ML14191B073
| ML14191B073 | |
| Person / Time | |
|---|---|
| Site: | Robinson  |
| Issue date: | 01/19/1989 |
| From: | Colon T, Merriweather N, Paulk C NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML14191B072 | List: |
| References | |
| 50-261-88-36, NUDOCS 8902030088 | |
| Download: ML14191B073 (8) | |
See also: IR 05000261/1988036
Text
&x
,pg
EG Ut
0
UNITED STATES
NUCLEAR REGULATORY COMMISSION
REGION il
101 MARIETTA ST., N.W.
ATLANTA, GEORGIA 30323
Report No.:
50-261/88-36
Licensee: Carolina Power and Light Company
P. 0. Box 1551
Raleigh, NC 27602
Docket No.:
50-261
License No.: DPR-23
Facility Name: H. B. Robinson
Inspection Conducted: December 19 -
22, 1988
Inspectors:
e___,___
N. Merriweather
Date Signed
C.R1
Da e Signed
Approved b
. E. Conlon, Section Chief
Date Signed
Plant Systems Section
Engineering Branch
Division of Reactor Safety
SUMMARY
Scope:
This special, announced inspection was conducted to followup on
recently identified operating events involving Non-EQ splices on
Containment Fans,
RCS High Point Vent System not on the EQ Master
List and not EQ,
and improper setpoints on Motor Control Centers
(MCCs) 5 and 6 Feeder Breakers.
Results: In the areas inspected, three violations were identified involving:
(1) Unanalyzed Loss of Safety-Related MCCs 5 and 6, paragraph 3.a
(2) HVH 1-4 Penetration Splices Being Non-EQ, paragraph 3.b.
(3) Non-EQ Reactor Vessel Head Vent System, paragraph 3.c.
No specific strengths or weaknesses were identified in the licensee's
corrective actions for these operational events.
However,
the
licensee has made a concerted effort through the Design Basis
Reconstitution Program and the EQ Program Assessment Project to
identify, evaluate and correct problems at the plant.
Management
seems to be supporting this effort at all levels.
(- - --
)3C)::
=C901 24
r-fnR
gAi:CK ()500026~'%1
REPORT DETAILS
1. Persons Contacted
Licensee Employees
- D. Baur, Supervisor, QA (Acting Director-QA/QC)
- S. Clark, Supervisor, Configuration Control
- J. M. Curley, Director, Regulatory Compliance
- C. R. Dietz, Manager, Robinson Nuclear Project Department
- W. J. Flanagan, Manager, Modifications and Configuration Control
- E..M. Harris, Director, On-Site Nuclear Safety
- M. Heath, Project Engineer-EQ Coordinator
- G. Shartzer, System Engineering Representative
- D. A. Sayre, Senior Specialist-Regulatory Compliance
Other licensee employees contacted during this inspection included
engineers,
operators,
security force
members,
and administrative
personnel.
NRC Resident Inspector
L. Garner
- Attended exit interview
2. Action on Previous Inspection Findings (92701)
a. (Closed)
Unresolved Item 50-261/88-30-01, Investigate Circumstances
Surrounding Improper Setpoints of MCC-5 and MCC-6 Feeder Breakers.
For discussion of findings see paragraph 3.a.
b. (Closed)
Unresolved Item 50-261/88-30-02,
Review Circumstances
Surrounding HVH 1-4 Penetration Splices Being Non-EQ. For discussion
of findings see paragraph 3.b.
3. Onsite Followup of Licensee Events at Operating Power Reactors (92700)
a. During reconstitution of the design basis for the electrical loading
of safety related MCCs,
the licensee discovered a potential overload
condition could exist under certain postulated accident conditions.
For example, should a LOCA occur while the plant is in hot shutdown
with off-site power available and there is a single failure of either
MCC-5 or MCC-6, the remaining MCC feeder breaker would have tripped
on overload. The licensee implemented compensatory measures prior to
confirming that the overload condition existed.
Once the overload
was confirmed, the proper reports were made to NRC.
2
The licensee determined that, prior to the compensatory measures, the
plant was operated with an unanalyzed condition. That is, the loss
of all 480V safety related MCC power:
MCCs-5,6,9 and 10,
and,
Instrument Buses 1,4,6 and 9. Most of the safety related loads are
motor operated valves that would fail as is on the loss of power. In
the above scenario, when the single failure occurs, redundant loads
are started on the other MCC resulting in the overcurrent condition.
This would result in the feeder breaker tripping shortly after the
transfer of loads, almost instantly.
Since the loss of both MCCs
occur very quickly, the motor operated valves would be de-energized
in their normal operating position. The result of this would be no
flow path available for Containment Spray, RHR, or Safety Injection
(except for the Accumulators). The pumps for these systems would be
running at shut off head without any area fan coolers operating.
Other equipment that would be lost includes:
the Iodine Removal
Units (HVE-3
and 4); steam supplies to the Steam Driven Feedwater
Pump (V1-8A and 8B); the Charcoal Filter Booster Fans (HVE-5A
and
5B); both Service Water Booster Pumps; AFW discharge valves; and the
Pressurizer Relief Isolation Valves.
On August 11,
1974, the licensee experienced an event which resulted
in the loss of MCC-5 and MCC-6,
possibly at the same time, as the
result of an overcurrent trip on the feeder breaker to MCC-6.
The
event was reported to the NRC via a letter dated January 3, 1975.
The letter detailed the accounts of the event. It took the operators
approximately two minutes to identify that MCC-6 was de-energized.
After an additional thirteen minutes, the operators discovered that
MCC-5 was de-energized. After re-energizing MCC-5, MCC-6 was found
de-energized approximately one minute later.
The licensee stated
that there was approximately a two minute period where both MCCs
possibly could have been de-energized at the same time, but it was
not confirmed. That time period was the first two minutes of the
accident.
The letter does not discuss whether the ECCS systems
functioned properly during the event. It does state that there was
no damage to systems or components,
no radiation exposures,
nor
radioactive releases.
As a result of the 1974 event, the licensee increased the overcurrent
trip setpoint from 600 amps to 720 amps.
Sometime later, the set
point was increased to 800 amps. In both cases, there was no docu
mentation to indicate that the setpoint was adequate and acceptable
since it was not controlled by any program at those times.
As can be seen from the list of equipment that was previously
discussed, in the event that a LOCA occurs with offsite power avail
able, concurrent with the single failure of MCC-5 or 6, and the loss
of the other MCC on overcurrent, a majority of the ECCS subsystems
would not be available. This event was not analyzed and could result
in safety systems designed to prevent or mitigate a serious accident not
being able to perform their intended functions under certain
conditions.
3
The compensatory measures discussed in NRC Report No. 50-261/88-30
were initiated in a timely manner by the licensee and have been
evaluated as acceptable for short term operation.
The licensee has
subsequently initiated plant modifications to transfer loads from
MCC-5 and 6 to other BOP MCCs. The loads that are to be removed are
the Reactor Support Cooling Fans (HVE-6A,B), the Concrete Shield Wall
Cooling Fans
(HVH-9A,B),
the turbine auxiliaries, and the duct
heaters for HVE-15A. These activities are scheduled to be completed
during the present outage.
If the duct heaters are not completed
prior to restart, they will be done at a later date since they are
not required during plant operation,
only during refueling
activities.
Based on the above, URI 50-261/88-30-01 will be closed and Violation
50-261/88-36-01, Unanalyzed Loss of Safety Related MCCs, is opened.
b.
On October 27, 1988, the licensee determined that the penetration to
pigtail splice configuration for the 480 volt power leads to the
Reactor Containment Fan Coolers
(HVH
1-4) were neither environ
mentally qualified nor qualifiable.
The licensee subsequently
declared the HVH 1-4 units inoperable and commenced reactor shutdown.
Hot shutdown was obtained at 2:03 a.m.,
on October 28,
1988.
The
unqualified cable splice configurations (Rayclad heat shrink tubing)
were replaced with qualified Raychem heat shrink and the plant
returned to power operation on October 29,
1988.
This issue was
initially investigated by the NRC Resident Inspector and documented
as Unresolved Item 88-30-02.
The main concern left outstanding
involved the root cause determination or reason why these particular
splice configurations were not included in the licensee's original
scope of corrective action to resolve item 50-261/87-10-02, Crouse-Hinds
Electrical Penetrations.
The inspector concentrated on trying to
determine the root cause for why these splices were not corrected and
to evaluate the safety significance of the splices being non-EQ.
Inspection Report 87-10 is the report which initially identified the
deficiency that the Heat Shrink Sleeve at the penetration to pigtail
splice on Crouse-Hinds electrical penetrations
(EPAs)
were not
environmentally qualified. The example cited in the report involved
generic qualification file 9.0 for Crouse-Hinds EPAs, Model 751.
The licensee attempted to correct all unqualified EPA splice
configurations by implementing Special Procedure SP-775,
Penetra
tion/Pigtail Splice Repair (May 1987). SP-775 typically replaced the
penetration-to-pigtail splices with individual
Raychem over
uninsulated butt splice connectors and the pigtail-to-field cables
splices with insulated butt splice connectors covered with an overall
Raychem sleeve.
This splice combination was later found to be
unqualified allowing moisture intrusion from the open ended cable
jacket at the penetration-to-pigtail splice, through the pigtail up
to the insulated butt splices, and consequently, the possibility for
4
a line-to-line leakage path was created. Modification M-928, Repair
of EQ Splices Installed under SP-775 was issued to address this
latter potential moisture intrusion problem.
In June 1988,
the licensee began a review of internal and NRC
generated EQ violations to assure that all necessary corrective
actions had been completed. On October 27, 1988, a review of records
for completed
EQ special procedures and modification packages,
revealed that EQ documentation did not exist to show that the HVH
splices had ever been modified to a qualified configuration or were
qualified by the penetration manufacturer, Crouse-Hinds.
After making this determination, the licensee made a containment
entry, at or near full power, to inspect the suspect splices.
The
results of the inspection confirmed that the splice configurations
were the original installations with Rayclad heat shrink tubing.
Engineering review and subsequent review by the PNSC determined that
the splices were not EQ qualified or qualifiable.
Recently, the licensee learned that qualification testing had been
performed on the Crouse-Hinds Penetration Splice configuration for
another utility. The licensee developed draft Engineering Evaluation
ENG 88-186 (December 16,
1988) to provide justification that the HVH
1-4 penetration splices were environmentally qualified, prior to the
repair. The NRC inspectors had serious concerns regarding the use of
this test report alone as establishing the basis for qualification of
the H. B. Robinson penetration splices. Concerns which the licensee
could not adequately address involved aging of test specimens versus
naturally aged splices at Robinson, similarity between test specimens
versus Robinson's as-found configurations (sleeves separating from
substrate), utilizing post testing aging to satisfy pre-test aging,
and applied test voltages and currents on specimens did not envelope
plant conditions.
All the above concerns, in particular, the
deteriorated condition observed for the as-found splice configura
tions at Robinson posed serious questions regarding the applicability
of this test report.
The licensee agreed that it might not be
possible to show qualification using the subject report. Therefore,
since the licensee did not have documentation in a file to establish
qualification for the subject penetration splice configurations these
items were considered unqualified and not qualifiable.
The HVH units provide containment pressure reduction and residual
heat removal from containment following a design basis accident. The
unqualified penetration splices on the
HVH units made them
technically inoperable, creating an unanalyzed condition that would
seriously degrade the plants capability for mitigating the
consequences of an accident.
The accident analysis described in the
FSAR considered two of four fan coolers being operable with one
containment spray pump.
5
Plant Technical Specification 3.3.2.1.c requires all four fan cooler
units to be operable whenever the reactor is made critical.
Tech
Spec 3.3.2.2.1 allows one fan cooler unit to become inoperable
for up to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, provided both containment spray pumps are demon
strated operable.
The failure of the licensee to establish environmental qualification
for the penetration splices for the HVH 1-4 Containment Fan Coolers
is an apparent violation of 10 CFR 50.49(f) and (k).
This lead to an
apparent violation of the above Technical Specifications because the
licensee operated the plant at various power levels and at hot
shutdown with all four fan coolers being inoperable due to a lack of
EQ documentation for the splices.
The root cause of the violation
appears to be the result of the licensee's failure to replace the
penetration to pigtail splices as part of the previous corrective
action for item 261/87-10-02.
This also is considered an apparent
violation of 10 CFR 50, Appendix B, Criterion XVI for failure to take
adequate corrective action for a known condition adverse to quality.
These concerns will be identified as Violation 50-261/88-36-02.
The
unresolved item (88-30-02) is now considered closed.
C. On September 14,
1988, while the plant was in cold shutdown for
maintenance, the licensee discovered that the Reactor Vessel Head
Vent System Solenoid operated valves were not properly installed to
meet EQ requirements and were not included on the Robinson Master
List of EQ equipment.
The licensee discovered this problem while
investigating some potentially generic EQ issues involving Target
Rock Solenoid valves that had been initially identified at the Harris
plant in August of 1988.
The extent of the deficiencies involved:
(1) failure to install
qualified penetration to pigtail splice configurations as discussed
in item b above; (2) failure to install conduit entrance seals at the
solenoid valve housing; and (3) failure to include the valves on the
Robinson EQ Master List.
The licensee took necessary corrective action by reworking the
solenoid valves to meet EQ requirements, installing cable entrance
seals, and reworking penetration splices. The inspector examined the
as-built configuration of the solenoid valves and confirmed that
cable entrance seals were installed.
Review of completed work
records confirmed that the solenoid valves were repaired and penetra
tion splices were replaced with a qualified configuration.
The Licensee Event Report (LER 88-020-00) indicated that the original
installation was intended to be fully EQ qualified.
The cable and
valves were purchased as EQ qualified.
However, at the time of
installation, the problems with containment penetration splices and
conduit entrance seals were not understood by the licensee to be
6
required to qualify the Target.Rock SOVs. Additionally, the licensee
indicated that the failure to include the NUREG 0737 Reactor Vessel
Heat Vent System in the EQ Program was an oversight since the Head
Vent System was installed prior to the compliance date for 10 CFR 50.49.
10 CFR 50.44(c)(3)(iii) requires the installation of high point vents
for the reactor coolant system, and the reactor vessel head to vent
non-condensable gases from the RCS which may inhibit core cooling
during natural circulation.
The licensee indicated that although
these valves may have been inoperable during a design basis event,
other methods would be available to remove voids from the reactor
vessel head area.
However, the failure to install the head vent
valves in a qualified configuration and the failure to include the
solenoid valves in the EQ Program is an apparent violation of both
10 CFR 50.44(c)(3)(iii) and 10 CFR 50.49(d), (f), (j) and (k).
This item will be identified as violation 50-261/88-36-03.
4. Exit Interview
The inspection scope and results were summarized on December 22, 1988,
with those persons indicated in paragraph 1. The inspectors described the
areas inspected and discussed in detail the inspection results listed
below. Although reviewed during this inspection, proprietary information
is not contained in this report.
Furthermore, dissenting comments were
not received from the licensee.
Violation 50-261/88-36-01, Unanalyzed Loss of Safety-Related MCCs 5 and 6,
paragraph 3.a.
Violation 50-261/88-36-02,
HVH 1-4 Penetration Splices Being Non-EQ,
paragraph 3.b.
Violation 50-261/88-36-03,
Non-EQ Reactor Vessel
Head Vent System,
paragraph 3.c.
5. Acronyms and Initialisms
BOP - Balance of Plant
CFR - Code of Federal Regulations
ECCS - Emergency Core Cooling System
EPA - Electrical Penetration Assembly
EQ - Environmental Qualification
FSAR -
Final Safety Analysis Report
HVH -
Reactor Containment Fan Cooler
LER - Licensee Event Report
MCC - Motor Control Center
PNSC - Plant Nuclear Safety Committee
7
QA - Quality Assurance
QC - Quality Control
SP - Special Procedures
URI - Unresolved Item