ML20035H335
| ML20035H335 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 04/27/1993 |
| From: | Fillion P, Shymlock M NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20035H331 | List: |
| References | |
| 50-302-93-10, NUDOCS 9305040244 | |
| Download: ML20035H335 (7) | |
See also: IR 05000302/1993010
Text
I
.
"I
.[(g" 4
g
UNITFD STATES
-
fg
NUCLEAR REGULATORY COMMISslON
.)
g
n
REGION Il
,j
101 MARIETTA STREET, N.W.
5
-y
,
l
f
ATLANT A. GEORGI A 30323
g
/
.....
I
Report No.: 50-302/93-10
Licensee:
Florida Power Corporation
3201 -34th Street, South
St. Petersburg, FL 33733
i
Docket No.: 50-302
License No.: DPR-72
i
Facility Name: Crystal River 3
-
Inspection Conducted: March 30 - April 2, 1993
4[BM93
Inspector:
/
PaulJ.Fiyion
Date' Signed
Approved by:
7
fffh2[$7
YW 7 -93
Milton B. Shymfock, Chief
Date Signed
Plant Systems Section
Engineering Branch
Division of Reactor Safety
SUMMARY
Scope:
This routine, announced inspection was conducted in the areas of engineering and
technical support and management response to operational events. The analysis
of and corrective actions for operational events was evaluated by the inspector.
Results:
In the areas inspected, violations or deviations were not identified.
In the
aftermath of a major storm on March 13, 1993, the stability of the offsite power
supply became an issue. The licensee's decisions and actions in relation to this
situation were reasonable.
A loss-of-power event occurred on March 29, 1993,
during an outage. Analysis of cause and corrective actions were adequate.
930504o244 930427
ADOCK 05000302
0
.
!
'
.
i
REPORT DETAILS
-
,
1.
Persons Contacted
Licensee Employees
N. Barbeito, Senior Standards Engineer, Substation Standards
D. Carrico, Supervisor, Systems Protection, Crystal River Complex
- M. Fitzgerald, Supervisor, Systems Engineering
- E. Froats, Manager Nuclear Compliance
- F. Fusick, Manager, Design and Modifications
- B. Hickle, Director, Nuclear Plant Operations
W. Morgan, Field Supervisor, Substation Maintenance
S. Puckett, Supervisor, Central Division Area Substation
R. Sweeney, Shift Technical Advisor
- R. Widell, Director, Nuclear Operations Site Support
- R. Wiemann, Senior Nuclear Electrical Engineer, Systems Engineering
- K. Wilson, Manager, Nuclear Licensing
Licensee employees contacted during this inspection included engineers
and technicians.
NRC Personnel
- R. Freudenberger, Resident Inspector
- P. Holmes-Ray, Senior Resident Inspector
- Attended exit interview
2.
Inspection Details
a.
Background Information
On March 13, 1993, Crystal River 3 was in their mid-cycle
e
maintenance outage. A major storm system lasting two days brought
high winds and high tides to the area surrounding the plant.
Consequences of these climatic conditions at the site included a
>
buildup of salt on the switchyard insulators and flooding of the
!
switchyard for one tide cycle. The buildup of salt on the
insulators was due to the fact that on the second day of the storm
there was high onshore winds carrying salt laden sea spray, but
there was no rain to wash off the insulators.
Some insulators in
the switchyard were damaged by wind induced vibrations or
missiles. Also as a result of the storm the entire transmission
system was degraded. About 1000 miles of transmission line were
deenergized. The buildup of salt on the insulators meant that
,
insulators would have to be washed. That work was done on March
r
17 for the 230 kV switchyard and on March 25, and 26, for the 500
kV switchyard.
i
-
.
-
.
.
_
.
.
2
Then on March 29, automatic tripping of all 500 kV circuit
breakers in the ring bus (refer to diagram) resulted in loss of
power to plant auxiliaries because the source of power at that
time was the 500 kV bus via the generator stepup transformer as is
normal for outage time.
Post-event analysis showed that the
circuit breaker tripping was caused by a short-circuit in a cable
trench.
Given this background, the inspection plan was to evaluate the
licensee's decision making process with respect to the timing of
the insulator washdown. The inspector would evaluate the
licensea's efforts in determining the root cause of the loss-of-
power.
Particularly NRC wished to determine whether or not the
March 29, event was caused by storm related damage.
Operational consequences of these power outages will be addressed
in the resident inspectors report covering the time period.
Preliminary, analysis indicates that all safety-related systems
performed as designed.
b.
High-Voltage Insulator Washdown
On the day of the storm, Saturday, March 13, flooding of the
switchyard had subsided by late day which allowed insulator
>
replacement work to commence. On Sunday, insulator replacement
work continued. Bushings on 230 kV transformers for Crystal River
3 were cleaned as part of normal outage scheduled maintenance. On
1
Monday, no major work was taking place. The Field Supervisor for
the switchyard observed some partial discharge on the surface of
the vertical bus bar insulators in the 230 kV yard. Partial
discharge is an electric discharge which only partially bridges
,
the insulation between conductors.
Partial discharges occur when
the local electric field intensity exceeds the dielectric strength
of the dielectric involved, resulting in localized ionization and
breakdown. He was able to observe this because partial
discharges are often accompanied by emission of light and sound.
The partial discharges were due to the fact that the high winds of
Sunday had deposited salt on the insulator surfaces. The 500 kV
yard had little or no visible partial discharge taking place
because the bus bars were overinsulated to 700 kV.
,
So the situation was that the 500 kV was relatively stable and it
was not in imminent danger of flashover. The plant was being
powered from the 500 kV. The 230 kV had visible partial discharge
,
indicating that insulator washing should be scheduled. The
partial discharges were not at a level that called for immediate
action.
The situation on Tuesday was basically the same as Monday.
9
At about 4:00 am, on Wednesday, a fine mist or drizzle of rain
,
commenced.
The combination of salt and moisture created a
,
.
-
~
.
.
3
condition of lowered effective insulated leakage distance along
the surface of the insulator which allowed an excessive leakage
current to flow which resulted in insulator flashovers in the 230
kV yard. Some circuit breakers went through their reclosing
cycle, which is a rapid opening and closing.
The 230 kV yard was deenergized by operator action on the
afternoon of Wednesday, March 17, and the insulators were washed.
t
It is feasible to wash insulators while energized, but the
deenergized method was chosen. The washing evolution took about
six hours. Once the washing was completed, the 230 kV bus was
reenergized and no further flashover or partial discharges were
experienced.
The insulators in the 500 kV yard were washed on March 25, and 26.
I
This was strictly precautionary because the 500 kV buses never
>
experienced severe partial discharge problems.
-
With respect to the preceding discussion and the main thrust of
the inspection, the relevant requirement is to provide reliable
-
sources of offsite power and to maintain due regard for shutdown
!
risk. The inspector reviewed the evolutions that took place in the
aftermath of the storm of March 13 in light of these
considerations, and concluded that the licensee's decisions and
actions were reasonable.
,
c.
Loss-of-Power Event of March 29
,
At approximately 1:50 am, on March 29, 1993, the plant experienced
a loss-of-offsite-power during a mid-cycle maintenance outage
while backfeeding through the generator stepup transformer. The
licensee formulated a theory on the sequence of events leading to
'
the loss-of-power.
First, a short-circuit occurred on either a
250 VDC cable or a 480 V cable. The short-circuit occurred in a
cable trench in the 500 kV switchyard. The 250 VDC and 480 V
i
cables were lying adjacent to each other.
The 250 VDC cable was a
nine conductor cable running between a relay rack in the
switchyard relay house and 500 kV motor operated switch MOS1873
(refer to diagram for location of this switch in the system.) The
480 V cable was a one conductor cable which was one phase of a
three phase circuit running between a motor control center in the
plant and a 230 kV transformer. Each conductor of a three phase
480 V circuit is 277 V above ground potential.
Second, regardless
of which cable shorted first eventually both cables were damaged,
'
conductors of each came in contact and as a result 277 Y was
,
superimposed on the switchyard DC System. This caused spurious
operation of auxiliary relays which caused tripping of all 500 kV
'
circuit breakers in the ring bus.
The inspector reviewed all available information, and concluded that the
i
licensee's hypothesis on the cause of the loss-of-pcwer event was valid
and the most probable cause that could be deduced from the evidence.
r
,
?
?
.
'
.
.
4
The following specific inspection activities were performed in this
effort.
l
(a)
Reviewed the substation sequence of events recorder
'
printout.
i
(b)
Reviewed the plant computer alarm printout.
'
(c)
Reviewed the control diagram for.MOS1873. This was relevant
j
because the switch partially opened and the sequence of
i
events recorder and plant computer show the switch rapidly
cycling. The recorder printout was actually a false signal
!
caused by the short-circuit, but it does indicate the time
i
that the fault occurred.
!
f
(d)
Examined the faulted cables.
(e)
Inspected the cable trenches.
(f)
Held discussions with the substation engineers.
,
The root cause of the cable fault has not as yet been determined with
any certainty. The damaged cables were sent to a cable manufacturer's
.
laboratory for root cause analysis. A report from the laboratory is
i
expected to be forwarded within two weeks of this inspection. The
i
preliminary conclusion is that the short-circuit is not directly related
to the flooding.
Submergence of the cables' was of-relatively short
duration and therefore by itself would not be expected to have caused
-j
the failure of a cable starting with good integrity.
Probabl, some pre-
'
existing cable defect interplayed in some manner with the flooding to
j{
-
,
produce a delayed reaction (16 days.) Cables away_ from the fault looked
normal and there has been no history 'of cable failures according to the
1
substation engineers, although they do not keep records.of all cable
faults.
.I
In order to maintain a record of the type of cable involved and the
>
method of installation, the following information is given. The control
!
cable was rated 600 V.
Each of the nine conductors had polyethylene
insulation and a polyvinyl chloride jacket. The assembly had_a
j
helically wrapped copper shield and an overall polyvinyl chloride-
1
Jacket. The conductors were No. 9 American Wire Gauge copper. The power
l
cable was rated 600 V and it- was single conductor 250 thousand circular
j
mils copper.
Insulation was type-THW/THHN/THWN as described in the
National Electrical Code. The control cable was installed about 17
i
years ago and.the power cable was-installed about two years ago. Both
q
cables run cold in normal operation because they carry relatively low
currents.- They were installed in a trench having a sand bottom, concrete
sides and a heavy gauge aluminum top. The purchase specification stated
'
that the cables would be installed in a trench, and they were Florida
i
Power Corporation's standard type cables.
1
?
!
!
!
I
-
.
-
-
-
- - -
. .
. - - - -
-
.
.
-
.
.
5
e
The licensee's corrective actions for the loss-of-power problem
included:
(a)
Inspection and insulation resistance measurement of all
!
cables in the trench.
(b)
Removal and replacement of any damaged cable or cable having
a relatively low resistance reading. About 15 cables were
replaced.
(c)
Testing of any circuits affected by the cable replacement
.:
work.
'
(d)
Inspection of all trenches in the 500 kV and 230 kV
'
(e)
Documentation of the cable replacements and circuit testing.
(f)
Evaluation of the root cause of the event.
(g)
Examined MOS1873 for possible damage. This was a
precautionary measure taken because there was some
possibility that the switch had opened under load. No
deterioration of the switch was found.
(h)
The event and corrective actions were reviewed by the Plant
Safety Review Committee.
(i)
A complete design review from the reliability perspective of
the switchyard was initiated. This was well beyond the
minimum actions necessary to respond to this event.
In relation to the inspection of this event, the most relevant NRC
requirement is the requirement to ascertain the root cause of events and
to implement appropriate corrective actions. The inspector's conclusion
was that the licensee met the requirements, because the event cause
analysis was correct and the corrective actions appropriate and
conservative.
!
3.
Exit Interview
The inspection scope and results were summarized on April 2,1993, with
those persons indicated in paragraph 1.
The inspector described the
areas inspected and discussed in detail the inspection results.
Proprietary information is not contained in this report. Dissenting
comments were not received from the licensee.
,
i
?
i
i
.
-
-
.
-
,
k
,
.
.
s
- m
sn
-
3
)
.
.
,
R(
an7
-
.
-
6
-
c
4
6'
1
5
1
+
1
S
.
l
O
,
l
'
P
-
-
-
-
.
A
-
-
D
,
I
_
R
-
.
,
D
5
-
.
L
.
8/
N
4
F
8
1
0
--
<
_
1
1
6
9
l.
15r
l
5
8
A
l
l
-
T
S(
S)
,
R
,
N
O
O
E
M
.
N
bi
_
,
C
9
.
3
(8
,
i
3
.
,
1
-
9
.
0 1e
1
,
5 f"j
-
-
3
2
,
8
H
8
N
_
i
R O
-
t
- s-
E
I
[
N
0
,
V T
^5
-
6
6
A )
I
6
-
1
R T S
-
1
S U
-
D
L B B
M
.
s
A U G
0
-
ki
T S N
66
os
S
I
1Sy
E
'
/. i
Y V R
3
D
G
R K
(
N
-
1
6
C 0
_
D
6
.
-
I
,
0
_
RE
-
C
v
O
2
5
_
O
w
M
.
.
s
R
e
-
_
.
i
B
-
.
>
.
1]-
_
c
r
-
N
4
!
.
I
2
8
.,
8
,
1
,
3
-
N
?-
,
7
I
8
-
C
,
I
%s
J_
-
1
t
'
n
z
e
-
S
(
_
.i
0
S
,-
9
2
8
$_
,
6
i
l*
l
-
S
I
-
-
O
,
S
M
-
O
, "
N
-
,
,
.
'
n
_
-
_
t
'
a
-
3
.
,
-
3
-
hv R.
3
-
.
w
~
e
,
C
-
-
.
,
.
- a ?!8w $a g*sia
,
3
A? M "
.
,e
,
- %
+
I
-
-