IR 05000293/1989012
| ML20006A959 | |
| Person / Time | |
|---|---|
| Site: | Pilgrim |
| Issue date: | 01/18/1990 |
| From: | Blough A NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20006A957 | List: |
| References | |
| 50-293-89-12, IEB-81-03, IEB-81-3, NUDOCS 9001310020 | |
| Download: ML20006A959 (34) | |
Text
....
.
,...,,,.
.............
..............
,,,.....
................
'
,
..
o
,
.
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
i Docket No.:
50-293
>
Report No.:
50-293/89-12 Licensee:
Boston Edison Company 800 Boylston Street Boston, Massachusetts 02199 Facility:
Pilgrim Nuclear Power Station
'
Location:
Plymouth, Massachusetts Dates:
October 2 - November 19, 1989
..
'
Inspectors:
C. Marschall, Senior Resident Inspector T. Kim, Resident Inspector C. Carpenter, Resident Inspector
}
>
T. Cerne, SRI - Seabrook J. Macdonald, RI - Vermont Yankee-M. Markley, RI - Yankee Rowe
'
L. Kolonauski, Project Engineer, RI D, Caphton, Senior Technical Reviewer, RI T. Rebelowski, Senior Reactor Engineer, RI
R. Paolino, Lead Reactor Engineer, RI
'
T. Koshy, Senior Reactor Engineer, RI l
T. Dragoun, Senior Radiation Specialist, RI i
D. Solorio, Reactor Engineer, NRR G. Wunder, Reactor Engineer, NRR P. Wen, Reactor Engineer, NRR J. Miller, Project Engineer, NRR J. Carter, Project Engineer, NRR T. Greene, Project Engineer, NRR G. Bethke, NRC Contractor G. Bryan, NRC Contractor M. Good, NRC Contractor
,
Approved by:
/~N'7D
-
A. Randy Bloup $' Chief, Reactor Projects Section 3A Date Inspection Summary:
Areas Inspected:
Restart staff inspection to assess licensee management con-trols, conduct of operations and the licensee's approach to investigation and resolution of events during the 100% power plateau of the licensee's Power Ascension Test Program and during the maintenance mini-outage.
9001310020 900119 ADOCK0500g3 FDR
/
_.. _.,...... _....
..
..
,
j e
.
-
,
.
Results:
Licensee performance during Phase II of the Shutdown Outside the Con-
-
trol Room Demonstration was exemplary.. Performance of the Instrument Line Flow Check Valve test was well. planned and. executed (section 5.2).
Licensee approach to ALARA during outages has greatly improved (section 6.3).
Overall, the outage was conducted in a very professional, well organized and conserva-
,
tive manner.
Particularly noteworthy were the conservative management deci-sions made at several junctures during the outage (section 8.0).
Operations cognizance of maintenance activities during the outage was superior to that noted during the previous months of the Power Ascension Test Program (section 8.0).
,
Shif.t and watchstander turnover, log keeping and operator cognizance of main-tenance during normal plant operations, as well as preservation of a question-ing attitude are areas which are still occasionally deficient and require addi-tional management emphasis (section 2.1).
The. lack of control of hazardous material in the cable spreading room and the lack of readily observable identi-
,
fication labels on several specific cable spreading room cable trays indicates a weaknets in the licensee's material condition inspection program (section 2.2).
During the outage, even though general plant housekeeping was excellent, housekeeping at several work locations was poor (section 8.0).
Failure to comply with procedural requirements for~ documentation associated with hanging and removal of tags resulted in administrative errors or omissions on over 80% of the completed tagouts reviewed (Section 2.4, VIO 89-12-01)..
-
<
l l
I l
,
..
_
_
._
,
..
.
..
!
i
.
,
,
!
TABLE OF CONTENTS i
!
PAGE
1.
S umma ry o f Fa c i l i ty Ac ti v i ti e s.......................................
2.
Operations (Modules 71707, 71710, 71715, 62703, 61726, 92701, 92702, j
93702,92700)......................................................
!
3.
Shutdown Outside Control Room (Phase !!) (Modules 71707, 61726,
)
40500)............................................................
4.
Maintenance and Modifications (Modules 62703,37828,62704)..........
J 5.
Surve111ances(Modules 61726,61700,61701)..........................
i 6.
Radiological Controls (Modules 71707, 93702, 83750, 83722, 83723,
<
83724,83725,83727)................................................
s 7.
Fi re Protecti on (Modul e 64704).......................................
!
8.
Outage Assessment (Module 40500)....................................,
]
9.
Followup on Previous Open and Unresolved Items (Modules 90712, 92700, 92701).............................................................
a 10. Management Meeti ng s (Module 30703)...................................
ATTACHMENT Attachment I - Persons Contacted i
,
.
.
.
l
.
,
,
DETAILS 1.0 Summary of Facility Activities At the beginning of this report period, the plant was operating at approxi-mately 75% power while the licensee awaited NRC approval to proceed to
';
full power.
On October 6,1989, Mr. William T. Russell, Region I Regional Administra-tor, approved the NRC Restart Assessment Panel s recommendation to release the licensee from the fifth NRC approval point (75% of rated power) in the licensee's Power Ascension Test Program.
The plant reached 100% power at 11:15 p.m. on October 10, 1989.
On October 7,1989 with reactor power at about 89% and subsequently on October 12, 1989 with reactor power at 100%, the "A" Recirculation motor-
,
generator (MG) set tripped when both the motor drive breaker and the gene-rator field breaker opened.
In both instances, reactor power stabilized at about 60% following the transient.
Details are provided in section 2.3.1.
On October 12 and 13, 1989 the licensee conducted an annual full partici-pation emergency preparedness exercise.
Details of the inspection are contained in NRC Inspection Report 50-293/89-11, issued on November 1, 1989.
The licensee successfully completed the second phase (hot standby to cold shutdown) of the Shutdown from Outside the Control Room (SDOCR) demonstra-tion on October 13, 1989. Details are provided in section 3.0.
,
l Following completion of the SDOCR demonstration, the plant was maintained i
in cold shutdown to perform scheduled maintenance and surveillance activi-ties until November 6, 1989. At 12:08 p.m. on November 6, 1989, the lic-ensee brought the reactor critical. The turbine generator was synchron-ized to the electrical power distribution grid at 5:43 p.m. on November 8, 1989.
,
At the close of this inspection period, the plant was at 94% power and limited to that power due to a lockup of the "A" recirculation motor-generator set scoop tube and current rod pattern.
NRC inspection activities during this report period were conducted by the onsite Pilgrim Restart Staff led by Mr. Charles S. Marschall, Senior Resident Inspector and Restart Manager.
The Pilgrim Restart Staff is com-
,
posed of the Pilgrim resident inspectors, NRC regional-based and head-quarters-based inspectors and an NRC contractor.
On October 6, 1989 Mr.
William T. Russell, NRC Region I Regional Administrator and Mr. Jon Johnson,
<
,
.
.
.
.
Chief, Projects Branch No. 3 were on site and toured the plant. On Octo-ber 11-12, 1989, Dr. Ronald R. Bellamy, Chief, Facilities Radiological
Safety and Safeguards Branch was onsite to observe portions of the emer-
'
gency preparedness exercise, i
At the beginning of this report period with the plant at 75% power, the Restart Staff was in extended shift coverage. On October 6, 1989 at 2:00 p.m. the staff began around-the-clock shift coverage. Around-the-clock shift coverage was suspended at 2:00 p.m. on October 11, 1989, consistent
,
with reduced testing activity. On November 5, 1989 at 10:00 p.m. the staff resumed around-the-clock shift coverage. Around-the-clock shift l
coverage was suspended on November 10, 1989 at 2:00 p.m.
The Restart Staff maintained extended shift coverage throughout the remainder of this inspection period. The inspection involved about 1090 hours0.0126 days <br />0.303 hours <br />0.0018 weeks <br />4.14745e-4 months <br /> of direct inspection effort.
2.0 Operations 2.1 Sustained Control Room Observations Control room activities were routinely observed during this inspec-tion period. Control room voice communications were comprehensive
-
and formal.
Operators used procedures properly. Coordination and mutual support were evident between operations and other licensee organizations.
Shift turnover briefings were generally suitably de-
'
tailed and conducted effectively with the exception that briefings by support groups (maintenance, chemistry, radwaste, health physics)
were often degraded by incomplete briefer preparation, speaker loca-tion and projection, and ambient noise.
Log keeping was occasionally deficient.
During the startup after the maintenance outage, on November 6 and 8, excess feed water flow re-suited in vessel level excursions from 28 to 46 inche.c (section 2.3.2).
Both excursions were terminated by timely response by the operators, but no Nuclear Operations Supervisor (NOS) log entry was made for the November 8 transient. After the inspector pointed this out, the Chief Operating Engineer made a Night Order Book entry re-minding control room operators to make log entries of unusual plant systems performance.
Another lapse in logkeeping and turnover occurred when control room speed indication of the "A" recirculation pump motor generator (MG)
became inoperable on November 14, 1989.
No NOS log entry was made for that failure. Technical Specification 3.6.F.1 requires that pump speeds be maintained withir spe ified limits of each other, as a function of power level; in addition, a daily surveillance is re -
quired. On November 15, 1989, the inspector questioned control room personnel on how they were meeting that surveillance requirement with the "A" speed indicator out-of-service.
None of the operators ques-tioned had been alerted to the problem. Although the day shift had
.
.
.
,
O
.
I until about 3:00 p.m. to complete the surveillance, the inspector
-
expected that the two previous shifts to have encountered the problem and alerted the day shift at turnover.
This expectation was based upon the fact that although the recirculation pump speeds are only logged daily, the Limiting Condition for Operation (LCO) requires
commencement of a shutdown within 30 minutes if the speeds cannot be brought within specification.
For that reason, operators check for mismatch several times per day, and always during speed adjustments.
Within an hour after the inspector's question, management had speci-fied an alternative means of obtaining "A" recirculation pump speed.
,,
This problem also illustrates a lapse in a questioning attitude on the part of the operators.
During the reactor shutdown, the turbine main field and exciter cir-cuit breakers appeared not to have tripped when the main turbine was tripped. When the SRO attempted to open the breakers from the con-trol room, both breakers appeared to remain shut.
Later, it was de-termined the that only the indication was faulty and the breakers did open as required. The next day however, the reactor operator on shift was unaware of the problem with the breaker (even though neither the red nor green indicating lights for the main field breaker were illuminated), was unaware of the status of the breaker,
-
and was unaware what was being done to fix the indication problem.
Shift and watchstander turnover, log keeping, and operator cognizance of maintenance during plant operations and a questioning attitude are
.
areas which are still occasionally deficient and require additional managenont emphasis.
The level of knowledge demonstrated by control room personnel was generally very good.
Senior personnel appeared to have a good under-standing of the limits placed on them by the Technical Specifications and by other regulatory requirements.
The Nuclear Watch Engineer
,
(NWE) readily identified the loss of a main steam line flow instru-t
'
ment, which gave one input to the Group I Isolation coincidence cir-i cuitry, as an entry into an LCO.
He then directed prompt corrective i
action and was able to resolve the problem.
'
Documents and procedures in the control room and control room annex l
l were maintained in an organized, usable way. The use of the LCO log to track plant conditions during shutdown and thus ensure the clear-ing of all LCO's before ttartup was considered an excellent practice.
The general attitude of plant employees was good. Operators con-ducted themselves in a highly professional manner.
Management atten-tion to operations was consistent and visible.
P I
.
.
.
.
.
2.2 Plant Tour nhservations t
Overall u e
59 in the plant continued to be a licensee strength, incu.
.o of management oversight of maintenance and
.
operations.
Some exceptions to normally excellent housekeeping occurred at job worksites, as discussed in section 8.0.
Detailed tours and inspections were made of the cable spreading room, switchgear room B, drywell and the reactor building with emphasis on electrical installations and applicable sections of the fire protec-tion program.
The inspector identified several material deficiencies (e~.g., cables retired in place but not identified as such and led so l
as to abraid the wires, a missing conduit cover, one set of infre-quently used test leads disconnected but left in place, missing bolts on several panel covers; several missing relay box covers in panels
,
C-1 & C-3; and a broken cable tray cover support). The inspector
'
also identified an unrestrained argon bottle on an unattended cart in the cable spreading room and several cable spreading room cable tray identification labels that had been obscured by flammastic spray application.
,
Some, but not all, of these findings were known to management and al-
-
ready scheduled for correction.
The inspector concluded that the
'
licensee has not been meticulous in identifying these types of de-
'
ficiencies in the cable spreading room.
.
The inspector's findings were discussed in detail with licensee man-agement. The unrestrained gas bottle and several material deficien-cies were rectified immediately.
The rer.aining items were addressed to the inspector's satisfaction, principally by initiating mainten-ance work requests.
The inspectors will routinely review these areas during plant tours.
2.3 Review of Plant Events 2.3.1
"A" Recirculation Motor Generator Set Trips On October 7, 1989 with reactor power at about 89% and sub-sequently on October 12, 1989 with reactor power at 100%,
the "A" recirculation motor generator (MG) set tripped when both the motor breaker and the generator field breaker opened.
In both instances, reactor power stabilized at about 60% following the transient.
On October 7, the M-G set tripped at 78% speed shortly after a small speed increase.
The " generator lockout" and
" drive motor breaker" alarms were received; however, no
!
targets were received. Af ter investigating potential root I
causes, the licensee adjusted the volts / hertz ratio and instrumented eight MG set relays, tachometer voltage, and
__
.
-
.
.
.
.
,
i
!
'
!
!
generator speed on a chart recorder to monitor trips which
could cause a lockout without setting targets on the MG
control panel.
i On October 12, during conduct of Recirculation Scoop Tube Position checks per Temporary Procedure (TP)87-265, the
,
"A" recirculation MG set tripped again. Just prior to the
!
.
trip, the reactor operator (RO) noted generator voltage i
divergent oscillations which reached 5KV at the time of the
'
trip. Again the " generator lockout" and " drive motor breaker" alarms were received. The chart recorder indi-cated that the generator loss of field relay (136A) caused
the generator lockout and subsequent drive motor trip.
Based on the RO's observations coupled with the generator loss of field relay pickup, the licensee concluded that the MG trip was caused by voltage regulator instability.
The licensee reviewed past maintenance records, conducted
electrical checks on all voltage regulator and exciter diodes, voltage regulator saturable core reactors, and machine windings; and made t.cmparisons of components be-
'
tween the "A" and "B" recirculation MG sets.
No discre-
-
pancies were noted.. Testing of voltage regulator capaci-l tors revealed two stability circuit capacitors which had a higher leakage current; they were replaced. A dynamic
.
check of the voltage regulator was performed. The silicon
-
controlled rectifiers (SCR) were checked, the voltage regu-lator was tuned, and the gain and stability potentiometers were adjusted.
On November 11, 1989 an automatic lockout of the "A" Recir-culation Pump MG set scoop tube positioner occurred during
steady state operation at approximately 95% of rated power.
Operators observed a spike on "A" recirculation pump con-
troller instrumentation.
However, troubleshooting did not identify the cause of the automatic lockout.
.
Although the licensee has extensively investigated the "A" Recirculation MG set trips, as of the end of this report period, troubleshooting was still in progress and the scoop tube remained locked, limiting the plant to 94% power after i
Xenon buildup. Management has established a Kepner-Tregoe (K-T) problem analysis team to evaluate work to date and to plan future efforts. Discussions with licensee personnel
.
on the events, troubleshooting activities and corrective
actions demonstrated their detailed knowledge of the sys-tem and adequate interim actions.
,
.
.
.
.
.
.
.
.
,
2.3.2 Group I primary containment Valve Isolation On November 6, 1989 at 3:39 p.m. with the mode switch in
Startup, reactor power at about 1% and primary pressure at about 9 psig, a group I primary containment valve isolation s
(main steam isolation valves, main steam drain valves and i
recirculation system sample valves) was actuated by high
Control room operators were initiating feed flow to the reactor vessel from a conden-sate pump via the startup feedwater regulating valve (FV-643) to supplement makeup from control rod drive water.
With FV-643 shut, the "B" train high pressure feedwater heater upstream block valve (MD-3477) was opened to provide condensate pressure up to FV-643. When MO-3477 was opened, reactor vessel level rose rapidly and the group 1 isolation actuated.
Since M0-3477 is a seal-in motor operated valve with a 60 second stroke time, operators could not close it soon enough to avoid the isolation signal.
Operators re-stored level by rejecting water via the reactor water cleanup (RWCU) system, power was held constant and the source of th problem was investigated. An ENS report was
-
made to the NRC operations center at 6:25 p.m. on November.
6, 1989.
Licensee investigation determined that manual bypass valve 6-HO-443 around FV-643 had been left open from the outage creating a flow path to the reactor vessel when the heater block valve was opened.
The licensee conducted a detailed investigation concerning the out-of position valve.
They determined that the. valve mis positioning had occurred earlier when operations person-nel had manipulated valves, first to establish conditions for maintenance and later to adjust reactor water during the outage.
The licensee identified the root causes of the group 1 isolation as: (1) changing tagout boundaries without fully
>
documenting the change, (2) a miscommunicated valve operat-ing instruction by the NOS (i.e., by location, rather than valve name or number), and (3) inadequate labelling of the fe?dwater system reach rods.
Immediate corrective actions included the completion of valve lineups on other systems which had significant work during the outage prior to pro-
.
ceeding with reactor startup. Additional corrective ac-tions are being planned and will be addressed in the forth-coming LER, such as improved labeling of reach rods and revised guidance on post-outage valve lineups prior to startup. No valve lineups were performed as part of the
.
startup in which this event occurred.
J i
i I
~
--
-
.
.
..
,
- O
.
.
'
.
After reviewing the licensee's critique and walking down
,
the feedwater system, the inspector raised two additional issues. The licensee had not determined where, by whom, i
and how the "B" train block valve (6-HO-444B) was opened between October 24 and November 6 nor noted that the t
travelling nut position indicator on the ROTO-HAMMER 444B valve reach rod was missing, making it impossible to visu-ally verify valve position from the operating station.
,
The inspector will review final corrective actions taken
!
for this event.
'
,
2.3.3 Partial Secondary Containment Isolation
,
On November 11, 1989, during a semi-annual Logic System l
Functional Test (LSFT), several reactor building ventila-tion dampers closed unexpectedly when test personnel were installing a test jumper.
Licensee investigation deter-mined that panel design was inadequate to support testing.
Since terminals on the exterior of the relay case were blocked by a structural support, it.as necessary to in-stall the jumper inside the relay.
During installation, a set of normally closed contacts were opened momentarily.
The resulting partial isolation of secondary containment was immediately recognized by control room personnel who halted the surveillance, ensured an actual condition re-quiring secondary containment isolation did not exist, and reset the isolation.
Operator action was prompt, conser-vative, and indicative of technical competence.
The licensee's root cause review was thorough and conducted
,
according to procedure. At the end of the inspection period, the licensee had not identified a more acceptable
'
method of jumpering the relay contacts for the obstructed relay.
Corrective action to prevent recurrence will be followed as a matter of normal inspection activities.
2.4 Review of Tagout Records The inspector reviewed procedures 1.4.5 "PNPS Tagging Procedure" and
!
1.5.9.1 "Lif ted Leads and Jumpers" (LL&J) (as applied to the tagout process), audited selected tagout documents generated during the October maintenance outage, and interviewed several personnel who are responsible for performing tagout administrative duties.
,
The audit consisted of about 50 completed tagouts and 5 active tag-l outs.
The completed tagouts had undergone final review in the con-trol room and were ready for shipment to the Pilgrim document control l
area.
Review of the tagouts revealed that over 80% of the tagouts i
~
contained some error or omission.
Procedure 1.4.5, Step 6.5[4](d)
.
.
.
.
,
.
'
.
f requires that "the NWE on duty at the time the tagout is to be con-
-
ducted shall give final approval prior to any tag placement... when all of the above criteria are met, the NWE shall sign on the tagout sheet." Step 6.5[6)(a)statesthat"theNOS/NWEshallsignto authorize tagout removal after all job supervisors have signed re-leasing the tagout." Review by the inspector determined that over 30*4 of the tagouts reviewed contained one or more significant errors.
- including tags hung or cleared without the NWE authorization signa-ture or tags hung or cleared by indivicuals other than those author-i ized on the tagout.
The inspector also reviewed ten tagouts which required entries in the LL&J log.
Procedure 1.4.5, Step 6.9[4]
states that "when 1 cads are lifted for isolation and included on the tag sheet, they shall also be logged in the lifted lead and jumper log per PNPS 1.5.9.1."
This inspection determined that over 50*4 of the required LL&J entries were not made; four involved replacement of
.
safety-related components.
'
The inspector provided information regarding the specific tagging deficiencies to the licensee.
The licensee confirmed the NRC find-ings in PNPS operations department memorandum on 11/13/89.
The memo aise identified possible causes and detailed a corrective action and
.
-
prevention program.
Included in this program were promulgation of a
-
new procedure requiring periodic tag audits and a Quality Assurance tegging audit to be completed early in 1990.
Failure to comply with existing procedures is a violation (VIO 50-293/89-12-01).
.
3.0 Shutdown Outside the Control Room (SDOCR)
On October 13, 1989, the augmented NRC Restart Staff observed the second of two phases demonstrating SDOCR. This demonstration was performed to fulfill a commitment in the Power Ascension Test Program (PATP) and was l
in accordance with USNRC Regulatory Guide 1.68.2 " Initial Startup Program l
to Demonstrate Remote Shutdown Capability for Water Cooled Nuclear Power
'
Plants."
The initial phase was completed on June 29,1989(Inspection Report 50-293/89-07 Section 3.2).
This phase began with the reactor shutdown and plant pressure established at about 140 psig from the main control room.
The test crew then took local control and lowered pressure to about 300 psig using the High Pres-sure Coolant Injection (HPCI) turbine.
One Safety Relief Valve (SRV) was e
opened, the Shutdown Cooling mode of Residual Heat Removal (RHR) was established, and moderator temperature was lowered by about 50 degrees F.
The pre-test briefings were adequate but the use of the control room for this briefing appeared inappropriate.
The large number of personnel pre-sent in the control room for the briefing represented an unnecessary bur-den on the normal shift operations crew.
.
.
_ -.
&
-
.
.
.
r
t
.
With few isolated exceptions, the test procedure, communications, and com-
mand and control were excellent.
Valve and breaker manipulations to
'
establish shutdown cooling were well planned and executed.
Test personnel were aware of their responsibilities and were skilled in their assign-ments. The test crew Nuclear Operations Supervisor (NOS) maintained con-trol and ensured that crew members were cognizant of the changing plant configuration.
'
On one occasion, the Shift Technical Advisor (STA)' incorrectly reported an excessive cooldown rate.
The Nuclear Watch Engineer (NWE), who had been prudently observing RHR heat exchanger inlet and outlet temperatures, immediately recognized the STA had shifted his monitoring from inlet to outlet temperature resulting in an erroneous conclusion about moderator
,
cooldown rate.
.
The procedure did not anticipate the control room key switch for the "A" Residual Heat Removal system (RHR) loop torus suction valve being left in the open position. When the test crew shut the valve locally, it immedi-ately re-opened.
The NWE, NOS, and maintenance crew reviewed logic draw-ings, properly concluded that the control room key lock switch was over-riding their actions, pulled the control fuse, and then closed the valve locally. This evolution demonstrated excellent plant knowledge, timely
technical evaluation, and team-work.
Since this test was conducted using a test procedure (versus the permanent plant procedure), and with a specially briefed crew, the test did not validate the permanent procedure, Abnormal Operating Procedure 2.4.143, nor did it show that a typical crew, as augmented in the latter stages of the evolution, could perform the shutdown.
To accomplish this, the licen-see will review the proven test procedure against 2.4.143 and upgrade the latter as applicable.
Crew training will also be reviewed to ensure that applicable portions of that training are included in the cyclic mainten-ance of operator training program.
From preparation through execution, this test was performed in a safe and satisfactory manner.
Phase I and II of the shutdown from outside control l
room demonstration were two of the more exemplary performances by the lic-
!
ensee's operations department during the PATP.
l 4.0 Maintenance and Modifications 4.1 Administrative Review of Maintenance Requests i
!
The inspector reviewed a sampling of maintenance requests to verify compliance with procedure 1.5.7, " Unplanned Maintenance," and 1.5.3,
" Maintenance Requests," including work plan, pre-job briefing, safety
'
l controls, radiation protection, authorized signatures and review.
No i
discrepancies were identified.
l
l
.
.-
l
'.
\\
..
.
)
.
i i
4.2 Core Spray Orifice Changeout In order to relieve recurring problems with meeting core spray loop TS surveillance test acceptance criteria, the licensee instituted a design change that increases the orifice size in the flow restrictor
.
located in the line to the reactor vessel from the "B" core spray pump. Since there currently is no similar surveillance problem with the "A" pump, the change was limited to only the "B" core spray loop.
.
The licensee stated that increasing the orifice size will increase j
the core spray pump margin to test criteria by about 5 psi. A General Electric Company review concluded that the additional flow would not cause a problem with Net Positive Suction Head (NPSH) or
pump power requirements during pump runout.
,
The inspector reviewed the design change, reviewed past surveillance test results, and discussed extensively with licensee personnel their rationale for the design change.
Discussion with the licensee's technical staff indicated that (1) analysis showed acceptable system flow with sufficient margin, but (2) the original selection of the acceptance discharge pressure was too close to actual operation. The pump characteristic cc.'ve shows that variations in measured pump head could come from pressure measurement uncertainties and variations in indicated flow.
There appears to have been no actual degradation of the pump, and the test failures are associated with problems with test acceptance criteria and measurement accuracy.
The inspector concluded that a change to test acceptance criteria, under development by the licensee, could resolve the problem. The inspector concluded, based on the above, that the design change to increase the orifice diameter did not appear to have been necessary and resulted at least in part from an insufficiently critical engi-neering review of the design change package.
However, no unsafe con-dition or safety degradation was introduced into the system by the design change.
,
4.3 ADS Check Valve Leakage
!
A routine check of the nitrogen system for the Automatic Depressuri-zation System (ADS) accumulators showed that pressure on accumulators C & D bled down rapidly.
The licensee suspected the nitrogen system supply line check valve was the source of the leakage and initiated a
'
Maintenance Request.
After replacing the check valve soft seats, the licensee determined
'
that the problem was not due to check valve leakage but resulted from-
'
seat and mechanical joint leaks on two relief valves, accumulator tank plug leaks, and drain valve seat leakage. The leaking relief
.
- - - - - - - - -
I
.
.
.
.
.
.
valves were replaced, the accumulator tank plug leaks were repaired and the accumulator drain valves were disassembled, lapped and re-tested satisfactorily.
,
The inspector considers that some of the repairs and post-work leak i
test failures could have been avoided had the problem been fully de-
'
fined initially.
However, licensee corrective actions with respect t
to the leakage were acceptable.
.
4.4
"B" RBCCW Heat Exchanger Repair
'
When the "B" Reactor Building Closed Cooling Water (RBCCW) heat ex-changer was inspected during the outage, extensive cracks were found near the welds between the service water divider plate and the chan-nel assembly. While the welds did not represent part of the heat exchanger pressure boundary, failure of the welds could have degraded heat exchanger performance.
Initial weld repair identified inclusions in copper-nickel base mate-
'
rial and welds and cracks propagating beyond the drilled stop holes.
The licensee revised work plan FRN-89-17-45 to replace the enti e
'
divider plate. A new divider plate was fabricated and installed.
The operations Quality Assurance (QA) department wrote five noncon-formance reports (NCRs) during the installation and testing of the
new divider plate. All NCRs were dispositioned. satisf actorily.
Quality Control involvement in these work activities was confirmed by noting QC hold point sign-offs on Joint Process Control Sheets, by
<
observing QC visual examinations of the completed welds and by in-terviewing the responsible QC inspectors.
The inspector also interviewed the Maintenance Department Mechanical Supervisor and reviewed the maintenance work plan, noting good con-trol of the relevant design documents, the work package and log sheets, Both the supervisor and craftsmen were fully aware of
planned changes to the original scope of work and correctly delayed
implementation of any change in the scope until the work plan had been properly updated with authorized revisions.
Offsite lab analysis of metal samples indicated intergranular micro-cracks possibly caused by intergranular corrosion, voids in the old weld, and grinding notches on the weld bead at the front of the plate. The cracks appeared to have propagated over a period of months. All indications were typical of fatigue and/or corrosion fatigue processes. The probable source of the crack propagation sites is a weld repair performed on the plate in 1981.
'
The post repair pressure test revealed leakage around 37 plugged heat exchanger tubes.
The leaking plugs were brass and had deteriorated.
Many of the plugs " crumbled" or cracked as they were drilled and
.
tapped for extraction.
Some silicone plugs were also found; none of
.
.
.
l I
those were leaking. The design of the Pilgrim RBCCW heat exchangers requires the use of copper-nickel tube plugs.
Because BECo could not procure and does not stock copper-nickel plugs, a temporary modifi-cation was written to allow interim use of replacement brass plugs and to replace the installed silicone plugs in kind.
The temporary modification required that the brass plugs be inspected / replaced within 18 months or by the end of Refueling Outage (RFO) 8, whichever occurs sooner.
The safety evaluation concluded that the brass plugs would provide a safe interim repair and would reduce the possibility of sea water contamination of the RBCCW water, thus preserving the safety function of the heat exchangers.
Concurrent with the "B" heat exchanger repair, an engineering analy-sis of the "A" RBCCW heat exchanger was conducted to determine the need for equivalent inspe: tion / repair.
Based on the results, the
licensee deferred "A" RBCCW heat exchanger inspection to the sched-uled spring 1990 outage.
No justification was located for the as found mixture of brass and silicone tube plugs. A Potential Condition Adverse to Quality has been written.
.
In summary, the replacement of the partition plate in RBCCW heat ex-changer E209B was well controlled and performed. Management atten-tion was apparent. The failure analysis was particularly impressive.
The inspector had no questions or concerns regarding the conduct, control or QC inspection of the RBCCW work.
4.5 Reactor Water Clean Up (RWCU) Modifications The licensee has experienced numerous group 6 primary containment isolations initiated by RWCU high flow sensors DPIS-1243 and DPIS-1244 The licensee concluded the probable cause was air in the in-strument lines because they were routed in one area with a slope which was susceptible to air trapping.
Past efforts to remove the air by backfilling or blowing down have been ineffective due to the excess flow check valves and instrument snubbers.
After testing to define the long term resolution, some modifications were initiated, Maintenance requests MR 89-12-86 and 89-12-107 were completed during this inspection period.
The inspector reviewed the work packages for TP 89-82, and TMs 89-20 and 89-21 associated with
,
these maintenance requests and noted that the required safety evalu-ations and ORC reviews had been properly performed. An instrument
',
snubber pin configuration which would provide a better performance
was identified and installed.
Subsequent RWCU flow sensor perform-ance was normal.
-
'.
.
.
.
.
The instrument snubber pin modification and backfilling of the in-strument lines appears to have solved the spurious isolation problem.
The licensee is continuing to evaluate this issue to confirm that a permanent solution has been found. The inspector had no further questions.
4.6 Furmaniting of Residual Heat Removal (RHR) Valves RHR valves MO-1001-50, Inboard RHR Shutdown Cooling Suction Valve,"
and H0-1001-33A, " Loop 'A' LPCI Manual Isolation Valve for Injection Piping," both developed body to bonnet leakage.
Although MD-1001-50 has an inboard manual isolation valve, it is in the only suction flow path for decay heat removal.
HO-1001-33A has no inboard isolation valve.
Disassembly and repair of either of these valves would re-quire the reactor vessel to be defueled.
In order to minimize the leakage without an unplanned defueling, the licensee initiated Plant Design Change (PDC)89-049.
The PDC re-quired that both valves be inspected and hot torqued during the end of outage reactor vessel pressurization test.
If hot torquing did not sufficiently rduce leakage, Furmanite sealant application was authorized and the valves would be scheduled for disassembly and re-
-
pair during the next refueling outage.
During the reactor vessel pressurization, with temperature about 200
.
degrees F and pressure at about 1000 psig, both valves were hot torqued. No movement was noted on any of the bolts. After the cold torque check of both valves, MD-1001-50 was still leaking at about one gpm, and HO-1001-33A at less than 1/16 pint per minute.
The de-cision was made to apply Furmanite to the MD-1001-50 valve and to build a catch containment around the H0-2001-33A valve.
One Furmanite fitting was installed on the M0-2001-50 valve. An attempt to inject the compound failed because the fitting hole had not penetrated into the void area below the seal ring. The plant was de pressurize.d the job re planned, and plant startup commenced.
Recalculation and ultrasonic scanning of the MO-1001-50 valve led to
a change in the injection plan involving injecting above the seal ring into the void between the seal ring and the retainer ring. A second Furmanite fitting was installed on the M0-1001-50 valve and a successful injection was performed with the plant at about 3% power and about 900 psig. The leak stopped immediately with about half of the calculated amount of Furmanite having been injected and with only one of four planned fittings having to be installed.
Total enclosure metal catch containments were constructed around both the MO-1001-50 and the HO-1001-33A valves with drain lines routed to the drywell equipment drain sump.
Both valves will be inspected during the next time the drywell is open or the spring mid-cycle outage, whichever comes firs e o
o
.
-
-
.
The inspector found the planning for this repair to be conservative and that the job was effectively accomplished.
Radiation exposure to
workers was below the "As low As Reasonably Achievable" (ALARA) bud-get.
4.7 Local Leak Rate Testing Ouring the October maintenance outage, many containment isolation j
valves underwent Local Leak Rate Testing (LLRT).
Check valves under-i went reverse flow / exercise testing to comply with the requirements of
~
ASME Section XI. One valve, the RCIC Vacuum Pump Discharge Check Valve (1301-40), failed the test.
The valve was disassembled, cleaned, and re-assembled four times be-fore it passed post work testing.
Failure and Malfunction Report (F&MR)89-392 concluded the original failure was caused by a build-up
of corrosion or particulate matter in the low point of the piping where the valve is located.
Failure of the repair attempts was not discussed in this MR but was probably attributable to top works mis-alignment during re-assembly; a separate F&MR has been issued to evaluate the repetitive repair failures and to define the long term resolution of the valve leakage problem.
While difficulties were encountered by the Maintenance Department in the conduct of reiairs to the valve, the management of this and the many other LLRTs by the four Senior Test Engineers in the Reactor t
Safety and Performance Division was a noteworthy strength.
4.8 Pre-Job Briefing for Repair of MO 1201-133
'
The inspector witnessed a pre-job briefing of electrical maintenance personnel prior to performing work under MR 89-12-100, Correct Leak-
<
ing Bonnet Seal on MO 1201-133.
The inspector observed that the electrical supervisor had prepared for the briefing using the proce-dure 1.5.3, "Maintenarce Requests," attachment 5, Pre-Job Briefing form. The supervisor was observed to be knowledgeable regarding the specific work to be done and fielded questions asked by the mainten-ance personnel. Upon completion of the briefing, the maintenance personnel appeared to have an understanding regarding the specific work to be performed.
The inspector noted that several items on the pre-job briefing form were checked "not required." The pre-job briefing instructions state if any issue is 'Not Applicable" to the specific work scope, the basis for this decision will be stated. The inspector noted that
,
there was no basis stated on the Pre-Job Briefing form for items checked "not required." Discussions with the supervisor and his man-ager identified some confusion over the intent of the instruction as to whether or not a basis should have been specified when "not re-quired" was checked on the Pre-Job Briefing sheet. The difference
'
.
.
,
.
,
between "not required" and "not applicable" was not defined in the instruction.
The supervisor committed to review the matter and re-
'
commend procedure changes as necessary to make clear the intent.
5.0 Surve111ances 5.1 Review of Surveillance Results The inspectors observed or reviewed the following sample of recently completed safety-related system surveillances to verify that the tests had been conducted in accordance with approved procedures.
This review included review of the work plan, pre-job briefing, authorized signature, tagging, acceptance criteria and review by cog-nizant personnel.
,
!
8.M.1-3 APRM Functional 8.M.3-2 Instrument Line Flow Check Valve test TP89-92 Reactor Vessel Pressurization and Temperature for Class 1 System Leakage Test and Excess Flow Check Valve Tests 8.M.1-4.1 APRM Scram Clamp Maneuvers TP87-265 Recirc Scoop Tube Position Readings 8.M.1-13 Main Steam Line High Radiation Calibration 8.5.2.2 LPCI Pump Flow Rate Test and MOV Timing 8.5.5.4 RCIC Valve Operability 9.17 Core Flow Evaluation and Jet Pump Calibration 8.5.2.10 RHR Piping Temperature and Pressure Monitoring APRM Sotdown Functional Test TP89-20 S$W "A" Loop Flow Test 8.5.6.2 ADS Subsystem 8.E.42 Recirculation System Instrument Calibration 8.4.2.1 Hydrodynamic Test for Measuring Possible Bypass Leakage of SBLC 8.M.2-3.1 Rod Block Monitor Functional 8.M.3-14 H202 Analyzer System Semi-Annual Calibration 8.M.3.9 Liquid Radwaste Effluent Discharge Radiation Monitor 8.M.1-4 APRM Flow Bias Signal Calibration 8.9.8 Battery Rated Load Discharge Test 3.M.2-5.4.3 LPRM Detector Tests 3.M.2-5.4 APRM Calibration Instructions TP 89-106 Post-Work Testing of Relay ESR-2 Based on direct test observations and/or discussions with licensee cognizant engineers, the inspector determined that the tests were performed in accordance with approved test procedures.
Test results met test acceptance criteria or test exceptions were properly evalu-ated and satisfactorily dispositioned.
,
.
u
- - -
..
'
.
.
i
'
.
l 5.2 Excess Flow Check Valve Tests
,
At the end of the October maintenance outage, the licensee performed a cold pressurization of the reactor vessel in accordance with Tem-porary Procedure TP 89-92, " Reactor Vessel Pressurization and Tem-perature Control for Class 1 System Leakage Test and Excess Flow Check Valve Tests." During the pressurization, all of the approxi-mately 80 instrument line check valves were tested in accordance with procedure 8.M.3-2, " Instrument Line Flow Check Valve Test," to meet the requirements of the ASME Code Section XI Subarticle IWV-3420
" Valve Leak Rate Test."
The licensee fielded three teams in parallel to perform the in plant valve manipulations and testing.
All but two check valves met the leakage rate criterton of less than 1.0 gpm. A small number of
,
valves did not initially reset (open) on completion of the leakage
,
rate portion of the test, but all were exercised, reset, and the associated instruments returned to service.
The licensee resolved the two leak rate failures through the appli-cation of a previous engineering review (Engineering Service Request (ESR)88-448) which thoroughly reviewed the design of Dragon and Chemequip check valves. One valve which failed was a Dragon and the other was a Chemequip. The ESR had noted that Dragon valves (of which Pilgrim only has two installed) require a much higher flowrate (greater than 6 gpm) to seat than is generated with the existing in-strument tubing size and test equipment.
The ESR had recommended that the two installed Dragon valves not be tested again until Re-fueling Outage No. 8, when either the instrument lines would be modi-fied, or weaker springs installed in the valves.
The Dragon valves
.
were inadvertently scheduled for testing during the October mainten-l ance outage. Since the ESR further evaluated failures of Dragon
!
valves as a testing problem, rather than a safety problem, no further i
action was taken to repair or test the one Dragon valve which had been inadvertently tested, The one (out of about 78) Chemequip valve which failed the leak test i
I had a flowrate of 1.2 gpm.
The previous ESR had noted that the manu-facturer's design leakage rate for these valves is 1.25 gpm and that the FSAR value is 2.0 gpm.
Based on this information, the valve was not repaired or retested. The licensee is contemplating relaxing the conservative 1.0 gpm leakage rate criteria which is an internal, ad-ministrative limit to 1.25 gpm (manufacturer's specification).
,
The inspector noted that no unexpected trip signals or isolations were received during the testing, and that all radiological controls practices at the instrument racks were proper.
From start to finish, with the exception of the testing of the Dragon valve, this set of tests was well planned and executed, i
l l
.
.
,
.
.
.
.
5.3 Salt Service Water (SSW) System Operability Testing per TP 89-20 Section 10.5.5.3 of the PNPS Final Safety Analysis Report (FSAR)
states that each Reactor Building Closed Cooling Water (RBCCW) heat exchanger (HX) is capable of removing the design Residual Heat Re-moval (RHR) system heat load during accident conditions assuming a minimum Salt Service Water (SSW) flow of 5000 gpm at 65 degrees F.
This information was also provided in the licensee's October 15, 1981 response to IE Bulletin (IEB) 81-03, " Flow Blockage of Cooling Water to Safety System Components by Corbicula (Asiatic Clam) and Mytilus (Mussel)," but is not an explicit requirement in the plant Technical Specifications.
In response to IEB 81-03, the licensee developed an administrative testing requirement to prove $$W heat removal cap-ability on a monthly basis and hcs implemented this requirement through a temporary procedure (TP) since 1982.
As described in NRC Inspection Report 50-29V89-10, Section 2.3.5, on September 25, 1989, the "A" RBCCW HX failed to meet the minimum SSW flow requirement specified by the current TP 89-20. Two separate heat exchanger cleaning and retesting evolutions were also unsuccess-ful.
The licensee initially suspected that the $$W pumps were de livering low flow rates, but this theory was disproved when the pump discharge pressures were compared to the manufacturer's pump curves.
The licensee then postulated that the flow element (flow orifice FE 6240) for the heat exchanger inlet might be fouled with mussels such that it acted more like a venturi than a flow orifice. The decreased flow indication from FE 6240 would be explained by the decreased pressure drop produced by a venturi when compared to a flow orifice.
The licensee verified the presence of mussels in the heat exchanger inlet through visual examination, and removed them manually and through hypochlorination.
To demonstrate that the heat exchanger was passing adequate flow and to ensure adequate heat removal from RHR system, the licensee elected to use previously obtained heat exchanger differential pressure versus flow data that were generated by computer codes and validated by actual field testing in 1982.
Since the "A" RBCCW heat exchanger has 51 (or approximately 2.5%) of its tubes plugged, the licensee developed a modified pressure drop versus flow curve to incorporate these losses. The licensee then verified this data using three dif-ferent analytical methods.
Revision 4 of TP 89-20 was issued to in-corporate this alternate flow determination method for the "A" RBCCW heat exchanger, and a successful surveillance was subsequently con-ducted.
The inspector found the alternate flow determination method to be acceptable because: (1) it was validated by actual field testing; (2) the licensee accounted for the plugged heat exchanger tube losses; and (3) the procedure mandates a heat exchanger cleanliness verification within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of the start of the test.
Also, it is
,s
.
.
.
.
.
not likely that the licensee would use this method indefinitely since i
the cleanliness verification is rigorous and requires entry into a seven day limiting condition for operation (LCO) per TS 3.5.B. " Con -
tainment Cooling."
In October, the licensee again experienced a series of TP 89-20 sur-(
veillance failures for both the "A" and "B" RBCCW heat exchangers-and developed Revisions 5 and 6 to TP 89-20 to allow use of the differen-
,
tial pressure method for the "B" heat exchanger, as well as introduc-ing the use of an ultrasonic flow meter.
However, licensee attempts
.
to utilize the ultrasonic flow instrumentation were unsuccessful.
On
'
November 1 and 2, 1989 both heat exchangers passed TP 89-20 using the origin 6l flow orifice method.
The licensee concluded that the tem-
-
porary loss of the flow orifices was caused by a combination of silt
clogging in the FE instrument lines that had later been blown clear, and the reduced inlet line mussel buildup af ter hypochlorination.
The inspector found this explanation to be plausible and had no fur-i ther questions.
5.4 Charging of the 250 Volt DC Station Battery Following " Battery Rated
)
Load Discharge Test" i
.
The licensee conducted procedure 8.9.8, " Rated Load Discharge Test,"
for the 125 VDC and 250 VDC safety-related batteries during the out-age.
Two problems were encountered in satisfactorily completing the
250 VDC battery charge, which followed the load discharge test con-ducted on October 15, 1989. As the battery reached the required ter-minal voltage, many of the cells had not reached acceptable specific gravity levels (i.e., cell electrolyte was stratified).
The strati-fication problem was solved by blowing air into all 120 cells (in accordance with the battery technical manual) to agitate the electro-lyte and by placing the battery back on charge. The second problem encountered was a failure of six cells to reach the minimum accept-able post-charge voltage. The six cells were adjacent to each other (physically and electrically) on the upper rack of the battery.
This
,
!
problem was solved by individually charging the cells for approxi-mately one day and then placing the battery on a 270 volt float
charge. The entire discharge and charge evolution for the 250 VDC battery took about 8 days,
,
The inspector verified that the six cells with low voltage were not
!
those showing the most amount of loose glass mat material and float-ing particulate matter in the electrolyte (as described in section l
9.1).
To the contrary, the six cells are among the cells in the bat-tery having the cleanest electrolyte.
The individual cell charging technique used to bring the cells into specification was one of two possible options available.
The alternative would have been to leave the entire battery on charge until the individual cell voltages in-creased to acceptable values.
l l
l l
.
e
,
,
,
.
.
,t
i
The inspector concluded that the licensee followed acceptable prac-I tices per the battery technical manual. The inspector had no further
,
questions.
6.0 Radiological Controls
,
6.1 Aydits The inspector reviewed the licensee's auditing e.v.eight of the radiological controls programs relative to requirements in Technical Specification B.8 Audits and 10 CFR 50 Appendix B Quality Assurance Criteria.
Performance was determined from interviews with personnel from the Quality Assurance and Radiological Controls Departn'ents and a review of the following audit reports:
,
Radiological Technical Section: RTS89-110, RTS89-128, RTS89-138 i
--
Quality Assurance Department: QAD 88-1247, QAD 89-258, Surveil-
-
lances 89-2,1-1, 89-2.1-3, 89-9.1-3, 89-9.1-4, 89-2.3-1, and 89-2.1-4
,
Within the scope of this review the inspector determined that audits
-
of the radiological controls program satisfied the baseline require-ments.
However, the inspector noted that the auditors had not been recently updated regarding state of the art radiological protection
,
practices; nor were outside technical experts hired for the audit teams.
The Station Director stated that personnel will be encouraged to visit other stations to observe performance of other programs.
This matter will be reviewed in future inspections.
6.2 Flat Bed Filter Spill i
On October 17, 1989, the B flatbed filter filtrate tank overflowed, f
A few gallons of contaminated water flowed out of the room into the i
corrider of the radwaste building.
The inspector reviewed the status
'
of the licensee's investigation of this matter through discussions with radwaste and RP personnel and a tour of the area.
- The licensee determined that the root cause of the event was multiple I
equipment failures in the air supply used to dry the filter media diatomateous earth.
Short term corrective action involved a shutdown
'
of the "A" and "B" filter train for inspection and repair. A tem-i porary curb has been installed in each room to contain any leakage in the future.
The inspector noted that the " spill" was nimilar in one respect to l
the event in the trucklock (Inspection Report 50-293/88-35) in that the work area was not designed to contain a spill of liquid.
How-ever, most radwaste areas appear to have adequate curbs and drains.
l The licensee stated that permanent curbing foe the filter room will
!
be considered during deliberations of long tr rm corrective actioli, e
P I
s
.
c.
r
.
20
.
A radwaste program improvement was noted in that the High Integrity Containers (HIC) for solid radwaste are no longer stored outside prior to shipment for burial. All filling and loading will be done inside the radwaste building.
Within the limited scope of this radwaste review, the inspector de-termined that the radwaste program is improving.
6.3 As Low as Reasonably Achievable (ALARA)
_
Earlier this year, the licensee transferred part of the ALARA group
,
from Health Physics to the Engineering Division. This improved the work and ALARA preplanning as evidenced by good performance during
,
the October 1989 outage.
For example, in the undervessel area, two Position Indicator Probes (PIP) were repaired and one Intermediate Range Monitor (IRM) was replaced. ALARA preplanning consisted of
,
historical review of previous performance, fabrication of special
'
tools, formal incorporation of specific ALARA steps in the work pro-cedure, and extensive prework briefings. The man-rem exposure for the IRM replacement was one third of the previous exposures. All other preplanned work was completed below original estimates. Only the " emergent work" identified during the outage which lacked pre-planning was found to exceed the exposure estimates.
The licensee approach to ALARA during outages has improved.
7.0 Fire Protection 7.1 Fire Protection / Prevention Program The inspector reviewed several documents in the following areas of the program to verify that the licensee had developed and implemented adequate procedures consistent with the Fire Hazard Analysis, Final Safety Analysis Report, and Technical Specifications. The documents reviewed, the scope of review, and the inspection findings for each area of the program Ore described in.the following section.
,
7.2 Program Administration and Organization The inspector reviewed the following licensee documents:
,
Technical Specifications, Section 6, Administrative Controls;
--
and Nuclear Organization Procedure NOP83FP1, dated August 1, 1985.
--
The scope of review was to ascertain that personnel were designated for implementing the program at the site and qualifications were de-
',
lineated for personnel designated to implement the program,-
e
.
,
.
.
.
7.3 Administrative Control of Combu_s,tibles and Ignition Sources e
The inspector reviewed the following licensee documents:
>
Procedure 1.4.2 revision 6, dated October 2, 1985 entitled "Per-
--
sonal Smoking;"
Procedure 1.4.3 revision 16, dated September 8, 1989 entitled
.
!
--
" Administrative Controls for Safe Use of Flammable / Combustible
.
Liquids, Ignitable Gases;"
Procedure 1.5.5 revision 17, dated October 10, 1989 entitled
--
" Cutting, Welding and Hot Work Fire Safety;" and Procedure 8.B.20 revision 2, dated May 12, 1989 entitled " Fire
--
Hazards Inspection."
The scope of review was to verify that the licensee had developed administrative controls which included: (a) special author 17ation for the use of combustible, flammable or explosive hazardous rnaterial in
,
safety-related areas; (b) prohibition on the storage of combustible.
'
flammable or explosive hazardous material in safety-related areas; (c) the removal of all wastes, debris, rags, oil spills or other com-
,
bustible materials resulting from the work activity or at th4 end of each work shif t, whichever is sooner; (d) all wood used in safety-
-
related areas to be treated with flame retardant; (e) periodic in-spection for accumulation of combustibles; (f) transient combustibles to be restricted and controlled in safety related areas; (g) house-keeping to be properly maintained in areas containing safety-related equipment and components; (h) requirements for special authorization (work permit) for activities involving welding, cutting, grinding, open flame or other ignition sources and that they are properly safe-guarded in areas containing safety-related equipment and components; and (1) prohibition on smoking in safety-related areas, except where
" smoking permitted" areas had been specifically designated by plant management.
No unacceptable conditions were identified.
7.4 Other Administrative Controls The inspector reviewed the following licensee documents:
Technical Specifications, Section 6, Administrative controls;
--
Plant Design Change / Technical Specifications Review Form (NOP-
--
83E1);
Nuclear Organization Procedure N0P-83FP1;
--
General Employee Training Indoctrination;
--
QA Manual Section 2.4.3 (Volume 2); and
--
Specification M-504.
--
.
-w
e s
,
.
.
.
.
The scope of review was to verify that the licensee had developed administrative controls which require that: (a) work authorization, construction permit or similar arrangement is provided for review and approval of modification, construction and maintenance activities which could adversely affect the safety of the facility; (b) fire brigade organization and qualifications of brigade members are de-lineated; (c) fire reporting instructions for general plant personnel are developed; (d) periodic audits are to be conducted on the entire fire protection program; and (e) fire protection / prevention program is included in the licensee's QA program.
No unacceptable conditions were identified.
7.5 Equipment Maintenance, Inspection and Tests The inspector reviewed the following randomly selected documents to determine whether the licensee had developed adequate procedures which established maintenance, inspection, and testing requirements for the plant fire protection equipment:
Procedure 8.B.1 revision 29, " Fire Pump Test;"
--
Procedure 8.B.11 revision 12 (Attachment A), " Fire Valve Oper-
--
-
.
ability;"
Procedure 8.8.13.1 revision 4, " Hydrostatic Testing of a Fire
--
'
House;"
Procedure 8.B.17.1 (Attachment B) revision 1, " Fire Door Inspec-
--
tions - Technical Specifications;"
Procedure 8.B.17.2 (Attachment B) revision 0, " Fire Damper In-
--
spection - Technical Specifications;"
Procedure 8.B.22 (Attachment A) revision 9, "Halon 1301 System
--
Cable Spreading Room;"
,
Procedure 8.B.3.1 (Attachment A) revision 4, " Interior Fire Hose
--
and Cabinets - Technical Specifications;"
Procedure 8.B.4 (Attachment A) revision 22 " Smoke and Heat De-
--
tection and Supervisory Circuitry Checklist;"
,
Procedure 8.B.9 revision 21, " Wet and Dry Pipe Sprinkler Sys-
--
tem;"
Procedure 8.B.2 revision 24, " Fire Water Supply Shutoff Valve
--
Inspection;"
Procedure 8.B.20 revision 2, " Fire Hazards Inspection;" and
--
Procedure 8 B.19 revision 5, " Fire Brigade Equipment Inspec-
--
tion."
No unacceptable conditions were identified.
7.6 Fire Brigade Trainino The inspector reviewed the following licensee procedures:
Nuclear Training Manual - Special Training Section 4.2, " Fire
--
Brigade Team," revision 11, dated June _8, 1989; and I
.'.
.
.
.
.
.
Procedure no.1.4.23 revision 10. " Fire Brigade Training Drill."
--
The scope of review was to verify that the licensee had developed administrative procedures which included: (a) requirements for an-nounced and unannounced drills; (b) requirements for fire brigade training and retraining at prescribed frequencies; (c) requirements for at least one drill per year to be performed on a "back shift" for each brigade; and (d) requirements for maintenance of training records.
No unacceptable conditions were identified.
The inspector also reviewed training records of fire brigade members for calendar years 1988 and 1989 to ascertain that they had attended the required quarterly training, participated in a quarterly drill, and received the annual hands-on fire extinguishment practice.
No unacceptable conditions were identified.
7.7 Periodic Inspections and Quality Assurance Audits l
The inspector reviewed the QA annual Audit Report No. 88-42, dated December 27, 1988, QA Biennial Acdit Report No. 87-14, dated June 19, 1987 and QA Triennial Audit Report No. 87-69, dated November 27,
1987. The scope of review was to ascertain that the audits were con-ducted in accordance with the Technical Specifications and that audit findings were being resolved in a timely and satisfactory manner.
No unacceptable conditions were identified.
'
7.8 Facility Tour The inspector examined fire prctect<cn water systems; including fire pumps, fire water piping and distribution systems, post indicator valves, hydrants and contents of hose houses. The inspector toured accessible vital and nonvital plant areas and examined fire detection and alarm systems, automatic and manual fixed suppression systems, interior hose stations, fire barrier penetration seals, and fire doors. The inspector observed general plant housekeeping conditions and randomly checked tags of portable extinguishers for evidence of periodic inspections.
No deterioration of equipment was noted.
The inspection tags attached to extinguishers indicated that monthly in-spections were performed.
No unacceptable conditions were identified.
r
)
- -
.
.
.
.
.
,
8.0 Outage Assessment
,
The inspector observed maintenance outage activities in the control room, the outage management office and the various in plant work locations.
Attention was directed more toward corrective maintenance activities than toward surveillance, with a higher percentage of observation of mechanical than of electrical and I&C activities.
Two weaknesses were observed dur-ing this outage. The first area was that of housekeeping lapses at plant work locations.
In several areas (RBCCW heat exchanger room, air compressor room, radwaste area below entrance to condenser bay) an excessive amount of tape, paper towels, buckets, tools, grinding wheels, and plastic bags were found adrift on the floor.
These housekeeping observations were made over the course of five days, both with work in progress and between shift changes.
Plant areas were returned to the normally observed high standards of v
housekeeping prior to start-up.
None of the housekeeping practices pre-sented a radiological or foreign material exclusion concern.
'
The second weakness area was that of protective tagging procedures.
In-spector findings on tagging are detailed in Section 2.4.
.
Overall, the outage was conducted in an extremely professional, well or-ganized and conservative manner.
Particularly noteworthy were the con-servative management decisions made at several junctures.
Examples included: (1) several reactor building (secondary containment)
doors were repaired or replaced during the outage. Although there was no Technical Specification or 10 CFR 50, Appendix J requirement to perform a secondary containment leak test as post work test for these repairs, lic-ensee management elected to conduct a full secondary containment leak rate
'
test; (2) the "B" RBCCW heat exchanger inlet divider plate was found cracked during initial open/ inspect work (section 4.4).
The initial approach taken was to grind out the cracked area and perform a repair using backing plates and weld repair.
During weld preparation, slag in-clusions from initial installation or previous weld repairs was dis-covered.
The licensee evaluated the inclusions, and rather than proceed with weld repair, the licensee again made the conservative decision to cut out the old divider plate and replace it with new material; and (3) one of the final evolutions in the outage was reactor vessel pressurization to perform a drywell instrument line check valve excessive flow test. The licensee made the conservative decision to have both trains of ECCS avail-able prior to conducting this test. When the aforementioned problem with the "B" RBCCW heat exchanger was discovered, the licensee could have de-cided to conduct the test with a single train of RBCCW. Again, licensee management made the conservative decision to wait until
"B" RBCCW train was returned to service (to provide ECCS pump and drywell cooling) before conducting vessel pressurizatio o.
,
.
.
.
.
Another noteworthy strength was the coordinated and controlled management of all outage activities from a special conference room. The office (con-ference room) was staffed around-the-clock with knowledgeable and profi-cient engineers, including tM outage management director. All mainten-ance, operations, health physics, chemistry and services departments appeared to understand the function of the office and supported its acti-vities with sound advice and work status feedback.
Briefings provided by outage management during control room turnover were thorough, informative and accurate.
Operations cognizance of maintenance activities during the outage was im-
-
proved from that observed during the preceeding nine months of the PATP.
The majority of the control room briefings and operations interface was provided by a newly hired licensee employee.
His grasp of outage manage-ment was particularly impressive in view of his recent tenure at Pilgrim.
The inspector interviewed several control room operators to determine their appreciation f or the need to control outage activities affect'ng decay heat removal (shutdown cooling).
Operators stated that they had all received training of this subject within the past two months.
Cont.ol room supervisory personnel (SR0s) were observed carefully considering the
-
opening of each new work package and the commencement of surveillances l
which could affect decay heat removal and reactor vessel level control.
The outage was conducted with a minimum of outside contractor support.
The relatively smooth conduct of the outage by a largely in-house team provided a positive indication as to the long term capability of the Pilgrim staff to maintain the plant following the Power Ascension Test Program.
9.0 F_ollowup on Previous NRC Open and Unresolved Items 9.1 (Updated) Unresolved Item (88-12-01): Battery Maintenance / Surveillance This item deals with three aspects of the battery maintenance / surveil-lance:
(1) the licensee procedure 8.9.8, " Battery Rated Load Dis-charge Test," had a deficient formula to compute the Battery capac-ity; (2) the use of uncontrolled hydrometer for measuring the speci-fic gravity and (3) inadequate control of water used for replenishing battery fluid.
The inspector reviewed revision 19 of the above referenced procedure.
The licensee had made some changes to the procedure to correct the formula for computing the battery capacity.
However, the temperature
'
corrections factor tabulated in attachment 13 did not agree with the recommended table in IEEE-450-1986, " Recommended Practice in Main-tenance, Testing, and Replacement of Large Lead Storage Batteries for
,
~~
,.
,
.
.
.
.
Generating Stations and Substations." This difference in correction factor will not lead to any significant errors. The licensee agreed to look into this matter and make changes as needed.
The inspector performed a physical inspection of the class IE bat-teries. Several positive terminals appeared to have discoloration, indicative of corrosion. The battery post seals were warped et several locations. A thick coat of a foamy substance was accumulat-ing on the top of the electrolyte.
Strands of glass mat separator material in the battery were peeling off. These concerns were dis-cussed with the battery manufacturer's representative, who was at the site during this inspection.
The manufacturer's representative stated that the warping of the post seal and corrosion signs at the positive terminals are the effect of oxidation under the seal due to the electrolysis process that occurs when electrolyte spilled on the cell top leaches under the seal.
The post seal expands resulting from the production of lead peroxide which requires a higher volume than lead.
The manufacturer has re-designed the post seal to eliminate this problem.
Pilgrim has a few cells with the redesigned posts, none of which show the warping prob-lem.
-
'
The inspector made independent measurements at the discolored termi-nals during the battery capacity test. The reading taken using the
,
infrared thermometer did not indicato any significant concerns as the temperature readings varied between 64 degrees F and 86 degrecs F.
The voltage drop readings taken during the capacity test showed a difference of 1 millivolt between a good bolted terminal connection and a discolored connection. The licensee procedure showed an acceptance value of 20-25 millivolts for inter cell voltage drop.
The inspector inquired as to the basis for this value and the ade-quacy of the DC system if all the connections were to have 24 milli-volts of voltage drop. The licensee agreed-to evaluate the accept-
,
ability of this value.
Presertly, the intercell voltage readings on battery A are much lower than 24 millivolts.
The licensee method of measuring voltage drop during a capacity test is not targeted towards specifically measuring the effect of corro-sion at the connections. As the corrosion of the battery terminal-has become a recurring problem, the licensee agreed to develop a
-
better method of measuring contact resistance.
The battery manufacturer's representative stated that the whitish gray powdery substan:o on the top of the electrolyte is glass mat and therefore not an objectionable material since it was generated from glass mat within the battery.
In order to assure the benign nature of this material the licensee agreed, af ter NRC prompting, to perform
.
--
-
s.
,
.
.
.
a chemical test and establish if it could become electrically con-ductive.
Subsequent analysis of this material confirmed that the material is particulate glass mat.
The material is therefore con-sidered unsightly but not detrimental to operation of the batteries.
In light of the questionable appearance of the class 1E batteries, the licensee had committed to perform a comprehensive test on one of the apparently questionable cells. The licensee representative stated that the cells that were shipped to the factory were dried and neutralized, which limited the laboratory testing to visual observa-tion and teardown analysis. No electrical tests were performed.
This analysis did not raise any significant concerns.
The inspector also inquired about the possibility of the dislodged fibrous material blocking the passage of gases to the top of the cell.
The manufac-turer's representative stated that the separator plates facing the positive plates have vertical groves to facilitate the gas bubble travel to the top.
This design feature should prevent the strati-fication of electrolyte which might cause varying specific gravity.
A previous NRC inspection report (50-293/88-12) noted that hydro-meters and battery water were f.ot being strictly controlled.
As of the end of this report period, the battery water is obtained in a controlled manner from the plant chemistry department, and the new electronic hydrometers are maintained under the measurement and test equipment (MLTE) control program.
No further problems with water and hydrometers were noted.
The inspector witnessed the capacity test conducted on 125V battery
"A" during October 17-18, 1989.
This was a constant current test at the rated capacity and the duration was measured until the terminal voltage reached 1.75 volts per cell.
The test demonstrated a battery capacity of 97.85 percent.
This result provides a reasonable assur-ance that the battery can perform its safety functien in the immedi-ate future, in spite of the concerns raised.
The concerns raised above need to be addressed to assure the long term performance of the batteries.
This item will remain unresolved pending the licensee's action on (1) justifying the correction factor for battery capacity and (2) a program to address an acceptable value and related mainten-ance program for the voltage drop at battery connections.
The remaining open issue from previous battery inspections by the NRC is the need for BEco to correct the battery gravity temperature cor-rection tables in their procedure so that they agree with the tables in IEEE-450-1936 " Recommended Practices for Maintenance Testing, and Replacement of Large Lead Storage Batteries for Generating Stations and Substations."
_.
..M 0-l
.
'
,
9.2 { Closed) Unresolved Item (89-01-01) Development of an Inspection Sched;1e for MSIV 4-Way Solenoid Valves
,
The licensee has documented, in the closeout of the root cause evalu-
ation (BECo SG89-488), that all eight of the 4-way solenoid valves d
will be removed, inspected and reinstalled during the next refueling-outage (RFO-8).
The Maintenance Requests (MRs) for performing the work have been scheduled on the licensee's outage scheduling system.
The licensee has also committed to continue quarterly trending of MSIV stroke times. This trend analysis will be used to revise the inspection schedule, if problems or degradation are noted.
This item is closed.
9.3 (Closed Unresolved Item 50-293/89-01-02 Missed Hourly Fire Watch.
>
This item pertained to a missed hourly fire watch in 4160 KV switch-gear room "A" and the adequacy of the reporting requirements.
Tech-nical Specification 3.12 requires the establishment of a fire watch upon loss.of detection, suppression or other fire barrier systems.
The watches can be hourly or continuous.
For hourly watches, the fire watch must sign and r he time in the area per Procedure 8.B.14.
For raissed houri i watches, the Fire Watch Supervtsor takes immediate correctiv.
Jon, documents the occurrence, notifies the Senior Fire Protecti'
sineer (SFPE), when practical.
Upon notification by the Fin
.i Supervisor of any variance from sta-tion procedure 8.B.14 -
iments, the SFPE reviews the documenta-
'
tion and corrective and determines if the Technical Specifi-cation requirements in.. tion 3.12.have been challenged.
If so, the SFPE notifies the Watch Engineer of a possible deviation, the Watch Engineer then determines whether it is reportable. The inspector determined this to be acceptable.
This item is closed.
9.4 (Closed) Unresolved Item 50-293/87-53-02.3 Event Classification and Notification During the AIT inspection 50-293/87-53, the team noted that the Emer-gency Plan provided four emergency condition classifications: Unusual Event, Alert, Site Area Emergency, and General Emergency.
The Un--
usual Event and the Alert classifications involve degradation of the safety of the plant and precautionary measures are initiated.should the situation degrade.
For loss of power conditions, an Unusual Event would have been declared if there was irradiated fue'l in the vessel and reactor cociant system (RCS) temperature was greater than 212 degrees F, and (a) there was a loss of all onsite ac power capacity or (b) there was a loss of all offsite power. An Alert (for loss of power conditions) would have been declared if there was irradiated fuel in the vessel and RCS temperature was greater than 212 degrees F, and (a) there was a loss of all onsite de power for 15 minutes or less, or (b) there was a loss of all offsite power coincident with loss of both emergency diesel generators.
!
e as a
+
,s2
.
.
t
.
.
The event at the Pilgrim station on November 12, 1987 involved a total loss of offsite power and the loss of the "B" emergency diesel generator (EDG). During the AIT interviews, licensee personnel stated that an " Alert" should be declared if the "A" EDG was lost.
'
The loss of the "A" EDG would have resulted in a total loss of off-site power and the total loss of onsite ac power.
However, as
pointed out by the AIT, based on criteria in the Pilgrim Emergency
.
Plan, an Alert would not be declared for this condition unless the RCS temperature was above 212 degrees F.
The licensee has re-evaluated the-emergency action _ level regarding i
~
'
loss of onsite and offsite power for situations where fuel is loaded in the reactor vessel and RCS temperature is less than 212 degrees F.
The licensee considered.that not requiring any emergency classifica-tion for any loss of power if the RCS temperature is less than 212
-
degrees F was not appropriate. This is because RCS temperature, in i
and of itself, is not indicative of conditions which preclude the possibility of degraded plant safety as a result of loss of power.
l Therefore, the Emergency Plan Emergency Action Level (EAL) document was revised regarding loss of power capabilities and is not contin-gent upon RCS temperature.
The revised Emergency Plan criteria re-quire the declaration of an-Unusual Event if either onsite or offsite
-
ac power capability is lost. An Alert is declared if both onsite and offsite ac power capability is lost. Based on the above, the licen-see has adequately addressed the AIT concern regarding re-evaluating the emergency action levels and this item is closed.
It should be
.
noted that this revision was part of a major revision to the Pilgrim EAL scheme.
That revision was recently ' submitted to NRC for review and comment.
9.5 (Closed) Unresolved Item 50-293/87-53-02.2 Prior to the November 12, 1987 loss of offsite power event at the Pilgrim Station, the station was in an extended outage.
The plant had significant components and equipment for a large number of sys-tems out of service for maintenance. This put the plant in a con-figuration that limited its operat unal flexibilities. Had this event occurred under other circumstances, such as substantial decay heat, the operational inflexibilities of the plant could have had impact on the operator's ability to cope with the situation. At the time of the actual event, however, decay heat was_very low and reac-tor safety was never a factor.
At the Pilgrim Station, there are three methods of supplying offsite power to the plant: startup transformer, shutdown transformer, and.
main transformer with main generator phase bus links removed (back-scuttle).
The startup transformer is the preferred ac_ power source when the station is shutdown which was the method being used prior to the event. The other two alternate methods are backup to the pre-ferred method.
However, both of these methods were not readily
~
.
M
-
- ---
-- -. -. - -
'T J
h g
.
.
.
.
'
'
.
available. The shutdown transformer was out-of-service because of maintenance and modifications which included the tie-in of a third diesel generator. The main transformer was not readily available-because the main generator phase bus links were still installed.
i Some of the other equipment out of service affected the availability of the instrument air system..Two of the three reciprocating air compressors were tagged out of service and the third had its control switch turned off. When the loss of offsite power occurred; both onsite powered screw compressors were lost, none of the three re-ciprocating compressors started, and a' total loss of instrument air occurred.
The licensee was requested to describe what considerations will be made in the future to assure that essential and nonessential equip--
ment removed from service for outage maintenance do net create undue operational inflexibilities. To address this concern, the licensee reviewed their work control system to determine where changes were needed to improve work coordination with the plant shutdown. As a result of this review, the licensee has issued two procedures: Main-tenance Requests, Procedure No. 1.5.3 and Maintenance Work Plan, Pro-cedure No. 1.5.3.1.
The Maintenance Requests Procedure describas the administrative controls for maintenance request used for itaplementing
-
plant modifications, corrective / preventive maintenance and repair of equipment. As part of the Maintenance Requests Procedure a work I
prioritization team (WPT) was formed.. The WPT consist of members
from Operations, Maintenance, CMG, P&OM System, Fire Protection, and a
Security. The WPT meets each day to review and approve maintenance.
The Maintenance Work Plan procedure establishes the measures to be
!
used for the preparation, review, approval and issuance of a main-
!
tenance work plan.
The Maintenance Work Plans are use to control and document plant maintenance, modifications, and repair / replacement activities for both "Q" and "Non-Q" components, systems, and struc-tures.
In the Maintenance Work Plan Procedure, the Nuclear Watch Engineer / Nuclear Operations Supervisor have been given, among other responsibilities, the responsibility to review plant / system impact-
matrix for impact on the plant. With the issuance of both the Main-
-
tenance Request Procedure, Procedure 1.5.3, and the Maintenance Work-Plan Procedure, Procedure 1.5.3.1, the licensee overall maintenance program has been improved.
The planning and coordination efforts
,
with Operations will' help to ensure that operational-coordination and I
work priorities are established and maintained. Based on the above,
,
this item is closed.
- !
10.0 Management Meetings At periodic intervals during the inspection period,' meetings were held with senior facility management to discuss the inspection, scope and pre-liminary findings of the resident inspectors. A final exit interview was conducted on December 15, 1989.
No written material was given to the licensee that was not previously available to the public.
,
.
- -
-
-
-
-
.c'a*.O
.
-...
s
'
,
s Attachment I to Inspection Report 50-293/89-12 Parsons Contacted R.
Bird, Senior Vice President - Nuclear K. Highfill, Vice President, Nuclear Operations and Station Director R. Anderson, Plant Manager
,
D.
Eng, Outage and Planning Manager
!
E.
Kraft, Deputy Plant Manager R.
Fairbanks, Nuclear Engineering Department Manager D.
Long, Plant Support Department Manager L. Olivier, Operations Section Manager N.
DiMascio, Radiological Section Manager J.
Seery, Technical Section Manager
-
G.
Stubbs,. Maintenance Section Manager T.
Sullivan, Chief Operating Engineer J. Neal, Security Division Manager W.
Clancy, Systems Engineering Division Manager B. Sullivan, Fire Protection Division Manager
!
!
l l
i I
l l
l l
L