L-99-112, Responds to NRC Questions Made During 990518 Meeting with Util Re LARs 220 & 88,for Bvps,Units 1 & 2.Copy of Ltr DLC-99-743,which Is non-proprietary Version of DLC-96-310 & Westinghouse Technical Bulletin ESBU-TB-96-07-R0 Also Encl
| ML20210A289 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 07/14/1999 |
| From: | Jain S DETROIT EDISON CO., DUQUESNE LIGHT CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| L-99-112, TAC-MA4616, TAC-MA4617, NUDOCS 9907220058 | |
| Download: ML20210A289 (22) | |
Text
- _ _ _ _ - _ _ _ _ _ _ _ _ - - _ - - - _ - _ _ _ _ _ . _ _ _ _ - _ _ _
, Duquesne Licjit Company g;;,v:a"e~rs'-
Shippingport. PA 15077 0004
^'"
!e"n$o"r VS/ President July 14, 1999 Faxf7N4Nes Ul'NLe'**o*m. ion L-99-112 U. S. Nuclear Regulatory Comrnission
/ Attention: Document Control Desk Washington, DC 20555-0001
Subject:
Beaver Valley Power Station, Unit No. I and No. 2 BV-1 Docket No. 50-334, License No. DPR-66 BV-2 Docket No. 50-412, License No. NPF-73 Response to Oral Request for Information Regarding License Amendment Request Nos. 220 and 88 (TAC Nos. MA4616 and MA4617)
On May 18, 1999, the Nuclear Regulatory Commission (NRC) staff met with representatives of the Duquesne Light Company (DLC) and Westinghouse to discuss DLC's License Amendment Request (LAR) Nos. 220 and 88 for Beaver Valley Power Station Units No.1 and 2, respectively. During this meeting, the NRC staff requested that DLC provide a copy of a letter, DLC-96-310, and Westinghouse Technical Bulletin ESBU-TB-96-07-R0, which are referenced in DLC's LAR Nos. 220 and 88. Enclosure Number 1 contains a copy ofletter DLC-99-743 which is a non-proprietary version of DLC-96-310. Westinghouse has modified the original version of the body of the DLC-96-310 letter during the preparation of the non-proprietary version. Three areas were modified in DLC-99-743 to provide additional clarification from the DLC-96-310 version. These areas have revision linBs with the letter "C". Two other areas were modified due to editorial changes and are indicated by revision lines and the letter "E".
Enclosure Number 2 contains a copy of Westinghouse Technical Bulletin ESBU-TB 07-R0.
If the NRC requires additional information concerning this subject matter, please contact Mark S. Ackerman at 412-393-5203.
Sincerely, h v_ J o W_
l~ Sushil C. Jain o c: Mr. D. S. Collins, Project Manager i Mr. D. M. Kern, Sr. Resident Inspector Mr. H. J. Miller, NRC Region I Admini:;trator OhlVERING.
Attachments E 9907220058 99o714 PDR ADOCK 05000334 P PDR
AFFIDAVIT FOR APPLICATION OF AMENDMENT COMMONWEALTH OF PENNSYLVANIA)
) SS:
COUNTY OF BEAVER )
Subject:
Beaver Valley Power Station, Unit No.1 and No. 2 BV-1 Docket No. 50-334, License No. DPR-66 BV-2 Docket No. 50-412, License No. NPF-73 Response to Oral Request for Information Regarding License Amendment Request Nos. 220 and 88 i
Before me, the undersigned notary public, in and for the County and l Commonwealth aforesaid, this day personally appeared Sushil C. Jain, to me known, who being duly sworn according to law, deposes and says that he is Senior Vice President, Nuclear Services of the Nuclear Power Division, Duquesne Light Company, he is duly authorized to execute and file the foregoing submittal or behalf of said Company, and the statements set forth in the submittal are true and correct to the best of his knowledge, information and belief.
i i
&Uo_Q Sushil C. Jain Subscribed and sworn to before me on this/ ay of , [99 k l ekana/2 N[/ary Public and " ~
Notarial Seat Tracey A. Baczek, Notary Public My s n xp'ir u.82 1 Member. Pennsytvens Assoc 6ation of Notanes
l l .
i 1
ENCLOSURE 1 1
l DLC-99-743 i 1
1 l
l
1 Westinghouse Electric Company LLC O
Nudsar Senrices DMsion Box 355 Pittsburgh, Pennsylvania 15:30 0355 June 11,1999 DLC-99-743 NSD-SAE ESI 99-234 Mr. W. R. Kline, Manager Nuclear Engineering Duquesne LigM Company Beaver Valley Power Station P.O. Box 004 Shippingport, PA 15077-0004 Attention: Mr. Mark Manoleras
)
Duquesne Light Company Beaver Valley Units 1 and 2 Overtemperature ATand overpower ATDynamic Compensation Tolerance Posillon Paper
Dear Mr. Kline:
Duquesne Light has recently requested that the Westinghouse Proprietary classification be removed from information provided in a letter (DLC-96 310) in support of a Technical Specification submittal. The subject letter was reviewed to determine if the proprietary classification could be removed. During this review, two areas were identified as needing additional clarification and were modified.
Please feel free to contact the undersigned if there are any additional questions.
Regards, WESTINGHOUSE ELECTRIC COMPANY MMM E.A. Dzenis Customer Project Manager JJD/kk cc:
R. A. Hruby K. J. Frederick BVRC Central File, SEB-1 M. Manoleras
i i
t
),
I Attachment ]
to 1 l
D LC-99-743
~
i I
t .. . . . .
4 Duquesne Light has recently contacted Westinghouse with questions on the uncertainties associated with instrumentation for the dynamic compensation terms that are included in the Ove temperature DT lh (OTDT) and Overpower DT (OPDT) reactor trip functions. Specifically, the dynamic compensation terms in question are the lead / lag on OTDT and the rate lag on OPDT. The Technical Specifications give the same values that were input in the safety analyses. The Beaver Valley Technical Specifications do not include inequalities. Also, the OTDT and OPDT hardware for plants with 7100 series equipment make it impossible to exactly input the Technical Specification values (per the functional requirements the hardware for the time constants is adjustable in increments such that any setpoint can be obtained within
- t10%). Thus, the settings for these plants have historically been set as close to the required values as is possible and within t10%. Note that Beaver Valley Unit i has 7100 series equipment and Beaver Valley Unit 2 has 7300 series equipment.
The purpose of this document is to explain the bases of the dynamic compensation settings assumed in the safety analyses and to provide a discussion of the acceptability of settings for prior and cunent -
practices. Also provided are recommendations for how the dynamic compensation terms could be set in i the future to avoid any reoccurrence of related questions. This document deals primarily with the i lead / lag on OTDT and the rate lag on OPDT but is equally applicable to other dynamically compensated protective functions at Beaver Valley. The low steam pressure protective function is dynamically compensated with a lead-lag function. This lead-lag is assumed in the analyses and the nominal values are assumed in the analysis and are listed in the Technical Specifications. No dynamic compensation is explicitly modeled for any other protective functions in the Beaver Valley non-LOCA safety analyses.
Backaround information - Currently, the lead and lag on the OTDT function and the rate lag on OPDT function at Beaver Valley as given in the Technical Specifications are:
Unit 1:
Lead (t,) = 30 seconds Lag (t 2) = 4 seconds Rate lag (t3) = 10 seconds Unit 2: .
Lead (t4) = 30 seconds Lag (ts) = 4 seconds Rate lag (tr) = 10 seconds The OTDT and OPDT setpoint equations are attached at the end of this document.
When these time constants are assumed in the safety analyses, the values modeled are the nominal values given above. This is consistent with all of the time response values (pure delays and dynamic compensation leads, legs and rate / lags) given in the Technical Specifications that are assumed in the safety analyses. The safety analysis values assumed for actuation setpoints are different from the Technical Specification actuation setpoints by an amount sufficient to cover the calculated static instrument uncertainties but the safety analysis and Technical Specification values for time response constants are identical i@
Compounding the quMion is the hardware used to set these values at Beaver Valley Unit 1. The hardware will not allow an exact value to be set and thus, the values at Unit 1 have always been set as
' close to the Technical Specification values as possible without necessarily being in the conservative direction with respect to the safety analyses. The hardware is designed to allow the settings to be set
- within 110% of a desired value.
As is mentioned eariier, similar questions could be asked about the other dynamically compensated protective functions. The only other dynamically compensated protective function explicitly credited in the safety analyses is the Low Steam Pressure function.- The lead-lag on Low Steam Pressure assumed in the accident analyses are as foll'ows:
Lead = 50.0 seconds Lag = 5.0 seconds The same values are used for both Unit 1 and Unit 2 in the safety analyses and are given in the Beaver Valley Unit 2 Technical Specifications. Note that these parameters are not included in the Beaver Valley
DLC-743 Unit 1 Technical Specifications. For both Unit 1 and Unit 2, these values have historically been set as close to the nominal values given above.' This is done for both Unit 1 and Unit 2 despite only being specified in the Unit 2 Technical Specifications.
Westinehouse Position - in non-LOCA safety analyses, most parameters are set to their nominal
- values. Selected key parameters which are determined to be important to the analysis results are identified and the values used in the analyses for these parameters are set in a conservative fashion to demonstrate that the applicable safety criteria are met. It is Westinghouse's position that this method yields a sufficiently conservative licensing basis and it is not necessary to be conservative for every parameter in every licensing basis analysis. As such, it is concluded that the analyses that model nominal values for the dynamic compensation terms at Beaver Valley are sufficiently conservative and that no additional analyses need to be performed..
If a more conservative position is desired, Westinghouse has provided guidance in the following areas in support of customer requests. One utility asked Westinghouse to identify the direction of conservatism for each of the constants associated with the OTDT and OPDT setpoint equations. They used this .
Information to revise their procedures to ensure that the settings at the plant are conservative with respect to the nominal values assumed in the analyses. Associated with this, a change to the. Technical )
Specifications was proposed to replace the equalities with inequalities to specifically allow these settings to be set in a conservative fashion. The Beaver Valley OTDT and OPDT setpoint equations along with the conservative direction for each of the terms are attached at tM end of this document. The i conservative direction for the load and lag on Low Steam Pressure are:
- Lead > indicated Value
_ Lag <_ Indicated Value A second utility requested reanalysis with a 5% uncertainty band assumed around each dynamic compensation term. Obviously, this resulted in a slight loss in DNBR margin.
Recommendation for Future Consideration - Given the increasing interest in confirming a clear design and operating basis for all safety and licensing analyses, Westinghouse recornmends consideration of the NUREG-1431 Technical Specification format with lacc. Mr%s and a procedural implementation which assures a conservative setting to prevent any further reoccurrence for the need to address this question. Note that the inequalities are recommended to ensure that literal compliance with the Technical Specifications can be demonstrated. The inequalities are not needed to support the safety analysis methodology.
Conclusion ,The safety analyses assume nominal values for all time constants included in the licensing l basis analyses. Uncertainties are applied to actuation setpoints and other key parameters in the safety j analyses with the uncertainties applied in the most conservative direction. This methcdology results in a conservative analysis. The safety analyses should continue to assume nominal dynar.n compensation values, the Technical Specifications should reflect nominal values with the equalities replaced with the {
appropriate inequalities and the dynamic compensation terms should be set at the plant on the l
conservative side of the nominal values to ensure that the Technical Specifications are met considering )
drift and other allowances as deemed necessary.
The tafety analyses performed in support of Beaver Valley are consistent with the methods discussed in l this document and, thus, conservatively demonstrate that all applicable acceptance criteria are met. The I results and conclusions presented in the Beaver Valley UFSAR remain applicable.
s
} a .
) es 5
- }
)
I A
s e osy K
- *
- I ( **** l eee A 3 eeee cl .
u uu ( f uuuu s n a K l l l aaa fi -
)
l l l l aaaa aa ,
} VVV - " VVVV t ey K 3
}
)
I ddd )
' T dddd s te ,
) A e ee P - eeee ef 2 I
A f(
i i i f f f i i i - ) i i i i f f f f de b as K f( -
t t nnn t
P s.
t i i i i t t t t nnnn d d e l 3
- ] eee (
+ eeee n uh ot K
(
) " ddd
- dddd e P
T I I I 3
I
(
I I I I m hh si t s n
sL s K /
)
sss2 m sn w i a
s - )
s +
1 o g n P
( T 5
KTT
. 3" (
KT tP.' 2 c
e it an ge e i
o * + ]
T T
- r y ei s t ht t 3 1
(
[ l g h s g a K /
) )
- n t n
,o i n
u q + 1
( s.
t K.
- t o
r i s s
n c i b
n s ai
] * * * * * + - **** m E
eeee s gt T T uuu u 1 eeee uuuu i n o
[ l l l l aaaa
( T] l l l l 1 Ko p 3 i c
t *
/
) aaaa R- s n )
s 3
VVVV 1
(
) VVVV 7 dt ne e
v i
o t K. dddd
- s. dddd 0- a s l o
p + - e e ee i i i i T eeee 3T v 1 ] ff f f i i i i T + i i i i f f f f i i i i 6
9- KD n t
e
(
/ T t t n.tnn t [
1 t t t t nnnn nT i
)
1
- e . ee (
/ eeee B iO o t
a S 1
(
)
s dddd I I I I 2
)
s 5
)
1
(
dddd I I I I T- t ce ht T U s T t i
n T +
t s
22s2 t i
n
+ *) 22ss B fut nh t
n o
A U *
[
1 3 3' U 1
( s, 3 3"
S E ph a i t
u P
i
)
(
) t Kt TT i
s r i
r t b
O l l
y e s2 T
)
1
( l y
e t
- s. +
1 ne op t
rm ou i
t r
s l
a + ( i s
Va + *)
t s t i d * /
- aa cn d l
- V ( 1 )
sy
- c e ae s n r /
e
( s 3
r e
- t ,
i f u i
e r to a a v
)
S i
+
t eeee v 2
(
[
eeee cl a e
pv Ts i
b a T uu uu a e K-
- 3 uuuu Dno d T B e + 1
(
l l l l aaaa VVVV B K l l l l aaaa Sd e Ti n a
A 1
(
/
)
s3 K i
VVVV l at a c
Ot a r ic n T
- 2
(
t dddd ee e e i i i i
{
- K dddd ee e e i i i i ic nid fi of c
i a
g O K
[
f f f f iiii t t t nnnn t
T, {*
ff f f i i i i t t t t nnnn hci n e r de ep s e
r y ,
K 3 ee ee dddd A T, eeee dddd To i f S i
i u
_ e K -
I I I I s A I I I I ed ht n t l na q e
l l
{
. ) u ec r" a
- K 22ss s s 222s f d n i
sT of o i
5
) s hc md V T, {,
A T 2 .
KKTT .' +
T s t
3 2
KKtT
. .' 1 s
- a i
e r n ea r s A 1
( + 2 c "m s e T
t TI" e ) s
/
) 1
(
2i if ih D ,
V s. )
1
( / e c t t a
ae +
t s.
t *
)
1
(
- l bap e s
v r i n T-Ts O
+ * * * ) * * *
- T ed s e dnrm B 1
(
/ 1
(
e uuu c e *e u 5
T 2
)
s2 e
uuuue e e u*e i noo p
n oi n ae
)
1 /
l l l aaala + l l l l aaaala dl ca t
( )
VVV V T VVVV V e e f n t
Tt 1
(
1
( + t oh oo Dna T
- ddd e ee d
/
) 1 si / ( dddd ee e e d n to n c Pt A T i i i f f f t e
)
i i i i f f f f t e
yt oe it u Osn he o i i i a t s iiii a l T
O A
P tnnni t t c eeed
+ t i tnnnni t t t c eeeed i
t n cio el a mv c O ddd n I I I I 1
( + dddd I I I I I n i l
paz t i d e t e fh ot
- 1
( xi " h t gd s2ss A
T
- s2s2s el a ao nn T T em u r h
ge t i l
a a ).
, 3 T, O AP i 5 , T, l ao ulb c sK KKTA 3 O KKT tA 3 VN oi ht ss ed Ap o l
h n
- Ta
ENCLOSURE 2 I Westinghouse Technical Bulletin ESBU-TB-96-07-R0 l
4 l
i l
l
i
\
ENERGY SYSTEMS Westinghouse BUSINESS UNrr Technical Bulletin An advisory notice of a recent technical development pertainisg ta the lastallation or operation of Westinghouse-supplied Nuclear Plant equipment. Recipients should evaluate the information and recommendation, and initiate action where rppropriate.
P.O. BOX 355 Piusburgh, PA !$230 Subject Number TEMPERATURE RELATED FUNCTIONS ESBU-TB 96-07-R0 System (s) Date RPS/ESFAS AND CONTROL SYSTEM TEMPERATURE ' November 5,1996 RELATED FUNCTIONS Affecmd Plaats 5.0.(s)
ALL PLANTS WITH AT AND T,ya RELATED FUNCTIONS References AIL,i.i. Safety Yes E Sheet See Below Related Equipment No 0 1 of 9
References:
Reactor Protection System / Engineered Safety Features Actuation System Setpoint Studies, RTD Bypass Elimination reports, Improved Thermal Design Pmcedure reports and Revised Thermal Design Procedure reports INTRODUCTION The Overtemperature AT and Overpower AT reactor trip functions (OTAT and OPAT) are relied upon for j many FSAR accident analyses. They are also the most complicated protection functions due to their definition, interactive nature and the number of parameters utilized in thek setpoint determinations. These functions rely upon temp::rature indication to define power (AT) and to define where the protection functions lie (T Ava). As a result several effects, e.g., hot leg streaming, changes in hot la satamuuLdue to burcup dependeelt radial, power redispibution and cold leg streaming (all of which have significant influence on temperature indicanon), have anaea complexity to the process of ensuring that these functions have been appropriately calibrated at the plant. The instrument uncertataty calculations and safety analysis assumptions for these tunctions have varied over the last several years as these effects were addressed.
Additional Informanor., if Reque.d, may be Oben.m.d from tb. Onsmasse Tel. phone 412-374-540s or (WIN) 2:4 5409 Onsa.mr Approval 7
C. R. Tuley, Fellow Egineer {/Wandrovich Safety Systems Operations Regulatory and Licensing Initiatives
- 5. FA-E. A. Dzenis, Manager /
Safety Systems Operations I
_s_-_ a - - m, - er n, ,_. . .,_t m . . ..m.- o, j
- .f.e w, a e t dw e. y. , m ,f., % .,d as..ba r af, .fs.a 1 afermanos.
14mLWFW3Ull5e6 i
ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 2 of 9 INPO Nuclear Network OE 7063,1/24/95 (see Attachment 1), (submitted by Pacific Gas and Electric) identified a plant specific instanc.s of a hot Icg indicating lower than actual temperature This had a detrimental influence on the associated OTAT and OPAT reacter trips and potentially, the rod control system. Westinghouse worked with PG&E to resolve this concern and wu successful with the implementation of an approach which normalized the two protection functions to the loop specific indicated values for AT and T,va. As a result of this e&rt, discussions with various plants and with the latest information available on streaming effects, it was suggested that a n. ors explicit consideratien of the effects of core burnup on changes in temperature distribution within the hot leg should be made in the uncertainty calculations. To assure plant operation consistent with the assumptions of the uncertainty calculations and safety analyses, this document has been published for plant consideration. Even if a l specific plant ha not observed the phenomena noted in this document in past or current fuel cycles, it may choose to consider the techniques noted in order to address possible future observations.
It should be recognized that the approach utilized in this document recommends changes to surveillance and normalization which, if the temperature variations occur, avoids the need for setpoint and/or safety analysis margin allocation. Alternatively, these effects can be accounted for through explicit assumptions in the setpoint uncertainty or transient analyses calculations.
BACKGROUND j Overtemperature AT Trip The OTAT trip protects the core from conditions that would cause DN3 and limits the range of s
applicability for the OPAT trip. Since DNB is a function of tempme- and pressure, the trip has '
historically been interpreted as an absolute trip with defined setpoints. The only parameter that was not considered t- he predetermined was AT,, which is the measured AT at full power conditions. A simplified yet ion of the basic equation,i.e., no lead /l'ag constants included,is:
ATg dromr = A T,[K - K 2(T - T') + K (P g
3 - P') - fg(M)]
where: AT,,,,om is the Overtemperature hip setpoint in *F, AT is the full power AT measured for the particular loop / channel, K iis the reference trip setpoint or AT multiplier, K 2is the penalty or benefit multiplier for deviation from reference Tava, T'is the reference T iva ia the analysis, K, is the penalty or benefit multiplier for deviation from reference Pressure, P' is the reference RCS pressure in the analysis (typically 2250 psia), and f i(AI) is the power shape penalty function. -
It should be noted in this equation and function that conditions that result in DNB margin will cause the trip setpoint to increase. Specifically, a lower T,yo or a higher RCS pressure will increase the trip setpoint.
l HawPwouitww l
)
J
ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 3 of 9 8
Overpower AT Trip The OPAT trip protects the core from overpower transient conditions that could result in excessive kw/ft in the fuel and to limit the range of applicability of the OTAT trip. This trip is redundant and diverse to the NIS Power Range - High reactor trip, which performs a similar function. A simplified version of the basic equation, i.e., no lead /las constants included, is:
j A r*lmyg.
s = AT,[K a - Kp) - Ks 7 - n - f2%
where: AT,,,,ol, is the Overpower trip setpoint in 'F, .
AT, is the full power AT measured for the particular loop / channel, K. is the reference trip setpoint or AT multiplier, K, is the penalty multiplier fer the rate of change in T ivo, K, is the pensity multiplier for deviation from reference Tava, T" is the nominal full power indicated (loop specific as measured or as found) Tava for the channel, and f 2(AI) is the power shape penalty function (typically set to zero).
It should be noted in this equation and function that conditions that result in margin will not cause the trip setpoint to increase. Specifically, all adjustments to the trip setpoint are reductions based upon deviation from the nominal operatag loop conditions.
Application and Calibration ,
1 Only the earliest plant protection systems (Foxboro process sacks) calculate the OTAT and OPAT trips as simplistically as the above equations indicate. These plants are differentiated by control board AT indication in 'F or 'C and intsrument spans of 75 'F. Later protection systems (7100 and 7300 series analog process racks and Eagle-21 digital process racks) normalize the AT span to percent power, thus in l
effect dividing both sides of the equation by AT,. This second set of plants is differentiated by control board AT indication and instnament spans of 150 % RTP and typically have more lead / lag functions in the equation definitions than the earlier plants.
Due to the nature of the trip f6=*ans, We tinghouse procedure recommendations were developed that required the determination and a4ustment of AT, and T" values for each A==1 The T' value in the
~ OTAT trip function was historically set based upon the safety analysis, which is typically the same Tava value as used in the control system. The T" value in the OPAT trip function was histoncally set to the loop specific indicated value.
i 840% WPW9ull/H6 i:
.I ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 4 of 9 -
E i
~ The following discusses how streaming occurs and what steps should be taken in the uncertamty calculations, the safety analyses and at the plant to address these efects.
Hot Leg Streaming The water heated in the contral regions of the core is directed into the upper internals where mixing occurs. This does not occur to the same extent at the core periphery where the water is not heated to the i same temperr ares. When directed to the vessel exit nozzles, the water exiting the core does not riix l completely, th as resulting in temperature gradients across the hot legs. The magnitudes of the gn iients vary, but have been observed to be in the range of 6 to 16 'F. The larger gradients have been noted in the more recent past and are attributed to the larger maximum to minimum core exit temperature diferentials associated with aggressive core designs. It has also become apparent that the longitudinal locations of the RTDs or scoops can afect the perception of the bulk average temperature due to the .
- mixing in the hot legs.
Radial Barndown Efects i
The general efect noted is that indicated Tm is typically higher than actual T , particularly at BOL.
However, at least one instance of indication lower than actual has been documented at PG&E (see Attachment 1). Also noted is a decrease in indicated Tm with constant power as the cycle progresses.
This is due to the relative decrease in the central core region water temperature from radial power flattening with increasing burnup. Over cycle life, the indicated Tm can decrease by several *F. This results in a corresponding decrease in indicated AT and indicated Ti vo also. With protection functions normalized at full power conditions at BOL (currently AT, for OTAT and OPAT and T" for OPAT), these indication decreases are in the non conservative direction. In addition, there has been no guidance previously provided on when re-normalization should take place to restore the protection functions to within the uncertainty calculation assumptions.
Cold Leg Streaming Another ofrect noted in temperature measurement is the streaming in the cold leg for loops with Model
- 93A reactor coolant pumps. Stressains is present in all cold less due to the diferent tube lengths and flowa in the Steam Generator U-tubes. The resultant gradient is then reduced by mixing in the cross-over leg and RCP impeller. However, for the 93A RCP, internal venes in the pump afect mixing such that the temperature gradient is not completely homogenized. - For this efect, indicated cold leg temperature may be less than actual tempperature. His is non conservative for a protection system because indicated T ava may be less than actual T,v. which generates inappropnate margin to trip. The magnitude and signire== of this efect is a function of the location of the centerline of the gradient and the placement of the cold les RTD or scoop. If the RTD is located at the centerline of the gradient and thus indicates the true bulk average tasaperature of the cold leg, there is no efect. If the centechne of the gradient is ofset, the indicated temperature and thus the inferred bulk average temperatuis, may be lower (or higher) than actual depending on the direction of tlw gradian 4
5
ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 5 of 9 All of these effects, streaming in the hot leg, the effect ofincreasing burnup on the streaming in the hot leg and streaming in the cold leg, make determination of absolute temperature to a high degree of accuracy difficult. Normalization of the protection functions can be an effective way to compensate for these effects and minimize operating and analytical margin erosion. The normalization of the various trip functions is discussed in detail below.
Normalization of AT, I
Overtemperature AT, Overpower AT and the Vessel AT Equivalent to Power (VATEP), which is used in the Steam Generator Water Level - Low-Low Trip Time Delay (TTD), are based on the definition of -
100 % RTP for a given loop. The loop specific full power is defined as the indicated AT at 100 %
system power as determined by a precision secondary side power calorimetric measurement. For OTAT and OPAT this value is defined as AT,. This determination is typically made at near 100 % RTP conditions at BOL. Since there can be loop to loop power and/or loop to loop flow variances due to several different causes, it is typical to have a different reference full power AT for each loop. The
~
important points to note are; 1) the value of AT, used in the scaling of these functions should be loop specific, and 2) AT, is an indicated value that should define the reference equilibrium full power condition. This value can change as a function of core conditions, since it is dependent on hot leg streaming characteristics, and should be determined on a loop specific bois each cycle. (For two loop plants with two protection channels per loop, the above is applicable on a 4===al basis. If the indicated AT varies between two channels for the same loop, each channel should be scaled per its specific characteristics. For three loop plants with separase control channels for AT, the control channels should be treated in a manner similar to AT in the protection channels.)
Normalization of T' and T" Since the OTAT and OPAT trip functions are dependent on the measured AT value for power and have become relativistic trip functions (m opposed to absolute trip functions) due to streaming in the hot and cold legs for T,va, it is important to include in the scaling process the indicated, full power, loop specific value for T,yo (T' in OTAT, T" in OPAT). This requirement is more apparent for GPAT since it is an overpower trip that can not recognize full power conditions without explicit normalization at those conditions. It is not as clear for OTAT, since it is primarily a DNBR trip. However, it should be noted that the safety analyses are performed for a given set ofinitial conditions for pressure, temperature, power and ftpw. It is W that the actual measurtd conditions of temperature and flow will be conservative with respect to these assumptions, i.e., measured flow is greater than Thermal Design or Minimum Measured Flow and indicated T,va is equal to the value assumed in the safety analyses. However, operation has been noted for several plants with ',v. significantly lower than assumed in the safety analyses. Also it is typical for the safety analy,,e to assume loop symmetry, i.e., all loops indicate the same value for T,y., while the actual plant oorditions may indmae= a loop to loop asymmetry. Add to these factors recent analyses for some plants a apportug operating temperature windows in the range of 10 to 30*F instead of the typical assumption of x single reference T,y program value (with an associated rod control accuracy about it) and it becomes more important to define T' and T" for the specific set of
. operatag conditions for that cycle for each loop. It is therefore important to utilize the loop specific, indicated, full power T,y as the input for T' and T".
mww.uwe. I
ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 6 of 9 Due to short term temperature oscillation phenomena, e.g., upper plenum anomaly (see NSAL-92-007),
variable power sharing between loops and flow cross-over between hot legs, the steady-state equilibrium values for Tum, Tem, AT and T va, i a determined by a time average over thase to five minutes (or a period consistent with the transients), should be used for determination of the appropriate scaling input values for AT., T' and T". These time averaged values should also be used for any loop to loop comparisons that the plant staff may wish to perform and comparison to any Technical Specification requirements.
Surveillance Interval for Verification of AT, The indicated value for Tum and thus indicated AT and T aya,is dependent on the hot les streaming characteristics at the time of normalization. Current plant data indicates that the hot leg streaming characteristics may change substantially over a cycle. Changes in indicated Tum in the non-conservative direction (resulting in a smaller indicated AT) over a given cycle have been observed. This is outside of typical Westinghouse uncertanty calculation assumptions. The magnitude of acceptable change in indicated Tue, AT or Tay, is a function of the uncertainty calculation assumptions and margin for the ,
protection functions. Since the change in indicated Tum is a function of radial power redistribution, quarterly surveillance of these three parnaeters is appropriate. More rapid changes in hot leg streaming (as a function of burnup) may require more frequent surveillance of these three parameters. These parameters should be verified to be within uncertainty calculation assumptions.
Verificaties of Tsor and Tc , When Normalising AT,, T' and T" When normalizing the OTAT, OPAT and VATEP functions to reflect the loop specific, indicated AT and Taya at full power conditions, it should be recognized that two other parameters (loop specific Tum and Teu) may also vay. Plant analyses typically assume the control of T va i (via the automatic rod control j system) follows the Tavs program (defined in the PLS document) up to the full power value (de' fined in the FSAR or RSAC) for the controlling loop. If the plant is in manual rod control, it is assumed that the operator emulates the characteristics and limits of the auto rod control system and follows the same Tava program. This will assure Tuor, T ea, AT (AT,) and Taya (T' and T") values within the analyses assumed ;
limits. Two means may be used to verify operation within the limits,1) verify the plant is in auto rod i control and that the control system is functioning appropriately and 2) via visual means (ifin manual rod control). This verification may be the use of control board, process compuser indication or the Man Machine Interface (MMI) after accounting for indication errors and acceptable process variation.
Operation outside the typical rod contml system ' accuracy or not following the Tava program may result in exceeding upper or lower limits on temperature
'5 Effect of Fall Power Operaties at h.dicated Loop AT Values Less Than Assamed la Uncertainty Calculations Instrument uncertanty e= lent =*ia== performed for OTAT and OF T are based on projected operation at the design vessel AT. The e= leal =*iana for the channel uncertanty (CSA) and the total allowance (TA) are a function of AT span. Typically the instrument span is orther a fixed value in *F or a percentage of power
(% RTP).
-mm.,
e ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 7 of 9 For tlwse calculations utilizing an instrument span based on a percentage of power the following is noted.
For a fixed set of input instrument errors (in 'F, psig and % AI), a decrease in the indicated full power AT results in a non conservative increase in the magnitude of the individual instrument errors in % AT span and thus the CSA? There is also an increase in the magnitude of TA, which is beneficial. But the rate of increase in the magnitude of TA is less than the rate of increase in the CSA. Thus operation at an indicated vessel AT less than that assumed in the uncertanty calculation is non conservative (from an instrument uncertamty point of view). Acceptability of operation at a reduced indicated AT is a function of the available margin in the OTAT and OPAT reactor trips and the magnitude of the difference between the assumed AT span in the uncertainty calculation vs the indicated AT span. For small changes in span, there may be sufficient margin in the uncertanty calculations to compensate. For larger changes, re- '
calculation is necessary. As the cuange is non-linear, a more definitive determination of the acceptable magnitude ofindicated AT change must be performed on a case by case basis.
For those calculations utilizing a span based on a fixed value, the decrease in vessel AT results in a decrease in the total allowance with a subsequent loss of margin. Again, for small changes in the ,
indicated AT, there may be sufficient margin to compensats. For larger changes, re-calculation is necessary As each case is unique, these calculations are typically performed on a case by case basis.
Thus it is important for the plant staff to verify setpoint acceptability based on the actual instrument span, vessel AT and channel specific scaling.on a periodic basis. This is particularly important when hot leg streaming characteristics are projected to change significantly over cycle life.
Verification of Acceptability for a Tsar er Tay. Window 1
In the performance of uncertanty ~1~1 dons for Precision RCS Flow calorimetric measurements, the calculation demonstrates a sensitivity to the value of Tuor due to the non-linear characteristics of the Steen Tables. The RCS Flow calorimetric measurament uncertanty is determined at a specific set of nominal full power conditions. Allowances are made for parameter measurement uncertamties and RCS Flow sensitivities are determined for a limited range about the nominal conditions. Significant variance from the assumed nominal conditions can affect the magnitude of calculated sensitivities. It has been '
determined that changes in Tn r can change the Tuor sensitivity magnitude and thus the RCS Flow calorimetric measurement uncertanty reported in the plant Technical Specific.ations. From an RCS Flow calorimetric measurement point of view, the extremes of a T uor or Taya window should be verified to be acceptable prior to' anticipated operation at those extremes. Changes in the minimum RCS Flow required to maintain consistency with safety analysis assumptions may be required. That is, the measurement uncertamty may increase over that previously assumed, thus requiring higher indicated flow values to compensate. ~
l Effects of T,v. Coastdown and Power Operation at Reduced T,y.
A T ,vo coastdown with a power reduction following the To reference profile, at the end of a fuel cycle, should not affect the OTAT, OPAT and VATEP protection funcaons since the normalization at full power ;
conditions should russin unaffected. However, it should be remembered that a reduction in T,yo will '
affect the NIS Power Range and Intermediate Range reactor trip and indication functions. With changes in Teoco, the indicated power on the NIS will change (due to increased downcomer water density acting u a shieldag increase), even at constant power. The operations staff should be sensitive to this effect when decreasing temperature c mmeuw.m
ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 8 of 9 A Tava coastdown without a corresponding power reduction, i.e., maintenance of full power by reducing TAva at the end of a fuel cycle, will affect OTAT, OPAT, NIS Power Range and Intermediate Range reactor trips. In this situation, T and T" should be renormalized as noted previously in this document.
The NIS Power Range and Intermediate Range indicated powers will also be affected, as noted above.
The rod control T,va reference profile should also be adjusted to reflect the full power operation at the reduced T,ya value if operation in Automatic rod control is desired. Otherwise, rod control should be placed in Manual and the operations staff must emulate the basic characteristics of the rod control system as if it were in Automatic. -
RECOMMENDATIONS To ensure operation consistent with the various assumptions of the safety analyses for the temperature related protection functions (OTAT, OPAT, VATEP) and to minimize operanonal margin impact, Westinghouse recommends that the following modifications to plant procedures be considered. Note that depending on the final choices made on operating practices and availability of setpoint margin (if explici,t allowances are included in the setpoint uncertamty or transient analyses calculations), modifications to the plant Technical Specifications may be appropriate.
- 1) The loop specific full power AT, should be extrapolated from measurements in the 75 to 90 %
RTP range and confirmed by measurement near 100 % RTP, The OTAT, OPAT and VATEP protection functions should be scaled, based on the extrapolation results and confirmed prior to extended operation at 100 % RTP. The valees for AT, should be determined acceptable via periodic surveillance of the loop specific, full power, indicated AT.
- 2) T and T", for OTAT and OPAT respectively, should reflect the loop specific, indicated, full power T ivo values determined at the beginning of each cycle and adjusted as necessary based on the periodic surveillance of Tava.
- 3) Tm, Teo, AT, T and T" surveillance and comparisons to Technical Specification values or other
. acceptance criteria should be performed utilizing time averaged values.
- 4) - Surveillance of T , Tea, AT,, T and T" should be performed on at least a quarterly basis.
- 5) At the time of normalm=*iam or surveillance of AT., T and T", the loop specific values for Tm and Tem hould s be venfied to be within the upper and lower bounds assumed in the safety analyses.
- 6) Prior to operanon at reduced full power loop AT values, with respect to OTAT and OPAT initial condition assumptions, the effects of such operation on the Channel Statistical Allowance and Total Allowance should be detenmined and confirmed acceptable.
B
' 7) Prior to operation at the extremes of a Tm or T,va window, i.e., at full power conditions sigah-tly ddierent than those assumed for RCS Flow calonmetric measurement uncertanty calculations, the effects of such operation on the RCS Flow calorimetric measurement uncertamty calculation should be detennined and the minimum RCS Flow required in the Technical Specifications adjusted as =~a==y.
vmmemm
t ESBU-TB-96-07-RO TEMPERATURE RELATED FUNCTIONS Page 9 of 9
- 8) The T,va reference profile for the rod control system should be adjusted prior to full power operation at T,va values less than that assumed in the safety analyses. If the T,vo reference is redetermined after reaching full power, rod control should be placed and maintained in Manual until adjustment of the T,vo profile is completed. The associated T' and T" values in the protection system should also be adjusted at the same time.
.e I
l i
14SSA.WFWDLflV446
i ATTACHMENT 1
~
INPO NUCLEAR NETWORK OE 7063,1/24/95 l
l 145eA.WPWUull/49el l l
Please note that access to NUCLEAR NETWORK is restricted to organizations authorimod by INPO. The information exchanged via this network is confidential and for the sole use of the authorized organization. Confidentiality is important to assure the open and frank exchange ofinformation among authorized organizations. These messages should not be published, disclosed, abstracted, or otherwise transferred
' 'in any form to any third party and their contents should not be made public without the prior written consent ofINPO.
OE 7063 I MERTOGUL 24-JAN-95 18:01 EST PACIFIC GAS AND ELECTRIC (PGE)- '
Subject:
Thermal Stratification in the RCS Hot LegsNNRA001 A NUCLEAR <
NETWORK 09/17/96' Message Retrieval
One message has been selected.
Units . . . . . . . . .. . . . Diablo Canyon 1 & 2 Doc Nos . . . . . . . . . . . 50-275/50-323
- Ratings. . . . . . . . . 1137 & 1164 MWe NSSS/AE . . . . . . .. . . . Westinghouse /PG&E Commercial Dates. . . . . . 5/7/85 & 3/13/86 Following elimination of RTD bypass piping from the reactor coolant system and installation of Eagle 21 process protection system equipment at the end of Diablo Canyon Power Plant (DCPP) Units 1 & 2 fuel cycle 6, thermal stratification in the RCS hot less was observed that was in excess of that observed previously by Westeghouse i
The efEset'of the strarem- was to introduce a non conservative bias in the measured Thot value in one loop of DCPP Unit 1. The nonanservative bias was in turn reflected in the calculated Tsve value. The effect of the nonenservatism was to calculate a Tsve value that was lower at 100 %
power than the actual Tave for the subject loop as i determined by primary and secondary calorimetric !
- calculations. This was doestmined not to be a 1 safety concern at DCPP because DCPP operates with the reactor control system referenos Teve (Tref) k
l.
t l'
l lower than the maximum value permitted by the 1 >
Technical Specifications.
In evaluating the effects of the stratification, -
i it was Jso determined that DCPP was operated with the T and T" setpoints in the Overtemperature l Delta T (OTDT) and Overpower Delta T (OPDT) trip .
l functions equal to the Rod Control System Tref, whereas Westinghouse assumed that loop specific reference values of Tave at 100% power were used for these terms. The ptarpose of using loop l specific values is to make the Tavg-T and Tavg-T" l
terms in the OTDT and OPDT trip equations nominally equal to aero at full power, in the l presence of any loop-to-loop variations in l measured Tave. Ifloop specific reference values l of Tave at 100% power are not used, there is a potential risk of calculating a non conservative i
l OTDT or OPDT trip setpoint; i.e., with indicated j
j 100% power Tave for a given loop less than the '
l reference T/f", the required trip could occur l late. This potential non conservative condition l
applies to all Westinghouse protection systems, because the observed loop-to-loop Thot variations 4
l are due to a combination of upper plenum flow !
asymmetry and hot les thermal stratification.
I DCPP operates with a control rod Tref of 4 degrees .
F lower than the maximum value permitted by the Technical Specifications. Operating at a lower -
i-Tave provided sufficient margin to account for the non conservative bias in one loop as well as not utilizing loop specific Tave terms in the OTDT and L OPDT trip functions. Therefore, the non-l conservative effects of not using loop specific
! T/f" values were bounded at DCPP. However, a non-l conservative condition could exist for other Westinghouse plants which operate with the T and T" values in the OTDT and OPDT trip equations equal to Traf, rather than loop specific values.
To avoid this potential non conservative condition, loop specific T and T" values, measured (and updated, if necessary) on a quarterly basis, must be used.
Westinghouse is aware of this issue and is currently evaluating any potential generic implications.
l ummmveu
Information
Contact:
John Hefier (415)973-9766/ Bill Bojduj (415)973-4568 e
4 0
4 14MA.WFWDull/496
.. - - _ _ _ _ - _