ML20054L459

Tech Specs for Operation of Fort St Vrain W/Region Outlet Temp Measurement Discrepancies

ML20054L459
Person / Time
Site:	Fort Saint Vrain
Issue date:	06/30/1982
From:	Alberstein D, Asmussen K GENERAL ATOMICS (FORMERLY GA TECHNOLOGIES, INC./GENER
To:
Shared Package
ML20054L444	List: ML20054L443 ML20054L448 ML20054L451 ML20054L453 ML20054L459
References
GA-C16781, NUDOCS 8207080140
Download: ML20054L459 (50)

Text

.

b,T[Q Sh&g I ams Mte 'rsv04r1

• t ,, I /

'E<1syySSe,', f

%3:$p%y:e:g r ado f

I;

,/

e

_ - . ,. :,= = w . -

[ 7 GA-C16781  :

i TECHNICAL SPECIFICATIONS FOR x OPERATION OF FSV WITH REGION j OUTLET TEMPERATURE  !

MEASUREMENT DISCREPAXCIES  ;

d f:

0 by  !!

D. ALBERSTEIN and K. E. ASMUSSEN I ,

O '

Prepared under  ! l Purchase Order No. N-3931 For Public Service Company of Colorado i

l l 0 1

~

l

., J

.1 o

l 3

1 f, f!

e

(

. 4 j JUNE 1982 3

r u

' "k' "

f_ - --'"~_;'

[*

• f.'. __7, y .e.

E _{.{A* :3. 'E_[ 3_,, _

_,,3 GENERAL - ~ =

ATOMIC COMPANY 8207080140 820706 PDR ADOCK 05000267 P PDR

GA-C16781 f

IJ l TECHNICAL SPECIFICATIONS FOR l OPERATION OF FSV WITH REGION

! OUTLET TEMPERATURE

! MEASUREMENT DISCREPANCIES a by p

j D. ALBERSTEIN and K. E. ASMUSSEN 1

f i

Prepared under ls Purchase Order No. N-3931 l

} For Public Service Company of Colorado i

(

I i i

L) 4 f  :

i ,

GENERAL ATOMIC PROJECT 1920 I JUNE 1982 GENERAL ATOMIC COMPANY

. .. n

I CONTENTS l

Page 1.0 SpMMARY AND CONCLUSIONS . . .. .. ... . . 1-1 2.0 TECENICAL SPECIFICATION CHANGES . . . . . .. 2-1 j 30 REFUELING REGION OUTLET TEMPERATURE . . . . . 3-1

'~

4.0 DISCUSSION OF LCO 4.1.7 - CORE INLET -

ORIFICE VALVES .. . . .. .. .. .. . .. 4-1 3.0 DISCUSSION OF SR 5.1.7 - REGION PEAKING t FACTOR SURVEILLANCE . . . . .. . . .. . .. 5-1 5.1 CALCULATED RPF SURVEILLANCE . . . .. 5-1 5.2 PERCENT RPF DISCREPANCY SURVEILLANCE .. . . .. . . . . . . . 5-3

5.3 CONCLUSION

S . . . . . . . . . . . . .. 5-6 6.0 COMBINED UNCERTAINTIES .. . . . ... . . . 6-1 6.1 TIME AVERAGE PEAK FUEL TEMPERATURES .. 6-2 6.2 MAXIMUM PEAK FUEL TEMPERATURES . . . . . 6-6

6.3 CONCLUSION

S . . . . . . . . . . . . . . 6-9

7.0 REFERENCES

. . . . . .. . . .. . . . . . . 7-1 w

\

i

FIGURES Page 4.1.7-1 Allowable Difference (Mismatch) Between Region Outlet Temperature and Core Average Outlet Temperature ....... 2-6 5-1 RPF Ratio as a Function of Cycle 3 Burnup : Control Rod Group 3A l Fully Withdrawn . . ........... 5-7 l

l 5-2 RPF Ratio as a Function of Cycle 3 l Burnup : Control Rod Group 4B Fully Withdrawn . . .. ......... 5-8 5-3 RPF Ratio as a Function of Cycle 3 Burnup : Control Rod Group 3D Fully Withdrawn . . ........... 5-9 5-4 RPF Ratio as a Function of Cycle 8 Burnup : Control Rod Group 3A Fully Withdrawn . . .. ......... 5-10 5-5 RPF Ratio as a Function of Cycle 8 Burnup: Control Rod Group 3B Fully Withdrawn . . .. ......... 5-11 5-6 Maximum Variation in RPF Ratio as a Function of Cyce Burnup (Cycle -3):

(Control God Group 3A Fully l Withdrawn) .. .. .. .. .. ..... 5-12 5-7 Time for ThC2 Kernel to Migrate 20 f Microns as a Function of Fuel Tem-perature - Upper 95% Confidence Fit f 5-13 to KMC Da ta ( Re f. 4 ) ..........

11

TABLES Page 6-1' Projected Equilibrium Cycle Maximum Time-Averaged Peak Fuel Temperature in NW Boundary Regions . . . ... . . . ... 6-10 6-2 Projected Equilibrium Cycle Maximum Peak Fuel Temperature in IN Boundary Regions . . . . . . . . . . . . . . . . . . . 6-11 6-3 Hypothetical Maximum Short Term Peak Fuel Temperature or 23000F in NW Boundary.

Region ...................

6-12 l

111

1.0

SUMMARY

AND CONCLUSIONS Extensive analyses of measured and computed region peaking factor (RPF) distributions during Fort St. Vrain Cycles 1, 2 and 3 have shown discrepancies (>10%) between the measured and calculated RPFs for the northwest (NW) boundary regions (Regions 20 and 32-37). The discrepan-cies in the NW boundary regions typically indicate a measured region outlet temperature lower than calculated, with the discrepancies in-creasing with core pressure drop. Other regions exhibit smaller dis-crepancies which are essentially independent of core pressure drop.

l Based upon extensive investigation, the core physics calculated RPFs have been discounted as a major source of the discrepancy. The major cause of l the RPF discrepancies in the NW boundary regions is a transverse flow of relatively cool helium from the core-reflector interface along the region

{ outlet thermocouple sleeve (Type II flow). This flow passes over the 1

t region outlet thermocouple assemblies of these regions and depresses the indicated region outlet temperature. The driving potential for Type II flow increases with increasing core pressure drop. This is consistent with the observed increase in discrepancy with core pressure drop. Only the seven NW boundary regions are susceptible to significant cool Type II flow induced outlet temperature measurement errors (Ref.1). These seven regions are the regions at which the core outlet thermocouple strings enter the core. Crossflow (jaws flow) may also contribute to the RPF discrepancies, but does not contribute to a region outlet temperature measurecent error.

To compensate for the NW boundary region outlet temperature measurement errors, special operating procedures have been provided for these regions to insure compliance with the original core design intent.

Appropriate technical specifications have been developed to govern operation with these measurement errors.

The purpose of this report is to review the new technical specifi-cations developed to address the NW boundary region discrepancies and to present analyses in support of those technical specifications.

1-1

In Section 2.0, the new technical specifications and their bases are presented. In Section 3 0, the method of determining the outlet temperature of the seven 1T4 boundary regions in the presence of Type II flow is discussed. In Section 4.0, the basis for the revised Technical l Specification LCO 4.1.7, Core Inlet Orifice Valves, is discussed in de tail. In Section 5.0, the basis for the new Technical Specification SR 5.1.7, Region Peaking Factor Surveillance, is presented.

In Sections 4.0 and 5.0 of this report, evaluations of fuel temperatures in the IT4 boundary regions are presented for two limiting cases. In order to determine the acceptability of these temperatures, they are compared to the following fuel thermal performance limits given in Section 3 2.3 3 of the FSAR:

Design maximum fuel temperature: 23720F J

Local short term peak fuel te mpera tu re , including hot spots: 27320F In Section 6.0 of this report, an estimate is made of the combined ,

l effect of uncertainties upon NW boundary region fuel temperatures. For these analyses, actual projected fuel temperatures in these regions are determined, and uncertainties are combined in the same manner as used in the FSAR hot spot evaluation.

It is concluded, based upon the b1 formation presented in this report, that the limiting conditions for operation and the surveillance

• equirements imposed by these technical specifications will result in fuel temperatures which are within the FSAR fuel thermal limits given above. Furthermore, under the method of operation governed by these technical specifications, fuel particle coating integrity will be maintained in a manner consistent with the Core Safety Limit. These technical specifications also define appropriate corrective action to be taken, if necessary, in time to assure that fuel particle coating integrity is maintained. The reactor will, therefore, continue to be operated without undue risk to public health and safety.

1-2 j

2.0 TECHNICAL SPECIFICATION CHANGES The new technical specifications (and bases) which will govern i operation with region outlet temperature measurement discrepancies are given in this section. In brief, the following changes have been made.

In Section 2.0, Definitions, a definition of individual refueling region outlet temperature has been added as Specification 2.21. This definition, when used in conjunction with the existing Specification 2.20, which defines core average outlet temperature, will enable the operator to ascertain compliance with the temperature limits of Spec-ification LCO 4.1.7. In addition, a new Specification 2.22, a definition of a comparison region, has been added.

Specification LCO 4.1.7 has been revised to limit the maximum mis-match between region outlet temperature and core average outlet tempera-ture to values more conservative than those previously contained in LCO 4.1.7. In addition, a limit on allowable RPF discrepancy in a comparison region is established, and appropriate corrective action, should the limit be exceeded, is specified.

Finally, a new surveillance requirement, SR 5.1.7, has been added to assure that the limit on comparison region RPF discrepancy in LCO 4.17 is met and that the values of RPF used to determine the outlet tempera-ture of the seven NW boundary regions are correct.

The new technical specifications and their bases are as follows:

a 2-1

2.21 Individual ReDjeling Region Outlet Temperature The individual refueling region outlet temperature is defi.. d as follows:

a) For Regions 1 through 19 and 21 through 31, the measured re-fueling region outlet temperature.

b) For Regions 20 and 32 through 37, whichever of the following temperatures is hottect: 1) the measured refueling region outlet temperature, or 2) the refueling region outlet tec:perature based upon the following quantities:

4g

1) The ratio of the relative power in each of these regions to that in their " comparison regions" as determined from physics calculations. j

2) The ratio of the helium flow rate through each of these regions to that through their " comparison regions" as determined based upon inlet orifice valve positions.

3) The measured refueling region outlet temperatures of their

" comparison region."

2.22 Comparison Region

, A comparison region is a core refueling region whose power, flow, and coolant outlet temperature characteristics are used to deter-mine the outlet temperature of a region for which the measured outlet temperature is unreliable. Experience has shown that Regions 20 and 32 through 37 have the potential for significant discrepancies between measured and actual region outlet tempera-tu re. These discrepancies are caused by a transverse flow of relatively cool helium from the core reflector interface along the region outlet thermocouple sleeve. This flow passes over the region outlet thermocouple assemblies of these regions and de-presses the indicated outlet temperature.

2-2

_ _ _ _ _ _ _ _ _ _ _ _ _ - l

Specification LCO 4.1.7 - Core Inlet Orifice Valves, Limiting Condition for Operation a) For a core average outlet temperature greater than or equal to 9500F, the core inlet orifice valves shall be adjusted to the following conditions: the individual region outlet temperature for the nine regions whose valves are most fully closed, and any region with control rods inserted more than two feet into the core, shall not exceed the core average outlet temperature by more than the limit (Mismatch B) shown in Figure 4.1.7-1. The individual region outlet temperature for the remaining regions shall not exceed the core average outlet temperature by more than the limit (Mismatch A) shown in Figure 4.1.7-1.

l b) For a core average outlet temperature less than 9500F, the individual I

region outlet temperature for all 37 regions shall not exceed the i core average cutlet temperature +400 0 F, and the conditions of LCO 4.1.9 must be met.

I f c) For any region being used as a comparison region, the percent " region f peaking factor (RPF) discrepancy,"% A RPF, given by

% ARPF = IRPF measured - RPF calculated) x 100%

\ l

( RPF calculated )

shall not be less than minus 10% (i.e., RPFmeasured less than 90% of RPF calculated) without corrective action as specified below.

Corrective action shall be initiated at the onset of a condition exceed-ing the limits stated in a) and b). If these limits are exceeded by 1) 1000F or more, an immediate orderly shutdown shall be initiated; 2) 500F or more, but less than 1000F, corrective action must be successfbl within two hours or an orderly shutdown shall be initiated; 3) less than 500F, 2-3

corrective action must be successful within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> or an orderly shut-down shall be initiated.

Corrective action shall be initiated upon discovery of a percent region peaking factor discrepancy exceeding the limit stated in c). If the limit is exceeded,1) a new comparison region which meets the limit shall be used or 2) the inferred coolant temperature rise (i.e., the individual refbeling region outlet temperature minus the core inlet temperature) in the region (20 or 32 through 37) being controlled by the comparison region shall be increased by a percent amount equal to or greater than that by which the limit stated in c) is exceeded.

Basis for Specification LCO 4.1.7 Experience gained during rise-to-power testing has shown Regions 20 and 32 through 37 have the potential for significant discrepancies between k measured and actual region outlet temperature. These discrepancies are caused by a transverse flow of relatively cool helium from the core reflector interface along the region outlet thermocouple sleeve. This flow passes over the region outlet thermocouple assemblies of these l regions and depresses the indicated region outlet temperature.

To compensate for these potential transverse flow-induced temperature measurement errors, Regions 20 and 32 through 37 shall have their region outlet temperatures determined by the power and flow characteristics of other regions in the core referred to as comparison regions. The comparison region method of operation was first developed for use during rise-to-power and fluctuation testing abovs 70% power in test procedure RT-500K. Experience obtained during that test indicates that, by use of comparison regions, Regions 20 and 32 through 37 can be operated in a manner consistent with the original reactor design intent and consistent with the criteria upon which Specificatior SL 31 is based.

If the measured region outlet temperature in Regions 20 and 32 through 37 is higher than that based upon the comparison region conditions, the measured region outlet te=perature is assumed to be correct.

2-4

Use of comparison regions requires that conditions in the comparison regions (power, flow, and outlet temperature) be well known. Region peaking factor discrepancies result fr,om combinations of errors or uncer-tainties in measured region outlet temperature, region flow inferred from orifice valve position, and calculated region power. Based upon an evaluation of data obtained during the rise-to-power testing program, region peaking factor discrepancies up to 10% (positive or negative) are not unexpected or considered to be excessive. Under the comparison region method of operation, excessively negative region peaking factor discrepancies in a comparison region could result in prolonged, high fuel temperatures in the region being operated with the comparison region.

The corrective action specified in LCO 4.1.7c will protect the fuel from such conditions.

During RT-500K testing, the difference (i.e., mismatch) between the i

measured region outlet temperature of any region and the core average outlet temperature at 100% power was maintained within the limits of Figure 4.1.7-1. The limits in this figure are more conservative than those used to develop Specification SL 3.1 and those contained in Speci-fication LCO 4.1.7 at the time RT-500K was conducted. In addition, Figure 4.1.7-1 directly limits the maximum region outlet temperature to 1,5550F, which is consistent with Table 3.6-1 of the FSAR. By requiring that the limits in Figure 4.1.7-1 be met, maximum fuel temperatures are kept within FSAR stated values regardless of the power level or the amount of core bypass flow which may exist.

During power operation with a core average outlet temperature less than 9500 F, sufficient overcooling of the core is provided with a +4000F deviation between the maximum region outlet temperature and the core average outlet temperature to assure that Specification SL 3.1 remains valid and that the integrity of the Dael particles is preserved.

The times at temperature exceeding the limits given represent conditions significantly below the core safety limit.

2-5

750 . _. .. .._, _ _ . _ . - . _ . -

. _- .d.=._ .: .x_xA.--...-._.-.

_  ; .. Ner t === {:_.

'= =._=_ _,tn__x_. _ __.th.N_ _ m =__.t ,- ,.t. =_

... q .. ......._..-.._....__..___-. ._g n__.__7 gi_m _r.r. _

in.unfx2t =F- - -- - t:= =* -t =-+----N _ _ _ t-

_r----*-

_u n. .. . . . m._.t=._ .=.._+t . _=_ ._. . - _ _ :_: =_ x_. ._ _ . . a. _ +t__._ 7 5 f'_*g > :__. _ 7 t=_.__-

. :...:fx

!xx!n=Nx irx x -n Nx = t = : x! = n ;=N = = =+x r n = :.N-r=f=:=tn c 'f-~~ n=n=n-\=

t== t N- .

rinun=}x--l=:_x=t__t_ _.t-+-M- - 100/ F H t N ----

700 ....__

== t___ut.=

. . . . . _ =_ .n: tNxt t: n: .m_____,__. .

\ = {+--

N. x : _.xq=_:=_ {=._:_ t=.__ =__t._n_._._n:.__ _ ._ t- ---- t- --

c==_ . ._

n,__ _.: .

- . _ _ _ _ _ - . . . . . ._.w._,_.__.

. .4-.

f x.

. u_r_r_.=_: t. _l.=_

_ . _=_.._ n_=_.!r__ _T7=w sg =N_xt-t--- i w . _

.IN!M55l55E!E5'lkU~'_U[E5EIN di

,5" _

.r == tt=:N ==n%=t-- t==={'=:=t 150/50:_A. uNy n__g----- 1 r--N  %---- t- - --- A=

- _t .

j 650

$ iRKEliMEWsTsgsid isi!=E=EsaW1 E%=___:iEE! . . _

~ = . - . .._...__;___.___p=.__.. t.__=. _

+-

. ,utr .. .._=._ = 7

._r.. ___l._____

___t- . .~__ I.

+

^~_iE g - -

e:  :;.  : __ t __Nx g T j:r_ up NxLA:t=n- m i = : = = =t--+ = N ...; t.X:n _ ._n =__.2.__.t t_ _. N.'g_ __. _ _t = = ,L

- N =i-*- g N s pV r

_"x200Tinni.-\=t=x=:trr,

..= t--- + -N, x;inc t==t=T. .-- jN.=r=t =: +- = _\rt--tr- t-z g -

t--- r+= u{='

W Nn=t---n=_W

.: xp_. t=:=tx--it. _' t- --nt=_==_xt=:= inn- _-t=r N--+- t --

5 000 3...._..._.._. ~-:t_ _x_ := . trun-N*.

. _ _ _ . ... _ t=__: r_.__,. . . _ . _ .

, . - t =. _. :=_ . [=. . . .:.x._=_ L: . ==. } =_-._n_- {= =. ; - _ n__

- _-t u -

~ -- "~-- '

g olmstimpi!ExmiaEgneaEgiesN =enis2 x ._

xx- =t=xx2t ==_x+ t ===n1N= :=n= :=

g . :-_ I===:txx }== n=txx -::==h===1=t- -+- N-- t-tr t----%r== t =

550

. ai::.:

fr-: =x+ =hnn--t =:xx t=:nntr-  : :t=:=[=N = xr:A =E_.nn .._.; 14_ t=:

g _= ._ 7 3 -- _ .

j

B = Mismatch limit applicable to - -_t - t= ~ N 9 regions whose orifices are l Eg7r= ===g most closed plus all regions . _. _ ; ; _ _ ._ . . g _ _.

f _ ;_ _N--+

with control rods inserted - - n:r more than 2 feet. !E. _ .i, E { Elf.i_4 t -E.E_ t t.E__

Q .

" A = Mismatch limit applicable to [EEEiE CE{g jg remaining reo, ions q -. . . . p .. .. __y p _ . . p .___3 = ..__g n_E_ ;.. _ ._.._..._

_ _(x ::=j _ ,.. n:r._ .; 7 _

660 700 755 AVERAGE TEMPERATURE RISE FROM CIRCULATOR INLET TO CORE OUTLET, *F Figure 4.1.7-1. Allowable Difference (Mismatch) Between Region Outlet Temperature and Core Average Outlet Temperature 2-6

Specification SR 5.1.7 - Region Peaking Factor Surveillance The calculated region peaking factors (RPFs) used in determining the individual region outlet temperatures for Regions 20 and 32 through 37 and the percent R?F discrepancy (see LCO 4.1.7) fbr Regions 1 through 19 and 21 through 31 shall be evaluated according to the fbilowing schedule for each refueling cycle:

a) Calculated RPFs: 1) Prior to initial power operation after re fueling.

2) At the equivalent of 20 (15) effective days at rated thercal power after re fueling.

3) At the equivalent of 40 (2 5) effective days at rated thermal power after re fueling.

4) At monthly intervals thereafter, provided that the core has accumulated an exposure of at least the equivalent of 10 effective days at rated thermal power since the previous evaluation.

If the core has accumulated an exposure of less than the equivalent of 10 effective days at rated thermal power since the previous evaluation, the evaluation may be deferred until the next applicable interval.

b) Percent RPF Discrepancy: Within a total elapsed time of 10 calendar days at reactor power levels above 40% of rated ' thermal power after the completion of any of the " Calculated RPF" evaluations required above with the following qualifi-cations:

2-7

1) A " Percent RPF Discrepancy" evaluation shall be performed prior to exceeding 40% of rated thermal power for the first time after refueling, but at a reactor power above 30% of rated ther-mal power.

2) If the total elapsed time at reactor power levels above 40% of rated thermal power does not exceed 10 calendar days prior to the subsequent " Calculated RPF" evaluation, the " Percent RPF Discrepancy" evaluation is not re-quired, but the total elapsed time at reactor power levels above 40% of rated thermal power between " Percent RPF Dis-crepancy" evaluations shall not exceed '

l 45 calendar days.

Basis for Specification SR 5.1.7 The calculated region peaking factors for Regions 20 and 32 through 37 and their comparison regiors will change during the refueling cycle as fission product inventories saturate, fissile material and burnable poison are depleted, and control rods are withdrawn from the core.

Evaluations based upon operating experience gained prior to completion of rise-to-power testing (i.e., Cycles 1 and 2 and part of Cycle 3) indicate that the ratio of the calculated region peaking factors in Regions 20 and 32 through 37 to the calculated region peaking factors in comparison regions as a function of control rod configuration, changes gradually in a predictable manner during a refueling cycle. A surveillance check of the calculated region peaking factors at the specified frequency will assure that the appropriate region peaking factors continue to be used in determining the region outlet temperature for Regions 20 and 32 through 37.

2-8

The calculated and measured region peaking factors for Regions 1 through 19 and 21 through 31 (candidate ce=parison regions) will change during the refbeling cycle as fission product inventories saturate, fissile material and burnable poison are depleted, control rods are withdrawn from the core, and region flow characteristics change. A surveillance check of the percent region peaking factor discrepancy will provide assurance that the requirements of LCO 4.1.7c are being met for comparison regions. The frequency for surveillance has been established based upon conservative evaluations of potential fuel kernel migration, which could occur if a region with an excessively large, negative region peaking factor discrepancy were used as a comparison region.

2-9

3.0 REFUELING REGION OUTLET TEMPERATURE As a result of the potential for Type II flow-induced region outlet temperature measurement errors in Regions 20 and 32 through 37, it is necessary to provide a definition of individual refueling region outlet temp era ture, Technical Specification 2.21. In Specification 2.20, core average outlet temperature is defined as the arithmetic average of the 37 individual measured region outlet tempera tures. Type II flow-induced region outlet temperature measurement errors in Regions 20 and 32 through 37 cause the core average outlet tempera ture calculated in accordance with this definition to be lower than the actual average by an amount which increases with reactor power level and is about 500F at 100% power.

Under the new de finition of individual refueling region outlet tempera ture, the outlet temperature for each of Regions 20 and 32 through 37 is given by:

l

[RPFi)[ flower \

11 l 6 To= Tin + ATer(RPFer) (fl0"i /

where T o = region outlet temperature for region being operated based on comparison region, Tin = core inlet helium temperature, aT er = measured comparison region temperature rise, RPFi = physics calculated region peaking factor (RPF) for region being operated based on comparison region, RPF er = physics calculated RPF for comparison region, flower = flow through comparison region, inferred from its inlet orifice valve position, 3-1

flowi = flow through region being cperated based on comparison region, inferred faam its own inlet orifice valve position, and 8 = factor to account for relative number of fuel colu=ns in co=parison region and NW boundary region, with values given by the following table:

Comparison Region Type 7-c olu=n 5-c olu=n 7-c olumn 1 7/5 NW Boundary Region Type 5-c olu=n 5/7 1 This equation enables one to determine the outlet tecperature of each NW boundary region from the measured outlet temperature and the power and flow characteristics of its co=parison region. The new definition also provides that if the measured region outlet temperature in Regions 20 and 32 through 37 is higher than that given above, the measured value is assumed to be correct.

For the remaining core regions (Regions 1 through 19 and 21 through

31) the region outlet te=perature is, as always, obtained by direct mea su rement. The core average outlet temperature is then obtained as the arithmetic average of the 37 individual region outlet temperatures, where for Regions 1 through 19 and 21 through 31 the individual temperatures a re mea su red , and for Regions 20 and 32 through 37 the individual temperatures are based upon comparison region characteristics or direct mea su rement, whichever is higher. Use of this new definition of individual refueling region outlet te=perature will eliminate the error in core average outlet temperature of up to 50 F at 100% power caused by Type II flow-induced outlet tecperature measurement errors.

3-2

4.0 DISCUSSION OF LCO 4.1.7 - CCRE INLET ORIFICE VALVES Operation of the reactor with errors in neasured region outlet tem-peratures for selected regions was first addressed in test procedure RT-500 K ( Re f. 2 ) . Throughout the various periods of testing from 40% to 100% of rated power per RT-500K, the use of comparison regions was demon-strated to be an effective manner in which to deal with potential region outlet temperature measurement errors. Based upon the operating experi-ence obtained during RT-500K testing, it is planned to operate each of the seven regions susceptible to significant outlet temperature measurement error, i.e. , Regions 20 and 32-37, via comparison regions.

This eliminates the need for frequent computer calculations, such as were done in RT-500K, to identify which of the seven regions requires the use of a comparison region or when such operation is required.

(

I Comparison regions typically have calculated power densities of magnitude and shape as a fbnction of control rod configuration similar to those of the corresponding region having an outlet temperature measure-ment error. Knowing the relative power densities (calculated) of the region susceptible to outlet temperature measurement error and that of the corresponding comparison region, and knowing the orifice valve position of the comparison region, the region susceptible to measurement error can be orificed to have an acceptable outlet temperature. Thus, the core inlet orifice valves of Regions 20 and 32-37 can be adjusted based upon the characteristics of their respective comparison regions such that their actual outlet gas temperatures are within the mismatch limits of Technical Specification Figure 4.1.7-1. Determination of the outlet temperature of a NW boundary region from its cc:parison region characteristics is discussed in Section 3.0 of this report.

Those regions which are not susceptible to significant cool Type II flow-induced outlet temperature measurement errors will continue to be operated in a normal manner based upon their measured outlet tempera-tu re s. However, to provide margin in addition to what was previously 4-1

provided in Technical Specification LCO 4.1.7 and therefore allow for the uncertainty associated with operation based in part on non-direct mea-su re ments , it is intended that the region outlet temperature mismatch limits will be restricted to those of Figure 4.1.7-1. This figure is based upon " Figure C" of RT-500K.

During RT-500K testing, the difference between the measured region outlet temperature of any region and the core average outlet temperature (i.e., mismatch) at 100% power was maintained within the limits of Figure 4.1.7-1. The limits in this figure are more conservative than those used to develop the Core Safety Limit, Specification SL 3.1, and those c on tained in Specification LCO 4.1.7 at the time RT-500K was conducted.

In addition, Figure 4.1.7-1 directly limits the maximum region outlet temperature to 1555 F, which is consistent with Table 3.6-1 of the FSAR.

By requiring that the limits in Figure 4.1.7-1 be met, maximum fuel temperatures are kept within FSAR-stated values regardless of the power level or the amount of core bypass flow which may exist. The corrective action requirements from the previous version of Technical Specification LCO 4.1.7 will be retained.

Operational flexibility is provided by the fact that more than one region can be used as a comparison region for each of Regions 20 and 32 -37. Any one of the comparison regions may be selected for use within the range of shim bank configurations for which it is best suited.

Regions affected similarly by changes in the regulating rod position (Region 1 control rod pair) are preferred as comparison regions.

Whenever possible, it is preferred that paired regions be of the same type, i.e. , 7-column regions compared with other 7-column regions and 5-column regions compared with other 5-column regions. However, it is l

1 not necessary that this be done. During RT-500K testing, provisions were made for comparing 5-column regions with 7-column regions, and vice v ersa. In one instance, such a comparison was success fbily used. If it becomes necessary to use a dissimilar region for comparison, the operator need only apply the appropriate correction factor (s) when determining the outlet temperature of the NW boundary region from the comparison region power and flow charac teristics. (See Sec tion 3.0 of this report. )

I 4-2

Uoe of comparison regions requires that conditions in the con-parison regions (power, flow, and outlet temperature) be well known.

Accordingly, LCO 4.1.7c includes a limit on the allowable percent region peaking factor (RPF) discrepancy in a cocoarison region.

Percent RPF discrepancy, % a RPF, is given by:

% ARPF = RPF measured - RPF calculated x 100%

( RPF calculated )

where RPF calculated = region peaking factor determined from physics calculations, and RPF measured = region peaking factor determined from measured coolant temperature rise across a region and flow through the region inferred from orifice valve position (i.e. , hea t balance) .

Based upon data obtained during the rise-to-power testing program, RPF discrepancies of up to 10% are not unexpected or considered to be exces-sive. ( An uncertainty in radial power of +10% was used in the hot spot analysis of FSAR Section 3.6.4.) RPF discrepancies result from combin-ations of errors or uncertainties in measured region outlet temperature, region flow inferred from orifice valve position, and calculated region power. The regions to be selected as comparison regions are not susceo-tible to significant cool Type II flow effects. Accord ingly, the in- <

dicated region outlet temperatures for these regions are considered to be reliable within the 500F range used to develop the Core Safety Limit, Specification SL 3 1. Based upon extensive investigations, the core physics calculated RPFs have been discounted as a major source of RPF disc rep anc ie s. It is believed, there fore , that the major cause of excessive RPF discrepancies, if any exist in candidate comparison regions, is uncertainty in flow through the region inferred fron inlet orifice valve position.

4-3

Under the co=parison region F.ethod of operation excessively negative RPF discrepancies in a co=parison region could result in prolonged, high fuel temperatures in the NW boundary region being paired with the comparison region. A negative RPF discrepancy would be obtained in a comparison region if the flow through the region were larger than the flow one would infer from orifice valve position (e.g. , due to cross flow entering the region). This additional flow would result in an actual lowering of the measured comparison region RPF, which is, in fac t ,

calculated from measured region outlet temperature and inferred region flow. If a region with an excessively negative RPF discrepancy were being used as a comparison region, one would infer that the cutlet te=perature of the NW boundary region, based upon co=parison region outlet temperature, power, and flow characteristics, is lower than it may be in fact. However, due to the effects of Type II flow upon the region outlet temperature measurement in the NW boundary regions, the operator would have no directly measurable indier cion of this possible discrep-ancy.

Therefore, LCO 4.1.7c limits the RPF discrepancy in a comparison region to minus 10%. Analyses have been conducted to determine the potential impact of negative RPF discrepancies in ec=parison regions upon fuel temperatures in Regions 20 and 32-37. Because n2el te=peratures are more adversely affected by variations in coolant flow than by variations in region power or region outlet temperature, it was conservatively assumed that the RPF discrepancy in the comparison region is entirely due to a discrepancy between the flow through the region inferred from orifice valve position and the actual flow through the region. Thus, for l example, if the actual flow through the comparison region were 10% higher than the flow indicated by its orifice valve position, the comparison region RPF discrepancy would be minus 10%. Furthermore, in quantifying the impact of negative RPF discrepancies in comparison regions on the fuel temperatures of NW boundary regions, it is conservatively assured that no cooling is provided by crossflow entering the NW boundary region.

4-4

A flow-induced RPF discrepancy of minus 10% in a comparison region can cause a fuel temperature increase in a NW boundary region ranging from about 1000F to about 1800F over the nominal value. The size of the increase is a function of the RPF/ intra-region power tilt combination in the NW region. Two limiting cases have been considered here.

In the first case, an RPF/ tilt combination necessary to maintain a high time-averaged fbel temperature was assumed to occur in a NW boundary region. Regions with the highest time-averaged fuel temperatures are expected to experience the most fuel kernel migration during normal opera tion. As noted in Appendix A of the FSAR, Section A.1.1.2.4, the fuel subject to the maximum time at high temperature in the core will experience a maximum temperature of 11500 C (2120 0 F) and an end-of-life temperature of 11000 C (2012 0F). The time-averaged temperature of this fuel is about 20500F. Less than 1% of the coated particles in the core experience these worst conditions. For a NW boundary region RPF/ tilt combination necessary to maintain the fuel at a time-averaged temperature of approximately 20500F, a minus 10% RPF discrepancy in the comparison region has been shown to result in an increase of 1500F in the NW boundary region fuel temperature. The fuel temperature in this small amount of fuel, therefore, may reach approximately 22000F, a value well below the 2372 F design maximum fuel temperature.

In the second case, an RPF/ tilt combination necessary to produce a peak NW boundary region fuel temperature of 23000F, the FSAR equilibrium cycle core maximum, was assumed. For this RPF/ tilt combination, a minus 10% RPF discrepancy in the comparison region would result in an increase of about 1800F in the NW boundary fuel temperature, resulting in a peak fuel temperature of about 24800F. However, core physics analyses have consistently indicated that the large intra-region power tilts necessary to produce higher fuel temperatures (such as the 23000F FSAR equilibrium cycle core maximum) do not persist for long periods of time and have indicated that such conditions (RPF/ tilt combinations) are usually found only in interior regions and not in the NW boundary regions. This 24800F peak fuel temperature is, nevertheless, well below the 27320F local short term peak fuel temperature limit in Section 3 2 3 3 of the FSAR.

4-5

There fore, it is concluded that the minus 10% limit on RPF dis-crepancy in Technical Specification LCO 4.1.7 will result in fbel tem-peratures which are, over the long term, well within the FSAR design maximum fuel temperature of 23720F, and over the short term, well within the FSAR local short term peak fuel temperature limit of 27320F.

4-6

5.0 DISCUSSION OF SR 5.1.7 - REG:0N PEAKING FACTOR SURVEILLANCE The calculated and measured RPFs will change during a refueling cycle as fission product inventories saturate, fissile material and burnable poison are depleted, control rods are withdrawn from the core, and region flow characteristics change. Accordingly, to assure that the appropriate RPFs are used in determining the region outlet temperatures of Regions 20 and 32 - 37, and to assure that the li=it on RPF discrepancy for comparison regions in LCO 4.1.7 is met, Surveillance Requirement SR 5.1.7, Region Peaking Factor Surveillance, has been established.

5.1 Calcula ted RPF Surveillance As discussed in Section 3.0 of this report, the region outlet tem-I perature in each of Regions 20 and 32 - 37 is determined from an expres-sion which includes the ratio of the physics calculated RPF in the the NW boundary region to that in the appropriate comparison region. The surveillance requirement on calculated RPFs (and thus on the RPF ratios) has been established to assure that appropriate RPF values are used to determine NW boundary region outlet temperatures.

The calculated RPFs for Regions 20 and 32 - 37 and their respective comparison regions will change during the refueling cycle as fission product inventories saturate, fissile material and burnable poison are depleted and centrol rods are withdrawn from the core. In order to characterize the variation in RPF ratios with burnup during a cycle, calculations have been done with the GAUGE code (Ref. 3). The change in RPF ratios (between NW boundary regions and various candidate comparison regions) with burnup was calculated for several control rod configur-a tion s. Analyses were done for both Cycle 3 and the equilibrium core cycles.

5-1

The variation in RPF ratio with Cycle 3 burnup, for various control rod groups fully withdrawn, is presented in Figures 5-1 through 5-3 as typical results from these analyses. Note that, for some pairs of regions, the RPF ratio decreases with burnup cr remains essentially c on stan t. In other cases, however, the RPF ratio increases with burnup.

Increases in RPF ratio with burnup can result in a nonconservative assessment of NW boundary region outlet temperature if these increases are not periodically taken into account (see equation in Section 3.0).

It should also be noted that, in general, the most rapid changes in RPF ratio occur early in the cycle.

The RPF ratio analyses indicate that these changes in RPF ratio with Cycle 3 burnup are somewhat larger than will be experienced in the equilibrium cycles. This result is illustrated in Figures 5-4 and 5-5, which show typical variations in RPF ratio with burnup during Cycle 8.

In general, the RPF ratio as a function of control rod configuration changes gradually during a refueling cycle. These variations in RPF ratio over the course of each surveillance interval can be predicted with the GAUGE code, and the calculations can be updated at each surveillance to reflect the actual operating history of the reactor over the preceeding surveillance interval.

The frequency of calculated RPF surveillance specified in SR 5.1.7 requires that calculated RPFs be checked more often early in the refuel-ing cycle (at beginning of cycle, at 20 1 5 EFPD, and at h0 1 5 EFPD),

when analyses indicate more rapid changes in RPF ratios. The largest increase in RPF ratio which has been projected to occur during these intervals is 12%, which would have occurred between beginning of cycle (assumed to be 5 EFPD) and 25 EFPD during Cycle 3 This case is illustrated in Figure 5-6. Variations in RPF ratio between surveillance checks later in Cycle 3 and during equilibrium cycles are smaller than this value, (e.g.

36%).

A series of analyses were conducted to assess the effects upon NW boundary region fuel temperature of variations in RPF ratio over the specified surveillance intervals. The effect of the variation in RPF 5-2 i __ ___ _ _ _ _ - - - .

ratio upon fuel temperature in the NW boundary region depends upon the region's RPF/ intra-region power tilt combination. As was done in Section 4.0 of this report, two limiting cases of RPF/ tilt combination in the NW boundary region were considered.

For the 12% increase in RPF ratio over a 25 EFPD period, cited above as the worst case, the resulting time-averaged error in RPF ratio is 6%. If this error is assumed to affect a NW boundary region with the RPF/ tilt combination necessary to maintain the maximum time-averaged temperature in the core, the fbel temperature increase is approxinately 950F. Since the maximum time-averaged Ibel temperature in the core is about 20500F, a 95 F time average temperature increase over a 25-day period will have a negligible impact upon Dael performance. Even at the end of the 25-day surveillance interval, when the RPF ratio has increased by 12% over its value at the beginning of the interval, the fuel temperature will be below 23720F.

If the 12% increase in RPF ratio were assumed to affect a NW boundary region with the RPF/ tilt combination necessary to produce a peak fuel temperature of 23000F, the FSAR equilibrium cycle core maximum, higher temperatures would result. As explained in Section 4.0, the conditions necessary to produce Dael temperatures near the FSAR maximum do not persist for long periods of time and are usually found only in interior regions rather than NW boundary regions. Nevertheless, if such conditions were assumed to exist in a NW region simultaneously affected by a 12% change in RIF ratio, a peak fuel temperature of 25200F would be ob tained. This value is less than the FSAR local short term peak fuel temperature limit of 27320F.

5.2 Parcant RPF Discrepancy Surveillance The frequency for surveillance of percent RPF discrepancy in Specification SR 5.1.7 has been established based upon conservative evaluations of potential fuel kernel migration. As stated in the basis of Technical Specification SL 3 1, the Core Safety Limit has been constructed to assure that a Dael kernel migrating at the highest rate in 5-3

the core will penetrate a distance less than the combined thickness (i.e., 70 microns) of the buffer coating plus inner isotropic coating on the particle. It is further noted in the basis of SL 3.1, that. the maximum fuel kernel migration expected for the Ibel with the most damaging temperature history is less than 20 microns. Thus, out of a total inner coating thickness of 70 microns, only 50 microns were assumed available in establishing the limits in SL 3.1. To establish an appropriate frequency for RPF discrepancy surveillance, the time required for a fuel kernel to migrate 20 microns as a function of fuel temperature was evaluated.

In the response to AEC Question 3 3, contained in FSAR Amendment 16, data on kernel migration coefficient as a function of temperature were presented. It was shown that ThC2 kernels migrate more rapidly than '

(Th/U)C2 kernels. Using the upper 95% confidence fit to the ThC2 da ta , a curve was presented which showed, as a function of temperature, time to migrate through the inner pyrolytic carbon coating (a distance of 70 mic rons) .

A similar analysis has been performed to establish the surveillance frequency for SR 5.1.7. The data on ThC2 kernel migration used in this

( evaluation were taken from Ref. 4. The upper 95% confidence fit to the data was used. The data in Ref. 4 were obtained with an experimental apparatus which provides a more accurate measurement of the thermal gradient across the fuel particles than that obtained in earlier experi-ments. In addition, more data are available now than were available at the time of the FSAR review. Consequently, the data used in this evalua-tion are believed to be the most representative data on ThC 2 kernel mig-ra tio1 available.

Assuming a thermal gradient of 0.0260F/ micron (the same thermal gradient as was assumed in FSAR Amendment 16 and in developing the core sa fety limit), time to migrate 20 microns was calculated as a function of fuel temperature. The results are shown in Figure 5-7.

5-4

Under SR 5.1.7, surveillance of RPF ' discrepancy will be conducted on a monthly basis. The longest time above 40% power between surveillances could be, on occasion, 45 days. As indicated in Figure 5-7, in order for fuel k'ernel to migrate 20 microns over a 45 day period, it must be exposed to a constant temperature of approximately 25300F.

In order for a NW boundary region to reach temperatures in excess of 25300F, the RPF discrepancy in its comparison region must be quite large and negative. If the RPF/ tilt combination in the NW boundary region were such as to produce the expected maximum time-averaged peak fuel temperature of 20500F, a comparison region RPF discrepancy of about minus 25% would be required to increase the fuel temperature to 25300F.

As discussed in Section 4.0, a minus 10% RPF discrepancy (the maximum allowed by LCO 4.1.7) could increase the peak fuel temperature by 1500F to approximately 22000F. This is less than the FSAR design maximum fuel temperature of 23720F and the 2530 0 F temperature at which a fbel kernel would migrate 20 microns over a 45 day maximum RPF discrepancy surveillance interval.

Another mechanism by which an RPF discrepancy might be imposed on a comparison region during a surveillance interval is by a region outlet temperature redistribution. If, as a result of a redistribution, the flow through the comparison region were to increase (by opening of a cross flow (jaws) path), a negative RPF discrepancy would occur. A review of all the data for redistributions obtained during RT-500 testing (Refs.

1 and 5) indicates that the largest decrease in region outlet temperature (indicative of the opening of a crossflow path) relative to the expected temperature change which has occurred in a candidate comparision region is 900F.' This region outlet temperature decrease occurred in Region 25

'It is also indicated by a review of all the data for redistributions obtained to date that only 3 non-NW boundary regions have ever participated in redistributions.

5-5

during the December 13, 980 redistribution between 58 and 61% power

( Ref. 5). Assuming an Bl F/ tilt combination necescary to produce the maximum expected time average fuel temperature (i.e., 20510F) occurs in a ff4 boundary region, then a decrease in the corresponding comparison region outlet tempem tur' of 900F, if allowed to go uncompensated, could result in a 1500F increa se in the fuel temperature of the 'f4 boundary region. Thus a peak fue . tempera ture of 22000F c ouw be obtained.

Again, this tempemture Ls below the FSAR design maximue Nel ter.'perature of 23720F. It is also below 25300F, the tempemture at which a N el kernel would migrate 20 microns during a 45 day maximum RPF discrepancy surveillance interval.

(It should also be notec that the !T4 boundary region's ficw may also increase due to the ope: ing of a crossflow path as a result of a redis tribution , in whict case the region's actual Nel tempemture would not increase as much, or, may in fact decrease. )

5.3 conclu sion s It is concluded that the specified surveillance frequency of SR 5.1.7 is sufficient to assure that the limit on RPF discrepancy in com-parison regions is met a'id that the appropriate RPFs are uced in determining the region outlet temperatures of Reginns 20 and 32 - 37.

The surveillance frequency is such that appropriate corrective action will be taken, if necessary, in time to assure that fuel particle coating integrity is maintained. Fuel temperatures will be kept, over the long term, below the tempem ture (i.e. , 25 300F) at which a fuel kernel would migmte 20 microns during a maximum 45 day RPF discrepancy surveillanno interval and below the FSAR design maximum value of 23720F. Over the short term, Nel tempemtures will be kept below the FSAR local short term pea:t temperature limit of 2732 F.

5-6

1 l y S46 6 8 R 222 2 2

/

- / // /

g 27 5

/

0 3

R 333 2 D

O 0 0 0

N a S I

3 R

NI O L O

OG R I E T G R E N RN O C

D O S EI  :

RR 0 P AA I 5 U PP 2 N MM R U

OO B CC

= = 3 j2 RR E L

C 0 Y I 0 C 2

s F D O P N F NW E O A

- I R T D P C H U N T 0

N U I 3- y I 5 1

R U

B F W A Y L

3 S L E A U L F C O Y I A C T 3 A

R P 0 U I 0 F O 1

P R R G 1

5 0 E s~ I 5 R U

G I

F 7

- - O 2 0 8 6 1 1 0 0 opg Y"

lllll l l l

l y 6 6 6 54 R 2 2 2 22

/ / / / / /

j S 2 0 67 D R 3 3 2 33 0 0 R A I 0 3 L A O R

M. T N

O C

g I 0

5 P

U 2 N R

U B

3 E

L C

0 Y I 0 C 2  :

D F P O F N E NW O A

- I R P T D

_ U C H

.. N N T R U I

. 0 F W

- A I 5 1

U B A Y

/

3 L E S L L A U C F Y O B

I

_ C T 4

_ A R P 0 U

_ I 0 F O 1

N P R

_ N OI

_ R G

_ OG I E G R .

E RN 2 D O 5 EI S -

_ RR 0 E AA 4- I 5 R PF MM f U G

I OO F CC

RR -

A-(

_ 0 6 4 2 8 1

G.

1 1 0 0 ,

9 &-

yco

l lfll 2 66 6 4 7 8 N R 22 /

2 2

/

2

/

/ //

j 25 0 3 4 D N O I R 33 2 3 3 0 OG 0 R I

G E E R A 1 4 I 0 3 L O

RN R DO S T

N EI RR O AA C PP MM  :

OO 0 P CC I 5 U

= = 2 N R

RR U B

3 E

L C

0 Y I 0 C 2

s F D O P N F NW E O A R

- I

- T D P C H U N T 0 N U I O I 5 R F W U

._ 1 B A Y L

3 S L E A U L F C O Y I D C T 3 A

R P 0 U I 0 F O 1

P R R G 3

5 E

," V I 0

5 R U

G I

F 6

1 4

1

- (2 1

0 1

8 0

6 0

0 o5c z u.

i@

1 l 1l ll

lI ll D

g 5 3 7 6 O s"

3 0 R N R 2 2 2 2 2 l 0

/ / / / / /

7 5 4 2 3 N OI R

3 0

2 3 3 3 L

O IOGE R

G R E

RN O 0 S

O ^ T N

O C

D O EI S RR  :

AA 0 P PP l 5 U MM 2 N OO R CC U B

8 RR E

L C

0 Y

' 0 C 2

F s

0 O P N F NW E O A I R

- T D P C l

l i

U i T 0 N U I F W O 0 _

. xT 5 1

R U A Y B L 8 S L E A U L F C O Y

I A T 3 C A R P 0 U

' 0 F O 1

P R R G 4

5 E

_ - m ^'- 0 5 R U

_ G

- I

_ F

- - - 4 y 0

6 4 2 8 1

0 6 1 1 1 O 0 e$5E y5 l1ll

i D

0 2 5 5 3 3 75 6 0 R R 2

/ /

2 2

//

2

/

22 2 l 0 j 3 7 5

// / 3 L 0 46 2 O R 3 3 3 2 33 3 R T

O O 0 ~ A N O

N C N IO OG  ;

I E G 0 P E R I 5 U RN 2 N R

DO S U

EI B RR AA 8 PP MM E L

OO C CC Y

= = 0 C I 0 2

RR s F D O P N F NW E O A I R

- T D P C l i

U N T N U I M' 0 . I 0

5 1

R U

B 8

E F W A Y S L A U L

L F C O

~ Y I B C T 3 A

R P 0 U I 0 F O 1

P R R G 5

5 E

R 70 I

0 U 5 G I

F 6 4 2 -

8 6 1 1 I 0 0 9 E*

b Y

,, y myygge j ggg 8 hN,U NB . .

~Uk .

js ..

, w a ~

1 -,

u } %. 9 y.

3 p .e D~ e p rza RGPp.+ : + aAggqi

ss ..4 A ru g %y= g& c z3 .;J,rey :. a

1. 3p Yp

..n' y..

f]:f '

y j ' , . )- $

j? " -$lk

. # ,[fd g% -

- ~;J yr , , ; .2 2 m4 f+r n$aQ(

+

h &xgn A;; eh h h.%-D 9 Qa-G' g ,. ,.

3,g wk W j V U *l6 u.

. ccw .&<9;.$#

.. .i %. < ' ,:

• ** m h -+ 4

- u, v

1 A

e. .._ . 1 ,4 77 1 9 a.- e.w 4 1 w n . w <.

k Mi,d&RV g. y y x

- w 3:

o:iQi'.g ,, g

,, *y p

a,naa,wn, e , + w m, u";p.;- , .

,3 .

a o o .  ;.m- 4 yz 7g ' *-gnn .

m.m, , 7.m..,e <.:.

4 . a < p r.=

• 9,, 9

..w p

. . .. +.

ga < 4'go o *

m. (9 ,% .y Q -(,1 gp[g. ~ w .

gy s ,

eg.

?, ,

fg ,= =4 ,

= * '

/w, 4

~% 3 y -

a, ;,. w .

,, ,. , .. .s ...

t

'[. .

I

.y gv wt .I w# f . 4,

.k% ;f D r w-w 3 ,

w&

' t- .-

w;,,,,, a ye. ao . 'yn 3 7 Y. g, . 2:Je: ~ 30f i -1lh h iM- ~

, c _ "m arreg

~l 'N -

-lW = p p. * ,

J& .na c- .

. . . s.m. a. .a.+n - x %.

RRBheMBei Eta..D..B2.3aaii811Riil8.ih.5 EtE3REERIEBEEE m --

9 2

/

5 3 0

• 0 3 E L

C Y

C F

O N

O 0 I

' 5 T 2 C N

U F

A S

A 0 O

' 0 I 2 T s A D R P

F F E P N R N IO P N OG U I I E G N E R * '

0 5 R N RN 1 U O DO B I T

EI S 3 A RR E I AA PP L R C A MM Y V OO C CC M U

= = 0 M

j2 I 0 I 1

RR X A

M 6

5 0 E I

5 R U

G I

F 0

0 8 6 4 2 0 2 1 1 1 1 1 9 icc Yn

0 0

0 2

0 0

0 1

)

4

- ' F E

R

(

A T

N A O D I

T C C M N K U

F 0

' T A

T S I A F

) S E S N C 0 Y O N 0 A R E 1 D C I

D

( I

' S M F N N O 0 O R 2 C CI E %

M T 5 0

A 9 2

R G R E I E T M P A P R O U T

Gi -

M L E E O N R

' T R U E E T M K A I R T 2E C P h M T E 0

T 1

R O L F E

' U E F M

I F

' T O

' 7 5

E R

U G

I F

- - - - - 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 7 5 3 1 3 2 2 2 2 2 Ch5%0s be 5

C 11l

l 6.0 COMBINED UNCERTAINTIES In previous sections of this report, the comparison region method of operation and the bases of Specifications LCO 4.1.7 and SR 5.1.7 were discussed in detail. The impact upon fuel temperatures in the NW boundary regions of three uncertainties associated with use of comparison regions were evaluated for two limiting cases. These uncertainties are

1) RPF discrepancy in the comparison region, which was conservatively attributed to uncertainty in flow through the comparison region, 2) variation with time of the RPF ratio (the ratio of the RPF in the NU boundary region to the RPF in the comparison region), and 3) the effects of a region outlet temperature redistribution upon the comparison region and the NW boundary region being operated with the corparison region. It was shown that, in each individual case, fuel temperatures will remain within the limits presented in the FSAR.

In this section, a realistic estimate is made of the combined effect of these uncertainties upon NW boundary region fuel temperatures.

In addition to the three uncertainties listed above, two other uncertainties are included: 1) uncertainty in the RPF ratio calculations (core physics calculations) performed at the beginning of a surveillance interval, and 2) uncertainty in the measured outlet temperature of the comparison region. These uncertainties were not addressed in the previous sections cecause they are not unique to the comparison region method of operation and because they are addressed separately in Section 3.6.4 of the FSAR. Their inclusion in this evaluation of total combined uncertainty is, however, appropriate.

These five uncertainties are considered random uncertainties since during the in-core residence time of any fuel element (6 cycles), they each may cause the peak fuel temperature to increase or decrease with essentially equal probability.

6-1

As discussed in Section 4.0 of this report, the core inlet orifice valves of all refueling regions are adjusted based upon their individual outlet temperatures so as to comply with the nismatch limits of LCO 4.1.7. For Regions 1-19 and 21-31 the individual outlet temperatures are measured, and for Regions 20 and 32-37 the individual outlet temperatures are determined using the expression given in Section 3 0 of this report or by direct measurement, whichever is higher.

In order to quantify realistically the impact of the various uncertainties on tne peak fuel temperatures of the NW boundary regions, calculations were performed which explicitly modeled each of these re-fueling regions. Core physics calculations were performed to predict the core power distribution for eight cycles of reactor operation. Based upon these calculated power distribution histories, thernal/ flow analyses were performed to predict the Dael temperature distribution histories.

The reactor was assumed to be at 100% power with a core inlet helium temperature of 7600F, and the outlet temperature of each region of interest was assumed to exceed the core average outlet te=perature by 500F, i.e., a 50 F mismatch. This mismatch is consistent with that allowed by the new Figure 4.1.7-1 of LCO 4.1.7.

6.1 Time Average Eeak Fuel Temoeratures From the results of the analyses mentioned above, the time average peak niel temperature for each column of Regions 20 and 32-37 was calcula ted. These peak fuel temperatures averaged over six cycles (Cycles 3-8) are representative of expected equilibrium conditions for these regions. The maximum time averaged peak Otel temperature calculated for any of these seven NW boundary regions is 1885 F.

The impact of each of the five random uncertainties on the maximum time average peak fuel temperature are discussed below.

6-2 t___ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ - . _

ihpsurement Error r

. The region cutlet temperature measurement error is taken to be 250 d F as was done in developing the Core Safety Limit, Specification SL 3.1. Analyses have shown that a 500F error in region outlet temperature results in a 2650F uncertainty in peak fuel temperature for regions having time average RPF/ tilt combinations such as those specifically predicted for the N'4 boundary regions.

RPF Ratio Calculation The uncertainty in the core physics calculated RPFs (and, hence, the RPF ratio uncertainty) is taken to be +10%, consistent with Section 3.6.4 of the FSAR. It has been shown that fuel temperatures are more adversely affected by variations in coolant flow than by variations in region power. Thus, it was conservatively assumed that the 10% RPF ratio uncertainty results in a 10% flow uncertainty. Calculations indicate that a 10% reduction in flow through any region having the time average RPF/ tilt combinations specifically predicted for the N'4 boundary regions results in an increase of <1200F in the peak fuel temperature.

RPF Ra tio Time Dependence As discussed in Section 5.1, core physics calculations show that the RPFs of Region 20 and 32-37 and their respective comparison regions change during the refueling cycle as fissile material and burnable poison are depleted. Thus, the RPF ratios used in the expression for region outlet temperature will change with time. During all but the first surveillance interval of a refueling cycle, the typical change in RPF ra tios between surveillance checks (at which time the RPF ratios are updated) is <6%, as discussed in Section 5.1. Thus, the time average change of RPF ratio during any surveillance interval is taken to be 23%.

A 3% error in region flow for any region having the time average RPF/ tilt combinations specifically predicted for the N'4 boundary regions results in an uncertainty in peak fuel temperature of < 300F.

6-3

RPF Discrepancy The core inlet orifice valves regulate the relative amount of coolant flow entering the individual refueling regions in a known and repeatable manner. Therefore, knowing the orifice valve positions of two regions, the relative amount of flow entering the two regions through their orifice valves may be reliably determined. The ratio of the region average coolant flow rates through the two regions is uncertain to the extent that the amount of cross flow (jaws flow), if any, differs between the two regions. As explained above (Section 4.0 of this report), this results in an adverse impact on DJel temperatures only if there is more cross flow entering a comparison region than the corresponding NW boundary region. If, on the other hand, there is more crossflow entering a NW boundary region than its comparison region, the effect is to reduce the peak fuel temperatures in the NW boundary region.

Crossflow entering a comparison region manifests itself in terms of a negative RPF discrepancy (see Section 4.0 of this report), which is limited by LCO 4.1.7 to minus 10%. (There is no restriction on positive RPF discrepancies in comparison regions, which could lead to lower fuel temperatures in NW boundary regions. ) The uncertainty in region coolent flow rate and hence in RPF discrepancy is thus taken to be 210 %. As sta ted in the discussion of RPF ratio uncertainty, a 210% uncertainty in region average flow for any region having the time average RPF/ tilt combinations specifically predicted for the NW boundary regions results in a 21200F 2ncertainty in peak fuel tempera ture.

Red is tribution The final uncertainty to be considered is the impact of region outlet temperature redistributions and the surveillance requirements of SR 5.1.7 on the peak fuel tempera tures.

As discussed in Section 5.2, the largest decrease in region outlet temperature (indicating an increase in crossflow) relative to the expected temperature change which has been experienced in any candidate 6-4

comparison region during any redistribution to date is 900F (Refs. 1 and 5). Analyses indicate that if a comparison region experienced such a change in crossflow and it went undetected, and there was no cross flow in the ff4 boundary region, the largest impact on any region having the time average RPF/ tilt combinations specifically predicted for the ?T4 boundary regions would be a 1400F increase in peak fuel temperature.

Region outlet temperature redistributions are not expected to occur during all surveillance intervals. However, if they did occur, the surveillance checks required by SR 5.1.7 would be expected to detect them. Assuming that redistributions of the largest magnitude experienced to date did occur during each interval and that they went undetected as much as 25% of the time, the probable uncertainty in the time average peak fbel temperature would be 350F (i.e. , 25% of 1400F).

Combined Uncertainties These random uncertainties are summarized in Table 6-1, where their combined impact on the peak fuel temperature is shown to result in a 1870F uncertainty. The uncertainties were combined using the root-sum-squares technique in the same manner as was applied in the hot spot evaluation in Section 3.6.4 of the FSAR.

When all of the above uncertainties are considered, the maximum time average peak fuel temperature for any ff4 boundary region during equilibrium core conditions is 20720F, as shown in Table 6-1. Fuel particles exposed to this temperature continuously would require more than 1800 days (6 cycles) to migrate 20 microns, as can be seen in Figure 5-7.

It is concluded, then, that under the comoarison region method of opera tion , taking into account the various associated uncertainties, fbel particle coating integrity will be maintained in a manner consistent with the basis of the Core Safety Limit, Specification SL 3.1.

6-5

6.2 Maximum Peak Fuel Tempera tures 6.2.1 Projec ted Maximum Peak Fuel Temeera ture The calculations which were performed to predict the fuel temper-ature histories for Cycles 3-8 are representative of expected equilibrium cycle conditions. The maximum peak fuel temperature calculated far any of the NW boundary regions during Cycles 3-8 is 20200F.

This peak fuel temperature is subject to the same five uncertain-ties as were discussed in Section 6.1 of this report. However, the On-pact of these various uncertainties on the Dael temperatures in the NW boundary regions is not of the same magnitude. The magnitude of the im-pact of the uncertainties is a function of the RPF/ tilt combination in the NW boundary region. The RPF/ tilt combination which produces the pre-dicted maximum peak fuel temperature of 20200F results in the uncertain-ties having a larger impact on fuel temperatures than did the time average RPF/ tilt combination appropriate for long term (time average) fuel temperature evaluations. (See Section 6.1)

The impact of each of the five random uncertainties on the pre-dicted maximum peak fuel temperature during an equilibrium cycle are discussed briefly below.

Measurement Error Again, the region outlet temperature measurement error is taken to be 150 F as was done in developing the Core Safety Limit, Specification SL 3 1. Analyses show that the 2500F error in measured region outlet temperature results in a 2800F uncertainty in peak fuel temperature for a region having a RPF/ tilt combination similar to that predicted for the NF boundary region having a peak fuel temperature of 20200F.

l 6-6

a. _ _ _ _ _ _ _ . _

RPF Ra tio Calculation The uncertainty in the core physics calcula ted RPFs (and, hence, the RPF ratio uncertainty) is again taken to be 310%, consistent with Section 3.6.4 of the FSAR. The impact of this uncertainty on peak fuel temperatures in NW boundary regions can be evaluated in a manner similar 1

to that discussed in Section 6.1 of this report. Such calculations indi-cate that this 10% uncertainty results in a 1500F uncertainty in the peak fuel temperature in any region having an RPF/ tilt combination similar to that predicted for the NW boundary region having the maximum predicted peak fbel temperature.

RPF Ratio Time Dependence As discussed in Section 5.1 of this report, core physics calcu-lations show that the RPFs of Regions 20 and 32-37 and their respective comparison regions change during the refueling cycle as fissile material and burnable poison are depleted. Thus, the RPF ra tios used in the ex-pression for region outlet temperature will change with time. The typical change in RPF ratios between surveillance checks (at which time the RPF ratios are updated) is 1,6%. Thus, for purposes of estimating the uncertainty in the maximum peak fuel temperature that might occur at any time between surveillance checks, the RPF ratio is taken to be uncertain by the full 6%. A 6% uncertainty in region flow for any region having an RPF/ tilt combination similar to that predicted for the NW boundary region having the maximum predicted peak fuel temperature results in an uncertainty of 1850F in peak fuel temperature.

RPF Discrepancy As discussed above (see Sections 4.0 and 6.1 of this report),

crossflow entering a comparison region manifests itself in terms of a negative RPF discrepancy which is limited by LCO 4.1.7 to minus 10%. The uncertainty in region coolant flow rate and hence in RPF discrepancy is thus again taken to be 210 %. A 210% uncertainty in region average flow 6-7

for any region having a RPF/ tilt ecebination similar to that predicted for the !T4 boundary region having the maximum peak fuel temperature results in a 2 500F 1 uncertainty in peak fuel temperature.

Red is tribu tion When considering the maximum peak Djel temperature which might occur in a NW boundary region between surveillance checks, it is conser-vatively assumed that a region outlet temperature redistribution, if it occurs, goes undetected. Assuming that a redistribution of the largest magnitude experienced to date did occur, the largest impact on the peak fuel temperature of any region having a RPF/ tilt conbination similar to that predicted for the NW boundary region having the maximum peak fuel temperature is 1500F.

Combined Uncertainties The random uncertainties are su=marized in Table 6-2, where their combined impact on the maximum peak fbel temperature is shown to result in a 2850F uncertainty. The uncertainties were combined using the root-sum-squares technique in the same manner as was applied in the hot spot evaluation in Section 3.6.4 of the FSAR.

When all of the above uncertainties are considered, the maximum peak fuel temperature for any NW boundary region during equilibrium core conditions is 23050F, as shown in Table 6-2. This maximum peak fuel tempera tu re is below the FSAR design maximum of 23720F.

6.2.2 Hypothetical Maximum Short Term Peak Fuel Temperature As was mentioned above, analyses were performed which explicitly modeled each refueling region's operating history during the course of eight cycles of predicted reactor operation. At no time during these eight cycles was the peak fuel temperature in any of the if4 boundary regions calculated to be as high as the FSAR (Section 3.6.3 3) equili-brium cycle maximum temperature of 23000F, However, a conservative estimate of the nnximum peak fuel temperature may be made by hypotheti- ,

6-8

cally assuming that an RPF/ tilt combination necessary to produce a peak temperature of 23000F exists in a NW boundary region. Core physics analyses indicate that the large intra-region power tilts necessary to produce this high temperature do not persist, thus these peak tempera-tures are considered short term.

The impact of uncertainties on the short term peak fuel temperature can be evaluated in a manner similar to that described in Section 6.2.1 of this report. The only difference being the increased magnitude of the impact of the uncertainties on the peak fuel temperature due to the RPF/ tilt combination necessary to produce a peak NW boundary region Dael temperature of 23000F.

The impact of each of the various random uncertainties on the FSAR equilibrium core maximum peak fuel temperature is summarized in Table 6-3 The combined impact of the uncertainties on the peak fuel temper-a tu re is shown to result in a 332 F uncertainty and a conservatively estimated maximum short term peak fuel temperature of 2632 F. Thus, the hypothetical maximum short term peak fuel temperature in a NW boundary region is below the 27320F local short term peak Dael temperature limit in Section 3.2.3.3 of the FSAR.

6.3 conc lu sion s From the information presented in Sections 6.1 and 6.2 of this report, it is concluded that under the comparison region method of op era tion , taking into account the various associated uncertainties,

1) 02el particle coating integrity will be maintained in a manner consistent with the bacis of the Core Safety Limit, Specification SL 3 1,

2) the projected maximum peak fuel temperature in any NW boundary region is below the FSAR design maximum of 23720F, and

3) the short term maximum peak fuel temperature in any NW boundary region is below the FSAR local short term peak fuel temperature limit of 2732 F.

6-9

1 TABLE 6-1 PROJECTED EQUILIBRIUM CYCLE MAXIMUM TIMF-AVERAGED PEAK FUEL TEMPERATURE IN NW BOUNDARY REGIONS Projected Maximum Time Average Peak Fuel Temp.* 18850F Impact of Uncertainties Measurement error 650F RPF ratio calculation 1200F RPF ratio time dependence 300F RPF discrepancy 1200F Redistribution 350F Combined Uncertainty (root-sum-squares) 187 F ,

Maximum Time Average Peak Fuel Temp. 20720F l l

l 1

' Maximum of Regions 20 and 32-37, average ever Cycles 3-8,100%

power,760 F core inlet, +500F mismatch.

6-10

c. _

l TABLE 6-2 l PROJECTED EQUILIBRIUM CYCLE MAXIMUM PEAK FUEL l

TEMPERATURE IN IN BOUNDARY REGIONS Projected Maximum Peak Fuel Temp.' 20200F g Impact of Uncertainties:

Measurement error 800F RPF ratio calculation 1500F RPF ratio time dependence 85 F RPF discrepancy 1500F Redistribution 1500F Combined Uncertainty (root-sum-squares) 285cp Maximum Peak Fuel Temp 23050F

' Maximum of Regions 20 and 32-37, average over Cycles 3-8,100%

power, 760oF core inlet, +500F mismatch.

6-11

i TABLE 6-3 HYPOTHETICAL MAXIMUM SHORT TERM PEAK FUEL TEMPERATURE OF 23000F IN NW BOUNDARY REGION FSAR Equilibrium Core Maximum Peak Fuel Temp. 23000F _

Impact of Uncertainties:

Measurement error 900F RPF ratio calculation 1800F RPF ratio time dependence 1000F RPF discrepancy 1800F Redistribution 1650F Combined Uncertainty (root-sum-squares) 3320F Hypothetical Maximum Short Term Peak Fuel Tenp 26320F 1

6-12

7.0 REFERENCES

1. D. W. Warembourg (PSC) letter to G. Kuzmycz (NRC) " Fort St. Vrain, Unit No.1, Meeting February 16, 1982, RT-500K Testing," P-82036, February 9,1982

2. Request for Test RT-500K, March 14, 1982 3 M. Wagner, " GAUGE, a Two Dimensional Few Group Neutron Diffusion-Depletion Program for a Uniform Triangular Mesh,"

GA-8307, March 15, 1968.

4. J. R. Sims, Jr. , and C. L. Smith, " Kernel Migration Measurements on Unteradiated BISO - and TRISO-Coated ThC2 Fuel Particles,"

GA-A13436, October, 1978.

5. K. E. As=ussen, et. al., " Testing at Fort St. Vrain After Installation of Region Constraint Devices," GA-C16277, February, 1981 - PSC submittal to NRC P-81312, December 10, 1981.

7-1

NN,f NERAL ATOMK GENERAL ATOMIC COMPANY P. O. BOX 81f;03 SAN DIEGO. CALIFORNI A 92138

-