ML19332A852
| ML19332A852 | |
| Person / Time | |
|---|---|
| Site: | Vallecitos File:GEH Hitachi icon.png |
| Issue date: | 09/15/1980 |
| From: | ENGINEERING DECISION ANALYSIS CO., INC. |
| To: | |
| Shared Package | |
| ML19332A849 | List: |
| References | |
| EDAC-117-258.03, NUDOCS 8009180477 | |
| Download: ML19332A852 (25) | |
Text
'
EDAC 117-258.03 v
GENERAL ELECTRIC TEST REACTOR RESPONSE TO ADDITIONAL IN:0RMATION REQUEST REGARDING SEISMIC SCRAM SYSTEM Prepared for General Electric C mpany Pleasanton, California 15 September 1980 ENGINEERING DECISION ANALYSIS COMPANY,INC.
450 CALIFORN!A AVE., SUITE 301 BURNITZSTRASSE 34 PALO ALTO CALIF. 94306 6 FRANKFURT 70. W. GERMANY 8009180 y y
A 4
3
. TABLE OF CONTENTS Page i
P R EFA C E............................
1 RESPONSES....:..............,.........
12 Request #1..........................
12 Request #2..........................
13 Request #3......................-....
14 Request #4..........................
15 Request #5'..........................
17 I
Request #6..........................
18 Request #7..........................
19 REFERENCES i
APPENDIX A - Description of Triaxial Sensor 2
2 i
l-r.
I 1
1 PREFACE This document presents the responses to the requests for additional information regarding the seismic scram system at the General Electric Test Reactor.
These requests were received on 28 August 1980, and were in relation to Item 2 of the 14 August 1980 GE submittal to the USNRC (Reference 1).
Further examination of records from historical earthquakes indicated that it would be more conservative to utilize three-component triggers (two horizontal and one vertical), rather than the current two component (two horizontal) triggers. Therefore, the infonnation provided in this dccument is based on the premise that three-component triggers will be used. The type of trigger currently contemplated is described in Appendix A.
Table 1 of Reference 1 has been revised to reflect the use of three-component triggers, and the r*evised version is presented in this document. Table 2 lists the sources of the digitized records used in Table 1.
The information shown in Revised Table 1 is represented graphically in Figures 1 and 2.
These two figu_res show the envelopes of the absolute values of the acceleration time histories for the first one second after detecting 0.01. The envelopes include all records listed in Revised 9
Table 1.
Also acte that the seismograph stations from which the records in Table 1 were obtained employed only two-component horizontal triggers, except for the Imperial Valley and Coyote Lake stations, which employed three-component triggers.
If the two-component stations had employed three-component triggers, it is possible that they would have been tripped earlier upon detection of vertical motions.
If so, the acceleration values shown in Table 1 are conservative for the two-component stations, and actual accelerations were likely less than shown for the intervals after detecting 0.01g.
2 Recall that the times of key events in the scram system operation are as follows (Ref. 1):
Incremental Total Elapsed Action Time, sec Time, sec Closure of Switches 0
0 at 0.01-0.03g Disengagement of 0.18 0.18 Control Rods Begin Opening Emergency 0.19 0.19 Cooling Valves Control Rods at or below 0.30 0.48 12.2. inch withdrawal position Full Insertion of 0.50 0.68 Contr.o1 Rods Full Opening of Emergency 0.80 0.99 Cooling Valves and Completion of Scram Examination of the data in Revised Table 1 and Figures 1 and 2 shows that the scran action, which requires a maximum of slightly under one second, will be completed before consequential horizontal or vertical accelerations occur.
i
3 TABLE 1 (REVISED)
MAXIMUM INSTRUMENTAL ACCELERATIONS AFTER RECORDING 0.01g l
Earthquake and Maximum Acceleration (g) ir, Interval No.
Recording Station Component 0.25 sec.
0.50 sec.
1.0 sec.
1 Imperial Valley Horizontal 0.020 0.025 0.066 5-18-40 Vertical 0.075 0.140 0.210 2
Eureka Fed. Bldg.
Horizontal 0.023 0.027 0.032 12-21-54 Vertical 0.0 83 0.083 0.083 3
Helena, Montana Horizontal 0.022 0.028 0.042 College, 10-31-35 Vertical 0.012 0.038 0.047 4
Golden Gate, S.F.
Horizontal 0.017 0.017 0.026 3-22-57 Vertical 0.010 0.014 0.021 5
Holister City Hall Horizontal 0.033 0.045 0.112 4-8-61 Vertical 0.011 0.024 0.024 6
Parkfield Horizontal 0.010 0.010 0.018 6-27-66 Vertical 0.029 0.038 0.078 7
San Fernando (Pacoima)
Horizontal 0.053 0.077 0.131 2-9-71 Vertical 0.086 0.138 0.243 8
San Fernando Horizontal 0.013 0.014 0.014 (646 So. Olive Ave.)
Vertical 0.018 0.018 0.024 2-9-71 9
San Fernando Horizontal 0.021 0.021 0.021 (3047 Sixth St.)
Vertical 0.087 0.087 0.103 2-9-71 10 San Fernando Horizontal 0.037 0.037 0.037 (633 E. Broadway)
Vertical 0.043 0.043 0.096 2-9-71 11 San Fernando Horizontal 0.063 0.063 0.063 (Castalic Old Ridge)
Vertical 0.037 0.046 0.069 2-9-71 12 Imperial Valley Horizontal 0.005 0.006 0.007 Array 7 Vertical 0.024 0.047 0.047 10-15-79
4 TABLE 1 (REVISED)
-Continued-MAXIMUM INSTRUMENTAL ACCELERATIONS AFTER RECORDING 0.01g Earthquake and Maximum Acceleration (g) in Interval No.
Recording Station Canponent 0.025 sec.
0.50 sec.
1.0 sec.
13 Imperial Valley Horizontal 0.003 0.004 0.006 Array 8 Vertical 0.029 0.128 0.133 10-15-79 4
i 14 Imperial Val 1ey Horizontal 0.006 0.009 0.009 Array 5 Vertical 0.016 0.016 0.016 10-15-79 i
J 15 Imperial Valley Horizontal 0.007 0.008 0.009 Diff. Array Vertical 0.028 0.028 0.037 10-15-79 16 Imperial Valley Horizontal 0.004 0.005 0.006 Array 10 Vertical 0.025 0.025 0.034 10-15-79 17 Imperial Valley Horizontal 0.005 0.005 0.008 A. ray 6 Vertical 0.0 38 0.038 0.081 10-15-79 18 Imperial Valley Horizontal 0.010 0.011 0.023 Array 4 Vertical 0.022 0.023 0.028 10-15-79 19 Imperial Valley Horizontal-0.003 0.003 0.004 Municipal Airport Vertical 0.020 0.020 0.026 10-15-79 20 Imperial Valley Horizontal 0.003 0.003 0.004 Calexico Fire Station Vertical 0.039 0.048 0.048 10-15-79 21 Imperial Valley Horizontal 0.002 0.004 0.005 Holtville Post Office Vertical 0.019 0.025 0.042 10-15-79 1
l i
5 TABLE 1 (REVISED)
-Continued-
_ MAXIMUM INSTRUMENTAL ACCELERATIONS AFTER RECORDING 0.01g E5rthquake and Maximum Acceleration (g) in Interval No.
Recording Station Component 0.25 sec.
0.50 sec.
1.0 sec.
22 Coyote Lake.
Horizontal 0.023 0.023 0.048 Array 1 Vertical 0.033 0.043 0.051 8-6-79 23 Coyote Lake Horizontal 0.020 0 021 0.058 Array 2 Vertical 0.120 0.126 0.150 8-6-79 24 Coyote Lake Horizontal 0.011 0.017 0.029 Array 4 Vertical 0.094 0.194 0.194 8-6-79 25 Coyote Lake Horizontal 0.016 0.026 v.048 Array 3 Vertical 0.121 0.121 0.122 8-6-79 26 Coyote Lake Horizontal 0.018 0.035 0.062 Array 6 Vertical 0.038 0.083 0.083 8-6-79 27 Coyote Lake Horizontal 0.014 0.021 0.028 Coyote Creek Vertical 0.030 0:034 0.038 8-6-79 28 Gazli Horizontal 0.015 0.023 0.033 5-17-76 Vertical 0.010 0.010 0.121 29 Santa Barbara Horizontal 0.003 0.006 0.009 UCSB, Goleta Vertical 0.026 0.026 0.026 8-13-78
~. -
6 TABLE 2 SOURCE OF_ EARTHQUAKE RECORDS Records Source 1.
Imperial Valley Strong Motion Earthquake 5/18/1940 Accelerograms, Digitized and Plotted Data, Volume II, Earthquake Engineering Research Laboratory, California Institute of Technology, March, 1973.
2.
Eureka Federal Building
,12/21/1950 3.
Helena, Montana College 10/31/1971 4.
Golden Gate, San Francisco 3/22/1957 5.
Holister City Hall 4/8/1961 6.
Parkfield 7.
SanFernando(Pacoima) 2/9/1971 8.
San Fernando 646 South Olive Ave.
2/9/1971 9.
San Fernando 3047 Sixth Street 2/9/1971 10.
San Fernando 633 E. Broadway 2/9/1971 11.
San Fernando Castalic Old Ridge 2/9/1971 m_
7 TABLE 2
-Continued-SOURCE OF EARTHQUAKE RECORDS Records Source 12.
Imperia Valley Preliminary Summary of the Array 7 U.S. Geological Survey, 10/15/1979 Strong-Motion Records from the October 15, 1979, Imperial Valley Earthquake, Report 79-1654, U.S. Geological Survey, Menlo Park, October, 1979.
13.
Imperial Valley Array 8 10/15/1979 14.
Imperial Valley Array 5 10/15/1979 15.
Imperial Valley Diff. Array 10/15/1979 16.
Imperial Valley Array 10 10/15/1979 17.
Imperial Valley Array 6 10/15/1979 18.
Imperial Valley Array 4 10/15/1979 19.
Imperial Valley Municipal Airport 10/15/1979
8 TABLE 2
-Continued-SOURCE OF EARTHQUAKE RECORDS Records Source 20.
Imperial Valley Preliminary Sunnary of the Calexico Fire Station U.S. Geological Survey, 10/15/1979 Strong-Motion Records from the October 15, 1979, Imperial Valley Earthquake, Report 79-1654, U.S. Geological Survey, Menlo Park, October, 1979.
21.
Imperial Valley Holtville Post Office 10/15/1979 22.
Coyote Lake Canpilation of Strong-Motion Array 1 Records from the August 6, 1979 8/6/1979 Coyote Lake Earthquake, Report 79-385, U.S. Geological Survey, Mer.lo Park, October, 1979.
- 23. Coyote Lake a
Array 2 8/6/1979 24.
Coyote Lake Array 4 8/6/1979 25.
Coyote Lake a
Array 3 8/6/1979
- 26. Coyote Lake Array 6 8/6/1979
- 27. Coyote Lake Coyote Creek 8/6/1979
g t
,Y6.
9 TABLE 2
-Conti nued-SOURCE OF EARTHQUAKE RECORDS Records Source
- 28. Gazli, Soviet Union Digitized Accelerogram 5/17/1976 from the Destructive Gazli Earthquake of 17 May 1977, by V. V. Steinberg et al.
Institute of Physics of the Earth, USSR Academy of Science.
- 29. Santa Barbara Processed data from the strong-UCSB, Goleta-motion records of the Santa 8/13/1978 Barbara Earthquake of 13 August 1978. California Division of Mines and Geology, Sacramento,1979.
u m
. e.
10 QQ a
g g
. _ =:-f S,_
Q r
<E U
Ew a
W'
~
3 Y
8
--- =
to u
N a
t s
I Q
w u
e u
o O
N w
=
C t-=
x cr
-c r 5 t
w E d U
~
Q x e
~
to W f 5 o
a
~
Q
- h. %
h O O O
/*
M bJ W m:c
- c. x oa x w m i
l
~Q 5 %
L E
r I
e e
m
~
e D
ci 8
-u.
!~
l 1
- - = =
OQ F
Q Q
W W
CD C
W m
m O
O i
l Q
l Q-Q
-Q Q
Q l
(3)
NCI168313338 1
(
.s
3 00 1
C
\\
A 1
l 0
D
=
,I L
l EV S
5 E
N 7
L O
A I
T D
R A
E R
C E
L C
E I
C R
C T
A L
)
I C
D T
5 E R
E S
V g
(
D 1
M 0 U
a E
M 0 M
I X
G 1
A N
\\
T M
I D
F R
O O f
C E
E 8
P R
i' 5
O L
R 2
E E
V T
0 N
F E
A l
V 2
E I
RU I
G I
l F
Y 00 D
5 D
5 D
5 D
2 2
1 1
D D
0 D
D D
D D
U~ zc 1
1 i
l' t
12 REQUEST #1
" Describe the source of -the computerized earthquake records used to determine the maximum instrumental accelerations in Table 's and provide data plots of the time histories illustrating acce:eration levels from seismic scram actuation through the time to complete control rod and valve operation."
RESPONSE
Table 2 provides the source of the computerized (digitized) earthquake records used in Revised Table 1.
Figures 1 and 2 provide envelopes of the absolute values of acceleration time histories as explained in the Preface.
4
4 13 REQUEST #2 "The earthquake threat comes from two main sources:
a strike slip event on the Calaveras fault and a thrust event in the immediate vicinity of the plant. The data set presented comes mainly from strike slip events. The strike slip data set should be expanded to include additional records from the 10/15/79 Imperial Valley event, and the August 1979 Coyote Lake event. These include additional records within one kilometer from the fault and more distant records.
The cut off distance (5,10, 20 km) is dependent upon the values of acceleration that are deemed necessary (see Question 4).
A data set for thrust type events should be presented.
Near-field accelerograms from the 1971 San Fernando,1978 Santa Barbara,1976 Gazli, Soviet Union, and -1978 Tabas (Iran) earthquake are examples of records which should be examined."
RESPONSE
The data set has been expanded as suggested (See Table 1).
Adequate records of the 1978 Tabas, Iran earthquake record, suitable for digitization, are not presently available.
i
14 REQUEST #3 "The vertical component of acceleration is not addressed in your submittal. Demonstrate that significant loadings due to the vertical component will not develop prior to completing rod and valve operation."
RESPONSE
As explained in the Preface to this document, a sensor will be installed which will be triggered by vertical or horizontal motions.
The building structure and reactor pressure vessel are essentially rigid in the vertical direction, and there will be no amplification of vertical motions. Therefore, the snall vertical accelerations shown in Table 1 will not produce significant loads on the structure or systems, and the rods and valves will operate properly.
1
15 REQUEST #4
" Demonstrate that the seismic scram and valve actuation circuitry, the core with control rods in motion and actuated valves while operating will satisf actorily complete their function at the acceleration levels based on your examination. Your response should include the acceleration levels to which this equipment has been demonstrated operable by test and/or analysis.
Specific reference to previous submittals may be used."
RESPONSE
The discussion in Reference 1 and the Preface to this document states that all the electrical and electronic scram circuitry operates within 0.18 seconds of the seismic switch closure within 0.01g. The largest horizontal and vertical accelerations in Table 1 at 0.25 second after trip were 0.0639 and 0.129, respectively. These accelerations are clearly very low and non-damaging. The GETR scram and valve actuation circuitry operated properly after horizontal accelerations of about 0.1g (Greenville earthquake, January 1980).
Therefore, since the horizontal acceleration levels at 0.2 seconds after trip are within the range of this value,and the small vertical accelerations are non-damaging because of the substantial margin of safety in the design for vertical loads, the scram circuitry will operate satisfactorily to scram the GETR and initiate the required valve operations in a seismic event.
Note also that the stresses in the fuel elements are extremely small (on the order of about 10 psi during actuation and about 70 psi for the 0.75g criterion event.);
therefore, the fuel element assemblies are not damaged.
As also discussed in Reference 1 and the Preface to this document, the control rods will be at or below the 12.2 inch witMraw (i.e. shutdown) position within 0.48 seconds after detection of 0.01g. To ensure that the rods would drop, the control rod assemblies were tested for operability with a 1.0g side load imposed statically.
The force to move the assembly with tnis side load was 0.26W where W is the weight of the rod. Grarity and flow forces are clearly greater than this value and, therefore, the control rod will continue to move into the core (within the guide tube of each assembly) with a lg side load. This latter load is well in excess of
16 RESPONSE TO REQUEST #4 - continued the loads which might occur up to 0.48 seconds, the time after the 0.01g trip at which the control rods reach the 12-inch position and the reactor is shut down.
The actuated valves, including those actuated by the seismic scram circuitry (emergency cooling power-operated valves, pressurizer valve and fuel. flooding system admission valves), have been proof tested (Ref. 2) and shown to be ooerable with a conservative envelope of vibratory motion that would be seen by each valve during the maximum postulated seismic event. Therefore, these valves will operate properly at any time during or after the seismic event.
,+
E 17 REQUEST #5 "The_ reactor core, as analyzed for design basis earthquake. loads, has control rods fully inserted. Therefore you should verify that the rods will be fully _ inserted before significant earthquake loading (i.e., acceleration which exceeds that level determined in 4. above)."
i
RESPONSE
~ The rods will be inserted before significant earthquake loadings as explained in the response to Request #4.
b e
b 4
s
~
r e
t,
18 REQUEST #6 "Please verify that the ' emergency cooling valves' referenced in your submittal include all valves which must operate to mitigate the consequences of a seismic event."
RESPONSE
The valves that must operate to mitigate the consequences of a seismic event were tested with a conservative envelope of vibratory motion that would be seen by each valve during the maximum postulated seismic event (Ref. 2 - Valve Test).
Included were proof tests that the valve would perfonc. the required operation (e.g., opening, closing, maintaining 2
pressure integrity, etc.). The energency cooling power-operated valves, pressurizer valves and fuel flooding system admission valves are the only valves for which initiating action is by seismic trip or scram circuitry.
The emergency cooling power-operated valves and the fuel flooding system admission valves begin to open and the pressurizer valve to close within 190 milliseconds after triggering of the scram system.
The remainder of the valve operation is complete within a total of one second from scram seismic trip; and, further, the valves have been qualified for. operation (for in excess of the vibratory motion to which the valve would be submitted for the maximum postulated seismic event.
e
o 19 REQUEST #7 "Since the seismic switches are located within the building your argument regarding the conservatism of using instrumental values of acceleration (last paragraph, Comment 2) is not clear.
Please explain."
RESPONSE
The seismic trigger is not located in the free field. Therefore, there is no advantage as described in the referenced paragraph, and the paragrdph should be deleted.
~
e o<
20 REFERENCE
- 1. ' General Electric (R. W. Darmitzel) letter to USNRC (D. G. Eisenhut), 14 August 1980.
Subject:
" Reliability and Response Action Time for the General Electric Test Reactor (GETR)
Scram System" 2.
Engineering Decision Analysis Company, Inc., " Qualification of Safety Related Valves, General Electric Test Reactor," prepared for General Electric Company, San Jose, California, EDAC-117-217.09, 30 June 1978.
3602
-or
.wr-
-m v,
y o-,
.. ~ --
APPENDIX A DESCRIPTION OF TRIAXIAL SENSOR
ie APPENDIX A DESCRIPTION OF THE KINEMETRICS TRIAXIAL SEISMIC TRIP SYSTEM The Kinemetrics triaxial seismic trip system consists of two major units.
These are a Model TS-3 seismic switch and a Model SP-1 seismic switch power supply.
The system output is a relay contact change of position when the seismic switch is accelerated either vertically or horizontally to a level greater than the preset level.
This relay will be connected into the GETR scram system in the same way as the present seismic switch output.
The seismic switch contains three orthogonal electromagnetic transducers.
The transducers are small moving coils that produce a voltage proportional to acceleration.
This coil voltage is amplified to energize the output relay.
The acceleration trip level is adjusted by changing the sensitivity of the amplifier.
The seismic switch power supply contains a de power supply and a battery.
The de power supply powers the seismic transducers and serves as a charger for the battery.
The battery is connected to float on the power supply output, thus it is a backup power supply.
Therefore, the de power supply and the battery provide redundant power to the triaxial seismic trip system.
The relay provides a contact change of position as a system output that requires a manual reset to return it to normal.
A test switch provides a voltage across the transducer coil that simulates an acceleration that would trip the system.
It is
... -. r r
APPENDIX A planned to set the trip points at 0.01g.
The Kinemetrics Company is willing to certify this equipment to operate as designed at acceleration levels up to and including 0.5g and in the environment in which it will be located.
- L'..
.