ML20207Q184
| ML20207Q184 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 01/08/1987 |
| From: | Wilson R GENERAL PUBLIC UTILITIES CORP. |
| To: | Heishman R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE) |
| References | |
| 5000-87-1131, NUDOCS 8701230170 | |
| Download: ML20207Q184 (10) | |
Text
- _ _ _ _ _ _ _ _ _ _ _
GPU Nucloar NggIgf 100 Interpace Parkway Parsippany, New Jersey 07054 201 263-6500 TELEX 136-482 Writer's Direct Dial Number:
January 8, 1987 5000-87-1131 Mr. Robert F. Heishman, Chief
{
Vendor Program Branch Division of Quality Assurance Vendor and Technical Trainirg Center Program Office of Inspection and Enforcement Nuclear Regulatory Commission Washington, D. C. 20555
Dear Mr. Heishman:
Subject:
Oyster Creek Nuclear Generating Station Docket No. 50-219 Inspection Report 50-219/86-08 Environmental Qualification (EQ) of Electrical Eauipment Important to Safety The purpose of this letter !s to describe actions taken by GPUN during and following the EQ inspection conducted by the NRC staff between March 24 and 27, 1986. The actions were taken to address the concerns raised by the staff during the inspection.
Subsecuently, the same concerns (five unresolved items and three open items) were documented by your letter dated August 8, 1986. Detatis of the concerns were riiscussed in Inspection Report No. 50-219/86-08 which was transmitted by the same letter.
We believe that our actions which are described in the attachment address all of the unresolved and open issues..
Ve trily yours,
'W gn 1
F RFW:YN:nk 4 F.
i son Vice President Technical Functions cc's on next page h
[
ho 0
5 I
i 16759 GPU Nuclear is a part of the General Pubhc Utihties System
P cc: Dr. Thomas E. Murley, Administrator Region I U.S. Nuclear Regulatory Commission 631 Park Avenue King of Prussia, PA.
19406 NRC Resident Inspector Oyster Creek Nuclear Generating Station Forked River, N.J.
08731 Mr. Jack N. Donohew, Jr.
U.S. Nuclear Regulatory Commission 7920 Norfolk Avenue Phillips Building, Mail Stop 314 Bethesda, Maryland 20014 1675g
ATTACMENT GPUN Res>onses to Unresolved and Open Items Identified in NRC Inspection Report 53-219/86-08 Potential Enforcement / Unresolved Item #1 Contrary to paragraph (f) of 10 CFR 50.49, the EQ files for five Limitoroue motor operators, model SBD-0, located in the drywell, did not adeouately establish oualification because of failure to demonstrate siellarity between the tested and installed motor operator configuration.
The files did not establish whether the grease relief valves that were used during the type. test were actually installed in these operators, and failed to adeouately demonstrate that the installation of "T" drains was unnecessary (Paragraph 4.F(1), Item 50-219/86-08-01).
Response #1a During the NRC EQ Audit, GPUN presented walkdown information which showed that some operators had grease relief valves. This provided a high level of assurance that all EQ operators inside the drywell had grease relief valves based on the reasons presented below.
The reason for the above GPUN position was that all EQ operators inside the drywell were purchased to the same Limitoroue EQ report (#600376A),
which oualified the operators with grease relief valves for inside drywell applications. All of the EQ operators inside the drywell were previously replaced and photographs of some of these operators clearly showed grease relief valves.
During the OC 11R outage, which followed the NRC EQ Audit, a walkdown of all EQ motor operators inside the drywell confirmed the presence of grease relief _ valves on all of the operators in ouestion.
GPUN has revised the appropriate EQ file to document the existence of grease relief valves on the affected Limitoroue operators.
Response #1h The GPUN position at the time of the NRC EQ Audit was and still is that l
the motors are environmentally oualified, for inside drywell applications, with or without "T" drains. Contrary to the NRC concern, the GPUN i
position was not based " solely" on an unplanned occurrence during the test l
described in Limitoroue test report #600376A. The GPUN position is based l
on the following:
i I
i i
l l
(
16889
~
.. o At the time of the NRC inspection, the five Limitoraue motor operators in the Drywell were installed with class "RH" insulation.
Some of the operators did not have "T" drains. Limitoroue test report #600376A oualifies the motor operator with class "RH" insulation with "T" drains while Limitoroue test report #600198 qualifies the motor operator with class "H" insulation without "T" drains.
In each of the above test reports, the test environments enveloped the applicable Oyster Creek environments with margin.
Limitoroue confirmed by their mailgram dated March 26, 1986 that the class "H" and class "RH" motors are similar in insulation and in mechanical configuration. Therefore, the tests only differ in the "T" drain configuration, o Limitoroue test report #B0058, which is Limitoroue's generic avalification report provides additional information applicable to both test reports, #600376A and #600198. Test report #B0058 states, in paragraph 4.1.3, "The mounting position of the actuator was chosen with the limit switch compartment up and the motor horizontal. This is considered the worst possible position because it allows any condensate that collects in the unit to flow through the motor to provide the most damaging effect to insulation".
o As a result of the unplanned occurrence described in Limitoroue test report 600376A, there was evidence that the entire test operator became submerged. Evidence of moisture was found on the inside of the motor end caps, mixed with grease at the front end.
During and after the above condition, the motor operator continued to function properly.
I Based on the above, GPUN concludes that the motors are environmentally qualified with or without "T" drains for inside the drywell applications.
Pending resolution of the facts stated above, GPUN has installed "T" 2
drains on all of the operators which did not have "T" drains inside the drywell during the llR refueling outage. Therefore, all Limitoroue operators inside the drywell are similar to the tested configuration and the EQ file has been revised to indicate same.
Potential Enforcement / Unresolved Item #2 Contrary to paragraphs (f) and (k) of 10 CFR 50.49, and Sections 5.1 and 5.2.2 of the 00R Guidelines, the EQ files for 26 Limitoraue motor
)
I operators, models SMB-1, SMB-00, and SMB-000, did not adeauately establish j
oualification because of failure to demonstrate similarity between the tested and installed motor operator configuration. The similarity analysis was inadeauate in that it did not establish aualification for the internal wiring of valve operators.
(Paragraph 4.F(2), Item 50-219/86-08-02).
1 16889
- A Response #2 2(a)
IE Notice 86-03, dated 1/14/86 " Potential Deficiencies in environmental cualification of Limitoraue Motor Valve Operator Wiring" clearly informed the industry of a specific concern related to the lack of environmental cualification of internal wiring in Limitorque operators.
GPUN had Certificates of Compliance from Limitoraue for each specific EQ operator, which identified the specific applicable EQ test report which avalified each operator. None of the l
certificates specifically excluded,-explicitly or implicitly, l
any sub-component-(including internal wiring) from cualification.
Upon receipt of the IE Notice, the GPUN section initiated the following actions.
1.
Contracted with Limitoraue to identify the specific internal wiring used in each EQ operator. Limitoraue informed us by letters dated January 23,'1986 and February 18, 1986 that they had only supplied 3 types of wires: Rockbestos SIS, Raychem Flamtrol, and TW (PVC) wire.
l I
i
.2.
For the Rockbestos SIS wire, GPUN had an EQ file demonstrating.its cualification. For the Flamtrol wire, GPUN obtained a cualification test report, evaluated same, and concluded that the Flamtrol wire is cualified for the specific locations where the various EQ operators were located. The analysis / conclusions resulting from the review of the Flamtrol test report were documented and presented to the NRC audit team during the audit.
For the TW (PVC) wire, a detailed aualification analysis l
presented to the NRC audit team was based on an accepted l
industry. standard (UL83),NEMAstandards,anIEEEpaper (generic test information on cable insulations for use in nuclear power plants), and generic PVC manufacturer's data.
The results of the generic analysis of PVC insulated wires were also documented and presented to the NRC audit team during the audit.
1688g
. 2(b) GPUN agrees that the cualification files did not address possible modifications to the operator wiring by valve manufacturers or GPUN installers. At that time, this was considered not necessary because Limitoraues Certificates of Compliance identified the specific EQ test report to which each operator was cualified, and, as previously stated, did not exclude internal wiring from the cualification.
Since the similarity analysis was judgeJ to be inadeauate by the NRC staff in that it was based on a generic aualification of PVC insulation and did not establish aualification for specific wire, i.e., manufacturer and insulation type, for internal wiring of the valve operators, GPUN performed an evaluation and determined operability for the 26 valve operators within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> of notification by the NRC Audit Team. This information was provided to the Audit Team.
Two valve operators (V-24-40 & V-20-15) that would experience steam conditions due to a postulated DBA were inspected and internal wiring was replaced with aualified wire on March 28, 1986 (the weekend following the audit) during an unscheduled outage. All remaining operators have been inspected and any operator containing internal wiring which was not identified as cualified, i.e., manufacturer and insulation type, was replaced with aualified wire.
Potential Enforcement /Unrosolved Item #3 Contrary to paragraphs (f) and (k) of 10 CFR 50.49, and Sections 5.1, 5.2.2, and 5.2.6 of the 00R Guidelines, the EQ files for Namco limit switches, model D1200G, did not adeauately establish aualification because of failure to base the cualification for severe temperature, pressure, relative humidity, and steam accident conditions on a type test.
(Paragraph 4.F(3), Item 50-219/86-08-03).
Response #3 The limit switch in auestion is installed on the nitrogen make-up isolation valve V-23-18, outside the drywell, in the reactor building above the 75'3" elevation. The valve is normally closed and is reauired to remain closed during a DBA.
Thus the 11. alt switch is not reauired to change state during a DBA.
In addition the valve is designed to fail in the closed position.
The above limit switch had been cualified to the reauirements of the 00R Guidelines which preclude the need for a steam test in non-steam environments, thus allowing for cualification by analysis.
16889
I l
~5-Section 5.1 of the D0R Guidelines states that type testing is the preferred aualification method for eauipnent located inside containment.
It further states that for steam service conditions, cualification should he based on type testing. The GPUN position, as presented during the NRC 1
EQ Audit, is that the internals of the above limit switch are not subject to a DBA saturated steam environment and therefore a cualification by analysis is considered appropriate for the following reasons:
1.
The short lived peak pressure the switches may be subjected to is 17.8 psia, which corresponds to a a saturated steam temperature of approximately 225'F.
2.
The peak temperature the switches will be subjected to is less than the above saturated steam temperature, not considering the inherent thermal lag of installed configuration.
3.
The switches are housed in NEMA 4 enclosures (water tight / dust tight-indoor / outdoor) and are directly connected to a terminal j
box (with weep-holes) via solid conduit connections. This installation configuration provides direct protection from the environment, thus minimizing moisture intrusion inside the limit switch, especially since the pressure transient is short lived.
i 4.
As previously indicated the switches are not reauired to change 1
I state as a result of a DBA.
GPUN implemented the following actions to address the NRC's concern, l
pending consideration of the above facts.
1.
A oualification test report (#CCL A-716-86 Test Report for l
Instruments subjected to HELB dated June 30,1986) was obtained, reviewed and used to' support the above cualification of GPUN limit switch position. This report subjected a similar type of un-sealed limit switch to an EQ test program including a steam test, with acceptable results.
l-The test report included secuential testing, i.e., thermal aging (6.1x10}enttomorethan40 years 985'F), irradiation (eauiva rads) and HELB tests (250*F-peak, 5 psig, 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> duration) of unsealed Namco limit switches. All of these parameters are well above the Oyster Creek normal and accident f
conditions of the limit switches in ouestion.
1 2.
Conduit seals were installed during the current 11R outage to preclude moisture entry into the limit switch.
)
As of November 30, 1985, the limit switch aualification (by analysis) was fully documented. As a result of the audit we have proceeded to provide additional nroof of aualification (the test report referenced in Item 1 above) and provision of conduit seals for a measure of defense in depth.
J 1688g
._. --. _ _._ - - __, - ~.. _.
.~
7 Potential Enforcement / Unresolved Item #4 Contrary to paragraph (f) of 10 CFR 50.49, the EQ files for Namco limit switches, model EA740, did not adeauately establish oualification because of failure to demonstrate similarity between (1) the tested and installed material of construction; and (2) the tested and installed eauipment configuration with respect to moisture seals. No supporting analysis was contained in the EQ file to justify revision level changes to the manufacturer's bill of material. GPU's similarity analysis was inadeauate to demonstrate eauipment operability if the limit switch internals were exposed to the harsh environment (Paragraph 4.F(4), Item 50-219-86-08-04).
4 Response #4 4a)
The Namco EQ File in auestion, did address the similarity between the tested model to the installed model as a checklist i
item within the EQ File, and while the basis for this determination was judged to be inadeauate by the NRC audit team, the purchase orders for the switches associate the tested model to the installed model via certification to the vendor's test report. This provides justification of the vendors confidence l
in the similarity between the installed and the tested model.
This was further confirmed during the NRC inspection with a discussion with the vendor in which it was indicated that the only changes between the test model "N and the installed model "P" involved minor mechanical part changes which do not affect oualification.
I The file has been revised to incorporate a letter from the vendor which documents the nature of the changes between the i
tested and the installed models and a similarity summary has I
been included in the file.
l l
4b)
There are a total of 14 limit switches in ouestion; 2 switches associated with valve V-6-395, and 12 switches associated with
(
MSIVs (3 switches per each of 4 MSIV's).
l The 2 switches for V-6-395 are only exposed to a maximum temperature of 120*F which results from a loss of HVAC. The SCEW sheet specifies 100% relative humidity as an assumed ultra conservative reauirement in the absence of any other accident data regarding humidity. However, since the HELB analysis postulates no significant temperature or pressure effects for the zone in wnich the switches are located, it follows that l
relative humidity will be minimal, certainly less than 100%.
j Furthermore, the switch is not subject to water spray as a result of the DBA since they are outside containment. The j
details of this evaluation have been incorporated into the EQ file which provides reasonable assurance by analysis that precludes the need for protecting these switches from humidity.
The limit switches associated with the MSIV's must be aualified to provide indication af containment isolation resulting from a g,}heswitchesareassociatedwithinboardandoutboard 16889
b.
The SCEW's reflect accident conditions conservatively based on a HELB occurring in the worst case location relative to the switchel in cuestion, i.e., for outboard switches, the accident is postulated outside the drywell. Realistically one pair of indicating switches will always be free of high moisture environment. Furthermore, existing Emergency Operating Procedures reouire the operator to confirm containment isolation by manual action to close the valves. Since these switches do
)
not contribute to the isolation function and only provide indication of position, the deficiency cited will not impact the safe operability of the plant.
Moisture seals however, have been provided for all the MSIV limit switches during the 11R outage and this configuration has been incorporated into the EQ File. This action was promptly initiated and scheduled as a result of the EQ Audit.
Potential Enforcement / Unresolved-Item #5 i
Contrary to paragraphs (f) and (k) of.10 CFR 50.49, and Section 5.2.5 of the DOR Guidelines, the EQ files for Rockbestos Firewall III and SIS caole, PVC splice tape, and GE type EB terminal blocks, did not adeouately demonstrate cualification because of failure to show that the eauipment i
functional performance reouirements were satisfied. Files for the Rockbestos cable and PVC splice tape did not contain any analysis of the functional performance recuirements and the functional performance 4
criteria and reauirements for GE EB terminal blocks were incorrectly addressed (Paragraph 4.F(5), Item 50-219/86-08-05).
i Response #5 GPUN has undertaken a study to address functional recuirements of cables and termination facilities. This study considers appropriate EQ circuit configurations whose performance may be impacted by a harsh environment.
Preliminary results indicate that the Rockbestos Firewall III and SIS cable meet performance reauirements for their intended service.
In addition, the file has been revised to address the degradation of the cable jacket.
With regard to PVC tape splices, based on a previous commitment relative to LER 50-219/82-25 (submitted on June 17,1982), GPUN had replaced six splices during the previous outage (10R). Under the commitment described L
by GPUN letter dated October 5, 1984, to Dr. T. E. Murley of NRC Region 1 office all remaining tape splices on control and instrument wiring have i-been removed during the current outage (llR). Programmatic controls have also been implemented to prohibit the future use of tape splices on instrument circuits. The EQ file has been revised accordingly.
The G.E. Terminal Block File has been revised to include the corrected calculation referenced in the audit inspection report.
j r
l 16889
. Open Items Open Items 86-08-06 and 07 have been addressed and the related documentation deficiencies as noted in paragraph 4.F(8) and 4.F(6) have all been resolved and the files and documentation have been revised accordingly.
Open Item 86-08-08, EQ Program Procedures - Maintenance Paragraph 4.B(1).
The inspectors concluded that Oyster Creek lacked a program for maintenance and surveillance of terminal blocks and terminal boxes.
During the refueling outage, GPUN performed an inspection of EQ Common Items. An inspection of EQ terminal blocks revealed no signs of deterioration or degradation due to aging or environmental conditions.
Two of the approximately 75 terminal blocks inspected showed signs of surfece corrosion due to what appears to be localized moisture impingement during normal operation. These results clearly support and reinforce GPUN's position, at the time of the audit, that regular scheduled maintenance for terminal blocks and terminal boxes are not necessary. The l
existing program which invokes inspection recommendations and guidance through plant maintenance procedures is sufficient but should not be interpreted as a mandatory, scheduled maintenance as the I&E reviewer suggested.
The paragraph 7.2.2 in Oyster Creek Procedure 105.3, (inadvertently referred to as paragraph 7.2.7 in the Inspection Report) was established as guidance to be implemented on a target of opportunity basis. This is indicated in paragraph 7.1 of the same procedure, which reads: "The following generic guidelines only apply when they are within the scope of the authorizing document.
l l
l 16889 3