ML20128D841

SER Accepting Util 720922 Request for Change to TS Re Rod Drop Accident

ML20128D841
Person / Time
Site:	Monticello
Issue date:	11/27/1973
From:	Reid R, James Shea, Ziemann D US ATOMIC ENERGY COMMISSION (AEC)
To:
References
NUDOCS 9212070400
Download: ML20128D841 (9)

Text

. , .,,~w  :.

l pg -

• '{

UNITED STATES ATOMIC ENERGY COMMISSION SAFETY EVALUATION BY THE DIRECTORATE OF LICENSING NORTHERN STATES POWER COMPANY DOCKET No. 50-263 R0D DROP ACCIDENT In a letter dated September 22, 1972, Northern States Power Company submitted a requent for-changes to the Technical Specifications for the Monticello reactor concerning the rod drop accident. .In response to our requests, Northern States Power Company submitted additional infor-mation in. letters dated January 18, March 2, April 11, and October 4, 1973. In addition, a meeting was held on May 17, 1973, with repre-sentatives of Northern States Power Company and the General Electric Company to review the calculational models and to discuss the input assumptions to be used in the change to the rod drop accident technical specifications. The change is based on new calculational models developed by the General Electric Company, presented in references 1, 2, and 3, and by a change in the assessment of the accident and scram reactivity shape. . These changes result in a reduction in maximum allowable in-sequence control rod reactivity worth from 2.5% to 1.3% delta k/k, and increase the assurance that .a control rod is not in an out-of-sequence position during low power operation.

The rod drop accident is one of the design basis accidents for boiling

water reactors. For calculational purposes it is assumed that a control

! rod blade separates from its drive, lodges in the core with the drive j withdrawn, and drops at the time which causes the most serious. power excursion due to. rapid reactivity insertion. The consequences of this accident are evaluated by determining the energy input to the fuel assuming 1

(1) Paone, C. J. , Stirn, R. C. , and Wooley, J. A. , " Rod Drop Accident

, Analysis for Large Boiling Water _ Reactors", NEDO-10527, March 1972.

(2) Stirn, R. C. , Paone, C. J. , : and Young, R. M. , " Rod Drop Accident Analysis for Large BWR's",-Supplement 1 - NEDO-10527, July 1972.

(3) Stirn, R. C., Paone, C. J., and Haun, J. M., " Rod Drop Accident Analysis for Large Boiling Water Reactors Addendum No.'2 Exposed Cores", Supplement 2 - NEDO-10527, January 1973. .

9212070400.731127~

PDR- ADOCK 05000263 PDR p ,

. - , . . . . c-. . . ..

that the reactivity worth of the dropped rod is the maximum which could occur. The maximum acceptable energy in the fuel is limited such that, in the event of fuel cladding failure, the energy input into the coolant will not result in a pressure pulse which might damage the core geometry or the reactor pressure vessel.

The analytical methods used by the General Electric Company (CE) to evaluate the consequences of the rod drop accident have been reviewed by the staf f and independent calculations have been performed by Brookhaven National Laboratory which show reasonable agreement with GE results. Based on these reviews, it is concluded that the analytical methods used by GE are acceptable.

Application of the GE analytical methods to operating reactors requires that the input parameters conservatively represent the reactor core over a broad range of operating conditions. The proposed changes to the Technical Specifications include, in the Bases, a set of boundary conditions which are used to calculate the maximum allowable reactivity worth of a control rod. It is not expected that these boundary conditions will be exceeded for reactor cores of current design. The boundary conditions include a maximum inter-assembly local power peaking factor, an end-of-cycle delayed neutron fraction, a beginning of life Doppler reactivity feedback, the technical specification control rod scram insertion rate, a control rod drop velocity of 3.11 ft/sec, and specified accident and scram reactivity shape functions. The. rod drop velocity of 3.11 ft/sec is based on tests with a " worst case" rod built with maximum clearances and features known to contribute the high rod drop velocities. The dif ference between the mean rod drop velocity and the 99.9% confidence limit for a group of production rods was added to the mean velocity obtained for the " worst case" control rod. We have included in the Bases the value 0.005 end-of-cycle delayed neutron fraction to further define the boundary assumptions that were used in the calculations.

In addition, we have added a statement to the Bases that each reload core must be analyzed to show conformance to the bounding assumptions.

The peak fuel enthalpy resulting from an in-seruence rod drop accident within the above boundary conditions is calculated not to exceed 280 cal /gm, which is acceptably below the peak i al enthalpy at which prompt fuel dispersal would occur based on the StERT ' tests. Based on the above, the resultant maximum allowable in-sequence rod worth of 1.3% delta k/k is acceptable.

Separate consideration is being given to the potentially adverse effect of compaction of boron carbide in the control rods on the rod drop accident in the event of inverted poison tubes. The evaluation of the effect of possible inverted poison tubes on the allowable in-sequence rod worth is currently in progress and if determined necessary, appro- '

priate changes to the allowable control rod reactivity worth will be made.

4

..,-,4 m

/

If a control rod is withdrawn out of sequence, a rod worth of greater than 1.3% delta k/k could result. In the event of rod drop accident associated with such an out-of-sequence rod, the peak fuel enthalpy could exceed 280 cal /gm. The rod worth minimizer (RWM) is designed as an operator aid to prevent an out-of-sequence rod withdrawal.

Current Technical Specifications allow the RWM to be bypassed if it is inoperable during a reactor startup provided that a second operator is assigned to monitor the rod withdrawal sequence. To increase the control on RWM availability during reactor startups, the technical specification is being changed to require that the RWM be operable for the withdrawal of a significant number of control rods. The ef fective date of the change in technical specifications concerning RWM operability is being deferred for six months to allow any necessary upgrading of the RWM to be accomplished.

Based on the above, we conclude that the proposed changes do not involve significant hazards considerations and that there is reasonable

' assurance that the health and safety of the public will not be endat.gered.

es J. Sh a Operating Reactors Branch f2 Directorate of Licensing qf fad' 'l'I' i er Robert W. Reid Operating Reactors Branch #2 Directorate of Licensing 2ivm h%w Dennis L. Zieman Chief Operating Reactors Branch #2 Directorate of Licensing Date; November 27, 1973 9

3.0 LIMITING CONDITIONS FOR OPERATION 4.0 SURVEILLANCE REQUIREMENTS (b) when the rod is withdrawn the first time subsequent to each re-fueling outage, observe discern-ible response of the nuclear in-strumentation. However, for initial rods when response is not discernbile, subsequent exercising of these rods after the reactor is critical shall be performed to observe nuclear instrumentation response.

2. The control rod drive housing 2 The control rod drive housing support syste-support system shall be in place shall be inspected after reassembly and ' c during reactor power operation and the results of the inspection recorded.

when the reactor coolant system is pressurized above atmospheric pres-sure with fuel in the reactor vessel, .

unless all operable control rods are fully inserted and Specification 3.3.A.1 is met.

3. (a) Control rod withdrawal sequence 9 3. (a) To consider the rod worth minimizer e shall be established so that the operable, the following steps must be maximum calculated reactivity that performed: ,

could be added by dropout of any increment of any one control blade (1) The control rod withdrawal s' 1 will not make the core more than for the rod worth minimizer 1.3% Ak supercritical. computer shall be verified as Correct.

s (ii) The rod worth minimizer compute on-line diagnostic test shall b successfully completed.

(iii) Proper annunciation of the sele- ,

error of at least one out-of-sequence control rod in each fully inserted group shall be verified.

1 1/ t. 11 WF

4.0 SURVEIIIANCE REQUIREMENTS 3.0 LIMITING CONDITIONS FOR OPERATION (iv) The rod block function of the rod worth minimizer shall be verified by attempting to with-draw an out-of-sequence control rod beyond the block point. ,

(b) If the rod worth minimizer is inoperal (b) Whenever the reactor is in the while the reactor is in the startup of startup or run mode below 10% run mode below 10% rated thermal pove:

rated thermal power, no control and the second independent operator rods shall be moved unless the or engineer is being used, he shal' -

rod worth ninimizer is operable verify that all rod positions are or a second independent operator correct prior to commencing withdrawa or engineer verifies that the of each rod group.

operator at the reactor console is following the control rod program. After May 1, 1974, the second operator may be used as a substitute for an inoperable rod worth minimizer during a startup only if the rod worth minimizer fails after withdrawal of at least twelve control rods.

4. Prior to control rod withdrawal for

4. Control rods shall not be withdrawn startup or during refueling verify for startup or refueling unless at that at least two source range least two source range channels have an observed count rate equal to or channels have an observed count rate of at least three counts per second.

greater than three counts per second.

5. Whenever the Engineer, Nuclear, deter-

5. Whenever the Engineer, Nuclear, deter- mines that a limiting control rod pattern mines that a limiting control rod exists, an instrument functional test pattern exists, withdrawal of desig- of the RWM shall be performed prior to nated control rods shall be permitted withdrawal of the designated rod (s) and only when the RWM system is operable, daily thereafter.

78 3.' 3/ 4. 3-4

K v10cd wfGnangh 11 Gt@ L1/1//IJ.

~

b Exhibit B (Continued)

Bases Continued 3.3 and 4.3:

A. Reactivity Limitations

1. Reactivity margin - core loading The core reactivity limitation is a restriction to be applied principally to the design of new fuel which may be loaded in the core or into a particular refueling pattern. Satisfaction of the limitation can only be demonstrated at the time of loading and must be such .that it will s,'

apply to the entire subsequent fuel cycle. The generalized form is that the reactivity of the ,

~

core loading will be limited so the core -can be made subcritical by at least R + 0.25% ok in .

the most reactive condition during the operating cycle, with the strongest control rod fully withdrawn and all others fully inserted. The value of R in % Ak is the amount by which the core the time of reactivity, at any time in the operating cycle, is calculated to be greater thanA at core which con-the check; 1.e., the initial loading. R must be a positive quantity or zero.

tains temporary control or other butr.able neutron absorbers may have a reactivity characteristic which increases with core lifetime goes through a maximum and then decreases thereafter. See Figure 3.3.2 of the FSAR for such a curve.

The value of R is the difference between the calculated core reactivity at the beginning of the operating cycle and the calculated value of core reactivity any time later in the cycle where it would be greater chan at the beginning. For the first fuel cy,cle, R was calculated to be 0.012 Ak. A new value of R must be 2 termined for each fuel cycle. ,

The 0.25% Ak in the expression R + 0.25% Ak is provided as a finite, demonstrable, sub-criticality margin. This margin is demonstrated by full withdrawal of the strongest rod and partial withdrawal of an adjacent rod to a position calculated to insert at least R + 0.25% Ak in reactivity. Observation of sub-criticality in this condition assures sub-criticality with not only the strongest rod fully withdrawn but at least a R + 0.25% ak ,

margin beyond this.

2. Reactivity margin e stuck control rods Specification 3.3.A.2 requires that a rod be taken out of service if it cannot be moved 82 ,

3.3/4.5-8

. . . . o -- - - a , s a : a.

Exhibit B (Continued)

Bases Continued 3.3 and 4.3 Section 6.5.3. This support is not required if the reactor coolant system is at atmospheric pressure -

since there would then be no driving force to rapidly eject a drive housing. Additionally, the support is not required if all control rods are fully inserted since the reactor would remain sub-critical even in the event of complete ejection of the strongest control rod.

3. Control rod withdrawal and insertion sequences are established to assure that the maximum in-sequence -

individual control rod or control rod segments which are withdrawn could not be worth enough to cause

- manner the coredefined to be more for the than Rod0.013 Dropdelta k sup(ercritical Accident. if they were

3) These sequences aretodeveloped drop out ofprior the to core in the initial operation of the unit following any refueling outage and the requirement that an operator follow these sequences is backed up by the operation of the RWM. This 0.013 delta k limit, together with the integral rod velocity limiters and the action of the control rod drive system, limit potential reactivity insertion such that the results of a control rod drop accident will not exceed a maximum fuel energy content of 280 cal /gm. The peak fuel enthalpy content of 280 cal /gm is below the energy content at which rapid fuel dispersal and primary system damage have been found to occur based on experimental data as is discumsed in reference 1.

Recent improvements in analytical capability have allowed more refined analysis of the control rod drop accident. These techniques have been described in a topical report and two supplements.(1)(2)(3)

By using the analytical models described in these reports coupled with conservative or worst-case input parameters, it has been determined that for power levels less than 10% of rated power, the -

specified limit on in-sequence control rod or control rod segment worths will limit the peak fuel enthalpy content to less than 280 cal /gm. Above 10% power even single operator errors cannot result in out-of-sequence control rod worths which are sufficient to reach a peak fuel enthalpy content of 280 cal /gm should a postulated control rod drop accident occur.

'(lY 'P'a'o'n'e',' J'C' ' ',S't'irm R C and Wooley, J A, " Rod Drop Accident Analysis for Large Boiling Water Reactors ," 6 NEDO-10527, March 1972.

(2) Stirn, R C, Paone, C J, and Young, R M, " Rod Drop Accident Analysis for Large BWR's," Supplement 1 - ,

NEDO-10527, July 1972.

(3) Stirn, R C, Paone, C J, and Haun, J M, " Rod Drop Accident Analysis for Large Boiling Water Reactors Addendum No. 2 Exposed Cores," Supplement 2 - NEDO-10527, January 1973. ,

84

Exhibit B (Continued) i Bases Continued 3.3 and 4.3

~

The following conservative or worst-case bounding assumptions have been made in the analysis used to ,

-determine the specified 0.013 delta k limit on in-sequence control rod or control rod segment worths.

- The allowable boundary conditions used in the analysis are quantified in reference 4. Each core 3

reload will be analyzed'to show conformance to the limiting parameters.

a. A startup inter-ass'embly local power peaking factor of 1.30 or less. (5) '

b. An end'of cycle delayed neutron fraction'of 0.005.  ;

c. A beginning of life Doppler reactivity feedback,

d. The Technical Specification rod scram insertion rate. ,

e. The maximum possible rod drop velocity (3.11 ft/sec). .

4 . . .

The design accident and scram reactivity shape function.,

f.

g. .The uoderator temperature at which criticality occurs.

l

- It .is recognized that these bounds are conservative with respect to expected operating-conditions. If 4

any'one.of the-above conditions is not satisfied, a more detailed calculation will be done to show

• compliance'with the 280 cal /gm design limit. . .

l In most cases; the worth of in-sequence rods or rod segments will be substantially less than 0.013 ., '

delta k. - -Further, the addition of 0.013 delta k vorth of reactivity as a result of a rod drop and in

'a conjunction with the' actual. values of the other important accident analysis parameters described i above would roost likely result in a peak fuel enthalpy'substantially less than the 280 cal /gm design _

i

< limit. However, the .0.013 delta k limit.:is applied in order to allow room for future reload changes '

i :and ease of verification without repetitive Technical Specification changes. .

v a .. ,

'(~4)' 'R'epor't' entitYed.," Technical Basis for Changes to Allowable Rod Worth Specified in Technical Specification - 3.3.B.3. (a)" transmitted by letter from L. O. Mayer (NSP) to J. F. O' Leary (USAEC) dated October 4,1973.

(5) To include the power spike effect caused by gaps between fuel pellets.- -. - . - .

84a ,

..._-,-.~--.u,.,,,. .

Exhibit B (Continued)

Bases Continued 3.3 and 4.3 Should a control rod drop accident result in a peak fuel energy content of 280 cal /gn, less than 660 (7 x 7) fuel rods are conservatively estimated to perforate. This would result in offsite doses twice that previously reported in the FSAR, but still well below the guideline values of 10 CFR 100.

For 8 x 8 fuel, less than 850 rods are conservatively estimated to perforate, which has nearly the same consequences as for the 7 x 7 fuel case because of the operating rod power differences.

The RW provides automatic supervision to assure that out-of-sequence control rods will not be with-drawn or inserted; i.e., it ilmits operator deviations from planned withdrawal sequences. Reference Section 7-9 FSAR. It serves as an independent backup of the normal withdrawal procedure followed by the operator. In the event that the RWM is out of service when required, a second independent operator or engineer can manually fulfill the operator-follower control rod pattern conformance function of the RW. In this case, procedaral control is exercised by verifying all control rod positions after the withdrawal of each group, prior to proceeding to the next group. Allowing substitution of a second independent operator or engineer In case of RW inoperability recognizes the capability to adequately monitor proper rod sequencing in an alternate manner without unduly restricting plant operations. Above 10% power, there.is no requirement that the RWM be operable since the control rod drop accident with out-of-sequence rods will result in a peak fuel energy content of less than 280 cal /gm. To assure high RWM availability, the RWM is required to be operating during a startup for the withdrawal of a significant number of control rods for any startup after May 1,1974.

4. The Source Range Monitor (SRM) system performs no automatic safety system function; i.e., it has no scram function. It 'does provide the operator with a visual indication of neutron level.

This is needed for knowledgeable and efficient reactor startup at low neutron levels. The 4

84b