ML18081B174
| ML18081B174 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 03/03/1980 |
| From: | Librizzi F Public Service Enterprise Group |
| To: | Ross D Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8003110573 | |
| Download: ML18081B174 (14) | |
Text
Public Service Electric and Gas Company 80 Park Place Newark, N.J. 07101 Phone 201/430-7000 March 3, 1980 Mr. Denwood F. Ross, Jr., Acting.Director Division of Project Management Off ice of Nuclear Reactor Regulation U.S. Nuclear Regulatory Corninission Washington, D.* c. 20555
Dear Mr. Ross:
MODIFICATION OF SMALL-BREAK LOSS-OF-COOLANT ACCIDENT OPERATOR GUIDELINES SALEM GENERATING STATION UNIT NO. 1 DOCKET NO. 50-272 In compliance with your letter of December 27,.197 9 to Cordell Reed, our response is attached.
If you have any questions, please do not hesitate to contact us.
Attachment CENTENNIAL OF LIGHT Ill.
,~',/
,,\\ c'"
' ~' '
1879 1979 Very truly yours,
~
Frank P. Librizzi General Manager -
Electric Production 8008110' -~'?3
I I
- 1.
- 2.
- 3.
- 4.
ATTACHMENT A copy of the detailed error analysis and its effects on the calculation of subcooling temperature margin is attached.
The application of the subcooling criterion for RPI termina-ti6n for non-LOCA events to existing operating procedures has been discussed in the Westinghouse Owner's Group Letter OG-29, dated January 23, 1980 from Cordell Reed to Mr. Denwood F. Ross, Jr.
Our calculations show that we have provided a minimum actual margin of 20°F in subcooling temperature indication after discounting the effect of possible maximum ins.trumentation error.
Therefore, we presently do not plan to add any additional instrumentation.
Sufficient instrumentation is available on redundant emergency power supplies to permit the operator to perform the actions defined in the procedure for the subcooling criterion for HPI termination.
ATTACHMENT-A:
- p. 1 of 12 ERROR ANALYSIS OF THE SUBCOOLING MARGIN READING:
- 1.
- 2.
- 3.
- 4.
5.
SALEM NUCLEAR GENERATING STATION PUBLIC SERVICE ELECTRIC & GAS CO.
CONTENT:
PrincipJe of Analysis Error Calculations in the Pressure Channels
. Error Calculations in the Temperature Channels Error Calculations in the Sub cooling Margin Tables: Analysis Summary NOTE: THE ERRORS IN THE SUBCOOLING MARGIN READINGS FOUND IN THIS ANALYSIS HAVE TO BE APPLIED.TO THE MINIMUM INDICATED SUBCOOLING MARGIN SET-POINT OF 50°F.
Prepared by Date
( Sanjit K. Bardhan)
Feb.22.1980 Page 2
4 7
11 12
(
No.
t
- p. 2 of 12 ANALYSIS OF ERROR IN SIGNAL CHANNELS:
Assumptions:(l) All the sources of error in any instrument ( or measure-( 2 )
ment subsystem ) are assumed to be random variables in the statistical sense. E.. represents the random variable lJ due to the ~th source of error in the ith instrument in a particular signal channel.
Each E.. is statistically characterized by a Normal lJ (Gaussian) distribution with zero mean (i.e. systematic or bias error is zero) and standard deviation d...
lJ (3) Some of the Eij 1 s may be statistically dependent. The statistical correlation is assumed to be 100% for the sake of conservativeness.
(4) Individual instrument ( or measurement subsystem )
errors are assumed to be statistically independent (Ei).
......, it will be normally distributed with zero mean and standard deviation d..
l
( 5) Any E.. 1 s contributing less than 1% accuracy to E lJ i
may be neglected.
Derivation:
For zero bias or systematic error in any E..
1 s, the accuracy is equal lJ to the precision.
- Let, e..
lJ accuracy (precision) for E..
lJ
- e.
= accuracy (grecision) for E.
l l
The random variable E with standard deviation d represents the overall signal channel error and is given by, E = E1 + E2 + *......
( 1)
The given accuracy figures eijs have confidence levels of better than
ATTACHMEllJjA:
- p. 3 of 12 95% i.e. the probability is greater than 95% that the true value will be within a range of ~eij with respect to the instrument reading due to E.. alone.
lJ e.. = 2d..
lJ lJ
( 2)
When Eik'Ei 1,.*.** are statistically dependent, the random variable defined by, E.k 1
= E.k + E.l +...*..*
l',...
l l
( 3) and is also normally distributed with standard deviation d.k 1 l
which is given by, d
ik,1,..* -
dik + ail + * * * * *
( 4) 2 2
2 di = dil + di2 + * * * *. + ( dik + di 1 +
2
..... )
2 2
(1/2) ( eik + eil + *.**. )
For a confidence level of at least 95% in E.,
l 2
2 2
2 2
e i = ( 2 di)
= e i 1 + e i 2 + * * * * * + ( e i k + e i 1 + * * * * * )
For a confidence level of 95% in the signal channel, (2a) 2 = (2d ) 2 +( 2d ) 2 +
1 2
e
=Ve f + e ~ + ****.
( 7)
( 5 )
( 6) i.e. the overall signal channel accuracy is equal to the square root of the sum of the squares of the individual instrument/or measurement subsystem accuracies ~nd is also defined as the root mean square (rms) error. It may be expressed either in units of a physical variable or in percentage of a reference span.
ATTACHMENT-A:
- p. 4 of 12 A. Errors in the Reactor Temperature Measurement Channel as used in the Subcooling Margin Calculations:
Reference Span = 700 °F Regular Channel:
Incore Thermocouple ~
Computer Signal Conditioner ~
~
Computer Normal Operating Condition:
Thermocouple:
Combined accuracy (e )
(Mr. Wassel,W, Tel. ~on. 2.7.80)
Computer Signal Conditioner:
Dependent accuracies:
Reference accuracy Drift (e 22 )
Independent accuracies:
Repeatability (e 23)
Resolution (e 24 )
e2 = + J ( e21 + e22 ) 2+
(e21) 2
+ e2 e23 24
= + J(o.05 + o.o5) 2+ 0.05 2+ 0.025 2 The channel accuracy is given by, e(nl)
== + J er + e~
+ J 0.7143 2 + 0.115 2 Small Break LOCA Condition:
+ 5 °F
+ 0.7143 %
+ 0.05 %
0.05 %
= +
== + 0.05 %
== + 0.025 %
+ 0.115 %
+ 0. 72 35 %
+ 5.1 °F The effects of a small break LOCA are assumed to result in a containment temperature of 165 OF and a corresponding pressure of 8 psig.
The channel accuracy is not affected by the changes in the environment as the components in this channel are unaffected by it. Hence, the channel accuracy e(al)
== e(nl)== + 5.1 °F Back-up Channel #1:
Incore Thermocouple ~
Precision Chart Recorder Normal Operating Condition:
Thermocouple:
Combined accuracy (e 1 )
(From Reg. Ch.)
== + 0.7143 %
e ATTACHMENT-A:
- p. 5 of 12 Test Instruments:
Combined accuracy (e8 )
Precision Chart Recorder:
Independent accuracies:
Accuracy (e 1 )
Resolution re 3 )
Reading error f e 33 )
(Mr. Andy Smith, Honeywell, Tel. Con. 2.13.80) e
= + I e2
+ e2
+ e2 3
v 31 32 33
= + J 0. 2 2 + 0. 0 7 2 + 0. 0 18 2 The channel accuracy is given by,
= + O.l oj,
= + 0.2 oj,
= + 0.07 oj,
= +(l/4)(1/2
= + 0.125 OF
- 0. 018 oj,
= +
= + 0.213 oj, OF)
= + 5.26 OF Small Break LOCA Condition:
The channel accuracy e(a2) = e(n2) = + 0.75 oj, Back-up Channel #2:
= + 5.25 OF Hot Leg RTD ~
Amplifier
~ Isolater --?i-Chart Recorder Normal Operating Condition:
RTD:
Independent accuracies:
Accuracy (e 41 )
= + 3 'fo (includes process measurement error,reference error per Mr. Dick Miller,W, Tel. Con. 2.7.80)
Calibration Curve-accuracy (e 42 )
= + 0.003 'fo Test Inst. accuracy (e 4 ~)
= + 0.1 %
= +
= + 3.002 'Ii Amplifier:
Combined accuracy (e 5 )
= + 0.1 %
Isolater:
Combined accuracy (es)
=+0.l'fo
ATTACHMENT-A:
- p. 6 of 12 The Chart Recorder:
Independent Accuracies:
e 7 channel e(n3) =
=
=
Accuracy (e71)
Repeatability (e72)
Deadband (e 73 )
Reading error (e 74 )
J e~l 2
2
+
+ e72 + e7 3 +
+ J 0.5 2 + 0.2 2 + 0.12 accuracy is given by, J e2 2
2 2
+
+ e5 + e6 + e 4
7 J 3.002 2 2
0.1 2
+
+ 0.1
+
Small Break LOCA Condition:
2 e74
+ 0.36 2
+ 0.656 The channel accuracy e(a3) = e(n3) = + 3.08 %
2
+ 0.5 %
= + 0.2 %
= + 0.1 %
"+(l/4)(10°F)
= + 0. 36 %
= + 0.656 %
= + 3.08 %
= + 21.56 °F 0
+ 21.56 F
ATTACHMEN.:
- p. 7 of 12 B. Errors in the Reacter Pressure Measurement Channel as used in the Subcooling Margin Calculations:
Reference Span = 3000 psig
- .Regular Channel:
Transmitter ~
Isolator ---7 Computer Signal Conditioner -
~
Computer.
Normal Operating Condition:
Transmitter:
Dependent Accuracies:
Reference accuracy (e11)
Drift ( e 12 )
Temperature effects (e 13 )
Independent Accuracies:
Deadband ( e 14 )
= + 0.5 %
= + 0.5 %
0.5 %
= +
= + 0.01 %
= + J (0.5 + 0.5 + 0.5) 2 + 0.01 2
= + 1.5 %
Isolator:
Combined accuracy (e 6 )
Computer Signal Conditioner:
Combined.accuracy (e 2 )
(From Temp. Ch.)
Test Instruments:
Combined accuracy (e8 )
The channel accuracy is given by, e(nl) e~ + e~ + e~
= +
Small Break LOCA Condition:
Transmitter:
Dependent Accuracies:
Reference accuracy (e 11 )
Drift (e
)
- Temperatti?e effects (e 13 )
Independent Accuracies:
Deadband (e 14 )
+ 0.1 %
= + 0.115 %
= + 0.1 %
= + 1.51 %
+ 45.35 psig
= + 0.5 %
= + 0.5 %
= + 3.47 %
= + 0.01 %
The ATTACHMEN~: p. 8 of 12 el = + J (ell + el2 + el3) 2 + ei4
= + J (0.5 + 0.5 + 3.47) 2 + 0.012 = ~ 4.47 %
- Temperature effects at 165 OF are being cal-culated as the square of the ratio of (165/280) multiplied by the accuracy at 280 °F because the accuracy is affected by the additional heat energy absorbed by the transmitter at an accident condition.
Isolator:
Combined accuracy (e 6 )
= + 0.1 %
Computer Signal Conditioner:
= + 0.115 %
Combined accuracy (e 2 )
(From Temp. Ch.)
Test Instruments:
Combined accuracy (e8 )
= + 0.1 %
Bias Error (e~
Due to rise in the containment pressure from 0 psig (approx) to 8 psig during accident condition, the gage pressure instruments will have bias errors of
(-)8 psig. This type of bias error can be factored out with the help of the containment pressure instru-mentation.
Otherwise it will tend to indicate a lower value £or the Subcooling Margin than the act-ual. For the sake of conservativeness the bias error will not be factored out.
channel accuracy is given by, e(al)
J e~
2 + e2 + e2
= +
+ e 6 2
8 J 4.47 2 0.12 2
0.1 2
= +
+
+ 0.115
+
= + 4.474 %
= + 134.2 psig Back-up Channel #1:
Transmitter ~
Isolator ----?> Chart Recorder Normal Operating Condition:
Transmitter:
Combined accuracy (e 1 )
(From Reg. Ch.)
+ 1.5 %
- p. 9 of 12 Isolator:
Combined accuracy ( e )
= + 0.1 %
6 Chart Recorder:
Combined accuracy (e7)
= + 0.656 %
(From Temp. Ch.)
Test Instruments:
Combined accuracy (es)
+ 0.1 %
The channel accuracy is given by, e(n2)
J e~
2 e2 2
+
+ e6 +
+ es 7
j 1.52 0.1 2 0.656 2 2
= +
+
+
+ 0.1
+ 1. 643
= + 49.3 psig Small Break LOCA Condition:
Transmitter:
Combined accuracy ( e )
+ 4.47 %
(From Reg. Ch.)
1 Isolator:
Combined accuracy ( e )
= + 0.1 %
6 Chart Recorder:
Combined accuracy ( e 7)
= + 0.656 %
Test Instruments:
Combined accuracy ( e )
= + 0.1 %
s The channel accuracy is given by, e(a2)
J e~
2 2
2
+
+ e6 + e7 + es j 4.47 2 0.1 2 0.656 2 2
= +
+
+
+ 0.1
= + 4.52
= + 135.6 psig
.Back-up Channel #2:
Transmitter ~
Isolator ~
Indicator Normal Operating Condition:
Transmitter:
Combined accuracy (el)
= + 1.5 %
(From Reg. Ch.)
L ATTACHMENa:
Isolater:
Combined accuracy (e 6 )
Indicator:
Independent Accuracies:
Accuracy (e 91 )
Linearity(e 92 )
Repeatability (e 93 )
Deadband (e
).**
Reading err§i (e 95 )
+ J e~l + e~2 + e~3 + e~4 + e~5
= +
Test Instruments:
Combined accuracy (e8 )
The channel accuracy is given by, 2
2 2
J e2 e(n3) = +
+ e6 + e
+ e8 1
9
= + j 1.5 2 + 0.1 2 + 1. 64 2 Small Break LOCA Condition:
Transmitter:
Combined accuracy (e 1 )
(From Reg. Ch.)
Isolator:
Combined accuracy Indicator:
( e )
6 Combined accuracy (e9 )
(From Norm. Cond.)
Test Instruments:
Combined accuracy (e8 )
The channel accuracy is given by, e( a3) = + J e~ + e~ + e~ + e~
2
+ 0.1
= + J 4.47 2 + 0.1 2 + 1.64 2 + 0.12
- p. 10 of 12
= + 0.1 %
= + 1.0 %
= + 1.0 %
= + 0.5
= + 0.5 %
= +(l/4)(50 psig)
= + 12.5 psig 0.42 %
= +
+ 1. 64 %
= + 0.1 %
= + 2.23 %
= + 66.8 psig
= + 4.47 %
= + 0.1 %
= + 1. 64 %
= + 0.1 %
= + 4.77 %
= + 142.9 psig
~-=-=-=-=~~- ------....,
e ATTACHMENT-A:
- p. 11 of 12 C. Errors in the Reactor Subcooling Margin Calculations:
The Subcooling Margin is given by,
= T(P)
(C-1) where, T(P) = A non-linear function of the coolant pressure in OF. It is found from the Steam Table (ASME) for a given pressure level.
Total Error in ~T
+ Error in T(P) + Error in Tact (C-2)
The worst case results from the maximum positive value of
( ~Treading -
~Tact). This corresponds to the situation when ep is positive and eT negative.
(_c - 3)
The same ep in psig will result in different ep's in °F correspon-ding to the different coolant pressure levels.
Let, ep(jk) =The coolant pressure error (psig) in kth pressure instrument channel (k=l,2,3) under jth operating condition (j= normal (n), accident (a)).
epi(jk)= Tsat error (°F) due to ep(jk)corresponding* to the coolant pressure level reading of Pi (i=l,2,3).
eT(jk) =Tact error (9F) in kth te~Rerature instrument channel (k=l,2,3) under j operating condition.
(C-4)
The different values for ePi(jk),
ep(~k) &_eT(jk) are being shown in Table-I. The different values for e4~~ are being shown in Table-II.
Sample Calculations:
Assumed Pressure Reading P1 = 1765 psig = 1780 psia Operating Condition: Small Break LOCA Accident (a)
Pressure Instrument Channel: #2 (BU#l)
=Temperature Instrument Channel: #1 (Reg.)
- From the Steam Table & egun(C-4),
ep1 (a2)= T(l780) - T(l780-135.6) = 619.47 - 608.88 = 10.89 OF (shown in Table-I) is given by, The total error in Subcooling Margin e~T(a2l)p 1 = ep1(a2) + eT(al) = 10.89 + 5.1 = 15.99 OF (shown in Table-II)
~-
- p. 12of12
> TABLE-I: TEMPERATURE &PRESSURE CHANNEL ERROR MATRIX AT DIFFERENT OPERATING CONDITIONS & COOLANT PRESSURES.
Assumed TEMP CH. ERRORS PRESS Cl:I
- ERRORS Operating Indicated.t\\e g *
~Uff.l BUffC.
Reg.
~U'l'f*.l BU'ff2 NOTE Cond,ition Pressure
( psi a )
0 OF OF OF OF OF F
psi psi nsi ePl(nl) ePl(n2)
-ep1(n3)
P1= 1780 3.57
~ 3.88 l0 5. 28 eT(nl) eT(n2) eT(n3) tTp;Cn1)~
co
' ep2l ni: I
. eP21 n3) p = 1880 3.42
-r.r. 3.72
..-..m 5.05
-c.o 2
r-1.q C\\J '<It l0 c.o Normal 5.10 5.26 121.56 i::
i::
i::
11 11
......... 11 p -
2000 2.73 Po!
- 3. 7 3 Po!
4.27 Po!
3-ep3(nl)
QJ ep3 (n2)
QJ e~~(n3)
QJ
- ePl (al)
C\\ ep1(a2) c.o ePl(a3)
O'l Small P,= 1780 10.77 10.8 9
. 11.49 Break eT(al) eT(a2) eT(a3)
~tii IJ) eP2(a3)
C\\J eP2( al) eP2( a2)
..-..!<)
..-..'<It LOCA P2= 1880
=
=
=
10.29 r-1 r-1 10.41 C\\J r-1 10.98 10 r-1 Cll Cll a:l Ac.cident
. 5.10 5.26 ~1.56 ep3l al)
-11 ep3(ai:1
-11 "p3l a3J
......... 11 Po!
Po!
Po!
P3= 2000
- 9. 98 QJ 10.09 QJ 10.64 QJ TABLE-II: TOTAL SUBCOOLING MARGIN ERROR MATRIX AT DIFFERENT OPERATING CONDITIONS & COOLANT PRESSURES.
MAXIMUM ERRORS IN PRESSURE & TEMPERATURE CHANNEL COMBINATIONS 8.67 8.8 3 25.13 16.03 32.33 P1 =1780
- 8. 98 9.14 25.44 16.15 32.45
- 10. 38
- 10. 54 26.84 16.75 32.45 8.52
. 8. 68 24.98 15.55 31.85 p 2~1880 8.82
- 8. 98 25.28 15.47 31.97 10.15 10.31 26.61 16.24 32.54 7.8 3
-7.99 24.29 15.24 31.54 P3=2000 8.83 8.99 25.29 15.35 31.65
- 9. 37 9.53 25.8 3 ep3(a3) 15.74 15.90 32.20