ML17289A260
| ML17289A260 | |
| Person / Time | |
|---|---|
| Site: | Columbia  |
| Issue date: | 01/03/1992 |
| From: | Johnson P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V) |
| To: | |
| Shared Package | |
| ML17289A257 | List: |
| References | |
| 50-397-91-42, NUDOCS 9201220090 | |
| Download: ML17289A260 (20) | |
See also: IR 05000397/1991042
Text
U.S.
NUCLEAR REGULATORY COMMISSION
REGION V.
Report No:
Docket No:
Licensee:
Facility Name:
50-397/91-42
50-397
Washington Public Power Supply System
P.O.
Box 968
Richland,
WA 99352
Washington Nuclear Project
No.
2
(WNP-2)
Inspection
Conducted:
November
4 - December
6,
1991
Inspector:
Approved by:
~ummar:
K. E. Oohnston,
Project Inspector
nson,,
ie
React
Projects Section
3
te
)gne
ection on November
4 - December
6
1991
Re ort No. 50-397 91-42
~td:
Th h'l
f th
l
p tl
t
p
f
l tl
of the ability of the standby
gas treatment
system
(SGTS) to perform its
safety function.
To support this evaluation,
the inspector performed
a system
walkdown in accordance
with inspection
procedures
71710.
Inspection
procedures
36800;
61700,
and 62700 were also used.
Safet
ssues
Mana ement
S stem
SIMS
Items:
None.
~esul ts:
Gene} al Conclusions
and
S ecific Findin
s
~Stree ths:
The inspector found the system condition and lineup accept-
able,
the system engineer
knowledgeable of the system
and its current
condition, operators
knowledgeable of system operations
and following
applicable operating procedures,
and with minor exceptions,
surveillance
and operating
procedures
up to date
and well written.
h
h <<:
Th
l
p t
tdl t
l t
thtt hl
tions were not thorough.
Paragraph
4.a of this report describes
a
Problem Evaluation Request
(PER) which addressed
a surveillance test of
the
SGTS heaters.
This
PER did not address
the potential of heater
damage
which might have resulted
from operation at higher voltages.
Paragraph
4.b describes
a design
change to allow greater
SGTS fan flow.
The analysis
supporting the design
change did not thoroughly address
the
capability of the
SGTS fan motors to support this change.
920i220090 920i03
ADOCK 05000397
8
~
'
jI
0
f
-2-
In both instances,
the licensee
was able to show that the equipment
could
continue to perform its design function.
However, in each
case,
the
design margins were considerably
reduced.
.Si nificant Safet
atters:
none
Summar
of Violations and Deviations:
One violation was identified
regarding the failure to follow the procedure
implementing the
requirements
contained in 10 CFR 50.59 (paragraph 4.b).,
0 en
tems
Summar
One new enforcement
item was opened
and one previous
enforcement
item was closed.
l
0
N
I
gETlgLS
1.
ersons
Contacted
Washin ton Public Power
Su
1
S stem
l
- A. L. Oxen,
Deputy Managing Director
- J.
W. Baker,
Plant Manager
- R. L. Webring, Plant Technical
Manager
- A. G. Hosier, Licensing Manager
- W. S. Davison,
Plant guality Assurance
Manager
- J. F. Peters,
Plant Administrative Manager
.
- D. J; Pisarcik, Assistant'ealth
Physics
and Chemistry Manager
- C. L. Fies,
Compliance
Engineer
- B. E..Pesek,
Plant Technical
Supervisor,
Balance of Plant
J.
V. Hanson,
Plant Technical
Lead Engineer,
Balance of Plant
- J.
E. Bekhazi,
Plant Technical
Engineer,
Balance of Plant
- L. D. Sharp,
Plant Support Engin'eering,
Principal
Engineer
W. D. Bainard,
Plant Support Engineering
uclear
Re ulator
Commission
- P.= L.'ng, Project Manager,
The inspector also talked to other licensee
employees
during the course
of the inspection.
"<<Attended the Exit Heeting
on December
6,
1991.
2.
Standb
Gas Treatment
S stem
Walkdown
The objective of the inspection
was to perform an evaluation of the
ability of the
SGTS to perform its safety function.
To support this
evaluation,
the inspector did the following:
system
walkdown in accordance
with NRC Manual Chapter
71710
system
walkdown with the Plant. Technical
system engineer
observation of system operations
and review .of operating procedures
review of surveillance
procedures
review of scheduled
maintenance,
selected
maintenance
procedures,
and vendor manuals
review of system description contained in the Final Safety Analysis
Report
(FSAR), training material,
and other design basis
documents
C
In summary,
the inspector found:
the system condition and lineup were acceptable,
e
'
-2-
the system engineer
was knowledgeable of the system
and its current
condition,
, operators
were knowledgeable of system operations
and were following
applicable operating procedures,,
- 'ith minor exceptions
di'scussed
below, surveillance
and .operating.
procedures'were
up to date
and well written.
During the review of the annunciator
response
procedures
for the
panels
(4.827.KI), the inspector noted that one of the steps for an
window did not make
sense
(Window 4-1, step
5 and 6).
The
inspector pointed this out to the licensee
who acknowledged
the error and
committed to change the procedure.
The inspector
found two instances
wherein engineering
work was 'less
than
rigorous.
These
examples,
one-regarding
the review of an out-of-specifi-
cation surveillance test
and th'e other regarding
a design
change,
are
discussed
in section.4 of this report.
To support the discussion
in
section 4, it was necessary
to provide the following brief system
description.
3.'GTS Descri tion
The
SGTS was designed to limit the release
of airborne radioactive
contaminants
from the secondary
containment
(the reactor building) to
atmosphere
following a design basis accident.
In additi.on, the system
was designed to enable
purging of the primary containment
through the
SGTS filters when airborne radiation levels inside the, primary contain-
ment are too high to permit direct purging to atmosphere.
The system is
required
by the Technical Specifications
(TS) to be operable
in Nodes I,
2,
and
3 and during refueling operations.
The system
was designed
to
remove
99 percent of iodides
and particulates.
The
SGTS is composed of two .full-capacity filter trains, originally sized
to provide
a reactor building pressure,
under design basis conditions,
0.25 inches of water
gage (w.g.) lower than atmospheric pressure.'ach
filter train consist of the following:
a moisture separator,
a prefilter, two high efficiency particulate-
air (HEPA) filters, and two 4" deep
banks of impregnated
charcoal
adsorber filters,
two banks of electric blast coil heaters,
rated at 20.7
kW each at
460V, provided to limit the relative humidity of the incoming air to
-70 percent,
and
two full-capacity centrifugal fans,
powered from separate vital 480V
buses.'he
SGTS will automatically initiate on high drywell pressure,
reactor
vessel
low water level,
and high radiation level in the reactor building
exhaust air plenum.
With an automatic actuation,
reactor building intake
valves,
fan intake valves,
and
SGTS discharge
valves to the elevated
0
Jl
t',
-3-
release
duct will open
and
one of the electric heaters will energize.
Following a 10-second
delay,
one fan on each train will start.
If it
does not operate
successfully
(achieve
a flow of 500 cfm), the second
fan
and electric heater will start.
System flow is controlled to maintain building pressure
negative with
respect to atmosphere
and is limited to a maximum flow.
The differential
pressure
is measured
on the four sides of the reactor build>ng at the
572'evel
(the reactor building basement
is at 422'nd the refueling
floor is at 606').
System flow is measured
by annubars
at the fan
.discharge.
It was recognized
in 1989 that by controlling the reactor building to
-0.25" w.g. with monitors
on the 572'evel, it would not be possible to
maintain pressure'negative
with respect to atmosphere
at all locations.
To achieve -0.25" w.g. in all locations,
the licensee
determined through-
tests
and calculations that building, pressure
ne'eded to be controlled to
-1.7" w.g. using the monitors
on the 572'evel.
It. was.noted that the
licensee
has led the industry in the study of atmospheric effects
on
reactor building differential pressures.
To establish
a more negative building pressure,
the
SGTS fans'pper
flow
limit was increased
from 4457 cfm (as measured
by a discharge
flow
annubar)
to 5600 cfm.
The issue
was discussed
by the licensee in a
Justification for Continued Operation
(JCO)
and was the subject of
Licensee
Event Reports
(LERs) 89-40
and 89-40, Revision
1 (dated. June
19,
1990).
4.
~indin
s
a.
Electric Heaters
0 crated at Hi
h Volta e Durin
Surveillance
~Testin
On April 15,
1991, shortly after plant shutdown for the
a refueling
outage,
a surveillance test
was performed
on
SGTS heaters
to measure
whether their heat generation
was
i'n compliance with the TS.
On two
heaters,
heat'generation
was found to be higher than the
TS limit.
This was
due to a high bus voltage condition.
To evaluate the con-
'ition and initiate corrective action,
a Problem Evaluation Request
(PER)
was initiated.
The inspector
reviewed the
PER (No. 291-250)
and found the following
weaknesses:
The actions
taken were poorly documented,
The review did not adequately
address.
whether the heaters
had
been
damaged,
and
The review did not address
whether the heaters
could operate
in
all voltage conditions
and still meet
TS requirements.
']
I
III
i
l
f
4
-"
escri tion of the Surveillance
and
P
R
TS surveillance 4.6.5.3.d.4
requires that every
18 months the
licensee verify that the
SGTS heaters
dissipate
20.7
+ 2. 1
kW.
TS
surveillance
procedure
TSS 7.4.6.5.3.4
was performed
on April 15,
1991 to meet this
TS requirement.
The'methodology of TSS 4.6.5.3.d.4
was to measure
the line current
and phase-to-phase
voltage of each of the nine heater coils which
make
up each heating unit.
The product of the current
and voltage,
cor rected for the coil configuration (three sets of coils wired in
delta) provided heat dissipated
in watts.
The measured
heat dissipation
was
as follows:
SGT-EHC-1A1
23.12
kW
(SGTS train A; Division
1 power supply)
SGT-EHC-1A2
-
22.21
kW
(SGTS train A; Division 2 power supply)
SGT-EHC-181
23.82
kW
(SGTS train B; Division
1 power supply)
SGT-EHC-1B2
22.79
kW
(SGTS train B; Division 2 power supply)
The heat dissipation
on SGT-EHC-1A1 and 1Bl were above the
TS limit
of 22.8
kW.
These results
could have
been anticipated.
Bus voltage
conditions,
as might be expected
during an outage,
were higher than
during normal operations
due to the reduced
bus loading.
Voltage
supplied to the heater cabinets
was measured
to be
9X higher than
nominal voltage
(460V).
Due to the exponential
relationshi,p
between
voltage
and power which exists in resistive loads
such
as heaters,,
the heat generation
was approximately
20X greater
than would have
been
expected
at 460V,
on which the 20.7
kW power rating was based.
The licensee initiated
PER 291-250 to address this issue.
The
surveillance
procedure
was revised to change
the methodology for
calculating heat dissipation.
Essentially,
the actual
heat output
was normalized to what would have
been expected
had the voltage at
the'eaters
been
460V.
The normalized heat outputs 'for SGT-EHC-1Al
and
1B1 were calculated to be 19.5 and 20.1 respectively.
Based
on
these results,
the licensee
concluded that the
TS surveillance
requirement
had been met,
and the
PER was closed.
Problem Evaluation
Re ort Documentation
The review and resolution of this issue
were not well documented
in
PER 291-250.
The Management
Review Committee
(MRC); which reviews
all
PERs,
selected
as the problem resolution method
"NONE/CLOSE
PER."
The comments section provided the details of the resolution:
"High bus voltage is normal
on plant shutdown.
Equip. previously
evaluated
in LER [Licensee
Event Report] to withstand high voltage.
This has
been evaluated
by compliance
and found to be acceptable.
Closed with the
PDF [Procedure-Deviation
Form] already written."
t
I
0
fl
/]
L
I
,(
1
Weaknesses
noted in this evaluatio~'ere:.
The
PER referenced
an
LER as supporting the ability of the
heaters
to withstand high voltages.'t
did not provide
an
LER
number or maximum high voltage allowed.
When questioned,
the
licensee
could not produce the supporting
LER.
The
PDF number
and date were not provided, nor was the date the
surveillance
was performed.
These
documents
were subsequently
retrieved
by the inspector.
The
HRC resolution
was misleading;
the
PDF was the resolution
method.
The lack of thorough documentation of the problem and path of
resoluti.on
appears
to have contributed to
a less
than thorough
technical
review as discussed
below.
The inspector discussed
these
weaknesses
with the licensee
at the exit meeting.
The licensee
committed to reopen the
PER and address this issue.
Haximum Heater
ut ut
The licensee
had not performed
an adequate
review of the ability of
the heaters
to withstand heat dissipation in excess of their TS
limit.
This demonstrated
an incomplete review of the technical
issues
raised
by the failed surveillance test.
The
TS surveillance
acceptance
criteria provide
10X allowance
above
and below the 20.7
kW rating of the heaters.
The 20.7
kW, according
to design
documents,
was
based
on 460V supplied to the heaters.
The
licensee
stated that the
IOX rating was typical of heaters.
The measured
output on one heater
was 23.8
kW, or 15X greater than
the design rating.
The inspector
questioned
whether this could have
damaged
the heater.
After discussions
with the vendor regarding
these
heaters,
the lice~see
found that under low flow conditions
a
watt density of 60 W/in
would have
been acceptable.
Higher watt
densities
could reduce the life of the heaters
or, if substantially
higher,
damage
the heaters.
The lic~nsee determined that 23.8
KW
provided
a watt density of 56.7 W/in, and concluded that the
heaters
were not damaged
(20.7
KW provides 49.3 W/in ).
The licensee
should
have included
an engineering
review of the
acceptable
watt density effects
on the heaters
in the resolution of
the
PER.
In this case,
the heaters
had additional margin.
0 eration of the Heaters
Under All Desi
n Conditio
s
Operation of the heaters
at elevated
voltages
does not appear to
have
been considered
in the plant design.
This point should
have
been recognized during the review of PER 291-250.
A complete review
should
have considered
the maximum voltage condition at which the
heaters
would be expected
to operate.
~
hl
k
t
Licensee
procedure
TSS 7.4.8.3.2,
which required that
bus voltages
be measured
on
a weekly basis,
allowed
a maximum voltage at the
bus
of 520 V. If a surveillance test found that
a heater dissipated its.
maximum acceptable
output of 22.8
KW at 460 V,'his heater
would
operate at a watt density (~ccording to calculations
performed
by
the inspector) of 69.3 W/in
at 520 V.
According to vendor data,
this worst case situation could result in heater degradation.
onclusion
The licensee's
disposition of PER 291-250
was not rigorous in either
technical
or administrative
aspects.
In this case,
adequate
margin
existed to prevent equipment
damage.
The licensee
committed at the exit meeting to reopen,
PER 291-250 and-
address
the issues
discussed
above.
an Motor
oadin
The inspector reviewed the design
change
which allowed the
SGTS fans
to provide
up to 5600 cfm.
The inspector found that the design
change did not adequately
consider
whether the installed fan motors
could operate
the fans at the greater
flow under all design
conditions.
In response
to this finding, the licensee
determined
that under worst case conditions,
the nameplate rating of the fan
motor and the full load current would be exceeded.
However, tak'ing
into account the fan motor service factor and the size of the fan
motor overload device,
the licensee
concluded that the fans would
operate
under the conditions described
in their Justification for
Continued Operation
(JCO) regarding the
SGTS, discussed
in LER
89-40, Revision l.
As discussed
in the Section 3,
"System Description," the upper flow
limit of the
SGTS fans
was raised
from 4457 to 5600 cfm in 1989 to
provide
a. greater
reactor building negative pressure.
To accomplish
this change,
the flow limiter setpoints for all four fans
(SGT-LNTR-
lA1, 1A2,
1B1,
and
1B2) were revised
by Instrument Setpoint
Change
Request
ISCR 952.
The design
change
review,
documented
in a
10 CFR 50.59 evaluation,
concluded the following:
The charcoal
residence
time remained within Regulatory
Guide
1.52 requirements,
Differential pressure
across
the
HEPA filters was within the
manufacturer's
tolerances,
and
The heaters
could maintain relative humidity less than
70X.
The evaluation fu'rther stated that
"Ho other elements of the design
are considered
impacted
by the reliance
on the higher than design
flow rate for a single. SGT train."
To support this conclusion,
a test
was performed for fan SGT-FN-1A1
which found:
'I
~
'
k
~,
l
tl
Motor current to be 25.5
amps at 5600 cfm
- ,
Leakage tests
and adsorption tests to be with specification
The inspector determined that the licensee
should
have considered
the following design basis
par ameters,
each of which could have
affected the maximum current drawn by the
SGTS fan motors:
~ddV1t:
C
t
qd
t
i
i tl
motors
under
degraded
voltage conditions.
The licensee
stated
that the design basis
degraded
voltage condition for the moto'rs
was 405
V and determined that this would increase
current draw
by 13.6X.
nstrument
Inaccuracies:
The overall inaccuracy of the flow
controllers
was
10X.
Testing performed
on June
5,
1991, in
accordance
with TS 4.6.5.3.2.d.l,
which requires verification
of system flow, found that for one of the fans the-actual
flow
was
9X greater
than the controller setting.
Cold Buildin
Tem eratu
s:
The fan performance
curves
were
established
for a building temperature of 104'.
A reduction
in building temperature
provides
a proportional
increase
in air
density
and motor amps required to move the
same
volume of air.
A 5'emperature
drop corresponds
to a
1X increase
in the motor
current requirement.
The difference in temperature
between the
performance
curves
and the minimum'building temperature
of,
50'
increases
current requirements
by 11X.
In response
to the inspectors
concerns,
the licensee
performed
a
calculation which considered
these three factors.
The results
indicated that under the worst-case
conditions,
motor current would
not trip the thermal overload devices.
The licensee
factored into
their analysis
the effects of low ambient temperature
on the
set~oints of thermal
overload trip devices.
These devices,
rated at
104
F, trip at higher currents at lower ambient temperatures.
The calculation performed
by the licensee
assumed
a maximum steady
state
SGTS fan flow of 5075 cfm versus the flow limiter setpoint of
5600 cfm.
This was
based
on actual
performance
tes'ts
completed in
September
1991.
These tests
concluded that to maintain the required
negative pressure,
full system flow (5600 cfm) would not be
required.
A factor in the tests
was that actual air leakage into the reactor
building (1475 cfm) was substantially less than the maximum leakage
the
TS allow (2240 cfm). 'he licensee's
JCO, discussed
in LER
89-40,
Revision
1, supported
the use of lower reactor building air
leakage.
The
LER stated that air leakage into the reactor building
would be limited to 1475 cfm until the reactor building negative
pressure
issue could be completely resolved.
~
i
0't'
Incom lete Desi
n Chan
e
Revising the
SGTS flow requirement
from 4457 cfm to 5600,cfm was
a
design
change
implemented
by ISCR 952.
The change
included
an
.
"unreviewed safety question" evaluation required by 10 CFR 50.59.
These evaluations
are governed
by licensee
procedure
PPH 1.3.43..The
evaluation did not consider the ability of the .fan motor to operate
the fan at the higher flow under all design basis conditions.
These
conditions included degraded
voltage,
instrument inaccuracies,
and
the effects of building temperature.
The failure to consider
these
conditions in the evaluation is an apparent'iolation
(Enforcement
Item 91-42-01) .
5.
ollowu
on 0 en
tems
The inspector
reviewed the actions taken
by the licensee
in response
to
the Notice of Violation,(NQV) tr'ansmitted with Inspection
Report
No.
91-24
and found them to be acceptable.
The
NOV addressed
the licensee's
failure to initiate a
PER to address diesel-starting air system high
'ressures.
The licensee's
root cause evaluation
and corrective actions
were discussed
in letters dated
October
14,
1991
and October 28,
1991.
In addition, the licensee's
commitment'o
improvements
in the Plant
Technical
system engineering
program,
discussed
in their October
14,
1991
letter,
were reviewed
and found to be appropriate.
Enforcement
Item 50-
397/91-24-01 is closed.
S.
~iN
An exit meetings
was conducted
on December
6,
1991, with the licensee
representatives
identified in Paragraph
1.
The inspector
summarized the
inspection
scope
and findings as described
in the Results section of this
report.
The licensee
did not identify as proprietary
any of the information
provided to or reviewed
by the inspector during this inspection.
0
L