ML19305A367

Submits Responses Re Startup Testing Requested by NRC 790109 & 790118 Telcons & 790206 Ltr.Current Schedule Calls for Refueling Shutdown to Begin 790323.Requests Issuance of Cycle 4 Tech Specs

ML19305A367
Person / Time
Site:	Arkansas Nuclear
Issue date:	02/27/1979
From:	Rueter D ARKANSAS POWER & LIGHT CO.
To:	Reid R Office of Nuclear Reactor Regulation
References
1-029-21, 1-29-21, NUDOCS 7903130404
Download: ML19305A367 (8)

Text

?

e

, H E L PI N G B UIL O AHhA%SAS r#-

ARK ANS AS POWER G LIGHT COMPA NY P O BOX 551

• LITTLE ROCK. AAKANSAS 72203

• C501) 371-4191 February 27, 1979 OONALD A. AUETER DIAECTOA TECHNICAL AND ENVIRONMENTAL SERVICES 1-029-21 Director of Nuclear Reactor Regulation ATTN: Mr. R. W. Reid, Chief Operating Reactor Branch #4 U. S. Nuclear Regulatory Commission Washington, D. C. 20555

Subject:

Arkansas Nuclear One - Unit 1 Docket No. 50-313 License No. DPR-51 Cycle 4 Reload Report Questions (File: 0242.6, 1511.1)

Gentlemen:

Our November 9,1978 submittal proposed Technical Specifications for Cycle 4 operation of Arkansas Nuclear One - Unit 1 (ANO-1) and provided to you our Cycle 4 Reload Report (BAW 1504) as supplemental information.

Your letter of February 6,1979, requested information discussed in our telephone conversations on January 9 and January 18, 1979, plus responses concerning startup testing. Our responses are attached.

Our current schedule calls for refueling shutdown to begin on March 23, 1979. We request issuance of the Cycle 4 technical specifications within one week following the beginning of shutdown to allow sufficient time for distribution of the license amendment and incorporation into procedures.

Very truly yours, O

lc ca & O^ / O([/7 Donald A. Rueter DAR:DGM:vb Attachment T @ 3090%

r.x ~ ,~a.,~vesranoo~eo 4t , ~1eueen u,oo<e sour- ur,c,1,es evereu

RESPONSES TO CYCLE 4 RELOAD QUESTIONS ARKANSAS NUCLEAR ONE - UNIT 1 ,

DOCKET NUMBER 50-313 Question 1 The current power distribution reliability factor, RF, shown in BAW 10119 is based on comparisons of measured and predicted power distri-butions of cores utilizing conventional three hatch, out-in, fuel manage-ment schemes. An in-out-in fuel management scheme has been proposed for cycle 4. Hence the current RF is not, a priori, applicable to cycle 4.

To support use of the current RF, confirmatory analyses should be pro-posed. Specifically, a statistical test, such as but not necessarily the F test or Bartlett test, and acceptance criteria should be proposed which will test the hypothesis that ANO-1 cycle 4 comparisons of measured and predicted power distributions are members of the family of comparisons which form the data base for the current reliability factor. Such comparisons and statistical testing should be made at at least monthly intervals (monthly surveillance is currently required) and a running tally maintained throughout the cycle.

Results of these tests need not be reported if acceptance criteria are met.

Response

We propose to periodically measure core power distributions and compare them to predicted power distributions in the following manner:

177 RMS = Z (100 Zj)2 5 7.5 1

177 where Zj is the difference between predicted and measured data for the ith assembly.

Power distributions at approximately 4, 25, 50, 100, 150, 200, 250, 254, 304, 354, and 387 EFPD will be taken and compared with predicted power distributions. We propose to use these times in cycle as they were used in the original Cycle 4 design. Results will be reported to NRC if acceptance criteria are not met.

Question 2 Please provide a description of your planned quality ass' rance u program to insure that the proposed reprogramming of control rods to altered bank designations will be successfully performed. Reference to specific approved procedures will suffice.

Response

Verification of correct control rod repatch following refueling is done by energizing and driving the control rods individually. While each control rod is driven, its absolute position indicator and relative position indicator are monitored to insure that the correct indicators move. Additionally, a technician is stationed on the Reactor head service structure to listen to the motors as they are energized and driven to insure that the correct control rods are moved as they are selected for motion from the control room.

Question 3 Provide a complete analysis of the moderator dilution accident.

Response

The moderator dilution accident is discussed in Chapter 15 of the FSAR.

Cycle 4 initial conditions would cause a specified dilution rate to result in a greater rate of reactivity insertion to the core than the FSAR values. For a moderator dilution rate of 500 gpm (the largest considered in the FSAR) the reactivity insertion rate is 1.227 x 10-5 a/k/k/sec for FSAR parameters (see Table 14-8) and is increased to 1.235 x 10-5 Ak/k/s for Cycle 4 parameters. Reactivity insertions of the order considered here result in a high RC system pressure trip. The higher insertion rate will cause a faster reactor trip. The average reactor coolant system temperature change would increase to <.04 F/sec and it can be inferred from Figure 14-10 that the peak system pressure could increase by less than 10 psi which is still s300 psi from the safety limit. It is therefore concluded that the effect of Cycle 4 parameters as initial conditions for the moderator dilution event are not significant changes.

Question 4 Provide the computational basis for the revision of the flux / flow setpoint.

Response

The flux / flow setpoint was revised o' ecause retaining devices will be installed on Burnable Poison Rod Assemblies (BPRA's) and neutron sources.

Introducing the retainer causes a small change in the core flow distri-bution as a result of the increased flow resistance in the fuel assemblies containing BPRA's and retainers. The increase in resistance was accounted for in thermal-hydraulic analyses by conservatively increasing the BPRA form loss coefficient as described in BAW 1496.

DNBR calculations were performed at the maximum design' conditions with the retainer assumed on the hot fuel assembly. These results showed that the retainer would reduce the hot assembly flow by less than 1% and that the reduction in minimum DNBR would be approximately two poirts (0.02). A comparison of Cycles 3 and 4 maximum design steady-state DNBR analysis is shown in Table 6-1 of the Reload Report. The flux / flow trip setpoint analysis for ANO-1, Cycle 4, demonstrated that the slight

A decrease in DNBR margin would be offset by reducing the trip setpoint from 1.06 to 1_.057 including margin for an 11.2% DNBR rod bow penalty.

Question 5 The ordinate of Figures 8-7, 8-8, and 8-9 should be labelled " Power, %

of Allowable." Please revise them.

Response

The ordinate axes in Figures 8-6, 8-7, and 8-8 of BAW 1504 and in Figures 3.5. D 2A, 3.5.2-2B, and 3.5.2-2C of our November 9, :978 sub-mittal are mislabeled and should read " Power, % of allowable". Please incorporate this change into the approved technical specifications.

Question 6 Extensive use of lumped burnable poisons to hold down excess react'vity and tailor power distributions, as has been proposed, is a potentially more difficult problem in a reload core than a fresh core. This potential problem has been addressed as question (1). An alternate apporach is to carefully monitor reactivity anomolies. Please provide a detailed description of your reactivity anomoly check, renormalization proce-dures, if any, and review criteria. It is understood that you employ a review criteria of 1/2%ap, as distinct from a Technical Specification limit. of 1%ap. Please confirm.

Response

Between the end of a cycle and the beginning of the next cycle, pre-dicted reactivity worth parameters are used to update the Reactivity Balance Calculation. These design calculations are verified during startup testing. During startup testing and early in cycle lifetime, data is gathered periodically at steady-state equilibrium operating conditions. When sufficient datn is gathered, the burnup dependent reactiv ? ?.y parameters are normalized to reflect actual core conditions.

As stated in Section 4.9 of the Technical Specifications, "Following a normalization of the computed boron concentration as a function of burnup, the actual boron concentration of the coolant shall be peri-odically compared with the predicted value." The anomaly check is reviewed by a Senior Reactor Operator and changes to the Reactivity Balance Calculation are approved by a Senior Reactor Operator and are reviewed by the onsite Plant Safety Committee. Administrative limits require further analysis if measured boron concentration is not within 1 0.5%ak/k of the predicted. The Technical Speci'fication limit is i 1.0%

ak/k.

Question 7 Please provide the predicted maximum batch and maximum assembly burnup at end of cycles 4, 5, and 6.

/

Response

The burnups listed below for Cycle 4 are taken from the final design.

The burnups listed below for Cycles 5 and 6 are based upon a preliminary design and are subject to change when the final design is completed for these cycles.

Maximum Batch f*2 Cycle N Assembly Burnup Burnup MWD /MTU MWD /MTU 4 32711 28282 5 35666 29677 6 37651 35208 Question 8 Table 7-2 of your submittal shows values of the maximum ejected and dropped roo worths for Cycle 4 are within the bounds of the reference analysis. Please confirm that associated peaking factors are also bounded and provide the predicted values.

Response

The peaking factor used for the FSAR analyses are design peaks and are bounding with respect to predicted Cycle 4 peaking values. The methods and techniques for rod ejection accidents, which have been approved by the NRC, are described in the FSAR. The approved techniques have been benchmarked against space-time-kinetics methods.

Question 9 Similarly, please show calculated values of the prompt neutron lifetime, and delayed neutron fraction, predicted for Cycle 4, and the correspond-ing values used in the reference safety analysis.

Response

The values of eff and neutron lifetime used in the FSAR are provided in the FSAR Table 14-27 and are repeated along with the calculated Cycle 4 values below.

FSAR ,

Cycle 4 peff BOL .0071 .00617 E0L .0053 .00517 Neutron Lifetime, BOL 24.8 24.

uS EOL 23.0 27.

Although the ejected rod accident analysis presented in the FSAR is based on larger p ff values, the FSAR analysis is still bounding for Cycle 4. This is a result of the larger rod worth (0.65% a k/k) assumed for the ejected rod in the FSAR analysis, as compared to 0.55% Ak/k

/

for Cycle 4. This is best illustrated if the reactivity is expressed in termsofdollars=p/peff. For the FSAR case, this is 3 = .0065/.0071 =

.92. For Cycle 4, reactivity added is $ = .0055/.00617 = .89, which is less than the FSAR value.

Question 10 Values at the predicted moderator temperature coefficient are shown in Table 7-2 of your submittal and shown to bound values used in the safety analyses. It is understood that the safety analyses do not employ values of these coefficients directly, but rather curve fits of reactivity vs density and temperature over a range of densities and temperatures spanning nominal and upset conditions. Please confirm that predicted values for cycle 4, over the full range of postulated states, are bounded by the values of reactivity vs density and reactivity vs temperature used in the safety analyses, and provide the cycle 4 predicted values and the values assumed in the safety analyses.

Response

Accident analysis is approached from the concept of evaluating bounding or " worst case" events. The initial conditions that result in worst case events are based on the extremes in moderator and Doppler coefficients throughout core life. Therefore, accidents are evaluated at t'he time in life (BOL or E0L) which would result in the more severe consequences with respect to the parameter of concern. Therefore, it is only necessary to compare the extremas in reactivity coefficients. The consequences of middle of life conditions are bounded by either BOL or E0L conditions.

The moderator reactivity feedback is modeled either as a reactivity coefficient which includes the effects of temperature and density, or a moderator density curve depending on the accident under evaluation and the range of 10% or less from nominal HFP conditions. The moderator reactivity coefficient (HFP) is a good approximation of the moderator density curve over the narrow range of density changes experienced during the transients. Since this approach was used in the F5AR analysis, it is appropriate to compare reactivity coefficients at HFP to determine validity of Cycle 4 over previous FSAR input for the narrow range of interest.

Question 11 Please commit to provide a physics startup test report similar to the report for cycle 3. ,

Response

We will submit a physics startup test report similar to the Cycle 3 Startup Test Report within 90 days following completion of startup testing.

/

Question 12 Section 9.4 of your submittal describes actions to be taken if accep-tance criteria are not met. The Control Rod Group Reactivity Worth Test description is of insufficient detail. Are the planned actions for this test the same as those for cycle 3, as stated in your March 20, 1978 submittal? If not, please state how they differ and why.

Response

The Control Rod Group Reactivity Worth Test procedures for Cycle 4 are similar to those used in Cycle 3. Each measured control rod group must agree within i 15% of the predicted worths and the sum of the measured control rod group worths (Groups 5, 6 and 7) must agree within i 10% of the corresponding predicted worth. -

The acceptance criteria of i 15% on each measured control rod group is for information only and no specific action will be required if the acceptance criteria is not met since no safety limits are associated with this limit. If the acceptance criteria of i 10% of the predicted worth for Groups 5, 6 and 7 is not met, an evaluation by the plant's nuclear engineering group will be performed with involvement of the fuel vendor and the onsite safety committee as needed.

Probable courses of action will be to re-perform the test with con-sideration for increasing precision or for measurement of an additional rod group (s) concurrent with re-verification of calculated values and generation of a calculated worth for the additional rod group (s) to be measured.

If after retesting and/or reanalysis involving the fuel vendor, results are still unacceptable, NRC will be notified since the shutdown analysis will be affected.

Question 13 Section 9.4 describes actions to be taken if Acceptance Criteria are not met. Please state who shall perform the " evaluation." Will the on-site safety committee review the evaluation prior to power escalation? What auditable records are to be maintained?

Response

The results of all tests will be reviewed by the plant's nuclear engi-neering group. If the acceptance criteria of the startup physics tests are not met, an evaluation will be performed by the plant's nuclear engineering group with assistance from general office personnel, Middle South Services and the fuel vendor, as needed. The results of this evaluation will be presented to the onsite Plant Safety Committee.

Resolution will be required prior to power escalation. If a safety question is involved, the off-site Safety Review Committee would review the situation and NRC would be notified if an unreviewed safety question exists. All recorded data work sheets and associated computer printouts specified by the test procedures governing the testing are retained.

/

Question 14 Section 9.2.4 describes the ejected control rod reactivity worth test.

It does not address " swap" of the symetric rods with the measured rod as you did during the cycle 3 startup. Do you intend to perform rod swap tests? If not, please explain why and describe alternate tests which will be performed to validate core symmetry prior to exceeding 5% of rated power.

Response

The rod swap test was performed at the beginning of Cycle 3. We plan to perform the rod swap test during Cycle 4 startup testing.

The ejected control rod reactivity worth test was also performed during Cycle 3 startup testing, and will again be performed during Cycle 4 startup testing. Our March 3,1978 response to Question Al describes our verification process for proper core loading and symmetry.