ML18046A818

From kanterella
Jump to navigation Jump to search
Proposed Revisions to Tech Spec Sections 1.0 & 3.10, Specifying New Limits for Radial Peaking Factors & Allowable Linear Heat Rates & Implementation of Excore Detectors for Use in Core Power Distribution Monitoring
ML18046A818
Person / Time
Site: Palisades Entergy icon.png
Issue date: 07/21/1981
From:
CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:
Shared Package
ML18046A817 List:
References
NUDOCS 8107280379
Download: ML18046A818 (137)


Text

CONSUMERS POWER COMPANY Docket 50-255 Request for* Change**to 'the *Technical* Specifieations Licens-e*DPR*20 For the reasons hereinafter set forth, it is requested that the Technical Specifications contained in the Provisional Operating License DPR-20, Docket 50-255, issued to Consumers Power Company on October 16, 1972, for the Palisades Plant be changed as described in Section I below:

/

I. Changes A. Add the following definitions to Section 1.0:

Axial Offset The difference between the power in the lower half of the core and the upper half of the core divided by the sum of the powers in the lower half and upper half of the core.

Narrow Water Gap Fuel Rod A fuel rod adjacent to the narrow inter-fuel assembly water gap (a gap not containing a control rod).

Narrow Water Gap Fuel Rod Peaking Factor - FN The maximum product of the ratio of individu~l fuel assembly power to core average fuel assembly power times the highest narrow water gap fuel rod local peaking factor integrated over the total core height including tilt.

B. Change the title of Section 3.10 from "CONTROL ROD AND POWER DISTRIBUTION LIMITS" to "CONTROL RODS."

C. Delete Section 3.10.3 in its entirety and replace with the following:

11 3.10.3 Part-Length Control Rods The part-length control rods will be completely withdrawn from the core (except for control rod exercises and physics tests)."

D. Changes to other sections of 3.10: In Section 3.10.7, change "3.10.3.f" to 11 3.10.3." In 3.10.8, change 11 3.10.5 and S-5-1" to "and 3.10.5 11 ; and 11 3.10.3 11 to 11 3.23."

E. Delete the following from the "Basis" section of Section 3.10:

(1)* The paragraph starting with "The limitation on linear heat ... "

an d en d ing

  • h II ... ra t e l'1mi.'t is wit
  • no t excee d e d . II (2) The first four sentences of the next paragraph, from: "When a flux tilt exists ... " to " ... local and total power distribution."

(3) The paragraph starting with "The limitations on FA ... " to " ... rod r

group insertion limits."

nu0781-0171a-43 8T072Slf:3~79-ST072] REGULATORY DOCKET FILE COPY nPDR

P ADOCK 05000255

2 F. Delete 11 (6) XN-NF-77-24" and 11 (7) XN-NF-78-16" from the "References" section of Section 3.10.

G. Delete Figures 3-9 and 3-10 (they will be replaced by Figures 3.23-1 and 3.23-2) .

  • nu0781-0171a-43

H. Delete Section 3.11 in its entirety and replace with the following:

3.11 POWER DISTRIBUTION INSTRUMENTATION 3.11.1 INCORE DETECTORS LIMITING CONDITION FOR OPERATION The incore detection system shall be operable:

a. With at least 50% of the incore detectors and 2 incores per axial level per core quadrant.
b. With the incore alarming function of th~ datalogger operable and alarm setpoints entered into the datalogger within the previous 7 days of power operation.

APPLICABILITY (1) Item a. above is applicable when the incore detection system is used for:

Measuring quadrant power tilt, Measuring radial peaking factors, Measuring linear heat rate (LHR), or Determining target Axial Offset (AO) and excore monitoring allowable power level.

(2) Items a. and b. above are applicable when the incore detection system is used for monitoring LHR with automatic alarms.

(Incore Alarm System.)

ACTION 1:

With less than the required number of incore detectors, do not use the system for the measuring and calibration functions under (1) above.

ACTION 2: With the alarming function of the datalogger inoperable, do not use the system for automatic monitoring of LHR (Inoperable Incore Alarm System) .

nu0781-0171a-43

POWER DISTRIBUTION INSTRUMENTATION 3.11.1 INCORE DETECTORS LIMITING CONDITION FOR OPERATION ACTION 2: (Contd)

Operation may continue using the excore monitoring system as specified in 3.11.2 or by meeting the requirements of 3.23.1.

Basis The operability of the incore detectors with the specified minimum .

complement of equipment ensures that the measurements obtained from use of this system accurately represent the spatial neutron flux distribution of the reactor core. The operability of the incore alarm system depends on the availability of the datalogger as well as the operability of a minimum number of incore detectors. Incore alarm

  • setpoints must be updated periodically based on measured power distributions. The incore detector Channel Check is normally performed by an off-line computer program that correlates readings with one another and with computed power shapes in order to identify inoperable detectors.

nu0781-0171a-43

POWER DISTRIBUTION INSTRUMENTATION 3.11.2 EXCORE POWER DISTRIBUTION MONITORING SYSTEM LIMITING CONDITION FOR OPERATION The excore monitoring system shall be operable with:

a. The target Axial Offset (AO) and the Excore Monitoring Allowable Power Level (APL) determined within the previous 31 days using the incore detectors, and the measured AO not deviated from the target AO by more than 0.05 in the previous 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
b. The AO measured by the excore detectors calibrated w.ith the AO measured by the incore detectors.
c. The quadrant tilt measured by the excore detectors calibrated with the quadrant tilt measured by the incore detectors.

APPLICABILITY:

(1) Items a., b. and c. above are applicable when the excore detectors are used for monitoring LHR.

(2) Item c. above is applicable when the excore detectors are used for monitoring quadrant tilt.

ACTION 1:

With the excore monitoring system inoperable, do not use the system for monitoring LHR.

ACTION 2:

If the measured quadrant tilt has not been calibrated with the incores, do not use the system for monitoring quadrant tilt.

. ' Basis The excore power distribution monitoring system consists of Power Range Detector Channels 5 through 7.

The operability of the excore monitoring system ensures that the assumptions employed in the PDC-II analysis(l) for determining AO limits that ensure operation within allowable 1HR limits are valid.

nu0781-0171a-43

  • POWER DISTRIBUTION INSTRUMENTATION 3.11.2 EXCORE POWER DISTRIBUTION MONITORING SYSTEM LIMITING CONDITION FOR OPERATION Basis (Contd)

Surveillance requirements ensure that the instruments are calibrated to agree with the incore measurements and that the target AO is based on the current operating conditions. Updating the Excore Monitoring APL ensures that.the core LHR limits are protected within the+/- 0.05 band on AO. The APL considers both LOCA and DNB based LHR limits, and factors are included to account for changes in radial power shape and LHR limits over the calibration interval.

The APL is determined from the following:

LHR(Z)TS APL = [LHR(Z)M ]

x V(Z) x Ep (Z) x 1.02 M'~

x Rated Power u

Where:

(1) LHR(Z)TS is the limiting LHR vs Core Height (from Section 3.23.1),

(2) LHR(Z)Mu is the measured peak LHR including uncertainties vs Core Height, (3) V(Z) is the function (shown in Figure 3.11-1),

(4) E (Z) is a factor to account for the reduction of allowed LHR in the peak rod p

with increased exposure (Figure 3.23.2) such that:

For fuel rod burnups less than 27.0 GWd/MT - E = 1.0 p

For fuel rod burnups greater than 27.0 GWd/MT but less than 33.0 GWd/MT -

E p

= 1.0 + 0.0064 x LHR For fuel rod burnups greater than 33.0 GWd/MT - E p

= 1.0 + 0.0012 x LHR

  • Where LHR is the measured fuel rod average LHR in kW/ft, nu0781-0171a-43

POWER DISTRIBUTION INSTRUMENTATION

3. 11. 2 EXCORE POWER DISTRIBUTION MONITORING SYSTEM LIMITING CONDITION FOR OPERATION Basis (Contd)

(5) The factor of 1.02 is an allowance for the effects of upburn, (6) The quantity in brackets is the minimum value for the entire core at any elevation (excluding the top and bottom 10% of core) co~sidering limits for peak rods, interior fuel rods and narrow water gap fuel rods. Ep (Z) is only applied if the minimum value is based on limits for the peak rod.

If the quantity in brackets is greater than one, the APL shall be the rated power level.

Reference (1) XN-NF-80-47 nu0781-017la-43

IJ:*

I. 18 I. 16

( . 9ij, I

  • I50)

( I . 0 , I . I 50 )

I. Iij I. 12 I. I0 (0,1.11) (0.77, I. I I) 1.08 N

> 1.06 I. Oti 1.02 1.00 0.98 o.o 0. I 0.3 0.5 0.6 0.7 o.a 0.9 I. 0 FRACTJON OF ACTIVE FUEL HEIGHT AXIAL VARIATION BOUNDING CONDITION Palisades FIGURE. 3.11-1 Technical Specifications

I. Add Section 3.23 as follows:

3.23 POWER DISTRIBUTION LIMITS 3.23.1 LINEAR HEAT RATE (LHR)

LIMITING CONDITION FOR OPERATION The LHR at peak power elevation Z shall not exceed 15.28 kW/ft x FA(Z) x FB(E) [the function FA(Z) is shown in Figure 3.23-1 and the function FB(E) where Eis the fuel rod burnup is shown in Figure 3.23-2]. The LHR at the peak power elevation in any interior fuel rod shall not exceed 14.16 kW/ft x FC(Z); and, the LHR at the peak power elevation in any narrow water gap fuel rod shall not exceed 15.02 kW/ft x FC(Z) [the function FC(Z) is shown in Figure 3.23-3].

APPLICABILITY: Power operation above 50% of rated power.

ACTION 1:

When using the incore alarm system to monitor LHR, and with four or more coincident incore alarms, initiate within 15 minutes corrective action to reduce the LHR to within the limits and restore the incore readings to less than the alarm setpoints within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> or failing this, be at less than 50% of rated power within the following 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.

ACTION 2:

When using the excore monitoring system to monitor LHR and with the AO deviating from the target AO by more than 0.05, discontinue using the excore monitoring system for monitoring LHR. If the incore alarm system is inoperable, within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> be at 85% (or less) of rated thermal power and follow the procedure in ACTION 3 below.

nu0781-0171a-43

POWER DISTRIBUTION LIMITS 3.23.1 LINEAR HEAT RATE (LHR)

LIMITING CONDITION FOR OPERATION ACTION 3:

If the incore alarm system is inoperable and the excore monitoring system is not being used, operation at less than or equal to 85% of rated power may continue provided that incore readings are recorded manually.

Readings shall be taken on a minimum of 10 individual detectors per quadrant (to include 50% of the total number of detectors in a 10-hour period) within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> and at least every 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> thereafter. If ~eadings indicate a local power level equal to or greater than the alarm setpoints, the action specified in ACTION 1 above shall be taken.

Basis The limitation on LHR ensures that, in the event of a LOCA, the peak temperature of the cladding will ~at exceed 2200°F. (1) In addition, the limitation on LHR for the highest power fuel rod, narrow water gap fuel rod and interior fuel rod ensures that the minimum DNBR will be maintained above 1.30 during anticipated transients; and, that fuel damage during Condition IV events such as locked rotor will not exceed acceptable limits. C2 )C 3 ) The inclusion of the axial power distribution term ensures that the operating power distribution is enveloped by the design power distributions.

Either of the two core power distribution monitoring systems (the incore alarm system or the excore monitoring system) provides adequate monitoring of the core power distribution and is capable of verifying that the LHR does not exceed its limits. The incore alarm system performs this nu0781-0171a-43

POWER DISTRIBUTION LIMITS 3.23.l LINEAR HEAT RATE (LHR)

LIMITING CONDITION FOR OPERATION Basis (Contd) function by continuously monitoring the local power at many points throughout the core and comparing the measurements to predetermined setpoints above which the limit on LHR could be exceeded, The excore monitoring system performs this function by providing comparison of the measured core AO with predetermined AO limits based on incore measurements. An Excore Monitoring Allowable Power Level (APL), which may be less than rated power, is applied when using the excore monitoring system to ensure that the AO limits adequately restrict the LHR to less

  • 4 than the limiting values. C )

If the incore alarm system and the excore monitoring system are both inoperable, power will be reduced to provide margin between the actual peak LHR and the LHR limits and the incore readings will be manually collected at the terminal blocks in the control room utilizing a suitable signal detector. If this is not feasible with the manpower available, the reactor power will be reduced to a point below which it is improbable that the LHR limits could be exceeded. The time interval of 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and the minimum of .10 detectors per quadrant are sufficient to maintain adequate surveillance of the core power distribution to detect significant changes until the monitoring systems are returned to service.

To ensure that the design margin of safety is maintained, the determination of both the incore alarm setpoints and the APL takes into account a measurement uncertainty factor of 1.10, an engineering nu0781-0171a-43

POWER DISTRIBUTION LIMITS 3.23.1 LINEAR HEAT RATES (LHR)

LIMITING CONDITIONS OF OPERATION Basis (Contd) uncertainty factor of 1.03, a thermal power measurement uncertainty factor of 1.02 and allowance for quadrant tilt.

References (1) XN-NF-77-24 (2) XN-NF-77-18 (3) XN-NF-78-16 (4) XN-NF-80-47 nu0781-0171a-43

er::::

s 0.9 2 *g*-~*******""******""""""'

' -.r 2:

u...

ACCEPTABLE 0 0 ** OPERATION z:

0 ti BREAK POINTS:

-.r er:::: I * ( * ~9 , I. 0) u.

er::::

2. (.7, .91)
3. (1.0, .67) **::::::..
5 0.7 LJ.J m

~ 0.65 0

4t

0. I 0.2 0.3 0.5 0.6 0.7 0.8 0.9 I. 0 LOCATION OF AXIAL POWER PEAK (FRACTION OF ACTIVE FUEL HEIGHT)

ALLOWABLE LHR AS A FUNCTION Palisades FIGURE 3. 23-1 OF PEAK POWER LOCATIQN Technical Speci.flca.tion$

I. IJ -----~~~~~~~~mw.~rnmrnrnrnm~~

-a:

c

....J

-er

E u..

0 0.8 z:

C>

l-(IJ3.60, 0.78) e.>

-er ACCEPTABLE OPERATION a:

u.. 0.6 a:

c

....J LLI

....J O.IJ all

-er g

....J

....J

-er 0.2 0 5 10 15 20 25 30 35 I.JO FUEL ROD BURNUP (GWD/MT)

ALLOWABLE LHR AS A FUNCTION OF BURNUP Palisades FIGURE 3.23-2 Technical Specifications

I -

0:::

c

...J

E
E

~ 0.9

E u..

0

z: ACCEPTABLE 0

l-OPERATION e..>

~ BREAK POINTS:

0:::

u.. I. ( . qg ' I. 0)

--- o. 8 0:::

2. ( I. 0'
  • 78)
c

...J La.J

...J co

~

~

...J

...J

~ 0.7 0 0.1 0.2 0.3 0.5 0.6 0.7 0.8 0.9 1.0 LOCATION OF AX IAL POWER PEAK (FRACTION OF ACTIVE FUEL HEIGHT)

ALLOWABLE LHR AS A FUNCTION OF ACTIVE FUEL Palisades HEIGHT FOR INTERIOR FUEL RODS Technical Specifications FIGURE 3.23-3

POWER DISTRIBUTION LIMITS 3.23.2 RADIAL PEAKING FACTORS LIMITING CONDITION FOR OPERATION The value of FA shall be less than or equal to 1.43 [ 1. 0 + 0.5(1-P)],

r The value of FT r shall be less than or equal to 1. 77 [ 1. 0 + 0.5(1-P)],

aH The value of Fr shall be less than or equal to 1.64 [1.0 + 0.5(1-P)], and N

The value of Fr shall be less than or equal to 1.74 [1.0 + 0.:5(1-P)] where p is the core thermal power in fraction of rated power.

APPLICABILITY: Power operation condition above 50% of rated power.

ACTION:

  • A T ~H N .

With (F ), (F ), (F ) or CF) exceeding its limit:

r r r r Within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />, reduce thermal power to less than:

FrA Cl) [1-2 C _ - l)] x Rated Power; 1 43 FrT C2) [1-2 ci. n - l)] x Rated Power; Whichever Is Lower F~H C3) [1-2 Ci. 64 - l)] x Rated Power; or (4) [1-2 F:r:N - l)] x Rated Power ci. 74 Basis The limitations on FA, FT, FaH and FN are provided to ensure that assumptions r r r r used in the analysis for establishing DNB margin, LHR and the thermal margin/low-pressure and high-power trip setpoints remain valid during operation.

Data from the incore detectors are used for determining the measured radial peaking factors. The periodic surveillance requirements for determining the measured radial peaking factors provide ~ssurance that they remain within prescribed limits. Determining the measured radial peaking factors after each

  • fuel loading prior to exceeding 50% of rated power provides additional assurance that the core is properly loaded.

nu0781-0171a-43

  • POWER DISTRIBUTION LIMITS 3.23.3 QUADRANT POWER TILT - T q

LIMITING CONDITION FOR OPERATION The quadrant power tilt (T ) shall not exceed 5%.

q APPLICABILITY: Power operation above 50% of rated power.

ACTION:

1. With the quadrant power tilt determined to exceed 5% but less than or equal to 10%, correct the power tilt within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> or determine within the next 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and at least once every 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> thereafter, that the radial peaking factors are within the limits of Section 3.23.2, or reduce power at the normal shutdown rate to less than 85% of rated power.
2. With the quadrant power tilt determined to exceed 10%, correct the quadrant power tilt within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after exceeding the limit or reduce power to less than 50% of rated power within the next 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.
3. With the quadrant power tilt determined to exceed 15%, be in at least hot standby within 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.

Basis Limitations on quadrant power tilt are provided to ensure that design safety margins are maintained. Quadrant power tilt is determined from excore detector readings which are calibrated using incore detector measurements. (l)

Calibration factors are determined from incore measurements by performing a two-dimensional, full-core surface fit of deviations between measured and theoretical incore readings and integrating the fitting function over each core quadrant. Values of LliR and radial peaking factors are increased by the value of quadrant tilt .

nu0781-0171a-43

POWER DISTRIBUTION LIMITS 3.23.3 QUADRANT POWER TILT - T q

LIMITING CONDITION FOR OPERATION References (1) FSAR, Section 7.4.2.2 nu0781:..0171a-43

  • J. Add Sections 4.18 and 4.19 as follows:

4.18 4.18.1 POWER DISTRIBUTION INSTRUMENTATION INCORE DETECTORS SURVEILLANCE REQUIREMENTS 4.18.1.1 The incore detection system shall be demonstrated operable:

a. By performance of a Channel Check prior to its use following a core alteration and at least once per 7 d~ys during power operation when required for the functions listed in Section 3 .11.1.
b. At least once per refueling by performance of a Channel Calibration which exempts the neutron detectors but includes electronic components.

4.18.1.2 The incore alarm system is demonstrated operable through use of the datalogger program out-of-sequence alarm. The out-of-sequence alarm is demonstrated operable once per refueling by performance of a Channel Check.

nu0781-017la-43

a. A target AO and excore monitoring allowable power level shall be determined using excore and incore detector readings at steady state near equilibrium conditions.
b. The excore measured AO shall be compared to the incore measured AO. If the difference is greater than 0.02, the excore monitoring system shall be recalibrated.
c. The excore measured Quadrant Power Tilt shall be compared to the incore measured Quadrant Power Tilt. If the difference
  • is greater than 2%, the excore monitoring system shall be recalibrated .
  • nu0781-0171a-43
  • 4.19 4.19.1 POWER DISTRIBUTION LIMITS LINEAR HEAT RATES SURVEILLANCE REQUIREMENTS 4.19.1.1 When using the incore alarm system to monitor LHR, prior to operation above 50% of rated power and every 7 days of power operation thereafter, incore alarms shall be set based on a measured power distribution.

4.19.1.2 When using the excore monitoring system to monitor LHR:

a. Prior to use, verify that the measured AO has not deviated from the target AO by more than 0.05 in the previous 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
b. Once per day, verify that the measured Quadrant Power Tilt is less than or equal to 3%.
c. Once per hour, verify that the power is less than or equal to the APL and not more than 10% of rated power greater than the power level used in determining the APL.
d. Once per hour, verify that the measured AO is within 0.05 of the established target AO.

nu0781-017la-43

  • 4. 19 4.19.2 POWER DISTRIBUTION LIMITS RADIAL PEAKING FACTORS SURVEILLANCE REQUIREMENTS 4.19.2.1 The measured radial peaking factors (FA, FT F~H and FN) r r' r r obtained by using the incore detection system, shall be determined to be less than or equal to the values stated in the LCO at the following intervals:
a. After each fuel loading prior to operation above 50% of rated power, and
b. At least once per week of power operation .
  • nu0781-0171a-43
  • 4.19 4.19.3 POWER DISTRIBUTION LIMITS QUADRANT POWER TILT - T q

SURVEILLANCE REQUIREMENTS 4.19.3.1 Calculate the Quadrant Power Tilt using the excore readings at least once per 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> when the excore detectors deviation alarms are inoperable .

  • nu0781-0171a-43
  • K. Add Item 16 and Note (h) to Table 3.17.4.

Table 3.17.4 (Contd)

Minimum Minimum Permissible Operable Degree of Bypass No Functional Unit Channels Redundancy Conditions 16 Excoie Detector None None Deviation Alarms (h) Calculate the tilt using excore readings once per 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> when the excore detector deviation alarms are inoperable.

nu0781-017la-43

II . DISCUSSION

  • The proposed Technical Specification changes are intended to accomplish four objectives:
1. To incorporate a burnup dependent linear heat rate limit for H, I and future fuel types.
2. To modify the radial peaking factor limits.
3. To adopt power distribution monitoring with the excore detectors as an alternative to the incore alarms.
4. To adopt the standard Technical Specifications format for power distribution monitoring and power distribution limits.

A. Burnup Dependent Linear Heat Rate Limit The basis for the burnup dependent LHR limit is in the Appendix to XN-NF-81-34. The allowable LHR is reduced at fuel rod average burnups greater than 27.25 GWd/MT to offset the adverse effects of fission gas

  • release on predicted clad rupture and flow blockage. This limit is implemented in the Technical Specifications by the additional multipli-cative term, FB(E) in the equation for limiting LHR. FB(E) is shown in Figure 3.23-2. Since the separate LHR limits for interior fuel rods and narrow water gap fuel rods are based on DNB considerations only, the burnup dependent limit is not applied to them unless they are also the peak LHR rods within an assembly.

B. Radial Peaking Factors New limits for Assembly Radial Peaking Factor - F!, Total Radial Peaking T ~H Factor - Fr and Total Interior Fuel Rod Radial Peaking Factor - Fr are proposed. A new limit, Total Narrow Water Gap Fuel Rod Peaking Factor -

FH, is also included. The basis for these limits are presented in r

  • Section 7.0 of XN-NF-81-34. Separate DNB analyses were performed for assemblies containing 208 fuel rods and 216 fuel rods, and the lower nu0781-0171a-43

limiting peaking factor between the two types was chosen for the Technical Specifications limit on fuel assembly, interior fuel rod and narrow gap

  • c.

fuel rod. The LOCA analysis is limiting for wide gap fuel rods, so that the previous limit of 1.77 is retained for the total radial peak.

Excore Power Distribution Monitoring The proposed Technical Specifications also include provisions for monitoring the limits on LHR.with the split excore detectors as an alternative to the incores when the datalogger is inoperable. The method employed is based on Exxon Nuclear Company's PDC-II as previously reported in XN-NF-77-57, "Exxon Nuclear Power Distribution Control for Pressurized Water Reactors, Phase 2," dated January 1978. In this method, a target axial offset and an allowable power level are chosen periodically based on incore measurements of margin to the LHR limits. Operation is then allowed within a narrow axial offset band around the target. The method was originally formulated for Westinghouse reactors, so an analysis was performed to verify its applicability to Palisades. The report of this analysis, XN-NF-80-47, "Palisades Power Distribution Control Procedures,"

dated 1980 is enclosed.

The Palisades instrument used to measure axial offset (the Power Ratio Recorder) does not have all the features of the corresponding Westinghouse system. (Neither the target offset nor the allowable offset band width can be varied automatically with power level, and there is no timer to automatically record time outside the target band.) Because of this, only a very limited portion of the procedure allowed under PDC-II is being implemented. There is only one allowable offset target band (+/- 0.05) and it does not vary with power level. If the datalogger is inoperable and, because of the need to maneuver or some other reason, the operator cannot maintain the axial offset within the band, the existing alternative has

  • been retained, wherein the reactor is limited to 85% power and incore outputs are recorded by hand.

nu0781-0171a-43

  • D. Standard Format It is currently planned to adopt the standard format for Palisades Technical Specifications. Therefore, this change reorganizes portions of the existing specifications into that format. It is expected that, with some renumbering of paragraphs, the new sections submitted herein will fit into the final form of the standard Palisades Technical Specifications with only a mino.r review required.

In the current change, the portion of Section 3.10 (having to do with power distribution limits 3.10.3) has been moved into a new section of its own, 3.23. Section 3.11, Incore Instrumentation, has been revised to include requirments for the excores as well as the incores and is now entitled "Power Distribution Instrumentation."

  • nu0781-0171a-43

The following is an itemization identifying the source of the proposed specifications:

New Section Old Section Discussion 3.11.1 INCORE DETECTORS LCD a. 3.11.a The requirement for operable instrumentation above 50%

power is now inherent in the applicability statement for linear heat rate limits (3.23). Since there are now five levels of incores, the requirement for 10 detectors per quadrant is redundant to the requirement for 2 per axial level per quadrant and has been deleted. (Since a description of the system is not present in the Standard Technical.Specifications basis, it has been deleted from the Palisades basis also.)

LCD b. 3.11.c & Determining alarm setpoints weekly is required as

3. 11. d before. An operational datalogger is required for incore alarms to be operable.

APPLICABILITY (1) 3.11.a The uses of the incores are specified in more detail than previously. Incore alarms are not necessary for measurement functions.

(2) 3.11.b The requirement for an operable datalogger is applicable when using the alarm feature when monitoring LHR.

ACTION 1 3.11.a Similar to Standard. ACTION 1 corresponds to APPLICABILITY (1).

ACTION 2 3 .11. b Similar to Standard. ACTION 2 corresponds to APPLICABILITY (2).

SURVEILLANCE REQUIREMENTS 4.18 .1.1 Similar to Standard Technical Specifications.

4 .18 .1. 2 The out-of-sequence alarm alerts the operator to datalogger problems.

3.11.2 EXCORE MONITORING SYSTEM LCD a. Required to implement excore monitoring of linear heat rate. Requirements conform to PDC-II methodology as outlined in XN-NF-80-47.

LCD b. 3 .11 Basis Axial offset must be calibrated.

nu0781-017la-43

New Section Old Section Discussion LCD c. 3 .11 Basis Quadrant tilt must be calibrat.ed.

APPLICABILITY (1) All calibrations are required for excore LHR monitoring.

(2) Axial offset calibration is not required if the excores are only being used to monitor quadrant tilt.

ACTION 1 Self-explanatory. ACTION 1 corresponds to APPLICABILITY 1.

ACTION 2 Self-explanatory. ACTION 2 corresponds to APPLICABILITY 2.

3 .11. 2 Basis The basis contains the method for selecting the excore monitoring allowable power level. The equation implements the procedure described in Section 3.2 and 3.3 of XN-NF-80-47. An additional factor is included that restricts the APL to account for the effects of the burnup dependent LHR. The coefficients in the equations for E are based on the amount the LHR p

limit could decrease in 31 days of full-power opera-tion as derived from Figure 3.23-2 of the new Technical Specifications.

SURVEILLANCE REQUIREMENTS 4.18.2.1 3.11 Basis Surveillance implements the calibrations required for operability .of the excores for the purpose of LHR monitoring and quadrant tilt surveillance.

3.23.1 LINEAR HEAT RATE LCO 3.10.3.a Changes from the current specification include:

1. The addition of the term FB(E) to the computation of peak IJIR and the new Figure 3.23-2.

2.* Reduction in the limit for interior rod LHR and addition of limit on narrow water gap rod in accordance with the DNB analysis in XN-NF-81-34 .

  • nu0781-0171a-43

New Section Old Section Discussion 3.23.1 LINEAR HEAT RATE (Contd)

LCO 3.10.3.a 3. Figures 3.9 and 3.10 have been renumbered 3.23-1 and 3.23-3.

4. The uncertainty factors are moved to the basis in accordance with the standard format.

APPLICABILITY 3.11.a The power level at which IJIR must be monitored is lower than that implied by the old specification, which allowed operation up to 65% power without incores.

ACTION 1 3.11. b The new specification is similar to the Standard Specifications.

ACTION 2 Axial Offset Monitoring as a backup to the incore alarms is proposed. If the Axial Offset cannot be maintained within the allowed band, operation may continue using the incores following the procedure in ACTION 3.

ACTION 3 3.11.e & The new specification is the same as the old.

3.11.f SURVEILLANCE REQUIREMENTS 4.19.1.1 3.11.c & The new requirement is similar to the old.

3 .11.d

4. 19 .1. 2 The surveillance ensures that axial offset, power level and quadrant tilt remains within the bounds of the PDC-II analysis.

3.23.2 RADIAL PEAKING FACTOR LCO 3.10.3.g ~ and F~H have been decreased to conform to the Cycle 5 DNB analysis, and a new limit on narrow water gap rod peaking factor has been added. The exception for H fuel in Cycle 4 has been dropped.

APPLICABILITY The applicability is the same as for linear heat rate

  • nu0781-0171a-43 limits.

New Section Old Section Discussion 3.23.2 RADIAL PEAKING FACTOR (Contd)

ACTION 3.11.g The new requirement is similar to the old.

SURVEILLANCE REQUIREMENTS 4.19.2.1 3.11.d& The new requirement is similar to the old.

3.11.g 3.23.3 QUADRANT POWER TILT LCD 3.10.3.e The limit of 0.05 is the same as. before.

APPLICABILITY The applicability is the same as for LHR limits.

ACTION

1. 3.10.3.e The action statement for 0.05 tilts is more conserva-tive than the old because it requires immediate corrective action. Reductions to 85% power is*

sufficient to assure safe operation with tilts less than 10%.

2. 3.10.3.c Again the action is more conservative than the old because the 24-hour delay has been eliminated.
3. 3.10.3.b The old requirement is unrealistic. Increasing the thermal margin low-pressure trip setpoint by 400 psi would trip the plant, so the result would be a shut-down in any case. Methods of measuring LHR are not valid for very large tilts.

SURVEILLANCE REQUIREMENTS 4.19.3.1 No specific surveillance requirements are in the current specifications .

  • nu0781-017la-43

ATTACHMENT A t'o Technical Specification Change Request L

I.

r

(

I PALISADES CYCLE 5 RELOAD FUEL SAFETY ANALYSIS REPORT MAY 1981 "RICHLAND, WA 99352 1.-

ElS{ON NUCLEAR COMPANY, Inc.

8107280386 810721 PDR ADOCK 05000255 p PDR I

r

  • 1 XN-NF-81-34 J

I

\

Issue Date: 06/10/31 PALISADES CYCLE 5 RELOAD FUEL I SAFETY ANALYSIS REPORT i

I

  • Prepared by: R. G. Gru1T1T1er~ ~

W. V. Kayser~~/}/

F. T. Adams ~~

Approved. by ef# /71;{£' ?(4~t;Jf

" . B. Skogen, nager G"/u/,f/

~ 1 i! PWR Neutronics Approved by: JWIJ. <:.,~ 0 R. B. Stout, Manager

.Jol(l.,,JI Neutronics & Fuel Management Approved by: l . ~ vl ~ . ,. . _

. N. organ, Manage 5-.;:JO . "If I Licensing &Safety Engineering CO~'TRDltfD COPY

/mar

  • E)${0N NUCLEAR COMPANY, Inc.

i XN-NF-81-34 TABLE OF CONTENTS Section Page

1.0 INTRODUCTION

. 1 2.0

SUMMARY

. . . 2 3.0 OPERATING HISTORY OF THE REFERENCE CYCLE . . .. . 5 3.1 CYCLE 4 STARTUP TESTS . . . 5 3.2 OPERATING STATUS OF CYCLE 4 (APRIL, 1981) 5 4.0 CYCLE 5 CORE DESCRIPTION . . . . . ...

. 10 5.0 RELOAD I FUEL ASSEMBLY DESIGN 16

. 6.0 NUCLEAR DESIGN . . . . . . . * . . . 19 6.1 PHYSICS CHARACTERISTICS . 21 6..1.1 Power Distribution Considerations 21 6.1.2 Control Rod Reactivity Requirements . . 22 6.1. 3 Moderator Temperature Coefficient Considerations .. 22 6.2 NUCLEAR DESIGN METHODOLOGY . . 23 7.0 SAFETY ANALYSIS ....... . 33 7.1 THERMAL HYDRAULIC ANALYSIS . 33 7.2 PLANT TRANSIENT ANALYSIS .* 35 7.3 ECCS ANALYSIS . . . . . . 35 7.3.1 Reload I ECCS Limits

  • 36 7.3.2 Spare Rods Assembly ECCS Limits .
  • 36

II

XN-NF-81-34 TABLE OF CONTENTS (Continued)

Section Page 7.4 ROD EJECTION ANALYSIS-. . . . . . . . 37 REFERENCES .... .... 42 APPENDIX A .... .... 45

  • AP.PENDIX B ..... 0 .... . . 48

iii XN-NF-81-34 LIST OF TABLES I

! Table Page

. i 2.1 Palisades Cycle 5 Summary of Core Characteristics 4 4.1 Fuel Assembly Design Parameters 12 4.2 Summary of Core Parameters 13 5.1 Fuel Design Summary . 18 6.1 Calculated Neutronics Characteristics of Cycle 5 Compared with Cycle 4 . . . . . . . . . . . . 24 6.2 Control Rod Shutdown Margins and Requirements for Cycle 5 * . * . . * * . . . . . . 25 7.1 Thermal Hydraulic Design Conditions . . . . . 38 7.2 Transient Events Considered in the Palisades Cycle 5 Plant Transient Analysis . . . . . . . . . . . . . . . 39 7.3 Important Core Kinetics Parameters Used in the Palisades Cycle 5 Plant Transient Analysis 40 7.4 Palisades Rod Ejection Accident . . . . 41 A-1 Palisades Exposure Sensitivity Results for H-Fuel at 2530 MWT . . . . . . . . . . . . . . . . . . . 46

iv XN-NF-81-34 LIST OF FIGURES .

Figure Page 3.1 Palisaqes Cycle 4 Operating History . . . . . . * . 7 3.2 Palisades Cycle 4 Critical Boron Concentration vs.

Exposure, HFP, ARO . . . . . . . . . . . . . . . . 8 3.3 Palisades Cycle 4, INCA Power Distribution (Measured) versus PDQ Calculated Relative Assembly Power, 100% Power . 9 4.1 Planned Cycle 5 Loading Pattern . . . . . . . . . . . 14 4.2 Cycle 5 Loading Pattern and Anticipated BOC Assembly Average Exposure Distribution . . . . . . . . . . . . 15 6.1 Palisades Cycle 5 Power Distribution 100 MWD/MT ARO, HFP. 26 6.2 Palisades Cycle 5 Power Distribution 4,000 MWD/MT, ARO, HFP . . . . . * * * . * . * * . * * . . . . . * * * * . 27 6.3 Palisades Cycle 5 Power Distribution 8,500 MWD/MT, ARO, HFP . . * . * * * * . * . . * * * . . . . * . . . . . . 28 6.4 Palisades Cycle 5 Core Average Power vs. Axial Pas it ion 0 MWD/MT . . . . . . . . . . . . . . . . . . * . . 29 6.5 Palisades cycle 5 Core Average Power vs. Axial Position 11,500 MWD/MT . . . . . . . . . . . . 29 6.6 Palisades Cycle 5 Power Distribution 75 MWD/MT, HFP, Group 4 Rods in 25% . . . . . . . . . . . . . . . . . . 30 6.7

  • Palisades Cycle 5 Power Distribution 10,000 MWD/MT, HFP, Group 4 Rods in 25% . . . . . . . . . . . . . . 31 6.8 Palisades Cycle 5 Critical Boron Concentration vs.

Exposure . . . . . . . . . . . . . . . . . * . . * . . . . 32

v XN-NF-81-34 Figure Page A-1 Palisades H-Fuel, F~ versus Peak Rod Burnup . . ..... 47 8-1 Palisades Gadolinia (Gd 2o3) Loading, Cycle 4 50 B-2 Palisades Cycle 4, INCA Power Distribution (Measured) versus PDQ Calculated Relative Assembly Power . . . . . . . 51 B-3 Palisades Cycle 4 Power Distribution Comparison Measured versus Calculated, 100% Power . . . . . . . . . * . .

  • 52 B-4 Palisades Cycle 4, INCA Power Distribution (Measured) versus PDQ Calculated . . . . . . . . . . . . . . . . . 53 B-5 Palisades Cycle 4 Gadolinia Assembly Power versus Exposure . . . . . . . . . . . . . . . . . . . . 54

1 XN-NF-81-34 I PALISADES NUCLEAR PLANT CYCLE 5 l

SAFETY ANALYSIS REPORT

1.0 INTRODUCTION

Exxon Nuclear Company (ENC) has performed a Safety Analysis of the Palisades Nuclear plant for the operation of Cycle 5. This consists of evaluating the proposed fuel loading configurations with regard to the.

power peaking, shutdown margin and transient and accident response. A preliminary analysis was reported in XN-NF-80-58 "Palisades Cycle 5 Fuel Cycle Design Analysis" December, 1980. The Startup and Operations Report to be issued later will confirm in more detail the safety related core parameters discussed in this report. The core in Cycle 5 will consist of sixty-eight (68) twice burned G assemblies, sixty-eight (68) once-burned H assemblies, and sixty-eight (68) fresh I assemblies. All assemblies in the core will have been manufactured by ENC.

The gadolinia program initiated in Cycle 3 will continue in Cycle

5. The use of gadolinia will be extended from an irradiation of thirty-two (32) fuel rods containing 1.0 w/o Gd 2o3 in Cycle 3 and thirty-two (32) fuel rods containing 4.0 w/o Gd 2o3 in Cycle 4 to an irradiation of ninety-six (96) fuel rods containing 4.0 w/o Gd 2o3 .

Operating history of Cycle 4 is given in Section 3. The Cycle 5 core is discussed in Section 4. Reload I fuel design is given in Section

5. Section 6 covers the nuclear design of reload I. Safety analysis is
    • discussed in Section 7.

. I I

I 2 XN-NF-81-34 2.0

SUMMARY

The characteristics of the fresh Batch I fuel and of the Cycle 5 relo'aded core result in conformance with existing Technical Specification Limits regarding shutdown margin, and moderator temperature coefficients.

This document provides the neutronic, thermal hydraulic, and control rod ejection analysis for the operation of Cycle 5. The ENC fuel assembly design for Batch I is similar to the Batch H extended burnup design(l, 5)_

Batch I contains 8 fuel assemblies made up of Batch I fuel rods and fuel rods left over from Batchs E and G. In addition Batch I fuel assemblies contain a lower enriched fuel pin in each corner than did the Batch H fuel assembly design. The ENC Plant Transient( 3), and ECCS( 4 , 5 ),

analyses for Palisades operations at 2,530 MWt are applicable to Cycle

5. A summary of the Cycle 5 plant parameters are compared to the core license limits in Table 2.1.

Since the extended burnup fuel, Batch H, will have pin exposures in excess of 30,000 MWD/MT by the end of Cycle 5 it will be necessary to implement the burnup dependent FQ limit reported in Appendix A( 2 ) into the Technical Specifications. Due to the replacement of D fuel assemblies, which contain 216 active fuel rods per assembly, with I fuel, which mostly contain 208 active fuel rods per assembly, the total number of rods in the core will be reduced by 1%. This corresponds to a 1% increase in the average linear heat generation rate (LHGR) for Cycle 5. While

3 XN-NF-81-34 the safety limits for allowable LHGR's in ENC assemblies with 208 active rods are unchanged, the reduction in the number of fuel rods in the Cycle 5 core necessitates a corresponding reduction in the allowable relative power peaking factors.

  • I l

4 XN-NF-81-34 Table 2.1 Palisades Cycle 5 Surrmary of Core Characteristics

(

\

Core Calculated License Parameters BOC EOC Limits Moderator 0 Tempera4ure Coefficient (1:!..p/ F x 10- )

HFP (no xenon) +0.20 -2.56 +0.5 to -3.5 Critical Boron Concentration HZP 1310 HFP 950 0 Shutdown Margin (%1:!..p) 2.40 2.33 >2.0 Power Peaking Factors FQ 2.35 1.84 <2.76*

FA 1.34 1.30 <1.43 R

Fl:!..H 1.62 1.53 <l.64 r

FT 1. 75 1. 59 <l. 77 r

  • Based on 208 active fuel rods per assembly and corresponds to the 15.28 kw/ft Technical Specification Limit on LHGR.

XN-NF-81-34 .

5 3.0 OPERATING HISTORY OF THE REFERENCE CYCLE The fourth power cycle has been chosen as the reference cycle with respect to Cycle 5 due to the similarity of the neutronic characteristics between t.he two eye l es. Cycle 4 operation began on May 24, 1980 and as of April 19, 1981, the core had accrued a cycle exposure of 7,548 MWD/MT. The plant availability and capacity factor is shown in Figure 3.1.

Appreciable quantities of gadolinia are currently being irradiated in the Cycle 4 core. Each of four (4) assemblies contain eight (8)

Gd 2o3-uo 2 rods with initial concentration of 4.0 w/o Gd 2o3 . The assemblies containing the gadolinia are located near the periphery of the core on the diagonal. The gadolinium bearing assemblies have performed as expected in the Cycle 4 core.

3.1 CYCLE 4 STARTUP TESTS The startup and low power physics tests performed at the beginning of life for Cycle 4 included boron end point measurements, isothermal temperature coefficient measurements, and rod bank worth measurements. The Palisades Cycle 4 Startup Report details the startup measurement~ and the comparisons to predictions.

  • 3.2 OPERATING STATUS OF CYCLE 4 (APRIL, 1981)

The Palisades core achieved a cycle burnup of 7,548 MWD/MTM on April 19, 1981, and has operated at or near full power for most of the cycle. Extept for the six week outage in November and December the

6 XN-NF-81-34 plant has had a good operating record. For operation through April, 1980, the cycle average capacity factor has been about 72%.

Comparison~ of predicted and measured boron concentrations and power distributions for Cycle 4 have been continously maintained.

Figure 3.2 displays the calculated and measured boron run down data for Cycle 4.

The calculated and measured power distribution at 6,950 MWO/MTM is shown in Figure 3.3, the standard deviation between predicted and measured values is less than 3%. On an assembly basis, comparisons of measured and predicted powers show deviations within 4.4%. Comparisons of calculated and measured assembly power during Cycle 4 for the assemblies containing 4 w/o Gd 2o3 are shown in more detail in Appendix B.

XN-NF-81-34 7

100 80

~

1.-

'O

.µ u 60 u

ttl c..

ttl u

40 1.-

0

+J u 1~

ttl l.J... .,...

+J ..c

.,... ttl u ,.....

20 <ti 0.

~,...

ttl u ct 0

May June July Aug Sept Oct Nov Dec Jan Feb Mar 1980 1981 Figure 3.1 Palisades Cycle 4 Operating History

1,000 0 Measured Calculated 800 (30 XTG Biased)

E 0..

0..

c:

0

  • r-

.µ ltl S-

.µ c: 600 QJ u

c:

0 u

c:

0 S-0 ca 400 200 ><

z I

.,,z I

co I

w

~

0 2 4 6 8 10 12 Cycle Exposure (GWD/MT)

Figure 3.2 Palisades Cycle 4 Critical Boron Concentration vs. Exposure, HFP, ARO

9

.883 1.048, *916 .905 1.052 1.065 .941 .91

.889 1.066 .922 .906 1.043 1.-048 .923 .94

- .6.7 -1.69 -.65 -.11 .86 1.62 I 1.95 -3.17 I

i

. 918 1.010 1.151 .941 .922 1.052 .92

.914 1.011 1.15J .935 .921 1.028 .94

.44 -.10 .64 .11, 2.33 -1.38 1.101 1.157* 1.226 .893 1.158 .75 1.094 -1.168 1.282 .864 1.175 . 77

.64 -.94 -4.37 3.36 "."1.45 -2.45

(

j

! .935 .954 1.064 1.058

! .914 .926 1.035 1.046 2.30 3.02 2.80 1.15.

e 1.284 *1.115 .719 PDQ at 7000 MWD/MT 1.310 1.102 .724 INCA at 6950 MWD/Ml

-1.98 1.18 -.69 {C-M) x 100 M

Figure. 3.3 Palisades Cycle 4, INCA Power Distribution (Measured) versus PDQ Calculated Relative Assembly Power, 100% Power

  • I l*

10 XN-NF-81-34 4.0

  • CYCLE 5 CORE DESCRIPTION The Palisades Cycle 5 core consists of 204 fuel assemblies, each having a 15 x 15 fuel rod array. The fuel rods consist of slightly enriched (in U~235) uo 2 pellets inserted in zircaloy tubes. The ENC assemblies have provisions for removable burnable absorber shims. Each ENC assembly contains ten zircaloy spacers with Inconel springs, nine of the. spacers are located within the active fuel region.

The planned Cycle 5 core loading arrangement is shown in Figure 4.1. The initial enrichments and burn~p distributions are displayed in Figure 4.2. The Cycle 5 core consists of twice burned Batch G and once burned Batch H assemblies scatter-loaded throughout the interior of the

  • core. The fresh Batch I fuel are located adjacent to and on the periphery of the core except for the eight assemblies containing boron carbide burnable absorber rods which are loaded in the interior. Batches G, H, and most of I (supplied by ENC) contain 208 fuel rods. The twelve (12)

Batch I assemblies containing gadolinia have 216 fuel rods.

Eight (8) Batch I assemblies contain removable boron carbide absorber shims. All eight of these assemblies are loaded in the interior.

The twelve (12) gadolinia bearing assemblies are located adjacent to the

    • peri~heral assemblies. The remaining 48 Batch I assemblies are located on the periphery and do not contain burnable absorber rods. Pertinent fuel assembly parameters are given in Table 4.1. In this table, the

l I

11 XN-NF-81-34

  • Batch I fuel is considered in four regions depending upon the assembly makeup.

A comparison of core parameters between Cycles 4 and 5 indicate a close resemblance between the two. As a consequence Cycle 4 is considered the reference cycle. Some of the main core parameters are summarized in Table 4.2.

L

Table 4.1 Fuel Assembly Design Parameters 12 Fuel Batch Gl G2 G3 Hl H2 HJ 11 .(Spare 13 14 Identification (Unshimmed) (B 4C) (Gd 2o3 ) (Unshimmed) (B 4C) {Gd 203) {Unshimmed)

  • Rods) {B 4C) {Gd 2o3)

Cycle Loaded 3 3 3 4 4 4 5 5 5 5 Initial Region Average Enrichment w/o U-235 3.00 3.00 3.00 3.27 3.27 3.24 3.26 3.23 3.26 3.24 No. of Assemblies 40 20 8 48 16 4 40 8 8 12 Pellet Density, % 94 94 94 94 94 94.0/94.75*** 94 94 94 94 Pellet to Clad Gap, {mil) 7.5 7.5 7.5 8.0 8.0 8.0 8.0 7.5/8.0 8.0 8.0 Fuel Stack Height 131.8 131. 8 131.8 131.8 131.8 131.8 131. 8 131.8 131.8 131.8 Region Average Burnup at BOC 5, MWD/MT 20,420 23,870 22,190 9,273 12,019 12,240 0 0 0 0

  1. to No. of Fuel Enrichments Per Assembly 3 3 3 2 2 3 3 7 3 3 ....

N No. of Fuel Rod and Enrichment (w/o) 60/2.52 60/2.52 60/2. 52 64/2.90 64/2.90 8/2.69 4/2.52 2/1. 87 4/2.52 4/2.52 4/3.01 4/3.01 4/3.01 144/3.43 144/3.43 64/2.90 60/2.90 8/2.40 60/2.90 8/2.52 144/3.20 144/3.20 144/3.20 136/3.43 144/3.43 9/2.81 144/3.43 56/2.90++

40/2.90 148/3.43 5/3.01 5/3.28 139/3.43 No. of Poison Rods Per Assembly 0 8 4* 0 8 8** 0 0 8 8*** ><

z I .

No. of Fuel Rods :z "Tl Per Assembly 208 208 208 208 208 208 208 208 208 216 I o::i I

Fixed 1 w/o Gd 203 in 3.20 w/o Fuel Rods w

    • Ffxed 4 w/o Gd~O~ in 2.69 w/o Fuel Rods ""'
      • Gadolfnia bear n rods only

++ Ffxed 4 w/0 Gd 2o3 in 2.52 w/o Fuel Rods

13 XN-NF-81-34 Table 4.2 Summary of Core Parameters Cycle 4 Cycle 5 Power Rating, MW Thermal 2,530 2,530 Expected Cycle Burnup (MWD/MT) 10,000 11,500 Beginning of Cycle Expected Core Average Burnup (MWD/MT) 11,000 10,500 End of Cycle Expected Core Average Burnup (MWD/MT) 21,000 22,000 Cycle Time (EFPH) 7,663 8,641 Moderator Temperat~re Coefficient

(~p/°F) x 10- (HFP, eq. xenon) +0.2 +0.3 Doppler Coeffic~ent

~p°F x 10- (HFP, BOC) -1. 25 -1.29 Delayed Neutron Fraction . .0061 .0061 Boron Concentration (ppm)

HFP equilibrium xenon 100 hr. sm. (at 100 MWD/MT) 743 945

f

  • I 14 XN-NF-81-34 .

N

\ **,--**

A B C D E f.f) G H Cf_). J N. ~ a* R ~T V (j}) X Z

! lir ,, 11 i I

.I-! . . - - 1021, ! !02 . !02 I !02 102 I !02~-.*-* j-1 HOO~I G111 J

'2 !I .  ! .

1!0271111!061 GDI* Li

-J__ 001 . :_ E-2 I G20 J-2 ,J N-2 I

HOig GD!* 106 11 102811 T-2 ~2 J

\  ! 2 i ,-:::'\

1 I*

I jro;i"f-!Hoo2 G1;-fGs4 J-1 J-8 I H~iWoo G-4 H-I' Gi9jf G26 Q-1 R-4 ! N-8. N-1 HOO'l~3oJ IL i

'1

--:)

5 I I 1

I

!031,1 !!032 ,GDI* ;1 IGOl IHOll Gl4 . HOO! HOO GlS ! H012 G02 ! GD!* I !033J:: !0341

_) J__j003_l), H-7 E-4 ~I K-S G-2_j,!J£2 I M-S 11 T-4 I Q-7 }l.£04 i . :,l-I 1 i; 5 c;-i  :

!ro;:!Fo"22

'I 1 .A-10 1G-8

!Go~lj~3o !Go;-1r-,.ro1 G~@1 r;-:~2 E-8 :K-10 I # K-41 i

M-4 **

HQ;4'(ffos i Ho25.{!o64\

M-10 T-8 1 R-8 1z-rn : I , *

\~:

8 jro3sjGDI*JllG47 IH014!l1~28 H6~*".; Gl3 Hoo jl H004 Gl6 H62*1G31; IH015jG49 :GDI*!I03.6! 8 i

~--- O~~ ~l~Q.JQ~LJ~ ,~-~1 E-~+/-7 K:;__~1. N-7 * ~S N:::J V-S ~1~~6 ! _j ~s>

10 J037 IH0131l!G38 IG2s llfo1s jG21 1 f'i G3o-

  • Ho1~11H23 j G3~ 'i G24 I 10161'1mz * ~4~ :1 (i~Q16 I I038! i~

1B-s , ID-7 IE-11 1-* 1G-10 H-8 H-2 1 Q-2 Q-8 R-10, ** I Ir-111 V-7 , lx-s !

  • 11

.._..\I0---39~\---G2--2 lef625ijH017 ~32 H026 \~j H029 H02L H0271 H036! IH0281. G33 I j Ro18\ Ro21! !G23 l I04Ui 11

! ,, ~ ~=4~~J.t.8 B-~Jt_11 -::..~8 H!_~S x-11~111 V-~1 X-7 z-.~.Jlt~- -~;;_':

. I 3 1rn41 IG43 :FHo38 Fo4o ~j4 rf 031 jl 1Ho37I *Ho4d 1\H04 71 H04sift H032I G35 !1Wo41 i H039i f !G44 1ID~ ~:

f

' 1 L.. B-141 IA.. 16 1~_-17 1 P--13 . -13. i B-16 H-19 . Q-19 1 X-161 Z-13 V-13 x~l7' z-161 ,x-14 1  ! _,

14 \I043 IH044 :11G59 r40 L_

I ~017 B-l9j .P:.__17 E-)]j I~* G-l!J). H-16 642 ii! GSS Hosq ~I HOSll GS6 '1 G4S I !018\ I 'G41 !G18 i :H052!1044 !

H-22, I Q-22! Q-161 !R-141 ** ;" T-13 V-17J:X-19 \ --~ ,~--

i

  • L..

~--- *,- - ~-- - - -- -~ --~ - ,---***-**C.

11 1Go1~ 'lfG511. o48J~37 1l H64~ G39 Ho49iGs2 :*i Gor*,ro46i~

16 lro4s 1 63!-"

1r1 Gs7 j Ho6 *11Ho661G63

. 007__ . ~-14 ' -19 g-13 -19 'J-17 K-2 IM-23 N-171 T19" N-131 1;V-19 1 Q-14 *~

y 1 I lro6s] ~033 ~06 .ll~o42 664.J 1\ ro19 I L ... --1 -14 .lG::l6J, E-~)_l~*

G6s l 1~~7 r220\ 11G68 I H043:J!G07 1Ho3s! 1!066 ! :

17 1 1 i K-20~201* :J_tM-14iT-16: _1R-l91Z-14 I.~ I

-?

Ii-*

~-;-lrn47\lfr048 Eoi="lf 009 to~lfG!lGsa 20 IHo~'f!Hos1I Gs;irfHOs41' G10lrfG0 K-19iG-2~1R-22M-19:.T-20Q-1710lO!

  • \GOI*: 1049/l~so 1

I i~

ilL_J ! ,"-; )

20 i

l J

j I

I IQSl ~060 G61 l_11Q.:23j-16 GS3 HOS91 IH0681 G66 G62 IH067J IQS2 I

!1~*-20 H-2~,lJ Q-23-, R-20iJ1N-l61 N-2ajJ . I l I I ,~~

~r


I1 --- - - --= - - . ~

. 22 l 1

. I0s3 r.!

1 jl'1 067 GDI*

t.nl

-YF 1':1 I Hoss G48 l1 Gso HOS8!-l GDI*1I068'l!10541 Asse:mbly ID ; . -;-~-

1 1E2~-:JJ-2~1 ! N-22 T-22! \ 012 ~I Cycl:e 4 1o~af1 ort'-

23-1I l

. . 1 I

oss 1

I I ros6L:ost*j i

ros8 IOS91 !060 i

I

- 23 A B © D E@ G H Q) J K. QM N @ Q R S T V @ X Z
  • 4.0 w/o gd 0
    • B c shimme6 ~ssembl_ies 4

Figure 4.1 Planned Cycle*5 Loarlinn Pattern

15 XN-NF-81-34 M Q T v x z H2 H2 Hl Gl Hl Hl Gl I1 Q-19 X-16* Z-13 V-13* X-17*** Z-16*** X-14 12,320 11,710 9 ,650* 21,490 10,620 7 ,710 21,640 0 H2 G2 Gl !3 Gl G2 . Hl I1 Q-22*** Q-16 R-14* T-13 V-17 X-19 11, 710 24,020 18,570 0 21,580 23,340 7,020 0 Hl Gl H3 Gl- Hl G2 14 I1 M-23 N-17***

18,550.

T-19 N-13* . V-19 Q-14 9,650 12,240 18,880 10,930 24,310 0 0 Gl 13 Gl H2 G3 Hl 12 M-20*** M-14*** T-16** R-16*** Z-14*

21,490 0 18,800 12,330 22,190 9,700 Hl Gl Hl G3 14 I1 11 19 R-22* M-19 T-20 Q-17*

10 ,620 21,560 10,930 22,180 0 0 0 Hl G2 G2 Hl 11 Q-23* R-20 N-16 N-23***

7 ,410 23,340 24,270 9,700 Gl Hl 14 12 11 Fuel Type N-22 T-22 Cycle 4 Location 21,640 7,020 0 0 0 BOC5 Exposure (MWD/MT)

  • .2 3 11 11 11 0 0 0 Region Initial Average Enrichment

--*- G 3.00 Rotations Counter-clock wise H 3.27 l 3.25

  • ~00
    • 180° Figure 4.2 Cycle 5 Loading Pattern and Anticipated.
      • 270° BOC Assembly Average Exposure Distribution

-.~~*--=- - -** *--....

16 *xN-NF-81-34

. J

\

5.0 RELOAD I FUEL ASSEMBLY DESIGN A description of the Exxon Nuclear supplied fuel design for the previous reload batches is contained in References 1, 7, 8 and 9. From

{

I a mechanical standpoint the reload batch I design is essentially identical

' \

to the reload batch H except for the eight (8) "spare rods" assemblies and the low enriched fuel pins in the corners of all the assemblies (2.52 w/o U-235). The "spare rods" assemblies contain fuel rods of the reload E/G design as well as that of the reload I design. These assemblies will, therefore, be required to operate within the design burnup envelope of the reload E/G design. The exposure limitation being applied to the "spare rod" assemblies will not restrict the operation of reload Batch I since the anticipated assembly exposure for the spare rod assemblies is below the batch I average which is less than the maximum assembly exposure

  • permitted for batch E. A comparison of the design parameters is presented in Table 5.1.

The gadolinia demonstration program witl continue in Cycle 5. The demonstration program began with the irradiation of thirty-two fuel rods containing 1 w/o Gd 2o3 in Cycle 3 and was followed by the irradiation of thirty two fuel rods containing 4 w/o Gd 2o3 in Cycle 4. In the latter assemblies, 2.69 w/o uo 2 fuel rods containing 4 w/o gadolinia replaced eight standard 3.43 w/o uo 2 fuel rods.

17 XN-NF-81-34 In Cycle 5 ninety-six (96) fuel rods will contain eight (8) 4 w/o Gd 2o3 fuel rods evenly distributed in twelve (12) assemblies. These gadolinia bearing fuel rods will be mixed with uranium enriched to 2.52 w/o. For these assemblies the guide tubes have been replaced with the high enrichment (3.43 w/o) fuel rods and the eight gadolinia rods replace four (4) 2.90 w/o and four (4) 3.43 w/o fuel rods.

Thermal margins have been calculated for the Cycle 5 core including the gadolinia demonstration program and it is concluded that the program will have no impact on the safety and performance of the Cycle 5 core and Reload I.

18 XN-NF-81-34 Table 5.1 Fuel Design Sullillary Reload* Design E/G H I Number of Assemblies 68 68 68 Initial Average Enrichment (%) 3.00 3.27 3.25 Pellet Density (% TD) 94.0 94.0/94.75* 94.0

' Pellet Clad Gap (in) . 0.0075 0.0080 0.0080 Fi 11 Gas Pressure (psia He) 300 321 321 Wall Thickness (in) .0285 .0295 .0295 Number of Assemblies with B4c-A1 2o3 Burnable Poison 20 16 8 B4c-A1 2o3 Rods/Assembly 8 8 8 Poison Loading, gm BlO/in 0.0204 0.0204 0.0204 Number of Assemblies with Gd 2o3 Burnable Poison 8 4 12 Urania-Gadolinia Rods/Assembly *4 8 8 Wt. %Gd 2o3 1.00 4.0 4.0 BOC 5 Batch Average Exposure (MWD/MT) 21,640 10 ,090 0

  • Gadolinia bearing rods only

19 XN-NF-81-34 6.0 NUCLEAR DESIGN i

The neutronic characteristics of the projected Cycle 5 core consis-ting of three regions of ENC fuel (Batches G, H, and I) are quite similar to those of the Cycle 4 core (see Section 4.0). The nuclear design bases for the Cycle 5 core a re as fo 11 ows:

1. The design shall permit operation of the Cycle 5 core at full power within the constraints established for the Palisades reactor.
2. The length of Cycle 5 shall be determined on the basis of a 2,530 MWt power rating and on an assumed Cycle 4 energy produc-tion equivalent to 10,000 MWD/MT.
3. The Region I assembly average enrichment shall be 3.25 w/o which is slightly less than the Batch H assembly average enrichment of 3.27 w/o. The Batch I Reload shall consist of 68 l assemblies (one-third core reload).
4. The Cycle 5 loading pattern shall be optimized to achieve non-1imiting power distributions and control rod reactivity worths according to the following constraints:
a. The peak LHGR shall not exceed 15.28 kw/ft including uncertainties in any single fuel rod throughout the cycle under nominal steady state full power operating conditions;
b. The peak assembly power shall not .exceed 17.78 MW in assemblies containing 208 fuel rods which corresponds to a peaking factor of 1.43 and 18.08 MW in assemblies

20 XN-NF-81-34 containing 216 fuel rods which corresponds to a peaking factor {F~) of 1. 46 in the corresponding assembly throughout the cycle.

c. The peak rod power for an internal rod shall not exceed 97.90 KW which corresponds to a peaking factor (F~~) of 1.64 for assemblies with 208 fuel rods. The peak rod power for an internal rod shall not exceed 95.86 KW which corresponds to. a peaking factor (F~H) of 1.67 for assemblies with 216 fuel rods throughout the cycle.
d. The peak rod power for all rods in batch G shall not exceed 104.3 KW which corresponds to a peaking factor (F~) of 1.75. The peak rod power for all rods in batches H and I shall not exceed 105.5 KW, which corresponds to a peaking factor (F~) of 1.77 and 1.84 for assemblies with 208 and 216 fuel rods, respectively. These peaking factor limits apply throughout the cycle under nominal full power steady state conditions .

. e. The N-1 scram worth shall not violate the HZP and HFP, BOC and EOC shutdown requirements;

5. The Cycle 5 core shall have a negative power coefficient.

The neutronic design methods are described in References 11, 12, and 13

  • 21 XN-NF-81-34
  • In order to simplify the Technical Specifications it is reco111T1ended that the peaking factor limits be established at 1.43, 1.64 and 1.77 for F~, F~H, and F~, respectively~ These values represent the lower peaking factor of the different assembly.types. Calculations show that by monitoring the core to the minimum allowable peaking factors, Cycle 5 0

will be able to operate at full power for the duration of the cycle.

6.1 PHYSICS CHARACTtRISTICS 6.1.1 Power Distribution Considerations Representative radial and axial power maps for the planned core loading are shown in Figure 6.1 through 6.5. Figures 6.1, 6.2 and 6.3 show the radial power maps at 100 MWD/MT, 4000 MWD/MT, and*

8,500 MWD/MT, respectively. The highest assembly power factor calculated for Cycle *5 is 1.30 and occurs at 100 MWD/MT. The corresponding core average axial profiles are shown in Figures 6.4 and 6.5. These power distributions were obtained from a three-dimensional analysis accounting for feedbacks including moderator density and Doppler.

The radial maps are representative of a hot full power, equi 1i bri um xenon core *configuration. The Cycle 5 1oadi ng pattern was designed to minimize F~H and F~. The largest calculated F~ was 2.03 diminishing to 1.59 at EOC conditions. The calculated xenon free; full power value of F~ is 2.06.

The axial power profiles are core average distribu-tions. For Cycle 5 the peak axial is predicted to remain at or below

22 XN-NF-81-34 i

. \

i.

' L26. This value is typical for a reload core like the Cycle 5 design

(

. \

and is nearly identical to the Cycle 4 core axial of 1.24.

Figure 6.6 and 6.7 show the radial power maps with the Group 4 regulating rods inserted to the power dependent inserti-0n limit at HFP (25% insertion) for BOL and EOL, respectively. Assembly powers and F~ are similar to the all rods out values shown in Figures 6.1 and 6.3.

Additional neutronic characteristics of the Cycle 5 core are compared with the Cycle 4 core in Table 6.1 for both BOC and EOC conditions. The Cycle 5 projected critical boron concentration as a function of cycle burnup is shown in Figure 6.8.

6.1. 2 Control Rod Reactivity Regui rements

  • Detailed calculations of shtudown margins for Cycle 5 are compared with the data for Cycle 4 in Table 6.2. For Palisades the minimum shutdown requirement at HFP and HZP is 2% l:lp assuming the most reactive control rod stuck out. A minimum excess shutdown margin of

.33% l:lp is indicated for Cycle 5 at HFP.

6.1.3 Moderator Temperature Coefficient Considerations The reference Cycle 5 design calculations indicate a*

HZP critical boron concentration at SOCS of 1,310 ppm. This value is 150 ppm higher than the measured HZP BOC 4 critical boron concentration where the moderator temperature coefficient was determined to be +0.09 x 10- 4 l:lp/°F. The moderator temperature coefficients at HZP and HFP for

23 XN-NF-81-34 BOC5 and HFP for EOC5 are shown in Table 6.L The BOC values fall well within the safety analysis limits of +0.5 x 10- 4 /J.p/°F ~ MTC ~ -3.5 x 10-4 /J.p/oF.

6.2 NUCLEAR DESIGN METHODOLOGY The methods used in the Cycle 5 core analyses are described in References 10, 11, and 12. In summary, the reference neutron*ic design analysis of the reload core was performed using a combination of the PDQ7(l 3)/HARMONY(l 4 ) depletion system and the XTG(l 5 ) reactor simulator system along with the XPOSE(l 6 ) and XPIN(ll) pin cell codes. For each model, the input isotopics data were based on quarter core calculations performed through Cycle 4 using the respective models. The fuel shuffling between cycles was accounted for in the calculations.

With the XTG reactor model, including 3-D effects such as moderator density and doppler feedbacks, values of Fq, Fxy' and Fz were.

studies. The calculated thermal-hydraulic feedback and axial exposure distribution effects on power shapes, rod worths, and cycle lifeti.me are explicitly included in the analysis.

In the PDQ model detailed pin-by-pin depletion analyses are performed. Local variations in power and isotopic distributions are explicitly calculated.

The in-core-measurement/calculation constants are obtained from the quarter core pin-by-pin calculational results.

.I 24 XN-NF-81-34 Table 6.1 Calculated Neutronics Characteristics of Cycle 5 Compared with Cycle 4 f

Cycle 4 Cycle 5 I Parameters BOC (2,530 MWt)

EOC (2,530 MWt)

BOC (2,530 MWt)

EOC (2,530 MWt)

Moderator Te~ij* Co9fficient -.63 -2.56 -.45 -2.56 at HFP (x 10 6p/ F) (ppm) (820) (0) (950) (0)

Moderator Te~ij* Cogfficient +.04 +.30 at HZP (x 10 6p/ f) (ppm) (1,200) (1,310)

Doppler Defect (% 6p) -0.76 -.62 -0.76 -0.62 Power Defect (Doppler + Moderator) -1.01 -1.49 -LOO -1.60 Delayed Neutron Fraction 0.0061 0.0051 0.0061 0.0052 Prompt Neutron Lifetime

( µ sec). 23.0 27.4 22.3 24.7 Inverse Boron Worth (ppm/% 6p) 102 92 95 82 Ejected Rod Worths 100% Power <0.20 <0.20 0.15 0.20 0% Power <0.90 <0.90 1.02 0.94 Peaking Factors Radial 1.26 1.27 1.30 1.25 Axial

( % 6p) 0.48 0.46 0.40 0.33

Total Minus Stuck Rod 4.69 4.69 5.22 5.22 4.59 4.59 5.28 5.28 Uncertainty (10%) 0.47 0.47 0.52 0.52 0.46 0.46 0.53 0.53 Net Shutdown Rod Worth (1) 4.22 4.22 4.70 4.70 4.13 4.13 4.75 4.75 Reactivit.}'. Insertion (% ~~)

Doppler Defect 0 0.76 0 0.62 0 0. 76 0 0.62 N l11 Moderator Temperature Defect 0 0.25 0 0.87 0 0.24 0 0.98 Moderator Void Defect 0 0.10 0 0.10 0 0.10 0 0.10 Axial Flux Redistribution 0 0.50 0 0.50 0 0.50 0 0.50 Required Shutdown Margin 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 Total Reactivity Allowances (2) 2.00 3.61 2.00 4.09 2.00 3.60 2.00 4.20 x

z Available for Maneuvering (1-2) 2.24 .061 2.70 0.61 2.13 .053 2.75 0.55 I z

I POil Rod Insertion 1.47 0.13 1.69 0.15 1.84 0.13 *2.14 0.22 co

~

I w

Excess Margin % ~P o. 77 0.48 1.01 0.46 0.60 0.40 0.56 0.33 -l!>o

I i

i . XN-NF-81-34 26 I

'* H*

1.18 M

H*

1.18 i~

H 1.28 Q

G I{

1.03 H

1.27 T

H 1.30 v

G

. 93 x

I

.88 z

l .'I H* G* G I* G G** H I 1.18 .91 1.01 1.24 .94 .88 1.17 .90 H G HG G H G* lG I 1.28 1.01 1.11. .95 1.04 .83 1.07 .67

~*-*

l / G I* G H* GG H IS 1.03 1.24 .95 .98 . 86 1.06 .92 H G H GG IG I I

1. 27 .94 1.04 .86 1.04 .93 .56 H G* G* H I Assembly Type 1.30 .88 .83 1.06 .93 Relative Assembly Power (PD Q)

G H IG IS I

.93 1.17 1.07 .92 .57

  • - - FQ = 2.03 (M-14)

I A (M-20)

. ' I I FR = 1.30

.89 .90 .67 F~H= 1. 54 (M-20)

F~ = 1. 67 (M-14)

  • B4c Shimmed G Gadolinia Shimmed S Sp9-re Rod Figure 6.1 Palisades Cycle 5 Power Distribution 100 MWD/MT ARO, HFP

, .. *.~4'f,.*,-* - ........ . .. .. . ~ - - *- ~*-* ----- ... ~ - . - .. *.----- .,_..,_.._.....,_,.. __._.._ . .:....... * . _. - *' ..r ./ .,.,,.. ...

I.

27 XN-NF-81-34 i*1 r: () R T v x z H* H* H G H H G I 13 1.14 1.12 1.18 .96 1.16 1.19 .90 .88 H* G* G I* G G* H I 1.12 .88 .97 1.20 .91 .88 1.16 .. 91 Hi H G HG G

. H G* iG I 1.18 .97 1.08 .95 1.05 .87 1.14 . 72

! I G I* G H* GG H IS

.96 1.20 .96 1.02 .92 1.12 .99 11 1 H G H GG IG I I 1.16 .91 1.05 .92 1.17 1.02 .63 H G* G* H I Assembly Type 1.19 .88 .87 1.12 1.02 Relative Assembly Power (PD Q)

G H IG IS I

.90 1.16 1.10 .99 .63

- FQ = 1.67 (M-14)

FA = 1. 20 ( R-14)

I I I I R F~H = 1.41

.88

.91

.. ____ .72 FRT -- 1.49

{V-19)

(M-14)

  • - B C Shimmed G - G~dolinia Shimmed S - Spare Rods Figure 6.2 Palisades Cycle 5 Power Distribution 4,000 MWD/MT, ARO, HFP

J 2a* XN-NF-al-34 I

M N () T v x z H* H* H G H H G I I .3 1.10 1.12 .93 1.10 1.13 .88 .87' 1.12 H* G* G I* G G* . H I 1.10 .87 .95 1.18 .90 .88 1.14 . 91 H G HG G H G* IG I 1.12 .95 1.06 .96 1.06 .90 1.22 .76 1; G I* G H* GG H IS

.93 1.18 .96 1.05 .96 1.14 1.02 1:J H G H GG IG I I 1.10 .90 1.06 .96 1.26 1.06 .67

~* H G* G* H I Assembly Type 1.13 .88 .90 1.14 1.06 Relative Assembly Power (PD Q) l..,. G H IG IS I

.88 1.14 1.22 1.02 .67 I I I FQ = 1.59 ( T-19)

.87 .91 .76 FA ( T-19)

R = 1.26 F~H= 1.42 (V-19)

  • B C Shimmed FRT = 1.51 (T-19)

G G~dolinia Shimmed S Spare Rods Figure 6.3 Palisades Cycle 5 Power Distribution 8,500 MWD/MT, ARO, HFP

  • ~*1:-* **-

- - -- - ** - - - * - - ** ,...*. - "'* ......

  • 4 * "* .,.. *;r. ~ * * , ,.,.._ ...

29 XN-NF-81-34 I *

  • t *-*
  • I I

i i' - ~*  ;' - i

~ ' ' -

QJ 3:

1.2 ' --1-- -- -1"----*;* - --..,;

0

+:1* . 1* : i i 0..  !

I QJ

+.>

1.0 - l- - - -~-1--- :...... j I -

'° QJ I

~ 0.8 i*- ... ,

I

'°x i

I f -

.:x;: 0.6 -

--- -:---4--

I QJ .  !

I C'l ----~.--..i. __ .

~

QJ 0.4 -------* -*-

> - ! - I *-

.:x;:

- -. - _. - *f-*--.'. ___L.__~--  !

aJ s... i* '. I . .

u 0 0.2 -- ----* --* --*-* - --- *- - ---4----L-...,-* -*

! l

- - * * - T.. *---L-_.:..; ____ ._.

0.0  ; l !

0 1 2 3 4 5 6 7 8 9 10 11 Top Bottom Axial Position (ft)

Figure6.4 Palisades Cycle 5 Core Average Power vs. Axial Position 0 MWD/MT

  • 1*** --1 I

I j

i

~

QJ 3

1.2 +---+-t'

i -
!

.+_J 0

0..

QJ

> 1.0 ~

-~~--~~~~r-* .. ***-- .... ------*** *-- --- ----+--------+-*-

,.... i - -L.- .... I QJ ,_

~

0.8 . ,

!I i

x 0.6

.:x;: - ---- **-*---...,---** i I ....... ***-***

I QJ C'l

-'° I

~

QJ 0.4  : - .. 1..

.:x;:

QJ

' i

~

0 0.2 '

  • - -----*--- -t**-*-- -

i u i

-.I.- -. .-.

0.0 0 1 2 3 4 5 6 7 8 9 10 11 Top Bottom Axial Position (ft)

Figure 6.5 Palisades Cycle 5 Cor.e Average Power vs. Axi a 1 Po~i tion 11, 500 ri~JD/t*iT* *

{

I . 30 XN-NF-81-34

  • I .~

H*

L19 H*

1.16 N

H

.Q 1.23 G

.99 H

1.24 T

H 1.28 v

G

.94 x

I

.99 z

14 H* G* G I* G G*. H I 1.16 .89 .97 1.22 .91 .88 1.19 .99 1 !.> H G HG G H G* I.G I 1.23 . 98 1.04 .90 1.00 .81 1.13 .73

~I I

' G I* G H* GG H IS

. 99 1.22 .91 .90 .80 1.03 1.00 H G H GG IG I I 1.24 .91 1.00 .80 1.04 .97 .61 H G* G* H I Assembly Type 1.28 .88 .82 1.04 .97 Relative Assembly Power (XT G)

G H IG IS I

.94 1.19 1.13 1.00 .61 I T J. I FQ -- 2.08 (M-14)

.99 .99 .73 FA = 1.28 (M-20)

R F~H= 1. 52 (M-20)

  • B C Shimmed G G&dolinia Shimmed FTR = 1. 64 (M-14)

S Spare Rods Figure 6.6 Palisades Cycle 5 Power Distribution 75 MWD/MT, HFP, Group 4 Rods in 25%

.....:"t.... - * - - *

  • 31 XN-NF-81-34 M Q R T v x z r .

l 1,)., H* H* H G H H G I 1.20 1.17 1.18 .98 1.14 1.16 .90 .91 l

14 H* G*. G I* G G*. H I 1.17 .94 1.00 1.22 .93 .89 1.13 .92

, r:

',.J H G HG G H G* IG I 1.18 1.00 1.07 .96 1.04 .87 1.21 .74 I.' G I* G H* GG H IS ..

. 98 1.22 .96 .98 .87 1.04 .99 H G H GG IG I I 1.14 .93 1.04 .87 1.16 .96 .63 H G* G* H I Assembly Type 1.16 .89 .87 1.04 .96 Relative Assembly Power (XT G)

G H IG IS I

.90 1.13 1.21 .99 .63

  • * - - - *- FQ = 1.91 (X.-16)

A I I I FR = 1.22 ( R-14)

.91 .92 .74 F~H= 1 ~44 (X-16)

F~

I.

=* 1.48 (X-16)

  • B C Shimmed G G&dolinia Shimmed S Spare Rods Figure 6. 7 Palisades Cycle 5 Power Distribution
    • 10,000 MWD/MT, HFP, Group 4 Rods in 25%

--~---. -~ --* *,y - - * - * - - - --:---* **

1200 llOO 1000 900 E

0.

0. 800

.C:

.µ 0

700 ltl

.µ s...

c:

QJ u

c: 600 w 0 N u

c: i I

0 s... 500 I 0 I ca 400 .l.

300 I

I ' ><

200 ..! i--. ~--- -: . ~ *- - -* - . -.. i:. - . -- . *. - . --- __ ,. -. 2 I

z "Tl i f- I 100 I ....... .
' . *1*.* i

. I'* ... - .- *- - . -- ---- --- *- .!..: .. - . . ---.!-

I i ....co I

I  ; lI w

~

i 0

1 2 4 5 6 7 8 9 10 11 12 13 Cycle Exposure (GWD/MT)

Figure 6. 8 Palisades Cycle 5 Critical Boron Concentration vs. Exposure

33 XN-NF-81-34 7.0 SAFETY ANALYSIS I. _Safety ana1ysis considerations for Cycle 5 is the normal cyc1e specific analysis. This analysis is described in each of the fo11owing subsections.

'I 7.1 THERMAL HYDRAULIC ANALYSIS l

The ENC Re1oad I fiie1 is dasfgned to be compatibl!'.i! with the Palisades *Reactor core aod with the existing fuel. The thermal hydrau1 ic design criteria for ENC re1oad fue1 at Pa1isades are:

o The maximum fuel temperature at 115% overpower shall not exceed the fuel me1ting temperature.

0 The minimum DNBR shall be greater than or equa1 to 1.30

  • at 115% of rated power based on the W-3 correlation (or an accepted equivalent) p1us correction factors which have been accepted by the NRC for the purpose of licensing the fuel design described herein.

o The c1adding temperature at nominal operating conditions (based on crud-free surface) sha11 be 1ess than:

850°F interna1 surface 675°F externa1 surface 750°F vo1ume average (1oca1) o The fue1 assemblies must be thermally and hydraulically compatib1e with the existing fue1 and the reactor core during the.design 1ife of the fue1 *

( 34 XN-NF-81-34

~

{

ENC reload fuel in the Palisades Cycle 5 core is calculated to I satisfy the thermal hydraulic design criteria for the following limits on assembly and interior rod power levels:

l o the maximum assembly average linear heat generation rate I

i is equal to or less than 7.78 kw/ft. for assemblies with I.

208 fuel rods~ and is equal to or less than 7.62 kw/ft.

for assemblies with 216 fuel rods.

! 0 The maximum interior rod linear heat generation. rate does i.

I not exceed 8.91 kw/ft. for assemblies with 208 fuel rods, and does not exceed 8.73 kw/ft. for assemblies with 216 fuel rods.

The line*ar heat generation rate limits above for Batch Hand I assemblies with 208 active fuel rods and the associated rod surface heat fluxes are unchanged from the previous analysis(S)_ The results above for assemblies with ~16 active fuel rods are new limits from thermal hydraulic analyses for Batch I assemblies with 216 active fuel rods.

Cycle 5 peaking factors which correspond to the above limits for assemblies with 208 *active rods and 216 active rods are given in Table 7.1. The relative peaking limits are slightly reduced from those in Cycle 4 reflecting the reduction in the total number of active fuel rods in the Cycle 5 core versus the Cycle 4 core.

The thermal hydraulic analysis for the Palisades Cycle 5 was performed in a manner consistent with applicable thermal margin analyses for the Palisades plant at 2530 MWt(S,G,l 9)_ The thermal hydraulic

  • . i

I 35 XN-NF-81-34 l

design conditions for this analysis are shown in Table 7.1. It is concluded that the performance of Palisades Cycle 5 falls within the thermal hydraulic design criteria .. The thermal hydraulic acceptability of ENC reload fuel for Palisades Cycle 5 operation is thus confirmed.

7.2 PLANT .TRANSIENT ANALYSIS The transient events listed in Table 7.2 were analyzed for Cycle 5 operation. Predicted Cycle 5 core kinetics parameters considered in the analysis appear in Table 7.3, and are identical to those used in the reference stretch power analysis( 3 ). The conclusion of this analysis is that MDNBR values for Cl~ss II and III events initiated during Cycle 5 operation will remain greater than the accepted minimum value of 1.3 .

.This analysis also concludes that the tlass IV locked rotor and main steam line break accidents will result in MDNBR values equal to or greater.than those reported in th~ reference analysis. Predicted operating thermal margin for Cycle 5 is therefore judged adequate to maintain the integrity of the fuel cladding within acceptable limits.

7.3 ECCS ANALYSIS Previous LOCA/ECCS analyses( 2,l 9 ) for Palisades E/G and Hfuel were.made with a maximum linear heat generation rate of 15.28 kw/ft at 102% of full core power (1.02 x 2530 MWt). This corresponds to an allowable assembly radial peaking limit of 1.45 and an FQ limit of 2.76.

These limits remain ~pplicable to the Palisades reload fuel design I and

(

36 XN-NF-81-34 to the eight (8) fuel assemblies built from spare rods left from fabricating reloads G and H. These assemblies will be loaded into the reactor in Cycle 5.

7.3.1 Reload I ECCS Limits Palisades reload designs .H and I have the same mechanical design but have slightly different neutronic designs. The neutronic bundle fuel design for reload I incorporated four (4) low enrichment rods in corner locations to reduce the local assembly peaking observed in the reload H design. Since the mechanical designs are identical,.

the hydraulic flow behavior for the I assemblies will be the same as that calculated in the LOCA analysis for the H assemblies. The ECCS limits astablished for previous reloads are therefore conservatively applicable to reload I. The reduction in local peaking for reload I will result in greater margin to ECCS limits.

7.3.2 Spare Rods Assembly ECCS Limits The fuel rods for reload I have a 2 mil larger clad outer diameter than fuel rods for reload G. In the ECCS analysis, the larger clad outer diameter results in improved r~flood rates.and in a

. larger surface heat transfer with reduced PCT's. The ECCS limits established for previous reloads are therefore conservatively applicable to the reload fuel assemblies fabricated from G and I fuel rods. For the bundles fabricated from G and I rods, the maximum calculated bundle local peaking (1.205) is lower than that used in the G/H (1.22) analyses.

I I \

37 XN-NF-81-34 Therefore the combined effect of lower bundle peaking, higher reflood rates and larger surface heat transfer areas will result in improved margin to ECCS limits for the spare rod bundles relative to that for the G bundle.

iI .

\

7.4 ROD EJECTION ANALYSIS A Control Rod Ejection Accident is defined as the mechanical failure of a control rod mechanism pressure housing, resulting in the ejection of a Rod Cluster. Control Assembly (RCCA) and drive shaft. The consequence of this mechanical failure is a rapid reactivity insertion together with an adverse core power distribution, possibly leading to localized fuel rod damage.

The rod ejection accident has been evaluated with the procedures developed in the ENC Generic Rod Ejection Analysis( 20)_ The ejected rod worths and hot pellet peaking factors were calculated using the XTG code. No credit was taken for the power flattening effects of Doppler or moderator feedback in the calculation of ejected rod worths or peaking factors. The calculations made for Cycle 5 using XTG were two-dimensional.

The pellet energy deposition resulting from an ejected rod was evaluated explicitly for BOC and found to be 164 cal/gm at HFP and 143 cal/gm at HZP. The results for EOC conditions were found to be 173 cal/gm at HFP and 126 cal/gm at HZP. The rod ejection accident was found to result in energy deposition of less than 280 cal/gm as required by Regulatory Guide 1.77. The significant parameters for the analysis, along with the results are summarized in Table 7.4

  • 38 *xN-NF-81-34 Table 7.1 Thermal Hydraulic Design Conditions Reactor Condttions Design Nominal Core Power (MWt) 2910 2530 Total reactor flow rate (Mlb/hr) 121.7 121. 7 Active core flow rate (Mlb/hr) 114.4 114.4 Coolant inlet temperature (OF) 542.5 537.5 Core pressure (psia) 2010 2060 Thermal Hydraulic Limits on Relative 208 Rod 216 Rod Power Factors Assemblies Assemblies
    • Assembly Radial Factor, FR Pin Peaking Factor (interior rod), FRxF t Pin Peaking Factor (narrow gap edge rod), FRxFt 1.43 1.64 1.75+

1.46 1.67

1. 74+

Pin Peaking Factor (wide gap edge rod), FRxRt 1.88+* 1.84+

Engineering Factor 1.03 1.03 LOCA/ECCS Limits on Relative Power Factors A11 Assemblies Assembly radial factor, FR 1.45 Pin Peaking Factor (all rods), FRxRt 1. 77 Total Peaking 2.76

  • The corresponding peaking factor for G fuel is 1.75.

+ At these peaking factors the interior rod remains limiting.

39 XN-NF-81-34 Table 7.2 Transient Events Considered in the Palisades Cycle 5 Plant Transient* Analysis

1. Uncontrolled Rod Withdrawals At 102% power (1.0 x 10- 5 ::._ 6p/s ::._ 1.4 x 10- 4 )

0 At 52% power (&.O x* 10- 5 ::._ !p/s ::._ 6.0 x 10- 4 )

2. Control Rod Drop
3. Four Pump Coastdown
4. Locked Rotor
5. Reduction in Feedwater Enthalpy
6. Increased Feedwater Flow (@ 52% power)
7. Excessive Load (from 102% power and 52% power)
8. Loss of Load
9. Loss of Feedwater
10. Steam Line Break (from 102% power and hot standby)
11. Single Rod Withdrawal

'{ 40 XN-NF-81-34

  • l Table 7.3 Important Core Kinetics Parameters Used in the Palisades Cycle 5 Plant Transient Analysis I

I t EOC BOC Moderator Co~fficient (t:.p/ F x 10 ) +0.5 -3. 50 Doppl 0r Coef5icient (t:.p/ F x 10 ) -1.09 -1. 38 Delayed Neutron Fraction, % 0.75 0.45 Net* Rod Worth (% t:.p)** -2.90 -2.90

  • Total rod worth minus stuck rod worth.
    • 2.0% at hot standby.

l 41 XN-NF-81-34 Table 7.4 Palisades Rod Ejection Accident BOC EOC HFP HZP HFP HZP F~ After Ejection 2.76 13.4 3.02 12.1 Ejected Rod Worth (%1'.lp) .15 1.02 .20 . .94 Doppler Coeff~c~ent

(%1'.lp x 10- I F) -1.29 -1.55 -1.49 -1. 73 Delayed Neutron Fraction .0061 .0061 .0052 .0052 Energy Deposition (cal/gm) 164 143 173 126

42 XN-NF-81-34 REFERENCES i

\

1. XN-75-29, "Generic Fuel Design for 15 x 15 Reload Assemblies for I Palisades, C. A. Brown, November 25, 1975.
2. XN-NF-81-34, Appendix A "ECCS Exposure Sensitivity Study for the Palisades Reload H Design", May 1981.
3. XN-NF-77-18, "Plant Transient Analysis of the Palisades* Reactor for Operation at 2,530 MWt", July, 1977.
4. XN-NF-77-24, "LOCA Analysis for Palisades *at 2,530 MWt Using the ENC WREM-II PWR ECCS Evaluation Model," July, 1977.
5. XN-NF-80-18, "ECCS and Thermal-Hydraulic Analysis for the Palisades Reload H Design", April, 1980.
6. XN-NF-77-22, "Steady State Thermal Hydraulic and Neutronics Analysis of the Palisades Reactor for Operation at 2,530 MWt", July, 1977.
7. XN-NF-77-59, "Palisades Cycle 3 Reload Fuel Licensing Data Submittal",

C. A. Brown, et al., December, 1972.

8. XN-NF-79-48, "Palisades Cycle 4 Reload Fuel Licensing Data Submittal",

L.A. Nielsen, et al., June, 1979.

9. XN-NF-79-48, Revision 1, "Palisades Cycle 4 Reload Fuel Licensing Data Submittal", L.A. Nielsen, et al.,September, 1979.
10. XN-75-27, "Exxon Nuclear Neutronic Design Methods for Pressurized Water Reactors", June, 1975.
11. Supplement 1 to Reference 10.
12. Supplement 2 to Reference 10.
13. WAPD-TM-678, 11 PDQ7 Reference Manual", W.R. Caldwell, January, 1965.
14. WAPD-TM-678, "HARMONY: System for Nuclear Reactor Depletion Computation",

R. J. Breen, et al., January, 1975.

15. XN-CC-28, Revision 5, 11 XTG - A Two-Group Three Dimensional Reactor Simulator Utilizing Coarse Mesh Spacing", Exxon Nuclear Company, R. B. Stout, July, 1979.

43 XN-NF-81-34

16. XN-CC-21, Revision 2, XPOSE - The Exxon Nuclear Revised LEOPARD",

Exxon Nuclear Company, April, 1975.

17. XN-NF-CC-26, "XPIN The Exxon Nuclear Revised Hambur Users Manual",

W. W. Porath, et al., December, 1975.

i8. Technical Specifications contained in Provisional Operating License DPR-20, Docket 50-255 issued to Consumers Power Company for the Palisades Nuclear Plant, October, 1972.

19. XN-NF-78-16, 11 Analysis of Axial Power Distribution Limits for the Palisades Nuclear Reactor at 2530 MWt", _June, 1978.
20. XN-NF-78-44, 11 A Generic Analysis of the Control Rod Ejection Transient for Pressurized Water Reactors 11 , R. J. Burnside, et al., February, 1978 .

44 XN-NF-81-34 APPENDIX A ECCS EXPOSURE SENSITIVITY STUDY FOR THE PALISADES H FUEL DESIGN In 1978, Exxon Nuclear Company (ENC) evaluated the performance of the Palisades H fuel design during a postulated loss-of-coolant accident (LOCA). The analysis used conditions reported in XN-NF-77-24(Al) for the limiting 0.6 DEG/PD break as boundary conditions for multiple fuel heatup calculations to establish a burnup dependent F6 limit for reload H fuel. The calculations included the effects of the NRC model for enhanced fission gas release and fuel rod internal pressure uncertainties.

The results of the calculations are shown in Figure A-1 which provides the maximum LOCA ECCS allowed peaking with exposure, normalfzed to BOC, for the Palisades H fuel design. Corresponding linear heat generation rates and ECCS results are given in Table A-1. The ECCS limiting F6 versus exposure curve for the Palisades H fuel design fs a constant F6 value of 2.76 (14.98 kw/ft total; 14.61 kw/ft heat release in the fuel) out to a peak rod burnup of 27,250 MWD/MTM. At higher exposures, up to a maximum exposure of 43,600 MWD/MTM, the F6 limit decreases as shown in Figure A-1 by about 20% .. The reduction in F6 is necessary to offset the adverse effects of fission gas release at high burnup on predicted clad rupture and flow blockage in the postulated

45 XN-NF-81-34 LOCA. The analysis shows that the Palisades reactor can operate with the H fuel design and satisfy licensing criteria specified by NRC 10 CFR 50.46 and Appendix K provided the F~ limits given in Figure A-1 are not violated.

A-1 Exxon Nuclear Company, 11 LOCA Analysis for Palisades at 2530 MWt Using the ENC WREM-II PWR ECCS Evaluation Model, XN-NF-77-24, July 1977, and XN-NF-77-24, Supplement 1, August 1977.

-1 Table A-1: Palisades Exposure Sensitivity Results for H-Fuel at 2530 MWT

---*-* - . . --~*- *-~-,

(0, l. 0) (27.25, 1.0) 0.8 (33.6, 0.820)

(43.6, 0.780)

I- O" l.J..

c:

  • o s....

Q.J 0.6 r-0.

.µ .

r-

s
E a.
I c:

s....

0.4

I co

..co LL.

0.2

z I
z "Tl CX>

10 20 30 40 50 I w

-Po Rod Average Burnup (GWD/MT)

Figure A-1 Palisades H-Fuel, F~ versus Peak R.od Burnup

48 XN-NF-81-34 i APPENDIX B

{

COMPARISONS OF CALCULATED AND MEASURED ASSEMBLY POWERS FOR THE PALISADES 4. 0 w/o Gd 2o3 DEMONSTRATION Currently there are a total of 64 gadolinia bearing fuel rods being irradiated at Pali~ades. The initial loading cif*gadolinia occurred at the startof Cycle 3 (Spring 1978). A total of 32 gadolinia bearing fuel rods containing 1.0 w/o Gd 2o3 were distributed among eight (8) assemblies. Additional gadolinia bearing fuel rods were loaded at the*

start of Cycle 4 (May 1980). In. this reload 32 gadolinia bearing fuel rods containing 4.0 w/o Gd 2o3 were distributed among four* (4) assemblies.

Comparisons of measured and calculated assembly powers indicated a variance of less than 3% during Cycle 3. This is typical for all assemblies and since no particular trends were observed it was felt that the calculational models adequately accounted for the effects of gadolinia.

A more severe test of the ENC calculational methodolOgy was initiated at the start of Cycle 4 wjth the frradiationof 4.0 w/o Gd 2o3 . Cycle 4 comparisons indicate a systematic bias between the measured and calculated assembly powers in the core locations where the gadolinia bearing assemblies are located. A relative constant difference of less than 3.0 percent has been observed between calculated and measured data through a Cycle 4 exposure of about 7,500 MWD/MT. F1gure B-1 shows the Cycle 4 fuel

49 XN-NF-81-34 assembly loading configuration. Figures B-2 through B-4 display quarter core power map comparisons at 500 MWD/MT, 2,200 MWD/MT, and 7,000 MWD/MT.

The Cycle 3 gadolinia bearing assemblies continue to show good agreement between the measured and calculated assembly powers. Figure B-5-shows a more detailed comparison of the power history for the gadolinia assembly.

Due to the close comparisons of measured and calculated assembly powers the ENC calculational methodology is adequately accounting for the presence of gadolinia in the core

  • i 50 XN-NF-81-34 1

. {

M N -Q R T v x z l

13 2 1 2 2 1 1 2 Fresh

_J l 14 1 2 1 1 2 2 1 Fresh

  • --r I i . .*.

16 2 1 1 1* Fresh 2 Fresh Fresh

. 17 2 1 1* 2 2 1 Fresh 19 1 2 Fresh 2 Fresh** Fresh Fresh 20 1 2 2 1 Fresh Fresh - No Burnup 1 - Once Burned 22 2 1 Fresh Fresh Fresh 2 - Twice Burned 23 Fresh Fresh Fresh

  • Contains 4 Gadolinia (1 w/o Gd?O~) Rods Per Assembly (Loaded Fresh in Cycle 3)
    • Contains 8 Gadolinia (4 w/o Gd 203 ) Rods Per Assembly Figure B-1 Palisades Gadolinia (Gd 2o3) Loading, Cycle 4

)

\

3, .

,A, r-- ......

51 1 p

...... I 3\

XN-NF-81..;34

.889 1.098 .921 .926 1.145 1.174 1.002 .984

  • 902 1.132 .947 . .942 1.132 1.140 .967 .990 I [ I
  • I. -1.44 -3.00 -2.75 -1.70 1.15 2.98 3.62 -0.61 I'

I

.917 1.028 1.229 .952 .928 1.117 .977

' i .934 1.067 1.235 .942 .916 1.075 .971

}

I

~1~02 -3.66 .

-0.49 1.06 L31 i

- **---, i,.-...:

3.91 0.62 i'iF --1 1~8F ..

--- .... ...I~

1

... ~

1.132 1.207* 1.193 .840 1.138 .744 1.171 1.232 1.231 -.813 1.123 .754 i-3.33 -2.03 -3.09 3.32 1.34 -1.33

.879 ~848 .971 1.002

.865 .822 .959 .985 1.62 3.16 1.25 1. 73 1-

1 '-I '

  • ~

~ A'

\......,.., PDQ at 500 1.078** 1.000 r~ .634 +

MWO/MT INC A at 498

.9~1 1 *. 080 .654 + MWD/MT

-0.19 0.60 -3.06 +

(C-

-11)

M x 100

  • Cycle 3 Gd 2o3
    • Cyc 1e 4 Gd 203 ** .

Figure B-2. Palisades Cycle 4, 'INCA Power Distribution*

(Measured) versus PDQ Calculated Relative Assembly Power

I I

I I

52 XN-NF-81-34

  • l l 3,.... ~----'t'-

.869 1.064

.. 872. 1.082

.907

.913 .913 Ip*,

.906 -- ~.126 ,__

1.090 1.lo5

_\3

.982

.956

.966

.989

-o*.3 -1. 7 -0.7 -0.8 +0.8 1.9 2.7 -2.3

-- ~.

.901 1.010 1.191 .942 .* 925 1.101 .969

.907 1.034 1.195 .932 .916 1.060 .976

-0.7 -2.3 -0.3 1.1 1.0 I 3.9 -0.7

~.'"'\

~BF

. c.. ....

' l.... F I

1 I

1 .,_,....I I -*---

--- ~,....

1.110 1.i83* 1.208 .866 1.164 .758 1.137 1.207 1.253 .835 1.162 . 773

-2.4 -2.0 -3.6 3.7 0.2 -1.9

.894 .887 1.019 1.045

.877 .856 1.000 1.0*33 1.9

..___;._ ,0..._ )

3.6 1.9

~ r--...-

A' 1.2 j

PDQ at 2,500 1.148** 1.059 .676 MWO/MT

+ INC A at 2,204 1.164 1.051 .692 + MWO/MT

-1.4 0.8 -2.3 +

{C- M x 100

_!1)

  • Cycle 3 Gd 2o3
    • Cycle 4 Gd 2o3 Figure 8-3 . Palisades Cycle 4 Power Distribution- Comparjson Measured versus Calculated,. 100% Power

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

53 XN-NF-81-34


1 ~/---* - - *--- '~ )-*---..--------=--\.._!_~-~-=--=--=--=-.-~.' ""-~1":"_-_-_-_****..,---

.883 1.048 .916 . .905 1.052 1.065 .941 . 91~

.889 1.066 0922 .906 1.043 1 .. 048 .* 923 .94~

- .6.7 -1.69 -065 -.11 .86 1.62. 1.95 -3.17

  • 9.18 1.010 1.151 .941 .922 1.052 .92E

.914 1.011 1.151 .935 .921 1.028 . 94]

.44 -.10 -- .64 .11 2.33 -1.38 I

---'~

"""'"I"'

~ ....i--~---..

1.101 1.157* 1.226 .893 1.158 *75E L094 1.168 1.282 .864 1.175

  • 77]

.64 -.94 ~4.37 3.36. -1.45 -2.45 0935 .954 1.064 1.058 0914 . 926 1.035 1.046 2.30 3.02 2.80 1.15


, .......~ '.. ., ,. .... ,-~

\... .- ,i......A.;.-'--.

L 284-A*l.115 . 719 PDQ at 7000 MWD/~

1.310 1.102 .724 INCA at 6950 MWD/I *

-1. 98 1.18 -.69 (cr-.r) x 100 Figure B-4 Palisades Cycle 4, INCA Power Distribution (Measured) versus PDQ Calculated Relative Assembly Power, 100% Power

1.35 I

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

1.30 i F' F ,.

L~L'l~ 'JU i r>

~... f1F'AC.

. *~ IF.'i="I;

.:_)[_... '~L/

,,' rev r* LAT,....,.....'

1.25 ut-iL,. U1_,, LU

~

QJ 3:

0 a...

QJ

.,.... 1.20

<ti r-QJ a::

>, 1.15 r-

..0 E

QJ VI en

~

1.10 1.05

z I
z 1.00 I CX>

0 1 2 3 4 5 6 7 8 9 10 11 12 .......

I w

Assembly Exposure (GWD/MT) ""'"

Figure B-5 Palisades Cycle 4 Gadolinia Assembly Power versus Exposure

55 . XN-NF-81-34 Issue Date: 06/10/81 PALISADES NUCLEAR PLANT CYCLE 5 SAFETY ANALYSIS REPORT DISTRIBUTION FT ADAMS JN MORGAN CA BROWN LA NIELSEN GJ BUSSELMAN GF OWSLEY GC COOKE PM O'LEARY A EV I NAY JF PATTERSON RL FEUERBACHER FB SKOGEN RG GRUMMER (2) GA SOFER JD KAHN RB STOUT MR KILLGORE PD WIMPY WV KAYSER HG SHAW/CPCO (10)

  • wL LAMBERT DOCUMENT CONTROL (10)

ATTACHMENT B to Technical Specification Change Request L

XN*Nf-80-47 0

PALISADES POWER DISTRIBUTION CONTROL PROCEDURES OCTOBER 1980 RICHLAND, WA 99352 1 .

E)${0N NUCLEAR COMPANY, Inc.

( 8107280394 810721 i PDR ADOCK 05000255 p* PDR

XN-NF-80-47 Issue Date: 10/03/80 PALISADES POWER DISTRIBUTION CONTROL PROCEDURES AhMA.-....n Prepared By: R. G. Grurrm;~t F. B. Skogen, Manager PWR Neutronics I

Approved By:

R. B. Stout, Manager Neutronics & Fuel Management Approved By:

G A. Sofer, Manager N clear Fuels Engineering i ,

  • EJ){ON NUCLEAR COMPANY, Inc.

XN-NF-80-47

  • TABLE OF CONTENTS Section

1.0 INTRODUCTION

1 2.0

SUMMARY

.. 2 3.0 POWER DISTRIBUTION CONTROL FOR PALISADES J 4 3.1 ESTABLISHING THE TARGET AXIAL OFFSET 5 3.2 VERIFYING THE POWER PEAKING PROTECTION 8

3. 3 ADDITIONAL PROTECTION FEATURES . . . . 9 4.0 VERIFICATION OF THE POWER DISTRIBUTION CONTROL PROCEDURES FOR PALISADES . . . 14 5.0 METHOD AND MODEL VERIFICATION 19 5 .1 PALISADES CYCLE 2 XENON OSCILLATIONS 19 5.2 INPUT PARAMETERS FOR THE PDC-II ANALYSIS .. 20

6.0 REFERENCES

. . . . . . . . . . . . . 24 APPENDIX A - EVALUATION OF PDC PROCEDURES FOR IMPLEMENTATION IN THE PALISADES REACTOR . . . . . . . . . . . . . . . . . 25 I

  • Use. reproduction, transmittal or disclosure o'f the above information is subject to the restriction on the first or title page of this document.

LIST OF TABLES Table 4.1 REACTOR OPERATING CONDITIONS ANALVZED WITH THE PALISADES 1D MODEL . . . ** . . . * * . 16

\* 5.1 KEY PARAMETERS FOR THE GENERIC REACTOR AND PALISADES 21 LIST OF FIGURES 3 .1 PALISADES TARGET AXIAL OFFSET BAND . . * . . ............. 11 3.2 PALISADES ACCEPTABLE OPERATION FOR ONE HOUR

. OUTSIDE OF THE TARGET BAND . . * ** . . . ......... 12

    • 3.3 PALISADES AXIAL VARIATION BOUNDING CURVE (V(Z))

4.1 PALISADES VARIATION IN LOCAL PEAKING AT 500 MWD/MT .

4.2 PALISADES VARIATION IN LOCAL PEAKING AT 8,500 MWD/MT 0 * .

  • 13 17 18 5.1 PALISADES AXIAL POWER VARIATION COMPARISON . .. . 22 5.2 PALISADES AXIAL POWER VARIATION COMPARISON .... 23 UH, reproduction, tr*n1mitt.I or disclosure of the *bove inform*tion ia subject to the restriction on the first or title peg. of this document.

L

XN-NF-80-47

1.0 INTRODUCTION

The two phase development of a PDC-II(l, 2, 3) type power distribution control procedure for the Palisades reactor has been completed. Phase I of the analysis demonstrated the applicability of a PDC-II type procedure

' for Palisades.( 4) For completeness the Phase I report is duplicated in

\

\

Appendix A. Phase II, reported herein, provides a procedure* for the implementation of a plant specific version of PDC-II for Palisades.

Phase II also demonstrates the validity of the Palisades calculational model.

The second phase of the evaluation of a power distribution control procedure for Palisades has been completed. The procedure developed sp~cifically for Palisades is derived from the current PDC-II procedure. (l, 2, 3)

Implementation of this procedure wi 11 a11 ow full power operation for up to one month after the incore detector system becomes inoperable.

Currently power must be reduced to 50% two hours after the incore monitoring system fails.

Protection of the power peaking limits is verified by evaluating the latest measured power distribution in conjunction with the expected .

variation in power peaking as determined by the PDC-II methodology and comparing to the Technical Specification limits on the power distribution.

The variation in power peaking is controlled by limiting the change in axial offset measured by the excore detectors. In this way, the reactor can be operated exclusively on the excore detectors for an additional month after the last full core power map, used in establishing the

~ target axial offset, was taken.

Use, reproduction, transmittal or disclosure of the above information is subject to the restriction on the first or title page of this document.

I

XN-NF-80-47 I

,I 2.0

SUMMARY

A power distribution control procedure, based exclusively on monitoring by the excore detectors, has been developed for Palisades. This Palisades specific procedure is derived from the Exxon Nuclear Company (ENC) PDC-II methodology and will allow full power operation for a period of about one month subsequent to a full core power map. Analysis of non-equilibrium conditions has shown that the PDC-II type procedure will assure that power peaking will be maintained within the specified power peaking limits. for Palisades. The Palisades Phase II analysis has shown that the model accu-rately predicts the change in power peaking for axial offsets beyond the range of expected operation.

The PDC-II procedure developed for Palisades is expected to be imple-mented only when the incore detector system is inoperabl_e. In the* event of a failure of the incore detector system, this procedure can be implemented without any interruption in power operation .. The incore detector power

  • peaking (F(z)) alann could be replaced by an alann on the excore axial off-set output; e.g., an alann which would be activated when the axial offset drifts outside the allowable operating band. If the axial offset has be~n within the band prescribed by the Technical Specifications for the la.st 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and the latest measured power distribution is within the prescribed limits, no change in power level will be required.

Use,* reproduction, transmittal or disclosure of the above information is subject to the restriction on the first or title page of this document.

XN-NF-80-47

      • 3.0 POWER DISTRIBUTION CONTROL FOR PALISADES The power distribution control procedure described in this report enables the Palisades nuclear plant staff to manage the core power distribution without the incore monitoring system such that Technical Specification limits on F6 are .not violated during normal operation.

Limits on MDNBR are also protected during steady state, load follow, and anticipated transients. The procedure Provides uninterrupted operation at full power for the Palisades in the event the incore detector system is not available. The PDC-11 type procedure will provide up to one month of operation with the incore detectors inoperable as compared to only two hours currently allowed by the Technical Specifications.*

This report provides the method for predicting the maximum F6 (z)

  • distribution anticipated during operation under the PDC-II procedure, taking into account the incore measured A comparison of this maximum F6 eq~ilibrium power distribution.

(z) distribution with the Technical Specification limit curve determines whether the Technical Specification limit can be protected. If such protection can be confirmed, the excore monitored axial offset limits will protect the Technical Specification F6 limits *. The ~axi~um possible variation in F6 (z) that can occur.

while operating* under the PDC-II procedures forms the bounding variation referred to as V (z). V (z) is the means by which the maximum anticipated F6 (z) is predicted .

  • Use, reproduct1on, transm1ttal. or disclosure of the above information 1s subject to the restriction on the first or title page of this document.

XN-NF-80-47 Power distribution is controlled by keeping the axial offset within a prescribed band. The band is centered on the axial offset of the measured equilibrium power distribution, the target axial offset, and the width of the band is determined by the power level at which the reactor is operating. Application of these criteria is described in the following sections.

3.1 ESTABLISHING THE TARGET AXIAL OFFSET The PDC-II methodology and resulting guidelines protect the core power distribution limits in the absence of incore measurements.

Protection is accomplished through control of the axial offset, measured with the excore detectors, with respect to a target axial offset. Core axial offset is defined as:

AO =

where PT = Power in the top half of the core P8 = Power in the bottom half of the core The target axial offset is established by measuring the core power shape at near equilibrium conditions with the incore detector system. Excore detectors are separately calibrated to reproduce the incore measured axial offset. Operation within PDC-II guidelines then allows axial power distribution changes within a band referred to as AOTB' The axial offset target band is defined as:

+ 5%

  • AOTB = PfAPL Use, reproduction. transmittal or disclosure of the above information is subject to the restriction on the first' or title page of this document.

XN-NF-80-47 where:

P = Operating reactor power (MWth)

APL =Maximum Power (MWth) allowed under the axial .power distribution constraints.

This band is shown graphically in Figure 3.1.

Below a relative power (P/APL) of 0.9, the axial offset is allowed to deviate from the target band for one hour out of each twenty-four con-secutive hours, provided that the measured axial offset remains within a /

broader, but specified, axial offset band shown in Figure 3.2. If this requirement is violated, the core relative power must be reduced be.low 0.5 of the APL where no restrictions on AO are imposed. Above a relative power (P/APL) of 0.9, the measured AO must remain within the allowed target band at a11 times.

The target axial offset (AOT) must correspond directly to an incore measured power shape. This power shape will be used to verify that operation under the power distribution control procedures will protect the plant FQ limits. The target axial offset is to be established after attaining a power level at which the reactor is expected to operate. This will ensure that the power distribution accurately represents the reactor conditions if the incore detectors are not available. The target axial offset should be determined after achieving equilibrium conditions for sustained operation .

  • Use, reproduction, transmittal or disclosure of the above information is subject to the restriction on the first or title page of this document.

XN-NF-80-47 As previously discussed, the target axial offset is tied explicitly to the core power distribution. The.+5% band about a target axial offset will protect the Technical Specification limits with sufficient margin in the power peaking to satisfy the 1l"rocedure constraints. This feature ~ermits the plant operator to alter the core po}'ler configuration, and -r~-establish a target axial offset in an attempt to satisfy the criteria on measured power distributions. This of course can only be done when the incores are operable l

as a measured power distribution must be available which corresponds to AOT.

From a long term operating standpoint it is not practical to establish a target axial offset for which the :t,5% band cannot be maintained for susta1ned

  • operation. Plant operate.rs should establi.sh a target axial offset consis-tent with the anticipated operating requirements. Generally, normal opera-tion of Palisades will result in target axial offsets and associated power distributions which will satisfy the ~riteria on measured power distributions.

The schedule for establishing a target axial offset should be

  • written into procedures instead of the Technical Specifications. This schedule should represent the maximum interval between measurements. The target axial offset should be evaluated *for every mea~urement taken at equilibrium conditio~s. Extended operation, in the event that the incore instrumentation fails can be maximized by keeping the target axial offset and associated power distribution current.

Use. reproduction, lransmittal or disclosure of the above information is subject to the restriction on the first or title page of this document.

XN-NF-80-47 3.2 VERIFYING THE POWER PEAKING PROTECTION Prediction of the maximum anticipated F6 is made possible by controlling the power distribution such that it does not increase more than the factor V (z). This is accomplished by maintaining the.core axial offset within the band prescribed in Section 3.1. The value of the V (z) factor is determined from analysis of plant operation during which the axial offset is mainta.ined within the band during various operating conditions. The V {z) factor is multiplied by the measured power distribution with measurement uncertaintities included. The resulting power distribution is compared to the Technical Specifications to determine whether the Technical Specification limits are protected by*

the procedure. Procedures for this analysis are:

(1) An F6 (z)eq distribution is determined along with an associated axial offset, denoted as th~ target axial offset (AOT), at equilibrium power and xenon. The F6 (z)eq distribution is the measured F~ (z) distribution multiplied by the measurement uncertainty factor of 1.10

  • and the engineering uncertainty factor of 1.03.

(2) The F6 (z)eq distribution is multiplied by the V (z) factor, shown in Figure 3.3, to obtain the maximum anticipated F6 (z)max which is compared to the Technical Specification limit, F6 (z)TS" If F6 (z)max does not exceed the F6 (z)TS limit, then operation may continue in the absence of incores .

  • Use, reproduct1on, 1ransm1ttal or disclosure of the above information is subject to the restriction on the first or title page of this document.

XN-NF-80-47 T T .

If FQ (z)max exceeds the FQ (z)TS reactor core power must be

l. reduced, in the absence of incore.s, to a power level equal to the minimum value of the ratio (F6 (z)Ts/F6 (z)max}.

(3) The maximum allowed power level (APL) is detennined to be the minimum value of the ratio (Fb (z) 15/Fb (z)max.l times t~e rated power. For case where the ratio is greater than one (1.0) the APL shall be the rated power.

I 3.3 ADDITIONAL PROTECTION FEATURES The PDC-II criteria also includes a provision to account for any degradation in power peaking margin resulting from 11 upburn 11

  • The tenn 11 upburn 11 is used to describe the phenomeno.n in which the peak radial power increases with exposure rather than diminishes as is usually observed. The phrase "with exposure" is significant because 11 upburn 11 idenUfies a specific phenomenon and should not be confused with observed changes in peak power resulting from actions such as .control r.od movement or power level changes.

Utilization of burnable poison is one recognized mechanism for producing

. II up burn . Th e 1. ntegrate d pea k p1. n power, . Ft.H II r , 1. s use d f or mom. t orrng. up burn.

When the target axial offset is established, the core power distribution

. i obtained from incore measurements is compared to the distribution associated

,/

  • with .the immediately previous target axial offset. As previously described the intent of the Technical" Specification with regard to "upburn" is to
  • Use, reproduction, transmittal or di5closure of the above information is subject to the restriction on the first or title page of this document.

XN-NF-80-47 provide protection with PDC-II in the event the power peak is increasing with exposure. The requirement that maps, to establish the target axial offset, be used to monitor for 11 upburn 11 provides a means to identify such an increase., If_ the F~H is observed to.increase between these maps, a pro-vision for possible upburn during the period of operation without detectors must be made. The provision is to apply to the measured FQ an additional 2%

uncertainty above the previously specified uncertainties.

Another phenomenon that may be respJns1ble for an increase in the1 peak radial power with increasing exposure is a change in the azimuthal

  • power tilt.* This can be monitored by the excore detectors. The integrated peak pin power, F~H, incorporates any azimuthal tilt at the time the power distribution is measured. During operation when the incore detector system
is not available the azimuthal power tilt must be limited to protect the technical specifications limit for F~H and FQ' Any significant change in the azimuthal tilt would be indicated by the excore detectors. Azimuthal tilt, as *measured by the excore.detectors, shall not exceed 3% while oper-ating under the guidelines of the power distribution control .procedures.

U.e, reproduction, transmittal or disclosure of the above information is subject to the restriction on the first or title page of this document.

--X~N._,~~*1f..-.80~.~p~. ------ --

_J i -- -

I.

i

{__ _______ ._ .._.J i

i------

1 -to-----------------------------------------------.------.-------+-----..,-----*----

.  :  ! I I f- -- - - - _____

  • _J _ _, -;---*-***- -*--*4---,--'---

' ' I

. I . :

'--------HlH---4-------- ... * - - - - --------:----1"---l--------------,----"------t---------,

_; ___ _j __ ~--- ------------ *-* *---

1 j . '

---*---*- -------- * - - - - * - - - + -

i

--* -*--- *i**------ ---

i'

r *
  • t I

L_

i


:----te- --1---------- **r-

.......... .i

' _J 1  :

i

~---- ~-*** - ----£(}-* -+---- ---- -. ---~-------- -~

I : --:-- . i r-....... ~*~-*~---------------*- ----~i _ _ _ _,_ ---+-------~-----+--------i-----f----~---

I I


iI i *~

I*

I

-CIJ

~

I

  • - - .. j.

-- *---.-+---"--- i

-*--*c:---- I i

---, l

,------+------ -- ----**-~*-*---*_J o_---*---- ----str-

.J.. ---~------ -- **------* -*

~----- --- --w-- ----,

i 1- ***-*--*

, -

  • I 1

l-- 10-- --**-------------- --* -* ________ ,_.=~-~

I I .I I L____ --o-~~--~----~--~--~--~~----~~~--~----~~--~------~~--~-----~-----

1 . -Sp -4P 2p - Ir

!----~------ -----~---Beviatiurrfrom--the Tar~et-Ax/tal .Offset(~):

Qi 10 . 2P 31>- i f- .

sb , .

. ' I  :  : i I . 1, *

    • - ------ *-*  ! - ---L-----;----- - - - - -----*-*-*----;-----

I

--~

' '  ; I i .

I Fi gar e----3-~T- ra1 isares Ta~ get Ajxta:1 ~fset ~dnF.  ;

I

~--- -- --- j_ ---------.----. _____ *-- _____ :furLAci:eplt.ablajlpe.ram_on____ ----._ _,___ __,_

  • 1 ' *  ; .. I ---,----~-

___, I

! I I:--*---*--*--- ----------------* I I I

l

-* -- . -- - - ---- : - -- . ;* ---- --;-- - ... -- ---*-- r- ----

, I . ,  :

~--

u.*. reproduction. trammittaf or. di1ckmnie of th* *" inf0m1etiof1! i1 ~*


- ---to-therestric:tion on the first or title pep of this document.

i

    • --*- ! . - - *-* -- .. *- .. ---t*' **-----*-:l I

~-----* -----------*- *---*** *----------=--12--"----~-------'-__,[\~.wN-~.~IF-f""4;-S0~-4t+7 I I ' i 1---. **1* ...  : 1' ' i i


*-t*-- ----.;

\ '1

- - - - * *---*--------*----------*------ --------~ ~

I j

i I . i r I ' I I  ! !I ... -1I .... ,!

iI  !*

L. . --~----*--: - *-"-*---L - *--!---.---+--: --!--:---i--- ~--~

I. .~-~--~---~- ... _*. --~-~~------ -~- ~--(-_ :* __ ,_ -*-;-*---;-: ' .~--t----i __ L I , '

_J ---*--- '*.

I I i  !

I  :

i i

1----'*--------------------------l----'---,----'----i----~----;--

i I f* . -** ---**-**J.*--** *- __ _: ____ *----!--* --+--*--+-- _ __,___ _-,-~*- ---- ~----*- - i- - - ---i

'-----* -- -----------------------------~---.,...---+---..-----+----1

\ . I . t *  :

1 **


r-:---,--

.j .

! **. - -------------~

> - - - - *~l:JH---'1----**---- ---- **-**------~----+------~---4----+--'--+----+---

. ---* -----*- ---r __  : -~---i --** -* --i----~-+--- .


*---1-----: .

Not Acceptable  : i

  • Ncit Acee t~ble  :  !

. ~------  !. -- .; -**--*- - . !i *---,--- -*---~--*------ .....

I 1--. -------$*

I  !

r-* ..

  • -t-----**- - ------~

r- --- ------- ----*------;

-* 1

. -* j i

I I

i *-*1 * **i ..

[ ...J

~~-- I 1 . j

  • I I 1

r---~'b

~--~---. 1-*- -:---r---;---~~-+-*--- ----~-------:-*

1 '

f-~-C.--~- I '

1  :  ; . I  : . . I , .

L ... -~ -- **- ~- . - - '

. ---~- ___ : *-* ---* . I

  • I J

--4------;-----**--r-*~.-.....,..--

  • -:~-:-- -----j

~ - -4tJ i

!r- --a..--- - *-----**----- *-+----r---+----L---_:

'I t . '

1 t

  • i

' --;--~--

i

~-*-* -*- --* ***-* . ---*-**-*---**--*-----*----------+-----+-----------,-------___,1----~-.

I i l -_: ___ *----1I r- - -- * - ----- -- ----* - --*--*--~.arget-~-t- -llfl'-~~--'----;----,--~.,.---+---,-- ___!I.

I i l i I

I

!* I- -I*

i 1

. .*_ - Hl-

-- ****- *-*-***-*-*-------------*-+-------~------.--,---*t-----:-+-----:-i

~ I i

.... i I  :

_;___ . : -*i-- --:--- f- *- .. -i-- -- .. ;

l~.,---:---- ' ~ i '  !

i I--* ~ .... ~  ::~D__ . ~-%)- . -_3Q ___ - _.p

~ . . .

~~O .. ___ 9,_ ---~-J&-* -~Q__. __30 l: ~l d

i~---'------------~ion

. . . . . fronj Target AO (j%) ' .. i* j..

~-- Figure 3. 2 *Peil i sades *Acc.eptabl"tOperati on ~or 'e~e Hour ,-------,-*--

!f------ ---,---- l '

  • a ..

,utsi:de :t=

.of 1a11ge~ ~!:land-:

. ~ . - I., 1

    • 1 .  ; \ .

~- - ~_ -~--~~r-:+- ~---~----:*-~--+-- ---~---:"-_;*j f- - -  :. - .. _!_ ___

.;.  : . . : ... : . -- ---!--.:~ ___J ___ :_ .. ~ __ _; _*1:*--:*-- . ------1 Ute. reproduction, transmittal or dilcl0sur1 of the abovt inform.ion :ii ~

  • - --* ----- --- -ioth* r11triction on th* first or title i>-. of thi1 document.

~-

--r11 I I T---r-1-: -i---J---i--:* I I  ;---T---~--~-- .

l :.

i---* I I i ~ I : . ! . : :. 1' . , I  !..  : . I . I . !

I .l I I

. I ' - '- - I .. (0 ' j!. *1*6*5 )1' i :- i I i I

II I i

--1.,,16-i ' I "T T- . .. ,.: _ _  !. . 1* . I I

i I

l '  !

I I* -:- 1,1

  • i ' I I '. i  !

I1!"*

i I  ; ' '

I I Ii I i I  !

l.i 14 l l ""'("' --1** -- - 1, ..  ! _,.

.I .

I 11 . I I *  : ' I ,. I 01it ,(.23;!1.-H>; *: , __ !I i.,. j .. 1h2* i* .! i I

-, I  : \.

~Ii I

I  ! +j1-0 ;  :

i ,.

I ,. *' (

' I

+- +--1  : I i i  ;

I I -*.

i ' .... 1 i

!!,~:L ' - J_1Joo' a 1  ! . *-* ** : L j H  ; -*J ' .. 1 I I

.i t -, T , -r -r-;- .~*-.-*11-- r-- - -- ---" -

I '  ! i :*

1 1 1 .1  !  : 1  : * ..

n_ :'. --1~ _:-. .:~ir~

1 1

1 . 1* -- .. 1 -; , *-1 .- ** * -t i i -

I ,-; . ---- ,-- - -'* ; . ---- - I f -T- ' --,-- l 1

  • -*~.*** -'
  • . fa!_ J~-- - .. *i ---- -~-. &2-l *.

I

.J 1

1 j I 1 1

1 J

- -- - ___ ..... - _J ___~ ~~-L=*- -::_-_ : ~--~-~-- ~ -~ T... --;.: -----~ l -~ . -l ---1 *.

I I I I I ' I 11_--~--:~l-~-~,;~--: .----

if:*

1 i  ! . \_  : .i - i  : i - i . ,:.

r ----:-- *- i*- *- ---1**--* *--i *- - *--f-*. -*--~ -*. -*- ;----- ___ .__ - ---1---~ --*-,--- .- *t**-L-----~ . --~-- ---r--- --**t . --

1 1 i , j' - -:- I ii-:- .. -*-:

  • L

~ iL--l -- ~ 1 * *- * * **

  • 1  ! ,  : ** 1 -r - J~*- - :~- ~-
  • . l * , .. *
  • toP :o * *** I* i . * .o..:4. -1° * *.. *1 1
  • 1
  • 0 *9 . - -
  • 1 io~ro;.;-- ! ** ;*-.
  • J~ * (~radti~rr- **IFu 1--~ei ht'.*- * ------!--

1

.... *-i* --* -*-+*- ---*:----,--* **** * *-!-A ia11*P siti - *: i-* *---- ;- -* -* -- ___ L .. --

  • II .
  • 1I ~ . I'I *  :! l I  : . I I : . i: ' ! i! '!
..... ! I '

L I

'1 i ,. i ...,,. : I .

....... I!* *L*** 1* *- **-r, --- -:_-*--1 -_-- ___:__ -**fi9*:,ur~*1;*.:3- --~li odes- rial~ari:ation /Bo di:ng Cotjdi ion - * --- .. ---~,; ___ :__ ]_ *-**-*-* -~-'.-**-- --**-***.

L _*.. * . . .J .. 1 _L 1

.i... 1 .. i _ I _i . J. .L ___ L _L_ _J _.! * *l_ *-- ...:_.cl.. _ [_ __ ....i .....J..

XN-NF-80-47 4.0 VERIFICATION OF THE POWER DISTRIBUTION CONTROL PROCEDURES FOR PALISADES A 10 XTG model was developed for the Palisades Cycle 4 and simulations of various non-equilibrium situations were conducted. Load follow simula-tion cases investigated are described in Table 4.1. These cases were deter-mined to be most limiting in the generic PDC-II analysis. The sensitive core parameters in the PDC-II model for Palisades were set equal to the parameters used in the generic PDC-II model as shown in Table 5.1. This I resulted in a more conservative model since the simulation with the generic parameters is more sensitive to perturbation and the axial offsets encoun-tered are more extreme than with Palisades specific parameters.

An analysis with the Palisades specif°ic parameters demonstrated that during load follow conditions the reactor would not reach the bounding axial offset, limits of the PDC;.. II procedures. Cri ti ca 1 parameters used to increase the axial offset swings were the Dopp1er broadening coefficient and the moderator scattering cross-section. By increasing the Doppler broadening coefficient and reducing the moderator scattering cross-section the axial offset swings were increased. This allowed the simulation to reach the upper and lower axial offset limits of the PDC-II guidelines. During normal operation including conditions such as load follow, the Palisades reactor should not approach the axial offset limits of the PDC-II guidelines .

  • Use, reproduction.transmittal or disclosure of the above information is subject to the restriction on the first or title page of this document.

XN-NF-80-47 Calculations with the Palisades specific model showed that the change in power peaking fell below the upper bound of the maximum c~ange in the linear heat generation rate established for the generic PDC-11 reactor .

.Variation in power peaking was generally below 8% at the beginning of the cycle but increased to 10% near the end of the cycle. Analyses in Phase 1 showed peaking variation as high as 13% near the top of the core. Calcu-lations for both BOC and EOC are shown in Figures 4.1 and 4.2. This analysis I

indicates that the PDC-II procedure, when used in Pali sades, wi 11 provide adequate protection for the limits on linear heat generation rate.

Use. reproduction, transmittal or di5closure of the above information is subject to the restriction on the first or title page of this document.

L

XN-NF-80-47 Table 4.1 Reactor Operating Conditions Analyzed with the Palisades lD Model Exposure (MWD/MT) Target Axial Offset Description of Operation 500 0.0 Operation*such that the axial offset is maintained within the target band

(+5% AO ta*rget)

I 500 0.0 Operation such that at full power the axial offset is maintained at the positive limit and at half power the.

axial offset is maintained at the negative limit

(+ (full)/~ (half))

8500 -2.5 +5% AO target 8500 -2.5 +(full)/-(half)

  • Use, reproduction, transmittal or disclosure of the above information is subject to the restriction on the first or title page of ihis document.

I L

,----r-*

  • . I **-**-**r. *--*

I I

I i

~

XN-NF-ao:...47 5.0 METHOD AND MODEL VERIFICATION The power distribution control methodology described in References 1, 2, and 3 utilized the XTG( 5) computer code in a one-dimensional configuration. The results obta1ned with the one-dimensional model have l; . been tested against standard ENC methods{s, 7,B) and operating data from the Ft. Calhoun, D. C. Cook #1 and Palisades reactors. Thermal hydraulics and Doppler feedback effects are accounted for in the model.

5.1 PALISADES CYCLE 2 XENON OSCILLATION Near the end of the second operating cycle of the Palisades

. core a significant xenon oscillation was observed. The measured axial offset oscillated between +20% and -20%. The 1ncore monitoring system was used to measure the power distribution throughout the oscillation.

INCA power maps were evaluated at the peak axial offset for both the positive and negative swings. These maps were used to demonstrate the adequacy of the one-dimensional XTG model in calculating the variation in the axial power distribution for large axial offsets.

A 1D XTG model was developed for Cycle 2 and comparisons were made with measured data. Compari~ons.of the variation from the equilibrium power distribution before the oscillation for power distributions at plus and minus 20% axial offsets are shown in Figures 5.1 and 5.2. The calculated variation in power distribution compares well with the measured in both cases. The maximum difference is 3% for the 25% swing in axial offset and 2% for the 14% swing .

  • Use, reproduction, transmittal or disclosure of the above information 1s subject to the restriction on the first or title page of this document.

... 20- XN-NF-80-47 The accurate representation of the variation in power distributions .at high axial offsets (outside the typical operating range for PWR's) in addi-tion to the large data base the model has been tested against in the oper-ating range, adequately demonstrates that the model accurately calculates the variation in axial power distributions for various axial offsets.

5.2 INPUT PARAMETERS FOR THE PDC-II ANALYSIS In the Palisades specific PDC-II analysis several key core para-meters were set equal to the values used in the generic PDC-II analysis. '

This was done in order to force the reactor simulation to calculate power distributions for the bounding axial offsets. The use of these generic core parameters in the Palisades analysis ~ncreases the confidence that the Palisades core can be protected by the PDC-II procedure. A comparison of some of the key parameters for Palisades and the generic reactor model is shown in Table 5.1.

Basic XTG input.parameters such as fuel cross-section sets, expo-sure distributions, and extrapolation lengths were those specifically developed for the Palisades core .. By using these parameters along with the generic ones, which increased the core sensitivity to perturbation, the model applies conservatively to Palisades and ensures that the Palisades reactor can be protected by following the PDC-II Guidelines .

  • Use, reproduction, transmittal or disclosure of the above information is subject to the restriction on the first or title page of this *document.

XN ... NF-80-47 Table 5.1 Key Parameters for the Generic Reactor and Palisades Generic PDC-II Palisades PWR PWR Rated Power 3,250 MWt . *2 ,530 MWt Power Density 100 kw/liter *79. 7 kw/liter Acitve Fuel Length 12 ft. *11 ft.

Coolant Flow Rate* 1. 43x10 8 lb/hr *1.269xl0 8 lb/hr Inlet Temperature 545°F *532.5°F Single Control Bank Worth 1.0 %p *0.45 %p Doppler Broadening Coefficient *0.0059 0.0035 Control Rod Insertion Limit 5.0% at Full Power *25% at Full Power Xenon Absorption Cross Section (MND). *3.10xl0 6 barns 2. 80x10 6 barns Moderator Scattering Cross Section *1.5 barns 1.6 barns

  • Denotes parameters used in the PDC-II analysis for* Palisades.

Use, reproduction, tr*n1mitt*I or dj1clo1ur1 of the above inform*tion is subject to the restriction on the first or title page of thi* document.

1.4 1.40  !. ------+-

1.35 0

~ Measured hAO 25.3 1.30 1.25 . ' 0 Calculated t.AO = 24.9

~  !

E

I s;..

1.20

..c

I 1.15 r::T LLI 1.10 *!*-A<f> 1 --t-** r-**- *i -**- --*----- *-----** *----**T*-------

s-

~ 1.05 c...

j- *-* -~--- -

i**

  • -*- -! r - -.. j __ j 4___ j _c T  :.,- +-- J+_~;>:~: .

1

. A.(f): __ -t_.~j _ -__ --~-~~.j---* ~

C1l

.,..... LOO

- __ , -r  !

i A tti. I 1 *.

,_Jr!

i, i, i

! ' . I '


:-----*r----j----------- N

.µ l'CJ i I . . 1

  • * - I I  !. ..... , I

.&--~-  :..+-:--+----*-t**--*-+*--- *-*-*----* ------~---*

r-C1l 0.95 . --1 *-- --,. -: -r--- -

~ : t:---r----r- .. ! ' *'**-**-- __ ; ---~ _____ j_-: ~1~

0:::

s- .

j : .!

C1l 3:

0 c...

0.')0 i .

I i

I . :

I -*--*-*-~*- -

      • ----*-***l-
  • + ;___T4
I I I i

.- * * , ; :_ : I I I

.J I .

C1l

> 0.85 -- ! . -- .. ! .. --* -*-- ---*--l*------**! ---* - ___ : _. ____ ;______**-t-----1-**------ --

.µ l'CJ

' :A<!> I i C1l 0.30 0:::

r --. ---*

-;*- *: *--~l . :

I '

0.75 i'

0.70 -f----+--

! I

  • -i--.

I ___ J._ _

I 0.65 I.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Top Bottom Axtal Posttton (Nodel Figure 5.1 Palisades Axial Power Variation Comparison U11, reproduction, trantmit.. I or di1clo1ure of the above information i1 subject.

to the restriction on the first or title page of this document.

1.40 1.35 I.

  • 1.30 11 Measured ~AO= 13.7

. Q *Calculated llAO = 13.9

-E

3 s..

1.25 1.20 - 1*- --- --*

..0 r-

  • r-
3 1.15 0-LL.I
  • .. i -

s..

QI 1.10  !

I

~ __ :. -- -

c..

1.05

  • -- -T-***

QI

  • r- . . _i

..... 1.00 ra

- QI 0::

s..

QI 3

0 c..

0.95 0.90 i

QI

..... 0.35 ra QI 0:: 0.80 0.75

z 0.70 .,,:::::

I I

0.65 ~

~ '

0,60 1 2 3 '4 5 6 7 8 q 10 11 12 13 14 15 lfl 17 18 19 21) 21 22 21 l4 Top Bottom Ax1~1 Positton (Node)

Figure 5.2 Palisades Axial Power Variation Comparison Uae, reproduction, tremmituil or di1do1ure of th* above information is subject to th* reatriction on th* first or title pag1 of this doeument.

XN-NF-80 ... 47

6.0 REFERENCES

1) XN-76.;40, "Exxon Nuclear Power Distribution Control for Pressurized Water Reactors", Exxon Nuclear Company, J. S. Holm, F. B. Skogen, Se_ptember 1976.
2) XN-NF-77-57, "Exxon Nuclear Power Distribution Control for Pressur-ized Water Reactors - Phase II", Exxon Nuclear Company, J. S. Holm, R. J. Burnside, January 1978.
3) Supplement 1 to XN-NF-77-57, Exx_on Nuclear Company, June 1979.
4) PWR-020-79, *"Evaluation of PDC Procedures for Implementation in the Palisades Reactor",*Exxon Nuclear Company, A. W. Prichard, August 1979.
5) XN-CC-28, Revision 5, "XTG - A Two Group Three-Dimensional Reactor Simulator Utilizing Coarse Mesh Spacing and Users Manual, Exxon.

Nuclear Company, R. B.. Stout, July 1979.

6) XN-75-27, "Exxon Nuclear Neutronic Design Methods for Pressurized Water Reactors", Exxon Nuclear Company, F. B. Skogen, June 1975.
7) Supplement 1 to XN-75-27, September 1976.
8) *Supplement 2 to XN-75-27, December 1977.

Use, reproduction, trensmittal or disclosure of the ebove informetion is subject to the r*st_riction on the first or title pege of this document.

XN-NF-80-47 APPENDIX A EVALUATION OF PDC PROCEDURES FOR IMPLEMENTATION IN THE PALISADES REACTOR I

Use, reproduction, 1ransm1ttal or disclosure of the abOYe information is 1ubj1e1 to the restriction on the first or title page of this document.

. XN-NF-80-47 1.D INTRODUCTION ANO

SUMMARY

A preli.minary evaluation (Phase I) of operating the Palisades Reactor within the Exxon Nuclear Company developed power distribution (POC-II control procedures has been completed. The implementation of a power distribution control procedure such as PDC-II in Palisades provides a means by which the reactor can be operated exclusively on excore detectors for a certain length of time in the event that the incore detectors fail. Power distribution control using PDC-II provides a separate system for ensuring margins to the safety limits such as LOCA (Loss of Coolant Accident) and MDNBR (Minimum Departure from Nucleate Boiling Ratio). The details of the ENC Power distribution control methodology can be found in Reference 3, PDC-I, and in References 4 and 5, PDC-II. Briefly, the PDC-I methodology protects a PWR generic FQ-limit of 2.32 while PDC-II, which is keyed to the actual measured power distributions at the plant, will protect a variable (in magnitude) FQ limit. The latter is accomplished by quantifying the variation in the core power distribution during anticipated load follow operations and combining this function with the measured distribution to establish the maximum anticipated power shape or conversely the lowest limit that can be protected.

The Phase I investigation indicates that the PDC-II methodology, briefly discussed above and in References 3, 4, and 5, can be incorpo-rated into the Pa 1i sades reactor control procedures. The PDC-I I system, Us*!, 1ttl1rut1uc11on, cransrn1ttal or <11st:1os.ure of the abuve 1nforma11on is sub1ect to the restnction on the first or t1118 P*Q* of thi' document.

-21 .. XN-NF-80-47 which depends only on excore instrumentation for normal operations and incore power maps for calibration, will benefit the Palisades reactor control system if implemented. The PDC-11 procedure will allow operation of the Palisades reactor for a certain length of time after the failure of the incore detector system. This length of time \vill dep~nd on ho1v well the excore detectors will remain calibrated. In some cases the recalibration interval has been as long as four (4) months. For Palisades this parameter must be defined. Currently the plant is allowed to operate only two hours at 100% power after the loss of the incore detectors ..

A plant specific PDC-IJ model for Palisades has been developed and compared to the generic ENC PDC-11 re.actor model. The Palisades model using conservative core input parameters from the generic model produce variation5 in LHGR results which are bounded by the generic PDC-11 limits.

Based on these variational results in conjunction with anticipated Palisades equilibrium power distributions it is anticipated that the power distributions allow~d by the PDC-II procedures will be bounded by .those used *in previous safety analyses. (lo)

A preliminary outline of the procedures for implementation of POC-

. II in Palisades has been included. In Phase II of this work a complete set of operating procedures will be issued. In addition, verification of measured and anticipated Palisades power distributions with regard to PDC-II will be conducted. The Phase II work will be ai~ed at incorpo-rating PDC-Il procedures into the Palisades Technical Specifications and*

the completion of all tasks required to obtain NRC approval for use of th~ procedures at the plant.

  • Us.o. repro<h~ct1on. tr*nsmitt.al or d1M:losure of the *bave ;ntor1nat1on rs sub1ect to,rh,? restriction on the first or title paQe ot U\1s ctocu1nent.
    • XN-NF-80-47 2.0 METHODS l The power distribution control methodology described in References 3, 4, and 5 utilized the computer code XTG( 6 ) in a one-dimensional configuration. The results of the one-dimensional model were tested against standard ENC methods(?,B, 9 ) and operating data from the Ft. Calhoun and D. C. Cook reactors. Thermal-Hydraulic and Doppler feedback effects are accounted for in the model.

Reactor power distributions can be bounded by the PDC-II procedure described in References 3, 4, and 5. The PDC-II operating guidelines p*roposed by ENC maintains the core power distributior. within limits by Plant Technical Specifications through the use of excore Hetectors. The total peaking factor, F , is protected by controlling 0

the axial power distribution and maintaining the difference in the Axial Offset, (the difference in the power in the top half and the power in the bottom half over total power) as indicated on the excore detectors, to within +5% of a target axial offset as determined by the last incore calibration. The significant feature of PDC-II is that it is viewed as controlling the variation in the axial power distribution rather than controlling the axial power distribution itself. The PDC-II procedural control of tre variation of the axial power distribution ensures margins to the reactor safety limits without being operationally unduly restrictive.

In the Palisades analysis Cycles l, 2, and 3 were analytically depleted in a three-dimensional quarter core model and normalized to

~easured data. The Cycle 3 model was then colla~sed to a one-dimensional Use, repruouct1on, tran11nirtal or drsl:losure of the above information is subject 10 the restriction on the first or title paoe of tn1s docurnent.

e i

XN-NF-80-47

. axial model for con;parison to the generic PDC-II model. Only cases found to be most limiting with PDC were run with Palisades data and compared to the generic PDC-II results .

  • UUt, reproduction, tr1nsmitt1l or disclosure of the above inform1tion is subject to the restriction on the first or title page of this document.

I 3.0 EVALUATION OF PDC-II CONTROL FOR PALISADE5 XN-Nf-80-47 Three .areas have been investigated to determine if PDC-II can be utilized to control the power distribution in Palisades. The first area of investigation was to determine if the Palisades power distribution l* behaves and responds similarly to the reactor model used in the g~neric I.

analyses performed in support of PDC-II. The second area of investigation was to determine if PDC-II procedures protE.Ct the MDN13R limits as specified in the Technical Specifications for Palisades. The third area was to determine if PDC-II would protect against exceeding the peak to average assembly power ratio limit (Fr) described in the Technical Specifications for Palisades. The means for showing that the Palisades reactor is similar to the generic PWR is discussed in Section 3.1. The work in this area concentrated on showing that the v~riations in the Palisades power distributions fell below the bounding limits of the generic reactor, and that the parameter selection of the Palisades PDC-II XTG models is conservative. The protection of the MDNBR limits for Palisades, discussed in Secticn 3.2, depends in part on the ability to ensure against exceeding the Fr limit. The method to ensure against exceeding the Fr limit is discussed in Section 3.3.

3.1 COMPARISON OF PALISADES TO THE GENERIC PDC-II REACTOR Comparison of the generic PDC-II reactor to the Palisades reactor was conducted by comparing the results of the PDC-II generic model to the results obtained with the PDC-II model developed speci-fically for Palisades. The load follow simulation cases investigated

  • are described in Table 3.1. These cases were determined to be the Use, reproduction, tr*nsm1ttal or disclosure of the above information is subject to the restriction on the first or title page of tn11 document.

XN-NF-80-47

  • most limiting in the generic PDC-11 analyses. In all cases the sensitive core parameters in the PDC-11 model for Palisades were set equal to parameters ,from the generic PDC-I I mode 1 and determined to be conservative.

In addition the axial peaking factors, as a function of control rod insertion and axial position, from the generic PDC-II analysis were used for the Palisades reactor. These parameters in Palisades are as much as 7% lower than thos~ used in the generic PDC-II analysis. The Use of generic core parameters in ttie Palisades analysis increases the confidence that the results are bounded by the PDC-II limits. Table 3.2 shows a comparison of some key parameters for Palisades and the generic reactor.

If this model is determined to be unduly conservative, the calculational model will be modified to correspond mo~e specifically to Palisades.

Preliminary power peaking calculations with the Palisades PDC-II model, with conservatisms, was determined to fall .below the upper bound of the maximum chinge in the linear heat generation rate peaking factor (FQ) for the generic PDC-II reactor, see Figure 3.1. From this analysis the POC-II procedure~ when used in Palisades, will provide adequate protection for FQ{Z) limits.

3. 2 PROTECTION OF MDNBR LIMITS The PDC-II proc~dure ensures that MDNBR limits are not exceeded by showing that reactor operation d.oes not exceed the limits set forth.

in the Technical Specifications. Reference 10 shows that ~t 115% power using peak pin and peak assembly limits, with a 65~ maldistribution of flow that MDNBR was above the limit of 1.30. The transient results, re~6rted in Referer.ce 10, also indicate an MDNBR of greater than the 1.30 limits.

Us*!, reproduction, tran11TI1ttal or d1sc..losure of th~ above information is subject

  • 10 thP. r.! 5u1c11on o~ the *first or tit le P*OO of this document.

-32 ... XN ... NF-80 ... 47 Analysis of the plant transients indicates that operation under PDC-11 would yield MDNBR values greater than or equal to those calculated in Reference 10. In the transient analysis the input values for Fr, Fz and F~ are all greater than or equal to the peakin9 values anticipated !Jy op~ra tion under the PDC-II procedure. The PDC-II procedure ensures that MONBR limits are not exceeded by maintaining core peaking factors bel~w operating limits.

3.3 PROTECTION OF Fr LIMITS USING EXCORE DETECTORS The PDC-II method does not directly ensure that the Fr limit is not exceeded. By imposing a quadrant tilt limit using the excore detectors, however, it is possible to ensure an Fr limit. If the excore quadrant tilt is calibrated with an incore measurement and a maximum quadrant tilt limit is imposed, the reactor should be allowed to continue operation with no real restrictions imposed under normal operating conditions.

The restrictive tilt limit is needed to protect the core against any transient or abnormal operating condition which may cause radial power increases accompanied with only minimal changes in the axial power dis-tribution, i.e., dropped rod .

  • Usf!. 1t!,HOduct1on, tranvn1ttal or d11Llosure of the above i11tormat1on is subiect 10 the restr1ct1on on the firit or title page of this. docum*nt.

XN-Nf-80-47 i .

Table 3.1 Reactor Operating Conditions Analyzed with the Palisades 10 Model*

  • Exposure Target*

MWD/MT Axial Offset Description of Operation 7,500 -2.6 Operated suc.h that the axial off set is maintained \*ii thin the target band. (+/-AOTB) 7,500 -2.6 Operated such that at full power the axial offset is maintained at the positive limit and at half power the axial affset is maintained at the negative limit.

I+AO(full)/-AO(half)]

. i '

  • Use, reproduction, transmittal or disc:losure of the above information is subjec:t to the. restriction on the first or title page of this doc:ument.

XN-NF-80-47 Table 3.2 Plant Characteristics Generic POC-II Palisades PWR PWR Power 3,250 MWt 2,350 MWt Power Density 100 k\oJ/l iter 79. 7 k\'l'/l i t~r Active Fuel Length 12 ft. 11 ft.

8 lb/hr Coolant Flow Rate l.43xl0 8 lb/hr 1. 269x10 Inlet Temperature 545°F 537.5°F Single Control Bank Worth 1.0 %µ 0.45 %p Doppler Broadening Coefficient 0.0059 0.00585 Control Rod Insertion Limit 50% at Full Power 25% at Full Power Use, reproduction, transmittal or disclosure of the ;bove information is subject to the restriction on the firn or title. page of this document.

XN-Nf-8'0~47

.\

4.0 IMPLEMENTATION OF THE PQC-II PROCEDURES IN PALISADES Implementing the PDC-II procedure in Palisades will impose tighter operating controls than do the incore detector and co~trol system proce-dures presently in use. The PDC-II procedure bounds operation using I

I excore detectors. PDC-II, being more restrictive, will require that ~t be in operation at all times to ensure that the condition of the core (Xenon distribution, Power Distr.ibution, etc.) is within PDC-II limits I in the event that the incore monitoring system is inoperative. The PDC-II procedure is based on maintaining the reactor axial offset \'lithin specific bounds to ensure that FQ(Z) and MDNBR limits are not exceeded.

The following is a brief description of procedures which will be requtred to i~plement PDC-II in Palisades:

1 Quadrant Power Tilt with excore detectors. This procedure is needed to ensure Fr limits when the plant is being operated exclusively on the excore detectors. Included in the procedure will be a new set of limits for Quadrant Power Tilt and an excore detector calibration procedure I

for Quadrant Power Tilt.

1 Power Operation using the excore detector power ratio alarms - This procedure will incorporate axial offset limits on the core which are required to confrom to PDC-II procedures using the excor~ detectors. Included in these procedures will be operating guidelines and cal-ibration procedure for excore detectors.

Use, reproduction, transmittal or di1clo1urt of the abov1 i:iform1t1on i1 subject to the restriction on the first or title page of this document.

L

XN-NF-80-47 5.0 PHASE l} WORK There are several additional objectives to be achieved in the Phase II of the PDC-II procedure implementation for the Palisades reactor. The first*

will be to compare the Palisades PDC-11 model tQ operating data dealing with xenon oscillations. This will calibrate the Palisades PDC-II model to operating data.

The second will be to do a more detailed analysis of acci-dent stiuations while -0peratirig under PDC-II limits to assure that MDNBR

.limits in Palisades are not violated. The third objective is to prepare a complete set of procedures for implementing PDC-II in Palisades with the intent of achieving on NRC operating license.

  • Uso. rtUJt0<1uc11on, 1ran51nirtal or d*Sl.IOsure of th* above information 1s subject 10 1he restr1c11on on the f1rit or title P*O* of this. document.

XN .. Nf .. 80-47 I

6.0 REFERENCES

1. Letter, Bruce D. Webb to Roland J. Ehlers,

Subject:

Palisades Physics Support, October 23, 1978, Webb 42-78.

2. Letter, R.J. Ehlers to W.A. Walker,

Subject:

Power Distribution Mcnitoring System Using Excore Detectors, March 30, 1979, RJE:061:79.

3. J.S. Holm F.B. Skogen, "Exxon Nuclear Povier Distribution Control For Pre~surized Water Reactors, XN-76-40, Exxon Nuclear Company, September, 1976.
4. J.S. Holm, R.J. Burnside, "Exxon Nuclear Power Distribution Control For Pressurized Water Reactors: Phase 2", XN-NF-77-57, Exxon Nuclear Company, June, 1978. *
5. Supplement 1 to Reference 4, June, 1979.
6. R.B. Stout, "XTG-A Two Group Three-Dimensional Reaetor Simulator Utilizing Coarse Mesh Spacing and Users Manual", XN-CC-28, Revision 5, Exxon Nuclear Company, July, 1979.
7. F.B. Skogen, "Exxon Nuclear Neutronic Design Methods For Pressurized Water Reactors", XN-75-27, Exxon Nuclear Company, June, 1975.
8. Supplement to Reference 7, September 1976.
9. Supplement 2 to Reference 7, December, 1977.
10. R.H. Kelly, "Analysis of Axial Pm*ier Distribution Limits For The Palisades Reactor at 2530 MWe"* XN-NF-78-16, Exxon Nuclear Company, June, 1978 .
  • U\P., repro<h1ct1on, tran51nutal or c11u:losure of the abova information is Sl..ibiect to the re!.tr1ct1on on the f1ru or title page of this. document.
  • -III. Conclusion Based on the foregoing, both the Palisades Plant Review Committee and the Safety and Audit Review Board have reviewed these changes and find them acceptable.

CONSlJIVIERS POWER COMPAI~Y By 0?!6 !'uwLXt-R B DeWitt, Vi~ President Nuclear Operations Sworn and subscribed to before me this 21st day of July 1981.

I Dempski, Notar Jackson County, Michigan My commission expires December l4, l983