ML20199C330
| ML20199C330 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 06/13/1986 |
| From: | Devincentis J PUBLIC SERVICE CO. OF NEW HAMPSHIRE |
| To: | Noonan V Office of Nuclear Reactor Regulation |
| References | |
| RTR-REGGD-01.075, RTR-REGGD-1.075 SBN-1107, NUDOCS 8606180071 | |
| Download: ML20199C330 (14) | |
Text
. _
l SEABROOK STATION Enginsering Office l e ) (-
.b Pub 5c SeMce of New HampW*e June 13, 1986 NEW HAMPSHIRE YANKEE DIVISION SBN-1107 T.F.
B7.1.2 United States Nuclear Regulatory Commission Washington, DC 20555 Attention:
Mr. Vincent S. Noonan, Project Director PWR Project Directorate No. 5
References:
(a) Construction Permits CPPR-135 and CPPR-136, Docket Nos. 50-443 and 50-444 (b) PSNH Letter (SBN-979), dated March 31, 1986,
" Electrical Separation Criteria", J. DeVincentis to V. S. Noonan (c) PSNH Letter (SBN-1100), dated June 10, 1986,
" Electrical f eparation Criteria", J. DeVincentis to V. S. Noonar Subj ect :
Electrical Separation Criteria; Additional Information
Dear Sir:
Reference (b) provided the results of a test program and analysis which justified electrical cable and raceway separation criteria less than the standard criteria provided in FSAR Appendix 8A, IEEE Standard 384-1974, and Regulatory Guide 1.75 Revision 2.
During various dis-cussions concerning this submittal, the NRC reviewer indicated that the Main Control Board and aluminum sheath (ALS) lighting cable portions of the analysis were acceptable but that there were some concerns with the test results involving power cables because of the fault current levels and test durations.
The reviewer requested additional information on these concerns.
Attached is our response to this request for additional information.
We believe these responses address the ERC reviewers concerns and should result in acceptance of the conduit-to-conduit and con-luit-to-cable tray separation criteria as presented in Reference (b).
8606180071 860613 PDR ADOCK 05000443 A
- PDR, Bwi Seabrook Station Construction Field Office. P.O. Box 700 Seabrook, NH O3874
\\
United Sttts2 Nuc1 car R2gulstory Commission Attention:
Mr. Vincent S. Noonan Page 2 As stated in Reference (c), Seabrook's Separation Criteria is in conformance with the requirements of FSAR Appendix 8A, IEEE Standard 384-1974, and Regulatory Guide 1.75 Revision 2, by use of a combination established by test and analysis as presented in Reference (b) and the attached additional information.
Accordingly, we do request that the acceptability of Seabrook's Separation Criteria be reflected in the next supplement to Seabrook's SER.
Very truly ours, j
John DeVincentis Director of Engineering Attachment cc:
Atomic Safety and Licensing Board Service List 4
Diane Curran, Esquire Calvin A. Canney Harmon & Weiss City Manager 2001 S. Street, N.W.
City Hall Suite 430 126 Daniel Street Washington, D.C.
20009 Portsmouth, NH 03801 Sherwin E. Turk, Esq.
Stephen E. Merrill, Esquire Office of the Executive Legal Director Attorney General UsS. Nuclear Regulatory Commission George Dana Bisbee, Esquire Tenth Floor Assistant Attorney General Washington, DC 20555 Office of the Attorney General 25 Capitol Street Robert A. Backus, Esquire Concord, NH 03301-6397 116 Lowell Street P.O. Box 516 Mr. J. P. Nadeau Manchester, NH 03105 Selectmen'a Office 10 Central Road Philip Ahrens, Esquire Rye, NH 03870 Assistant Attorney General Department of The Attorney General Mr. Angie hachiros Statehouse Station #6 Chairman of the Board of Selectmen Augusta, ME 04333 Town of Newbury Newbury, MA 01950 Mrs. Sandra Gavutis Chairman, Board of Selectmen Mr. William S. Lord RFD 1 - Box 1154 Board of Selectmen Kennsington, NH 03827 Town Hall - Friend Street Amesbury, MA 01913 Carol S. Sneider, Esquire Assistant Attorney General Senator Gordon J. Humphrey Department of the Attorney General 1 Pillsbury Street One Ashburton Place, 19th Floor Concord, NH 03301 Boston, MA 02108 (ATTN: Herb Boynton)
Senator Gordon J. Humphrey H. Joseph Flynn, Esquire U.S. Senate Office of General Counsel Washington, DC 20510 Federal Emergency Management Agency (ATTN: Tom Burack) 500 C Street, SW Washington, DC 20472 Richard A. Hampe Esq.
Hampe and McNicholas Paul McEachern, Esquire 35 Pleasant Street Matthew T. Brock, Esquire Concord, NH 03301 Shaines & McEachern 25 Maplewood Avenue Thomas F. Powers III P.O. Box 360 Town Manager Portsmouth, NH 03801 Town of Exeter 10 Front Street Gary W. Holmes, Esq.
Exeter, NH 03833 Holmes & Ells 47 Winnacunnet Road Brentwood Board of Selectmen Hampton, NH 03842
'RFD Dalton Road Brentwood, NH 03833 Mr. Ed Thomas FEMA Region I Peter J. Mathews, Mayor 442 John W. McCormack PO & Courthouse City Hall Boston, MA 02109 Newburyport, MA 01950 Robert Carrigg Town Office Atlantic Avenue North Hampton, NH 03862
SBN-1107 ATTACHMENT 1 Request for Additional Information on Separation Testing Reference - PSNH Letter (SBN-979), " Electrical Separation Criteria",
dated March 31, 1986, J. DeVincentis to V. S. Noonan
Background
The above referenced letter provided an analysis of a test program which justified electrical cable and raceway separation criteria less than standard criteria given in the regulatory standards (IEEE-384 and Regulatory Guide 1.75).
During various discussions with the NRC, the reviewer requested additional information on certain portions of the test program and analysis involving the level and duration of fault currents utilized on the power cables tests, on cable surveil-lance testing, and on a conduit-to-conduit test anomaly.
Our responses to these requests are given below.
Discussion A.
Fault current levels and durations.
The fault current levels for the power cable tests were based on the locked rotor amperes (LRA) of the largest motor that a particular cable was sized to supply at Seabrook. The LRA was selected because it produced a current in the range that previous testing had snown was worst case in terms of heat generation during a f ault.
In addition, the LRA failure condition is the only failure mode th5t could result in a sustained overload current flow and would result in a relatively long term temperature interaction during a fault with the most probability for producing potential detrimentel inter-actions.
For added conservatism, the primary protective device was assumed to fail and 10% margin wsn added to the LRA values.
During one of the 500MCM triplex f ault cable tests (see page 12 of of referenced letter), target cable damage occurred as a result of fault cable ignition when the fault current was permitted to flow until the f ault cable open circuited.
After evaluation of actual field installations, it was concluded that the fault current would not flow for this long and would be interrupted when the motor pigtails melted open (see Appendix C of Enclosure 1 to referenced letter).
Since the analysis showed that the motor pigtail would melt open before the fault cable ignited, we concluded that the target cables would not have been damaged and the test would have been acceptable.
SBN-1107 ATTACRMENT 1 (Continued) i The NRC reviewer had a concern with this approach, and felt that a cable fault could occur upstream of the motor (between the motor and circuit breaker) which could potentially cause detrimental interaction.
For this fault location, credit could not be taken for the motor pigtails melting open to limit the fault duration. We acknowledged that a cable breakdown could occur which could result in a leakage current flowing through the cable insulation and jacket as a conductor to conductor or conductor to ground f ault.
However, we did not feel that a cable fault of this type where only the insulation and jacket would limit the current could support the relatively high current flows (1760A and 2116A) involved in the test without quickly degrading to a short circuit which would cause the protective devices to open.
Although no specific time can be calculated for this degradation to a short circuit, it is our judgement that it would occur much faster than the time we have included for pigtail melting such that our analysis is very conservative.
Further, we have discussed this type of failure with the two cable manufacturers who supplied power cable to Seabrook ( Anaconda and Okonite) and the test lab who conducted our test program (Wyle) for an independent expert opinion.
They agreed with us that a breakdown could occur but that it would not support a high current flow of the levels used in the test program for the time frame experienced in the testing without degrading to a short circuit.
Based on the above discussion, the reviewer agreed that a cable fault could not support the high current flows mentioned above for long periods of time without degrading to a short circuit which would operate the backup protective device. However, it was requested that we verify that there would be sufficient fault current available to trip the backup device for a cable at the end of any feeder cable.
A typical circuit is shown in Sketch No. 1.
All feeder cables to loads, for example motor control centers, unit substations, motors, etc. have been evaluated to determine the available fault current for a cable fault just before the point of termination at the load.
It has been determined that the minimum available f ault current is 3904A. This represents a phase-to-phase fault which produces less than the 3 phase fault (the minimum 3 phase fault current is 4487A). The long time setting of the largest backup protective device is 2400A.
Since the minimum fault current exceeds the backup protective device setting, the device will trip for the postulated cable fault. At the 3904A level, the backup protective device will trip in the range of 105 to 225 seconds.
SBN-1107 ATTACHMENT 1 (Continued)
The above indicates that the backup device will operate to isolate the cable fault.
The NRC requested that we demonstrate that there will be no detrimental effect on target cables during the time required for the backup device to open.
We attempted to demonstrate the above utilizing analytical methods but the results were inconclusive.
Therefore, we decided to perform a test designed to simulate the phase to phase fault and determine the ef fects on the target cables.
The test setup used a 500MCM fault cable installed in a cable tray as shown in Sketch 2.
A conduit containing target cables was installed 1/2" above the cable tray side to evaluate the effects of the cable fault on target cables. The test procedure specified to stop the test at 250 seconds which represents the 225 second operating time of the backup protective device plus 10%.
The applied f ault current was 3904A (phase-to-phase fault current from above). About 94.28 seconds after applying the fault current, the cable conductors shorted together and the test set current was increased to 7830A (actually the maximum current available from the test set) to simulate additional current that would occur in the field as a cable continually degrades. At about 157.6 seconds, the fault cable ignited.
At about 167.37 seconds, the fault cable open circuited.
The total insulation burn time was about 9 minutes.
The target cables conducted current throughout the test.
The post test insulation resistance and hipot functional tests were acceptable. The hipot test on the 2/C-14 cable was 2200 VAC for 1 minute and on the instru-ment cable was 1600 VAC for 1 minute. No physical damage was observed on the target cables.
The cable insulation remained pliable and flexible throughout the test.
The maximum temperature on the outside of the conduit was about 672.5'F.
The maximum temperature on the target cables in the conduit was about 488.4*F with a duration above 480*F of 1 minute 20 seconds which was short enough to not damage the target cables.
This shows that the target cables were not degraded by the cable fault. We consider these results acceptable.
A summary of the test results is given in a letter from Wyle Labs dated June 12,1986 provided herewith.
The reviewer also requested that we document the surveillance testing performed on cables which could detect potential cable degradation. This information is provided as Item B.
Futhermore the reveiwer requested that we address the measures that will be taken by the plant staff in the event of a cable ignition in a tray.
This fire will be considered an Operational Event and station procedures, Station Information Report (SIR) will report and evaluate the event. The SIRS are distributed to all plant departments for evaluation and remedial actions (replacement of damaged cables etc).
In addition, depending on the duration of the fire and the potential effects on safety systems, this event will be reported to the NRC as part of the Seabrook Station Emergency Plan.
SBN-1107 ATTACHMENT 1 (Continued)
The NRC reviewer has indicated that the above discussion is acceptable to address the concerns on cable faults and that this results in the 1/2 inch conduit-to-cable tray separation criteria being acceptable.
B.
Cable Surveillance Testing The following list provides the frequency of surveillance testing for cables. This testing will be insulation resistance (megger) measurements.
MOTORS Feeder cables to all safety and non-safety related motors 50HP and larger are tested every refueling.
Feeder cables to all satety related motors smaller than 50HP are tested every refueling.
Feeder cables to non-safety related motors smaller than 50HP are tested every 5th refueling.
UNIT SUBSTATIONS Feeder cables to safety and non-safety related unit substations are tested every 3rd refueling.
MOTOR CONTROL CENTERS Feeder cables to safety related Motor Control Centers (MCC) are tested every 3rd refueling.
Feeder cables to non-safety related Motor Control Centers are tested every 5th refueling.
MISCELLANEOUS Feeder cables to miscellaneous loads (except Motor Control Centers) fed from Unit Substations are tested at an interval not to exceed every 5th refueling.
C.
Conduit-to-Conduit Test Anomaly During discussion of the conduit-to-conduit separation tests, the reviewer indicated that the test results may not be acceptable for Test C2-T1 because of the reference to pigtail melting for limiting the fault current duration.
We indicated that there was a Notice of Anomaly report which further discussed the problems encountered in this test.
The reviewer requested a further discussion of this anomaly.
SBN-1107 ATTACHMENT 1 (Continued)
Test C2-Tl (see Sketch 3) used a 500MCM triplex f ault cable installed in a rigid conduit with a rigid and flexible conduit containing target cables mounted 1/4 inch vertically above and parallel with the conduit containing the fault cable. The f ault cable ignited during the test and caused target cable degradation.
As documented by the Notice of Anomaly report, the target cable degrad-ation occurred because of the test setup and the physical proximity of the ends of the target and fault conduits (i.e., the ends of the target conduits were directly above the end of the fault conduit.
For the rigid target conduit, the flames from the fault conduit im-pinged on the target cables as they exited the target conduit.
The heat from the flames heated up the target cables and caused the 2/C-14 cable to short to the conduit at the point where the cables exited the conduit (see Sketch 3).
This situation is not representative of actual field installations because the target conduit and fault conduit would not be terminated in close proximity to each other such that a possible ignition would not effect the target cables. As part of the post test functional testing, the degraded section of the 2/C-14 cable was cut off and the remaining section of 2/C-14 cable within the conduit passed the high pot and megger tests. The triaxial and twisted pair target cable also within the rigid conduit were not effected by the flames and had acceptable meggers and hipot functional tests.
Since the cables passed the electrical tests, we considered the test acceptable to demonstrate 1/4 inch conduit-to-conduit separation.
For the flexible target conduit, the heat from the f ault conduit caused the outer coating of the flexible conduit to melt and drop onto the f ault conduit which in itself did not appear to be a problem. However, the flames out the end of the rigid f ault conduit impinged on the end of the flex conduit and caused the flexible conduit coating to ignite.
The direct burning of the coating caused the temperature inside the flexible conduit to increase to a level that damaged the target cables.
Although this situation is not representative of actual field install-ations, it was not conclusive on whether the target cables would have been degraded had the flexible conduit coating not ignited. Therefore to further evaluate flexible conduit separation, we included a flexible to rigid conduit separation setup as part of Test Cl-T2.
As discussed on pages 10 and 11 of Enclosure 1 of reference letter, test Cl-T2 produced acceptable results to justify 1/4 inch conduit separation.
Based on the above discussion, the reviewer indicated that the conduit-to-conduit separation criteria of 1/4 inch was acceptable.
SBN-1107 e
seemme ese iese a e,cs e amou.
June it, le88 Reference No: 48381M-002 Publie Servlee of New Hampshire Company e/o United Engineers at Constructors Seabrook Station Seabeook, NH 03874 Attention Randy Jamison
Reference:
Purchase Order No. 9763.011-44145 C/03 Gentlemen:
Me foI1owing preliminary results are provided for tiiw Conflgurullos Numlier 7 Lost conducted for the Public Servlee of New Hampshire Company's Seabrook Nuclear Power Plant.
The test consisted of a Triplex 500 MCM fault cable mounted in a emble tray with 1-inch conduit mounted 1/3-inch above the top of the sideralls eroaning perp.Jieuler to the fault oable at the emble tray centerline. Three target cables (a triasdal, T.P.
16 AWG, and 2/C 14 AWG) were energized throughout the test and ran through the condult.
Two arrays of thermoeouples were am mounted above the fault cable approximately 24 Inches to either side of the conduit.
1A Purpose The purpose of this test was to determine the amount of heat delivered to the environment from a fault cable subjoeted to a short-circuit current for a duration
!!mited by the beekup protective devlos.
I.8 6
The cables tested and the electriest powering were es follows:
omhan asse runetson Isomtson Powertug 3-1/C 500 MCM Pault Cable Tray 430A, 700A, 3904A, 7830A*
T.P.16 AWG Target Conduit 50 VAC,1A Triaxial Target Conduit 50 VAC,1A 2/C 14 A WG Target Conduit 120 VAC,10A
- Initial warmup current, final warmup current, fault current, and short-circuit current.
SBN-1107 r
Public Service of New Hampshire Reference No 48841H-002 Seabrook Station June 12,1986 Attna Randy Jamlaon Page 2 S.0 Reedte Results of the test are triefly summarized below:
Matrimaam CaMe Sise _
Funetion Loestion Tenessatures 3-1/C 500 MCM Pault CaMe Tray 1500.4*F 2/C 14 AWG Target Inside conduit -
488.4'F Thermocouples were mounted 4" to both sides of the fault eable T.P.18 AWG Target inside Conduit -
454.3'F Thermocouples were mounted 3" to both eldse of the fault emble 1/C 18 AWO Target Inside conduit -
302.9'F Thermoeouples were mounted 8" to both sides of and above the fault esbia N/A Conduit Exterior surface 872.S*F facing fault cane N/A T/C Array 1/2" above fault cable 1822.5'F N/A T/C Array 1" above fault caWe 1828.6'F
~
N/A T/C Array 8" above fault eaWe 1434.9'F 0
N/A T/C Array 8" above fault eaMe and 828.9 F 8"lateraDy distant N/A T/C Array 12" above fault cable 1189.9'F N/A T/C Array 24" above fault cable 768.0'F Time to ignition:
157.8 seconds Burn time:
9.0 minutes Time to open-etroult:
187.37 seconds. Fault current was 3900 amperes initiaRy.
Sheet-efecult occurrent after 94.28 manande at this autrent.
Short-elroult current was 7830 amperes.
b\\
SBN-1107 Public Servloe of Now Hampshire Reference No: 48361H-003 seabrook Station June it,1988 Attn: Randy Jamison Page 8 Target osble performance:
Instrumentation cables (triaxial and T.P.16 AWG) conducted 1A at 50 VAC throughout the Overeurrent Test. The control cable (2/C 14 AWG) conducted 10A at 120 YAC throughout the Overeurrent Test. AR target cables passed Post 4}vereurrent Test Punettonal Teste.
Although the 2/C 14 AWG and T.P.16 AWO target cable Jackets egeoeded their rated tempernture and were above 480 P for 85 seconds, their performance was not affected. No damage was observed durtrg the visual Inspection following the Overeurrent and Post-Test Functional tests. The cable remained pliant and flexible and performed sastisfeetorily thmughout testing and it's performance is therefore judged not to have been degraded.
4.3 r wishimsmme k.
The above results generated the conclusion that design / construction is acceptable for a physleal, vertleal separation distance of 1/2-inch between a rigid steel condult containlig low-energy embles and the top of a power cable tray siderall, when a short-circuit level fault occurs in the power cable.
Other data was collected to determine the heat released to the environment by a short-circuit current of duration limited by the backw protective devlee.
4 Should you have any additional comments or require additional Information, please feel free to contact the undersigned.
81nately.
WYLE RA'IORM Bassess tiens ik
_jN WY
% 'CG"I2'%
{ fFge y Nauer
- 5. M. MeMicken 5F. Contreets Administrator Project Engineer
/jmk I
r
~.
SBN-1107 O
1
}
m hv__ a m p. w _ <~<-c - -s u [
4,,.a) _.
p l
a
=
L8 Ad l
i
= 4
.. - -. -...-=.
l 4
U #
.m..
SBN-1107 smua f _e.. m u m ca 3rato me
... ~
~
m+
m R;e:ACo.Au;t f
%s i%%es m
A A
q..
i i
SIC-l'l 8"j l*
rPt6 W 6 n
ijc. )15 TecI Ax14 L m
2 5"
1 B"
i v..ac.u
( " d ",
j Odn'dsd M'.cus o$ TC. LocaT..ons o nIa,9eT to'o\\e.'r,3;ac Rie;d Co~duit rt o n ou b de of co*4=it
(
'Tcuget as u c.au /
ty2 t
e O@O i
cm-y~'
- t. va c..we.
s o 9-a
SBN-1107 i'
% tek 3 M d 6 'E l--
- .o.p.mm+
- ^d. %sd cAkle dkump4L
~
~
f,.rgd
..__.f Q e 5
\\
g e,4 ;\\, __ _
dN g)
G.\\t C.M._
~
c.o wa a:+
n
>/Y
'22d C 0Wd a'si i
l