ML13330B430
| ML13330B430 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 11/19/1988 |
| From: | Medford M SOUTHERN CALIFORNIA EDISON CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NUDOCS 8811220430 | |
| Download: ML13330B430 (13) | |
Text
Southern California Edison Company P. 0. BOX 800 2244 WALNUT GROVE AVENUE ROSEMEAD, CALIFORNIA 91770 M.0.MEDFORD November 19, 1988 TELEPHONE MANAGER OF (818) 302-1749 NUCLEAR REGULATORY AFFAIRS U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D.C. 20555 Gentlemen:
Subject:
Docket No. 50-206 Nuclear Instrumentation System Upgrade San Onofre Nuclear Generating Station Unit 1 The purpose of this letter is to provide additional information which supports the NRC staff's review of the Nuclear Instrumentation System (NIS) upgrade.scheduled to be installed during the Cycle 10 refueling outage. This information has been previously discussed with and provided to the NRC staff during telephone.discussions on November 17 and 18, 1988, with SCE and Westinghouse. Provided as enclosures is the information responding to the staff's concerns. provides the discussion on the battery power supply. It discusses the discharge testing that will be performed during the Cycle 10 refueling outage and the details of the loading profile for the first minute on the battery. The design calculation, DC 1604, which includes the battery load profile was provided to the NRC on November 18, 1988. Enclosure 2 is a discussion which provides information on the routing of the cables for the existing power supplies from the battery to the vital buses and to the regulated buses. Enclosure 3 provides a discussion of the compliance with Regulatory Guide 1.75 for the upgraded NIS from the detectors to the NIS cabinets and to the control room panels. It discusses the interfaces with the existing plant systems. provides information prepared by Westinghouse regarding the NIS hardware and software. It provides in detail the similarities and differences between the standard Westinghouse supplied NIS and the SCE NIS, the Regulatory Guide 1.97 flux monitoring system at Vogtle and the SCE system, and the microprocessors utilized in each system. Discussions on the isolation devices used in the NIS, the software verification and validation program and the noise and fault tests conducted for the field installation are also provided.
Also as discussed with the NRC staff, the RPS single failure analysis submitted March 11, 1987, which is being revised to include the NIS replacement, uses the applicable sections of IEEE 279. The analysis addresses each module or tag numbered device. It looks at the credible failure modes for these devices on both the input and output side of the 8-' 11220430 119 FPDR ADOCK 05000206 PDC
Document Control Desk November 19, 1988 devices. Specifically, it addresses channel common or interface devices and power supplies to determine if there are any channel common failure susceptibilities. Results are tabulated in standard FMEA format for each of the four power range channels, two intermediate range channels and channel common features. The aspects of concern in this analysis are reactor trips (including P7 and P8 permissives) and dropped rod rod stop because these features are specific assumptions or are credited by the accident analyses. The revision is scheduled for completion by November 21, 1988.
The physical separation of the vital bus and regulated bus power supplies has also been the subject of discussions with the NRC staff. SCE has indicated that this subject has been raised as an issue from the Systematic Evaluation Program (SEP), in that SEP Topic VI-7.C.2 addresses the physical separation of the vital bus and regulated bus power supplies and the associated cables.
SCE' s resolution of this issue has been delayed due to potential impact resulting from the single failure analysis, the environmental qualification and Regulatory Guide 1.97 efforts undertaken over the past two years. As indicated in our letter dated November 18, 1988, the report resolving the SEP issue is presently scheduled to be submitted to the NRC on May 31, 1989.
The submittal will resolve the issue of the power supply physical separation not only as it relates to the NIS and Regulatory Guide 1.97 but to other safety related systems and components that are powered by the vital buses and regulated buses.
If you have any questions regarding this, please call me.
Very truly yours Enclosures cc: C. M. Trammell, NRR Project Manager, San Onofre Unit 1
- 3. B. Martin, Regional Administrator, NRC Region V F. R. Huey, NRC Senior Resident Inspector, San Onofre Units 1, 2 and 3
0nclosure1 RESPONSE TO NRC ISSUES ON BATTERY *1
- 1. Testing of Battery #1 Battery #1 is scheduled to have both a performance test and a service test performed during the upcoming Cycle X outage. The performance test is scheduled because of the 60 month interval requirement. A service test is being performed as a result of the load profile changes. Battery voltage will be recorded during the service test at required intervals. The number of tests expected to be run on this battery during its installed life is approximately 20 which is well below the manufacturer's recommendation of no more than 2 discharge tests per year or a total of 40 for the 20 year life. Therefore no degradation of battery capacity is expected.
- 2. Basis for First Minute Loading on Battery #1 The battery loading calculation for Battery #1 was performed in accordance with the guidelines of IEEE Std 485. The method used for identifying the load for the one-minute period is per IEEE 485 Section 4.2.3, where a discrete sequence is established and the maximum current at any instant is the one-minute period load. The one minute load profile was generated by reviewing elementary diagrams and vendor data to establish the correct time sequencing. As shown in the load profile in the calculation (Table 5.1.142, page T107), the maximum current (1285 amps) occurs at the 1.000 second to 1.249 second interval of the 0-1 minute period. The majority of the sequence changes are from the 4160V and 480V buses' breaker operations. This sequencing is governed by the Load Sequencer controls which are tested periodically for proper time intervals. The remaining load intervals are based on motor acceleration times and relay operation times.
The one-minute load capacity at the 85% replacement value is calculated to be 1426 amps(Attachment 1). This provides a margin of 131 amps with respect to the calculated one minute load of 1285 amps. If the one-minute loads that are not controlled by the Load Sequencer are conservatively assumed to operate during-the one-minute peak interval(i.e. Gen/Xfmr Relay Bus-56A, Oscillograph-3.5A, DG Fuel Oil Pump-10.5A), this additional 70 amps would still be within the 131 amps margin. Also, the peak load of 1285 amps decreases to below 1000 amps at the 8-second mark which provides additional margin.
The battery terminal voltage expected during the first
,one-minute of the load profile was calculated (Table 3.3, page T60) to be 108.3 volts, which is above the minimum end-of-discharge voltage of 106 volts.
In summary, there is sufficient margin available in the battery to safely supply the one-minute load requirement while maintaining the required minimum battery voltage.
ATTACHMENT 1 125V DC BATTERY NO. 1 SIZING CALCULATION DC-1604, REV. 5
- 1. Maximum Current During First Minute Period Assuming no margin left and the battery is at its end-of service life condition, i.e., 85% capacity which corresponds to the replacement value, using Table 5.3.1A, page T115:
Needed size = 17 x 0.85 =
14.45 pos. plates Uncorrected size = 14.45 / 1.11 = 13.02 pos. plates Maximum section size = 13.02,- 0.29 = 12.73 pos. plates If the first minute is the section that governs the cell sizing:
Maximum first minute load = 12.73 x 112 = 1426 amps Physical Separation of Existing Vital/Regulated Bus Supply Cables The Train A 120 VAX vital bus systems is configured such that each vital bus can receive power from either of two redundant train sources of power via an automatic transfer switch. The normal source of power is supplied by Train A 125 VDC Bus #1 thru an inverter for each vital bus and the alternate source of power is provided by Train B MCC #2 thru a stepdown transformer. Each regulated bus is supplied by its respective vital bus.
The inverters are housed in separate enclosures. The automatic transfer switches and vital buses share a common enclosure and the regulated buses share a common enclosure.
The power for the intermediate range channels is supplied by Regulated Bus
- 1 and #2. Supply cables from 125 VDC Bus #1 to Inverters #1 and #2 are routed together in the same cable trays. Supply cables from Inverters #1 and #2 to Automatic Transfer Switches #1 and #2 are routed together in common cable trays. The supply cable from the alternative source of power to the automatic transfer switches is routed in a raceway separate from the inverter output cables. Connections between automatic transfer switches and vital buses are made within the common enclosure. Supply cables from Vital Buses #1 and #2 to Regulated Buses #1 and #2 are routed together in common cable trays.
In summary, existing supply cables for the regulated buses are routed together and share common raceways.
01130 R.G. 1.75 Compliance Physical Separation of New NIS Channels Description of Proposed NIS Raceway Configuration Beginning at the detectors, a separate and dedicated existing embedded conduit is utilized for each NIS channel to exit the bioshield. New channelized conduits and short sections of new dedicated channelized trays are provided from the bioshield to the penetrations. A new dedicated penetration is provided for each NIS channel.
New channelized conduits are provided from the penetrations to the NIS cabinet in the control room.
New dedicated channelized conduits are provided for each channel of power from the existing regulated buses to each bay of the NIS cabinet.
All NIS channelized outputs are provided new channelized conduits from the NIS cabinet to the interface with existing plant electrical/control panels and to the new coincidentor cabinets. The non safety, NIS outputs are routed in separate new conduits and/or new enclosed trays. These outputs are electrically isolated, from the safety related portions of the NIS. The covered trays ensure existing plant systems are not degraded. The coincidentor safety outputs are routed in train aligned conduits up to the interface with existing plant control panels.
Physical and electrical separation of NIS channelized circuits per R.G.
1.75 is achieved except as follows:
- 1. New channelized conduits do not fully meet the IEEE-384 separation distances to existing plant cable tray. However, new dedicated conduit provides a much greater level of protection than the existing NIS cable routing and the separation distances have been optimized to the extent achievable given existing plant structural/electrical configuration.
- 2. Channelized NIS protection outputs and R.G. 1.97 channelized indication signals once inside the existing main control room 3-console (C03) are bundled into separate channelized wiring bundles. The internal (to the 3-console) routing is optimized to the extent practicable to separate the bundles. Also the replacement mode selector switch improved the separation distances between channels compared to the existing mode selector switch. However, the installation of barriers between channels at the mode switch and at the R. G. 1.97 indicators is not practicable and are not provided.
In summary, the proposed NIS upgrade separation achieved is a significant improvement over the existing NIS since channelized circuits are routed in new dedicated conduits and separation at interface locations with the existing plant are optimized to the extent practicable.
01080 WESTINGHOUSE STANDARD 4-BAY NIS VS. SAN ONOFRE 4 BAY NIS The following tables provide information that describes the differences and similarities between Westinghouse standard 4-Bay NIS and the San Onofre Unit 1 4-Bay NIS. Information relative to the drawer modifications (if any), the field cabling, external signal amplification and sensors (detectors) are provided in Table 1. In addition to this information, a comparison study is provided to detail the similarities and differences between the Georgia Power Vogtle 1.97 Neutron Flux Monitoring System (NFMS) and the San Onofre NIS Intermediate Range - Wide Range 1.97 System. This information is provided in Table 2. On Table 3 a comparison is provided for the microprocessors that are installed in the Vogtle 1.97 (NFMS) and the San Onofre NIS Intermediate Range 1.97 System. Details are provided on the testing performance and code used in conjunction with the microprocessor.
TABLE 1 4 Bay System (Standard) Class IE SCE 4 Bay System (with R.G. 1.97) Class IE Cabinet Standard 4 Bay Cabinet Standard 4 Bay Cabinet Source Range:
- BF3 detector (2 channels)
- BF3 detector (2 channels)
Preamplifier
- Preamplifier (unchanged)
- Source Range Drawer
- Source Range Drawer (Pulse Amplifier and Signal (Pulse Amplifier and Signal Conditioning Circuits, Bistable Conditioning Circuits, Bistable Amplifiers, Isolation Amplifier)
Amplifiers, Isolation Amplifiers, Start-up Rate Amplifier*)
Power Range:
- 2 Section Ion Chambers (4 channels)
- 2 Section Ion Chambers (2 channels)
- 2 Section Fission Chambers (2 channels)
- Power Range Drawers (A&B)
Power Range Drawers (A&B)
(Summing and Level Amp, Rate Amp, (Summing and Level Amp, Rate Amp, Bistables, Isolation Amp)
Bistables, Isolation Amp)
Intermediate Range:
- Compensated Ion Chamber (2 channels) -
Shares Power Range 2 Section Fission Chamber (2 channels)
- N/A
- Wide Range Preamplifier
- Triaxial In-Containment Cabling Special In-Containment Cabling*
- Intermediate Range Drawer
- Intermediate Range (R.G. 1.97) Drawer (Log Amplifier,*** Bistables, (Wide Range Amplifier,*** Bistables, Isolation Amplifier)
Isolation Amplifier, Start-Up Rate Amplifier*)
Tested in conjunction with intermediate range drawer at Cornell.
Nickel quartz hardline up to the bioshield, quadaxial cable from bioshield to penetration (used for noise rejection).
The major difference between the Westinghouse standard 4 bay NIS and the San Onofre 4 bay NIS is the amplification circuitry for the logarithmic function that is performed. The differences are noted as below.
Log Current Amplifier:
Wide Range Amplifier:
Totally analog logarithmic amplifier (uses 66% analog using transistors/op-amps to transistors and op-amps).
amplify signals that will be processed by Used to amplify and logarithmically process microprocessor (CPS/MSV, TSG).
input signals from analog bistables.
33% digital microprocessor board using Intel 8095 microprocessor. Performs logarithmic functions (in software) of input signals from analog circuits (,AP).
TABLE 2 Georgia Power R. G. 1.97 Neutron Flux Monitoring System (NFMS)
San Onofre R.G. 1.97 Fission Chamber (1 Section)
Fission Chamber (2 section)
In-containment cabling from detector to penetration --
In-containment cabling from detector to quadaxial cable (copper sheathed triaxial cable) bioshield -
Nickel Quartz Hardline; From Bioshield to penetration Quadaxial cable (same as Georgia power 1.97)
Preamplifier Preamplifier (same as Georgia Power R. G. 1.97)
Signal Processing Cabinet comprised of:
Hide Range Amplifier (in Intermediate 1 - Intel 8086 microprocessor Range Drawer) comprised of:
1 - Data translation A/D Intel 8095 microprocessor 1 - W CPS/MSV* board
- H CPS/MSV* board 1 -
TSG* board TSG* board 1 - Contact output board (Uses intermediate range drawer 1 - Hi-level analog isolator standard isolation amplifier and 1-Data translation D/A bistables for outputs.)
-CPS Counts Per Second MSV -
Mean Square Voltage TSG -
Test Signal Generator
TABLE 3 MISCROPROCESSOR SIMILARITIES/DIFFERENCES Georgia Power R. G. 1.97 NFMS San Onofre R.G. 1.97 8086.uP used on Intel 86/05 board 8095 chip contains A/D converters, serial port, and high speed counters (these functions are in addition to the functions in the 8086 microprocessor)
Code written in P/LM-86
- Code written in P/LM-96 Operating experience from Georgia Vogtle 1,
- Testing performed at Cornell test reactor CP&L Shearon Harris, and Vandellos in Spain.
to check operation of modified drawer No problems reported to Westinghouse with (with wide range) with real neutron fluxes the 8086 microprocessor or coding.
over entire operating ranges and with post-accident gamma fluxes when shutdown.
One way communication - Read only One way communication - Read only Diagnostic capability Diagnostic capability Setpoints cannot be adjusted with MMI interface.
_ Setpoints are adjusted via analog Coding cannot be changed.
adjustments (bistables). No adjustments can be made to the coding in the microprocessor.
Diagnostic capability available from drawer face.
- Diagnostic capability can only be accessed from cabinet drawer (wide range amplifier), no interface is available on the cabinet face.
-5 Isolation Devices All outputs from the San Onofre Unit 1 NIS upgrade cabinets that interface with non Class 1E systems are routed through isolation amplifiers. The isolation amplifiers utilized are manufactured by Westinghouse and are identical to those utilized in the standard 4-bay Westinghouse NIS system currently installed in many Westinghouse NSSS plants. The isolators utilize active amplifier circuits which divert input faults to dissipating loads, thus protecting the Class 1E NIS circuits.
The tests that were conducted on the isolation cards are described in the following letters to the NRC on the Diablo Canyon docket.
January 16, 1975 Letter from Philip A. Crane Jr. to NRC April 7, 1975 Letter from Philip A. Crane Jr. to NRC November 24, 1975 Letter from Philip A. Crane Jr. to NRC These reports were subsequently submitted to the NRC on the South Texas Project docket in letter ST-HL-AE-2035 from M. R. Wisenburg to the NRC on April 30, 1987.
The NRC accepted the test program in a letter from Olan D. Parr to Pacific Gas and Electric Company dated April 22, 1976. The acceptance letter stated that other applicants referencing the report will be required to provide justification "that the tests reported encompass the potential electrical faults or interference reflecting into the systems tested as a result of the particular plants design."
San Onofre Unit 1 has confirmed that the highest voltage associated with the NIS system does not exceed the maximum credible fault conditions of 118VAC and 25OVDC (nominal) that were coupled directly to the isolation amplifier output cables.
-6 Verification and Validation No specific Verification and Validation (V&V) Plan was written for the San Onofre Unit 1 Nuclear Instrumentation System (NIS) upgrade. The V&V plan submitted on the South Texas Project Qualified Display Processing System (QDPS) via letter ST-H1-AE-1859 from HL&P to Vincent S. Noonan, dated December 23, 1986 was adopted for use on the NIS upgrade with the following clarifications:
Verification: the same process was utilized on the NIS upgrade as described in the QDPS V&V Plan. The number of units tested was 31 which resulted in a total of 12 clarification reports and the identification of 8 failed software units.
All trouble and clarification reports were successfully closed out.
Validation:
the QDPS V&V Plan for the Validation process consisted of the following phases:
a) Functional Requirements Testing b) Abnormal Mode Testing c) System Prudency Review d) Man-Machine Interface Testing Sections of the system functional requirements were identified which were implemented on the microprocessor hardware of the NIS upgrade. Because of the simplicity of the system implemented on the microprocessor (31 units versus 1300 on the QDPS), no detailed functional requirement decomposition document was generated. The validator reviewed the system Factory Acceptance Test (FAT) and compared it to those portions of the functional requirements applicable to the microprocessor. Based upon the review, the validator identified one functional requirement which was not addressed in the FAT. The FAT was supplemented to include testing in this area. The Validator reviewed the FAT test results and did not identify any areas that did not meet the system functional requirements.
A total of approximately 20 test measurements were taken to test the microprocessor portion of the NIS Upgrade.
Because of the simplicity and function of the software, no system prudency review and man-machine interface testing were deemed necessary.
All records of the verification and validation test results are on file at Westinghouse.
The South Texas Project QDPS V&V plan was also utilized on the Plant Vogtle NFMS.
-7 Noise & Fault Tests MIL-N-19900B Noise Susceptibility Tests were used as the basis for identifying the field installation criteria of the NIS.
This information is provided via the tests that were conducted for Diablo Canyon.
Westinghouse Control and Electrical System (C&ES) Standard, Section 4.1 determines the minimum, separation of the conduit containing the NIS field cabling from potential electrical noise sources. The Standard states that "minimum separation of 2 feet shall be maintained from conduit to electrical noise sources such as power circuits of 118 volts 10 Amps and greater, fluorescent light fixtures, or circuits with inductive and/or switched loads such as Relays, triacs or SCRs.
Minimum separation of 6 feet from 4160 volt circuits shall be maintained." In addition to these guidelines, the Standard provides separation criteria for instrumentation circuit cables (4-20 mA DC) to be installed near the NIS conduits.
The criteria stated in this Standard is based upon Westinghouse testing (MIL-N-19900B tests) and experience with numerous installations.
This criteria has been applied to all Westinghouse supplied NIS NSSS's.
Operating plant data and customer feedback demonstrates that an installation of the NIS using the guidelines prescribed in C&ES Section 4.1 yields acceptable results.
San Onofre was issued a copy of C&ES Section 4.1, Rev. 7 for use in installing all cabling that would be used in the NIS. This revision includes separation distances from possible noise sources (both for conduit and penetration), maximum cable lengths to reduce noise susceptibility, conduit shielding and grounding requirements, conduit insulation requirements, and cabinet grounding requirements.
From the system comparisons it can be determined that the San Onofre 4 Bay NIS does not differ significantly from the Westinghouse Standard 4 Bay NIS (see Tables 1, 2, and 3).
Therefore it can be determined that the San Onofre 4 Bay NIS is equivalent to the Standard Westinghouse 4 Bay NIS that contains a field modification to provide for Post Accident Monitoring. Hence, it can be concluded that the San Onofre NIS will not be susceptible to noise and electromagnetic interference if installed in conjunction with the criteria stated in C&ES Section 4.1 to the maximum extent possible.
01100