ML17157A372
| ML17157A372 | |
| Person / Time | |
|---|---|
| Site: | Susquehanna  |
| Issue date: | 10/12/1990 |
| From: | Gramm R, Jeffrey Jacobson, Lanning W Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML17157A371 | List: |
| References | |
| 50-387-90-200, 50-388-90-200, NUDOCS 9010240232 | |
| Download: ML17157A372 (79) | |
See also: IR 05000387/1990200
Text
U.S.
NUCLEAR REGULATORY CO/MISSION
OFFICE
OF NUCLEAR REACTOR REGULATION
Division of Reactor Inspection
and Safeguards
NRC Inspection Report:
50-387/90-200
50-388/90-200
Docket Nos.:
50-387
50-388
License No.: NPF-14
NPF-22
Licensee:
Power and Light Company
Facility Name:
Susquehanna
Steam Electric Station Units
1 and
2
Inspection
Conducted:
August 13-17
and 27-31,
1990
Inspection
Team:
NRC Consultants:
Jeffrey B. Jacobson,
Team Leader,
Jul io Lara, Region I
Roy Mathew, Region I
Shashikant
V. Athavale,
Richard Jolliffe, AEOD
Atty E. Almond,
Omar Hazzoni,
AECL
Alek Josefowicz,
AECL
Jim Beaton,
AECL
Prepared
by:
Je
r
a
son,
Team Lea er
Team
n
c
n Development Section
C
Specia
In
ection Branch
Division f Reactor Inspection
and Safeguards
Office of Nuclear
Reactor Regulation
loh~J o
Date
Reviewed by:
o
r
.
ramm,
~e
Team Inspection
Development Section
C
Special
Inspection
Branch
Division of Reactor Inspection
and Safeguards
Office of Nu lear Reactor Regulation
Approved by:
ay
e D.
La
ing,
>e
Sp
ial Inspection
Branch
Division of Reactor
Inspection
and Safeguards
Office of Nuclear Reactor Regulation
eosoahaisz
vosoisl f
ADOCK 05000387
6
veau
I
>
~ '
(
I
EXECUTIVE SUtNARY
During August
13 through
17 and
27 through 31, 1990,
a Nuclear Regulatory
Commission
(NRC) inspection
team conducted
an electrical distribution system
.functional inspection
(EDSFI) at Susquehanna
Steam Electric Station
(SSES)
Units
1 and
2 to determine if the electrical distribution system
(EDS) wou1d be
capable of performing its intended safety functions as designed,
installed,
and
configured.
The team also assessed
the licensee's
engineering
and technical
support of EDS activities.
The team conducted plant walkdowns
and technical
reviews of the calculations
and associated
documents pertinent to the
EDS and
interviewed corporate
and plant personnel.
The team identified 14 findings that were discussed
with the licensee
during
the exit meeting
on August 31, 1990.
Weaknesses
in the licensee's
discrepancy
management
program may have contributed to the findings related to the unquali-
fied Limitorque valve motor actuators,
the inadequate
setpoints for degraded
grid relays,
the inoperable level indication for the diesel fuel oil storage
tanks
and the inadequate
evaluation of relay setpoint calibration data.
Although the licensee
had identified these deficiencies before this inspection,
,it had failed to adequately
and effectively correct the problems.
Weaknesses
or errors in the original plant design
may have contributed to the
findings related to the lack of overpressure
protection for the diesel
emergency
(ESW) piping, the overvoltage condition on the
dc system,
and the misapplication of an undervoltage
relay.
Certain of the findings were of particular concern
because
of their potential
safety significance.
These
included the inadequate
setpoints for the degraded
grid relays,
the unqualified Limitorque valve motor actuators,
the inadequate
testing of dc circuit breakers,
and the inoperable level indication for diesel
fuel oil storage
tanks.
Specifically, at their current setpoint,
the degraded
grid relays
cannot ensure that adequate
voltage will be available to all
safety-related
loads.
During the inspection,
interim operator
procedures
were
implemented which mitigated this concern until proper setpoints
can be
established.
In addition,
numerous safety-related
Limitorque valve motor
actuators
are not environmentally qualified because
they are operated with
250-volt control power and nontested
motors.
The testing of dc circuit
breakers with an ac current source
has resulted
in uncertainty
as to their
expected field performance.
Finally, the inoperability of level indication for
the diesel fuel oil storage
tanks could lead to uncertainties
in actual fuel
levels during accident conditions.
The team also noted strengths
in the plant and corporate organizations.
Among
these
were good coordination
between
corporate engineering,
the site,
and
nuclear licensing;
good control of instrument setpoints;
and
a high quality of
engineering
associated
with recent modifications.
The Pennsylvania
Power and
Light Company
(PPSL)
team that interacted with the inspection
team was very
well prepared,
technically competent,
and organized.
In addition, internal
gA audits were thorough
and identified some
key programmatic
concerns
although
corrective action for these
concerns
were found to be slow or lacking in
several
instances.
TABLE OF
CONTENTS
1.0
INTRODUCTION..................................................
Pa>ac
2.0
ELECTRICAL DESIGN.............................................
2.1
2.2
Offsite Grid and 13.8-kVac Systems.................
4160-Vac Class
1E System...........................
2.2.1
Switchgear Short Circuit Ratings............
2.2.2
Protective Relaying.........................
2.2.3
Transfer of 4160-Vac
ESF Buses
Between Offsi
~ ~ ~ ~ ~ ~
te Sou
~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~
rces.
1
2
3
3
3
2.3
2.4
2.5
2.6
Emergency Diesel Generators........................
480-Vac Class
1E System............................
2.4.1
4160-Vac/480-Vac
1E Transformers Sizing.....
2.4.2
480-Yac Switchgear Short Circuit Ratings....
2.4.3
Yoltage Regulation/Review of ASDOP Program..
120-Vac Class
1E System............................
250/125-Vdc Class
1E System........................
2.6.1
Yoltage Drop................................
2.6.2
Battery Sizing/Short Circuit Study..........
2.6.3
DC System Overvoltage.......................
2.6.4
Motor Control Center Circuit Breakers.......
5
5
5
6
7
7
8
9
10
2.7
2.8
2.9
2.10
Protective Coordination............................
120-Vac
1E Distribution System.....................
Penetration Sizing.................................
Degraded
Voltage and Loss of Yoltage Relays.......
10
11
12
12
3.0
MECHANICAL DESIGN........... "... -. -."...... -.... -... -.... - .. -. - ..
15
3.1
3.2
3.3
3.4
3.5
Heating, Yenti lation and Air Conditioning
Power Demands for Major Loads............
Emergency Service Water System...........
Diesel Generators
and Auxiliary Systems..
Fire Protection System...................
Systems.............
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ t ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
15
16
17
18
20
4.0
ELECTRICAL EQUIPMENT TESTING AND SURVEILLANCE......................
21
4.1
4.2
4.3
4.4
4.5
4.6
Diesel Generator Testing.......
Setpoint Calculation
and Contro
Circuit Breaker Testing........
Fuse Control...................
Inverter Testing...............
Battery Testing................
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~
~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
1
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ I ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
21
21
22
23
24
24
5.0
ENGINEERING AND TECHNICAL SUPPORT....................
~
~
5.1
Equipment Modifications.........................
5.2
Discrepancy
Management System...................
5.2.1
Deficiency Control System................
5.2.2
Summary of Discrepancy
Management
System
6.0
GENERAL CONCLUSIONS..................................
25
26
~ ~ ~ ~ ~ ~ ~
Review.
26
32
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
33
~Pa
e
~ o ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
5
APPENDIX A - Deficiency Sheets............................
APPENDIX
B - Personnel Contacted..........................
A-1
B-1
1.0
INTRODUCTION
During recent inspections,
the Nuclear Regulatory
Commission
(NRC) inspection
teams
observed that the functionality of safety-related
systems
had been
compromised
as
a result of design deficiencies
introduced during design modifi-
cations of the electrical distribution system.
Inspection
teams also discov-
ered that the actual configuration of the installed equipment
does not adhere
to the original design.
Consequently,
the
NRC initiated the electrical distri-
bution system functional inspections
to determine if the electrical distribu-
tion systems
(EDSs) at operating nuclear
power plants are adequate.
The two main objectives of this inspection were {1) to ensure that
EDS power
sources
and equipment
were adequate
to support the operation of the plant's
safety-related
equipment
and (2) to assess
the licensee's
engineering
and
technical
support associated
with this system.
The team reviewed calculations
and associated
documents
to ensure that electrical
power of acceptable
voltage,
current,
and frequency would be available to safety-related
equipment
powered
from the station
EDS.
The review included portions of the onsite
and offsite
EDS including the station startup transformers,
13.8-kVac system,
the 4160-Vac
Class
1E system,
the emergency diesel generators,
the 480-Vac Class lE system,
the 120-Vac Class lE system,
the station batteries,
and the 250-Vdc and 125-Ydc
Class
lE systems.
The team also reviewed the mechanical
systems
which inter-
face with the
EDS, conducted
an onsite walkdown, and reviewed maintenance,
calibration,
and surveillance activities for the above mentioned listed sys-
tems.
In addition, the team reviewed selected modifications and the licensee's
discrepancy
management
program to assess
the capabi lity and performance of
engineering
and technical
support.
The team verified conformance with General
Design Criteria
(GDC) 17 and
18 and
appropriate criteria of Appendix 8 to 10
CFR Part 50.
The team reviewed plant
Technical Specifications,
the Final Safety Analysis Report,,
and appropriate
safety evaluation reports to verify that technical
requirements
and licensee
commitments
were being met.
The specific areas
reviewed and the team's find-
ings are described
in Sections
2 through
5 of this report.
A summary of the
conclusions,
strengths,
and weaknesses
is given in Section 6.
Each finding
addressed
in the report is provided in Appendix A and is categorized
as either
an open or an unresolved
item.
A list of personnel
contacted
is provided in
Appendix
B and persons
attending the exit meeting are indicated with asterisks
before their names.
2.0
ELECTRICAL DESIGN
To obtain
a clearer understanding
of the electrical design,
the team examined
system descriptions,
design reports, electrical
design calculations
{including
system loading, fault level, protection settings
and coordination, voltage
regulation,
and equipment sizing), design
changes,
nonconformance
reports,
and
equipment specifications.
The specific areas of the design review are dis-
cussed
below.
2.1
Offsite Grid and 13.8-kVac Systems
The power system grid provides for two independent offsite power sources.
One
source is established
by tapping the Montour-Mountain 230-kV line north of the
I
~
I
~
Susquehanna
plant to startup transformer
10.
The second offsite power source
is supplied at 230-kY from a tie between
the 500-kV and 230-kV switchyards at
the south
end of the plant and connects
to startup transformer
20.
The two
step-down transformers
supply power from the offsite grid to two separate
non-Class
lE 13.8-kVac buses.
These
buses
supply the systems that are not
designated
Class lE and also the Class
1E systems
through two engineered
safety
feature
(ESF) step-down transformers
per bus.
The team asked the licensee
about the
1975 to 1985 data contained in the
(page 8.2-10), which indicated that for the Montour-Mountain 230-kV line, 1.96
outages
had occurred every
100 circuit miles.
The licensee
responded
that the
FSAR reliability data is no longer accurate
because
considerable
improvements
had
been
made during the last
4 years.
The licensee
said that the
FSAR will be
revised to include the latest data,
which reflects the improved reliability as
a result of licensee's
upgrading surveillance
and maintenance
of the line.
The transformers
were sized to provide for all load requirements for the most
stringent operating condition, which is a loss of coolant accident
(LOCA) in
one unit with the other unit carrying full load and with one of the two offsite
sources
out of service.
The kilovolt ampere capability, connections
to the
safety buses, field installation, protection, testing
and surveillance,
and
voltage regulation of the offsite power source
and
ESF transformers
were
adequate.
The transfer
scheme at the 13.8-kVac level allows for the supply of power to
the plant auxiliary systems that are not designated
Class
1E during plant
startup
and shutdown.
A synchronizing relay is used to allow the transfer of
the two sources
of power within predetermined
limits of phase
angle.
The
licensee
had selected
the setting of the synchronizing relay to avoid undue
protective relay tripping under the inrush current introduced
by the transfer.
However,
no consideration
had been given to the need to verify that the voltage
drop through the transformer would be acceptable.
In response
to the team's
concern,
the licensee
performed preliminary calculations that showed the
voltage drop would not exceed
25 percent.
This was acceptable
to the team.
2.2
4160-Vac Class
1E System
The onsite
power sources
consist of four load group channels.
Each load group
channel
can be supplied
by one of four emergency
diesel generators.
Any three
out of the four load group channels is capable of meeting the design-basis
requirements.
The 4160-Vac buses
supply the large motors
and
one
power center
per bus;
each is rated at 750
kVA and feed the smaller loads at 480 Vac.
The
5-kVac cables
connect the 4160-Vac buses
and transformers.
All redundant safety
loads are divided between division
1 and
2 buses.
The
4160-Vac safety
buses
and their connected
loads
had adequate
load current and
short circuit current capabilities,
protection,
and cable connections
between
loads
and buses.
Testing
and surveillance
and compliance with the single-
failure criterion appeared
adequate.
In addition, the fast bus transfer
scheme
met applicable separation
requirements for safety systems.
a
~I
2.2.1
Switchgear Short Circuit Ratings
The licensee
does not have
a calculation index,
or other system for controlling
the use of non-valid or superseded
calculations.
As a result, three calcula-
tions were found to exist for the short circuit ratings of the lE switchgear.
These calculations
were produced during 1980 and
1982 and,
even though the
calculational results
and assumptions
were not the
same,
the licensee
consid-
ered
them valid and had put them into effect simultaneously.
These calcula-
tions were found to contain several
nonconservative
features
including those
listed below.
There were errors in the application of breaker interrupting factors.
\\
Maximum possible voltage was improperly considered.
The addition of ESF transformers
installed before plant startup
was not
incorporated into two of the calculations.
One of the calculations
was considered
"non-g" and
had not received
appropriate
technical
review.
In response
to the team's
concerns,
the licensee
performed
a fourth calculation
(SC-I) and gave it to the team to review in preliminary form on August 31,
1990.
This calculation indicated that most of the team's
comments
and concerns
had been addressed.
However, the available margin between
duty and rated
values for the short circuit capability was minimal.
The licensee
committed to
perform formal short circuit calculations to allow for a proper evaluation of
this issue.
This item is identified as Item 1 of Open Item 90-200-01, of
Appendix A to this report.
2.2.2
Protective Relaying
The team identified that the ground sensors for the 4160-Vac system were set at
2 amperes with an accuracy of 2 amperes.
This would give a maximum pickup trip
of 4 amperes.
The team was concerned that the protective relaying
schemes
and
settings
would be insensitive to ground faults involving fault impedance,
as
well as those occurring inside windings of machines.
To verify protection for
these
cases,
the team performed its own calculations to reevaluate
the maximum
ground fault current.
The team found that when capacitive
components
are
included, the ground fault current would be more than
4 amperes,
which would be
adequate
to actuate
the ground sensor relays.
2.2.3
Transfer of 4160-Vac
ESF Buses
Between Offsite Sources
During
a loss of one offsite power source,
the
EDS is designed
so that the
feeder breaker to the
ESF transformer will trip and the alternate
feeder
breaker will close to supply voltage.
This is a slow bus transfer
because all
4160-Vac bus
loads will experience
an undervoltage trip before the alternate
source provides
power to the buses.
Therefore,
the transfer
scheme
would
preclude the possibility of. closing in on an out-of-phase
condition and
no
undue transient effects
on
1E motors or undue voltage drops would be caused
by
this transfer.
No other concerns
were noted with the 4160-Vac transfer.
'f
~
2.3
Although the emergency
diesel
generators
(EDGs)
had
an adequate
kilowatt
rating, appeared
able to start
and accelerate
the assigned
safety
loads in the
required time sequence,
was adequately
protected,
and had sufficient voltage
and frequency regulation under transient
and steady state conditions, the team
identified problems in the diesel
loading tabulations
and with the diesel
overcurrent relays.
Details of the teams findings in these
areas
are discussed
below.
Cable losses
had not been included in the loading tabulations
in the calcula-
tions pertaining to the loading of the
EDGs.
As a result, the licensee
per-
formed
a calculation for the largest
EDG cable connection to the bus
(EDG E)
and found that the associated
losses
were approximately
8
kW.
The team consid-
ered these
losses sufficiently important to warrant
a revision of the loading
tabulations
although the existing loading margin of approximately
7 percent
(326
kW) for overall diesel capacity would still be sufficient after all the
cable losses
were included.
The licensee
committed to perform complete calcu-
lations
and to revise the A-E diesel
loading tables accordingly.
This item is
included as part of Open Item 90-200-01,
Item 2 in Appendix A to this report.
In the modeling of the transient conditions
imposed
by the restart of the
largest
load [2000-HP residual
heat
removal
(RHR) pumpj, the bus preload
was
assumed
to be 2000
kW when the actual
bus preload could be as high as 2600'W.
When the team discussed
this discrepancy with the licensee,
the licensee
contacted
the diesel manufacturer
who performed
an additional
computer analysis
for the case with the bus preload at 2600 N.
The analysis
showed that the
diesel
generator
can successfully restart
an
RHR pump with a bus preload
of.
2600
kW.
Therefore,
the
EDGs had adequate ability to successfully start
and
accelerate
the assigned
safety loads.
The additional diesel
loading requirements for a Unit I design-basis
accident
(DBA) and
a Unit 2 forced shutdown scenario
are given in Emergency
Operating
Procedures
(EOPs).
These required accident
and shutdown
loads are based
on
Loading Calculation
E-RGF-002
and
FSAR Table 8.3-1a to preclude
any overloading
of the diesel generators.
The
do not require
any manual tripping of
loads.
All DBA and shutdown loads,
except motor-operated
valves,
are
conservatively
assumed
to be connected
to the diesel generator
as
shown in the
FSAR.
Procedure
AR-016-001 requires control room operators
to verify and shut
down
diesel
generator
loads if an overcurrent alarm condition occurs.
This alarm is
provided by an inverse time overcurrent relay.
However, the team found that
the overcurrent relay settings
were not adequate for diesel generator
E and
that all diesel relays exhibited excessive drifting, which was not properly
accounted for in the relay setting. calculations.
The team was concerned that
because
of the improper settings
and the drifting of setpoints,
the overload
alarm might not be initiated if a diesel generator
was overloaded.
This item
is identified as Unresolved
Item 90-200-02 in Appendix A to this report.
2.4
480-Vac Class
1E System
The 480-Vac Class
1E distribution system including the load center
(LC) trans-
formers, switchgear
and motor control center
(NCC) short circuit ratings,
ground fault protection,
motor overload protection,
and voltage regulation
was
~
~
I
reviewed
and appeared
adequate.
The 120-Vac Class
1E distribution system
protection coordination
and the design features of the 120-Vac
1E inverters
also appeared
adequate
and met applicable
requirements.
Specific details of
the team's
review are described
below.
2.4.1
4160-Vac/480-Vac
1E Transformers Sizing
Bechtel Calculation E2020.02,
Revision 1, related to the loading of the Unit 1
NCCs and
LCs was reviewed
by the team.
The team observed that under worst-case
conditions for LCs 1B230 and IB240 there is a 5-percent
margin between the
rating and the loading of the transformers,
which is adequate
but would limit
future increases
in loads
on these
buses.
The 1B240
LC had
no spaces
available
for additional circuit breakers
and
1B230
LC had only one spare
space.
In
addition,
2 of the 20 Unit 1 NCCs
(OB136 and OS146) were found to be loaded to
86 percent
and 80 percent,
respectively,
which would limit the future increases
in loads for the areas
serviced
by those
NCCs.
The team noted the drawings for
the two NCCs had warning notes describing the loading limitations.
2.4.2
480-Vac Switchgear Short Circuit Ratings
The team identified that Bechtel Calculation E2005.03,
Revision 0, which
related to the ability of the
NCC circuit breakers to accommodate
postulated
fault currents,
did not include cases
where the 4160-Vac/480-Yac transformers
and
LCs were loaded to the maximum.
These
cases
were described
in Calculation
E2020.02,
Revision 1, for sizing of the transformers
and were used
as input to
the
ASDOP computer program.
The calculation only applied to Unit 1, which also
feeds the station
common loads,
and the calculation
used
a nominal system
voltage value instead of the highest voltage.
The team's
review of the existing calculation,
considering the missing require-
ments,
indicated that the rating of NCC circuit breakers
would not be exceeded
under the worst-case
conditions (i.e., highest
system voltage, highest trans-
former loading,
and negating
the cables
between
the
The licensee
indicated that the calculations will be updated
under
a program for review and
revision of all calculations.
Bechtel Calculation E2006.01,
Revision 4, documented
the basis for selection of
480-Vac power
cables
between
the
LC to NCC with regard to short circuit ratings
of those cables.
However, there
was
no calculation for the selection of the
cables
from the
NCCs to the ultimate loads although there
was
a set of require-
ments in the design-basis
calculation
document ("Electr ical Protective Devices
- DBC-1") that governed the selection of the settings for protective devices.
The protective devices
were set
so that insulation for the cables
would not be
damaged
by thermal effects of fault currents.
2.4.3
Voltage Regulation/Review of ASDOP Program
The licensee
used the
ASDOP computer program to calculate
a voltage profile for
the ac system to the 120-Vac level and calculated
short circuit fault levels
and protective relay settings
by hand.
Because
the computer program could not
model the emergency
diesel generators
(EDGs), the licensee
had not used it to
verify the behavior of the
EDGs during the starting of large loads or blocks of
~
~ * i
~
~
loads following a loss of offsite power
(LOOP).
However, the licensee
had
verified the voltage regulation following a
LOOP during an actual
run of the
EGGs.
The initial settings of the stepdown transformer s had been selected
so that the
loads would riot be exposed to voltages
10 percent higher than their nominal
rating anywhere in the system
(13.8-kVac to 480-Vac).
It was
assumed
that the
on-load tap changers
of the 230-kVac/13.8-kVac
stepdown transformers
would
maintain the system voltage in the desired
range
and that the 4-kV bus voltage
would be maintained
by the
EDG automatic voltage regulator
when the Class
1E
system
was operated
from the
EDG.
As a result of the
ASDOP analysis,
the
licensee
had changed
the load restarting
sequence
to maintain the 480-Vac bus
voltages at or above
80 percent of nominal rating.
Table E-57, Revision 5, and Bechtel voltage drop Calculations
E2006.01
(Revision 0) and E2006.04 listed the
maximum distance to the load and maximum
permissible
load (horsepower/amperes)
for each
cable size in the 480-Vac
Class
IE system.
These were design guides written to ensure
the voltage would
remain at or above
80 percent of the nominal rating for starting the
EDG and at
90 percent for other loads.
Even though this was not a verification of the
actual
design,
the team found this approach
of documenting
the voltage drop
study acceptable.
The
LC protection for the
HCC feeder circuit breakers
was
set to permit simultaneous restart of all motor loads with all non-motor loads
energized
in accordance
with design-basis
calculation "Electrical Protective
Devices - DBC-1."
2.5
120-Vac Class
1E System
The Class
1E instrument circuits required for post-LOCA monitoring are supplied
from recently installed Class lE 120-Vac Topaz inverters that are backed
by the
Class
1E batteries.
Instrumentation
and control (ISC) load groups of the
engineered
safety features
system
and other safety-related
systems
are fed from
four independent,
electrically and physically isolated Class
lE 120-Vac buses.
These
buses
are not supported
by battery-backed
inverters; if a,loss of offsite
power occurred,
the instruments
would be disabled until powered
by a Class
1E
emergency
diesel generator
(EDG).
In the case of a
LOOP, components
of each
instrument
loop could fail either at the high- or low-side of their range.
The
team noted that spurious actions of equipment
could occur because
of differ-
ences
in reset times for each
component of the loop.
Although effects of such
a scenario
were not analyzed
by the licensee,
safety
loops have
been tested
by
actually shutting off the power during each refueling outage
and
no spurious
actions
were noted during the performance of this testing.
The licensee
used the
DAPPER computer
program to analyze voltages in the
120-Vac system.
The program was used for voltage calculations of portions of
. the distribution network connecting
the 120-Vac buses to various
load groups
ard individual loads. 'he available-voltage
value used
as input for the
DAPPER
calculation
was the result of another voltage study and
a
2 percent correction
factor was used
because
of calculational uncertainties.
The inputs for load
data
and for bus data were taken from vendor supplied manuals for the applica-
ble equipment, while the input for the cable data
was derived from the station
raceway
schedule.
Actual cable lengths obtained through
raceway walkdowns were
used.
Resistances
of fuses
and breakers
were neglected.
Results of the
DAPPER study were validated
by the licensee
against
long-hand
calculations
and were found to be within a margin of 0.2 percent.
An evalua-
tion of the
DAPPER study results
indicated that
some
120-Vac loads
may not be
supplied with adequate
voltage under all design-bases
conditions.
As a result,
the licensee is planning to install 120-Vac regulating transformers
to bolster
the available voltage.
In addition, the licensee is currently evaluating-
changes
to the degraded grid relay setpoints
(see Section 2.10 of this report)
which should also serve to ensure
adequate
voltage at the 120-Vac level.
2.6
250/125-Vdc Class lE System
The 125-Ydc system
has four independent
channel
bus arrangements,
each supplied
from a Class
1E battery
connected
to a Class
1E charger.
These four redundant
and independent
buses
are grouped into two power trains.
Loads
can be trans-
ferred from one channel of the 125-Ydc system to another
channel of the
same
division during service testing of the battery and/or testing of the charger
during
a refueling outage.
During normal operation,
no two battery
systems
are
connected to each other.
The 250-Vdc system
has
two independent
redundant
bus
arrangements,
with each
bus connected
to a Class
1E 250-Vdc battery
and charger
unit.
The specific aspects
of the 250/125-Vdc Class IE system review are
discussed
below.
2.6.1
Voltage Drop
The voltage drop calculations for the 125-Vdc system consist of two parts:
The
first part determined
the voltage drop between
the battery
and the load center,
and the second part calculated
the voltage drop for the circuit network between
the load center
and the Class
1E distribution panel(s).
The resulting values
of the voltages
on the Class
1E distribution panel
were compared to the desired
minimum values required to achieve acceptable
voltages at the terminals of the
Class
lE devices.
Calculation E-AAA-391, Revision 1, "Unit 1 125 Vdc system,"
used conductor
resistance
values derived from tables of the National Electric Code (1990)
and
compensated for a maximum ambient temperature
of 50'C.
Actual measured
cable
lengths
were used
and the end of life (EOL) battery voltage of 105 volts was
used
as the lowest terminal voltage.
Results of these calculations
indicated
that the available
minimum voltage was inadequate for some loads.
To resolve
this problem, the licensee
changed
the
EOL battery voltage from 105 volts to
109.2 volts, and the plant maintenance
group was instructed to include the
new
EOL value in the acceptance
criteria portion of the station battery test
procedures.
The calculations
were recently performed only for the A battery;
however,
the licensee
has
a program to perform such calculations for the
remaining batteries
in the near future.
Voltage drop calculations for the 250-Ydc circuits also consisted of two parts.
The first part computed
the voltage drop between
the Class
1E batteries
and the
250-Vdc motor control centers
(liCCs).
Calculation E-AAA-255, Revision 0,
July 19, 1990,
showed that
a minimum voltage of 210 volts was required at the
HCCs to provide adequate
voltage at the load terminals.
The second part of the
voltage drop was originally done by the architect engineer
who assumed
a
maximum voltage drop of 20 percent
between
the
MCCs and the connected
loads
and
calculated
the maximum cable length of the connecting
cable to achieve the
~ ~
20-percent
drop.
The installed lengths of the cables
were compared to the
allowed maximum length to verify that the installed lengths
were either less
than or equal to the calculated
maximum allowable length.
Calculation E2015.03,
Revision 3, Hay 17, 1983,
"250-Vdc Motor Feeder
Cable
Sizing," and Calculation E2015.01,
Revision 2, April 8, 1983, "Available Fault
and Voltage Drop at Station 125- and 250-Vdc Systems,"
indicated that there
was
no margin between
the calculated
maximum cable length
and the installed
cable length for motor-operated
valve
(NOV) HV-E51-2F013.
In addition,
a
nonconservative
resistance
correction factor of 40'C was used instead of 50'C.
Using the correct value of maximum ambient temperature,
the minimum terminal
voltage at this valve would be lower than that required by the original design.
As a result, the licensee
performed
a
new calculation which demonstrated,
on
the basis of motor current, that the operating torque developed
by the operator
motor at this reduced voltage would be adequate.
The licensee
informed the team that it is aware of problems resulting from
various inconsistencies
in its calculations
and is in the process of recalcu-
lating voltages
using proper temperature for resistance
correction, actual
installed cable lengths,
and
a higher
EOL voltage for the 250-volt batteries.
The current
EOL voltage for the 250-volt batteries is 210 volts, which is
inadequate
to maintain 210 volts at the NCCs.
The voltage drop calculation
indicated that the
EOL voltage of these batteries
must be raised to a minimum
of 220.8 volts to.maintain
210 volts at the YiCCs.
In order to maintain the
battery voltage at 220.8 volts during accident conditions,
some
loads that are
not designated
Class
1E are required to be stripped.
Although the licensee
had
revised
Emergency
Operating
Procedure
EO-100-30 accordingly, the acceptance
criteria for the battery surveillance testing procedures
had not been revised
to change
the value of the minimum acceptable
terminal voltage to 220.8 volts.
The licensee
said that procedures for the surveillance test would be revised in
the near future to include an
EOL voltage greater
than or equal to 220.8 volts.
2.6.2
Battery Sizing/Short Circuit Study
The licensee is currently in the process
of revising old calculations
using
more conservative
bases.
At the time of the inspection,
the licensee
had
completed the revision for the Unit 1 125-Vdc battery
A (Calculation E-AAA-391,
Revision 1) and planned to revise similar calculations for all the remaining
125-Vdc and 250-Vdc battery/charger
combinations.
The calculation for the
A battery
has
a 6.72-percent
margin for its worst-case
loading.
A minimum
electrolyte temperature
equal to 60'F was used
as the basis for the sizing
calculation which was performed in accordance
with the guidelines of IEEE
Standard
485.
The duty cycle from the Technical Specification tables
was used,
which is more conservative
than the actual
duty cycle.
However, the actual
duty cycle calculation did not account for the nameplate
rating of the invert-
ers
and used
a value approximately
equal to 30 percent of the rating. If the
inverters are
loaded
beyond the assumed
30-percent
loading (by a high impedance
fault on its output side),
the condition would go unnoticed
because
there is no
indication or annunciation for such conditions.
The licensee
revised the
sizing calculation to account for the higher potential inverter loading which
reduced
the battery margin from 6.72 percent to approximately
4 percent.
The
team found the revised calculation acceptable.
The short circuit current calculation for the distt ibution network of the
125-Ydc battery
was performed using
a maximum electrolyte temperature
of 90'F,
and the actual
length of the cables.
The team found this approach
conservative
and acceptable.
However, fault calculations for the 250-Vdc distribution
network did not use the higher electrolyte temperature
or the actual
length of
the cables.
During the inspection,
the licensee
revised this calculation using
actual installed cable lengths
and
a temperature
of 25'C and demonstrated
that
the resulting values of short circuit current were still within the ratings of
the breakers
and buses.
2.6.3
DC System Overvoltage
During regular maintenance
and after periodic testing,
the station batteries
are
charged with an equalizing current while the batteries
are connected to the
dc distribution system.
On-line equalizing
imposes
a higher voltage
on equip-
ment connected
to the load side of the system.
The voltage could be as high as
142.8
Ydc for the 125-Vdc system
and 285.6
Vdc for the 250-Vdc system.
Overvoltage
could shorten
equipment life and could prevent the affected equip-
ment from performing its intended design function.
In 1985 the licensee
had
a
consultant
perform an evaluation of the effects of such overvoltages
on the
capability of Class
IE dc system equipment.
The study indicated that the
following equipment
could be affected
by overvoltages.
DC Cutler-Hammer Rela s:
Normally closed contacts of these relays
have
been
used
or
ypassing
the thermal overload trip contact of Class
1E
motor-operated
valve
(NOY) circuits.
These
normal,ly deenergized
relays
are energized
during periodic testing of NOYs to allow the valve motor to
be tripped in the event of an overload condition.
Because failure of this
relay at any time would keep the bypass active
and would not have
any
affect on the safe
shutdown capability of the systems,
the licensee
has
not taken
any corrective action for these relays.
250-Vdc
Pum
Motors:
The list of the affected motors consists
of Class
1E
motors
an
a
ew non-Class
1E motors.
These motors are nominally rated
for an operating voltage of 240 Vdc and could be degraded if operated with
an overvoltage condition for an extended duration.
As a corrective
action, the licensee
revised the
EOPs to restrict operation of these
motors during battery equalizing.
However, this measure
is not effective
for the 1P215 high-pressure
coolant injection and 1P220 reactor core
isolation cooling condenser
vacuum
pump motors which are started automati-
cally, therefore,
the licensee
intends to implement additional corrective
action for these motors.
To az Inverters:
These
Class
1E inverters supply 120-Yac power to instru-
ments require
for post-LOCA monitoring.
The inverters were specified for
an input voltage of 105 to 140 Ydc.
Higher input voltage could damage
and
disable the inverters.
In its
stu@
, the consultant
recommended
that the
maximum setting for the high voltage
be reduced
and annunciation for this
condition be provided.
The licensee
lowered the high-voltage setting but
did not provide any alarms to indicate tripping of inverters
as
a result
of overvoltage conditions.
Westin house
Rela
s MG-6
and AR:
These relays are used
as auxiliary
re ays in
C ass
1E contro
circu ts.
The study indicated that exposure to
over voltages
could shorten the life of these relays from 40 years to
17 years
and
recommended
replacement within this time.
The licensee
does
not have any program to track and replace
such relays, but intends to
issue
an engineering
work request
(EWR) in the near future to expand its
equipment qualification program to track and replace
the affected relays.
Circle Seal Solenoids:
Circle Seal
solenoid valves were specifically
manu acture
or 125-Vdc operation
and were qualified for a life of
40 years.
The maximum and minimum voltage ratings for these
solenoids is
plus or minus
10 percent,
and they were subjected
to a maximum of 130-Vdc
during qualification testing
by the valve vendor.
The overvoltage
evaluation
study indicated that these valves should be replaced.
The
licensee
has replaced
two of the three valves identified.
Since the third
valve would be required only after shutdown of the plant, the licensee
has
decided to replace the third valve at
a later date.
Backdraft Isolation
Dam ers:
The
ASCO solenoids for the isolation dampers
were qua i
e
or a maximum of 140 Vdc.
Nonconformance
Report
(NCR)84-994 dated August 14, 1984, indicates that these short-duty-rated
solenoids
are continuously energized
because
of their location and could
fail with no prior indication.
The study
recommends
replacement of these
solenoids.
The licensee
does not have
any program to replace
these
solenoids,
but intends to issue
an
EWR for replacement of the solenoids
for affected isolation dampers
in the near future.
2.6.4
Motor Control Center Circuit Breakers
During a walkdown, the team noted spare
breakers
in the 250/125-Vdc control
centers
and load centers
were left in the drawn-out position as
a permanent
arrangement.
This partially drawn-out position of the breakers
is a configu-
ration that was not analyzed for seismic forces.
In response
to the team's
concern,
the licensee
racked in all spare
breakers
in the respective
breaker
cubicles until further analysis is completed regarding
seismic qualifications
of the affected
load centers.
This item is identified as
Open Item 90-200-04
in Appendix A to this report.
The team also noted that 250-Vdc load center breaker
(GE AK 2-25) nameplate
rating is only 250-Vdc.
The team, raised
a concern regarding the qualification
of the breaker during higher voltage conditions of float/equalize voltage
(264-285
Vdc)'.
Subsequent
discussions
with the licensee
and with the manufac-
turer showed that the 250-Vdc nameplate
value is nominal,and the breaker is
rated for 300-Vdc maximum.
2.7
Protective Coordination
Adequate protection
and coordination
was generally found for the 480-Vac
circuits, including feeders for a non-Class
1E load connected
to a Class lE bus
and for a Class
1E load from a Class lE bus.
Coordination
between
the down-
stream breakers
and the upstream
load center breakers
and trip settings
was
adequate
for the non-Class
lE load turbine bui lding stack vent vacuum
pump
~
~
1P160.
Coordination of the downstream breakers,
the upstream
load center
10
~
~
b'reakers,
and with the 4160-Vac switchgear breakers
and the trip settings
was
acceptable for the Class
1E load to
RHR pump sunction shutoff valve
HV-E11-1F009.
The principles governing the settings of the electrical protective devices
were
listed in design basis calculation
DBC-1 "Electrical Protective Devices."
The
protective relays were generally set
so that normal load would not exceed
80 percent of the relay setting range, to allow for drift and other tolerances.
The protection
was set to stay within the long term and emergency rating of
cables
and transformers.
Motor overload relays,
where not blocked, were set
not to trip during acceleration
and operation of motors with terminal voltage
equal to 100 percent or 80 percent of nominal.
The relays were set to account
for both the reduced voltage
and the motors'ervice factor of 1.15.
Mhen a non-Class
lE load was energized
from a Class lE MCC there
was
no coordi-
nation between
the two breakers
in series [e.g., in MCC 1B216 breakers
81 (1E)
and
52A (non-lE)], but there
was full coordination
among either of the two
breakers
and the
LC breaker feeding the
MCC, which agreed with FSAR Section
8.1.6.1.H.5
and Calculation
DBC-1.
The ground fault protection of the 480-Vac lE system
had been blocked
due to a
lack of coordination
between various protective devices.
The 480-Vac distribu-
tion system
had
a solidly grounded neutral.
However, the circuit breakers
(CBs) rated less
than
50 amperes
or motor starters
rated less
than the National
Electrical Manufacturers Association's
(NEMA) size
2 in the MCCs, did not have
a ground fault protection (device 50/G),
and the magnetic trip units on these
CBs had been set higher than the setting of the ground fault protection of the
LC CB feeding the
MCC.
Since this had effectively eliminated the possibility
of maintaining coordination
between
the two protective elements,
the ground
fault protection in the
LC had been blocked.
This created
a situation where
the fault on loads fed by those
flCC circuits must be cleared
by the overload
protection device, or the magnetic trip element in the
CB, after the fault
persisted
long enough to change
from phase-to-ground
to phase-to-phase.
However, this may not work in case of an arcing ground fault with the fault
current lower than the setting of the CB's magnetic trip.
The licensee indi-
cated that the blocking of the ground fault protection at the
LC level was the
original decision taken
as part of a policy that low voltage ground fault
protection
was
done primarily for economic reasons.
The team accepted
the
licensee's
position on this point.
2.8
120-Vac lE Distribution System
t)o deficiencies
were found in the protective coordination of the 120-Vac
Class
1E system
and the control power circuits wer e adequate.
The portion of
the 120-Vac Class lE system
used in control
and annunciation circuits is
supplied from a 480-Vac Class lE MCC through
a 480-208/120-Vac transformer
and
is an interruptible system.
The 120-Vac single-phase
system
was protected with
circuit breakers
in the main power distribution panels.
The circuit breakers
supplied
power to fuse distribution panels.
The system
was arranged hierar-
chically with up to four protective devices in-line from the main distribution
panel to the individual control and annunciation circuits.
The protective
devices
were selected
in a sequential
manner (i.e.,
a 20-ampere circuit breaker
fed
a circuit with a 10-ampere
fuse that fed a number of circuits with 6-ampere
11
fuses that fed a number of circuits with 3-ampere fuses.)
The. licensee indi-
cated that there were
no requirements for, and
no coordination between,
those
fuses for response
to fault currents.
The criteria for selecting
the fuses
had
been strictly for the protection of circuit integrity during overload condi-
tions.
Additional fuses
were installed in the grounded neutral wires for
circuits going through containment penetrations
to provide redundant
overload
protection.
Bechtel Calculations
E2010.04,
Revision 1, and E2010.05,
Revision 0, provided
design
guides for selecting
maximum allowable length of wires
(AWG 14)
and size
of control transformers
(depending
on the
NEHA starter size
and the number of
relays) in the
HCC control circuits.
The wires and control transformers
were
selected
to maintain the voltage at above the minimum pickup voltage of 102-Vac
for all devices.
Calculation E2010.04 allowed
a 20-percent
margin in the
length of the control circuit.
Host, but not all, of the lengths of the wires
in the control circuits had been verified to an as-built condition.
Even
though the design guides did not provide verification of the actual design,
this approach of documenting
the sizing of the control transformers
and voltage
drop study was acceptable.
The licensee
had
a procedure to verify the length
of wires in a control circuit whenever the circuit was modified.
The control transformers
selected
using Calculation E2010.05
had sufficient
capacity.
The criteria listed in the calculations
were valid for steady state
conditions.
During transients
such
as starting of large loads,
the
HCC bus
voltage could dip below 432-Vac and the control circuit voltage could be below
the relays
and contactors
pickup voltage; therefore, it would not be possible
to successfully start
HCC loads until the voltage recovered
to the 432-Vac
level.
The licensee
stated that sufficient indication was available in the
control
room to alert operators of such
a condition and that the
HCC loads
would either automatically attempt to restart or would be restarted
by an
operator after the bus voltage
had recovered.
This explanation
was accepted
by
the team.
2.9
Sizing
The 4160-Vac system penetration protection for the reactor recirculation
pumps
feeder
and the 480-Vac system penetration
protection for the hydrogen
recombiner were reviewed
by the inspection
team.
The values of maximum short
circuit currents
and settings of breaker trips for these penetrations
were
acceptable
to the team for penetration protection.
2.10
Degraded
Voltage and Loss of Voltage Relays
The Susquehanna
degraded grid and loss of voltage protective
scheme
consists of
different levels of protection to preclude
equipment
damage
and to transfer
Class
1E loads to the emergency
power supply whenever the available voltage to
the Class
1E 4160-Vac buses is not sufficient to power the required safety
loads.
Each 4160-Vac bus is provided with undervoltage
relay protection to
either transfer
loads to the alternate
power source or to the emergency
diesel
generators
(EDGs).
The following relays are used to provide these
levels of
protection:
12
~
~
~
Undervoltage relay 27AI is set at 96.5 percent of the nominal
bus voltage
to monitor the availability of the incoming offsite power supply.
This
relay also is used to initiate an alarm after
a 10-second
delay.
Undervoltage relay 27A initiates
a bus transfer to the alternate
power
source
on a loss of voltage to the 4160-Vac buses.
It is set to drop out
at 20 percent of the rated
bus voltage.
If the alternate
power source is
not available,
the
EOG is automatically started after
a 0.5-second
delay.
27B1, 27B2, 27B3, and 27B4 provide backup protection
for initiating bus transfers
and undervo'itage
alarms if degraded
voltage
conditions occur.
Devices 27Bl and 27B2 are set to drop out at 84 percent of rated
bus voltage
and initiate an alarm after
a 10-second
delay.
If the degraded
condition
exists with no
LOCA signal present,
the relays will initiate a bus transfer
after a 5-minute delay.
The delay is provided to allow sufficient time for
operator action to restore
the voltage to acceptable
levels. If the degraded
condition exists with a
LOCA signal present,
the relays will initiate a bus
transfer after
a 10-second
delay.
The 10-second
delay is provided to prevent
motor starting transients
from initiating undesired
bus transfers.
Devices
27B3 and 27B4 are set to drop out at 65 percent of rated
bus voltage
and initiate a bus transfer after
a 3-second
delay.
The setting
and delay is
provided to prevent
bus transfers
during voltage dips that can result from
fault currents that
can occur prior to overcurrent relay operations.
Degraded voltage and loss of voltage protection
schemes
are provided to prevent
spurious trips of the offsite power supplies,
to provide protection for equip-
ment operating at low voltages
so that the equipment will not be damaged,
and
to ensure that equipment
can operate at the lower voltages to perform its
design functions during design-basis
events.
The undervoltage
setpoints
determine
the level of reduced voltage at which the 4160-Vac buses
are isolated
from the offsite power system
and are connected to the
EDGs.
Therefore,
the
relay settings
should be set
so that the safety-related
equipment is able to
perform its intended functions under conditions that are above the undervoltage
setpoints.
During a previous inspection,
the
NRC noted that the licensee's
protective
scheme for the 4160-Yac buses
allows the system voltage to be
significantly lower than the nominal value for the buses without the
relays actuating (i.e., at 84 percent).
The licensee
subsequently
determined that approximately
93 percent of rated
bus voltage must be available
at the 4160-Vac buses to ensure that all safety-related
equipment
has suffi-
cient voltage to operate.
Therefore,
the current relay setpoints
do not
provide sufficient protection to ensure that all equipment will operate
during
degraded
voltage conditions during which the available voltage is less than
93
percent but greater
than 84 percent.
The licensee's
original basis for these
relay setpoints
was based
on the assumption that all postulated
system distur-
bances either at the grid or within the plant would result in actuating
the
protective
scheme relays.
Therefore,
considering
the desire to minimize the
probability of spurious trips of the offsite power supplies,
the relay
setpoints
were
chosen at the present
value.
13
~
~
In Hay of 1990, the licensee identified a condition in which a single failure
~
~
~
~
~
~
~
~
~
~
~
~
during
a
LOCA could result in degraded
system voltages without the undervoltage
protection relays actuating.
This condition could occur
when all Class
1E
buses
are supplied from one startup bus.
The licensee
evaluated
the postulated
event
and concluded that the probability of the event was minimal because
the
Susquehanna
power system is very stable.
Furthermore,
the'icensee
contended
that the postulated
single, failure had to occur during
a very narrow timeframe
at the beginning of a
LOCA to cause
a degraded
voltage condition and the
probability of this event was very low; therefore,
no immediate action was
necessary
and the condition was determined
not to be reportable to the
NRC
under the criteria of 10
CFR 50.72 or 50.73.
During this inspection,
the team
became
aware of the significant number of Class lE loads at the 480-Vac and
120-Vac buses that would not operate
dur ing the postulated
event.
The licensee
was requested
to provide
a safety
assessment
addressing
the adequacy of the
current undervoltage
protective
scheme to enable
equipment to perform its
intended functions.
The licensee
stated that
a detailed study was in progress
and that the following actions
are planned in the interim to minimize the
probability of being in a sustained
degraded
voltage condition:
The dropout setpoints for 27Bl and
27B2 undervoltage
relays will be
increased
from 84 percent to a yet undetermined
value.
The 27B3 and 27B4
dropout setpoints
also will be increased
from 65 percent to a yet undeter-
mined value.
These setpoints will be based
on the results of an ongoing
voltage study.
The time delays for these relays are expected to remain
the
same.
The licensee
committed to submit
a technical specification
change
request
to the
NRC by September
30, 1990.
Operation procedures
have been revised to incorporate additional operator
actions to be taken if the 96.5-percent
alarm is received.
These actions include:
If the undervoltage
alarm is received
and the alternate
power supply
is available,
the operators
wi 11 initiate a transfer to the alternate
source.
If the undervoltage
alarm is received
and the alternate
power supply
also is in a degraded
condition, the operators will initiate manual
start of the
EDG.
After the
EDG has reached
rated
speed
and voltage,
the operators will initiate a loss of voltage to the affected
4160-Vac buses
by tripping the normal feed circuit .breakers,
which
will result in the
EDG automatically closing its output circuit
breaker onto the 4160-Vac bus thereby restoring
power to the bus.
The above operator actions
were incorporated into operator
procedures
during
this inspection.
Though these corrective actions collectively provide greater
assurance
that
equipment will be able to perform their intended functions with a stable
power
supply from either offsite or emergency
sources, it does require operator
actions,
which is an undesirable
method to ensure
equipment functionality upon
14
receipt of the
LOCA signal.
This item is identified as Unresolved
~
~
Item 90-200-05
in, Appendix A to this report.
3.0
MECHANICAL DESIGN
The team review included
a walkdown of the reactor building heating, ventila-
tion, and air conditioning
(HVAC) system
and detailed review of engineering,
licensing,
and plant operations
documents
associated
with mechanical
systems
in
support of the
EDS, including the following:
The
HVAC systems that provide the required operating
environment for the
safety-related
equipment.
This included local air coolers for high-
pressure
coolant injection (HPCI) and core spray
pumps,
the ventilators
for the diesel
generator
rooms
and the pumphouse
[housing the essential
(ESW) and residual
heat
removal service water
(RHRSW)
pumps],
and the battery
room and emergency
switchgear
room
(ESWGR)
systems
including the interfacing control structure chilled water system
and the dedicated direct expansion
(DX) units for the Unit 2
ESWGR HYAC.
The power demands for major loads required
by the core spray,
ESW,
residual
heat removal
(RHR), and
RHRSW pumps following a
LOOP.
The ability of the
ESW to provide cooling to safety-related
equipment
following a design-basis
event.
The ability of the diesel generators
to provide the standby
power supply
required to operate
the safety-related
equipment
needed for a safe plant
shutdown following a design-basis
event, including the following auxiliary
systems:
fuel oil supply
lube oil supply
jacket water cooling
air start
combustion air intercooler
In addition, the assumptions,
input data,
design bases,
methodology,
and output
results of selected
calculations
were spot
checked for consistency
between
design
documents
and the thoroughness
of the major electrical
loads
on the
diesel
generator
buses
and
compared to the load lists in the design
documentation.
3.1
Heating, Ventilation, and Air Conditioning Systems
The licensee
found that
a single failure of an
NCC could prevent all venti la-
tors in one train of the
ESW pumphouse.from starting, which would prevent
cooling for the
ESW and two
RHRSW pumps.
However, further analysis
indicated
that if the operator were to manually open the pumphouse
louvre vents within
5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> of the event, natural circulation would remove
enough heat to prevent
motor failures, but the resulting
pumphouse
temperature
would exceed
the limit
in the
FSAR.
Therefore,
pending completion of the Engineering
Work Request
(EWR) H80540, the licensee
plans to update the design documentation
and
FSAR.
15
During a walkdown of local
room coolers,
the team discovered that the thermal
insulation
had been
removed from the
HPCI pump, booster
pump,
and crossover
piping.
The licensee
determined that this insulation was
removed for mainte-
nance in July of 1984
and inadvertently
was not reinstalled.
In addition, the
calculation of the heat
load for the
pump room had not been revised to
analyze the effect of the additional heat load.
This item is identified as
Unresolved
Item 90-200-06 in Appendix A to this report.
Limiting conditions for operation
(LCO) are not delineated for HVAC systems
because
these
systems
are not explicitly addressed
in the Technical Specifica-
tion.
The unavailability of an
HVAC system could limit the ability of the
associated
safety-related
equipment to perform its function.
During a walkdown
of the Unit 2
HVAC equipment,
the team noted that the loop
B direct
expansion unit was out of service for maintenance.
Although only one division
was available,
no specific outage times, limiting conditions, or compensatory
measures
had been established.
In addition, although the unavailability of the
reactor building circulation fans could impair the performance of the standby
gas treatment
system,
no procedures
had been generated
which would govern the
acceptable
outage
times for these
components.
The team considered that appt o-
priate procedures
should be provided for the acceptable
outage
times of one or
both trains of support systems,
which are not specifically addressed
by the
plant technical specifications.
During the inspection,
the licensee
provided
Technical Specification Interpretations
(TSI) for the
ESSM ventilation and
Control Structure
HVAC.
A team work plan was also provided which forms the
basis for an internal process
of prioritizing work on critical equipment.
This
is the procedure
used to expedite work on the non-technical
specification
equipment listed above until the TSIs, which are currently under development,
are completed
and issued.
3.2
Power Demands for Major Loads
Although the power demands of the major pump motors powered
by the diesel
generators
following a
LOOP were generally
adequate for the core spray,
ESW,
RHR, and
RHRSW pumps,
the following concerns
were noted.
There are four ESM
pumps,
each of which deliver 50 percent of the required flow.
Two pumps are in
each division of the
ESM system.
Each
ESW pump is powered from one of the four
diesel generators.
The latest revision of FSAR Table 9.2-3 indicates that when
both
pumps are operating in parallel, the system flow is 9800
gpm (4900
gpm per
pump).
Each
pump requires
410 bhp in this mode of operation.
If only one of
the two
ESW pumps in a loop starts,
the flow through the
pump increases
to
approximately
7600
gpm.
The
pump power is about
503 bhp under these
condi-
tions, exceeding
the 450 bhp (rated motor power)
assumed for the diesel
genera-.
tor loading.
Consequently,
the single failure of one
ESM pump could cause
increased
loading on the diesel generator
which supplies
the other
ESW pump.
Although the team felt this oversight should be corrected in the
FSAR diesel
generator
loading tables,
the power demands for other loads were sufficiently
conservative to compensate
for this oversight.
The second
concern
was that
with three out of four ESM pumps operating,
a required
mode of operation,
the
motor is at 110 percent of its rated capacity of 450 bhp.
Although this
operating
mode is considered
marginal, it is acceptable
because
the motor is
designed with a minimum 15 percent
margin and the motor protective devices
are
16
~
~
set at
a minimum of 132 percent of the maximum load at rated voltage
and are
not expected to trip under this load.
The four core spray
pumps are arranged
in two loops, with two pumps that can
each deliver 50 percent of required flow in each
loop. If only one
pump in a
loop starts, it is estimated that the
pump would produce approximately
150 percent of its normal flow.
The net positive suction
head
(NPSH) required
by the
pump increases
to about
17 feet, which exceeds
the
HPSH available in the
system.
Operation of this large
pump (700 bhp) under cavitating conditions
could result in excessive
pump vibration and extensive
damage to the piping
supports,
and the potential for damage is increased
by the relatively low
operating
temperatures
of less than 250'F.
However, Operating
Procedure
OP-151-001
does caution the operator against the high system flow rates
and
instructs the operator to throttle the
pump discharge
valve to limit system
flows to 7900
gpm with two pumps
and 3950
gpm with one
pump per loop.
The team
agreed that the damage, if any, would be limited to one loop only, which is
within the design basis of the plant.
3.3
Emergency Service Water System
The
ESW system
draws water from a reservoir .to provide cooling to Class
1E
equipment.
Little can be done to control the water chemistry in an open-loop
cooling water system.
Furthermore,
the
ESW system is normally in a standby
condition; therefore,
sedimentation
can occur
as
a result of stagnant
cooling
water left in the piping and heat exchangers.
As a result,
excessive
fouling
of the heat transfer surface
(0.003 hr-ft'-'F/Btu versus
0.002
as originally
specified)
and tube plugging of tubes
(due to wall thinning caused
by erosion,
corrosion,
and buildup of sediments),
can decrease
the capacities
of the heat
transfer equipment serviced
by ESW.
The licensee
has evaluated
the maximum allowable
number of plugged tubes in
procedure
t31453
on the basis of the actual
heat exchanger
performance at the
site.
When the number of plugged tubes for the
D emergency
diesel generator
jacket water cooler
exceeded
the allowable limit of 4 percent,
the licensee
issued
Nonconformance
Report
HCR 90-0117
and stated that the tube bundle will
be replaced
next year (1991).
The licensee
did not evaluate
the allowable
number of plugged holes for the condensers
that were purchased
as part of a
commercial
package to account for the increased fouling, which was in excess of
the fouling assumed
in the original design.
Increased fouling is of particular
concern in applications with two-phase
heat transfer in which fouling can be
the dominant parameter
in the relatively high overall heat transfer coeffi-
cient.
Currently,
5 to 6 percent of the tubes in the
CSCW system
condenser
are
plugged,
and 4 percent of the tubes in the direct expansion unit condenser for
the Unit 2
HVAC system are plugged.
The effects of increased fouling and
the reduction of available surface
area
as
a result of tube plugging could
affect the operability of these
components
and hence the operability of the
Class
lE equipment
supported
by them.
The licensee
has
committed to evaluating
the surface
area available in the
CSCW
system
condenser
and assessing
the operability of the equipment affected
by the
number of tubes
plugged to date.
In addition, if the assessment
of CSCW
condenser
proves to be marginal or unacceptable,
the licensee
committed to
perform a similar assessment
of the equipment affected
by the direct expansion
17
unit condenser for the Unit 2
ESMGR HVAC.
This item is identified as
Unresolved
Item 90-200-07 in Appendix A to this report.
The sections of ESW piping between
the supply
and return isolation valves to
each diesel generator
and the secondary
side of the emergency
diesel
generator
(EDG) auxiliary coolers (intercoolers,
jacket water cooler,
lube oil coolers,
fuel oil cooler) are not protected
from overpressurization
as required
by the
ASHE Code Section III, ND-7000.
The
ESM system
has capacity to provide cooling
to four of the five EDGs.
As
a result, the standby unit is isolated from main
ESM piping and the isolation valves are normally closed.
Under the following
operating conditions,
energy is added to the isolated section of the system
and
overpressurization
could occur.
To minimize the time required to place
an
EDG back in service,
the standby
lube oil and jacket water heaters
and circulating
pumps are operated
to
prewarm the equipment while the
ESW valves are closed to isolate the
cooling water.
Overpressurization
of that section of the piping has the potential to impair
both loops of the
ESW system since
common connections
from both the A and
B
loops are provided to each diesel generator.
The license
has committed to
performing procedural
or hardware modifications as necessary
to eliminate the
overpressure
concerns.
This item is identified as
Open Item 90-200-08 in
Appendix A to this report.
The
FSAR states that the two independent
loops of ESW are separated
by barriers
and trenches.
The PAID shows
two pipe trenches,
each with a supply and return
line for one loop, running outside the diesel
generator
rooms
and branch lines
from each
loop entering
each diesel generator
room.
However, during
a walkdown
of the diesel generator
rooms, the team noted that this
1oop separation
had not
been provided.
The four pipes
(two supply and two return lines) run beside
each other from one diesel generator
room to the other in the basement of the
diesel generator
rooms.
The licensee
stated that the design documentation
and
FSAR would be updated to reflect the as-built configuration.
3.4
Diesel Generator
and Auxiliary Systems
An actuated
control valve with a temperature
feedback
loop was recently
added
to the
ESW return line from the diesel generator intercooler.
A downstream
butterfly valve is used
as
a throttling valve.
The butterfly valve flapper is
fixed in a permanent position to ensure that other
ESW loads are not affected
if the temperature
control valve fails open.
During
a walkdown of this modifi-
cation, the team heard cavitation noise in the piping near
the throttling
valve.
The vibration and erosion associated
with the cavitation process
could
result in potential failures in these
valves
and downstream piping.
The
licensee
is planning to disassemb'le
this section of piping and to inspect the
temperature
control valves
(TV-01124A-E), the butterfly valves
(011042, 44, 46,
68,
and 011509),
and the associated
downstream piping for indications of
excessive
erosion
and to inspect the butterfly valve stem for fatigue damage.
18
The licensee
was in the process
of revising the Technical Specifications
and
associated
operating
and surveillance
procedures
to ensure
the instrumentation
setpoints
and normal
day tank fuel levels were in accordance
with NRC
Regulatory
Guide 1.137
and ANSI Standard
N195
(ANS 59.51).
The ANSI standard
requires that the day tank have
enough fuel below the fuel transfer
pump start
level setpoint to allow the diesel generator to operate for a minimum of 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />
at 110 percent of its rated capacity.
The team identified two areas of concern
with the calculations for the level setpoints.
The required fuel capacities
are based
on
a specific fuel consumption rate
(BTU/BHP-hr) as verified by the manufacturer's
test results.
The quantity of
chemical
energy required for 1-hour operation at 110 percent rated capacity is
converted to an equivalent
volume that can be verified in the field.
The
licensee
advised that the fuel oil can range in specific gravity from 0.82 to
0.95, with an associated
range of heat content,
BTU/gallon.
The day tank
capacities
have
been calculated
using
a median specific gravity of 0.87 which
is not conservative.
Furthermore,
the quoted specific gravities are for fuel
at 60'F.
Accounting for thermal
expansion of the fuel to typical operating
room temperatures
(100-105'F)
could require another
2 percent in the required
capacity.
The licensee
committed to a thorough review of the day tank calcula-
tions and setpoints
to resolve these
issues.
During
a walkdown of the five diesel generators,
the team identified inconsis-
tencies
on the nameplates
of the diesel
motors
and the generators.
For exam-
ple, the 2000-hour rating was not given on the nameplates for diesel
generators
A through
D.
The Instruction and Operating
Manual states
the
2000-hour rating is 4700
kW.
The licensee
was able to confirm with the manu-
facturer that the peaking
power for the A through
D diesel
motors is equivalent
to the 2000-hour rating.
The manufacturer also provided information to justify
the 4700-kM 2000-hour rating for generators
A through
D.
Although the name-
plate data
was incomplete
and did not reflect the design requirements for the
equipment,
the actual capabilities
of the components
were satisfactory.
The
FSAR shows diesel generator
loading following an accident
as high as
4400
kW, (which is 110 percent of the rated capacity of diesels
A through D).
The team felt that the intent of the day tank sizing criteria in ANSI Standard
N195 was not satisfied
by calculating the fuel requirements
based
on the
nameplate rating when the normal operating condition exceeds
the rated value.
The team expressed
their opinion that the 2000-hour rating (i.e., the nominal
rating applicable for the post-accident
operating conditions) or, as
a minimum,
the maximum diesel
loading to be sustained for at least
1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, should be used
instead of the nameplate rating for determining the day tank capacity.
Although the team disagreed
with the way the diesel
day tank levels were deter-
mined,
no specific safety concern
could be identified with this issue.
The Technical Specification
bases for diesel fuel testing includes the guidance
of Regulatory
Guide 1.137.
Regulatory
Guide 1.137 states that prior to adding
fuel to the fuel storage
tanks, testing should
be performed to evaluate
the
fuels specific gravity, water content,
sediment content,
and viscosity.
The
licensee
s technical specification (4.8.1.1)
and diesel fuel oil handling
procedures
did not include the requirement for testing the
specific gravity of the sample.
The licensee
has agreed to update the testing
requirements.
19
The portion of the diesel
generator
fuel supply system below ground level and
in the diesel generator
room is seismically qualified and protected
from damage
by a tornado.
However, the fuel storage
tank vent and its associated
flame
arrester is not qualified for these
design-basis
events.
There are two failure
modes of concern:
1.
If the vent is crushed,
a vacuum would be created
in the tank as fuel is
transferred
to the day tank.
As the available
NPSH falls, the transfer
pump cavitates
preventing further flow of fuel to the engine.
2.
If the vent is broken off, debris would enter the opening into the tank
and block the transfer
pump inlet.
In addition, there would be
a poten-
tial fire/explosion hazard
because
of the loss of the flame arrester.
In response
to this concern,
the licensee
said that the overflow line from the
day tank to the storage
tank would provide
a vent for the storage tank.
In the
case of diesel generator
E, there is a 2-foot trap in the overflow line.
Even
with the partial
vacuum required in the storage
tank to clear this trapped
fuel, there is sufficient NPSH available for continued transfer
pump operation.
The licensee believed that the probabi lity of debris entering the storage
tank
through
a broken vent stack
was very low and the layout of pump suction line
would inhibit debris from enter ing the
pump suction line, which is further
protected
by a strainer.
In addition,
a temporary flame arrester
would be
installed over the opening before trucks or other ignition sources
were allowed
in the vicinity of the storage
tanks.
The team found this explanation
acceptab le.
Contrary to the
FSAR, the level indication and low level annunciators for the
diesel fuel oil storage
tanks are not being continuously monitored or indicated
in the diesel generator building.
The licensee is currently using
a dip-stick
method to check level in the five diesel fuel oil tanks monthly and following
each diesel start.
This item identified as Unresolved
Item 90-200-09
and is
discussed
in further detail in Appendix A.
3.5
Fire Protection
System
The fire protection
system is not seismically qualified.
Although the sprin-
kler pipes are normally air filled, the deluge valves
and related controls are
not seismically qualified and can be assumed
to fail.
Therefore,
nonqualified
sprinkler systems
could spray Class
1E electrical
equipment following a seismic
event.
The team requested
that the licensee
demonstrate
that the fire protec-
tion system would not impair the operation of the diesel generators
following a
seismic event.
The licensee
provided
a summary sheet of the activities conducted
in accordance
with Bechtel Specification 8856-f)-388, which was
a guide for assessing
the"
effect of failures of equipment that is not safety-related
on Class
1E equip-
ment.
The diesel generator
room signoff sheet stated that the fire protection
system
was designed
and installed in accordance
with Bechtel Specification
N-343, which thus ensured
the integrity of the piping.
However, the team
found, that the spacing of many pipe supports
exceeded
the allowable spans
given in Appendix
B of the specification
by 50 to 100 percent.
The licensee
performed calculations for these
nonconforming sections of pipe, taking into
20
I
0
account the water filled pipes,
the stress
intensification for threaded joints,
the seismic accelerations
with 2 percent
damping,
and the as-built spacing.
These calculations
indicated the stresses
to be below the maximum allowable.
The team was concerned that the guideline practices
used to design
a system
were not followed for many of these
pipe supports.
The licensee
confirmed that
an engineering
discrepancy
report
(EDR G00154)
had been
issued to revise the
specifications
and drawings to reflect the as-built layout.
This item is
identified as Unresolved
Item 90-200-10 in Appendix A to this report.
4.0
ELECTRICAL EQUIPMENT TESTING AND SURVEILLANCE
As part of its overall evaluation of the electrical distribution system,
the
team performed
a review of selected
testing
and surveillance
procedures
related
to the diesel generators,
the batteries,
the Class
1E inverters, circuit
breakers,
and relays.
In addition,
a review was conducted of the licensee's
fuse control.
Details of these
reviews are contained
below.
4.1
Diesel Generator Testing
The diesel generator testing procedure for classifying starts
as valid or not
valid did riot specifically address
the case of a maintenance
run.
The licensee
currently classifies
successful
maintenance
runs
as valid tests to reduce the
number of starts that the diesel is subjected to.
This is acceptable
as
long
as failures of maintenance
runs are also classified
as valid tests if the
failure results from a problem not related to the maintenance activity and if
the failure would have prevented
the diesel
from performing its emergency
function.
A failure of a maintenance
run should not automatically
be classi-
fied as
a non-valid test simply because it was
a failure.
A review should
be
conducted of test failures of maintenance
runs
and otherwise to determine if
the failure was part of a valid test or not valid, and this review should
be
adequately
documented
in the diesel start logs.
The licensee
has
agreed to
revise its diesel test procedures
to clarify this issue.
The readings of insulation resistance
and polarization index from insulation
resistance
tests
on the generator windings were compared to the minimum accept-
able values
and recorded.
However, there
was
no trending of the results,
which
could be beneficial to tracking insulation life deterioration
and predicting
remaining life.
4.2
Setpoint Calculation
and Control
The licensee's
program for establishing
and controlling safety-related
instru-
ment setpoints
is described
in Nuclear Department Instruction NDI-QA-15.2.9,
"Setpoint Selection
and Control." It delineates
the responsibilities of the
groups responsible for establishing
and controlling setpoints.
The nuclear
plant engineering
and system operation
groups
share
the responsibility for
establishing
setpoints for various instruments
depending
on the system voltage
level.
Setpoint
changes
are required to be evaluated for their effect on the
FSAR, Technical Specifications,
and design description
documents.
Setpoint
change
packages
(SCPs)
were developed
as
a means of controlling instrument
setpoint
changes.
Relaying setting
change notices
(RSCNs)
wer e the design
21
E
~
l
~
output documents that provide the controlling means for ensuring that setpoints
are maintained in accordance
with their bases.
Setpoint criteria for electrical devices
were established
and described
in
various relay section calculations
(DBC-1,2,3).
These calculations specified
the setpoint criteria for protective devices to protect electrical equipment,
such
as motors, timers,
and undervoltage
and overcurrent relays,
through
selective tripping of loads.
These criteria were used to establish
setpoints
in setpoint calculations.
The calculation results
were transferred
to relay
setting
change notices
(RSCNs) that are
used to maintain
and control instrument
setpoints.
Once
a device requires testing or calibration, the applicable
RSCN is reviewed
to note the previously calculated
required relay setpoint.
This information,
along with applicable tolerances,
is translated
into the applicable calibration
procedure
data sheet before calibration.
The data sheet requires that as-found
and as-left data
be recorded.
The team reviewed relay calibration records to verify that the device setpoints
conformed to the specified setpoints
in setpoint calculations.
The setpoints
were specified along with a tolerance
band for acceptable
values.
The as-left
setpoints
were within this specified tolerance
band.
However,
8 of 16 data
sheets
reviewed for both undervoltage
timers
and overcurrent relays indicated
that. the as-found settings
were outside the setpoint tolerance
band.
These
settings
ranged from+2.8 to -21 percent
beyond the tolerance
band.
Although
these
readings
were outside the specified tolerances,
there
was
no evaluation
to determine the cause for the instrument drifts.
The licensee
stated that
there
was
no formal program for evaluating
instrument drifts to identify
instruments that may require more frequent calibrations.
Additionally,
have not been
made for those instruments
in which
the excessive drifts could have
compromised their safety functions.
This item
is identified as Unresolved
Item 90-200-11 in Appendix A to this report.
The setpoint for all 27A undervoltage
relays is required to be 24 volts.
One
of the 27A relays installed in the Unit 2 4160-Vac bus 20104
was of a different
type and had
a manufacturer's
setpoint
band of 36-45 volts.
The licensee
installed this relay and adjusted
the setpoint to 24 volts.
The manufacturer's
setpoint
band envelopes
the settings at which the relay can be set
and still
perform as desired with accuracy
and repeatability.
Although the licensee
knew
the relay setpoint
was outside the manufacturer's
band,
the relay was evaluated
as being satisfactory.
The team asked the licensee
to provide an evaluation
for setting the relay outside the manufacturer's
specified setting
band.
Instead of providing an evaluation,
the licensee
replaced
the installed relay
with one with an appropriate
band of 18-30 volts.
Although this issue is
closed, it is identified as Unresolved
Item 90-200-12 in Appendix A to this
report.
4.3
Circuit Breaker Testing
The team reviewed the licensee's
program for periodic testing of Mestinghouse
4160-Vac
1200A switchgear breakers,
Brown Boveri
K600S 480 Vac load center
breakers
and
GE AK-2-25 250-Vdc load center breakers.
The team identified that
Procedure
NT-GE-006, "Load Center Breaker Relay Logic and Primary Current
Testing,"
does not provide adequate
information for testing
dc circuit breakers
and there is no acceptance
criteria specified in the procedure
except for a
22
data
sheet to record the test values.
During the review of test
documents for
250-Vdc breakers
72-66222
and 72-662013,
the team noted that the licensee
was
using
an ac test current source to test the dc circuit breakers.
Discussions
with the manufacturer
revealed that the breaker trip response
would be differ-
ent if they are tested with an ac current.
In response
to the team's
concern,
the licensee
sent
two sample
GE AK breakers
to the manufacturer's facility (GE)
for testing with ac and dc current.
The preliminary assessment
of the test
results
indicated that the time-current characteristics
of the breakers
varied
significantly between
the ac and dc tests.
This item is identified as Unre-
solved Item 90-200-12 in Appendix A to this report.
A complete overhaul of 4160-Vac breakers
is performed every
3 years in accor-
dance with Procedure
HT-GE-005, "Circuit Breaker
and Switchgear Inspection
and
Maintenance of 5 and
15
kV Breakers."
A routine relay logic and system'func-
tional test also is performed every refueling cycle.
This procedure reflects
all the necessary
maintenance
suggested
by the manufacturer.
However, the
existing maintenance
program does not require
any periodic overcurrent trip
testing of 480-Vac breakers
or requirements
to perform this test after
a
breaker
overhaul is performed.
The team noted that
a lack of maintenance
testing of circuit breakers
related to fire protection equipment
(10 CFR Part 50, Appendix R) was identified during
a previous
NRC inspection,
Inspection Report (88-21/88-24).
The licensee
stated that
a formal breaker
maintenance
testing
program to address all of load center
and motor control
center breakers
480-Vac and below is being developed to ensure
proper breaker
operation.
This program will determine
the appropriate
frequency of future
preventive maintenance
from trending past test results
and proposed
sample
tests.
It will also consider industry and manufacturer's
recommendations
and
procedures
to be revised accordingly.
A pilot test program consisting of
approximately
10 percent of Unit 1 and
common (to both units) Class
1E breakers
of 480-Vac and below is scheduled
to be conducted
during the next refueling
outage.
This program is scheduled
to be in place at the end of 1991.
4.4
Fuse Control
During a walkdown (control
room panel
IC614; dc panels
1D631, 632, 634, 661,
662 and 274; switchgear
1A203; load center
)B230 and motor control center
1B236), the team noted that the ampere rating of the fuses
agreed with the
applicable design
documents
except for the fuses
(ID Nos.
6A and 7A) for
inverter 821B-K801A in control panel
1C614.
These fuses
were 20-ampere
Bussman
type KTK-20 and did not agree with the 10-ampere
fuse rating identified in
Design
Document E16, Sht.
11, and vendor Drawing N1-B21-98.
The licensee
issued
NCR 90-0174 to address this discrepancy
and determined that the
installed fuses
are correct for the inverter application.
The design drawings
were never updated to reflect the as-built condition.
The licensee
took
appropriate
actions to update
the affected drawings before the end of the
inspection.
Control schematic
drawings reviewed typically specified only the current rating
and not the type of fuse.
The licensee
stated that during the pre-operational
phase of the plant, the engineering
department
performed
a walkdown to verify
as-built conditions.
However, the walkdown documents
did not include an
evaluation of the acceptability of various types of installed fuses.
Without
verification that the proper type of fuse is used,
a loss of electrical
23
~
~
~
~
coordination could occur.
In response
to this concern,
the licensee
stated
that Operation
Procedure
OI-AD-037 establishes
administrative control of fuse
removal
and replacement.
In addition, the licensee
stated that it is taking
steps to implement
a formal fuse control program that will include creating
a
database,
conducting walkdowns, verifying engineering
design,
and revising
applicable administrative procedures.
The fuses installed in 250-Ydc control centers
and load centers
are rated for
250-Vac.
This equipment is operated at
a higher dc voltage of 264-285
Vdc
because
of battery float/equalizing requirements.
The affected fuses
are Gould
OT fuses,
BUSS
NON fuses,
and
GE CF6 fuses.
In response
to the team's
concern
about the acceptability of these fuses,
the licensee
provided documentation
of
the qualification of these
fuses for a 250-Vdc rating.
However,
no documenta-
tion was available for the qualification of these
fuses at higher voltages.
The licensee
issued
an engineering
discrepancy
report
(EDR G00107) to have all
existing underrated
250-Ydc fuses
replaced with fuses of a higher voltage
rating.
This item is identified as
Open Item 90-200-03 in Appendix A to this
report.
4.5
Inverter Testing
The preventive maintenance
work authorizations
(MA) written to calibrate
and
test Topaz inverters in accordance
with procedures
IC-dc-100 and 400 and the
operating
and instruction manuals for static inverters
appeared
adequate.
This
testing is performed to verify inverter low- and high-voltage trip setpoints
and includes calibrations
every refueling cycle.
During initial installation,
the inverters were subjected
to a performance test from no-load to full-load
conditions to verify the output response.
The inverter setpoints
are selected
so that inverters will not trip at input voltages
between
100 and
140 Vdc to
provide 115-Vac output.
The low- and high-voltage trips are set at 95-Vdc and
147-Vdc, respectively, with resets at 108 Vdc and
132 Vdc.
The dc input
voltages to the inverter during battery float and equalize periods are
132 Vdc
and 142.8 Vdc, respectively.
The calibration
and trip setpoint records also
seemed
adequate.
Significant Operating
Occurrence
Report 1-90-177
was initiated to address
an
inverter trip which occurred while the batteries
were
on equalizing voltages.
The licensee
concluded that the dc overvoltage,
which caused
the inverter trip,
appeared
to be the result of the battery charger
response
to a voltage dip
during the start of a heavy ac load or a possible potential inverter drift.
As
a result, the corresponding
battery charger
was returned to float operation.
This caused
the inverter to reset
and reenergize
the lost instrumentation.
The
licensee is planning to trend the setpoint calibration data from the next
scheduled
preventive maintenance
tests
and determine whether the setpoints
or
battery equalizing voltage
have to be changed to address
this issue.
The team
concluded that the licensee
has
a program in place to monitor and test the
inver ter adequately.
4.6
Battery Testing
The team reviewed service testing procedures
SN-102-A04,A03
and SM-188-103-2
for Class
1E 125-Vdc and 250-Vdc batteries.
The test procedures
were adequate
to verify the design function capability of the Class
1E battery
system.
24
~
~
r
However, the procedures
did not account for the inaccuracies
of instruments
used during testing.
The licensee
believed the test procedures
have
enough
margin to account for instrumentation
inaccuracies,
but said it would revise
the procedures.to
address test instrumentation
inaccuracies.
The completed
test records for the Class
1E 125-Vdc battery bank 10610
had conservative test
acceptance
criteria and were in accordance
with requirements
of the station
Technical Specifications
and
IEEE Standard
450.
5.0
ENGINEERING AND TECHNICAL SUPPORT
5.1
Equipment Modifications
The team found the engineering
and technical
support associated
with selected
modifications related to the electrical distribution system to be of high
quality.
These modifications are discussed
below.
l.
Addition of a Fifth Diesel
The modification package for the addition of the fifth diesel generator
and the
team's
walkdown of the related installation were adequate.
Cable length and
separation,
compliance with the single-failure criterion,
and the short circuit
ratings of the associated
equipment
were also adequate.
2.
Temperature
Control Valve for Diesel Generator
Super charger Intercooler
The combustion air supplied to the diesel
generator is cooled downstream of the
supercharger
before entering the engine manifold.
The cooler was sized for ESW
at 95'F
and maximum outside air temperature.
A butterfly valve was adjusted to
establish
the required flow for these
bounding rated conditions
and locked in
position.
In the winter months,
excessive
heat
was
removed from the combustion
air and the temperature fell below the recommended
range of 95 to 125'F.
The
manufacturer felt that low air temperature
was
a contributing factor to recent
diesel failures.
The licensee
decided to add
a temperature
control valve and
a temperature
feedback control loop to modulate the
ESW flow and maintain inlet air tempera-
ture within the specified range.
The modification had been
completed in two of
the five diesels at the time of the inspection.
The design
change
package
(DCP-3009) for this modification describes
the changes
to be implemented
and
provides the supporting calculations
and setpoint evaluations.
The team was
impressed with the quality of the engineering
process
in preparing the design
documentation.
Although several
questions
were raised
concerning
the input
data
and boundary conditions,
the root source of the discrepancies
were easily
identified and corrected
because
the calculations
were well referenced
and the
excerpts
from the relevant design
documents
were included in the modification
package.
3.
120-Vac lE Inverter Addition
The primary function of the inverters
was to supply power to post-accident
monitoring
(PAN) instrumentation.
The calculated
load per inverter was below
1450
VA.
The inverters were energized
from the 125-volt Class
1E batteries,
channels
A and B.
Each inverter was rated at 2000
VA and consisted of two
25
1000
VA inverters in a master-slave
configuration.
The units were of the
ferroresonant
transformer output type.
The team found the input voltage with
the minimum and maximum values available from the 125-volt dc system
and output
voltage regulation acceptable for the application.
The rating of the invert-
ers, current limits for overload
and short circuit conditions,
and the protec-
tive fusing in the distribution system energized
from those inverters
were
adequate.
4.
Replacement
of Class
lE Battery
Modification package
DCP 87-9128 for the Unit 2 125-Vdc station battery capaci-
ty upgrade
was issued to replace the old 742-ampere-hours
Class lE batteries
with new 825-ampere-hours
CLD batteries.
The
new batter ies were installed
on
the old battery racks, which were seismically reanalyzed.
The safety evalua-
tion, post-modification testing,
and installation instructions
were technically
acceptable.
5.
Relay Replacement
Modifications
,Two plant modifications
(DCP 89-9017 A,B,C,D and 89-9132)
were associated
with
the replacement
of instrument relays
used in safety-related
applications.
Both
modifications required the dedication of commercial
grade relays for the
intended safety-related
application.
The associated
design
change
packages
verified the critical characteristics
required to ensure
the desired perfor-
'ance.
In addition, these relays were to be functionally tested after instal-
lation.
The licensee's
process
appeared
to adequately
address
the
critical characteristics
for these relays.
5.2
Discrepancy
Management
System
As part of the team's
review of the licensee's
engineering
and technical
support,
a review was conducted of the discrepancy
management
system
used to
identify, document,
evaluate,
and report deficiencies at Susquehanna.
5.2.1
Deficiency Control System
The licensee's
mechanisms
to control deficiencies at Susquehanna
include
nonconformance
reports,
engineering
discrepancy
reports, significant operating
occurrence
reports, audit findings, quality assurance
surveillance findings,
and deficiency reports.
1.
Nonconformance
Reports
Nonconformance
Reports
(NCRs) are
used to identify, document,
process,
and
control deficiencies with regard to a characteristic,
documentation,
or proce-
dure that renders
the quality of an item unacceptable
or indeterminate.
Nonconformances
include physical defects, test failures, incorrect or inade-
quate documentation,
and fai lure to comply with prescribed
processing,
inspec-
tion, or test procedures.
document the cause of the nonconformance,
the
evaluation of its effect on plant operability, the determination of its report-
abi lity to the
HRC, and the corrective actions taken
and to be taken to pre-
clude recurrence.
26
Procedure
AD-gA-120 requires
NCRs to be evaluated
to determine if the noncon-
fornance affects plant operability and if it has to be reported to the
NRC
within 2 days.
The procedure also requires that the report be dispositioned
within 45 days of the date of issue.
Contrary to these
requirements,
the
following open
NCRs, which were issued before June 7, 1990, were, evaluated for
operability and reportability from 4 to 86 days after their date of issue:
NCR Number
Number of Da s from Issue to Evaluation Dates
86-0931
86
86-0930
85
87-0883
,
31
90-0003
27
88-0660
19
89-0435
15
89-0362
11
88-0524
10
89-0066
8
89-0473
6
87-0453
5
87-0021
4
90-0117
Evaluation undated
Two PP&L audits of corrective action (Audit Report 90-019, Finding No. Ol, and
Audit Report 88-100, Observation. No. 5) identified similar conditions.
To control the "hold" placed
on a nonconforming item, authorization is needed
to have the item conditionally released
to permit its use before obtaining
a
disposition or completion of corrective action.
The conditional release
is
authorized for a specific length of time and is based
on an engineering justi-
fication that includes the extent to which the nonconforming item may be
installed,
used, or operated.
Although no nonconforming
items were currently installed in the plant beyond
their conditional release
expiration dates,
several
nonconforming
items
(NCRs
87-0021
and -0336; 88-0085,
-0327, -0328,
and -0493;
and 89-0659
and -0660)
had
been previously installed in the plant beyond their conditional release
expira-
tion dates.
This problem
may have
been
caused
because
there
was
no procedural
~
requirements
to designate
who is responsible for tracking
NCR conditional
release
expiration dates
and, specifically, to designate
what action is to be
taken before the
NCR conditional release
expiration date is reached.
2.
Engineering Discrepancies
A fairly new program is in place to control engineering
discrepancies (i.e.,
differences
between
an engineering
requirement
and its implementation or a
conflict between
engineering
documents).
The governing procedure,
EDM-(A-122,
Engineering Discrepancy
Management,
was originally issued
on December
12, 1989.
Revision
1 was issued
on June
29
ai>d Revision
2 on July 19, 1990.
The latest
procedure revision requires
engineering-discrepancy
items to be documented
by
using either an engineering
discrepancy
report
(EDR) or by assigning
an
type classification
and priority classification to an existing deficiency
control mechanism or an engineering
wor k request
(Eh'R).
The item, its assigned
27
EDR type classification
and priority classification, is entered
in the computer
tracking system for engineering
work requests.
However, several
items with a
priority 2
EDR classification (to be dispositioned within 14 days from the
origination date) did not have the "Date Disposition Needed" block on the
form completed
and
no date
had been entered
in the "Date Needed"
column of the
EWR tracking system.
Thus, ft was unclear
when the disposition
was needed.
The
EWR tracking report, titled "Weekly Engineering
Discrepancy
Report," sorts
entries
by priority class.
The issue of August 7, 1990, contained
138 entries,
and
no date
had been entered
in the "Date Needed"
column for 75 of those
54 percent).
This "Date Needed"
column is apparently
the disposition
due date
based
on the
EDR priority classification)
and not the implementation
due date.
Procedure
EPM-gA-122 requires
EDRs classified
as "Nuclear Safety" or
"Regulatory" to be reviewed within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to determine if a reportabi lity,
operability,
and plant safety assessment
is required.
Procedure
EPH-gA-122,
requires
the Discrepancy
Review Committee
(DRC) to document the results of its
reportability, operability,
and plant safety assessments,
and the basis for its
conclusions,
but does not specify where and
how these
assessments
are to be
documented.
These reportability, operability, and plant safety
assessments
were not being documented
on the
EDR form but in the minutes of the
DRC meet-
ings, which were not indexed
by
EDR number.
The minutes
and other appropriate
documents
are not being filed with the associated
EDR and are not readily
retrievable.
During the inspection, reportability, operability,
and plant
safety
assessments
could not be retrieved from the
DRC meeting minutes,
was not
on the
EDR Form, or elsewhere
in the
and G-00039.
Although the procedure
states that engineering action to close
EDRs be
completed within 180 days of the disposition date,
several "old" EWRs,
designated
and tracked
as
EDRs, remain
open (e.g., H-50612, October 4, 1985;
M-60758, June 23, 1986; M-71248, October 6, 1987; H-79310,
May 28, 1987).
EWR,
ID No. EIR-100304,
dated
September
15, 1982, containing
a "Disposition
Required
By" date of October 15, 1982,
was still open,
had been designated
as
an
EDR, and was being tracked in the
EWR tracking system
as
EDR X-03186.
3.
Significant Operating Occurrence
Reports
Significant Operating Occurrence
Reports
(SOORs) are used to identify and
document significant, potentially significant, or near miss occurrences
that
may negatively affect operation of the plant, adversely affect personnel
safety,
be of significant interest to the public, and require
management
attention.
Procedure
AD-gA-424 requires
a timely resolution of SOORs.
The
team identified several
old SOORs that have remained
open since
1987 (e.g.,
1-87-179, July 8,
1987
1-87-209,
August 4, 1987; 2-87-076, April 29, 1987;
and
2-87-104, July 4, 1987I.
4.
guality Assurance Audit Findings
and Surveillance
Findings
equal ity assurance
audit finding sheets
(AFSs) are used to document
adverse
findings identified during operational quality assurance
audits that require
corrective action.
They contain the controlling document
and requirement,
a
description of the deficient finding, the action required to correct the
28
L E
~
~ I
~
~
~
deficiency, the corrective actions
taken to resolve the deficiency,
and
verification of the adequacy of the corrective actions
completed.
The audit report that includes the AFSs contains observations
and recommenda-
tions and
an assessment
of the effect each audit finding has
on quality as
a
means of signifying the importance of each finding as well as
an overall
assessment
of the effectiveness
of the program being audited.
guality assurance
surveillance findings (gASFs) are used to document deviations
from specified requirements
or procedures.
(ASFs contain
a description of the
requirement,
the nonconforming activities or items,
a description of the action
required to correct the condition and to prevent recurrence,
and action per-
formed to verify the implemented corrective action.
The reviewed
gA audit reports
and surveillance
reports
appeared
to be well
written and effective in identifying problem areas.
5.
Deficiency Reports
Deficiency Reports
(DRs) are used to identify, document, control, and disposi-
tion significant conditions adverse
to quality.
Significant conditions adverse
to quality include major program breakdowns,
license
and regulation violations,
repetitive nonconforming conditions, trends
adverse
to quality, and adverse
generic conditions.
The processing
of DRs includes reporting significant
adverse
conditions to appropriate
levels of management,
correcting deficient
conditions, determining root causes
of the conditions,
and implementing correc-
tive and preventive actions to preclude recurrence
of the condition.
Of the five DRs that have been issued,
the team had concerns
about
DRs89-001
and 89-002.
DR 89-001
was issued
on April 24, 1989, to identify that adequate
configuration control requirements for safety-related
cable
and raceway design
drawings
have not been defined
and implemented.
This deficiency was identified
during the performance of gA Surveillance 88-184.
Cable
and raceway design
data were contained
in circuit and raceway
schedules
that were outputs of an
architect/engineer
computer program.
The architect/engineer's
computer was not
compatible with PPAL's computer equipment,
had
no calculation ability, and did
not maintain data related to 10 CFR Part 50, Appendix R.
The licensee's
computer program, which currently is used to determine
the as-engineered
or
as-built status of these
drawings,
was inadequate
because
the appropriate
software
gA controls
had not been applied to the program.
Furthermore,
the
process
required to accurately
determine
the as-engineered
or as-built status
of these
drawings
had not been defined.
The existing drawing update
process
in
use at the plant was not compatible for revising these
drawings.
The responsi-
bilities and special
process
required to revise these
drawings
had not been
established
and documented.
Thus, these
drawings were not being revised
on
a
routine basis.
Updated
FSAR Section 8.3.3.1 requires
a review to be performed
when
a cable
tray becomes
more than
30 percent full by cross-sectional
area to determine the
adequacy
of the design.
gA Surveillance
88-184 identified instances
where this
review (engineering
calculations)
was not performed
when the 30-percent-full
limit was exceeded.
Other instances
were identified where results of the
design
adequacy
review were questionable
as
a result of errors in the computer
29
program database.
These errors include cables
and raceways
having duplicate
identification numbers,
cable lengths not provided,
cable size
codes
not
defined,
and invalid cable size codes.
The conditions identified in QA Sur-
veillancee
88-184
have existed since
1984 when the licensee
assumed
responsibi l-
ity for the design of cable
and raceway from the architect/engineer.
The
licensee
established
an interim cable
and raceway tracking plan to correct
these conditions.
However, only part of this plan has
been
implemented.
The
licensee currently is re-evaluating
the required corrective actions
necessary
to resolve this issue.
DR 89-002 was issued
on November 11, 1989, to identify repetitive nonconforming
conditions with regard to the implementation of key elements of the Susquehanna
Equipment Qualification (EQ) Program.
These conditions were identified during
the performance of the
1989 annual
QA audit of the
EQ Program (Audit 89-075).
Several audit findings identifying EQ deficiencies
remained unresolved,
includ-
ing one
open finding from Audit 86-034
and three
open findings from Audit
88-021.
These conditions include as-qualified verses as-installed
configura-
tion control; identified
EQ maintenance
and surveillance
requirements
not
having been adequately
implemented;
and previously identified conditions/
deficiencies
reported
via NCRs, equipment qualification evaluation
requests
(EQERs), audit findings; and at least
one
NRC commitment.
These previously
identified
EQ deficiencies
remain unresolved
and more are continuing to
accumulate.
Specific deficiencies
include
Hardware found installed in the plant in an unqualified configuration.
This hardware
included standby
gas treatment
damper actuators
and
Rosemount transmitters.
Conflicting, obsolete,
and inaccurate
information that exists in
binders
as
a result of over 400 unincorporated
change notices
posted
against
89
EQ binders.
Inconsistences
between actual
EQ binder configuration and the design
document
management
system,
which contains
the official record of the
plant configuration.
Maintenance
requirements
specified in qualification test reports not being
identified to the Maintenance
Department.
Implementation of several
EQERs that provide
a mechanism to request
improvements or corrections to
EQ maintenance
requirements
were postponed
by as
much as 2-1/2 years.
Correction of these identified
EQ program deficiencies
are being addressed
by a
special action team.
6.
NCRs Related to Environmental Qualification
Three of the
NCRs reviewed pertained to the lack of qualification for numerous
Limitorque motor actuators
and target rock solenoid valves.
Of specific
30
I
0
concern
were
NCRs 88-0181
and 88-0520, which were originated
on March 24 and
July ll, 1988, respectively.
NCR 88-0181 identifies the concern that 21 motor actuators
in each unit are
equipped with Reliance
dc motors that were not subjected
to Limitorque qualifi-
cation testing.
The qualification testing related to these
motor actuators
was
performed
on Porter/Peerless
dc motors, which have not been
shown to be similar
to the Reliance
dc motors installed at Susquehanna.
Although the evaluation to
NCR 88-0181 identified a similarity analysis
performed
by Myle Labs for the
Shoreham nuclear plant, which compared
Reliance
125 Vdc and Reliance
480 Vac
motors, the licensee
had not shown its applicability to their 250-Vdc Reliance
motors.
In addition, the licensee
had not shown qualification for motor
actuators fitted with Peerless
Class
8 motors.
NCR 88-0520 identifies the concern that
31 motor actuators
in each unit are
operated with 250-Vdc control power, which is twice as
much as the 125-Vdc
control power used in the Limitorque qualification testing of these actuators.
The 250-Vdc control power is routed though the motor actuator limit and torque
switches that have exposed terminal connections.
In an accident environment,
these
connections
could be subject to insulation breakdown
as
a result of
moisture intrusion.
Although the licensee
provided
some evidence that these
components
could be environmentally qualified, the evaluation
was found to be
weak because
It relied partially on
a test report (F-C3271) that included
no
pre-accident
aging or radiation.
Low-resistance
readings
have
been recorded for fibrite torque switches,
even at 120-Vdc.
the Limitorque motor actuator is not a sealed
device
and
some moisture
intrusion is expected.
Although both
NCRs 88-0181
and 88-0520
have
been evaluated,
over 2 years
has
elapsed without resolution of these
issues.
Furthermore,
extensions
to both
NCRs cited
a lack of resources
as the reason resolution of these
NCRs has not
been accomplished.
As a result,
numerous Limitorque motor actuators
have still
not been qualified and are currently in an indeterminate
status.
This issue
has
been
documented
in Region I
NRC Inspection Report 50-387/90-17
and
50-388/90-17.
In addition, the three
NCRs discussed
below identified nonconforming environ-
mentally qualified equipment installed beyond their conditionally released
expiration dates.
NCR 88-0493 identified two installed, environmentally qualified electrical
relays that had not been replaced
by a qualified life expiration date specified
in their preventive maintenance
schedule.
These relays were installed for
17 days beyond their conditionally released
expiration date until a reanalysis
indicated that the relays could remain installed until a new date that was
specified in a revised preventive maintenance
schedule.
This
NCR was verified
closed
on November 17, 1988.
31
~
~
NCR 88-0659 identified
1 of 2 primary containment
hydrogen/oxygen
analyzer
sample
pumps installed in Unit 2 contained nonqualified
Buna-N rather
than
Neoprene 0-rings.
This condition constituted
an
EQ nonconformance
since
Buna-N
0-rings
had not been tested or analyzed for use under post-accident
conditions.
NCR 88-0660 'identified a similar condition on 2 of 2 primary containment
hydrogen/oxygen
analyzer
sample
pumps installed in Unit 1.
The three installed
pumps were permitted to operate for 2 months beyond their conditionally
released
expiration date which were documented
on the NCR's.
This condition
was identified during the performance of the 1989 annual
QA Audit of the
Program, Audit 89-075.
The nonconforming Unit 2 pump was replaced with an
acceptable
spare
pump on 'October 13,
1989
and
NCR 88-0659
was verified closed
on November 24, 1989.
Eight spare
hydrogen/oxygen
analyzer
sample
pumps were
determined
unacceptable.
Five were sent out for repair; the remaining
3 were
retired from service.
The
NCR for the Unit 1 pumps
was reevaluated.
It was determined that these
pumps could remain installed for an indefinite period of time.
These Unit 1
pumps are scheduled
to be replaced with qualified pumps during the upcoming
refueling outage
scheduled
to begin in September
1990.
DR 89-002 was issued
on
November 30, 1989, to inform SSES
management
personnel
of this condition.
5.2.2
Su+nary of Discrepancy
Management
System
Review
Although the licensee
has effectively identified and documented
the deficien-
cies at Susquehanna,
the deficiencies
have not been evaluated,
controlled,
processed,
and resolved in a timely and effective manner.
To address
and improve this situation,
PPKL issued
Nuclear Department Policy
Letter Number 90-003, Revision 0, on August 13, 1990,
on the subject of Defi-
ciency Control.
This policy letter
includes
PPSL standards
of performance for
the identification, control, and closure of deficiencies at the Susquehanna
plant.
Conditions adverse
to quality, plant safety,
and reliability are to be
promptly identified, reported,
and corrected.
All work activities associated
with the identification and closure of
deficiencies will receive proper management
attention
and priority.
These
activities include determinations
of operability and reportability,
disposition,
and corrective action implementation.
PPSL
has established
a Deficiency Control and Corrective Action Program.
As
part of this program,
PPKL will review the backlog of open deficiencies
issued
before August 13, 1990, to determine
the need for an accelerated
and/or differ-
ent disposition plan for all open nonconformance
reports,
engineer ing discrep-
ancy reports, significant operating
occurrence
reports, quality assurance
audit
findings, quality assurance
surveillance findings,
and deficiency reports
by
the target date of November 15, 1990.
PPSL will integrate provisions into its deficiency control processes
to ensure,
with limited exceptions,
that the lifetime of any corrective action is limited
to one cycle of operation.
32
'l
e
~
~
If this program is properly implemented,
the teams
concerns with the
PPEL
discrepancy
management
program should
be resolved.
6. 0
GENERAL CONCLUSIONS
The inspection
team concluded that generally the
SSES electrical distribution
system would be capable of performing its intended safety functions.
With the
exception of the
14 specific findings identified in the report; the batteries,
emergency
diesel generators,
switchgear,
and other components within the
EDS
were found to be adequately
sized
and configured.
Separation
between
redundant
trains or divisions was found to have
been adequately
maintained,
and
an
adequate
design basis exists
and is being upgraded
and maintained for the SSES.
The general quality of engineering
and technical
support appeared
to be good,
and the licensee
had
a good self-assessment
(quality assurance
audit) program.
Weaknesses
were identified by the inspection
team in voltage regulation at the
lower ac voltage levels,
and in timeliness of corrective actions to self-
identified deficiencies.
33
APPENDIX A
Ins ection Findin
FINDING CATEGORY AND NUMBER:
OPEN
ITEM 90-200-01
FINDING TITLE:
Calculation Errors and Discrepancies
DESCRIPTION OF CONDITION:
The team identified several
problems
concerning calculations
and the control of
calculations.
Some of these
problems
are discussed
below.
1.
The licensee
does not have
a calculation index or other system for con-
trolling the use of non-valid or superseded
calculations.
As
a result,
three calculations
were found to be in effect for the short circuit rating
of Class
lE switchgear.
The three calculations
did not have the
same
assumptions
and results.
Even so, the licensee
considered
them valid and
had put them into effect simultaneously.
These calculations
contained
several
nonconservative
features,
including those listed below.
There were errors in the multiples used for breaker interrupting
times.
Maximum possible voltage
was improperly considered.
The installation of new
ESF transformers
before plant startup
had not
been accurately
considered
in modeling of the system.
One of the calculations
had been designated
"non-g" and lacked proper
review and auditabi lity.
In response
to the team's
concern,
the licensee
proceeded
to perform a
fourth calculation
(SC-1), which was given to the team for review in
preliminary form on August 31,
1990.
The team's
review of this calcula-
tion indicated that most of the
comments
and concerns
had been addressed.
However, the available margin between
the duty and the rated values for
the switchgear short circuit capability
was practically nonexistent.
The
licensee
committed to perform formal calculations to allow for proper
evaluation of this issue.
The team was not concerned with the safety implications of the minimal
available
margin because
the conditions
when the short circuit rating
could be exceeded
are not continuously present.
These conditions are only
present
when the diesel generator is in test
and is paralleled with the
grid.
2.
The team determined that cable losses
were not included in the loading
tabulations for the emergency
diesel generators.
These
losses
were
calculated
by the licensee to be approximately
8
kW.
Although the team
did not have
an immediate safety
concern
because
approximately
a 326-kW
margin exists for the diesel capacity,
the team considered
these
losses
sufficiently important to warrant revision of the loading tabulations.
The licensee
has
committed to perform complete calculations
and revise the
loading tables accordingly.
REQUIREMENTS
ANSI NQA-1-1979, Supplement 6S-l, requires that design inputs
and documents
be
properly prepared, filed, and controlled.
ANSI Std C37.010-1972,
requires that circuit breakers
be rated
above the
required interrupting duty.
10 CFR Part 50, Appendix B, Criterion III, Design Control, requires
design
control measures
to be provided for verifying of checking the adequacy of
design.
Criterion III of 10 CFR Part 50, Appendix B, requires that measures
be estab-
lished to ensure
the design basis is correctly translated
into specifications,
drawings,
procedures,
and instructions.
REFERENCES:
2.
4 ~
Calculation E2004-01,
"System Fault Duty Calculations,"
Revision 3,
Narch 17, 1981, performed
by Bechtel.
Calculation GP-19, "Calculation for Minimum ES Transformer
Impedance,"
Revision 1, October 22, 1982, performed
by PP&L System Operating
Department.
Calculation GP-18,
"4
8
13
kY Swgr. Units
1
E
2 Duty Gale.," Revision 1,
October 2, 1980, performed
by PPhL.
Calculation SC-l, preliminary, August 31, 1990, performed
by PPSL.
A-2
FINDING CATEGORY AND NUMBER:
UNRESOLVED ITEM 90-200-02
FINDING TITLE:
Over load Alarm
DESCRIPTION
OF CONDITION:
The diesel generators
incorporate
an inverse time overcurrent relay to provide
an alarm in the control
room to alert operators
of a diesel overload condition.
The team found that the overcurrent relay settings
were not adequate for diesel
generator
E and that all diesel relays exhibited excessive drifting, not
properly accounted for in the relay setting calculations.
For diesel generator
E, the relay minimum pickup was
105 percent
above the maximum generator rating,
which would mean that overloading of up to 5 percent
could be present for an
indefinite period of time without any alarm to indicate this condition.
In
addition, review of the surveillance report for the overload relays disclosed
that excessive drifting was occurring in the pickup value of the relays,
as
depicted
by the following examples:
Ref.
MAP83526,
November 4, 1989,
51 relay for DGA, drift from previous
test
on September
9,
1985 was 250 percent of setpoint.
Ref.
MAP00555, March 16,
1990,
51 relay for DGC, drift from previous test
on October 31,
1988 was
100 percent of setpoint.
Ref.
HAP83528,
November 18, 1988,
51 relay for
DGC, drift from previous
test
on February 22,
1984 was 25 percent of setpoint.
Ref. HAP83527,
February 21, 1989,
51 relay for DGB, drift from previous
test
on February
15,
1984 was
50 percent of the setpoint.
The
51 relay instruction sheets
(GEI-28818D) indicated that the above drifting
values
are not unexpected for values
close to minimum pickup.
Despite the very
high level of drifting found in the periodic surveillance, drifting had not
been considered
in the relay setpoint calculation.
The team was concerned that improper settings
and drifting of setpoints
could
prevent initiation of the overload alarm if the diesel generator
was
overloaded.
REIEIIIREIIEHTE:
IEEE Standard
308, Alternating Current Power Systems,
requires that protective
devices
be provided to limit the degradation of the Class
1E power systems.
10 CFR Part 50, Appendix A, Criterion 17, requires
the design of the electric
power system to be capable of permitting functioning of components
important to
safety.
10 CFR Part 50, Appendix B. Criterion III requires
measures
be established
to
ensure
the design basis is correctly translated
into specifications,
drawings,
procedures,
and instructions.
A-3
REFERENCES:
~
~
~
~
1.
PPSL System Operating
Department
Relay Setting
Change Notice 1381(,
January
31, 1981.
2.
Calculation 1-20204-5,
"Relay Setting Calculation for Diesel
Gen A
(B,C,D,E) T.O.C. Alarm 5 Diesel
Gen
E Def. T.O.C. Alarm," Revision 1,
January
14, 1987,
performed by PPSL.
3.
Relay Surveillance
Work Authorizations, for Card 4DGA-1-A4, Card
ODGB-I-A4, Card fDGC-l-A4, Card fDGD-l-A4, Card EDGE-l-A4.
4.
Relay Instructions,
GEI-28818D, Applicable to Type
IAC66A Relay.
5.
PP8L Single Line Heter
5 Relay Diagram 4.16
kV Diesel Generator
Common,
Dwg. No. E-5, Sheet 4, Rev. 9, January ll, 1988.
FINDING CATEGORY AND NUMBER:
OPEN
ITEN 90-200-03
FINDING TITLE:
Overvoltage
On 250-Vdc System During Battery Float/Equalize
Conditions
DESCRIPTION OF CONDITION:
The licensee's
fuse control program
showed that several
control fuses installed
in 250-Vdc load centers
and motor control centers
(MCCs) are rated only for
250 Vdc.
This equipment is operated at
a higher dc voltage of 264 to 285 Vdc
during the application of a battery float voltage
and equalizing
charge.
The
250-Vdc rated fuses that were verified during the field walkdown in load
centers
and
l'ICCs are Gould
OT fuses,
Bussman
NON fuses,
and
GE CF6 fuses.
The
team raised
concerns
regarding
the acceptability of these
fuses at higher
voltages
and the licensee
issued
engineering
Discrepancy
Report
G00107 to
address
this issue.
Preliminary assessment
of overvoltage
on fuses
showed that during short circuit
fault conditions greater
than
375 amperes,
the battery charger would regulate
the output voltage to the battery
and would eliminate any potential overvoltage
problems.
However, the licensee
could not provide sufficient justification for
the 250-Vdc fuses capability to interrupt during normal overloads at higher
voltages.
The licensee's
preliminary analyses
showed that if a sustained
overload existed for a long period of time, the upstream breakers
or fuses
would interrupt to protect the circuit, causing
a loss of one division of dc
power.
The licensee
stated that it is planning to replace all underrated
fuses in the 250-Vdc system
as
soon
as possible.
REIEII REIIRIIE:
10 CFR Part 50, Appendix B, Criterion III, "Design Control," requires in part
that measures
be established for the selection
and review for suitability of
materials
and equipment that are essential
to the safety-related
functions of
the systems.
REFERENCES:
1.
FSAR Section 8.3.2,
"DC Power Systems."
2.
"Component Failures
Caused
By Elevated
Control Voltage."
3.
Engineering
Discrepancy
Report G100107.
A-5
FINDING CATEGORY AND NUYiBER:
UNRESOLVED ITEtl 90-200-04
FINDING TITLE:
Questionable
Seismic Qualification of 250/125
Vdc Load
Center Breakers
In Racked-Out Position
DESCRIPTION
OF CONDITION:
During a walkdown, the team found all the spare breakers
in the 250/125-Vdc
load centers
were in a partially drawn-out position, which led the team to
question
the seismic qualification of this equipment in this configuration.
The licensee
stated that the effect on the seismic qualification of the
affected
load centers
is indeterminate
because
the test configuration was with
the breakers
racked in.
The licensee
subsequently
reported this unanalyzed
condition to
NRC in accordance
with 10
CFR 50.9.
As a result of the team's
concerns, all spare circuit breakers
were racked in until the analysis to
determine
the acceptability of the seismic qualification is completed.
RE UIREHENT:
GDC 2, "Design Bases for Protection Against Natural
Phenomena,"
requires that
structures,
systems,
and components
important to safety
be designed to with-
stand the effects of earthquakes.
REFERENCES:
None
A-6
~
g
~
FINDING CATEGORY AND NUMBER:
UNRESOLVED ITEM 90-200-05
DEFICIENCY TITLE:
Degraded Grid Relay Setpoints
DESCRIPTION
OF CONDITION:
The current setpoints for the undervoltage
relays to actuate
under degraded
grid conditions
do not provide adequate
protection for safe operation of all
Class
1E loads at the 480-Vac and 120-Vac voltage levels.
The setpoints for
the relays are at 84 percent of rated
bus voltage for the 4160-Vac buses.
To
ensure
adequate
protection
and operation of all Class
1E loads, calculations
indicate that the system voltage must not drop below approximately
93 percent
of rated voltage at the 4.16-kV bus.
As a result,
should the 4160-Vac bus
voltage fall below 93'percent,
but remain
above the degraded
grid relay
setpoints of 84 percent,
adequate
operation of numerous
Class
1E loads could
not be ensured.
The licensee is in the process of developing
new relay setpoints to provide
greater protection of'quipment and to ensure that sufficient voltage is
available at the required
loads.
These
new settings require that the facility
Technical Specifications
be changed
and these
changes
require
NRC approval.
The licensee
has
committed to submitting
a technical specification
change
request for raising these setpoints
by September
30, 1990.
REIEUERBiENTS:
~
~
~
Criterion III of 10 CFR Part 50, Appendix B, requires that measures
be estab-
lished to ensure the design basis is correctly translated
into specification,
drawings, procedures,
and instructions.
Criterion XVI of 10 CFR Part 50, Appendix B, requires
measures
be established
to ensure that deviations
and deficiencies
are promptly identified and
corrected.
REFERENCES:
1.
FSAR Section 8.3.1.3.6,
"l1anual
and Automatic Interconnections
Between
Buses,
Buses
and Loads,
and Buses
and Supplies."
2.
Technical Specifications,
Section 3/4.3.3,
"Emergency
Core Cooling System
Actuation Instrumentation."
A-7
~
~
~
~
E 'g'INDING
CATEGORY AND NUMBER:
UNRESOLVED ITEN 90-200-06
FINDING TITLE:
Pump Thermal Insulation
Removal
DESCRIPTION
OF CONDITION:
During
a walkdown of the reactor building HVAC system,
the team found that the
thermal insulation
on the
HPCI pump, booster
pump,
and crossover piping had
been
removed.
This insulation
was
removed during July 1984 for a maintenance
activity and inadvertently
was not reinstalled.
In addition, the reactor
building
HVAC calculations
had not been revised to consider the result of the
insulation removal.
During the inspection,
the licensee
performed
a calcula-
tion and determined that there
was
a sufficient margin in the
HPCI room coolers
to handle the additional
heat load.
Two nonconformance
reports
were generated
as
a result of this concern,
and the licensee is planning to reinstall the
insulation in the near future.
REEEIIIREIIERTR:
Criterion
V of 10 CFR Part 50, Appendix B, requires activities affecting
quality be performed in accordance
with appropriate
procedures.
The procedures
should contain appropriate
acceptance
criteria for determining the activities
have
been satisfactorily accomplished.
REFERENCES:
i.
2.
3.
Calculation 176-18,
"Reactor Building Heat Load," Revision 4, performed
by
Bechtel.
PPKL Nonconformance
Report No. 90-0185,
August 27, 1990.
PP8L Nonconformance
Report No. 90-0186, August 27, 1990.
~
I
FINDING CATEGORY AND NUMBER:
UNRESOLVED ITEN 90-200-07
FINDING TITLE:
Insufficient Heat Transfer Surface
Area in Condensers
DESCRIPTION
OF CONDITION:
The essential
(ESW) system
draws water from a reservoir to
provide cooling to Class
lE equipment.
Little can
be done to control the water
chemistry in an open-loop cooling water system.
Furthermore,
the
ESW system is
normally in a standby condition; therefore,
sedimentation
occurs
as
a result of
stagnant
cooling water left in the piping and heat exchangers.
As a result,
excessive fouling of the heat transfer surface
(0.003 hr-ft'-'F/btu versus
0.002
as originally specified)
and plugging of tubes
due to wall thinning,
decreases
the capacity of the heat transfer equipment serviced
by
ESW.
Currently, 5,to
6 percent of the tubes in the control structure chilled water
(CSCW) system
condenser
are plugged,
and
4 percent of the tubes in the direct
expansion unit condenser for the Unit 2
HVAC system are plugged.
The
effects of increased
fouling and the reduction of available surface
area
as
a
result of tube plugging could affect the operability of these
components
and
hence the operability of the Class
1E equipment
supported
by them.
As a result
of this finding, the licensee is performing
an evaluation using
a fouling
factor of 0.002
on the refrigerant side
and of 0.003 (hr-ft'-'F/BTU) on the
side.
Should the assessment
of CSCW condenser
prove to be marginal or unac-
~
~
ceptable,
the licensee
has agreed to perform a similar assessment
of the
DX
units for the Unit 2
REIERIREMERI:
Criterion III of 10 CFR Part 50, Appendix B requires that measures
be estab-
lished for the selection
and review for suitability of materials
and equipment
that are essential
to the safety-related
functions of the systems.
REFERENCE:
1.
PPEL Specification
N1453 General Specification for Heat Exchanger
Tube
Plugging.
A-9
FINDING CATEGORY AND NUMBER:
OPEN ITEN 90-200-08
FINDING TITLE:
Lack of Overpressure
Protection for ESM Side of Diesel
Generator Auxiliary Heater Exchangers
DESCRIPTION
OF CONDITION:
The sections of emergency
(ESM) system piping providing cooling
water to the auxiliary coolers of each diesel generator
(intercoolers,
water
jacket,
lube oil and fuel oil coolers)
can be isolated from the main system
piping.
Overpressure
protection
has not been provided for this piping and the
secondary
side of the heat exchangers
as required
by ASHE Code Section III,
The plant has five diesel generators,
four of which are aligned to Class
1E
buses while the fifth is on standby.
The
ESW system
has sufficient capacity to
provide cooling to only four diesel generators
at one time.
Therefore,
the
isolation valves in
ESW lines to the standby diesel generator
are normally
closed,
which allows for the following potential
source of overpressurization:
1.
To minimize the time required to place
a diesel generator
back in service,
the standby
lube oil and jacket water
pumps
and heaters
are operated to
prewarm the equipment while the cooling water is isolated.
This condition introduces
energy
sources
to the primary side of the auxiliary
heat exchangers
with no heat sink available.
There is a potential for
common-mode failure of both divisions of the safety-related
ESW because
both
loops of ESW are connected
to each diesel generator.
The licensee
has
committed to perform procedural
or hardware modifications
as
required to eliminate the overpressure
concerns.
REIEUIRE ~ENT:
ASME BPV Code,Section III, ND-7000, requires that "a system shall
be protected
from the consequences
arising from the application of conditions of pressure
and coincident temperature
that would cause
the Design Pressure
specified in
the Design Specification to be exceeded"
(where system refers to a component or
group of components for which overpressure
protection is provided).
REFERENCES:
l.
ASNE Boiler and Pressure
Vessel
Code,Section III, ND-7000,
1971 Edition
up to and including Minter 1972 addenda.
2.
E106216 Sheet (H-ill) Emergency Service Mater System
PAID.
A-10
FINDING CATEGORY AND NUMBER:
UNRESOLYED ITEN 90-200-09
FINDING TITLE:
Diesel
Fuel Oil Storage
Tank Level Indication Inoperable
DESCRIPTION
OF CONDITION:
During a walkdown of the diesel generators,
the team observed that the alarm
for low level in the diesel fuel oil storage
tanks
was annunciated
in two of
the five diesel generator
rooms.
The licensee told the team that there are
currently
no level indicators or low-level annunciators
in service for the five
diesel fuel oil storage
tanks
because
incorrect assumptions
were
made in
calculating the instrument setpoints
and, with the high-error rate in the
instruments,
they could not be properly calibrated.
Although two of the five
level instruments
have been replaced with new, more accurate
instruments,
they
are not operating properly because
of elevated
temperatures
in the cabinet
where they are located.
The licensee
is currently using
a dip-stick method to
check the level in the tanks monthly and after diesel starts.
This condition
has existed for over a year with tank
E and for several
months with tanks
A
thr ough D.
This condition is inconsistent with the
FSAR, which states that the fuel oil
storage
tank level is continuously monitored
and indicated at the local control
panel in the diesel
generator building.
By the current method of monitoring
level,
a fuel oil leak between diesel starts
would go undetected.
The licensee
plans to replace
the three remaining old level indicators
and to make
a modifi-
cation to the
new instruments
so that they will be operable.
The licensee
has
committed to check the diesel fuel oil storage
tank levels
every
7 days by the dip-stick method until the
new instruments
are operable for
all five tanks.
The team found this acceptable.
RE IEII R El l EIII:
FSAR, Chapter 9.5.4.4, states
in part:
"The fuel oil inventory in the
storage
tanks
and day tanks is continuously monitored
and the level is
indicated at the local control panel in the diesel generator building."
10 CFR Part 50, Appendix B, Criterion XVI, requires
measures
be established
to
ensure that deviations
and deficiencies
are promptly identified and corrected.
REFERENCES:
1.
FSAR, Chapter 9, Revision 40.
FINDING CATEGORY AND NUMBER:
UNRESOLVED ITEM 90-200-10
FINDING TITLE:
Fire Protection Piping in Diesel Generator
Room
DESCRIPTION
OF CONDITION:
The fire protection
system is not seismically qualified.
Although the sprin-
kler pipes are normally air filled, the deluge valves
and related controls are
not seismically qualified and
can
be assumed
to fail.
Therefore,
the
nonqualified sprinkler systems
could spray Class
1E electrical
equipment
following a seismic event.
The team requested
that the licensee
demonstrate
that the fire protection
system would not impair the operation of the diesel
generators
following a seismic event.
The licensee
provided
a summary
sheet of the activities conducted
in accordance
with Bechtel Specification 8856-M-388,
Engineering
Program for the Job Site
Review and Disposition of Safety Impact Items,"
a guide for assessing
the
effect of failures of equipment that is not safety qualified on Class
1E
equipment.
The diesel
generator
room signoff sheet stated that the fire
protection
system
was designed
and installed in accordance
with Bechtel Speci-
fication M-343, which thus ensured
the integrity. of the piping.
However, the team found that the spacing of many pipe supports
exceeded
the
allowable spans
given in Appendix
B of the specification
by 50 to 100 percent.
As a result of this finding, the licensee
performed calculations for these
nonconforming sections
of pipe, taking into account the water filled pipes,
the
stress
intensification for threaded joints, the seismic accelerations
with
2 percent
damping,
and the as-built spacing
and found the stresses
to be below
the
maximum allowable.
However, the team was concerned that the guideline
practices
used to design
a system,
were not followed for many of these
pipe
supports.
The licensee
confirmed that an engineering
discrepancy
report
(EDR
G00154)
had been
issued to revise the specifications
and drawings to reflect
the as-built layout.
Rtt ENTS:
Criterion III of 10 CFR Part 50, Appendix B requires
measures
be established
to
ensure
the design basis is correctly translated
into specifications,
drawings,
procedures,
and instructions.
Criterion
V of 10 CFR Part 50, Appendix B requires activities affecting quality
be accomplished
in accordance
with appropriate
procedures.
USNRC Regulatory
Guide 1.29, "Seismic Design Classification," Section
C-2
requires that "those portions of structures,
systems,
or components
whose
continued function is not required but whose failure could reduce the function-
ing of any plant feature
included in items l.a through 1.q above to an unac-
ceptable
safety level or could result in incapacitating injury to occupants
of
the control room should be designed
and constructed
so that the
SSE would not
cause
such failure."
~
S
FSAR Section 9-5.1.1-19 requires structural integrity of the portions of the
~
~
~
~
~
~
~
~
fire protection
system in the vicinity of safety-related
structures,
systems,
and
components
during
a safe-shutdown
REFERENCES:
1.
Bechtel Specification 8856-M-388, "Engineering
Program for the Jobsite
Review and Disposition of Safety
Impact Items."
2.
Bechtel Specification 8856-M-387, "Safety Impact List for SSES."
3.
Bechtel Specification 8856-M-343, "Technical Specification for Deluge and
Sprinklers for SSES," Appendix B, specifies
the type of pipe supports to
be used
and the
maximum spans
between supports.
4.
FF 108930-8856-M343-74-5,
Fire Protection
System Layout Drawings.
5.
FF 108930-8856-M343-75-5,
Fire Protection
System Layout Drawings.
6.
FF 108930-8856-M343-76-5,
Fire Protection
System Layout Drawings.
7.
E-106227
(M-122), Sheets
1, 2, 3, and 10, Fire Protection
System
P&ID.
8.
Design Description,
Chapter 44, Fire Protection
System.
9.
FSAR Section 9.5.1, Fire Protection
System.
A-13
~
~
~
FINDING CATEGORY AND NUMBER:
UNRESOLVED ITEM 90-200-11
FINDING TITLE:
Instrument Drift and Lack of Trending
DESCRIPTION
OF CONDITION:
The team identified several
relays
and timers that were found to be out of
calibration during testing.
These relays included overcurrent relays
and
protective circuit timers.
These
instruments
are calibrated currently
on a
3-year interval.
The team reviewed the data sheets for 16 instruments
and identified 8 of them
with as-found readings
outside the instrument tolerance
band.
Bus
1A 27A
UV Relay (+2.9 percent)
Bus
1B 27A UV Relay (+3.3 percent)
Pump lA 50/51 Alarm Relay (+4.6 percent)
Pump
1B 50/51 Relay (+2.8 percent)
RHR Pump
1A 50/51 Relay (+6.3 percent)
62-27-A IX1-20201 Timer (-7. 5
ercent)
62B1-20102 Timer (-21 percent
62Bl-20202 Timer (-21 percent
Although, during the calibration testing the instruments
were reset to within
the acceptable
band, there
was
no formal program to evaluate
"as-found" data
~
~
which was outside the tolerance
band or to trend the associated
instrument
drift.
As a result,
instrument drift values
used in,setpoint calculations for
these
instruments
have not been validated,
and
may be nonconservative.
Additionally, operability determinations
have not been
made for those
instruments
in which the excessive drifts could have
compromised their safety
functions.
R ttUIR IIERTS:
10 CFR Part 50, Appendix B, Criterion III requires that measures
be established
for establishing
the adequacy of a design.
10
CFR Part 50, Appendix 8, Criterion XI requires that testing which is
designed to demonstrate
system or component
performance
be performed in accor-
dance with written test procedures
that incorporate
the acceptance
limits
contained
in the applicable design
documents.
REFERENCES:
None
A-14
c I ~
J
I
~
4
FINDING CATEGORY AND NUMBER:
UNRESOLVED ITEM 90-200-12
FINDING TITLE:
Improper Application of Undervoltage
Relay
DESCRIPTION
OF CONDITION:
The team identified an installed undervoltage
relay for which the dropout
setting
was outside the manufacturer's
specified range.
The relay setpoint
calculation indicated that the installed
27A relay had
a manufacturer's
setpoint
band of 36-45 volts.
The licensee installed this relay and adjusted
the setpoint to 24 volts, which is outside the manufacturer's
operating
band.
The manufacturer's
setpoint
band envelopes
the settings at which the relay can
be set
and still perform as desired with accuracy
and repeatabi lity.
The
licensee
replaced
the installed relay with one with an appropriate
band of
18-30 volts.
RE UIREMENT:
10 CFR 50, Appendix B, Criterion III, "Design Control", requires in part, that
measures
be established for the selection
and review for suitability of appli-
cation of equipment that are essential
to the safety-related
functions of the
structures,
systems
and components.
REFERENCES:
1.
Calculation 1-20200-2,
"27A Relays,
March 4, 1982, performed
by PPGL.
2.
Work Authorization V00492, August 24, 1990.
A-15
~ a5 +
~
r
a
~
FINDING CATEGORY AND NUMBER:
UNRESOLVED ITEYi 90-200-13
FINDING TITLE:
Inadequate
Testing of 250/125-Vdc
GE AK Circuit Breakers
DESCRIPTION
OF CONDITION:
The General Electric AK 2-25 250-Vdc circuit breakers
are tested
using
a
multi-ampere
ac current source.
The team questioned
the accuracy of testing
dc
circuit breakers with an ac current source.
Subsequent
discussions
with the
licensee
and the manufacturer
indicated that the breaker trip response
would be
different if they are tested with an ac current source.
In response
to the
team's
concern,
the licensee
sent two breakers to the manufacturer to perform
an overcurrent trip test using ac and dc currents.
Preliminary test results
showed that when
a breaker is tested
using dc current, the trip unit will pick
up 15-40 percent
sooner than with an equivalent
ac current.
The
iicensee
has
not analyzed
how coordination of the existing 250/125-Vdc breakers will be
affected
by the difference in the actual
pickup values.
In addition, the
licensee
does not have adequate
data to indicate that the dc trip curve main-
tains the
same
shape
and size as the ac trip curve.
Furthermore,
the existing
maintenance
procedure,
NT-GE-006,
does not provide adequate
information for
testing the circuit breakers
and there is no acceptance
criteria mentioned in
the procedure
except for the data sheet to record the test values.
The
licensee
has
committed to a complete evaluation of this issue.
REIEUI RE I ERIE:
10 CFR Part 50, Appendix B, Criterion XI, "Test Control", requires,
in part,
that
a test program be established
to assure that all testing require to
demonstrate
that structures,
systems
and components will perform satisfactorily
in service is identified and performed in accordance
with written test proce-
dure which incorporate
the requirements
and acceptance
limits contained
in
applicable
design
documents.
SSES Technical Specification, Section 6.8.1 states
in part, "that written
procedures
shall
be established,
implemented
and maintained
covering the test
activities of safety related equipment."
REFERENCES'.
Procedure
NT-GE-006, "Load Center Breaker Relay Logic and Primary Current
Testing," Revision 7.
2.
Test data for circuit breakers
and 1D662-013,
December
12,
1989
and July 15, 1982.
4
~ 4$ )t ~ g 4
~
~
(4
APPENDIX
B
Persons
Contacted
The following list contains
those
persons
contacted
by the team during the
inspection.
Those persons
marked with an asterisk also attended
the exit
meeting.
J.
P.
- J
- R.
- B
- p
- S.
- R.
- P
- C.
- D
'K.
- P
E.
- P
- C
M,
- 'g
- P
M.
- F
- P
S.
G.
G.
- A.
D.
+G.
- C
- P
T.
- R.
- V
P.
- J
J.
D..
- J
A.
- M.
- B
Sleva
A.
Personnel
A
s
Bartel
Berger
Bogar
Bozarth
Brady
Bry 1 ins ky
Byram
Cardinale
Coddington
Davis
Deange
1 i s
Dyckman
Eustice
Filchner
Galbrai th
Gulliver
Hecht
Heffelfinger
Hiedorn
Jones
Koste1 ni k
Kuhn
Kuzynski
Maertz
Male
McGann
Morris
Myers
Nudge
Oldenhage
Paley
Reel a
Riley
Rimsky
Robinson
Roth
Rothe'abol
She lbner
Skorus
ND, Electrical
Operations
Site Engineering
Electrical Maintenance
EDSFI
Team
EDSFI Team
Electrical Testing
VP, Nuclear Operations
NPE,
Licensing
Bechtel
Electrical
Al1egheny E1ectri c
Coordinator,
NQA
ND Electrical
EDSFI Team
Supervising
Engr.
NQA
Exec.
VP, Operations
NQA
Bechtel
EDSFI Team
Technical Supervisor
Operations
Manager,
Nuclear Plant Engineering
Compliance
Consultant
Manager,
Nuclear Projects
ND, Electrical
Tech Section
Tech Section
Super visor,
NQA
Tech/Site
Transmission
Senior Compliance Engineer
Electrical Supervisor
Manager,
NQA
Site Engineering
NP Electrical
Electrical
l
~
~ /y
PPKL Personnel
(cont.)
- G. Stanley
- B. Veazie
D. Weatherly
- R. Wehry
- G. Wetel
- C. Whirl
NRC Personnel
AAA
<<C. Anderson
- S. Athavale
- G. Barber
- J. Beaton
- R. Gramm
- W. Hodges
- J. Jacobson
- R. Jolliffe
- A. Josefowicz
",J. Lara
"B. Liaw
- R. Hathew
- 0. Hazzoni
- J. Stone
Plant Superintendent
EDSFI Team
Relaying
Compliance
Nuclear Plant Engineer
Asst. Hanager,
NgA
Region I
Senior Resident
Inspector
Consultant
Region I
Consultant
Region I
Region I
Consultant
B-2
'+W
~ QQ-
f. '*