license amendment request compares calculated accident doses at the existing licensed power level of 1670 MWi to those calculated at the analyzed terate power level of 1880 MWt S W 11 D C Al%Of i K i '% iNw OltD 11 kil't All siili (tC 34

A comparison of accident dose calculations at terate power level using currently approved methods to doses calculated at 1670 MWt in the FSAR and USAR would provide inconsistent results and would not be meaningfulin the context of the current methodology, Following NRC approval of terate power operation at Monticello, the USAR will be updated to reflect the new radiological analyses performed by General Electric using currently approved methods. These analyses employ methods previously approved by the NRC and are available for inspection onsite, A formal and more encompassing commitment to update the USAR will be made in NSP's revised terate license amendment riquest.

45 Descnbo those plant changes (both operational and hardware changes) made to accommodate the proposed power terate that will have an effect on the calculated EAB, LPZ, and control room doses following any one of the following design-basis accidents: loss of cooling, reluchng, control rod drop, and steamline break MP Respjng Two basic hardware changes were performed to the plant in order to support radiological dose evaluations for the power rerate. The first change was the installation of blanking plates on the Control Room Ventilation System inlet ductwork. This will ensure that inleakage into the protective envelope from the isolation dampers is zero. This is consistent with the control room operator dose calculations used in the power rerate program.

The offsite and control te am operator dose assessmerds used to support ferate use the BWROG methodotogy (GE Report NEDC 31858P) with regards to plateout in the main stream lines and the condenser, This required that the steam lines, steam line auxiliary piping, stet m lano drains and main condenser bo verified to withstand a seismic event in accordance with the methodology. Doing this venfication, a number cf improvements and modifications to steam line auxiliary piping and drains were identified. These modifications included changing pipe from carbon steel to stainless steel to reduce corrosion / erosion concems, changing and adding piping support; and eliminating a steam line drain loo 9 seal to increase the effective size of the main condenser credited in the calculations, Some operational changes have resulted from the Power Rerate program. As mentioned in question 43, the credited filter efficiencies of the Control Room Filtration System and the Standby Gas Treatment System have been changed. This required changing the acceptance criteria of these filters during normal surveillance tests. Because of the plates installed on the control room ventilation system, the filter trains run more often resulting in the testing of the charcoal beds as required by Technical Specifications after every 720 hours0.00833 days <br />0.2 hours <br />0.00119 weeks <br />2.7396e-4 months <br /> of operation. This increases the frequency of charcoal bed testing from once every 18 months in the past to about once every 0 9 months in the future.

The Technical Specification allowed dose equivalent iodine has been reduced from 5 microcuries/ml to 0.25 microcuries/ml. See NSP's licen',e amendment request dated July 26,1996 and revised on April 11,1997,

' Reactor Coolant Equivalent Radioiodine Concentration and Control Room Habitability,'

e m n a c wsommwiwommimmini tw 35

Probabilistic Risk Assessmenj 40.

On page 10-8. the last paragraph states *retato analysis did require two SRVs (safety-rehof valves) to open to avoid reactor overpressure whereas only one SRVis adequate for the 100% powerlevel case.' How was this change reflected, if any, in the nsk analysis and how (much) did it contnbute towart the eshmated l

increase in plant core damage frequency (CDF)?

NSP Resp 2ng 1he likelihood of all eight SRVs faihng is very similar to at least seven out of eight SRVs failing. The effect on the estimated CDF would be negligible so the PRA model was not modified to retlect the change in the SRV over pressure protection success enteria. This change was only addressed qualitatively.

47.

On page 10 9, 2nd paragraph under the section, *iime Available for Operator Action

• states that *the most important post-inttiator human enors were recalculated using the method descnbed in NUREGICR-4772

(*A:cident Sequence Evaluation Program Human Rehabably Analysis Procedure *) for deriving nominal human enorprobabihty estimates (ASEP(Accident Sequence Evaluation Program) method), Please descnbe the increase in human error roles of the inost impacted operator actions due to the power relate by providing their

• current

• human onor rates as well as the 'new' human enor rates that were estimated using the above method.

For example, on page 10-15. the first paragraph states 'the time for the operator to initiate SBLC is reduced from approximately 21 minutes to 13 minutes In spite of the reduction in hmo to perform this actron, the hkchhood of the operator conectly performing this achon is stsli high.* Picase provide the change in human error rates associated with the change in requirement to initiate SDLC from 2t minutes to 13 minutes and show how this change impacts the analysis results.

As another example, on page 1014, the fourth paragraph states 'a large portion of the CDF due to high pressure core damage sequences result from intemal flood initiator events.* On page 10-15, the fourth paragraph states *this is due to the decrease in the time available for the operator to blowdown the vessel before the ente becomes uncovered,' Howis two thirds (approximately 1.6E-6/Yr) of the CDF increase attnbuted to operators' ability to respond to these sequences?

Table 471 provides the

• current

• and *new' human error rates for the operator actions that could have the greatest impact on the change in CDF due to terate. They are given in the columns called

• base case

• and

• rerate'. The human error rate associated with a change in the requirement to initiate SBLC from 21 minutes to 13 minutes is in the row with the Basic Event name SLCTOPY, less than one third of the 17.5%

change in CDF is due to this change.

Two thirds of the CDF increase occurs in high pressure core damage sequences and is due to the decrease in time available for the operator to depressurire the vessel. Although there is a relatively small decrease in time associated with this operator action, it is a sensitive parameter in the PRA model (operator action with the highest Bimbaum). Table 471 provides the

• current

• and *new* human error rates for failure to depressurize in the row for Basic Event XRPVBLDWNY. See the response to question 4g for more discussion on emergency depressurization.

mt an c wasomcremanemim tec 36

4 Table 471 Important internal Events HEPs Basic Events Description Probability Base Case Rerate XRPVBLDWNY Operator fails to depressurize reactor for 1,1 E 3 1.4E 3 low pressure makeup XRPV10 Operator fails to depressurize reactor in 2.0E 2 2.6E 2 10 minutes (Medium LOCA)

FRFPOPERAY Failure to control feedwater after a 2.4E 3 2 4E 3 scram UH Operator dilute boron by f ailing to 1.0E 2 1.0E 2 control reactor water level MCONVENTY Operbtors f ail to vent containment 3 0E 5 3 0E 5 SLCCOPY Operator falls to initiate SLC dunng 2.4E 3 4.9E 3 ATWS (Loss of Main Condenser)

SLCMOPY Operator fails to initiate SLC dunng 2.4E 3 4.0E 3 ATWS (MSIVclosure feedwater operates initially until hohvell Iow level)

SLCFOPY Operator fails to initiate SLC dunng 2.0E 3 2.0E-3 ATWS (MSIV closure no feedwater)

SLCLOPY Operator fads to initiate SLC dunng 2.0E 3 2.0E 3 ATWS (LOSP event)

SLCTOPY Operator fails to initiate $LC dunna 4.4E 3 4.04E-2 ATWS (Turbine Trip with bypass)

NOOSP30 Failure to recover OSP in 30 minutes 0.64 0 66 NOOSP2 Conditional non recovery of OSP in 2 0.45 0 44 hours5.092593e-4 days <br />0.0122 hours <br />7.275132e-5 weeks <br />1.6742e-5 months <br /> given no OSP recovery in 30 minutes NOOSP6 Conditional non recovery of OSP in 6 0.16 0,153 hours0.00177 days <br />0.0425 hours <br />2.529762e-4 weeks <br />5.82165e-5 months <br /> given no OSP recovery in 30 minutes NODG2 Failure to recover EDG in 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> 0.66 0 68 NODG8 Failure to recover EDG in 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> 0.35 0.36 ABKF OXXXXY Failure to back feed 0013 to essential 05 0.5 loads fo!!owing loss of all AC sources WREC48 HRS Failure to recover RB equipment (long 0.24 0.28 term) 48.

On page 1011, Section 10.5.3 2, 'Intemal Events PRA Level 2 (Containment Analysis),'the nrst paragmph stales that the requanbned results of the Level 2 portion of the PRA update was not available at the tome of this analysis. Please provide the quantitative results of the Level 2 analysis. If available, please also provide the quanhtative results of the risk analysis for extemal events.

NSP Response Part 1 Quantitative Results of Level 2 Analysis The results of the updated level 2 analysis for the Monticello PRA are summarized in Figure 48-1. Two means of binning the accident sequences of the Level 2 PRA are used in this figure. Plant damage states, defined in Table 48-1 1, establish the reactor status, containment failure mode, and containment failure timing (from an emergency planning perspective) of any sequence. Release modes, defined in Table 481 2, were used to categonze the accident sequences from the standpoint of magnitude of release. Consistent with the definition used in the implementation of the Maintenance Rule, the following was used to categonze large, early release sequences.

noi n a r uw nn wwem rutet.arra ni noc 37

The term Release implies failure of containment to retain fission products folicwing a core damage event in the form noble gases, volatiles (Cs/l) and non-volatiles (Te, Sr, etc ). The term Earfy refers to the timing of releases from containment relative to implementation of protective ac' ion guidelines associated with the emergency plan For the purpose of this discussion, the definition of early releases is simplified to be those releases which occur before the offsite protective action recommendations under the emergency plan can be effectively implemented se., approximately 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> after declaration of a general emergency). The term large includes two aspects: the volumetric release rate and the amount of fission products released. Large releases are considered to occur only if the release path is sufficiently large that release rates would be significantly greater than those permitted by Technical Specifications, and if the release is not filtered through a pool of water or sprays to retain a significant fraction of the fission products inside containment.

Given these definitions, the accident sequences which are considered to lead to large early releases are those that are binned into release mode categories C4 through C12 which also have a timing plant damage state designation 'E' (for early). In addition, release modo categories E1 and E2, which represent bypass and containment isolation failure sequences, are considered to lead to large early releases.

From Figure 481, it can b3 Seen that the majonty of sequences do not fall in the large early release category either because they do not result in containment failure, they are vented or released through a pool, or they occur late, many hours into the accident. From the baseline Level 2 PRA results, the potential for a large early release therefore is small, on the order of 3% of the total core damage frequency. As with other Mark I containments, large early releases for Monticello are dominated by ATWS and interfacing LOCA sequences. Hydrogen combustion during accident sequences in which the containment is deinerted also contributes to early releases because a conservative assumption was made that periodic burns will not limit hydrogen concentration. The liner melt through containment failure niode is not likely to occur at Monticello because the drywell sumps are large compared to the size of the core, so these sumps can retain most of the debris that would be released from the vessel at the time that the bottom head is breached by molten debris. The debris la therefore not expected to come in contact with the containment liner.

Following rerate, the sequences leading to large early releases rise roughly proportionally with the core damage frequency to remain at approximately 3% of the new core damage frequency. Virtually all of the changes in the Level 2 quantification are a result of the changes made to the Level 1 accident sequence analysis in the form of reduced time available for operator action in preventing core damage. ATWS sequences make up the bulk of the increase in large early releases due to the shorter time available to the operator to initiate standby liquid control and effect shutdown in the Level 1 portion of the quantification.

Little modification to the Level 2 analysis was necessary Only station blackout sequences were affected due to e slightly reduced amount of time available to recover offsite p0wer. The major contnbutors to large early releases remain the same as indicated in the baseline analysis and include ATWS, hydrogen combustion, and interfacing LOCA nequences. As in the base case analysis, the majority of the Level 2 accident sequences either do not result in containment failure, are vented or released through a pool, or are estimated to occur many hours into the accident. Figure 48-1 presents the distribution of Level 2 accident sequence results for the terate sensitivity analysis.

HW M ft C 'M50FFKTVlWOR!7TrMPt. ATPlillIOC 38

Table 481 1-Plant Damage States Reactor Pressure at Vessel Containment Failure Mode Timing of Release Failure R-Recovered in vessel XX-Containment intact X-Containment intact L-Vessel pressure low at lower VS-Containment vented through L-Late release head penetration pool

-24 hrs H-Vessel pressure high at lower VB-Containment vented I-Intermediate release head penetration bypassing pool

>6 hts OD-Overpressure failure due to E-Early release steam from decay heat or

<6 hrs noncondensible gas generation OT-Overtemperature failure OA-Overpressure failure due to steam generation from ATWS OH-Overpressure failure due to hydrogen combustion EC-Containment failure due to early severe accident challenges LM-Liner melt through Cl-Containment isolction failure BY-Containment bypass J

• 197 ft b C WiM4 FK l' %IWOllD TL%tPL AltTIIi1 LMC 39

Table 4812-Release Modes Debns Radionuclide Small Failure Large Failure Cooling locations prior Before or as Delayed Before or as Delayed after Systems to vessel failure a result of after Vessel a result of Vessel Vessel Failure Vessel Failure Failure Failure Through CSI WWA/DW Suppression WWW Pool No CS +

WWA/DW B13 B23 RPV in)

WWA No CS +

WWA/DW No RPV inJ WWW Dypass CS WWA/DW C1 014 C2 D24 Suppression WWW C3 C4 Pool No CS +

VNVA/DW C5 D34 C6 D44 RPV Inj WWW C7 C8 No CS +

WWA/DW C9 D54 C10 D04 No RPV in)

WWW G11 C12

~

Dypass CS WWA/DW Containment WWW No CS +

WWA/DW E15 RPV Inj WWW E2 No CS +

WWA/DW No RPV inj WWW Other Release Modes:

A1 Recovered in vessel, containment leakage only A2 Recovered in vessel, vent through the suppression pool A3 Recovered in vessel, vent through the drywell A4 Core debns in containment. containment leakage only AS Core debris in containment, vent through the suppression pool AB Core debris in containment, vent through the drywell 1 CS = Containment spray 2 DW = Drywell, WWW = wetwel' below waterline, WWA = wetwell airspace 3 Debris cooling systems and radionuclide location before vessel failure do not affect the release category definition due to the large suppression pool decontamination factor. Timing of release vs, vessel failure is not important to the release category definition as releases are principally limited to noble gases.

4 Distinction between WWA/DW and WW is not significant to the Release Mode definition because of the long time for natural aerovo' removal (gravitational settling and gradual flow of steam and aerosols to the suppression pool prior to containment failure are assumed to be effective in removing early aerosol releases from the vessel).

5 Ei = Containment isolation Failure, E2 = Containment bypass 4W ll15 C AIVOlft HINWultD !!MPLAllTililInW' 40

n[

2K

>e 7i Oz

i. 5 c!

s

_.'-T.''C

.f.',T.'l

'C

+

_ g = =uy

=7.._

\\\\

r,

\\\\

s\\

'I g

1, ss s

s

'\\\\

\\l l,;{,. 9m;>.

is;;w _n

' v.' l

's

~

Oi?-QL^;.q;;_ ~^-

l ki

'$b 4 (4. ~g.

h_?gg I

y' g

}

sPnsw c:..........f.. :... ~'y

+

5.kk NiA i

z.

\\

s 0

.:.w i

)

i

'1 t

b i.

l.

s

\\,h:

f[

\\

,, \\

i

~

a ue sa s.

e j

\\

x!

7 1

\\\\ l

[ 8lh[ h

.E 3

E p

,c m

w r,-~

t

! M[k)[.1 7

Er;= -."

l p

i

%,k (f 4 /

2E:5

[

[ }

OM

(E~; f g

.......m.....ae............

Im

!!j J

$ (l g

o Ei

\\8 4

T t

\\"

• S fl

..p. e 4

'50.. (

,- e, s 3

4 f

6;i I

h- _ _.

g; l

4 3

E I

WiH Jt B C MMH KTi%1N%ORD f tAtPL Af 011II DOC 41

Part II. Risk Analysis Results for IPEEE The external events review was conducted for three distinct categories. seismic, internal fires, and othcr events. This is similar to the manner in which the iPEEE was performed. The methodology used to perform the analysis for both internal fires and tomado missiles involved quantification. The results of the new quantification performed as a part of the rerate analysis are provided below.

Internal Firea NSPNAD-92003, Rev. 0 (IPE) and NSPNAO 95001, Rev.1 (IPEEE) discuss the classification of core damage sequences into functional categories based upon charactoristics of the accidant sequences with respect to reactor and containment conditions at the time core damage is assumed to occur. Thesc functional categories ar6 called

• accident classes?

The potential types and frequencies of accident scenarios at a nuclear power plant cover a broad spectrum. In order to limit these sequences to a manageable number, sequences with similar functional cf aracteristics are grouped together. Three such functional classes were defined for the Monticello fire IPEEE:

Class 1 A. Transient-initiated events in which all high pressure injection systems become unevailable and depressurization of the reactor to allow low pressure injection is not accomplished. Core damage is assumed to occur with the reactor at high pressure for these sequences.

Class 10. Transient-initiated events in which all high and low pressure injection systems become unavailable. Depressurization of the reactor is successful for these sequences. Core damage is assumed to occur at a low reactor pressure.

Class 2 Coro damage events which occur as a result of the inability to remove decay heat from the containment. All means of heat removal are assumed not to function for this accident class, including the main condenser, containment venting, and RHR in shutdown cooling, suppression pool cooling, and wetwell and drywell spray modes. This accident sequence takes days to develop, saturation of the pool taking more than eight hours, pressurization to containment design on the order of a day, and closure of SRVs prohibiting low pressure injection at least thirty hours into the event. High pressure systems must also fail to result in core damage for this accident class.

These accident classes are typical of other PRAs and are a subset of those used in the Monticello intemal events PRA. Other acudent classes that were not considered to be applicable to the fire PRA include:

Class 1B. Station blackout. No single fire area is likely to result in a loss of all AC power at Monticello.

Class 3 LOCAs. No fire inrtiator was identified that could credibly lead to a loss of coolant accident.

Class 4 ATWS. No fire initiator was identified that could oedibly lead to a failure of the reactor protection system The simultaneous, independent failure of the reactor protection system or of control rod insertion during a fire is probabilistically insignificant.

The human error rates for the important human errors in the fire PRA are provided in Table 48-2-1. A comparison of the quantification results for the baseline case and the rerate case are provided in table 48-2-2. The rationale for the increase in CDF associated with Class 1 A is similar to that for the internal events PRA. But, unlike class 2 of the intemal events evaluation, a relatively large portion of the increase in CDF from the internal fires is a result of accident sequences in which containment heat removal is lost. Fires contributed more to accident class 2 because there are a few locations in the plant that contain support equipment for multiple trains of decay heat removal systems (such as RHR and the hard pipe vent). The increase in CDF due to the terate results from a reduction in time to repair failed decay heat removal equipment (from 27 hours3.125e-4 days <br />0.0075 hours <br />4.464286e-5 weeks <br />1.02735e-5 months <br /> to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />).

m n a r atmmawrwomumtnim roc 42 j

Tomado Missiles Tornado generated missiles have a potentialin striking and penetrating certain esposed areas of the plant and thus damaging safety-related equipment that are required for accident mitigation. Although the power terate does not alter the tornado missile sinke probabilities, it does affect the time fut operators to mitigate the various accident scenarios that may occur. To assess the effect of the shorter response time available to the operators, the affected human error events were revised accordingly and the accident sequences were re quantified. The resulting CDF of tha base case and the terate case for the areas that are vulnerable to a missile strike is presented in Tat,le 48 2 3.

Table 48.21 Internal Fire HEPs Basic Events Description Base Case Rcrate Probability Probability XRPVBLDWNY ~ j'essel Depressunzation 1.1E 3 1.4E 3 ASDS asilure to man ASDS panel 3 4E-3 3 4E 3 RLOTORCLGY Failure to align torus cooling 2.4E5 2.4E 5 SUP MC Suppression in the Main Control Room 1.0E 2 1.0E 2 SUP-CS Suppression in Cable Sprearting Area 5 OE 2 5 OE 2 DCD40X,XXXY Failure to align attemate battery charger 1.0E-2 1.0E 2 D40 to RCIC PUMP-4BHR Failure to repair failed pump in the reactor 2.4E 1 2.8E 1 building (Class 2)

ELECT-6HR Failure to repair failed l&C component 3.7E 1 3.9E 1 (Class 1A) s Table 48 2 21.ovel1 Core Damage Frequencies by Accident Classes (Base and Rerate Cases)

Internal Fire Accident Class Core Damage Core Damage ACDF rel to ACDF rel to Frcquency baseline Frequency.112%

overall baseline baseline Accident (per year)

Power Lerel(per power level CDF Class year) 1A 2.91 E-06 3.19E-06 3.36%

9.62 %

10 3.25E-06 3 25E 06 0 00 %

0.00 %

2 2.18E 06 2.36E-06 2,16%

8.26 %

Overall CDF j 8.34E-06 8.80E-06 5.52 %

5.52 %

4t97118 C WWM TKT% tNuoRDT[htPt.All'illi(Ms0 43

Table 45 2 3 Core DamaDe Frequency Due to Tornado Generated Missiles Tornado Generated Missiles Vulnerabilities Plant Areas that are Affected by a Core Damago Core Damage Delta CDF Missile Strike (Associated Fire Area)

Frequency.

Frequancy 112%

relevant to baseline Power t.evel(per overall baseline (par year) year) power level CDF Walls / roof TB 951 Turbine deck (FIRE-1.63E 09 1.0$E 09 0.17 %

Xil)

East face RB 985, louvers (FIRE 3A) 6 80E-10 6.82E 10 0 02 %

South face RB Ground, double doors 7.60E 09 7.00E 09 0 00 %

-=

South face RB Ground, double doors 5.79E 11 0.03E 11 0 02 %

(FIRE X)

West face RB Ground, access door 1.20E 09 1.20E 09 0 00 %

West face RB 946,louvera 1.85E 09 1.85E-09 0 00 %

West face RB 955, louvers 1.85E 09 1.85E-09 0 00 %

West face TB 945, electrical 2.07E 10 2.07E-10 0 00 %

penetration (FIRE 14A)

West face T B 935, electrical 1.26E 11 1.31E 11 0 00 %

penetration (FIRE X)

DG roof, diesel intake / exhaust lines Screened Screened 0.00 %

Overall CDC 1.51 E-08 1.51E-08 0 20 %

49.

On page 1014, the second paragraph states

• Human Reliability Analysis (HR t),. This portion of the PRA involves some of the largest uncertainty in failure probabiltty estimates..

• In view of such uncertainty in the HRA and since the CDF increase of 2 4E-6Nr (or 17.5 percent increase from the Monticello baseline CDF) is not considered insignificant, the staff needs to review the uncertainty analyses to understand how uncertainties were addressed, both quantitaticaly and qualitatively, in the decision vnaking process NSP Response, The largest effect that terate has on the PRA is due to the time available for operators to respond to conditions occurring in a variety of accident sequences. Several types of uncertainty were considered in reviewing the impact of terate on the operator actions included in the PRA. The following addresses the reviews performed to examine these uncertainties. These reviews identified no new operator actions that are important to the rerate that were not already considered in the original evaluation. The model uncertainty discussion below concludes that the contribution of operator actions to nskt associated with the rerate may be, in fact, less than that indicated by the PRA.

Numerical Uncertainty To determine the effects of mcertainty associated with operator actions, sensitivity studies were performed in which the failure probability for each operator action was varied over the full range of probabilities from 0 mm n n c,Mmericrewosum un en n tn 44 n

h

to 1. The results of this analysis were used to generate importance measures for each operator action As noted in the probabilistic risk assessment section of the power rerate liconse emendment request, NEDC-32546P, those operator actions to which the results were most sensitive were subject to fulther review as to the effects cf reducing the time available to perform these actio,is due to the ferate. Any operator action which, by itself, could contribute to accident sequences totaling more than 1F 6/ year were candidates for fur'her review, The operator actions identified as important as a part of this study are presented in Table 1 of question 47.

Varying 'he failure pr >babilities of each operator action over the full range bounds the effect of any numertal uncertainHs associated with their failure probabilities. However, it was recognized that if accident sequence cut :ets existed in which more than one operator action occ/Jrred, there could b6 a multiplicative effect on the overall core damage frequency that would not be recognized by varying the failure probabilities one orerator action at a $me. For this reason, additional sensitivity studies were performed to identify those out sets containing multiple post initiator operator actions. In addition to examining multiple operator actions in the retained cutsets, the sensitivity studies included a search for cutsets that would have had mult9 e operator actions if they had not been truncated from the results. The t

results of this sensitivity study are contained in Table 491. No new operator actions were idenused as a result of this evaluation that would be affected by the ferate that were not already noted in Table i et question 47.

Model Uncertainty Operator actions to initiate emergency depressurization during transients and initiate SLC in response to an ATWS make up the bulk of the change in risk due to the terate. In reviewing the outcome of the PRA evaluation,it should be noted that there are some modeling assumptions that have an effect on these actions.

For transient events requiring emergency depressurization, few sequences credit the ability of the CRD pumps to makeup to the reactor, thereby extending the time assumed to be available to depressurize the reactor. Furthermore, the SRVs in their depressurization mode are essentially modeled as a manually initiated system due to an assumption that ADS is always inhibited on low reactor water level in accordance with the EOPs. In fact, ADS is automatic. That the operators recognize the need to inhibit ADS reflects their awareness of low reactor water level conditions and increases the likelihood that emergency depressurization will be initiated successfully on reactor inventory reaching the top of active fuel. Each of these two modeling assumptions artificially increase the contribution of failing to initiate emergency depressurization on the overall CDF.

With respect to SLC initiation, examination shows that the accident sequences which are affected most by the terate are those in which feedwater continues to run, maintaining reactor level and hence reactor power relatively high. In quantifying ATWS sequences, an assumption has been made that the reactor initialfy is operating at 100% power. No attempt has been made to use historical data to distribute initiating events between those that occur at partial vs. full power, Furthermore, complete failure of control blade insertion is assumed. That is, an attempt to quantify the potential for partial rod insertion, reducing initial power, has not been made. Both of these 6asumptions minimize the time available for the operator to respond to an ATWS. The actual contribution to core damago resulting from failure to initiate SLC may be less than quantmed in the PRA, as a result Consideration of Unmodeled Accident Scenanos The quantitative analysis performed with the PRA include intemal events, intemal flooding and intemal fires. Qualitative evaluation was per*ormed for initiators not explicitly modeled in the PRA, such as earthquakes and other external events. The ferate is expected to have little or no effect on shutdown risk.

In the PRA section of the rerate license amendment request, it was concluded that the potential for seismic w n a c mmmiwom mmmium tw 45

and other external event initiators and response of plant equipment to these events was unaffected by the rerate. Further, as the types of acc. dent sequences expected for these initiators are similar to that modeled in the internai events PRA, the timing associated with these accident sequences would be $1milar to that found in the intemal events PRA. As a result, uncertainties in important operator actions for seismic and other external events are similar to that discussed above. That is, important operator actions have been identified as a part of the internal events ORA and are found to have only limited impact on the results of the PRA.

Table 49-1 Multiple Operator Actions Tranuent CDF for this combination Both actions already identified as potentially Restore FW after a high of actions = 12E.7/yr important and included in sensitivity study to level trip determine effects of reduced timing due to Emergency terate depressurization LOCA CDF for this combination Doth actions already identified as potentially Restore FW after a high of actions = 4E 8/yr important and included in sensitivity study to level trip determine effects of reducett timing due to Emergency torate depressurization Align Service Water to the CDF for this combination Alignment of SW to the main condenser has Main Cond of actions = BE 8/yr been given only limited credit in the baseline Emergency CDF calculation due to anticipated difficulty depressurization in perfor.. ting this action (failure probability a

.75). Little impact on this action or the CDF is expected due to terate.

Place standby bat charger CDF associated with thin, Alignment of backup battery chargers is in service combination of actions <

initiated in response to de trouble alarms. It Emergency 1E-10/yr is important to accomplish over the time depressurization frame in which battery depletion would occur and is therefoto independent of terate affects.

Align Service Water to the CDF associated with this CSTs are sufficiently large that makeup Main Cond combination of actions <

would be an issue only during medium LOCA Makeup to the CSTs 1E 10/yr sequences, a small part of the CDF. Further, alignment of SW to the main condenser has been given only limited credit in the baseline CDF calculation due to anticipated difficulty in performing this action (failure probability =

.75). Little impact on this action or the CDF is expected due to rerate.

Initiate torus cooling CDF associated with this Herate affects only the diagnosis time for Repair failed heat removal combination of actions <

alignment of torus cooling which is significan' i

systems 1E.10/yr (on the order of a day) Snd therefore plays little or no role in determia.ing the potential for successfully performing this actsrs initiate torus cooling CDF associated with th:s Rerate affects only the diagnosis ;me for initiato containment venting combination of actions <

alignment of torus coohng which is significant 1E 10/yr (on the order of a day) and therefore plays little or no role in determining the potential for successfully performing this action, h47 ilIl 0 MSOtflCEwiWOMtti hiM Arttiili(KK' 46 l

1

Containment Systerg 50.

It is Indicated in USAR Table 5.2 4 thet maximum drywellpressure is 42 0 psig r.>unded off to the nearest

- psi In NEOC-32546P (Power Rotate Safety Analysis Reporf for Monticello Nuclear Generating Plant,'

proprietary infonnation e not publicly available) Table 4 f it is stated that peak drywell pressure is 4 f.0 psig at f 02 percent of f 670 MM using Mark Ilong-term program (L TP) method and 39 0 psig using Mark I LTP method with break flow from moto detailed RPV Model. Please discuss the reasons for the difference between the USAR and MARK Inumbers. Also discuss the reasons why the pressure goes up by only f psig to 40 psig when power is (sised from 1670 MM to 1880 MM using the same method.

Please confirm that if pressure is rounded off. it is rounded to the nert higher number. Please indicate key input parameterir besides power teleted that are different from the USAR and the effects on peak pressure.

NSP Response The difference between the peak drywell pressure shown in USAR Table 5.2 4 (42.0 psig) and NEDC. 32546P Table 41 (41.0 psig) is due to the values assumed for the initial containment conditions. The USAR value was based on initial conditions that maximize the peak drywell pressure While the power ferate value was based on initial conditions consistent with the Mark i Long Term Program (LTP) short term loads evaluations. The peak drywell pressure entries in Table 41 of NEDC 32540P have been revised in the table below based on using a consistent set of input assumptions that maximize the calculated peak drywell pressure. These initial conditions, with the exception of initial containment pressure discussed below, are the same as those used in the Reference i USAR analysis. An additional benchmark case for the change in initial containment pressure has been included to more clearly show the progression from the current USAR value to the power rerste peak drywell pressure. The values in the table below are shown to three

- significant figures with the last figure rounded up from the calculated value.-

The USAR peak drywell pressure of 42.2 psig was calculated using the Mark i LTP method and t Jsumed an initial containment pressure of 1.0 psig. Increasing the initial containment pressure to 2.0 psig resutted in an increase in peak drywell pressure of 0.7 psi to 42.9 psig (again using the Mark i LTP method).

Changing the break flow model to the more detailed RPV model reduces the peak drywell pressure more than 2 pal to 40.7 psig. The change in reactor operating conditions from a power level of 1070 MWt to 1880 MWt result in a small reduction in the peak drywell pressure to 39 6 psig (again using the break flow from the more detailed RPV model). The 1070 MWt and 1880 MWt entries had been inadvertently switched in the original Table 41. In all cases, there is significant margin to the 50 psig containment design pressuto -

limit.

The break flow model in the original Mark i LTP method significantly overpredicts the break flow during the first couple of seconds. The critical break flux is determined by the initial pressure and enthalpy conditions in the t essel downcomer and is assumed constant until all the subcooled water in the recirculation piping and vessel downcomer has been depleted. The break fi'w is determined by the sum of the flows through the recirculation outlet (suction) nozzle and reverse flow through the jet pump nozzles and out the recirculation discharge piping. Choked flow is assumed to occur almost immediately at the outlet nozzle.

The break flow from the recirculation discharge piping is initially assumed to be choked at the pipe end, with the flow area determined by the pipe area. Choked flow at the jet pump nozzles is assumed to occur only after the recirculation piping has been emptied. These assumptions lead to a very high break flow dar;ng the first few seconds of the event. In addition, the br9ak flow calculated in this manner is very sensitive to changes in the initial vessel pressure and downcomar enthalpy. The high initial break flow causea a high l drywell pressurization rate. The peak drywell pressure occurs quickly (at about i second), and because the peak occurs quickly, the peak value is dependent on the vent clearing characteristics. The break flow assumptions, therefore, make the calculated peak drywell pressure sensitive to the initial conditions.

The more detailed RPV model calculates a more realistic break flow during the early part of the event The RPV model has separate nodes for the downcomer and recirculation pipe. The critical flur is calcu!ated i

separately for flow between the nodes and flow out the break. The critical flux is calculated based on the pressure and enthalpy conditions during the transient. The critical flux decreases as the vessel pressure falls and the subcooling lessens, which reduces the break flow, in addition, choked flow at the jet purnp nozzles occurs much more quickly, further reducing the break flow. The initial break flow is sbout the same

-MH Jin CanOHirit%INwoltDTEMitAliallit rWT 47

ms that calcuW with the Mark I LTP method, but the break flow falls off more quickly, resulting in a slower drywell p'essun:ation rate and later peak drywell pressure. The peak drywell pressure oce9ts at about severs seceos, which is well after the vents have cleared. Thcrofore, the calculated peak value is not sensitive to the vont riaaring charac%ristics. The peak drywall pressure occurs about the time that the vessel oowncomer has emptied, which means the peak drywell pressure is primarily dependent on the ma*s and energy stored in the vessel downcomer water and the initial reactor pressure which determines the overall rate at which the downcomer emptina. The more realistic break flow calculation, therefore, makes thg calcylmed peak d'ywell p, essure let9 sens$ve to changes in the initial cond$ons than the Mark i LTP method.

1hc peak orbM preasure is p,imarily deoendent on the mest and anergy stored in the vessel downcomer water and the irdal reac'Or pressure which determir'en the overall rate at which the downcomer empties.

The peak drywell presso ecors bo 7 ickiv L oe affected by heat transfer from the core or changes in decay pcwer remling hm Wge N bdiai reactor power. The reactor pov.,r hat, a small effect on the energy stored in the doWWP'if )ute the reactor power determines the amount of feedwater flow to the vessel. The downcomer enthalpy k. determined by the mixture of the $3turated water % lrculated in ths vessel and the colder feedwatar flow. At a constant acre flow, an increase in reactor powta results la an increase in the steam flow leavi"g the vessel, slighuy warmer feedwater flow returning to the vessel (the increased steam flow provides more feedwater heating) and a decrease in the saturated recirculated water (essentially total core flow minus steam flow), T5e increased feedwater flow and temperature is offset by the reduced recirculation flow, resulting in a slight decrease in downcomer enthalpy.

For MNGP, there is o change in the reactor pressure assumed for power terate. Similarly, there is no change in the initiat Jeactor water level (mass in the downcnmer). Therefore, power terate only cht get the energy stored in the downcomer, As shnwn in Table 12 of NEDC 3254GP, the core inlet enthelpy (essentially the same as the downcomer e Malpy)is reduced by almost i Blu/lb for a power increase to 1775 MWt; the reduction in downcomer enthalpy is about double for a power increase to 1880 MWt. Power rerate, therefore, results in a slight decrease in the stored energy in the vessel downcomer. This reduction in stored 3nergy is reflected in the small decrease in the peak drywell pressure at power rerate shown in the table below.

The only key input parameter in the powcr terate analysis that is different from the USAR (besides changes related to the increase in reactor power) is the initial pressure in the drywell and wetwell. The USAR analysts assumed an initial pr6ssure of 1.C psig, while the power terate analysis assumed that the initial pressure was at the drywell pressure scram setpoint of 2 0 psig in order to provide an analytical basis for the future implementation of Improved Technical Opecifications. The increase in initial pressure causes the peak drywell pressure to increase by an amount somewhat less than the initial pressure increase. As shown in the reviseu entries for Table 41 below, the 1.0 pal increase in initial containment pressure resulted in an increase in the peak drywell pressure of 0.7 psi.

Ta'olo 41 LOCA CONTAINMENT PERFORMANCE RESULTS Parameter 102% of 1670 MWt 102% of 1880 MWt

- Limit Peak Drywell 42 20) at 1.2 seconds 56W Prescure (psig) 42.9(2) at 1.2 seconds 40 6(3) at 7.0 seconds 39.6(3) at fA9 seconds~~

Notes (1) Mark i LTP method,1 psig initial containment pressure (from Reference 1)

(2) Mark I LTP method,2 psig initial containment pressure (3) Mark I LTP method with detalled RPV model,2 psig initial containment pressure (4) Containment design pressure -

Reference ~

(1) NEDO-32418, *MNGP Containment Pressure and Temperature Response for USAR Update, Dec.1994 m an t encmimenwummuve 48

51.

It is indicated in 4. f. f. f for emergency core coohng system (ECCS) not positive suction head (NPSH) that the decrease in NPSH due to the increase in the long-term bulk suppression pool ternperature at upteled power will be offset by the suppression pool tem ('erature increase. Please provide the specinc numbers.

NSP Response Revised calculations for the containment pressure and suppression pool temperature have t,een provided to the NRC in Exhibit E (GE-NE T23007312) of NSP s License Amendment Request dated June 19,1997 which was approved by Staff SER dated July 25,1997. GE-NE-T23007312 provides the detailed input parameter values, a66umptions and results for a variety of conditions The Staff a SER and NSP'e associated letters provide justification for adequate NPSH at 1880 MM.

The NRC haw determined that an uncertainty adder of 2a (95% confidence interval) is necessary for the use of the ANS 5.1 1979 decay heat model. In the ANS 5.1 1979 standard, uncertainty is expressed as a two sided in. Since the opper bound of the normal distnbution is the parameter of interest for power terate, it is reasonable to construct the confidence interval from the one sided upper tall of the normal distribution. For this distnbution, an uncertainty of 1.645o corresponds to 95% percentile. T hat is, a vahd statistical inference with 95% confidence that the actual decay heat will be less than tne calculated value if the samphng difference is 1.645o. The 1880 MWt decay heat profile used for power rerate analyses bounds the decay heat profile of 1775 MWt with an 1.645a adder. Given the above,it is reasonable to conclude that the 1880 MWt decay heat profile bounds the actual 1775 MWt decay with 95% confidence.

52.

Please provide the confirmatory calculations valtdating the results from the analyses using the SHEX computer code.

NSP Responso The confirmatory calculations vahdating the results from the analyses using the SHEX computer code have been provided to the NRC in Appendix A of Exhibit D (GE NE.T23007312) of NSP's license amendment request dated June 19,1997 with supplements dated July 16 and July 21,1997. These confirmatory calculations and analyses vahdated the use of the SHEX computer code for performing containment calculations that maximize the suppression pool temperature and containment pressures. The results of the confirmatory calculations are presented in Table A 1 below. A comparison of the peak suppression pool temperatures obtained with the SHEX code to the values obtained with the HXSIZ rode (used for MNGP's previous licensing basis containment calculations) show that there is httle difference (about 1*F) in the peak suppression pool temperature predicted by both codes with the use of eithet May Witt or ANS 5.1 decay beat. A comparison of the peak long term secondary containment pressure (nett time of peak suppression pool temperature) shows close comparison (<1 psi) between the results obtained with HXSl2 and SHEX.

A second benchmark calculation was provided in response to Question S in a letter from NSP to the Staff dated July 10,1997. This benchmark calculation validated the rebults frorn the analyses using the SHEX computer code for containment calculations that use containment tprays or analysis assumptions (such as break flow thermal mixing efficiency) to minimize the containmen' pressure. The benchmark calculation for the wetwell pressure and the SHEX results are in close agreement (within one percent).

SW n h C *i40FIMFw twoltD TIMPt Aff wililint 49 s

TABLE A 1. SHEX CONFIRMATORY CALCULATION RESULTS (from Appendix A of Exhibit D)

CASE A1 CASE 1 A-2 CASE A 2 REF. A 1 REF. A 1 Code SHEX M3CPT/

SHEX M3CPT/

HXSIZ HXSIZ Rated Power * (MWt) 1670 1870 1670 1670 Decay Heat ANS 5.1 ANS 5.1 Ma gtt May Witt RHR Heat Exchanger K 143.1 143.1 143.1 143.1 (BTU /sec *F) total Initial Drywell & Supp.

15,7 15 7 15.7 15.7 Chamb. Airspace Pressure (psia)

Pool Temp at 600s (*F) 142.3 145 0 1446 146 0 Peak Suppression Pool Temperature (*F) 184.8 184.0 190.7 195.5 Secondary Suppression Chamber Airspace Pressure 31.4_

31.3 30 8 30.3 Peak (psla)

• Analyses performed at 102% of initial core thermal power.

REFERENCE A 1 NEDO-32418,'Monticello Design Basis Accident Cor.tainment Pressure and Temperature Response for USAR Update,' December 1994 Reactor Systems.

53.

In Section 3 2 of Exhibit E. did the overpressure analysis assume 102 percent of terated power,105 percent versted steam now, and an SRV opening tolerance of 3 percent?

NSP Response The overpressure analysis in Section 3.2 assumed an initial reactor power of 102% of 1775 MWt,1040 psia initial; team dome pressure,105% of rated core flow and an SRV opening pressure of 1142.3 psig, which includes an opening tolerance of 3 percent.

54.

In Section 3.5 of Eahibit E, the licensee should commit to performing the vibration monitoring of the reactor recirculation system (RRS) as stated in the GE generic report in Section 5.5.1.3 and the review of the plant operating data as specified in Section 5.6 2 to conntm that the RRS will accommodate the uprated Row conditions.

NSP Response Monticello will not operate the Reactor Recirculation System at any increased flow conditions due to uprate.

The maximum system flow is constrained by other limitations. Presently the recirc system flow rate is limited by the following parameters.

Recirc pump motor current limit of 390 A and winding temperature lir$t of 248'F MG Set drive motor current limit of 590 A and winding temperature limit of 230'F.

Recire pump dp limit of 143 psid.

4197 KD 04tsOHKt3.wtNw(*pitAtttATfillit rWK 50

ICF operation has been approved for Monticello. During normal plant operation with operation near the end of cycle, recirc pump speed is increased in order to increase the reactor power output. In doing so, the recirc pump dp hmit is usually reached first with the recirc pump motor current limit being approached as weli. As a result, the recirc system has been operated in the past at the maxircuin achievable flow rate as determined by the above mentioned parameters with no associated problems with the Recirc System of the reactor vesselinternals. These limits will remain in force at terate conditions and the ferate does not involve an increase in pump speed. Therefore, recirc system flow will not increase above present conditions.

9191 H b 0 thot tlCl3%l'i%1*D fl MPLAf Dillt Inst 51

_ _ _ -