ML20155D428

Safety Evaluation Supporting More Negative End-of-Life Moderator Temp Coefficient Tech Spec for Millstone Unit 3

ML20155D428
Person / Time
Site:	Millstone
Issue date:	09/30/1988
From:	Fetterman R, Foley J, Savage M WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19297H067	List: B13045, Forwards Nonproprietary & Proprietary WCAP-11951 & 11946, Safety Evaluation Supporting More Negative End-of-Life Moderator Temp Coefficient Tech Specs for Millstone Unit 3. Proprietary Rept Withheld (Ref 10CFR2.790) ML20155D403 ML20155D428
References
WCAP-11951, NUDOCS 8810110251
Download: ML20155D428 (53)

Text

-

Westinghouse Class 3 WCAP-11951 SAFETY EVALUATION SUPPORTING A MORE NEGATIVE EOL MODERATOR TEMPERATURE COEFFICIENT TECHNICAL SPECIFICATION FOR THE NILLSTONE NUCLEAR POWER STATION UNIT 3 September, 1988 R. J. Fetterman M. C. Savage J. V. Foley P. W. Rosenthal M

~ br o /**

APPROVED: /

~

APPkOVED: /

H. Mutyala, Manager

~ P. A. L3ftus Manager Core Design Nuclear Safety Transient Analysts I 8310110251 801005 PDR ADOCK 05000423 P

PDC HESTINGHOUSE ELECTRIC CORPORATION Comercial Nuclear Fuel Division P. O. Box 3912 Pittsburgh, fennsylvania 15230 02931:6/880920

ABSTRACT This repert proposes a relaxation of the Limiting Condition for Operation and Surveillance Requirements values of Moderator Temperature Coefficient for the end of cycle, rated thermal power condition.

Relaxation is sought in order to improve plant availability and minimize disruptions to normal plant operation, while continuing to satisfy plant safety criteria. A methodology for establishing Technical Specification end of cycle Moderator Temperature Coefficient values that are consistent with the plant safety analyses is described herein.

Specific application of the methodology to the M111 stone Nuclear Power Station Unit 3 provides Technical Specification Moderator Temperature Coefficient values which are proposed to replace the existing values.

This proposed Technical Specification EOL MTC Limiting Condition for Operation and 300 ppm Surveillance Requirements are applicable to both 4 Loop and 3 Loop (corrected to rated thermal power) operation.

1 02931:6/880920

TABLE OF CONTENTS Section Title Ease

1.0 INTRODUCTION

1-1

1.1 Background

1-1 1.2 Basis of Current EOL HTC 1-1 LCO and SR Values 1.3 Operational Considerations:

1-3 EOL HTC Tech Spec SR Value 1.4 Operational Considerations:

1-4 EOL HTC Tech Spec LCO Value 2.0 METHODOLOGY FOR M00!FYING HOST NEGATIVE HTC TECH SPEC VALUES 2-1 2.1 Conversion of Safety Analysis 2-1 HDC to Tech Spec HTC 2.2 Conservatism of the ARI to ARO 2-2 MTC Conversion 2.3 Alternative HTC Conversion 2-2 Approach 2.4 Determining SR HTC from 2-4 LC0 HTC 2.5 Benefits of the Alternative HTC 2-5 Conversion Approach 3.C HOST NEGATIVE FEASIBLE HTC APPROACH APPLIED TO MILLSTONE UNIT 3 3-1 3.1 Hillstone Unit 3 Accident Analysis 3-1 HDC Assumption 3.2 Determination of Host Negative 3-2 Feasible HTC Senu tivities 3.3 Maximum Allowed Deviations from 3-5 Nominal Operating Conditions 11 02931:6/880920 A

TABLE OF CONTENTS (cont'd.)

Section li_ tit Eagt 3.0 3.4 Overall "Delta MTC" Factor for 3-12 Millstone Unit 3 Reloads 3.5 Proposed Millstone Unit 3 3-13 Tech Spec EOL MTC LCO Value 4.0 SAFETY ANALYSIS IMPACT OF MOST NEGATIVE 4-1 FEASIBLE MTC APPROACH 5.0 DETERMINATION OF MOST NEGATIVE FEASIBLE 5-1 MTC SURVEILLANCE VALUE

6.0 CONCLUSION

S 6-1 REFERENCES R-1 APPENDIX A DETERMINATION f MOST NEGATIVE FEASIBLE MTC SENSITIVITa A-1 A.1 MTC Sensitivity to Moderator A-4 Temperature and Pressure Variation A.2 MTC Sensitivity to RCCA Insertion A-5 A.3 Sensitivity to Axial Flux A-7 (Power) Shape A.4 Sensitivity to Transient Fission A-9 Product (Xenon) Concentration A.5 Three Loop Operation A-10 til 02931:6/880920 j

LIST OF TABLES IAhlt lLtle EAge 4.1 FSAR Chapter 15 Events That Assume 4-2 a Constant 0.43 ak/gm/cc Value of HOC LIST OF ILLG;ToaTIONS Elgure

. Title Eage_

3.1 4-Loop Control Rod Locations 3-15 3.2 Expected RCCA Insertion Limits for 3-16 Hillstone Unit 3 Reload Cores A.1 Change in HTC with Increase in A-12 T-Average above Nominal T-Average A.2 Corn Average Axial Burnup versus A-13 Core Height at EOL A3 Axial Power and Moderator Temperature A-14 Versus Core Height A.4 Delta HTC versus Axial Flux Difference A-15 at EOL, HFP, ARO 1

1 iv 02931:6/880920

r I

i

1.0 INTRODUCTION

1.1 Background

'T For FSAR accident analyses, the transient response of the plant is dependent on reactivity feedback effects, in particular, the moderator temperature coefficient (MTC) and the Doppler power coefficient.

Because of the sensitivity of accident analyses results to the MTC value assumed It is important that the actual core MTC remain within the bounds of the Ilmiting values assumed in the FSAR accident analyses. While core neutronic analyses l

will have predicted that the MTC is within these bounds, the Technical j

Specifications require that the core MTC also be confirmed by measurement, as j

verification of the accuracy of the neutronic predictions.

These MTC measurements are performed:

f j

1.

At beginning of cycle, prior to initial operation above 51. rated

]

thermal power, and

)

2.

Within 7 EFPD after reaching an equilibrium boron concentration of l

300 ppm.

l a

i 1.2 Basis of Current EOL MTC LCO and SR Values l

In order to ensure a bounding accident analysis, the MTC is assumed to be at I

j its most limiting value for the analysis conditions appropriate to each j

accident.

Currently, the most negative MTC limiting value is based on EOL l

conditions (specifically with regards to fuel burnts ud boron concentration),

full power, with rods fully inserted.

Host accident analyses ute a constant moderator density coefficient (MDC) l designed to bound the MDC at this worst set of initial conditions (as well as i

i at the most limiting set of transient conditions).

This value for HDC forms the Itcensing basis for the FSAR accident analysis.

02931:6/880920 1-1

Converting the MDC used in the accident analyses to a MTC is a simple calculation which accounts for the rate of change of moderator density with temperature at the conditions of interest.

In this report, the convention followed is to discuss the moderator feedback in terms of MTC, rather than MDC.

Nevertheless, it is important to note that the accident analyses actually assume a constant MDC value, rather than making any explicit assumption on MTC.

Technical Specifications place both Limiting Condition for Operation (LCO) and Surveillance Requirements (SR) values on MTC, based on the accident analysis i

assumptions on MDC described above.

The most positive MTC LCO limit applies to Modes 1 nd 2, and requires that the MTC be less positive than the 1

specified 'imit value.

The most negative MTC LCO limit applies to Modos 1, 2, and 3, anc requires that the MTC be less negative than the specified limit value for the all rods withdrawn, end of cycle life, rated thermal power condition.

The Technical Specification SR calls for measurement of the MTC at BOL of each cycle prior to initial operation above 57. rated thermal power, in order to demonstrate compliance with the most positive MTC LCO.

Similarly, to demonstrate compliance with the most negative MTC LCO, the Technical l

Specification SR calls for measurement of the MTC prior to E0L (near 300 ppm equilibrium boren concentration).

However, unlike the BOL situation, this 300 ppm l

SR MTC value differs from the 0L LCO limit value.

Because the HFP MTC value will gradually become more negative with further core depletion and boron f

concentration reduction, a 300 pom SR value of MTC should necessarily be less l

negative than the EOL LC0 limit.

The 300 ppm SR value is sufficiently less negative than the EOL LCO limit value to provide assurance that the LCO limit will be met when the 300 ppm surveillance criterion is met.

l l

(

l 02931:6/880920 1-2 i

l.3 Operational Considerations:

EOL MTC Tech Spec SR Value It is becoming increasingly probable that reload cores will fail to meet the 300 ppm surveillance criterion associated with the E0L LCO limit.

The primary factors causing nre negative HTCs near EOL are higher core average operating temperature and higher discharge burnup.

Failure to meet the surveillance criterion does not by itself imply a failure to meet the actual EOL MTC limit stated in the LCO, but invokes the requirement that the HTC continue to be measured at least once per 14 EFPD during the remainder of the fuel cycle.

This repeated surveillance is performed to demonstrate that the actual LCO limit on EOL MTC is not violated.

The drawbacks to the current EOL MTC Tef.inical Specification are:

1.

The current and planned fuel management strategy is expected to yield MTC values which will be more negative than the existing 300 ppm surveillance criterion.

This would result in repeated MTC measurements every 14 EFPD.

In addition, the EOL HFP ARO MTC values for these anticipated designs will approach or possibly be more negative than the existing LCO limit.

l 2.

If repeated measurements are ner-sary, they can require that load swings be performed, causing temperatures to deviate from the programmed reference temperature - situations which are never preferable to nominal steady state operation.

l 3.

The repeated measurements require the resources of multiple l

operations personnel for roughly an entire shift, and require greater i

water processing for measurement via the boration/dtlution method.

Hestinghouse-designed PHR$ which conform to Standard Technica' Specification (STS) format generally feature a 300 ppm SR MTC value which il 9 pcm/'F less J

negative than the E0L LC0 limit on MTC.

Given the disadvantages of repeating the MTC plant measurements, it is logical to inquire if this difference I

between the SR value and the LCO value is overly large, and whether this conservatism would invoke repeated measurements that are unnecessary.

02931:6/880920 1-3

Examination of both plant-specific characteristics and fuel management effects on the difference between the 300 ppm HFP ARO predicted MTC and the EOL (0 ppm) HFP ARO predicted HTC indicates that the 9 pcm/*F difference applied to Westinghouse-designed STS plants is very conservative.

This implies that a failure to satisfy the 300 ppm surveillance criteria can occur, yet the actual EOL MTC value could show margin to the LCO limit.

It is concluded that relaxation of the difference between the SR limit value and the LCO limit talue should be investigated, so as to preclude unnecessary MTC testing at full power conditions.

1.4 Operational Considerations:

EOL MTC Tech Spec LCO Value Relaxation of the SR limit value may provide only temporary relief from the repeated MTC measurement situation.

With longer operating cycles and increased fuel discharge burnups, future reload core designs may eventually challenge the EOL MTC LCO limit.

The reload core design process would detect the fact that the design value of EOL MTC could exceed the Tech Spec LCO limit long before a reload core were to begin operation, but the design measures that can be taken to produce a less negative EOL HTC negatively impact the desired energy production for the reload.

The FSAR accident analyses that form the plant's licensing basis have assumed a MDC value which, when converted to a MTC at full power pressure and temperature, translates to a HFP HTC value that is more negative than the LCO limit value of the Technical Specifications.

The difference between the value of most negative MTC (most positive MDC) assumed in the accident analyses and that presented as the LCO of the Tech Specs is substantial, and offers a potential avenue for relaxation of the Tech Spec EOL HTC LCO value.

The thrust of such an effort to relax the EOL MTC Tech Spec LCO limit must continue to bound the accident analysis assumptions, and should establish a reasonable basis for the difference between the safety analysis value of most negative MTC and the Tech Spec LCO value.

02931:6/880920 1-4

i l

4 l

1 1

i q

2.0 METH000 LOGY FOR MODIFYING MOST NEGATIVE MTC TECH SPEC VALUES f

(

i 2.1 Conversion of Safety Analysis MDC to Tech Spec HTC As stated previously, the FSAR accident analyses have assumed a bounding value

(

of the moderator density coefficient (hDC) which ensures a conservative result j

for the transient analyzed.

The process by which this alcident analysis most positive MDC is transformed into the most negative MTC LCO value is stated in I

STS BASES section 3/4.1.1.3:

i j

"The most negative MTC value, equivalent to the most positive moderator

[

)

density coefficient (MDC), was obtained by incrementally correcting the MDC '

l used in the FSAR accidvnt analyses to nominal operating conditions.

These l

corrections involved subtracting the incremental change in the MDC l

j associated with a condition of all rods inserted (most positive MDC) to an

{

l all rods withdrawr condition, and a conversion for the rate of change of l

moderator density with temperature at PATED THERM /'. PONER conditions.

This value of the HDC was then transformed into the limiting MTC value."

f In the process of converting the accident analysis MDC into the Tech Spec HTC, l

j the conversion for the rate of change of moderator density with temperature at rated thermal power conditions involves conventional thermodynamic properties l

and imposes no undue conservatism on the resulting HTC value.

The additional conversion made is to correct the above MDC (HTC) value for the change j

associated with going from a condition of ARI to one of ARO.

That is, the

}

accident analysts MDC (HTC) assumes a coefficient determined for a condition l

of EOL HFP O ppm with all control aad shutdown banks fully inserted.

This

{

}

accident analysis MDC (MTC) is corrected back to the ARO condition, in order l

l to produce a Tech Spec limit which permits direct comparison against measured j

values.

The effect of the presence of all control and shutdown banks is to f

l make the MTC markedly more negative than a MTC at the ARO condition, hence j

i this conversion has a substantial Impact.

[

1 i

i 1

i l

\\

02931:6/880920 2-1 1

i

i i

i 2.2 Conservatism of the ARI to ARO MTC Conversion i

The use of a substantially negative MTC (positive MDC) value for the transient l

]

accident analyses is prudent, in that it produces a more severe result for the j

transient, which makes the analysis inherently conservative.

The drawoack to I

the ARI assumption is that when the conversion to the ARO condition is made, the resulting Tech Spec HTC value is dramatically less negative than the value i

corresponding to the transient safety calculations, and is even less negative than expected best estimate values of EOL MTC for high discharge burnup reload l

l cores.

In the worst case, maintaining the EOL MTC Tech Spec limit at its i

present value could result in requiring that the plant be placed in hot shutdown when, in fact, there exists substantial margin to the moderator I

a coefficient assumed in the accident analyses.

Such a situation is i

unnecessarily restrictive, and results primarily from the ARI to ARO j

adjustment made between the accident analysis MDC value and the Tech Spec HTC l

l l

value.

l l

In addition to being unnecessarily restrictive, the HFP ARI assumption is j

l inconsistent with Tech Spec requirements for allowable operation, wherein i

I l

shutdown banks are not permitted to be inserted during power operation and j

)

control banks must be maintained above their insertion 11mi'J.

1 1

l 2.3 Alternative MTC Conversion Approach t

l t

l If the ARI to ARO basis for converting from the accident analysis MDC value to

(

j a Tech Spec LCO MTC value is overly restrictive, what would constitute a more

[

)

meaningful, yet inherently conservative basis? The concept herein proposed as l

an alternative to the ARI to ARO conversion is termed the "Most Negative l

Feasible MTC" approach, This approach maintains the existing accident analysis assumption of a bounding value of moderator coefficient, but offers i

l l

an alternative method for converting to the Tech Spec LCO HTC value.

[

i J

t i

I I

f i

l l

i i

t 02931:6/880920 2-2 L

f

d The Most Negative Feasible MTC approach seeks to determine the conditions for which a core will exhibit the most negative MTC value that is consistent with operation allowed by the Tech Specs.

As an example, the Most Negative Feasible MTC appreach would not require a conversion assumption that all rods be fully inserted at HFP conditions, but would require a conversion assumption that all control banks are inserted the maximum amount that Tech Specs permit, so as to make the calculated EOL HFP MTC more negative than it would be for an unrodded core.

The Most Negative Feasible MTC approach determines EOL MTC sensitivity to tnose design and operational parameters that directly impact MTC, and attempts i

to make this determination in a such a manner that the resulting sensitivity j

for one parameter is independent of the assumed values of the other parameters.

As a result, parameters which are mutually exclusive but permissible according to the Tech Specs (such as an assumption of full power operation and an assumption of no xenon concentration in the core), and which serve to make MTC more negative, will have their incremental impacts on MTC combined to arrive at a conservative and bounding condition for the most i

negative feasible MTC.

The paramete'es which are variable under normal operation, and which affect HTC are:

l soluble boron concentration in the cootant moderator temperature and pressure RCCA insertion axial flux (power) shape i

transient fission product (xenon) concentration e

The Most Negative feasible MTC approach examines each parameter separately, and assesses the impact of variation in that parameter on E0L MTC. The l

assessment is performed for multiple core designs that feature combinations of I

fuel design, discharge burnup, cycle length, and operating temperature expected to envelope future core designs of the plant of Interest.

l l

r i

i i

02931:6/880920 2-3

I e

When the assessment is complete, the MTC sensitivity associated with each of the above parameters has been identified.

One then determines the maximum deviation from "nominal" conditions (AR0, HFP, equilibrium xenon, Tavg on the reference temperature program) that the Tech Specs permit, and multiplies that deviation by the appropriate MTC sensitivity to arrive at a "delta MTC" factor associated with the parameter.

4 J

For example, suppose it is determined that MTC becomes 1 pcm/'F more negative for l

each 1*F increase in core average operating temperature above nominal (the MTC "sensitivity" is -I pcm/'F/'F).

If the Tech Specs permit a maximum increase in Tavg of 4*F above nominal core Tavg then the moderator temperature "delta MTC" j

factor is:

l 1

(-1 pcm/*F/*F) x 4'F = -4 pcm/'F.

Bounding "delta MTC" factors are determined in this way for each of the above parameters, and these factors are then added to arrive at an overall bounding l

I "delta MTC" factor.

This overall "delta MTC" factor states how much more negative the MTC can become, relative to the nominal EOL HFP ARO MTC value, for normal operation scenarios permitted by the current Tech Specs.

The conditions of moderator temperature, rod insertion, "non, etc., which defined the Most Negative Feasible MTC condition become the conversion proposed as a replacement for the ARI to ARO conversion of the current MTC Tech 5...

The conversion for the Most Negative Feasible MTC condition is applied in the same way that the current ARI-to-ARO conversion is applied, in order to arrive at an EOL ARO HFP HTC Tech l

Spec limit that remains based on the accident analysis MDC assumption.

1 i

l 2.4 Determining SR MTC from LCO HTC l

1 l

Under the Most Negative Feasible MTC approach, the 300 ppm surveillance value is l

l determined in the manner curren*1y stated in the BASES for STS plant MTC Tech Specs:

h i

"The HTC surveillance value represents a conservative value (with f

e

)

corrections for burnup and soluble boron).t a core condition of 300 ppm

)

equilibrium boron concentration and is obtained by making these corrections I

f to the limiting MTC LCO value."

i.

i 02931:6/880920 2-4 l

i

i

]

That is, the 300 ppm surveillance value is derived by making a conservative adjustment to the E0L ARO HFP MTC limit value that accounts for the change to MTC with soluble boron and burnup.

Plant-specific examination of the difference between 300 ppm HFP MTC and EOL (0 ppm) HFP HTC suggests that a sw. aller correction is justified than the 9 pcm/'F which has historically been applied to Westinghouse-designed STS plants.

2.5 Benefits of the Alternative HTC Conversion Approach The Host Negativa Feasible HTC approach is considered to be superior to the ARI-to-ARO conversion specified by current STS plant Tech Specs for the following reasons:

1.

The Most Negative Feasible MTC approach does not require an unduly positive 300 ppm survelliance value that would result in repeated MTC surveillance measurements, These repeated measurements are j

undesirable in that they entail perturbations to normal reactor operation.

1 1

2.

The Most Negative Feasible MTC approach does not alter the FSAR l

transient accident analysis bases or assumptions, and hence, does not affect the accident analysis conclusions.

It retains the concept of a conversion between the accident analysis MDC assumption and the Tech Spec LCO MTC value that assures that the plant cannot experience j

i a MDC which is rnoce severe than that assumed in the accident analyses.

l j

3.

The Most Negative Feasible HTC condition is a conservative but reasonable basis to assume for a HTC vtlue of the reload core prior i

to a transient, and is consistent with operation as defined by other i

sections of the Tech Specs (whereas the ARI-to-ARO conversion is l

overly conservative and makes assumptions which are inconsistent wtth other sections of the Tech Specs).

I l

02931:6/8809:0 2-5

i d

Additionally, the Most Negative feasible HTC approach retains the "built-in j

safeguard" of a requirement for a 300 ppm surveillance measurement to be l

performed in order to verify that the reactor is operating in a regime that is bounded by the accident analysis input assumptions.

i I

1 l

I i

i I

t l

t l

)

i i

a i

f j

1 l

f d

1 L

i 4

4 n

I l

t l

i l

l

)

f I

l 02931:6/880920 2-6

1 3.0 MOST NEGATIVE FEASIBLE MTC APPROACH APPLIED TO MILLSTONE UNIT 3 3.1 Millstone Unit 3 Accident Analysis HDC Assumption The FSAR accident analyses upon which the Tech Spec E0! HFP LCO MTC limit is

+

l based have assumed bounding values of moderator density coefficient in order

(

to ensure a conservativ>. simulation of plant transient response for the Hillstone Nuclear Power Station Unit 3.

For those transients for which analysis results are made more severe by assuming maximum moderator feedback, a moderator density coefftetent (MDC) of 0.43 ak/gm/cc has been assumed to q

exist throughout the transient.

l i

When discussing the Tech Spec EOL LCO limit on moderator feedtsack, it is simpler to talk in terms of moderator temperature coefficient (MTC) than HDC.

For this reason, the Millstone Unit 3 accident analysis MDC assumption o' l

0.43 ak/gm/cc is converted to its equivalent MTC.

This conversion depends l

on the density change-to-temperature change relationship which prevails for the conditions of interest.

For this discussion, the conditions of interest

]

are the core temperature and pressure (hence, density) experienced under l

normal operation at which the MDC assumes its most extreme (positive) value.

l These temperature and pressure condit'. N are the Hillstone Unit 3 rated l

1 thermal power (RTP), full flow nominal operating condttions of 590.5*F acd 2250 psia, respectively.

)

At these nominal RTP operating conditions, the accident analysts MDC value of 0.43 ak/gm/cc is equivalent to a HFP MTC of -55.47 pcm/'F.

For simplicity,

{

this value of MTC will often be referred to as the "accident analysis MTC", in the discussion which follows. However, it should be remembered that the I

applicable accident analyses actually assume a constant MDC value of f

L 0.43 ak/gm/cc and make no explictt assumption about MTC.

1 I

(

l 02931:6/880920 3-1 i

3.2 Determination of Most Negative feasible MTC Sensitivities As stated previously, there are a limited number of core operational parameters that directly affect MTC and are variable under normal core operation. The list of parameters is as follows.

soluble boron concentratton in the coolant moderator tempera b.e and pressure RCCA insertion axial flux (power) shape transient fission product (xenon) concentration The radial flux (power) shape can also vary under normal core operation and will affect MTC.

However, the operational activities that directly affect radial power shape do so through withdrawal or insertion of control rods and through xenon concentration; therefore, the impact of radial flux distributton variation on MTC will be an implicit part of the HTC sensitivity to these other parameters.

Soluble boron concentratie i is certainly variable under normal core operation.

However, it eliminated as a source of sensitivity for this analysts.

This is because the EOL HFP ARO MTC Tech Spec limit value is assumed to be essentially a O ppm limit by virtue of the definition of EOL.

The most negative HTC value will always occur at a boron concentration of 0 ppm, and ther efore, a O ppm boron concentration is assumed as the basis of the E0L HTC Tech Spec Ilmit under the Most Negative Feasible MTC approach.

For the remaining parameters, sensitivity analyses were performed by perturbing the parameter of interest in such a way as to induce a change from its nominal E0L value, and then performing a MTC determination with the parameter held in the perturbed state.

A further perturbation was induced and the MTC calculation repeated.

This sequence was repeated until sufficient HTC data values were generated to reliably determine the trend of MTC change wtth 1

{

variation in the value of the independent parameter, i

l 02931:6/880920 3-2 1

I It should be noted that the discussion regarding the MTC sensitivities is based on 4 Loop operation unless specifically noted.

It is shown in Sections 3.3 and A.5 that the proposed Technical Specification will be applicable to 3 Loop and 4 Loop operation.

The LCO snd SR values of most negative MTC are for the Rated Thermal Power, ARO condition 1 if a measurement is taken at a different condition, it should be adjusted to RTP, ARC conditions prior to comparison with the Technical Specification value.

In order to establish trends in MTC that are appropriate and bounding for the t

Millstone Unit 3 reload cores, these sensitivities were determined for five different reload cores.

These cores exhibit design features that are expected to be incorporated in future Millstone Unit 3 reload cores (such as increased discharge burnup, longer cycles, and advanced fuel product features).

I i

A brief description of the five reload ccre designs follows:

RELOAD A:

This core is the currently operating reload (Cycle 2) of Millstone Unit 3.

It utilizes the Westinghouse 17xl? standard rod diameter fuel design, feeds 84 assemblies in a low leakage loading pattern (L3P) and assumes a region average discharge burnup of 36000 MHD/MTU.

This reload also has 448 pyrex glass bas and a nominal cycle length of 15800 MWD /MTV.

The control red absorber material is hafnium.

The nominal core everage temperature assumed in this analysis is 590.5'F for 4 Loop operatim and 582.7'F for 3 Loop operation.

This is the first cycle for implementation of the revised Tech Spec.

RELOAD B:

This is the proposed Cycle 3 reload design for Millstone Unit 3 using an L3P design feeding 76 assemblies with axial blankets, approximately 4500 part-length IFBA, high enrichments, and a region average dischaege burnup of 45000 MHD/HTU.

The cycle length is 16500 MWD /MTU and all other core operating parameters are assumed to be the same as Reload A.

02931:6/880920 3-3

RELOAD C:

This reload is a conceptual equilibrium 24 month (21830 MHD/MTV) cycle design with advanced fuel product features including the features in RELOAD 8 plus Zr grids and Intermediate Flow Mixer (IFM) grids. A thimble plug removal analysis is also assumed which increates the core bypass flow and raises the core average modcrator temrerature to 591.3'F for 4 Loop operation and 583.5'F for 3 Loop operation.

RELOAD D:

This reload core is an annuti cycle L3P desiga using the Westinghouse 14 foot length 17x17 standard fuel rod diameter design.

The discharge burnun for this design is necessarily low since this is the first reload core for this particular plant.

This design has 416 pyrex glass bas and uses hafnium as the control rod absorber material.

The nominal core average moderator temperature of 596.5'F at full power.

RELOAD E:

This reload core is a Westinghouse 4 loop plant with 17x17 standard diameter fuel. An L3P design is used for a cycle length of 17000 MHD/MTV and a regicn average discharge burnup of 38000 SHD/MTU.

This design also feeds 84 assemblies and uses 1728 pyrex glass bas.

The control rod absorber material is silver-indium-cadmt.um.

The nominal core average moderator temperature is 575.3'F.

Reloads /. B. D. and E were used in the calculation of the sensitivities to the four parameters described previously.

To provide additional information regarding the effect of extremely long cycles and high discharge burnups on the MTC sensitivity to moderator temperature and pressure and xenon concentration. Reload C was also used in the determination of those sensitivities. Given the nature of Reload C. it is expected that the impact on the sensitivities to RCCA insertion and axial flux shape would be bounded by the other reloads.

02931:6/880920 3-4

All five of these reload cores feature the same control bank configuration I

that will be used in Millstone Unit 3 reload cores, shown in Fiqure 3.1.

1 The core neutronic models of these five reload cores were derived using

)

standard Westinghouse procedures and computer mathods.

The ARK code, which I

has evolved from the LEOPARD < " and CINDER (2) codes, was used to perform the fast and thermal spectrum calculations and is the basis for all cross sections, depletion rates, and reactivity feedback models. ANC( ' and 54)

PALADON

, nodal analysis theory codes used in two and three dimensions, I

were used for core neutronte calculations to determine MTC sensitivity for the l

five reload cores. APOLLO, an advanced version of PANDA (5), was used as an axial neutronic model of the reload cores to determine MTC sensitivity to j

varying axial flux sh?pe.

The neutronic calculations and evaluations performed for the five reload core designs established HTC seriitivities for each of the parameters listed i

above.

The detailed description and results of this analysis are provided in l

Appendtx A.

)

3.3 Maximum Allowed Deviations from Nominal Operating Conditions l

i j

The concep.t of maximum allowed deviation from nominal operating conditions is j

employed to determine the extent to which reactor parameters can vary under i

normal operation so as to cause MTC to become more negative.

This comM nation f

of parameter statepoints then defines the worst allowable initial condition

[

]

for a transient that employs a most negative HTC (most positive MDC) l assumption.

It is also necessary to demonstrate that the parameter changes l

that occur throughout the transient do not result in a MTC value which is f

unbounded by the moderator coefficient assumption used in the accident a

j analysts. The adequacy of the consta t HDC accident analysts assumption to bound MTC values that occur throughout the transient is examined in Section 4.

l i

l t

j l

.l 02931:6/880920 3-5 i

The bases for the maximum allowed deviation from nominal operating conditions are Technical Specifications that limit the extent of moderator temperature increase, RCCA insertion, and axial power skewing.

The deviations permitted by present Millstone Unit 3 Tech Specs and possible future perturbations to those Tech Specs values are discussed in the following sections:

tMdentor TempenturemLEnune_De31Ltioni Tech Spec 3.2.5 estabitshes the LCO values of th DNR parameters reactor coolant system Tavg and pressurizer pressure.

For Millstone Unit 3, Tech Spec 3.2.5 states for 4 Loop operation a minimum allowable indicated pressurtzer pressure of 2226 psia, and the maximum allowable indicated RCS average temperature of 591.2*F.

These values have accounted for instrumentation uncertainties of 21 psi and 2*F.

Therefore, the maximum allowable RCS temperature and the minimum allowable pressure assumed in the safety analysts are 593.2'F and 2205 psta, respectively.

Because the current nominal design RCS temperature for the Millstone Unit 3 plant is 587.I'F, the 593.2*F safety analysis limit represents a 6.1'F maximum allowable Tavg increase over nominal conditions.

The current ncminal design pressure for the Millstone Unit 3 plant is 2250 psia; therefore, the 2205 psia safety analysis limit represents a 45 psi maximum reduction from nominal system pressure.

Note that these pressure and temperature deviations from nominal HFP values match those deviations from nominal conditions which are assumed as initial conditions in the Hillstone Unit 3 FSAR transient accident analyses.

Because Tech Spec 3.2.5 limits deviations from nominal condition RCS temperature and pressure to 6.1'F and 45 psi, respectively, it also indirectly places a limit on the maximum allowable deviation of RCS moderator density from nominal.

These maximum temperature and oressure deviations are applied to the HTC Sensitivity to temperature and pressure, which is described in Appendix A, to obtain a "delta MTC" factor associated with RCS moderator temperature and pressure deviations from nominal.

The resulting "delta MTC" is [

]C pcm/'F.

02931:6/880920 3-6

RCCA_JniertionJnittion The nominal condition assumption for RCCA placement is complete withdrawal (ARO).

This assumption is underscored by the requirement in Tech Spec 3.1.1.3 that the LCO value of EOL HTC is for the "all rods withdrawn" condition.

Because some RCCA insertion is allowed during full power operation, and because RCCA insertion will generally cause HTC to be more negative than it would be otherwise, the RCCA insertion deviation is simply that maximum allowable RCCA insettion permitted by the Tech Specs.

Tech Specs 3.1.3.5 and 3.1.3.6 place limits on allowable RCCA insertion.

Tech Spec 3.1.3.5 precludes Shutdown RCCA insertion in Modes I and 2, and Tech Spec 3.1.3.6 limits Control Bank insertion via the Rod Insertion Limits (RILs).

Control rods can be inserted as a function of power level according to the R!ls, and all RCCAs can be inserted at HZP coincident with reactor trip. With greater RCCA insertion, MTC becomes more negative relative to the ARO HTC, all other parameters being held equal.

However, Tech Specs do not allow all other parameters to be held equal. With deeper RCCA insertion, power must be reduced and Tavg will be reduced accordingly. The reduction in Tavg serves to make the MTC more positive, and at EOL 0 ppm conditions, this positive Tavg effect will entirely offset the negative RCCA effect on HTC.

For example, for the Hillstone Unit 3 first reload core, complete ir.sertion of both Control Banks D and C at EOL 0 ppm HFP c,ditions (a condition not permissible under normal operation) will make the MTC (

)***C pcm/'F more negative than the ARO HTC.

However, in going from a nominal HFP Tavg to a nominal HZP Tavg at EOL and 0 ppm, the MTC for this same core becomes

(

J+a c pg,j.F more positive.

This (

lC pcm/*F more positive component of the HTC that results from the moderator temperature (density) change in going from HFP to HZP will not only compensate for the negative HTC component associated with 9CCA insertion permitted by the RILS, but will also more than compensate for the negative MTC component that arises from total RCCA insertion with trip.

02931:6/880920 3-7

l In this respect, the Millstone Unit 3 first reload core design is typical of reloads for Hestinghouse-designed PHRs.

Because the rate at which decreasing moderator temperature makes MTC positive exceeds the rate at which allowable RCCA insertion makes MTC negative, the most negative MTC situation will always exist at HFP Tavg with RCCAs inserted to the extent allowed by the HFP insertion limits.

For this reason, the maximum RCCA deviation from nominal conditions allowable by the Tech Specs needs to be assessed at only the HFP condition.

Figure 3.2 shows the RCCA insertion limits to be used for Millstone Unit 3 reload cores.

These are typical insertion limits for Hestinghouse-designed 4 loop PHRs with the five control rod lead control bank of Figure 3.1. Figure 3.2 shows that at full power the lead control bank can be inserted to a depth of 164 steps withdrawn.

Hcwever, strict application of these current R!Ls in determining the "delta MTC" factor associated with RCCA insertion may prove to be restrictive if minor changes to the RILS become necessary in the future.

For this reason, the HFP RCCA insertion assumed for this analysis is

(

lC steps withdrawn.

This additional insertion is expected to bound minor RIL adjustment which may be necessary for optimizing core performance characteristics of future Millstone Unit 3 roloads. However, applicability of

{

the ultimate EOL MTC LCO value derived from these sensitivities will be confirmed on a cycle-by-cycle basis as part of the reload design, process, therefore, a RIL adjustment to lead control bank insertion beyond [

lC steps withdrawn will not necessarily invalidate the revised EOL MTC LCO value.

This limiting H7P RCCA insertion of (

]+a c steps withdrawn forms the basis of the determination of MTC sensitivity to HFP RCCA insertion, which is described in Appendit A.

The resulting "delta MTC" factor associated with RCCA insertion was determined to be (

3**'Cpcm/*F.

Avi.a Lflui jPowerL 5hape_Dev.tation As indicated earlier, MTC is affected by the asial flun shape which exists in the core, primarily as a result of the influence that ti,s axial flut shape has

/

on the rate ac which the trederator is heated as it moves up the core.

The detailed @ ape itself is not so important, but rather the "balance" of the 02931:6/880920 3-8

flux shape, in terms of how much moderator heating occurs in the lower half of the core versus the upper half of the core.

The influence which axial power shape has on HTC can, therefore, be captured by quantifying this axial flux "balance", and this balance is best quantified by the core's Axial Flux Difference (AFD).

The discussion of the HTC sensitivity to axial flux (power) shape presented in Appendix A establishes that the more negative the AFD becomes, the more negative the HTC will become. The axial flux (power) shape deviation is, therefore, determined by how negative the AFD is allowed to become under normal full power operating conditions.

The initial Hillstone Unit 3 reload core employs a CAOC Tech Spec which sets the allowable full power AFD limits at +3% and -12%.

To assign a "delta HTC" factor attributable to axial flux shape, one need only examine the HTC effect associated with the -12% "deviation" from a most negative expected target AFD value of approximately (

)*8'C%.

However, to account for possible future changes in the most negative HFP AFD limit, an AFD value of (

]* * ' C. i s 7

selected as the basis of the axial flux (power) shape deviation.

This

(

)+a c% AFD deviation is applied to the HTC sensitivity to axial flux (power) shape, which is described in Appendix A, to obtain a "delta HTC" factor associated with AFD deviation from 4 perfectly balanced axial flux shape.

The resulting "delta HTC" factor is (

lCpcm/'F.

Ira n sie ntli nioLErod u cLGenonLCont e n t ra tio tDex11 tion Xenon is the most significant transient fission product in terms of effects on core reactivit/ and flux distribution; therefore, its possible impact on HTC are investigated to compute the final "delta HTC" factor to include in the Most Negative Feasible HTC approach. While Tech Specs plac6 no limitations on either xenon distribution or overall concentration, the AFD limits discussed above, in effect, place a limitation on the amount of axial xenon skewing that can occur, and the physics of xenon bulldup and decay place practical limits on the concentration.

02931:6/880920 3-9 a

Because axial xenon distribution directly 'mpacts axial flux shape, this 1

aspect of xenon effect on MTC is implicitly included in the axial flux (power) shape deviation discussed above.

Haat remains to be quantified is the impact of the overall xenon concentration in the core.

Taking the EOL HFP ARO equilibrium renon concentration to be the nominal xenon condition for the core, it was determined for low leakage core designs of the

}

type anticipated for Millstone Unit 3 reloads, the MTC would become enore negative with a reduced xenon concentration.

Accordingly, the most negative HTC results when there is no menon in the core.

It was established in the discussion on moderator temperature and pressure deviation and on RCCA insertion deviation, that the condition for the most negative HTC requires maximum allowable temperature (minimum allowable density) and, therefore, occurs at full power conditions.

While the assumption of achieving full power operation with no senon in the core is certainly a conservative assumption, the possibility of steady power escalation after an extended shutdown period presents a reasonable scenario for full power operation with a comparatively low xenori concentration in the core.

For this reason, the "xenon deviation" to be used in conservatively determining the "delta HTC" factor attributable to transient fission product is a change from HFP ARO equilibrium xenon to no menon in the core.

The resulting "delta HTC" factor is (

pa.c cm/*F, Ihree_ Loco Opentien l

For this report, unless otherwise stated, the mp imum 3 Loop poner level assund is 751 rated thermal power (RTP).

This is conservative since the current Tech Specs limit 3 Loop operation to 65% RTP and, as discussed earlier, a reduction in core power will make the HTC fore positive, for a corresponding nominal operating condition of 582.7'F and 2250 psia, the accident analysis MDC o? 0.43 ak/g/cc is equivalent to an HTC of -52.46 pcm/'F, or a change of 3.01 pcm/*F frce the equivalent 4 Loop safety analysis HTC.

It has been observed for Reload A that the E0L 3 Locp 75% RTP AR0 HTC is

(

pa,c penj.F more positive than the correspending EOL 4 Locp HFP ARO 02931:6/880920 3-10 I

7 HTC.

The (

J+a, cpg,j.F change in the calculated MTCs between 3 Loop and 4 Loop is larger than the [

]+a. cpg,j.F reduction of the safety analysts 1470.

Therefore, demonstrating tht.t the 4 Loop total "delta-HTC" is applicable to 3 Loop operation will in turn imply that the Tech Spec determined for 4 Loop operation is also applicable to 3 Loop operation.

The change in HTC due to moderator temperature and pressure deviations for 3 Loop oceration will be smaller than the corresponding 4 Loop derivative.

This is because the nominal 3 Loop core aversgo moderator temperature is lower than the 4 Loop core average moderator temperature and the lower moderator temperature provides a smaller change in moderator density for a given change in temperature.

The 3 Loop change in HTC as a function of the xenon concentration will be smaller than the corresponding 4 t. cop change due to a reduction in the equilibrium xenon concentration during 3 Loop operation, This reduction in the change of the HTC is on the order of ( )+a.c. as discussed in Section 7

A.5 of this report.

The change in HTC due to RCG insertion.to the R!L will be larger for 3 Loop operation than for 4 Loop operation.

This is due to the deeper D bank insertion allowed by the 3 Loop R!L and a:nounts to the 3 Loop delta-tiTC due RCCA insertion being about (

3+a c pcm/'F more negative than the N Loop value.

The delta-HTC as a function of axial offset (AO) for 3 Loop operation is smaller than the 4 Loop value.

The calculated value for reload A Loop operation is about [

]***C of the 4 Loop delta-HTC used in setting the Tech Spec LCO, or about (

)*8'C pcm/'r, This is because the 3 Loop total temperature rise is less than the 4 Loop temperature rise.

In sumary, the 3 Loop delta-MTC due to RCCA insertion is more negative than the 4 Loop delta-MTC by (

]+a.c pg,f.F, but the 3 Loop delta-MTC as a function of axial offset is more positive than the 4 Loop delta-HTC by

(

J'3'C pcm/'F.

Also the delta-MTCs due to menon and moderator denstty l

02931:6/880920 3-11

will also be slightly more positive for 3 Loop operation.

Therefore the 3 Loop total delta-MTC is more positive than the 4 Loop total delta-HTC and is therefore bounded by the 4 Loop value.

3.4 Overall "Delta MTC" Factor for Hillstone Unit 3 Reloads The preceding section has concluded that the most adverse operation possible, in terms of achieving the most negative EOL HTC under current and proposed Millstone Unit 3 Tech Specs, would feature the following values of key parameters:

- Core Moderator Temperature:

6.1*F above HFP nominal

- Core Moderator Pressure:

2205 psia

- HFP RCCA insertion:

(

J+a,c steps withdrawn

- HFP most negative AO:

C lC %

- HFP xenon concentration:

0%

When these maximum allowable deviations from a nominal condition of EOL HFP ARO, with equilibrium xenon, and 0 ppm boron are applied to the individual parameter sensitivities discussed in Appendix A, the overall "delta MTC" factor is determined.

This overall factor for Hillstone Unit 3 15 computed as follows:

Core Moderator Temperature and Pressure Factor:

(

)+a,c pc,f.7 HFP RCCA Insertion Factor:

(

)***C pcm/*F Axial Flux (Power) Shape Factor:

(

)*"'C pcm/'F Xenon Concentration Factor:

t

)+a.c_ccmL'E Overall "Delta MTC" rattor:

(

)*"'C pcm/*F The interpretation of this overall ' delta MTC" f actor is as follows.

The Tech Spec LCO value of EOL HTC is based in the explicit conditions of unrodded full power operation.

This is an appropriate condition for performing a MTC experiment and obtaining results that can be meaningfully compared to design predictions.

It is not, however, the condition under which the HTC can 02931:6/880920 3-12

b achieve its most negative value under normal operation scenarios permitted by the Tech Specs.

The conservative "delta HTC" formulation has concluded that the actual core HTC can be as much as (

)**'C pcm/'F more negative than the EOL HTC LCO value defined by the Tech Specs.

The individual components of this (

)*"'C pcm/'F overall "delta HTC" factor h4ve been determined on a conservative basis and are evpected to bound the values predicted or Hillstone Unit 3 reload cores in the future. Hhtte an individual component could conceivably exceed the value cited above, such an occurrence would not invalidate the Most Negative Feasible HTC approa h, as long as the total of all the components remains bound by the (

)+a,c pcm/*F overall "delta MTC factor, Implementation of a revised EOL HTC Tech Spec based on the Most Negative Feasible HTC approach will require that any reload core's overall "delta MTC" factor be demonstrated to be bounded by the licensing basis value, rather than specifically addressing the individual components.

Such validation of the Most Negative Feasible HTC approach on a cy le-specific basis would be performed as part of the reload core design process described in Reference 6.

3.5 Proposed Hillstone Unit 3 Tech Spec EOL HTC LCO Value As was pointed out in Section 3.1, Hillstone Unit 3 FSAR accident analyses have assumed a HDC value which, when converted to a HTC at nominal HFP conditions, is equivalent to a MTC of -55.47 pcm/'F.

At no time may the actual core be allowed to experience a MTC more negative than -55.47 pcm/'F, as this would invalidate an assumption of the accident analyses.

The Host Negative Feasible HTC approach guarantees that such a situation will not occur by subtracting from this -55.47 pcm/'F HTC value the (

)+4 c pg,f.7 "delta HTC" factor determined for Hillstene Unit 3.

The resulting value of

(

)+4.c pg,j.7 g3,,he Tech Spec EOL LC0 value of HTC under the Host Negative Feasible HTC approach.

As an additional measure of conservatism, this value is further increased to -47.5 pcm/'F, and proposed as the EOL HFP ARO Tech Spec HTC LCO value for Hillstene Unit 3 reload cores, replacing the current LCO value of -40 pcm/'F.

02931:6/880920 3-13

1 1

The -47.5 pcm/*F proposed limit provides relief over the -40 pcm/'F limit l

associated with the current Tech Spec ARO-to-ARI conversion requirement, yet still represents a conservative formulation.

The scenario of deep RCCA insertion, coupled with high Tavg low system pressure, and no xenon, i

represents a compounding of worst case events which can be considered Independent, yet are not treated as such in the Most Negative Feasible MTC

]

formulation.

Determination that the core MTC is less negative than l

-47.5 pcm/'F at EOL HFP ARO conditions provides assurance that the assumption l

on initial condition MTC made in the plant accident analyses ismains bounding, j

]

Additional assurance that the MTC (MDC) will not b)come more limiting at any time during a transient is also needed, in order to demonstrate that the i

accident analysis conclusions remain valid.

This additional assurance is the l

primary subject of Section 4.0.

I I

)

l i

1 l

1 l

I t

1 1

l i

l l

c l

I t

I 1

1 i

4 I

f 02931:6/880920 3-14 f

R P

N M

L K

d H

G F

E D

C B

A 180*

1 2

B C

B 3

4 0

0 5

A A

6 B

C A

C B

7 8

90' A

D A

C 270'

~

9 10 B

C A

C B

11 A

A 12 D

D 13 14 8

C B

15 0'

FIGUR E 3.1 4 LOOP CONTROL ROD LOC ATIONS 3-15

(FULLY WITHDRAWN)

(028.226)

(0.78.228)

/

200

- BANK B Z

(0.Q164)

E 150

/

b OANK C i

/

100 o

f O

BANKD 50 (0.0,50)

(028,O)

O O

0,2 0.4 0.6 0.8 1.0 (FULLY INSERTED)

FRACTION OF RATED THERM AL POMR FIGURE 3.2 EXPECTED RCC A INSERTION LIMITS FOR MILLSTONE UNIT 3 RELO AD CORES 3-16

4.0 SAFETY ANALYSIS IMPACT OF MOST NEGATIVE FEASIBLE MTC APPROACH The accident analyses conservatively model the various reactivity coefficients to produce a bounding analysis.

As discussed in Section 3.1, the applicable analyses assume a constant MDC of 0.43 ak/gm/cc to bound the predicted mcderator reactivity insertion.

The events which assume this value for EOL MDC are listed in Table 4.'

.yeld be noted that the Millstone Unit 3 FSAR accident analyses include consideration of both 4 loop and 3 loop operating conditions.

The Most Negativa Feasible MTC approach determines the conditions for which a core will exhibit the most negative MTC value that is consistent with operation allowed by the Tech Specs.

Thus, the value for the Most Negative Feasible MTC provides the basis for a conservative initial condition assumption.

Changes in the parameters identified in Section 2.3 could take place during a transient in such a way as to make the MTC more negative than that allowed under normal operation.

However, the most adverse conditions seen in these events will not result in a reactivity insertion that would invalidate the conclusions of the FSAR accident analyses.

Therefore, the 0.43 ak/gm/cc assumption used as the basis for the Most Negative Feasible MTC Tech Spec will not change.

As discussed in Reference 6, the reactivity coefficients assumed can have a strong influence on accident analysis results.

Since the moderator coefficient can be affected by a reload, the conservative nature of the accident analysis assumption must be confirmed on a cycle-specific basis using the methodology discussed in Reference 6.

This includes verification that the most adverse accident conditions discussed above do not invalidate the conservative nature of the accident analysis assumption. Inis process ensures the ability to verify that the applicable safety limits are met for each reload design and, consequently, that the Tech Specs are met.

02931:6/880920 4-1

TABLE 4.1 FSAR Chapter 15 Events That Assume A Constant 0.43Ak/gm/cc Value of MDC 15.1.1 Feedwater System Malfunctions that Result in a Decrease in Feedwater Temperature 15.1.2 Feedwater System Malfunctions that Result in an Increase in Feedwater Flow 15.1.3 Excessive Increase in Secondary Steam flow 15.2.2 External-Loss of Electrical Load 15.2.3 Turbine Trip 15.2.8 Feedwater System Pipe Break 15.4.2 Uncontrolled Rod Cluster Control Assembly Bank Hithdrawal at Power 15.4.4 Startup of an Inactive Reactor Coolant Pump at an Incorrect Temperature 4

02931:6/880920 4-2

1 l

5.0 DETERMINATION OF MOST NEGATIVE FEASIBLE MTC SURVEILLANCE VALUE Section 1.3 pointed out the potential conservatism in the separation of 9 pcm/*F between the Tech Spec 300 ppm HTC SR value and the E0L HFP AR0 MTC LCO value.

Typical 17x17 reload designs exhibit a predicted difference between the 300 ppm HFP design MTC and the EOL HFP ARO design MTC which is much less than 9 pcm/'F. However, in order to justify the use of a value which is smaller than 9 pcm/'F for a given plant, the specific design predictions of the plant must be examined.

Design predictions for the initial Millstone Unit 3 reload core were reviewed in order to determine the difference between the predicted 300 ppm HFP MTC and the predicted E0L HFP MTC.

The difference was found to be [

]+8'C pcm/'F.

In reviewing the differences between predicted 300 ppm and E01. MTC values for the reload cores of 4 loop plants which are similar to Millstone Unit 3, an important trend was discerned.

It was observed that the higher core average enrichments associated with increasing discharge burnup' tend to decrease the magnitude of MTC difference.

As the Millstone Unit 3 reload cores will ner.essarily increase fuel discharge burnup levels beyond that of Cycle 2, it is anticipated that the difference between the HFP 300 ppm MTC and the HFP EOL MTC will become less in the future.

This [

]**'C pcm/*F should, therefore, bound the 300 ppm MTC-to-EOL MTC differences for future Millstone Unit 3 reload corce The proposed Ter:. Spec SR value for Hillstone Unit 3 reload cores is -40.0 pcm/*F.

This value is 7.50 pcm/'F less negative than the EOL LCO MTC value proposed in Section 3.5.

The 7.50 pcm/*F was chosen to bound the maximum

(

1+a,c pcm/'F difference predicted for the Millstone Unit 3 Cycle 2 reload ccre, yet afford relief from the 9 pcm/*F difference applied by the current Tech Specs.

While the -40.0 pcm/*F 300 ppm Tech Spec SR value is expected to be bounding for future Millstone Unit 3 fuel management, the validity of this SR value will be confirmed on a cycle-by-cycle basis, as part of the reload core design process described in Reference 6.

02931:6/880920 5-1

6.0 CONCLUSION

S The present Millstone Unit 3 Technical Specification values of -40 pcm/'F for the EOL HFP ARO MTC LCO and -31 pcm/*F for the 300 ppm HFP AR0 SR conservatively reflect the FSAR accident analysis MDC assumption, but are considered to be overly restrictive by potentially requiring repeated deviation from nominal plant operation. An alternative adjustment procedure is proposed which is based on a conservative determination of the extent to which a nominal EOL HFP ARO MTC value can be made more negative under the most extreme values of certain operational parameters that are permitted by othe.-

Tech Specs.

This Most Negative Feasible MTC approach assumes that these largely independent extreme situations occur simultaneously, and in the worst case, serve to make the E0L HFP MTC C

]+a,c pcm/'F more negative than it would be at nominal conditions. When this value is subtracted from the HTC equivalent of the accident analysis assumed MDC value, the resulting HTC is

[

]+a,c pcm/'F.

The slightly more conservative value of

-47.5 pcm/'F is, proposed as the E0L HFP MTC Tech Spec LC0 limit under the Most Negative Feasible MTC approach.

Examination of the difference between the 300 ppm HFP equilibrium boron concentration MTC value and the EOL HFP MTC values concluded that a bounding expected difference between these two MTC values for Millstone Unit 3 reload cores is -7.50 pcm/'F.

This difference is subtracted from the proposed

-47.5 pcm/'F EOL HFP MTC Tech Spec limit to arrive at a proposed Tech Spec 300 ppm HFP HTC SR value of -40.0 pcm/*F.

It is concluded that the Tech Spec EOL MTC LCO and 300 ppm SR values proposed under the Most Negative Feasible HTC approach do not impact conclusions of FSAR accident analyses, because they do not affect the accident analysis assumption of MDC.

In addition, the validity of the above-stated LCO and SR HTC values, as well as the plant's ability to comply with them, will be examined for each reload cycle as part of the normal reload design process.

This proposed Tech Spec E0L MTC LCO and 300 ppm SR values are applicable to both 4 Loop and 3 Loop (corrected to rated thermal power) operation.

02931:6/880920 6-1

The new E0L MTC LC0 and 300 ppm SR HTC values and the revised basis for adjustment overcome the problems inherent with the present version of Tech Spec 3/4.1.1.3, yet still afford protection.

Tech Spec 3/4.1.1.3 continues to require that surveillance be performed, so that any deviations between the operating core and design predictions that might threaten the validicy of accident analysis assumptions can be detected, and continued surveillance and appropriate action undertaken.

02931:6/880920 6-2

REFERENCES 1.

Barry, R.

F., "LEOPARD - A Spectrum Dependent Non-Spatial Depletion Code for the IBM-7094," HCAP-3?69-26, September 1963.

2.

England, T. R., "CINDER - A One-Point Depletion and Fission Product Program," HAPD-TM-334, August 1962.

3.

Liu, Y.

S., et al., "ANC: A Hestinghouse Advanced Nodal Computer Code,"

WCAP-10965-P-A, September 1986. (H Proprietary) 4.

Camden, T. M., Kersting, P.

J., Carlson, H.

R., "PALADON - Hestinghouse Nodal Computer Code." HCAP-9485-P-A, December 1978. (H Proprietary]

5.

Barry, R.

F., "The PANDA Code," HCAP-7048-P-A, April 1967. (H Proprietary]

6.

Davidson, S.

L., Kramer, H. R., ed., "Hestinghouse Reload Safety Evaluation Methodology," HCAP-9272-P-A, July 1985. (H Proprietary]

02931:6/880921 R-1 l

APPENDIX A DETERMINATION OF MOST NEGATIVE FEASIBLE MTC SENSITIVITIES 02931:6/880920 A-1

i Investigation of the sensitivity of MTC to core operational parameters that are variable under normal core operation is a fundamental requirement of the Most Negative Feasible MTC approach.

Of the parameters discussed in Section 3.2, those that required detailed evaluation are:

- moderator temperature and pressure

- RCCA insertion

- axial flux (power) shape

- transient fission product (xenon) concentration for each of these parameters, the sensitivity analyses were performed by perturbing the parameter in such a way as to induce a change from its nominal E0L value, and then performing a MTC determination with the parameter held in the perturbed state. A further perturbation was induced and the MTC calculation repeated.

This sequence was repeated until sufficient data was obtained to reliably determine the trend of MTC change with variation in the value of the parameter.

In order to establish trends in MTC that are appropriate and bounding for the reload core type of interest, the sensitivity calculations were performed for five different reload cores.

These cores exhibit a spectrum of design features (such as cycle length, fuel lattice design, etc.) that permit the MTC sensitivity results to have broad application for Westinghouse-designed 17x17 4-loop cores. A brief description of the five reload core designs follows:

RELOAD A:

This core is the currently operating reload (Cycle 2) of Millstone Unit 3.

It utilizes the Westinghouse 17x17 standard rod diameter fuel design, feeds 84 assemblies in a low leakage loading pattern (L3P) and assumes a region average discharge burnup of 36000 MHD/MTU.

This reload also has 448 pyrex glass bas and a nominal cycle length of 15800 MHD/MTV.

The control rod absorber material is hafnium.

The nominal core average temperature assumed in this analysis is 590.5'F for 4 Loop operation and 582.7'F for 3 Loop operation.

This is the first cycle for implementation of the revised Tech Spec.

02931:6/880920 A-2

RELOAD B:

This is the proposed Cycle 3 reload design for Millstone Unit 3 using an L3P design feeding 76 assemblies with axial blankets, approximately 4500 part-length IFBA, high enrichments, and a region average discharge burnup of 45000 MHD/MTU.

The cycle length is 16500 MHD/MTU and all other core operating parameters are assumed to be the same as Reload A.

RELOAD C:

This reload is a conceptual equilibrium 24 month (21830 MHD/MTU) cycle design with advanced fuel product features including the features in RELOAD B plus Zr grids and Intermediate Flow Mixer (IFM) grids.

A thimble plug removal analysis is also assumed which increases the core bypass flow and raises the core average moderator temperature to 591.3*F for 4 Loop operation and 583.5'F for 3 Loop operation.

RELOAD D:

This reload core is an annual cycle L3P design using the Hestinghouse 14 foot length, 17x17 standard fuel rod diameter design.

The discharge burnup for this design is necessarily low since this is the first reload core for this particular plant.

This design has 416 pyrex glass bas and uses hafnium as the control rod absorber material.

The nominal core average moderator temperature of 596.5'F at full power.

RELOAD E:

This reload core is a Westinghouse 4 loop plant with 17x17 standard diameter fuel.

An L3P design is used for a cycle length of 17000 MHD/MTV and a region average discharge burnup of 38000 MHD/MTU.

This design also feeds 84 assemblies and uses 1728 pyrex glass bas.

The control rod absorber material is silver-indium-cadmium.

The neminal core average moderator temperature is 575.3*F.

As stated in Section 3.2, all of these core designs feature the control bank configuration that will be used in Millstone Unit 3 reload cores, which is shown in Figure 3.1.

02931:6/880920 A-3

The core neutronic models of these five reload cores were derived using standard Westinghouse design procedures and computer methods.

The ARK code, whic.) evolved from the LEOPARD (I) and CINDER (2) codes, was used to perform the fast and thermal spectrum calculations and is the basis for all cross sections, depletion rates, and reactivity feedback models. ANC(3) and PALAD0N(4), nodal analysis theory codes used in two and three dimensions, were used for core neutronic calculations to determine MTC sensitivity for the five reload cores.

APOLLO, an advanced version of PANDA (5), was used as an axial neutronic model of the reload cores to determine MTC.e sitivity to varying axial flux shape.

The neutronic calculations and evaluations performed for the five reload core designs established MTC sensitivities for each of the parameters listed above.

The sections which follow provide details of the calculations performed and the MTC sensitivity results obtained.

A.1 MTC Sensitivity to Moderator Temperature and Pressure Variation The decrease in moderator density which accompanies moderator heatup has the effact of reducing neutron moderation.

With a low soluble boron concentration in the moderator, this results in a negative moderator temperature coefficient. An increase in coolant temperature, keeping density constant, ieads to a hardened neutron spectrum and results in an increase in resonance absorption in U238, Pu240, and other isotopes.

The hardened spectrum also causes a decrease in the fission-to-capture ratio in U235 and Pu239.

Both of these effects make the HTC more negative.

In addition, the hardened neutron spectrum results in a larger fast-to-thermal flux ratio which increases the leakage of the core.

Again, the effect of higher leakage is to make the MTC more negative.

Since water density changes more rapidly with increasing temperature, and because of the spectrum hardening effects mentioned above, the HTC becomes progressively more negative with increasing temperature.

The sensitivity of MTC to increasing temperature was determined for each of the three reload cores by increasing core reference moderator temperature slightly above the nominal HFP value, while holding pressure constant at 2250 psia, and then 02931:6/880920 A-4

performing a MTC calculation that induced small changes in core K-effective via changes in moderator temperature and density about the reference values.

The effects of changes in moderator temperature and density were considered together.

After the MTC value was computed, core reference moderator temperature was further increased, and another MTC calculation performed. This process was repeated until the trend of MTC with increasing core reference moderator temperature was clearly established.

Results were recorded for the five reload cores in terms of change in MTC from the nominal HFP MTC as a function of increase in reference moderator temperature above the nominal HFP moderator temperature.

The results are shcwn in Figure A.I.

As expected, Reload 0 exhibits the strongest sensitivity of MTC to increases in moderator temperature, due to its higher nominal HFP reference temperature.

A curve which conservatively bounds the MTC sensitivity results of the five reloads is also shown in Figure A.I.

This bounding curve is used in determining the "delta MTC" factor, r

To use the bounding MTC sensitivity information of Figure A.I. the maximum allowable temperature and pressure (and, therefore, density) deviations permissible under operation that complies with Tech Specs must also be determined.

These deviation values, presented in Section 3.3, are combined with the sensitivity data to arrive at a "delta MTC" factor associated with moderator temperature and pressure (and, ther efore, density).

For the 6.l'F temperature deviation cited in Section 3.3, figure A.1 Indicates that the corresponding "delta MTC" due to temperature increase is (

]+a.c pcm/*F.

The "delta MTC" due to the 45 psi pressure deviation cited in Section 3.3 was conservatively determined to not exceed [

]+a c pc,f F.

The combined "delta MTC" factor is, therefore, [

]+a,c pcm/*F.

A.2 MTC Sensitivity to RCCA Insertion Hith constant moderator temperature, pressure, and boron concentration, insertion of control rods makes MTC more negative.

This trend in MTC arises from three effects.

The first is that RCCA insertion makes the overall flux spectrum slightly harder, which makes MTC more negative, as discussed in Section A.l.

The second effect is that RCCA insertion will increase core 02931:6/880920 A-5

leakage, which again makes HTC more negative.

The third effect arises from the impact of RCCA insertion on axial flux (power) shape.

This effect is treated separately in Section A.3.

Control rods can be inserted as a function of power level according to the RCCA insertion limits (RILs), and all RCCAs can be inserted at HZP coincident with reactor trip.

With greater RCCA insertion, MTC becomes more negative relative to the ARO MTC, all other parameters being held equal.

However, Tech Specs do not allow all other parameters to be held equal.

With deeper RCCA insertion, power must be reduced and Tavg will be reduced accordingly.

The reduction in Tavg serves to make the MTC more positive, and at E0L 0 ppm conditions, this positive Tavg effect will entirely offset the negative RCCA effect on MTC. The overall result is that the most negative HTC that can exist in the core occurs at HFP; therefore, the MTC sensitivity to RCCA insertion need only be determined at HFP conditions for HFP allowed RCCA insertion.

To calculate the EOL HFP HTC sensitivity to RCCA insertion, four of the reload core models had the lead control bank Inserted the maximum applicable amount determined from Section 3.3 ((

]+a,c steps withdrawn), at HFP, with no soluble boron in the core.

The MTC value for this condition was then determined by inducing small changes in core K-effective via changes in I

modeiater temperature and density about their reference values.

This MTC value was compared to the HTC determined at the same conditions, but with all RCCAs removed from the core.

l Of the four reload cores analyzed, it was determined that the maximum change to the EOL 0 ppm HFP AR0 MTC which occurred as a result of HFP RCCA insertion to a depth of (

]+"'C steps withdrawn was (

)+a,c pg,f.F.

Because some minor adjustment to RIls may be desirable for optimization of future core designs, it was considered prudent to further increase this MTC sensitivity

]+a,c. is considered sufficient to bound HFP RCCA factor. An increase of (

7 worth changes that would accompany anticipated RIL changes, therefore, the bounding "delta MTC" factor associated with allowable HFP RCCA insertien becomes (

)+a,c pg,f.7, 02931:6/880920 A-6

A.3 Sensitivity to Axial Flux (Power) Shape MTC is not so mucr directly affected t,y axial flux distribution itself, but is affected via the impact which the axial flux distribution has on the rate at which the moderator is heated as it moves up the core, and via the importance weighting which the axial flux shape imparts to different regions of the core.

In general, the accumulated burnup in the bottom half of the core exceeds that in the top half of the core, a; indicated in Figure A.2 for EOL of Reload A.

Other things being equal, higher burnup results in a more negative MTC as a result of isotopic impacts on flux spectrum.

A more negative axial flux (power) shape allocates a greater "importance weighting" to the lower regions of the core where burnups are greater, thereby accentuating this effect.

A greater effect is the impact which axial flux (power) shape has on heating rate of the moderator as a function of axial elevation.

Figure A.3 shows, in the top diagram, three distinct axial power shapes - one which is skewed toward the bottom of the core, one which is skewsd toward the top of the core, and one which is balanced, with an axial offset near zero.

The lower diagram in Figure A.3 shows the core moderator temperature as a function of core height for these three different axial power distributions.

While the same temperature rise through the core occurs for all three power shapes, it is evident that a more bottom-skewed axial power distribution will give rise to a higher average moderator temperature.

This results from the greater heating of the moderator in the lower core elevations for the bottom-skewed case.

As energy is added to the moderator at higher elevations, the temperature still remains highest for the bottom-skewed power case because of its initial "head start" in the lower elevations.

The temperature differences gradually decrease as a result of the differing heating rates occurring in the upper core regions among the three shapes.

Both the importance weighting effect and the moderator axial heating rate effect indicate that a more bottom-skewed flux shape (more negative Axial Flux Difference) will result in a more negative MTC.

This effect was investigated for four of the reload cores at EOL HFP O ppm conditions with no xenon in the core (xenon was removed so as to not complicate flux skewing strategy). A 02931:6/880920 A-7

specific axial flux shape was induced and then, holding this flux shape approximately constant, the MTC was determined by observing the small changes in core K-effective which resulted from variation in moderator temperature and density about their reference values.

A different axial flux l

shape was induced, and the MTC calculation repeated.

This process was rep?ated until the behavior of MTC with variation in axial flux shape (as quantified by Axial Flux Difference) was clearly identified.

Curves of "delta MTC" as a function of Axial Flux Difference (AFD) for the four reload cores are shown in Figure A.4.

Note that a zero AFD is taken as the reference point, therefore, "delta MTC" is fixed at zero for an AFD of zero.

Because more negative AFD values result from RCCA insertion, this axial flux shape MTC sensitivity implicitly captures part of the RCCA MTC sensitivity not included in the "delta MTC" factor of the previous section.

Section 3.1 concludes that a negative value of HFP AFD that is expected to bound future Millstone Unit 3 reload cores is [

]+a,c%.

Using a value which bounds the most conservative trend of Figure A.4, the "delta MTC" factor corresponding to [

]+a,c% t.FD is [

]+8'C pcm/'F.

Figure A.3 indicates that for a markedly negative AFD, the core average moderator temperature could b? as much as [ ]+a,c'F higher than that seen l

for a core with a balanced axial power shape (AFD near 0).

Recalling the MTC sensitivity to moderator temperattre of Figure A.1, one would expect a much I

greater MTC sensitivity to AFD than is indicated by Figure A.4. While the l

volume-weighted moderator temperature for a very aegative AFD may increase I

significantly above that of the balanced flux shape case, the power-weighted moderator temperature increase will be very modest, and this will result in rather weak MTC sensitivity to AFD.

l l

To further 111uttrate this point, examination of Figure A.3 shows that the very negative AFD power shape imparts a significant "importance weighting" tc the bottom portion of the core where moderator temperature is lowest, but in the top portion of the core, where moderator temperature is greatest, the relative importance weighting is low.

This power "importance weighting" 02931:6/880920 A-8

aspect serves to riegate a great deal of the "volume-weighted" temperature effect described above, and makes the "effective" moderator temperature l

increase for very bottom-skewed power shape rather small.

Again, this causes I

the MTC sensi,tivity to extremet of flux (power) shape to be rather weak.

A.4 Sensitivity to Transient Fission Product (Xenon) Concentration Xenon is the most sigalficant transient fission product in terms of effects on core reactivity and flux distribution, therefore, its possible impacts on MTC were investigated to compute the final "delta MTC" factor to include in the Most Negative Feasible MTC approach.

While Tech Specs place no limitations on either xenon distribution or overall concentration, the AFD limits discussed in Section 3.3, in effect, place a limitation on the amount of axial xenon skewing that can occur, and the physics of xenon buildup and decay place practical limits on the cont atration.

Because axial xenon distribution directly impacts axial flux shape, this aspect of xenon effect on MTC is implicitly included in the Axial Flux Shape "delta MTC" factor discussed in Section A.3.

What remains to be determined is the sensitivity to overall xenon concentration in the core.

Calculations to determine this sensitivity were performed with the ANC(#' and PALADON j

l codes, taking the E0L HFP AR0 0 ppm MTC value with an equilibrium l

concentration of xenon as the reference value of HTC.

A number of differing xenon concentration scenarios were modeled, and the MTC value associated eith each scenario was determined, l

For all five reload cores, the most negative HTC resulted when all xenon was removed from the core.

The largest "delta" from the reference (equilibrium xenon) MTC that occurred when all xenon was removed was (

pa.c pcm/*F.

This value becomes the final "delta MTC" factor attributable to xenon.

No further uncertainty is added, simply because the scenario of operating at full l

power with no xenon in the core is itself sufficiently conservative as to be bounding.

02931:6/880920 A-9

\\

f l

A.5 Three Loop Operation l

As discussed in Sectio's 3.3, the change in the MTC corresponding to the safety analysis MDC is less '.han the change in the calculated MTCs at the nominal core conditions for 4 and 3 Loop operation.

The following table is summarizes the EOL MTCs based on the differences discussed in Section 3.3.

Note that f

both 4 and 3 Loor shfety analysis MTC limits assume the same MOC of I

0.43 ok/g/cc but at different core average moderator temperatures.

Tech Spec Corresponding (100% RTP Values) 3 Loop (75% RTP) 300 PPM Sury.

-40.0 pcm/*F

(

J+a c pcm/'F 0 PPM Tech Spec (T.S.)

-47.5 pcm/*F

[

)+a,c pg,j.7 Safety Analysis Limit (S.A.)

-55.47 pcm/*F

-52.46 pcm/*F Delta (S.A. - T.S.)

- 7.97 pcm/*F

[

J+"'C pcm/*F The (

)+a, cpg,j.F change in the calculated MTCs between 3 Loop and 4 Loop is larger than the [

3+a cpg,j.F reduction of the safety analysis MTC.

Therefore, demonstrating that the 4 Loop total "delta-MTC" 15 applicable to 3 j

Loop operation will in turn inply that the Tech Spec determined for 4 Loop operation is also applicable to 3 Loop operation.

The 3 Loop MTC sensitivity will be less than the 4 Loop MTC sensitivity for the same deviations in the moderator temperature and pressure conditions.

This is because during 3 Loop operation the core is at a lower average moderator temperature, and the change in moderator density for a l'F change in moderator temperature will be smaller at the lower core average temperature.

This in turn will make the change in the MTC for a l'F change in temperature smaller for the 3 Loop core conditions.

The 3 Loop MTC sensitivity to changes in the moderator temperature and pressure will be approximately [ ]+a,c%

less than the 4 Loop sensitivity, or approximately (

J+a,c pcm/F.

The 3 Loop MTC sensitivity to RCCA insertion will be larger than the 4 loop sensitivity due to the increased D bank insertion allowed by the 3 Loop rod Insertion limits (RIL) in the Tech Spec versus the 4 Loop RIL at HFP (132 steps withdrawn for 3 Loop versus 164 steos withdrawn for 4 Loop).

The i

02931:6/880921 A-10

i l

increased D bank insertion is expected to make the 3 Loop HTC sensitivity to RCCA insertion approximately [

]+a,c y,f.F more negative than 4 Loop value, or approximately [

]+a,c pcm/' F.

i l

The 3 Loop sensitivity to the axial flu; (power) shape was calculated for j

Reload A and found to be [

]+"'C pcm/*/ for a (

3+a,c1 Axial Offset (AO),

Section A.3 reports a bounding 4 Loop value of (

J+a,c pcm/'F for a(

]+a c% A0, however the actual calculated 4 Loop value for Reload A was

[

]+a,c pcm/*F.

If the 3 Loop sensitivity is scaled up by the same amount.as the 4 Loop value, then the 3 Loop bounding HTC sensitivity to the axial flux (power) shape is [

]+a.c pcm/*F.

4 The 3 Loop MTC sensitivity to the transient fission product (xenon) concentration is also less limiting than the 4 Loop sensitivity.

This is due to the reduction in the 3 Loop xenon concentration co,apared to the N Loop HFP i

j equilibrium concentration.

The 3 Loop equillbelum xenon worth is approximately ( 3+"'C% Icss than the 4 Loop equilibrium worth at HFP, E0L core conditions.

Then an approximate value of the 3 Loop sensitivity to the I

transient fission product concentration is (

)+C% of the 4 Loop value f

given in Section A.4, or (

J+a,c pggf.F.

Finally, an estimate of the overall "delta-MTC" factor for 3 Loop operation for Millstone Unit 3 reloads is given by:

i Core Mooerator Temperature and Pressure Factor:

(

]+a,c cm/*F 4

RIL RCCA Insertion Factor:

(

j+a,c cm/'F Axial Flux (Power) Factor:

(

3+a,c C,j.7 Xenon Concentration Factor:

(

j+a,c f.7 Overall 3 Loop Delta-MTC Factor:

(

J+a,c pg,j.F If this value is further increased by (

)+a.c% for addittorial conservatism, l

the resulting 3 Loop "Delta-HTC" of (

]+a.cpggf.F is still less than the l

4 Loop "Delta-HTC" of (

]+a, cpg,j.F given in Sections 3.4 and 3.5.

)

Therefore the safety analysis MDC of 0.43 ak/g/cc is mot for 3 Loop j

operation for Hillstone Unit 3 reload cores and furthermore the 4 Loop Tech j

Spec is also applicable to 3 Loop operation.

02931:6/880920 A-11

--.---,u---y,-e

.,gn

.,_..,.,- 3 m w m_.

y.,,.

w w,,,

_ e n. +g

4 FIGURE A.1 CHANGE IN MTC WITH INCREASE IN T-AVERAGE ABOVE NOMINAL T-AVERAGE l

O

~

, l I

i i ll!

liii i lli i !!!

i!ii I

l l

Ill l

I ll Ill Illi l

ll lll l!

Ii lll I!ll l

Ill ll ll Illl

!II I

i l

Illi Illi i l li i

illl 11!i

~'

IIII I!!!

!III IlI!

Ii!I IIIl iIll ll l

illi lill I

i l llll liil Illi

-l l l

llll Ii illi illl iIlI ii!!

llll lIll lli I

ll lill 1l ll Il!!

l c

iili ii!I Iiil I

i l IiIi lili

!1!!

-2 l

s ij l

llll lI ll-

!Ill il l!

l l li

!I l

lll!

!!Il

!III

!! Il l l ll il!!

Ei I!

I l lll

!l !!

lIll

!i!!

i;4 !

s iIII Illl I,Ii il1' il Ii

'I I

Ii! i llll II II._!'l!

l Ii l!l!_

!!!l

!lll ll1I IIII i!

I!I I: I!

l!il l !

l l

Ill

!1 !

!!ll

!li!

!i!!

i I+'

Il ll l

_!_L i ll l!

Illi I

j.

Il II iiI

!l: i i : Ii l !l i l !

1 l

s l III llll l l lI l! i I

!Ii!

_i!Il _!'

lll ll lI ll ll lll l Iill i l !

l l 1

l!!l !!!

!!ii lI!!

!l!!

dd l

!!!L !ill Illi l l!!

Ji I

iiii i!l l l ! l;

!il l!1!

!l u

~

0 1

2 3

4 5

6 7

INCREASE IN T-AVG AB0VE HFP NOMINAL (OF) 02931:6/880920 A-12

FIGURE A.2 CORE AVERAGE AXIAL BURNUP VERSUS CORE HEIGHT AT EOL e

a

+a,C 30000 i

i i

e i

i i

i 6-

~

25000

~

~

~

J0000 B

R.

~

I

~

15000 M

-J 10000 N

1 I

l

(

i i

i i

i e

i e

i i

i i

i-5000 0 20 40 60 86 100 CrAt E LCM 1 (M Rctat) 80ff0W TOP 02721:6/880906 A-13

FIGURE A.3 AXIAL POWER AND MODERATOR TEMPERATURE VERSUS CORE HEIGHT RELATIVE AXtAL POWLR VERSUS CORE HEIGHT

+a,c 58 i

i 1.3 E

5 Q.so E

5W 40 l,

d

,,,-l I

I l

l l

l I

I l

t i

I l

I I

CORE CICHT (PEMENT)

BOTTOW TOP MODERATOR TEMPERATURE VERSUS CORE HEIGHT

+3,c i

4 i

i i

_i 1

p t.

E g...

Ea st9 CORE k iCHT (PERlENT) 907 TOW top 02721:6/880906 A.14

FIGURE A.4 DELTA MTC VERSUS AXIAL FLUX DIFFERENCE AT EOL, HFP, ARO

+a,c 1.0 11 l

l l

ll l

ll lll l

l l

ll-l l l'

l l

l i

i li ',I ll l

j i

I l-I li I

i i

i; 1

I i

l I

I I

I l-I I

I

,1, I

i l

'l I

.50

,l t

!!;t ll I

l l'

I ci i

i ilil I

l l

.L

'l!

l i

i i

l-

! l.

I 0.00 !!!!

lill llll !!!!

lll1 ill il!I I li li l

Il Ill! llll 11

!_L!. I!!! lill

!!I I!

!!I Illt l'

11-1 l!

.! Ill!

~

Il Ill!

I-ll l

I il i Ill 5

i_

i lIl i

l I il li il il 1111 i

i 111~

6 il

.I i

ill Ili l-lit:

I ll illi Il til i II illi liI ill

~

S' ill l

I !!

Il Ill!

I II Illi

!Ill

!!I 1111 I il !!I I

I l Il I!!!

I !!

11 l

11 0

l i

lll l'I I I!I

~i:1 l!'

I!

I I

I I

I I

Ill l

I lI l

!!! il i Il

!! il I

I I

I' l

I:ll i 11 Ill Ill Ill I il llIl ji-l ll l

Ill lill I

I Il I ll I

I

! ll I

!!I I{ 1

!I I

l LL !!!!

ll _

ll i

i l

l

! Il I

Il l

!Il 11!

ll _

l 'l I I!

I il

!! I!!!

I l!

1 l

!.!I I

I

! II

!ll l'l' I

I II

I 'I I

' I

-1.5 I

l l.

l 1.1 l!!

l l1

-l

!I!!

I il 1

I 1

1.!

lil i

i I

I I

I

!__li I

I I

l l

Ill I

ll!!

I!

l I

l l

I I'I' III I

II' I

IIII I 'II I

I

-2.0

-35 25 15 -10

-5 0

5 10 15 20 25 30 35 AXIALFt.UXOlFFERENCE(PERCENT) 02931:6/880920 A-15

- _ - _ _ - _ - _ _ _ _ _.