IR 05000277/1983023
| ML19303B905 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 08/22/1983 |
| From: | Blough A, Lester Tripp, Williams J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML19303B906 | List: |
| References | |
| 50-277-83-23-EC, 50-278-83-23, 50-278-83-23-EC, NUDOCS 8310210278 | |
| Download: ML19303B905 (4) | |
Text
.
.-
.
.
.
'
U. S. NUCLEAR REGULATORY COMMISSION
,
Region I 50-277/83-23 Report No. 50-778/99 99 50-277 Docket No. 50-278
-
-
License No. DPR-56 Priority __
Category c
Licensee:
Philadelphia Electric Company 2301 Market Street
.
Phila'delphia. Pennsylvania 19101 Facility Name: Peach Bottom Atomic Power Station Meeting at: USNRC, Region I, King of Prussie, Pennsylvania Meeting conducted:
August 11, 1983
..
K. bipdgn, dr. ne510ent _
8/22ff3 NRC Personnel:
.(
Adr)
..
Anspector date s~igned
'
!Ldho 8hs/0
. H. WiMiams, Resident Inspector da'te 'si gned date signed
/
. b. JN 7[hf(3 Approved by:
,
Re'ack*or P ec s ection 3A Enforcement Conference on August 11,1983 (Report 50-277/83-23;50-278/83-23)
Meetino Sumary:
Special enforcement conference convened to discuss findings of Region I Inspections 50-277/83-16 and 50-278/83-16, relative to unsealed penetrations in various fire barriers.
Senior Philadelphia Electric Company and NRC Region I management personnel attended this one-hour meeting at the Region I office.
..
-
Region I Form 12-1' $ 31 f Edkh
\\j d,hp,
(Rev. August 77)
.
- -
.
.
.
%
DETAILS
,
1.
Attendees Philadelphia Electric Company
.
J. S. Kemper, Vice President Engineering and Research J. W. Gallagher, Manager, Electric Production Department M. J. Cooney, Superintendent, Nuclear Generation Division R. S. Fleischmann, Station Superintendent, Peach Bottom Atomic Power Station S. A. Spitko, Site QA Engineer T. P. Gotzis, Assistant General Superintendent, Construction Division C. J. Gerdes, PECO Engineer G. M. Morley, Supervisory Engineer, Mechanical Engineering Division G. J. Reid, PECO Engineer U.S. Nuclear Regulatory Commission R. W. Starostecki, Director, DPRP R. R. Keimig, Chief Projects Branch 3 D. J. Holody, Enforcement Coordinator, Region I A. R. Blough, Senior Resident Inspector J. H. Williams, Resident Inspector 6. M. Gutierrez, Regional Counsel, Region I 2.
Meeting Purpose To discuss fire protection program events associated with sealing and inspecting penetrations in fire barrier walls, to emphasize their importance, and to obtain licensee views on causes and corrective actions.
,
3.
Events of Concern The NRC identified the following violation as being of concern.
l 3.1 Failure to have functional fire barriers and to perform adequate
inspections (see NRC:RI Report 277/83-16,278/83-16).
'4.
Cause and Safety Significance The NRC stated that the violation appeared to be caused by inadequate construction. QA and incomplete inspections of the fire barriers. The
'
NRC was concerned that the fire barrier penetration discrepancies 'h'ad~
not been identified earlier.
5.
Licensee Discussions S.1 Causes Philadelphia Electric Company management acknowledged the construction and procedural defects associated with the fire barriers.
It was noted that the fire barriers were constructed as ordinary " block" l
.
.. -
.
,,
walls with no special requirements for inspections and no detailed documentation of inspection required. This was stated to be normal construction practice at the time.
Inspections in 1977 by PECO and a contractor during development of the Fire Protection Program Report (March 1977) did not uncover the problem because the defects were difficult to observe.
Philadelphia Electric Company review indicated that the problems stemmed from initial construction rather than from
. subsequent modification activities.
The licensee stated that there have been three 18-month inspections with Surveillance Test ST 16.17.
He indicated that the procedural problems stemmed from (1) incomplete definition of what constitutes a fire barrier and (2) failure to specify a "close-up" inspection.
Photographs were shown to illustrate the need for a very close range inspection to identify the fire barrier deficiencies that existed.
5.2 Corrective Actions 5.2.1 Immediate Corrective Actions.
Firewatches were posted as needed.
The events were reported to NRC.
A detailed inspec-tion of wall surfaces was initiated and is about 95 percent complete. The surveillance test was revised to. provide much more detail on how to conduct inspections and on acceptance criteria.
Corporate engineering personnel contributed to the technical upgrading of the procedure.
'
5.2.2 Long-term Corrective Actions. The detailed fire barrier inspection now in progress will be completed.
Firewatches will remain as required until penetrations are sealed.
Seals meeting 10 CFR 50 Appendix R criteria will be installed by March 1, 1984.
The modification process, although not involved in this event, is being reviewed to determine if the process needs upgrading in-light of 10 CFR 50 Appendix R.
As part of the fire barrier upgrade.to meet Appendix R requirements, a computer program is being used to record data on all seals. After seals are up-
.
graded, the final data will be made available to site personnel for evaluation of whether further inspection procedure improve-ments are needed.
6.
Conclusion
,
The licensee also presented infomation concerning the safety signfficance of the missing penetration fire barriers.
As a result of licensee /NRC discussions, it was concluded that the degradation of fire wall barrier integrity was limited because of the following considerations. The I
majority of the missing fire barriers were small and located high above
.
I
-,. -
-
-
.
-
.
.,
.-
.
-
.
,
khe floor (at the wall to ceiling juncture), thus, they were unlikely to contribute to spreading of the types of fires that might occur in the
>
affected rooms.
In those rooms with Cardox fire suppression systems, preoperational tests had demonstrated the abili'ty to maintain adequate
,
carbon dioxide concentrations with the missing barriers.
Each affected room had fire (heat and/or smoke) detectors.
Thus, in the event of a fireinoneoftheaffectedrooms,theeffectofthemissingbarrier(s)
would be to allow the increased spreading of smoke and combustion pro-ducts with only a small increase in the possibility of the fire spread-ing to other areas.
,
The NRC thanked the licensee for his input and indicated that NRC review would be facilitated by the licensee-provided information.
I i
.
j i
!
'
.-
'
-
.
r.-
l c
.
,
BRIEFIkG PAPER
,
' Subject:
ENFORCEMENT CONFERENCE WITH PECO, AUGUST 11, 1983
)
'1.
Background As part of fire' protection program reviews, NRC issued a Technical Specifi-i cation Amendment in February 1978 which requires fire barrier penetrations
-
protecting the following areas to be functional:
Cable Spreading Room, Emergency f uitchgear Rooms, Diesel Generator Rooms, Battery Rooms, and the Control! Room.
In event of inoperable barriers a fire watch must be estab-lished within one hour.
Visual inspection of all fire barriers is required
.
on'an eighteen month interval.
Original penetration seal designs do not
!
meet.all 10 CFR 50 Appendix R requirements. The licensee is committed to i
upgrading penetrLtion seals by March 1,1984.
At that time additional
.
Technical : Specification Amendments will become effective and require oper-
.able fire barriers for.all areas protecting safe shutdown capabilities.
-
The. licensee has a contractor force of about 90 persons working on the
~
-barrier upgrade.
,
,
2.
Event Overview On May 11, 1983; con % ctor persornel involved.in the fire barrier upcrade s
,
identified an unsealed penetration above a ventilation duct through an
-
-
'Er:ergency Switchgear Room wall.
Subsequent detailed licensee inspections have identified additional unsealed or improperly sealed penetrations in other Emergency Switchgear Rooms, Battery Rooms, the Control Room, the Cable Spreading and Computer Rooms,;and Diesel Generator Rooms. These
findings were reported to the NRC.
Fire watches were posted as required-eleven fire watches _are still in place.
For additional detail see Enclosure
(1), excerpt of inspection report 277/83-16, 278/83-16.
'3.
Cause
{
3.1-Inadequate ionstruction 0A of fire barriers.
These barriers were not
envisioned as a Technical Specification item, but were required by general construction specifications to be sealed.
3.2 Inadeouate fire barrier inspections, possibly from the following:
!
poor human. factor considerations in the test, use of inexperienced inspectors, inadequate supervision and followup of inspections.
The licensee was asked in the cover letter of the inspection report to provide additional insights at the meeting.
-
4.
Safety Significance
,
i A fire could more quickly spread to adjacent areas if fire barriers are irioperable. ' All areas involved contain safety-related equipment.
.
i r
'
-
,
-
.
,
-
- ^
..
,.
.
.
..
'
-
'
- T e selection of the areas as Technical Specification barriers in 1978 was not. based on a safe shutdown analysis.
The licensee's safe shutdown analysis is still in progress.
.
.
Also, degraded fire barriers could result in accelerated dilution of carbon
.
-dioxide.after initiation in those rooms so protected (Control Room, Cable Spreading.and Computer Rooms, Diesel Generator Rooms).
Because the degraded barriers in those rooms were small and were located high above the floor,
'
this effect would probably be small.
In a previous inspection the resident inspector reviewed Diesel Generator Cardox System preoperational test results.
Each room maintained adequate carbon dioxide concentrations with a signifi-cant margin, despite the apparent presence of the degraded barriers.
The affected rooms were provided with fire detection and suppression features
,
as shown below.
Fire Detectors Cardox Fire Suppression
Prior to 1/82 After 1/82
,
Control Room 4 (Smoke)
10 (Smoke)
Manual Cable Spreading s
and Computer Rooms 6 (Smoke)
27 (Smoke)
Manual (before 1/82)
Automatic (since 1/82)
,
Diesel Generator
,
Rooms 4 per room (Heat) No change Automatic Emergency Switchgear 1 per room.
no change
Rooms-8 total (Smoke)
Battery Rooms
--
2 per room 8 total (Smoke)
,
5.
Licensee Corrective Autions
.
Licensee-corrective actions thusfar have' included posting fire watches,
'
temporarily sealing those penetrations for which a temporary seal was feasible, starting to design and install permanent seals, and conducting
a thorough inspection and review of all Technical Specification fire
i barriers. As 'noted above, this has resulted in identification of additional degraded barriers.
This inspection effort has been on-going, involving several man-months thusfar, and is nearly complete.
The licensee is
apparently formulating additional long-term corrective actions. The licen-
see was asked in the cover letter of the inspection report to describe his corrective actions at the enforcement conference.
%
,
...
-
',,.
'
.
.
6.
R sident Inspector Perceptions As a result of previous events, SALPs, and enforcement actions, the licensee has made broad efforts to improve worker and supervisor performance, adher-ence to procedures, and attention to. detail. Most of these efforts occurred since the last fire barrier surveillance.
Some (such as meetings of station personnel with the Station Superintendent, special training sessions in
,
Administrative Procedures and Technical Specifications, and a Vice Presiden-
'
tial letter to all-supervisors) have direct applicability to this event;
'
others, such as (1) Quality Control Reorganization, (2) problem analysis training for supervisors (3) formation of an independent safety engineering group, (4) establishment of a Fire Protection Assistant on the site staff, and (5) use of a." Quality Team" to study the problem of failure to follow
,
procedures, have potential for long term beneficial impact.
Based on licensee managsment's commitment to improve personnel performance
and supervision, as evidenced in his recent responses to SALP (July 17,1983)
and EA-46 (July 12, 1983), the resident inspector feels that a' civil penalty
- in this case would be primarily punitive in nature.
>
!
i u
,
!
,
i
I i
I
)
,=
s
g Y
-
-
.
,
e ENCLOSURE (1)
(Exi:erpts from Inspection Report 50-277/83-16; 50-278/83-16)
9.2 On-site Followuo For LERs selected for onsite review (denoted by asterisks above), the inspector verified that appropriate corrective action was taken or responsibility assigned and that continued operation of the facility i
was conducted in accordance with Technical Specifications and did not constitute an unreviewed safety question as defined in 10 CFR 50.59.
Report accuracy, compliance with current reporting requirements and
.
applicability to other site systems and components were also reviewed.
l 9.2.1 LER 2-83-12/IP and IT 9.2.1.1 Background.
As part of fire protection program
-
reviews, NRC issued Technical Specification Amendment
,
- 39 on February 28, 1978 which requires fire barrier
,-
penetrations protecting the following areas to be i
functional:
Cable Spreading Room, Emergency Switch-i gear Rooms, Diesel Generator Rooms, Battery Rooms,
!
and the Control Room.
In event of inoperable barriers a fire watch must be established within one hour.
Visual inspection of all fire barriers is required on an eighteen month interval.
Original penetration seal designs do not meet all 10 CFR 50 App;ndix R h
requirements.
The licensee is committed to upgrading penetration seals by March 1, 1984.
At that time additional Technical Specification Amendments will
,
become effective and require operable fire barriers for all areas protecting safe shutdown capabilities.
9.2.1.2 Event Description.
On May 11, 1983, fire barrier penetration inspections during upgrading of penetra-tion seals to 10 CFR 50 Appendix R requirements.
identified an unsealed penetration above a ventilation duct running through an Emergency Switchgear Room wall.
The licensee posted a. fire watch and began additional inspections.
Two additional holes above ventilation ducts, as well as a lack of sealant in small openings between several walls and ceilings, were found in other Emergency Switchgear Rooms and in Battery Rooms.
Additional fire watches were posted and a more detailed inspection initiated.
The inspections are completed for switchgear and battery rooms.
Several additional unsealed small-diameter penetrations, as well as other fire barriers not meeting original design specifica-tions, were found.
For other areas, the inspections are ongoing.
On June 28, four small unsealed penetra-tions were found in the Control Room wall, above the dropped ceiling level, leading to the Turbine Building.
A fire watch was posted until the penetrations were tem-porarily sealed per licensee procedures.
.-
_ _. _ _.. _.. _
. _...
,
\\7 5
.,,, _ _ _
. ~,
..
!
...
q
-
9.2.1.3 Licensee Corrective Actions.
In addition to posting i
fire watches and conducting detailed inspections as
.
'
mentioned above, the licensee is designing permanent seals, qualified to 10 CFR 50 Appendix R requirements, l
for the penetrations.
Sealing of'some penetrations
-
began during this inspection period.
The licensee stated that additional corrective actions were being formulated.
For example, he was considering revising y
,
the inspection procedure to make it more explicit and also dividing the test into smaller portions.
As of the end of the inspection period, the licensee's cor-
.
'
rective action plans were not finalized.
9.2.1.4 Inspector Observations.
The inspector frequently verified that fire watches were posted as required.
i The inspector also reviewed fire barrier inspection records and interviewed licensee engineers.
Surveil-l
'
lance Test ST 16.7, Visual Inspection of Fire Barriers, i
had been fully performed on three occasions since the Technical Specification requirement became effective:
-- August 24, 1979, using surveillance test Revision 0
.
of hugust 9, 1979;
.w
.
,
'
-- June 4, 1981, using surveillance test Revision 1 of
.
May 30, 1980; and
-- September 28, 1982, using surveillance test Revision
~1'of May 30, 1980.
The test was performed in one case by a Fire Protection
Engineer, and in the two other cases by Junior Technical
Assistants (OTAs).
In some cases, minor deficiencies were identified during the surveillance.
These were re-paired and re-inspected. The site QA Engineer stated that, in the cases where JTAs did the inspections, their training included a detailed briefing on the procedure
.and how to perform it, plus in-plant training on how to inspect the various types of seals.
Although the pro-cedure allows temporary sealing of degraded seals by
>
the inspec+or. the JTAs were specifically instructed
)
not to attempt repairs and were not provide _d with seal-ing materials.
The inspections generally took 60 to 80
man-hours to complete.
None of the individuals who did the inspections were still working for the licensee; teerefore, they were unavailable for interview by either the licensee or the NRC inspector.
The individual who e
performed the more 'recent, detailed inspections stated that it had taken about 160 man-hours (two persons for
,
two weeks) to inspect the Switchgear and Battery Rooms.
!
!
l J.
,
.
b
~..a o.
.,
-
,
The inspector observed the various fire barrier defi-
. ciencies in-plant.
All the deficiencies are over ten
.
feet above floor level.
In most cases, the penetra-
'
tions must be observed at close distance to note de-ficiencies.
This requires use of a ladder, and often climbing on top of ducting or switchgear cabinets.
However, once observed at close range, many of the i
deficiencies are obvious.
The licensee does not know l
how the previous inspections were conducted, but believes they were probably done from floor level.
The procedure is specific with respect to what con-stitutes an acceptable seal but is not specific on
"
how to inspect.
None of the unsealed penetrations observed by the inspector appeared to have been previously sealed, i.e.,
,
there were no visible remnants of sealing materials.
.
The licensee stated that none of the unsealed openings appeared, by his inspection to have been previously
,
sealed.
The largest penetrations, those adjacent to l
ventilation ducts had been present since initial con-struction.
The licensee was investigating whether j
some of the other penetrations might have resulted i
from later modifications, but had not identified any
-
'^
such cases.
j-
'
The licensee reviewed the original construction re-
'
quirements for fire barrier penetrations.
Drawing
'
E-1315 discusses how to seal typical electrical pene-
,
trations.
The inspector reviewed this drawing.
Ven-
~
tilation system installation specification, M-307,
,
references NFPA Code 90A for sealing of vent duct penetrations.
The licensee has not been able to deter-mine why the construction specifications were not met.
l
+
The inspector reviewed automatic fire detection capa-bilities for the areas with degraded fire barriers.
i, All switchgear rooms and the control room have had early warning smoke detectors since construction.
Two smoke detectors were installed in each battery room in January 1982.
Each Battery Room is adjacent to a Switchgear Room.
Smoke detectors are required by Technical Specifications to be operable.
9.2.1.5 Conclusions.
The inspector agreed with the LER state-ment that the event was caused by errors during the construction phase combined with inadequacies in the fire barrier inspection program.
As stated above, Technical Specification Limiting Con-dition for Operation 3.14D requires fire barrier penetrations to be functional in various areas, inclu-dino the Control Room, Emergency Switchgear Rooms, and
.
!
,.
.
-
. - -
\\,
... -.
.. _..
...
..
.-
,
..
-
,
Surveillante Requirement 4.14D requires s
Battery Rooms.
visual inspection of fire barrier penetrations once per Failure to have functional fire i '
eighteen months.
barriers and to perfcrm adequate inspections is a
,
Violation (277/83-16-03, 278/83-16-02).
l.-
,
,
.
'
e I
+
P i. -
~
.
t
,
-
- '.
t YA
-
=
-
-
_
_
--