ML20151A983
| ML20151A983 | |
| Person / Time | |
|---|---|
| Site: | FitzPatrick  |
| Issue date: | 07/07/1988 |
| From: | Fox E, Lazarus W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20151A975 | List: |
| References | |
| RTR-NUREG-0737, RTR-NUREG-737 50-333-88-07, 50-333-88-7, NUDOCS 8807200188 | |
| Download: ML20151A983 (24) | |
See also: IR 05000333/1988007
Text
.
..
_
.
.-
,
.. .
.
4
.
-
i
U.S. NUCLEAR REGULATORY' COMMISSION
REGION.I-
Report No.
50-333/88-07
Docket No.
50-333
License'No. OPR 59
Priority
Category
Licensee: New York Power Authority
P. O. Box 41
Lycoming, New York 13093
Facility Name: James A. Fitzpatrick Nuclear Power Plant
Inspection At:
Lycoming, New-York
Inspection Conducted: May 1672_03 1988
Inspectors:
'
Y Fox,"S
..rgency Preparedness
Date
r
S
ist, F SB, DRSS
Accompanying Personnel
'
C. Conklin, EPS
K. McBride, PNL
.
J. Ransdell, PNL
'
G. Andrews, PNL
l
G. Bryan, PLN
'
G. Martin, PNL
O
Approved by:
_
k
-?
7
W. J. Lazarps, Chief
Date
EmergencVPreparedness Section, DRSS
I
Inspection Summan :
Inspection on May. 16-20, 1988 (Report No. 50-333/88-07)
Areas Inspected:
Scope: This special, announced inspection was an emergency
response facility appraisal. Areas examined during the appraisal included a _
review of selected procedures, supporting documents and representatives records,
4
the ERFs and related equipment and interviews with licensee personnel. Selected
'
activities were observed during the 1988 annual exercise to ascertain the
adequacy of the ERFs and related equipment.
Results: No violations were identified.
One -(1) Deviation was identified
involving the failure to meet the commitments to NUREG-0737, Supplement 1, Part
8 in that the technical support center will be (1) environmentally controlled
h72OO18e89o797-
a
ADOCK 05000333
-
-
- - .
-
-
-
.
- - -
- -
- -
- -
.
'
.
2
and (2) provided with radiological protection necessary to assure that
radioactive exposure to any person working in the TSC would not exceed 5 rem
whole body or its equivalent in any part of the body for the duration of the
accident. Areas needing increased licensee attention include the need to revise
dose calculation procedures to consider calculating offsite doses for credible
unmonitored accidental release pathways; provide validation and verification
documentation for the Class A Dose assessment model; and provide a formalized
-
method for controlling and documenting maintenance of the dose assessment
models.
i
l
1
_
.
TABLE OF CONTENTS
1.0 Persons Contacted
2.0 Assessment of Radioactive Releases
2.1 Source Term
2.2 Dose Assessment
3.0 Meteorological Ir. formation
4.0 Technical Support Center (TSC)
4.1 TSC Data Availability
4.1.1
Documentation of Regulatory Guide (RG) 1.97 Vertables
4.1.2
RG 1.97 Variables Availability and Sufficiency
4.1.3
Computer Data
4.1.4
Manual Data
4.1.5
Data Adequacy
4.2 TSC Functional Capability
4.2.1
TSC Power Supplies
4.2.2
TSC Data Analysis
4.3 TSC Habitability
4.4 TSC Data Collection, Analysis and Displays
f
4.4.1
Methods of Data Collection
5.0 Emergency Response Facility (ERF)
5.1 ERF Data Displays
5.2 ERF Time Resolution
5.3 ERF Signal Isolation
5.4 ERF Data Communication
5.5 ERF Processing Capacities
5.7 ERF Model and System Reliability and Validity
5.8 ERF Reliability of Computer Systems
5.9 ERF Mutual Systems
5.10 ERF Environmental Control Systems
5.11 ERF Regulatory Guide Report
.
-
,
.
,
)
2
6.0 Emergency Opsrations Facility (E0F)
6.1 E0F Habitability
6.1.1
Backup EOF
6.1.2
E0F Reliability
6.2 EOF Functional Capabilities
6.2.1
Data Analyses Adequacy
6.2.2
E0F Data Availability
6.2.3
EOF Data Collection Storage Analysis and Display
?
7.0 TMI Action Plan Requirements
8.0 Exit interview
.
i
m
,
.
.
. .
-. .
.
..
..
.
'
.
_
.-
.
,
,
.
.
Details
1.0 ' Persons Contacted
L. Bsale
Computer Software Specialist (JKF)
I
- D. Bell-
'
Emergency Planning,'(NYPA)
-R. Burns:
- T. Coffey
Sof tware Specialist-(JKF)
-(
R. Converse
Resident Manager (JKF)
.
- G.~Varg
Radiological Engineering. General Supervisor'(JKF)'
- A.
Zaremba
EP Coordinator (JKF)
{
S. Murawski
. Computer Systems Consultant (JKF)
!
- M. Prarie
Assistant EP Coordinator (JKF)
D. Berry
Communications (JKF)
~
- J.
Elmers
Supervisory Licensing Engineer (NYPA)
S. Horvath
Technician (NYPA)
- 0.
Lindsey -
Operations Superintendent (JKF)
- S. Pobutkiewicz
Supervisor Computer Specialist (JKF)
J. Street
Technical Services (JKF)
4
R. Sullivan
Reactor Operator (JKF)
-i
i
K. Rao
Superintendent (NYPA-EE)
j
T. Galleta
Staff Meteorologist (NMP)
.
,
C. Faison
Supervisor Nuclear. Emergency Preparedness (NYPA)
,'
- W.
Fernandez
Superintendent Operations-(JKF)
,
- T. Herrman
SNR Plant Engineer (NMP)
O. Johnson
Orill Cc, troller (JKF)
- B. Mathe
Emergency Planning (NYPA)
- E. Mulcahey
RESS (JKF)
G. Tavick
Supervisor Quality Assurance (JKF)
'
W. Robinson
Quality Assurance Engineer '(JKF)
- V. Waltz
Superintendent Technical Support (JKF)
4
i
G. Tavick
Supervisor Quality Assurance (JKF)
j
Other licensee personnel were ' contacted or observed either during the-
exercise or the appraisal.
- Denotes those who attended exit meeting on May 20,.1988.
!
!
4
'
J
_
.
.
.-
..
,
FITZPATRICK ERF INSPECTION
2.0 Assessment of Radiological Releases
2.1 Source Term
The primary monitored gaseous release pathways to the environment-are
the turbine building, reactor building, radwaste building, and plant
stack. The reactor building and the refueling floor exhaust through
a common vent.
In the event the normal ventilation or stack monitors
are inoperable or off scale, the stack, turbine building, and radwaste
buildings are equipped with high-range effluent monitors. > Procedures
are in place to calculate source term information for releases via
all monitored release pathways.
The primary unmonitored release
pathways are via the pressure ' relief panels on_ the reactor building,
the heat actuated fire dampers on the turbine' building, and through
reactor containment.
Radionuclide mixes have been taken from_ the Final Safety Analysis
Report (FSAR) to characterize five classes of accidents. These in-
clude:
steam line break; loss of coolant; containment design. basis
accident; refueling; and control rod drop.
These serve as the de-
fault radionuclide mixes for the dose assessment methods. A review
of FSAR section 14.6, Analysis of Design Basis Accidents, shows that
+he five accident sequences considered result in minimal fuel damage
.
and are therefore not representative of accidents resulting in major-
fuel damage. The exercise conducted during the week of this inspec-
tion resulted in a source term which did not fit wellLinto one of the
five default accident classes.
The default accident-class used re-
suited in an isotopic mix higher in radiciodines than was realistic.
Isotopic mixes can also be entered into the models from post accident
sample and grab sample results.
Procedures EAP-18, Protective Action Recommendations, and RTP-46,
'
Core Damage Estimation / PASS, contain precalculated relationships
between various plant parameters and percent fuel damage. Section
)
4.10.5 of procedure RTP-46 contains a reference to curves of hydrogen
i
concentration versus percent cladding failure, however, the procedure
does not contain these curves nor does it direct the user to where
they can be found. Procedure EAP-18 provides a relationship between
the drywell high-range radiation monitoi' and percent fuel damage.
However, no procedure was found that allows the use of the high-range
drywell radiation monitors, drywell technical specification leak
rate, or temperature and pressure to determine a source term for use
i
in dose calculations. Discussions with licensee personnel indicate
that they had already recognized this need and are in the process of
developing a procedure to rddress this.
'
Based on the above review this portion of the licensee's program
is adequate,
j
i
I
J
-
- -
-
-
, , , ,
__
,
.
.
.
..
.
.
2
.
The fellowing item will be reviewed during a subsequent inspection..
Complete work on a' procedure which will. allow the use of real
time plant parameters to calculate an offsite dose due to-leakage
through the drywell boundary.
Some of the parameters to consider
are drywell highfrange radiacion monitors, drywell temp'rature,
e
and drywell pressure. (50-333/88-05-01)
The following items are suggested for program improvement:
Consider developing radionuclide mixes for use as source _ term
input to the dose assessment models which are more.representa-
tive of the range of accidents normally considered for_ emergency
planning purposes.
For instance, . consider mixes which cover the
range of severe accident sequences resulting-in major fuel
damage to minor accidents resulting in releases near the normal
levels of the reactor coolant (such as the release presented in
the exercise scenario).
In Procedure RTP-46, Core Damage Estimation / PASS, either include
the hydrogen versus percent. fuel damage curves' referred to in-
section 4.10.5 or include a referen'ce to wheie they can be located.
2.2 Dose Assessment
.
<
At present there are four dose assessment methods available to licensee
personnel.
The primary model for use in the Technical Support . Center
(TSC) and Emergency Operations Facility (E0F) is a Class A atmospheric
transport and diffusion model which is part of the Meteorological
Monitoring and Radiological Assessment System (MMRAS).
The backup
model and the primary method for use in the Control Room (CR) is the
Initial Dose Assessment Calculation (10AC) model. The dose assessment
calculator provides a manual means of performing dose assessment and .
a n.anual method based on output ' om IDAC is also available.
Proceduce
EAP-4, Dose Assessment Calculations, deals with all four methods.
All of the models examined employed well-accepted dose calculational
methods. During the review of~ procedure EAP-4, it was noticed that
the manual method based on the IDAC model outputs, contained in tables
~
4.7-4.lb, was not described or referenced in the body of the procedure.
Also, section 4.4.3., Back Calculation of Release Rate, states that
it is possible to back calculate the release rate from field data,
however, it does not describe the-method or provide an appropriate
reference.
A review was conducted of the valiaition and verification documenta-
tion for the MMRAS dose assessment model. The current documentation
consists of several sets of test cases. The test cases are mostly
handwritten and poorly organized. .It is not clear'what has been
tested in each case or what the results of the test were,
The docu-
mentation contains no description of the validation and verification
-
-
'
.
.
3
.
methodology or a description of an overall test plan.
In general,
the documentation is incomplete ano it was not possible from the
review to determine whether the methodology used was adequate to
validate and verify the model.
Some limited calculational comparisons between IDAC and MMRAS and
between IRDAM and the dose calculator have been performed by the
licensee.
However, no fully documented comparison between the primary
and all backup dose assessment models has been made nor has there-
been a comparison between the primary model and the State and NRC
models. Documentation resulting from such a comparison should include
a discussion of the differences identified and relate them to the
differing methodologies employed. Discussions with licensee person-
nel revealed that no formalized method existed for controlling and
documenting the maintenance of the dose assessment computer source
code (i.e. MMRAS and 10AC).
The Meteorological Monitoring and Radiological Assessment System (MMRAS)
contains an atmospheric dispersion computer code developed by Energy Im-
pact Associates. The model is based on the plume element concept (as
discussed in Regulatory Guide 1.111) with a wind field that varies temp-
orally and spatially.
Plume elements are described using a three-dimen-
sional puff. The model has the capability of handling continuous or non-
continuous elevated and ground level releases.
Plume diffusion is computed
using a split-sigma approach.
The split-sigma approach uses the standard
deviation of the wind direction (sigma theta) over a specified period of
time for the horizontal diffusion and the vertical temperature gradient
(delta-temperature) for the vertical diffusion.
This apprcach is encour-
aged as it provides a better estimate of the horizontal diffusion of the
plume elements. Ground-level relative conc?ntrations are computed using
the standard Gaussian equation adjusted for the puff concept.
The model also computes the Thermal Internal Boundary layer (TIBL)
which is common to most co mcal areas during onshore flow.
In general,
the TIBL delineates the boundary between the stable marine atmosphere
and the unstable land atmo phere.
Defining the TIBL is essential to
providing credible diffusion parameters to the plume elements. Also,
the model has a number of features to handle effects on the plume.
Some of them are:
plume rise during stable, neutral, and unstable atmospheric
-
conditions;
- buoyant plume rise;
- adjustment in the stability class at the lake / land 'nterface
(e.g.,TIBL);
adjustment for night stability and meandering;
-
building wake effects;
-
- dry and wet deposition; and
cooling tower entrainment.
-
-
.
.
.
-
..
_
_--
'
,
.
.
.
,.
.
.
4
.
Graphical output from the atmospheric dispersion model displays all the
necessary information needed to interpret plume movement._ However, the
,
time of day for which the plume plot is_ valid is not included on the -
graphical output. Strong con:;ideration should be given tc ' adding .the time
of day to the graphical output for which the plot is' valid. This may
prevent misinterpretation of the valid time of the plume plot.
Based upon the above review this portian of the licensee's-program
is.adequste. The following items will be revuwed in 3 subsequent
-
inspection:
Insure that a thorough validation and verification has 'been
performed for MMRAS and provide comprehensive documentation
(50-333/88-05-02); and
Develop and implement a software control procedure to insure
+
adequate control and documentation is maintained for the dose
v
assessment software models throughout their useful lifetimes-
(50-333/88-05-03).
The following items are. suggested for program improvement:
Revise procedure EAP-4, Dose Assessment Calculations, to include
in the body a description.of, and reference to, the m aual
.r.alculational method contained in tables 4.7-4.10.
Perform and document a calculational comparison between primary
and backup dose assessment models and.between these models and
the State and NRC models.
The resulting documentation should
include a description of the differences identified and a dis-
cussion relating them to the differing methodologies employed.
3.0 Meteorological Information
Onsite meteorological-data for the James A. FitzPatrick (JAF) nuclear
power plant are monitored from primary, backup, and inland towers. . The
primary tower is located between the James.A. FitzPatrick and Nine~ Mile
Point (NMP) nuclear power plants and approximately one half mile from the
shore of Lake Ontario.
The two plants are adjacent to each other.
Sensors
on the primary tower measure the following atmosnheric phenomena:
30-ft
100-ft
200-ft
Wind Speed
Wind Speed
Wind Speed
Wind Direction
Wind Direction
Wind Direction
Sigma Theta
Sigma Theta
Sigma Theta
.
Temperature
Atmospheric Pressure
l
Dew Point
'
In addition, there are two delta-temperature (Vertical ~temperatur grad-
J
ient) sensors for the 30-ft to 100-f t levels and the 30-ft to 200-f t
1
levels.
The backy tower is located on JAF plant grounds approximately
j
1
2
.
'
.
5
100 to 200 yards from Lake Ontario.
The backup tower measures wind speed,
wind direction, and sigma theta at the 90-ft level. The inland tower is
located at the Fulton Airport next to the Emergency Operations Facility
(E0F) approximately ten miles inland
The inland tower measures wind
speed, wind direction, and sigma theta at the 30-ft level.
The immediate surroundings near the towers are clear of obstructions
(e.g., foliage and buildings) and sensors on all the towers are well
exposed, thus providing representative and reliable measurements of the
"free" atmosphere near the JAF site.
All of the towers are grounded for protection from lightning and contain
current-surge protectors to protect sensors from large spikes of electri-
cal current.
Tower electronics are housed in sheds at the base of the
towers.
The sheds prov.de adequate environmental control for protection against
adverse weatter.
Electronics placed outside of the sheds contain heaters
to prevent f"eezing during the winter months.
Voltages from the sensors
are cor erted to digital form at the sheds.
The digital data are then
sent via dedicated lines to Microcat computers located in the Technical
Sup.nort Center (TSC) where the data are validated and formatted.
Valida-
t1on software was checked and it appeared sufficient for validating the
meteorological data. The meteorological data are averaged for two time
periods:
fifteen minutes and hourly.
The data are stored in fif teen-minute
and hourly databases on a Data General (OG) MV8000 located with the Micro-
cat computers in the TSC. At least two years of fifteen minute and hourly
meteorological data are kept on-line for immediate use.
Meteorological datr are avv iable in the Control Room from a terminal
which is connected ts +he DG MV8000. As a backup to the terminal, wind
speed and direction and Celta-temperature information at various levels
j
are available from the # mary tower on strip chart recorders.
The read-
'
outs are available in the TSC. The delta-temperature information are
difficult to read off the strip chart due to the plotting of other vari-
ables (e.g., temperature).
In the event the computer averaged
l
delta-temperature data is inaccessible during an emergency, it would be
'
difficult L determine precisely a fifteen minute or hourly average for
the delta-temperature.
Serious consideration should be given to dedicat-
ing one strip chart to plot the delta-temperature.
Wind direction displays on strip chart recorders in the TSC and Control
Room are inadequace during meteorological conditions that yield wind
directions from the north (i.e., 360 degrees).
Apparently, Niagara Mohawk
has modified their wind direction sensors resulting in an increase in
voltage level.
The current strip chart recorders (Tigraph 100A) are
unable to properly handle this change in voltage level.
The resulting
trace on the strip chart is one that cwers the entire scale thus making
-
._.
. . _ .
_
-
'
.
,
.
.
,
.
6-
'
it impossible to determine a fifteen minute or hourly average during an
emergency. TheNewYorkPowerAuthority(NYPA)isawareofthisproblem
and has 'taken action to correct.it (see JAF Nuclear Safety Evaluation.No.-
However, at the time of the inspection'the problem still
, existed.
Historical records (1985 through 1987) showed that the. meteorological data
from all the towers are reliable.
In most cases, the data. exceeded the
ninety-five percent data recovery level 1.for 1987. This implies the meteor-
ological sensors and related instrumentation are well maintained.
In
addition, the meteorological sensors and instrumentation are calibrated on
a semi-annual basis. Historical-calibration records ~were checked and=the
.
records correspond with the semi-annual schedule.
Meteorological data from other sources are available through Weather
'
Services Incorporated (WSI). The WSI-service is available in the EOF.
WSI provides access to all National Weather Service products for the
United States. This servica is more than adequate for providing accurate
current and forecast meteorological data for dose assessment _ purposes.
,
During an eniergency requiring dose assessment, the NYPA corporate meteor-
ologist interprets WSI forecast data from his White Plains,'New York
office and provides that information to the TSC and E0F via telefax.
Also, he has the capability to review plume projections produced by the
Meteorological Monitoring and Radiological Assessment System.
Based upon the above review, this portion of the licensee's program is
adequate.
'
The following items are suggested for program improvement:
i
One strip chart should contain just the delta-temperature information to
improve readability.
It may be necessary to obtained a fif teen minute
or hourly average when the primary source -is not available; and
'
The problem with,the wind directions on the strip chart recorders in-
the TSC and Control' Room should be corrected as soon as possible in-
order to provide adequate backup to the delta-temperature information..
4.1 TSC Data Availability
4.1.1
Documentation of Regulatory Guide (RG) 1.97 Variables
The Safety Evaluation Report <(3/14/88) determined that the
licensee conformed to RG 1.97 Rev. 2, or adequately justi-
fied deviations therefrom, except in the case of neutron
flux. There, the licensee has installed category 2 instru-
,
M
.
.
'
.
.
7
.
mentation, some of which is not environmentally qualif ted,
whereas RG 1.97 specifies category 1 instrumentation.
Existing instrumentation was accepted for interim use
pending implementation of a fully qualified system.
4.1.2
RG 1.97 Variable Availability and Sufficit.acy
~
The licensee committed to Revision 2 of RG 1.97, with some
exceptions.
For instance, no in-core thermocouples are
installed. When the Revision 2 requirement for BWR in-core
thermocouples was deleted from RG 1.97 Rev. 3, the licensee
committed to Rev. 3 for this parameter, thus resolving the
issue.
Meteorological data is available from trend recorders and
the MMRAS computer systems.
Remaining RG 1.97 variables
are available in the TSC from computer systems, principally
the Emergency and Plant Information Computer (EPIC), except-
for:
Variables not readily connected to EPIC (e.g., chemistry
-
sample results, data from portable radiation survey
meters, etc.); and
'
Variables to be added to EPIC prior to the end of the
-
next refueling outage at the end of CY 88 (e.g.,
containment isolation valve position and cooling water
flow to ESF components).
Based upon the above review this portion of the licensee's
program is adequate.
,
i
4.1.3
Computer Data
)
At the time of the inspection, the licensee was operating
two computer systems in parallel on some applications
'
following the relatively recent implementation of the EPIC
i
system.
The Honeywell Plant Process Computer System will
be phased out after successful' parallel operations.
Cur-
4
rently, it is not a TSC/ EOF data source.
EPIC uses a fiber
optic cable to link the data acquisition equipment, a VAX
,
11/785, to the EPIC host, another larger VAX 11/785.
Two
j
redundant trains feed the failover~ monitor switch. Output
is provided to the E0F and the programmer's terminal via
dial up access, to the Control Room and the Reactor Engineers'
office by fiber optic links, and to the computer room and
the TSC by direct wire linkage. Output is also provided to
,
disk and tape on off-line storage devices.
The EPIC and
'
the Safety Parameter Display System (SPDS), a subset of
EPIC, were operated by the licensee and the inspectors
t
.
.
_
.
.
,
.
,
.
.
.
-
4
.
.
..
.
.
8-
.
during the appraisal. Although'the systems were not evalu-
ated in depth, the inspector found them generally user.
friendly, except that some displays were cluttered. -System
'
time response was excellent; most screens were painted in
2-4 seconds. Computer points which were invalid or of poor-
quality had been removed from scan to preclude misleading:
operators by erroneous data. No incorrect alarm displays'
were noted.
EPIC can be switched to non plant data-for use-
in drills.
Except for an inconspicuous screen flag,-that~-
shift.is transparent to the user. Therefore, the licensee
has adopted the practice of' posting portable signs.at each-
terminal when the system is switched to non plant-_ data.
Although the inspector concurred with this asian interim
solution, it is recommended that the screen flag indicating
shift to non plant data be made more conspicuous. The
Safety Evaluation Report (SER) (3/18/88) determined there
were no serious safety questions with the SPDS design. The
SER also directed that containment radiation, containment
combustible gas, and source range monitor' input be added to
the SPDS.
Containment radiation is currently-available on
the system and containment hydrogen'(and other- gases re-
quired) will be added during the first half of 1989.
Based upon the above review portion of the licensee's
'
program is adequate.
!
The following item is suggested for program improvement.
.
The EPIC screen flag indicating that the display
contains non plant (drill or exercise)-data should be
made more conspicuous.
4.1.4
Manual Data
Telephone communications and status boards provide an
adequate backup to the primary computer based information
systems.
4.1.5
Data Adequacy,
2
All RG 1.97 variables are available in the TSC. The wind
direction trend recorders range from 0-360 degrees true.
As a-result, when the wind oscillates.near 0 degrees, the
'
recorder is useless. A design change has been_ authorized
I
to eliminate this problem by extending the scale to.540
degrees tre e. The inspector determined that the' variables
provided ts the TSC were adequate to allow the staff'to
perform its function.
)
- - - - - - -
- - -
--
-
- - - - -
-
-
- --
- - - - -
- - - - - - -
-
-
.
.
-
.
.
.
9
Based upon the above review, this portion of the licensee's
program is adequate.
4.2 TSC Functional Capabilities
4.2.1.
TSC Power Supplies
TSC electrical power is supplied from lighting panel LPA3.
Two alternate non-vital power paths exist.
The preferred
path is from the 115 KV switchyard service transformer T2,.
through non-vital 4160 VAC bus 10300 and transformer T9 to
600 V bus 13300 and MCC C332, then.to.11ghtirig panel LPA3.
Alternately, power n.ay be fed from 115 KV transformer T3
via 4160 VAC bus 10400 and transformer T10 to 600 V bus
13400, then across to bus 13300 and MCC C332. No diesel-
generator backup is provided. The EPIC computer. system is
posered from a dual unit uninterruptable power supply (UPS)
installation-located in the Administration Building HVAC
room. At the request of the inspector, TSC power was
killed.
No attempt was made to shift to alternate power.
The EPIC and its TSC terminals survived without rebooting.
The balance of the TSC went dead with the. exception of;
battery powered emergency lighting.
Lighting was more than
adequate to allow the TSC staff to relocate safely. With
the exception of the New York 1 telephone T-1 off-site con-
nection lines,:all phone circuits were provided with on-
site backup battery power supplies including the alternate
to T-1, the SLC-96 system.
The inspector.noted one' serious
,
,
weakness which jeopardized'the reliability of the plant
telephone system. All telephone voice and data communica-
tions circuits pass through the Administration Building PBX
room. The room is equipped with a wet stand pipe fire
suppression system.
If the PBX room were flooded or gutted
by fire, the FitzPatrick site would experience a communica-
tions blackout, except for one outside security phone at
the gate house, GAITR0!!ICS (plant page system), UHF radio,
and Messengers.
Although the EOF would survive, it's effectiveness would be
severely degraded by the loss of communications with the
site.
The inspector found that the TSC power supplies were
adequate to insure a high probability that the TSC would
operate uninterrupted and that the TSC could perform EOF
functions until the EOF were activated.
Based upon the above review, this portion of the licensee's
program is adequate.
The following item is suggested for program improvement:
.
.
.
.
-
.
.
.
~ '
.
.
'
.
4
.
.10 :
,
.
-Selected telephone lines should be rerouted to bypass
the PBX 1 room in order to enhance survivability in-the
event of a loss of the PBX room.
4.2.2
TSC Data Analysis
In addition to the systems already described, the TSC is-
outfitted with a complete set of plant drawings, reference
material, and other data analysis aids.
Based upon-the above review, this portion of the licensee's.
program is adequate.
4.3
TSC Habitability
'
The only documentation supplied concerning TSC habitability was
contained in a report titled, Post Accident Shielding Analysis (Deems
Report) - Response to NRC NUREG-0578 Item 2.1.6.b,. dated March'29,
1981. The report contains a general statement that.the TSC shielding
is adequate to meet NRC requirements and guidance.
The report does
not contain the supporting calculations nor does it describe the
specific methodology used to arrive at the statement of adequacy.
Neither the FSAR or the shielding report provide detailed information
concerning TSC construction.
The inspectors did note'a large number
of windows in the west wall of the Administration Building which
contains the TSC.
It was clear from the review of the shielding
1
i
report that, although radiation sources due to plant systems were
considered in the analysis, the immersion of the TSC in an infinite
cloud of radionuclides was not.
The licensee was unable to provide
>
further supporting documentation during the inspection.
Due to the
.
concerns over TSC construction, the methodology used.in the shielding
analysis report which appeared not to consider the glass west wall
and the add on shield wall for the condensate storage tank outside
the TSC, and the lack of supporting documentation for conclusions
,
reached in the study, it was not possible.for the inspectors.to
determine the adequacy of the-TSC shielding..
The TSC is in a separate habitability envelope from that of the
,
control room.
The inspector reviewed OP-59B, the TSC_HVAC procedure,
examined the TSC boundaries and'the HVAC system,'and witnessed repet-
itive mode shifts from normal to emergency ventilation. The follow-
ing was noted:
I
'
The TSC ventilation system was inoperative in the emergency
-
mode. As a result of damper failures, the system inducted
-identical amounts of external air via the unfiltered path in
both modes of operation;
HVAC components were poorly labeled with felt pen; no open/ closed
i
-
damper markings are provided;
.
-
-
-
-
-
-
-
.
-
- -
-
-
-
-
-
-
-
.
'
'
'
'
.
,.
.
11-
-
One motor operated damper (MOD 107) could not be located,-
although it's operation was verified by air flow changes;
.
,
-
Damper A00 146 is contained entirely inside the duct; no'exter--
nal position. indication is available. A ladder is required to
open the inspection plate to verify damper position. No ladder
is available on the third floor ~of the Administration Building;
-
Although the HEPA and charcoal filters are tested periodically,
the total system is not included in a-praventative maintenance
program nor is it subject to' periodic testina;
The elevator is not marked as a TSC boundary, although all other
-
accesses are so marked; and
The procedure provides sufficient instruction to permit shifting
-
ventilation modes but provides no method to verify that the
shift has been successful. The change in pressure alone.is
insufficient to verify the change over to emergency operation
since the system displayed the same pressure differential in
both modes of operation.
.
Based upon the above review, this portion of the licensee's program is
not adequate.
The following items will be reviewed in a subsequent inspection:
TSC shielding analysis documentation is insufficient to deter-
mine the adequacy of.TSC shielding; and
The TSC HVAC system is not currently being maintained in an
adequate state of readiness (50-333/88-05-04). This is a
deviation.
The following items should be considered for program improvement:
Place the TSC HVAC system on a routine preventative maintenance -
I
progran; and
'
Insure that the TSC HVAC system is fully operationally checked on a
periodic basis.
4.4 TSC Data Collection, Storage, Analysis, and Display
Licensee system hardware and corresponding documentation was reviewed
to determine whether Emergency Response Facility (ERF) functions
would be adequately supported.
Data collection, storage, display,
and communications were found to be adequate.
'
.
-
- -
-
.-
12
4.4.1
Methods of Data Collection
Real-time data acquisition, display, and storage to support
ERF functions are performed by a distributed computer
system called the Emergency and Plant Information Computer
(EPIC).
The distributed system includes a data gathering
front end based on Motorola 68000 microprocessors, a Data
Acquisition System (DAS) based on a DEC VAX 11/785, a host
data display system based on a DEC VAX 11/785, and a Micro-
VAX II to perform MONIC0RE calculations (core performance
computations).
The data acquisition, host, and MicroVAX II systems have
redundant processors for continuous backup.
Switchover to
backup computers can happen in three different ways:
(1) a
"watchdog" process running on the backup system monitors
the primary system - if the primary system fails to respond
to requests within a fixed time period, the backup system
automatically assumes primary system functions; (2) the
primary and backup systems can be switched using software
commands; and (3) computer panel switches can be traqually
switched to swap the primary and backup systems.
The configurations and primary functions of the EPIC com-
puters are as follows:
COMPUTER PRODUCTS INC. (CPI) - Front Ends
Motorola 68000
32 kilobytes (KB)
no disk and no tape drives
one of these units is used for each 2000 plant sensors
Function: Collect data from plant sensors, perform signal
conditioning, multiplex data, and transmit sensor data in
binary format to the OAS via a serial link
Digital Equipment Co. (DEC) - Data Acquisition System
VAX 11/785
4 Megabytes (MB) Random Access Memory (RAM)
1 - 456 MB hard disk unit (with 1 spare 456 MB disk drive)
no tape drives
Function: Collect data from CPI front ends, convert binary
values to values with engineering units, store data, and
transmit data to the Host computer for further processing
-
^
y
e ,
-
.
.
.
.
,
-
-
~
ll
13
H
'
DEC - Host System
VAX 11/785
d
-12 Md RAM
2
456 MB h'ard disk units (with 1 spare 456 MB disk drive)
2 tape drives-(1600/6250 bits per inch)
~
Function:
Receive data from DAS,. compute' calculated points
(points that are defined using different combinations of'
plant sensor data and mathematical. equations to compute:
values called C Points .or calculated _ points).
DEC - MONICORE Calculations
DEC MICROVAX II
13 MB RAM
1 - 71 MB-hard disk
1 - 680 MB hard disk
1
.95 MB tape drive
1 - 290 MB tape drive
Function: Collect data from the OAS and calculate data
values to update displays showing current reactor core
.
status (e.g., reactor state parameters,-thermal limits,
radial and axial power distributions, etc.). Also, perform
data archival to disk and tape.
The bulk of the ERF software is written'in FORTRAN 77.with
some routines written _in assembly language.
Supporting
documentation (e.g. , user's guide, programmer's reference
manual, and test acceptance documentation) was found to be
comprehensive and professionally'done.
The following is a table of analog (continuously variable)
and digital (2 state) plant sensors routinely sampled and
used to assess plant safety _' status
)
Analog
Digital
Computed
Total
Sensors
Sensors
Points
Sensors
i
1487
2009
612
4108
Based upon the above review, this portion of the licensee's-
program is adequate.
5.0 Emergency Response Facility (ERF)
5.1 Data Displays
Data displays supporting ERF functions' include:
j
.l
.
.
_.
1
-
.
.
,
t
14'
-
,
Control Room (CR)
,
2 - 19 inch CRTs for CR operators
3
.19 inch CRTs for shift supervisors
2
'25 inch slave units-(can be set to specified display)
TSC (Technical Support Center)
2 - 19 inch CRTs-
2 - color copiers (hard copy units)
1 - 600 line per minute printer
EOF (Emergency Operations Facility)
1 - 19 inch CRT
i
1 - 20 inch slave CRT
1 - color copier
1
printer
Display generation is a strong feature of this ERF supporting system.
Users are given the option of selecting displays by:
(1) pressing
function keys, (2) typing in display selection parameters on a key-
board, and (3) using a track ball (this is a ball in a socket that
can be manually rotated to control a screen cursor used to make menu
selections). Also, the display menu includes a "help" feature to
provide additional information en the use of the display system.
Displays were generated on the CRTs in 1 to 5 seccnds with no signif-
icant delays in response to display requests.
Based upon the above review, this portion of the licensee's program
is adequate.
The following item is suggested for program improvement.
J
It is the practice of the licensee to place printed labels on
ERF display CRTs during system tests or plant exercises'to
inform the user that the data displayed is computer generated
and is not real plant data.
It is suggested that displays up-
dated by computer generated data be tagged by the computer as
"Simulated Data" or any other appropriate label to avoid poss-
ible confusion.
5.2 ERF Time Resolution
.
ERF supporting computers read, analyze, and store to hard disk data'
from 3496 analog and digital sensors.
The sampling rate for. data
sets (there are 40 different scan classes) varies between 1 second
and 60 seconds for ERF related plant sensors. The data sampling rate
is considered low to' moderate speed.
The data acquisition, tasks are
"
i
- -
-
. - -
-
-
.
.
-
.
-
- -
-
-
..
r
,_
-
<
.
,
.
,
.
15
'
assigned a high priority and are not delayed by waiting on other
supporting tasks.
Further, data acquisition tasks are partitioned
1
between computer systems.
For example, signal conditioning and~
,
multiplexing is done by the CPI front ends,-data conversion to en-
gineering units is done by the Data Acquisition System'VAX 11/785,
calculated point computations and display devices are driven by the
Host VAX 11-785, and MONICORE calculations are performed by the DEC
MicroVAX II.
Based upon the above review, this portion of the licensee's. program
is adequate.
5.3 ERF Signal Isolation
,
Adequate isolation is documented by-a letter "Safety Parameter Dis-
!
play System", to J. C. Brons of the Power Authority of the State of
.
g
New York from H. I. Abelson of the NRC, dated March 18,' 1988.
Sec-
'
tion 4 of this letter states "the SPDS is suitably isolated".
Based upon the above findings, this portion of the licensee's
program is adequate.
5.4 ERF Data Communications
'
Data communications capabilities were reviewed, including the front
ends, the VAX 11/785 DAS, the VAX 11/785 Host, and the MicroVAX II
communications.
It was reported that error checking and correcting
d
functions were performed.
'
.
Modem firmware and operating system software for EOF telecommunica-
2
tions support was reported as using error detection and correction or
c
request for re-transmission on error detection. Data communications
betweta ;rocessors was reported to use high speed data links (approxi-
mately t0 megabits /second).
l
t
Data communications between the Host VAX and the EOF display devices
was via telephone modem and was reported to be 4800 bits per second.
,
This data rate is viewed as adequate, but the licensee should con-
]
sider obtaining higher speed telecommunications equipment to support
high data rates in case of emergency situations.
Based upon the above review, this portion of tr.e licensee's program
is adequate.
5.5 ERF Processing Capacities
1
The DEC VAX's, MicroVAX's and peripheral computer systems were con-
figured to support plant safety monitoring and reporting needs.
Processing is based on multitasking to allow several software func-
,
tions to be processed concurrently. Data acquisition and storage
j
.
..
.
.
.
.
.
.
_
_
..
.
.
b
.
.
16
,
.
tasks are high priority tasks and are executed before supporting
tasks. Licensee contacts reported the OAS VAX and MicroVAX to be.
H
60% loaded and the Host VAX to be 75% loaded for normal operation.
Based _upon.the above review, this portion of the. licensee's program
is adequate.
The following item is suggested for program improvement.
Further loading of the Host VAX through the addition of displays or
calculated points is expected to result in performance degradation at-
peak load times.
It is suggested that the licensee either limit
software enhancements to the current' level (don't add other calcu -
lated points or displays) or acquire additional computing capability
for the Host VAX.
5.6 ERF Data Storage Capacity
Historical data can be stored to disk such that any time period of
data will be saved.
The amount of data is limited by disk resources.
Routinely, 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> of historical data are available for trending.
Utility personnel interviewed reported that every 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />, 24 hourly
sample sets of historical data are saved to magnetic tape. . This
-
process continues for 60 days after which the oldest data tape;is
.
overwritten by a new 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> data set. This wrapping process of data
storage was reported to continue. indefinitely.
Based upon the above review, this portion of the licensee's program
,
is adequate.
l
5.7 ERF Model and System Reliability and Validity
i
i
i
Documentation for the model algorithms was reviewed. Specifically,
the "SPOS Algorithm Design Report for New York Power Authority James
A. FitzPatrick Nuclear Power Plant Emergency and Plant Information
'
Computer", report numbers CORL L01 or. NY0-83-134 < dated February 1986
)
and revised October 1986 by Energy Inc. Also, the New York Power
!
Authority conducted a "SPOS Algorithm Design Report Review", Dec. 30,
1987 and made recommendations for correction.
It was reported that
the recommended corrections were being included in the final SPOS
Algorithm Design Report.
]
Based upon the above review, this portion of the licensee's program
is adequate.
5.8 ERF Reliability of Computer Systems
Computer system availability was documented by the utility as greater
than 90% from October 1987 through the current month,
,
. .
.
. .
.
.
. .
..
.
,
.
17
..
Based upon the above review, this portion of the licensee's program
is adequate.
'
5.9 ERF Manual Systems
See Section 5.1.4
-
5.10- ERF Environmental Control Systems.
'
L
An air conditioning unit was located in the computer room and tempera-
ture and humidity were monitored and controlled. The air condition-
ing unit installed was reported to have approximately twice the
required capacity for the computer hardware being protected.
Based upon the above review, this portion of the licensee's program.
is adequate.
'
5.11 ERF R9gulatory Guide 1.97 Report
The licensee report of the implementation of RG 1.97 was available
and was adequate as an appraisal information resource.
Based upon the above review, this portion cf the licensee's
l
program is adequate.
6.0 Emergency Operations Facility (E0F)
6.1 E0F Habitability
Since the licensee's primary E0F is located outside of the 10 mile
emergency planning zone (Ep2) there are no specific habitability
requirements.
,
6.1.1
Backup EOF
i
No backup EOF is required nor is one available.
6.1.2
EOF Reliability
The EOF is powered from the Whitaker substation which is
fed from the Clay and Oswego power stations. The site is
)
,
powered from Nine Mile point #4 and Lighthouse Hill #3.
Since the EOF and the site are separated by at least one
intermediate substation, each is adequately protected from
any single localized grid disturbance affecting the other
facility.
No alternate EOF power supply is required or
installed. The system is designed to accept an emergency
diesel generator. Although none is installed, agreements
exist with two sources to provide a diesel generator upon
,
request.
The inspector concluded that the EOF was equipped
to support overall licensee response to accidents at the
<
i
plant,
-
- .
_-
_-
..
..
.
.
'
,
i
-
,
.
.
. . .
18
x.
Based upon'the above review,-this portion of the licensee's-
program is adequate,
,
6.2 EOF Functional Capabilities
6.2.1
Dat'a' Analysis Adequacy
By choice, the E0F depends almost entirely upon the TSC
'
staff.in plant assessment matters. An' EOF technica1 Estaff
member interfaces between the TSC and the.E0F. In all
other areas,-EOF data analysis is adequate to' support the
d
EOF functions.
Based upon the above review, this portion of .the licensee's
program is adequate.
6.2.2
EOF Data Availability
,
The E0F uses the same computer equipment as-the TSC.
Therefore, the comments of Section~03.04 apply to the EOF
as well as to the TSC.
The EOF telephones are supported.
from a dedicated PBX with battery backup. The majority of.
.
.
the status boards are identical to those of the TSC; where
,
there.are differences, the EOF boards tend to provide more
detailed coverage of the dose projection, PARS, and state
i
actions in response to the PARS.
Based upon the above review, this portion of the licensee's
program is adequate.
4
6.2.3
E0F Data Collection. Storage, Analysis, and Display
The same computers supporting TSC activities ~ support.the
EOF.
These systems and details of their functions have
been described above.
'
Based upon the above review, this portion of the licensee's
!
program is adequate.
7.0 TMI Action Plan Requirements
.
With regards to the TMI Action Plan Requirements III.A.1.2 (Upgrade
Emergency Support Facilities) and III.A.2 (Improving Licensee Emergency
!
Preparedness - Long-Term), the items identified in this report must be
4
resolved prior to the closure of these requirements. .
I
!
I
,
!
i
_ _ -
. .
.
~
.
.
19
.
8.0 Exit Interview
The NRC team met with the licensee representatives listed in Section 1,0
of this report at the end of the inspection.
The team leader summarized
the appraisal scope and specific areas evaluated and discussed in detail
all findings and recommendations presented herein,
Licensee management acknowledged the findings and indicated that
appropriate action would be taken regarding the identified open items.
At no time during this inspection did the inspectors provide any
written information to the licensee,
.