ML20042F868
| ML20042F868 | |
| Person / Time | |
|---|---|
| Site: | Prairie Island  |
| Issue date: | 04/30/1990 |
| From: | Larson C NORTHERN STATES POWER CO. |
| To: | Davis A NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| References | |
| NUDOCS 9005100097 | |
| Download: ML20042F868 (5) | |
Text
{{#Wiki_filter:. /p/ w9 Northem Steles Power Company 414 Nicollet Mail Minneapolis Minnesota 55401 1927 Telephone (612) 330-5500 April 30, 1990 Mr A Bert Davis Regional-Administrator Region III U S Nuclear Regulatory Commission 799 Roosevelt Road Glen Ellyn, Illinois 60137 PRAIRIE ISLAND NUCLEAR GENERATING PLANT Docket Nos. 50 282 License Naa. DPR 42 50 6 DPR 60 306 Supplemental Information on Response Time Testinc of Instrumentation
References:
(a) Letter, C E Larson (NSP) to A B Davis, NRC Region III Regional Administrator, dated October 31, 1988, " Response Time Testing of Instrumentation". (b) Letter dated August 22, 1988 from Mr W L Axelscn, Chief. Projects Branch 2, Region III, USNRC, Inspection Reports 50-282/88012(DRP) & 50 306/88012 02(DRP). (c) Letter, C E Larson (NSP) to A B Davis, NRC Region III Regional Administrator, dated May 31, 1989, " Supplemental Information on Response Time Testing of Instrumentation". In Reference (a), we committed to perform additional confirmation and testing for Prairie Island Unit 1 and Unit 2 which was required to fully address the. response time testing issues identified in Unresolved Items 282/88012 01 and 506/88012 02 of Reference (b). In response to that commitment and in an attempt to confirm the instrument delay times assumed in the plant safety analysia., testing was performed on the Prairie Island Unit 2 Reactor Protection System during the refueling outage in the spring of 1989. In Reference (c)', we reported the results of that testing and described the changes we have made to the Prairie Island responso time test procedures. We have concluded the testing on Unit 1 during the recent refueling outage. The j[ results are described in the attachment. No additional changes to the current g Prairie Island response time testing program are necessary, /' / m U 900510o097 900430,, ADOCK 0500 g g FDR l 0 \\ ~ MAY 21990
P' ~ ~ { T a r p 'j ..g 4 <.y q.
- sr
. j.. '.f .I ^ z ~ v u 1 g'- .j... 8 A B Davis April 30, 1990' ' Page 2 - y -Please contact us if you have'any questions related to.our response, j t-1 1 4 .l 1 l 4
- y.
C E 1Ars "I .R '- Vice Presi ent-k -Nuclear' Generation p I i ? c:. Senior: Resident Inspector,' NRC-NRR Project Manager., NRC - G Charrioff 7, a n F :
Attachment:
Response to Response Time Testing, Issues j cg. a.. . j p- .y. 4'I n s, t v. ..[
- [q s
'b -' i. 1 -L .[ Y'i
- -lN ;,, '. - { '.
h-i l ue i i A, f f s - L 'k i k i-. I i .-? <..s .y a, 3.f '-{ s -- ! '" 'r - -. o..e i; l -! .h ? 4 7 h I' t k ,( l t t,{ 'g 4 ;. ;P _g ,n " .,q.- n. -,,.. a.. .. ~ 4,.
? .e 6 ATTACHMENT PRAIRIE ISLAND NUCLEAR GENERATING PIANT RESPONSE TO RESPONSE TIME TESTING ISSUES RESULTS OF UNIT 1 TESTING In response to the concerns raised in Unresolved Items 282/88012-01 and 306/88012 02,. portions of the Prairie Island Units 1 r.nd 2 reactor protection systems were tested in an attempt to confirm the instrument delay times assumed t-in the Updated Safety Analysis Report and the plant Reload Safety Evaluations i (RSEs). The best estimate results of the Unit 2 testing were presented in an earlier report (May 31, 1989). This report documents the best estimate.results J l of the Unit 1 testing. Where possible, sensor, logic and motion delay times were measured. The L, protection channels tested, the resulting delay times, and the delay times used in the safety analysis are summarized below and are shown in Table 1. Nuclear Instrumentation The nuclear instrumentation sensor delay times were not measured because there is no proven method for testing those types of sensors (Standard ISA-S67.06 Section 12.3). The nuclear instrumentation sensor delay times are assumed to be negligible with respect to the total trip delay time. The values for the Unit I nuclear instrumentation logic delays shown in Table 1 are the average of the Unit 1 measurements taken during each refueling outage and include the measurements taken during the most recent Unit 1 outage. It can be seen that the total best estimate delay times (for high neutron flux and high neBative flux rate) are bounded by the values assumed in the RSEs. Overpower and Overtemnerature Delta T The overpower delta-T and overtemperature delta T sensor delays are made up of two parts. First is the transport delay required for a change in the coolant I temperature at the core exit to propagate to the RTD manifold. A very conservative value of 2.0 seconds is assumed. The second portion of the sensor delay time is the RTD response time.- The overpower delta T and overtemperature delta T trips are used to mitigate the effects of slow transients which have a gradual increase in reactivity and coolant temperature. Specifically, these two trip functions mitigate the effects of the Uncontrolled Rod Withdrawal (slow rate) and the Uncontrolled Boron Dilution'translents. It was determined that for these two transients, the measured RTD time constant j is very nearly equal to the RTD response time. The best estimate Unit 1 RTD 4 response time, which is based on the average of the measured value;, is.3.071 i seconds. Therefore, the best estimate Unit 1 sensor delay for the overpower j delta T and the overtemperature delta T is 5.071 seconds. The values for the overpower and overtemperature delta T logic delays shown in Table 1 are the average of the measurements taken during each refueling outage and include the measurements taken during the most recent Unit 1 outage. i 1
t 9 i 4 Attachment i Page 2 of 3 As can be seen in Table 1, the total overpower and overtemperature delta T trip delays are bounded by the values assumed for the RSE. Low Pressurizer Pressure l t The best estimate value of the Unit 1 low pressurizer pressure sensor delay time is 0.340 seconds. This value is the third fastest sensor delay time out of the four sensors tested. The logic for this trip is 2 out of 4 Taking the third fastest conservatively assumes that the fastest sensor is out of service. The Unit 1 logic delay value provided in Table 1 is the average of the logic delay times measured during the initial plant startup and the most recent Unit-1 measurements. The total best estimate low pressurizer pressure delay time is bounded by the value assumed for the RSE. Low Reactor Coolant Flow s The best estimate for the Unit 1 low reactor coolant flow sensor delay, which is based on the averago of the measured values, is 0.410 seconds. The best estimate Unit 1 logic delay time shown in Table 1 is the average of the logic delay times measured during the initial plant startup and the most recent Unit 1 measurements. The total best estimate low reactor coolant flow total delay time is calculated to be 0.581 seconds, which is higher than the 0.45 second total delay assumed in the RSE. A similar result occurred from'the Unit 2 testing which was documented in the previous report (May 31, 1989). Unit 2 was found to have a measured best estimate delay of 0.491 seconds. As a result of the findings for Unit 2, the locked rotor event was re evaluated using a conservative bounding value of 1.05 seconds for. the low reactor coolant flow total trip delay. Tne results of this study were documented in the May 31, 1989 report. The conclusion of that study was that for both Units l'and 2, the current RSEs vere still valid. Assuming a 1.05 second bounding trip delay did not reduce'the margin of safety for either unit. In the May 31, 1989 report, it was stated that the Unit I delay' time would be measured to insure there were no-changes in the conclusions of that report. t Since the Unit 1 delay time is well below the bounding value of 1.05 seconds, the conclusions drawn in the May 31, 1989 report remain valid for both units. For i future RSEs, an appropriately conservative value for the total delay time will be used such that the best estimato delay time is bounded. Morion Delav The motion delay is defined as the interval from the time a trip signal is sent to the trip breaker to the time that the trip breaker opens. The best estimate l Unit 1 motion delay shown in Table 1 is the average of the measurements taken l during each refueling outage and include the measurements taken during the most recent Unit 1 outage, i
o. .= . ~... Attachment ~ Page 3 of 3 TABLE 1 Prairie Island Unit 1 Trip Delays (seconds) Best Estimate Trip Function Sensor Delay logic Delay Motion Delay Total Trip RSE Value Delay High Neutron Flux High Setpoint - updated: 0.0 0.030 0.082 0.112 - previous: 0.0 0.028 0.082 0.110 0.35 Low Setpoint - updated: 0.0 0.028 0.082 0.110 - previous: 0.0 0.028 0.082 0.110 0.35 High Negative Flux - updated: 0.0 0.172 0.082 0.254 Rate - previous: 0.0 0.171 0.082 0.253 0.35 Overpower Delta-T - updated: 5.071 0.201 0.082 5.354 - previous: 4.30 0.200 0.082 4.582 5.85 Overtemperature - updated: 5.071 0.259-0.082 5.412 Delta-T - previous: 4.3 0.256 0.082 4.638 5.85 Low Pressurizer - updated: 0.340 0.117 0.082 0.539 Pressure - previous: 0.4 0.120 0.082 0.602 0.85 Low RC Flow - updated: 0.410 0.089 0.082 _0.581 - previous: 0.2 0.064 0.082 0.346 0.45*
- See Low Reactor Coolant Flow discussion on page 2.
._}}