ML20147B294

From kanterella
Jump to navigation Jump to search
Responses to NRC Questions Re Fast Response Narrow Range, Thermowell Mounted Resistance Temp Detectors
ML20147B294
Person / Time
Site: South Texas STP Nuclear Operating Company icon.png
Issue date: 12/31/1987
From: Rice W, Tuley C
WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML16223A662 List:
References
WCAP-11702, NUDOCS 8801150193
Download: ML20147B294 (10)
v • d • e


Text

._

I WCAP-11702 e

i t

i RES10EEE 'IO NRC QUESTICtG RD3ARDING FASI RESPDNSE, NARROW RANGE, 'IHERGELL MXWrED RIDS -

IECENBER 1987 i i

C. R. 'IULEY

, TECHNICAL SPECIFICATICH SERVICES l W. R. RICE FIDID, RADIATIN AND SUPPJRT SYSTDS i

i i

r h

WESTINGHOUSE ELECIRIC CDRIGATICN j ICWER SYSIDE RJSINESS UNIT  ;

P. O. BOX 355 t PITISBURGH,PA 15230-0355 (

i I

i I

0 PDR D K0 P l

0 INIEX 0

2 REFERDiCES 7 O

I l

l 1

I l

l 1

l l

e

Qu_ggtion 1 Regartling the prchle of drift of the RID response tim identified in NURD3-0809 (Referen 1), hibe (a) the mthod(s) for checking RID response time, and (b) the safety a)lowanoe or other m thods to provide assurance that the imponse times do not drift outside acceptable limits between the 18 month checks.

Answer la

'Ibe response time of the individual RIDS will be checked using the Irep Orzent Step Response (IfSR) test prior to initial criticality. 'Ihe RID response time will also be checked as part of the data reduction fran the full power plant trip test which will be conducted during initial pcuer operations.

Answer Ib

'Ibe plant safety analysis has ==M an ce,arall protection channel response tim that has been shown to provide adequate protection frun design basis transients. 'Ibe various elements which make up this assumed response time were chosen so as to bourri the actual performance of each hart 3 ware constituent. When the channel respcnse time is measured it can be cestpared to the a=M value and the diffarence will be the em margin that ,

1 exists.

l 1

The present 18 mnth surveillance schedule is designed to monitor such critical hardware as the RIDS arrl determine if adverse trends may be developing. If such a trerri were to exhibit itself with this model RID /thernowell then at that time appropriate steps would be taken. The l present surveillance interval is appropriate for identifyirg those trends and no further steps need to be taken at this time.

ouestion 2 There is evidence of a slow drift in RID rm<iings with time, possibly all in one directim. Discuss how lorg tu nu drift would be detected before it l

l reached unacceptable levels. We staff has accepted one licensee's l

, ccmitment to replace two RIDS after each of the next two refueling outages to check the calibration of the renoved RIDS.

l*

Answer 2 Westinghouse reccanends the performance of an RID crossdibration during the heatup after each refueling. This procedure requires rultiple measurements at three or four different tetperatures. Tb date, Westir@.cuse l, l

has evaluated the data fmu over four hundred RIDS usirg this tedmique, far more than any other recent testinf. Several repeat tests performed one to 1

three years aput have not shown any indication of drift in only one direction. The results of the tests indicate that the RIDS drift less than

[ Ja,c a= =vi for uncertainty calculations for the protection systan. The 2

procedure scanitivity is sufficient to discern a randca drift of less than 1.0 degree F by one or several RIDS. At that time, either the calibration of the F/E(s) for the affected RID (s) would be adjusted to account for the shift, or the RID (s) wculd be declared inoperable and would be replaced.

Question 3 Failure of an RID in a particular loop would be signalled and alarmed by the deviation of the Tavg and delta T values for that loop ompared to the values for the, other three loops.

Question 3a What are the deviation values that will activate the alarm?

Answer 3a

'Ihere presently exists as part of the plant electronics both a Tavg arri a Delta T deviaticn alarm. This alarm ocupares the four signals; either Tavg or Delta T, one frca each channel, to a pre-set threshold value. This value is nominally set to + or - 2 degrees F and is adjusted during startup and subsequent operation such that it is just beyond the range of normal operatiry variaticos.

i 3 1

. l l

l I

I l

Question 3b What is the mininum taperature error in a sigle failed RID that

. will cause the deviation alarm to go off?

Answer 3b Tavg is calculated by aMig the average hot leg taperature (Thot) arri cold leg tecperature ( h ld) and dividim the sum by two. If we assune the deviation threshold renains at + or - 2 degrees F, then the deviation alarn will enunciate when the That average value has shiftal four (4) degrees F.

However, Thot average is first calculated by averaging three (3) independent RIDS. This Thot signal is then used to calculate Tavg. Since Thot relies on three (3) RIDS instead of one (1) the shift in any one RID could be tW.lve (12) degrees F. This still represents the same That shift of four (4) degrees that would have existed if the RID Bypass Elimination had not been inplemented and provides the same level of warning that the That channel is experiencing prtblems.

The same logic can be applied to tJ;e Delta T deviation alarm. Delta T is calculated by : That - 7 bold. In this case, Thot average is aM W to shift by two (2) degrees F before tho + or - 2 degree F threshold value is met.  !

Since there are three (3) RIDS used to generate Thot, any one RID cculd shift six (6) degrees F.  ;

4 i I

I

\s

'Ihe allowable channel to channel deviation remains the same as the original design. '1he additicnal shift that indivMm1 RIDS am allowed is a direct

. result of the averaging natum of the modification. 'Ihe impact of any one RID changing is r=+v=4 by a factor of three (3) and, themfom, its allowable shift can be larger.

Question 3c How often will individual RIDS be monitored in normal operation?

Answer 3c

'Ibe inilvidual RIDS will be monitored during the normal periodic surveillance routine when the bias daa is recorded and updated.

Question 4 Inlicate your plans to check and confinn the accuracy of the new method for hot leg tenperature measurements by caparison against other plant measurenents. 'Ibe staff has accepted one licensee's ccanitment on a plant which is rmovirg their existing RID bypass system, to obtain confirmatory information by caparing pre-installation and post-ins $allatico calorimetric  ;

data on RID tenperature measurments in their plant for matdting cperating coniitions. 'Ibe staff intends to review this confirmatory information.  ;

i 5 l

. l l

w

l o l Answer 4 i South Texas will enter initial operation with the themowell RID

. taperature measta==it only. There will be no prior operating data to use as a ocmparison to the present taperature measurements. The plant will be perfoming the normal intensive startup test program which is designed to evaluate plant system performance. This testing is mxh more involved than the nomal post-refueling startup testing that operating plants routinely perform. It is expected that the testing to be performed at South Texas will uncover any gross instrumentation errors prior to the plant entering carercial cperation. The successful ocmpletion of the startup test prcgram will provide confimation that the integrated plant systems, including primary system tenperature meastumedit, are perfomirg in an acmptable manner. No other special tests or emparisons are newM.

6

t BEFERENCES

1. NURD3-0809, Safety Evaluation Report, Review of Resistance 'hmperature Detector Time Resporse Characteristics, August 1981.
2. NURD3/CR-2928, Degradation of Ibclear Plant 'Dimperature Sensors, June 1987 f
3. K.R.Carr, An Evaluation of Industrial Platinan Resistance 'Ihermaneter

'hrperature - Its Measurment and Control in Science and Industry, ISA  ;

Publication, Vol. 4, Part 2, 1972, Pages 971-982. [

4. B.W.Mangum, 'Ibe Stability of Small Industrial Platinum Resistance

'Ibernaneters, Journal of Wrth of the NBS, Vol. 89, No. 4, '

July-August 1984, Pages 305-350.

7  :

I

[

l l

,

Retrieved from "https://kanterella.com/w/index.php?title=ML20147B294&oldid=2853276"