ML20215D751
| ML20215D751 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 05/12/1987 |
| From: | Hooks K, Konklin J, Lloyd R, Stein S NRC OFFICE OF SPECIAL PROJECTS |
| To: | |
| Shared Package | |
| ML20215D725 | List: |
| References | |
| 50-327-87-18, 50-328-87-18, NUDOCS 8706190131 | |
| Download: ML20215D751 (17) | |
See also: IR 05000327/1987018
Text
.
U.S. NUCLEAR REGULATORY COMMISSION
OFFICE OF SPECIAL PROJECTS
Report Nos.:
50-327/87-18 and 50-328/87-18
Licensee: Tennessee Valley Authority
,
6N 38A Lookout Place
1101 Market Street
'
Chattanooga, Tennessee 37402-2801
Facility Name:
Sequoyah Units 1 and 2
Docket Nos.
50-327 and 50-328
License Nos.
Inspectors:
5/12/f7
i
E. Konklin, Team Leader
Date Signed-
LAM
es arr
K.
Hooks, Assistant Team Leader
Fat'e Si'gned
afY
Shh7
,
R. L.
yd, Mel:hp6'
ipline Lead
Date Signed
I
-
I
_
$W2 ] ~
'
%. R. Stein, Electrical Discipline Lead
Date Signed
-1
Consultants:
R. Compton and M. Juister (Mechanical)
i
D. Ford and M. Good (Electrical)
Approved By:
f/fE/T7
/TeamSupport&IntegrationSection
mes E. Konklin, Chief
Difte Signed
SUMMARY
Inspection on March 16-27, 1987 and April 13-17, 1987.
(
Areas Inspected:
This special announced testing inspection was conducted in
'
the area of post-modification testing, and included review of test procedures
and test results, observation of work activities associated with functional
<
and surveillance testing, review of system evaluation reports associated with
I
the Design Baseline Verification Program (DBVP), and a review of the
separation of electrial circuits.
i
"
8706190131 870610
ADOCK 05000327
G
i
.
Results:
In the areas inspected, one violation was identified with regard to
the control of post-modification testing.
The examples included in the viola-
tion are discussed in Sections 4 and 5 of the report.
1
l
\\
_ - _ _ _
,
.
REPORT DETAILS-
1.
Key Licensee Personnel Contacted
NAME
ORGANIZATION
J. Dorris
Design Basis Verification Program (Electrical)
- 6T. Flippo
Quality Surveillance Supervisor
- 6T. Hughes
DNE Principal Electrical Engineer
GN.'Kazanas
Director, Nuclear QA
- G. Kirk
Compliance Licensing Manager
C. Lagasse
Craft Foreman
9J. LaPoint
Deputy Site Director
- R. Mooney
Systems Engineer Supervisor
G. Newman
Electrical Craft Superintendent
D. Peters
Modification Supervisor
- 6M. Purcell
Licensing Engineer
- sA. Qualls
-Plant Manager'(on detail'to SQN)'
- H. Rankin
Project Manager
- M. Sedlacik
Modification Group A Supervisor
- J. Staub
DNE Electrical Engineering Branch Supervisor .
E. Steinhauser
Design Basis Verification Program (Mechanical)
- R. Stockton
Modification Group A Engineering Supervisor
Other licensee employees were contacted, including supervisors, operators,
shift engineers, electricians, and maintenance personnel.
The NRC
resident inspection staff was represented at both exit meetings.
- Present at exit meeting on March 27, 1987.
9 Present at exit meeting on April 17, 1987.
2.
Followup on Previously Identified Items
The inspectors reviewed two unresolved items at the request of the
resident inspector,
a.
(Closed) 327,328/86-08-08: This item concerned WP 12365, which was
issued to replace Cable 2V2451A to 2-FCV-68-333, for ampacity con-
siderations.
The cable was routed from the Reactor MOV Board 2Al-A,
Panel 9D to Penetration 15 at Azimuth 101 on the 693 level.
FCR 5142
was written to correct cable routing cards since the work plan
required the cable to change trays.
The tray change was not speci-
fied on the routing cards that were prepared to support the work
plan.
The installation was inspected by the NRC inspectors in the
480 Board Room 2A and the Unit 2. Mechanical Equipment Room in the
Auxiliary Building. The cable was routed from Tray HT-A, Node 126 to
Tray HU-A, Node 123 in the Mechanical Equipment Room and then from
.HU-A, Node 122 back to Tray HP-A, Node 127.where'it connects with
Tray HT-A in the 480 Board Room 2A.
The trays were separated by
about 12 inches vertically and the cable routing was appropriate.
No
deficiencies were noted. This unresolved item is closed.
f
b.
(0 pen) 327,328/86-49-06: This open item concerned potential cable
tension and tension on cable terminations in the control room because
of long vertical runs of Flamemastic coated bundled cables in the
cable spreading room. The inspectors found no problems with the 68
j
cables that were inspected in the risers of control room panels. All
'
cables had adequate slack, and the terminal lug to wire connections
were not under tension.
During the inspection, it was found that TVA
is conducting an independent review of an employee concern in this
area. Therefore, this item was not closed by the inspectors.
1
3.
Scope of Inspection
The inspection team reviewed selected documents which specified test
requirements, including seven surveillance instructions (sis), ten
maintenance requests (MRs), four workplans (WPs), and two technical
instructions (tis). The inspectors also observed testing activities and
verified the reporting of test results for a total of twelve functional
tests, ten surveillance tests and two instrument calibration tests.
In
addition, two safety-related cable pulls and the makeup of six Raychem
splices were observed.
In conjunction with the above test inspection activities, the inspectors
also reviewed sixteen system evaluation reports generated by TVA as part
of the Design Baseline Verification Program (DBVP), and inspected a sample
of 75 cables in the control room and the auxiliary building for potential
secondary bridging violations. The results of the system evaluation
report review are discussed in Section 6 of this report.
The inspection
of cables for secondary bridging is discussed in Section 7.
4.
Electrical Inspection Results
The NRC inspectors monitored the performance of selected plant functional
and surveillance tests in the areas of electrical and instrumentation
modifications.
Sequoyahworkrequestpackages(WRs)providethebasic
criteria for the component or system modifications and functional testing.
Surveillance tests are generally accomplished in accordance with appli-
cable surveillance instructions. Tests asse sated with five WRs were
observed. The review of the WRs concentrav.d on the functional testing
aspects of the packages.
In addition, functional tests associated with
j
selected modification workplans (WPs) and maintenance requests (MRs) were
reviewed and observed.
The observation of functional testing by the NRC inspectors indicated that
weaknesses exist in the licensee's testing program.
Based upon the
deficiencies observed, the NRC inspectors concluded that Sequoyah func-
tional testing associated with MR/WR work activities lacked sufficient
controls to assure that modified components are tested for full perform-
ance of their intended functions.
In the foliowing sections a brief
description of each test is provided along with the inspectors' observa-
tions regarding the adequacy of associated test documentation, test
performance and test results evaluation.
l
-2-
)
.
a.
Functional Testing of CVCS Valves.
WRs B217164 and B217156 details instructions for the functional test
of Chemical and Volume Control System (CVCS) Letdown Isolation Valves
2-FCV-62-69A and 2-FCV-62-70A.
Testing of these components was
required because of the modification of existing system circuits
to facilitate the replacement of Raychem splice materials. Addi-
tionally, surveillance testing of valve stroke and timing was per-
formed in accordance with Sequoyah Surveillance Instruction (SI)
166.6.
The functional tests were performed by operating the valves from
Local Control Stations 1361-8 and 1365-B.
Indicating lights were
observed to verify valve travel to the " fully open" and " fully
closed" positions.
The inspectors' review of the above WRs verified that functional
testing had been performed in accordance with requirements.
Valve
positioning and confirmation of associated relay pickup were accom-
plished as specified in the WR test instructions.
Surveillance
testing was accomplished through remote signal initiation from the
indicating switches located in the control room.
Test results
1
indicated that timing and valve stroke data were in compliance with
the criteria specified in SI 166.6
No deficiencies were identified during performance of this test.
However, the inspectors noted that over 50 percent of the indicating
lights located in Local Control Stations 1361-B and 1365-B were
,
broken, missing or burned out.
To perform the functional test, the
cognizant engineer provided the required indicating lights by remov-
ing bulbs from other local control stations in the area. Although
this method of replacement accommodated the needs of the specific
test, it is poor test practice.
Faulty or missing bulbs should be
replaced with new bulbs to assure adequate indication at all control
stations.
b.
Functional Testing of Containment Isolation Valves.
WRs B219442 and 8219444 provided instructions for the functional
testing of solenoid-operated Containment Vacuum Relief Isolation
Valves 2-FSV-30-46A and 2-FSV-30-468.
Testing of these components
was required because of the modification of existing system circuits
to facilitate replacement of Raychem splice materials.
Functional
and surveillance tests were performed by the cognizant engineer and a
reactor operator using the remote indicating switches located in the
control room.
l
1
During the observation of the functional test the inspectors identi-
fied the following concerns:
i
1.
The WR package contained formal instructions which detailed the
steps required for successful completion of the functional
test.
Several handwritten changes had been made to the approved
test package; specifically, the annotation of splice and test
instructions without signatures, initials or subsequent approvals
-3-
_
l
.
as required by Sequoyah Procedure SQM-2 " Maintenance Management
System."
2.
Step 1 of the functional test instruction for each of the two
containment vacuum relief isolation valves incorrectly speci-
fied that the valve solenoid actuator should be de-energized
when the valve is placed in the fully closed position. The NRC
inspector's review of applicable TVA Schematic 45N630-10 indi-
cated that the solenoid should energize at valve closure.
Consequently, the test instruction provided information which
conflicted with approved design documents and which was techni-
cally incorrect.
3.
Following completion of the functional test, the inspector
reviewed the test instruction and schematic with the cognizant
engineer and discussed the conflict noted above.
The cognizant
engineer later modified the approved test instructions by elimi-
nating portions of Steps 1 and 2 in order to reflect actual test
results. This action is contrary to the requirements of the
controlling procedure, SQM-2, and indicates a lack of
familiarity with plant administrative requirements.
These examples of changes to approved test packages without required
reviews, inclusion of incorrect requirements in a test instruction,
and after-the-fact changes to a test instruction to reflect the test
results, indicate a lack of test control as required by Criterion XI
of 10 CFR 50 Appendix B (50-327/87-18-01; 50-328/87-18-01).
c.
Functional Testing of CVCS Isolation Valve 2-FCV-62-91A.
WR 219655 details instructions for the functional test of valve
2-FCV-62-91A. Testing of this component was required because of
modification of existing system circuits to facilitate the replace-
ment of Raychem splice materials.
No deficiencies were observed in
this area.
d.
Inspection of Damper Covers for the Fifth Vital Battery.
WP 10968 included the manufacture and installation of two ventilation
damper covers associated with the installation of the fifth Vital
Battery.
No deficiencies were noted.
e.
Phase Testing and Motor Bump Tests for the Component Cooling System
Swing Pump.
WP 12337 covered the replacement of cables from the switchgear to the
component cooling swing pump for both the normal and alternate power
supplies.
Inspectors observed the phase testing at the normal /
alternate power transfer switch, lead makeup, and the motor bump test
at the swing pump. The motor had been decoupled from the pump as a
precaution to prevent potential reverse rotation of the pump during
the motor bump tests should the wiring be in error. The following
concerns were identified by the inspectors:
-4-
-
.
.
1.
Phase checking of the normal supply to the motor could not
initially be accomplished since the transfer switch normal
supply breaker would not close.
The breaker mechanism would not
engage with the handle. The breaker was finally closed by
manipulating the interlock mechanism behind the breaker. A
review of WR B127722, which was posted on the breaker, indicated
that repair of the breaker mechanism was a priority one work
request that had been outstanding for six months.
During phase checking of the alternate transfer supply, the
alternate transfer switch breaker could not be closed without
immediately tripping.
Investigation revealed that the remote
switches were misaligned.
This also occurred during attempts to
shut the alternate supply breaker at the 480 VAC switchboard.
The apparent cause was that the test procedure in the work-
plan was not detailed enough to specify a proper switch align-
ment prior to shutting off the alternate supply breaker and the
transfer switch breaker. The test discrepancy was not recorded
by test personnel.
2.
During motor bump tests the swing pump motor was inadvertently
started and ran for approximately 20 seconds until it was
secured by test personnel with the local control pushbuttons at
the motor. The inadvertent start of the decoupled motor threw
potentially contaminated grease from the coupling outside the
contamination zone onto two personnel observing the test.
Although no personnel were injured, the potential for personnel
injury existed because of the close proximity of personnel to
the motor / pump coupling. The apparent cause was that the test
procedure was inadequate.
No other deficiencies were noted.
The examples of inadequate procedural requirements, and a failure to
record a test discrepancy, discussed above, are considered to be
further examples of a lack of proper test control, as designated in
Section 4.b above.
f.
Calibration of Upper Head Injection System Level Setpoint Switches.
WP 12385 involved removing level switch channel modules from the
control room, conducting a channel calibration in the instrument
shop, replacing the modules and then conducting a "zero" and " span"
adjustment in the system.
The calibration was conducted on two BLH
Electronics Tranducer Indicators (Model 450). Transducer input was
simulated with a Precision Calibrator Model 625, Serial 435659, and
power supply checes were done using a Fluke 8600A digital multimeter
Serial Number 432927. All test instruments were in calibration.
Power supply checks were also observed. The module was warmed up but
did not stabilize, indicating a problem with the module.
The power
supply leads and cables were checked by test personnel, but no
problem was foun,i. The calibration was later run satisfactorily on
the same module. A loose cable was suspected, however, no root cause
of the problem wis dets mined and the problem was not logged on the
work plan. No otaer deficiencies were noted.
-5-
-
n
.
l
g.
Functional Testing of Safety Injection Valves.
,
The inspectors observed. functional and stroke timing tests on Boron
Injection Tank Isolation Valves 2-FCV-63-25, 2-FCV-63-26,.2-FCV-63-39
.
and 2-FCV-63-40. The functional tests were completed under
j
Maintenance _ Request Packages B219665 through B219668. Stroke timing
j
tests were conducted using SI 166.4.
The functional test was con-
trolled from the Control Room and involved mid-positioning each
valve, applying power to the valve and ensuring it did not move,
cycling the valve from the local control station and observing proper
operation, and then performing the stroke timing test from the
,
control room. The Senior Reactor Operator decided to complete the
test one valve at a time since the valves were the only normally-
closed valves between the charging pumps and the reactor vessel. The
following was observed by the NRC inspectors during the testing:
1.
Three of the four Boron Injection Tank isolation valves were
energized prior to placing the valves in the mid-position. This
was contrary to the signature-verified steps in the functional
test procedure.
3
2.
The Senior Reactor Operator operated one valve with a hold order
tag still attached to the valve control switch in the Control
Room.
3.
Steps in the functional test procedure were not signed off as
each step was completed, which may have contributed to the
procedural violations identified above.
These examples of failure to follow procedures are considered to be
further examples of a lack of test control, as designated in Section
4.b above.
j
l
h.
Installation of Raychem Splices at Penetration No. 37.
4
Raychem splices were installed on four conductors of Cable 2V55998-
under MR 8217399.
The splices were installed at Penetration No. 37
(inboard containment penetration). The conductors had been pre-
viously prepared for splicing by a different work package.
During
installation, a butt connector was installed on the wrong lead for
one conductor but was promptly recognized by the electrician and a
quality control inspector. Work was stopped while a new connector
was obtained. After certification of the new connector by the
quality control inspector, the job was satisfactorily completed. No
deficiencies were noted.
1.
Functional and Stroke Timing Test of Reactor Building Instrument
Room Air Conditioning Isolation Valve.
A functional and stroke timing test of the Reactor Building Instru-
i
ment Room Air Condition Isolation Valve, 2-FCV-31C-232, was conducted
j
under WR B212284. Two-way radio communication between the Control
l
Room and the remote station during the test was marginal and resulted
in having to. stroke the valve three times in order to properly time
the valve for the test.
No deficiencies were noted.
-6-
i
..
i
l
i
j.
Observation of Spent fuel Pump Motor Bump Test.
WP 12424 replaced the power cable to the Spent Fuel Cooling System
Pump because of ampacity considerations. At the time of this inspec-
J
tion, the new cable had been routed in the raceway but had not yet
1
been terminated at the motor.
The motor bump test was conducted
.
with the original power cable to determine correct motor rotation
to assure proper phase connection when the new cable is connected to
the motor. This was being done because the test procedure called
for a motor bump test or for observation'of a permanent rotation
indicator on the motor and pump assembly, and test personnel were
unaware that an indicator existed. However, during setup for the
test, the test engineer found a permanent direction arrow on the
pump which made the motor bump test unnecessary.
No deficiencies
'
were identified.
t
k.
Functional and Stroke Time Test of the Upper Head Injection Test
Line Isolation Valve.
i
A functional and stroke time test of the Upper Head Injection Test
Line Isolation Valve, 2-FCV-87-8 was conducted under WR B217159. The
functional test and stroke timing test were conducted satisfactorily,
except that signature steps were not signed off as each step was
completed. All steps were signed off after the entire test had been
done.
No deficiencies were noted.
1.
Functional and Stroke Time Test of the Containment Standpipe
Isolation Valve.
A functional and stroke timing test of the Containment Standpipe
Isolation Valve 2-FCV-26-240 was conducted under WR B219647. Motor
leads were meggered prior to the functional test. The tests were
specified because of replacement of the Raychem splices on the motor
leads. The functional test involved hand cranking the valve to the
mid-position, applying power to the valve and ensuring that it did
not move, cycling the valve from the local control station, and then
performing the stroke timing test from the local control station.
During the testing, the inspectors identified the following discrep-
ancies:
1.
The indicator dial on the valve was used to mid-position the
valve; however, the indicator dial on the valve was misaligned
'
by about 180 degrees.
Consequently, the dial indicated about 65
percent when open, through "0" to about 63 percent when shut.
The functional test procedure did not specify what indication to
use when mid-positioning the valve.
2.
Valve indicator lights on the local control panel (Box 2921
Train A) did not work when power was applied to the valve.
Investigation by the Auxiliary Unit Operator found that more
than half of the bulbs on that panel were missing or broken.
Bulbs from other valves were scavenged to complete the test.
-7-
1
.
,
I
i
This is a further example of the poor test practice noted in
)
Paragraph 4'a above.
]
.
3.
The functional test procedure did not specify how to check that
the valve was full open after it had been stroked from the local
control panel.
The test had separate acceptance criteria for
" valve goes full open."
The above cases of inadequate procedural requirements for determining
valve mid-position and for checking the valve full-open position are
considered to be further examples of lack of proper test control, as
designated in Section 4.b above.
During observation of the test the inspectors noted improper thread
engagement on four body to bonnet bolts on the safety-related valve,
and one missing limit switch cover. A number of installation dis-
crepancies were also noted on other modifications associated with
recent valve work.
Some specific deficiencies noted included loose
fittings on Cables 2M24-668, 2M24-678, 2M24-68B, 2M24-728, and
2M24-758 to the Containment LOCA H2 Detector at the 734 level; and,
in the 690 level pipe chase, loose or missing conduit straps between
Valves 2-FCV-26-240, 2-FCV-62-63 and 2-FCV-77-10; loose conduit
.'
connectors on Valves 2-FCV-63-156, 2-FCV-77-10, 2-FCV-63-8, 2-FCV-63-40,
2-FCV-70-85, 2-FCV-72-41, 2-FCV-43-69, 2-FCV-63-44, and 2-FCV-26-240;
a conduit run partially blocking a handwheel on Valve 2-FCV-70-92;
broken flexible conduit on Valve 2-FCV-26-245, missing LB box covers
near Valve 2-FCV-77-20, near a Boron Injection Tank, and on cable to
2-TIS-63-36; missing bolts on a safety-related transmitter (2-PT-63-35)
body to mounting plate, and on the transmitter mounting plate to Wall
i
Bracket 2-L-346; and missing hangers on rigid conduit to Valve
2-FCV-26-240.
These discrepancies were brought to the attention of-
appropriate licensee personnel.
m.
Cable Installation.
(
The inspectors observed the in-process installation of safety-related
cable 2PL4957A. The installation was specified by WP 12392 to
facilitate the replacement of an existing cable from the 480V
Shutdown Board 2Al-A to the 480V control / auxiliary building Vent
Board 2Al-1. The inspectors monitored installation work activities
and reviewed the controlling installation procedure, M&Al 4, and the
associated cable pull tension calculations provided in the work plan.
No deficiencies were identified in this area.
5.
Mechanical Inspection Results
During the two week inspection only a few mechanical tests were performed.
l
The inspection team was able to observe three mechanical tests,
a.
Operability Test of Boric Acid Transfer Pump 2A-A
The inspectors witnessed the performance of Surveillance Instruction
304 on Boric Acid Transfer Pump (BATP) 2A-A after maintenance had
-8-
,
.
'
.
been performed.
This safety-related pump is included in the
Inservice Test (IST) program governed by ASME Section XI which
requires verification of continued operability after maintenance
activities. The test involved duplicating system conditions pre-
viously documented when the pump was known to be operating properly
and verifying that pump. characteristics were the same as before
(within tolerances). The inspectors identified the following
i
concerns:
,
1.
Suction and discharge gauges for the pump, which was already
running when the test began, were not functioning.
In fact,
gauges for two of the other three BATP's (running) were also
On Pump 1A-A, the suction and discharge gauges
indicated a differential pressure that 'did not meet the opera-
bility requirements of the IST program, however, this pump was
j
not being tested at the time.
It was not clear whether Pump.
1A-A was " operable" or " inoperable." To meet the accuracy
requirements of ASME Section XI, temporary gauges were installed
on Pump 2A-A to perform the test.
In summary, although operable
gauges and lines were used to conduct the. test, there was no
indication of pump suction and discharge pressures available to
plant staff during normal operations.
2.
In investigating the gauge concern, the inspectors noted that
each BATP had two sets of suction and discharge pressure gauges;
a labeled set at a panel six feet from the pump and another set
immediately off the suction and discharge piping.
For Pump 1B-8
the nearer gauges had been removed and the tubing capped.
Flow
Diagram 47W809-5 in the control room showed additional gauges
,
i
installed on Unit 2 only, per Temporary Alteration Change Form
(TACF) 84-200262, although TACF 84-20062 was no longer listed in
the Shift Engineer's TACF log.
The gauges were not shown
installed for any BATPs on the as-constructed system flow
diagram. Also noted was that the piping class boundary designa-
tions on the flow diagram were inconsistent in that the boric
acid tanks are indicated as Safety Class "C" in some cases and
as Class "D" in others.
3.
Because the test resulted in a low differential pressure in
the " action" range of the IST' program, the pump was declared
.
The deficiency log attached to the completed test
report showed that the initial valve lineup was inadequate and
that a procedure change would be issued and the test redone.
It
appears that the test, which was run in accordance with an
established SI that has been used on a quarterly basis for
several years, may be invalid because the procedure was incor-
rect. This raises a concern as to whether previous pump
operability tests to the same procedure have been valid.
4.
One of the pump parameters verified during the operability
tests was vibration of pump bearing housings.
The inspector
noted that readings were taken on Pump 2A-A at locations dif-
ferent from those shown on the sketch in the SI. The discre-
pancy between field markings and procedure sketches was pointed
out to licensee personnel.
-g-
--_- _ _,
.
As a result of this discrepancy, the inspectors reviewed sis and
verified target markings on 17 additional safety-related pumps
involved in the IST program. Technical Instruction (TI) 96.1,
" Vibration Monitoring-Pumps and Motors," states that, for
,
repeatability, the pickup location should be marked on each
j
bearing cap with a paint stick. The inspectors found no paint
marks on the two Unit 2 Motor Driven Auxiliary Feedwater (AFW)
Pumps, Unit 2 Turbine Driven AFW Pump 2AS, or on the top of Unit
1 Turbine Driven AFW Pump 1AS.
RHR Pump 188 had no markings at
the south and east orientations as shown in the SI sketch.
1
Conversely, SI-37 for the Component Cooling Pumps did not
'
contain a sketch showing pickup locations although pumps were
marked in the field.
5.
To obtain suction and discharge pressure readings on the tem-
porary gauges for the test, the instrument mechanic had to
j
do extensive cleanout of boric acid crystals from the instrument
j
tubing. This involved the " rodding" of tubing, and disassembly
j
of additional tubing joints and of a bolted flange. None of the
I
additional activities were described on any work requests
associated with the test or gauge installation / removal process,
4
and no inspections or second party verifications of joint makeup
i
were performed,
b.
Review of Previous SI Test Records
The inspectors reviewed surveillance test results for the BAT Pumps
,
(SI-304) f rom February 1985 through September 1986 and for the
i
Component Cooling Pumps (SI-46) from April 1986 through July 1986.
With the exception of one test, the data and evaluations appeared to
I
be in accordance with procedure and Code requirements. The one
discrepancy noted involved the SI-304 test en September 12, 1986 for
BATP 1B-B, which showed a suction pressure tf 2.5 psi while the
minimum allowable was 4.3 psi. The out of specification reading was
not noted by the test personnel, and apparently no evaluation or
corrective action was performed. This is considered to be an
q
isolated occurrence.
j
c.
Functional Testing of Diesel Generator (DG) Room Exhaust Fan Low
Flow Switches
j
The NRC inspectors witnessed the performance of the functional test
on six fans in three of the DG rooms. During the first test on
DG 28-B, a step verifying that red lights on the fan breaker were
energized was signed off when in fact the panel lights were burned
out.
A subsequent search by test personnel to obtain bulbs from
other panels indicated that approximately 50 percent of the panel
.
lights were missing or burned out.
These local lights indicate the
i
energized or de-energized status of plant equipment,
d.
Monthly Functional Testing of Diesel Generator 2A-A
The part of this test, which was observed by the inspection team,
indicated that DG 2A-A had a load dropping problem.
Discussions
with test personnel revealed that load dropping was related to the
-10-
.
rate at which loads.were picked up.
By the end of the NRC.inspec-
tion, TVA had not yet determined what corrective action was neces-
sary.
The examples of inadequate test conditions, inadequate test procedure
requirements, discrepancies in test documentation, and failure to follow
procedures discussed in Section 5.a above, are considered to be further
examples of lack of proper test control, as designated in Section 4.b
above.
6.
Design Baseline Verification Program (DBVP) Post Modification Test Review
The DBVP post-modification test review was initiated by TVA to review
Sequoyah modifications which had been performed since plant licensing.
The TVA review was limited to those modifications that were'within the
boundaries of 37 FSAR Chapter 15 systems required to shutdown the reactor
All applicable engineer-
or to mitigate the conseq)uences of.an. accident.and field change notices (FCNs)
ing change notices (ECNs
plant licensing were reviewed by TVA for Unit 2 and Unit 1/2 common
systems to verify the adequacy of post-modification-testing. The NRC
inspection team reviewed sixteen of the DBVP system evaluation reports
(SYSTERS) to determine whether an adequate-screening of ECNs and WPs had
been accomplished by TVA. The inspectors identified isolated deficien-
cies, which were discussed in detail with licensee personnel during the
inspection and were provided to licensee management during the exit
meeting. The overall conclusion of the NRC inspectors in this area was
that the licensee had applied the review criteria in a reasonable manner
in determining testing requirements for previous modifications.
7.
Separation of Electrical Circuits
a.
In response to questions raised during the previous NRC Safety
Systems Outage Modifications Installation Inspection (SS0MI), the
team inspected the routing of cables in a number of Flamemastic
coated cable bundles which contained both essential and nonessential
circuits.
The Sequoyah FSAR, in Section 8.3.1.4.3, states, in part... " Cables
for nonessential circuits may be run on cable trays.with those for
essential circuits with the following restrictions. When nonessen-
tial cable is routed in a tray with essential (GSPS) cables, that
cable or any cable in the same circuit has not been subsequently
routed onto another tray containing a different division of separa-
tion of essential cables." For the purpose of this report, the FSAR
requirement which prohibits the routing of any nonessential circuit
into raceways containing both divisions of essential circuits will be
referred to as " secondary bridging."
During the previous SS0MI inspection, it had been determined that
free-air cable installations are not controlled by the computer
generated routing scheme established for plant design.
Consequently,
the bundling of essential and nonessential cables in this manner
represented a potential for inadvertent secondary bridging.
-11-
,
..
.
. - - - . . --
..
..
.
_ _ _ . . . . . . - . . .
. . . . - . . .
..
.. -- . . . .
. . ,
- - - - -
During this inspection, the inspectors identified a' sample of. .
nondivisional cables in the Units 1 and 2 spreading rooms which were
bundled with divisional cables. The nondivisional cables were then
identified above the fire barrier in the control room panel risers.
The inspectors then checked the computer generated routing of each .
identified nondivisional cable to ensure that it was not routed with
,
'
a redundant division in other parts of the plant. A total'of 68.
cables in the control room and seven cables from various locations in
the auxiliary building were selected. The cables are identified in
,
Table I.
The licensee had performed a similar effort just prior to
1
this NRC inspection and sampled 165 nonsafety cables bundled with
safety cables in the spreading rooms. Of the 75 cables in the NRC
sample, 31 overlapped TVA's sample.
A review of the-computer routing records for the cables selected
indicated that none were routed across' safety divisions. An audit.of
the routing list for the 165 cables identified by TVA also indicated
-
no routings across divisional boundaries.
However, two instances of
misrouted cables were identified with the cables. selected by the NRC
in the Auxiliary. Building. These two examples, which are discussed
below, call into question the accuracy'of.the cable routing lists:
1.
Safety-related Conduit IPP673B had three cables that' exited the
i
conduit and were then routed into divisional Tray LM-B and one
essential cable that was routed into nonessential Tray EAZ.- The
,
cable ran about 4 feet in Tray EAZ and then entered divisional
i
Conduit IV2943B. A review of the routing record for the cables
and conduit indicated that all four cables, IPP664B, IV2944B,
IV2963B, and IV29438, should have been routed in divisional Tray
LM-B.
Cable IV2943B should have been routed in divisional Tray
LM-B rather than non-essential Tray EAZ.
Sequoyah DNE personnel
verified the installation with NRC inspectors and indicated that
a Condition Adverse to Quality Report (CAQR) would be prepared
to document and correct the misrouting by moving the cable into
the appropriate divisional tray.
,
!
2.
Nonsafety-related Conduit FE-4832 had three cables exiting.the
conduit. After exiting the conduit, two of the cables were
i
routed into divisional Tray NJ-A, and one cable was routed into
j
non-essential Tray JAM. A review of the conduit and cable
j
routing records indicated that all three cables, FE4968, FE4969,
!
and V1004 should have been routed in nondivisional Tray JAM.
i
j
Since the two misrouted cables entered the Flamemastic bundle in
divisional Tray NJ-A, finding'the true routing of the cables was'
not possible. Sequoyah DNE-staff verified the misrouting with
the NRC inspectors.- The NRC inspectors were informed that a.
CAQR would be written to document and evaluate the misrouting.
Since the route for the two cables in Tray NJ-A was unknown,
the potential exists for secondary bridging by these cables.
1
In summary, the question of secondary bridging at Sequoyah will
require further evaluation by the licensee and NRC personnel. The
routing deficiencies identified during this inspection indicate a
3
potential for inadvertent problems in this area.
l
(
4
-12-
i
1
.
b.
During the entrance meeting with the licensee the NRC inspectors
raised questions concerning the use of Flamemastic 77 as an approved
fire barrier for protection of safety-related circuits from inter-
nally generated fires.
In response to these questions the licensee
stated that, with one exception, Flamemastic 77 had not been used for
this purpose. The exception is the annulus area where cables leaving
various electrical penetrations are dispersed into cable trays or
conduits for routing through the shield wall to other areas of the
plant.
This exception is found in TVA Design Criteria SQN-DC-V-12.2,
Section 4.2.9, which states in part... "Where it is impracticable to
provide enclosed trays, then the cables shall be coated with Flame-
mastic 77 or its equivalent."
During evaluation of Sequoyah FSAR requirements pertaining to elec-
trical separation the inspectors noted NRC Question 8.20 and a TVA
response which relate to this issue.
The NRC question states:
"Several places in Section 8.3.1.4.2 the use of fire resistant
coating materials such as 'Flamemastic 71A' is proposed as an
1
alternative to the use of solid barriers.
This is not an
1
'
acceptable alternative and we will require the use of adequate
solid barriers that will not permit the transfer of sufficient
heat energy between trays to raise the temperature of the cables
contained in the redundant tray above their safe temperature
rating."
TVA's response to the NRR question was as follows:
1
" Fire resistant coating material such as 'Flamemastic 71A' in
!
lieu of solid barriers has not been used.
Subparagraph
j
8.3.1.4.2 has been revised to delete this reference."
1
The NRC inspectors discussed this issue with TVA personnel in an
attempt to determine to what extent Section 4.2.9 of the design
criteria had been used at Sequoyah.
The discussions indicated that
design drawings for the annulus area specify the installation of tray
1
covers, and that no Flamemastic was used in lieu of solid barriers
j
in the annulus area. Additionally, the licensee provided two Quality
Information Requests (QIRs), SQN-EEB-86145 and SQN-EEB-86172, which
j
detail the walkdown of plant cable trays, including the annulus area,
I
to assure that tray covers have been installed. Based upon this
information the NRC inspectors concluded that, while portions of the
design criteria make provision for use of cable coating in lieu of
solid barriers, it does not appear that this provision was exercised
in field installations.
Based on the conflict which exists between the Sequoyah FSAR and the
TVA design criteria the inspectors concluded that the provision for
use of Flamemastic cable coating, as specified in Section 4.2.9 of
Sequoyah Design Criteria SQN-DC-V-12.2, does not provide an acceptable
alternative for protection of safety-related circuits and appears to
permit installations which violate FSAR requirements.
c.
The Sequoyah FSAR, in Section 8.3, establishes separation criteria
for independence of safety-related equipment and circuits. These
-13-
_-
.
.
criteria are delineated in SQN-DC-V-12.2, titled " Separation of
Electric Equipment and Wiring." Specific criteria are provided for
physical separation of cable trays in order to protect the enclosed
safety-related circuits. However, the FSAR provides no criteria for
physical separation of conduits which contain safety-related circuits
and, while this subject is referenced in Section 4.2.2.2 of the
Design Criteria, no specific details are provided.
During the
inspection of installed components, the NRC inspectors observed
several installations in which the physical separation between
redundant divisional cable trays and conduits was less than the FSAR
)
specified distances for cable tray.
As an example, Division A
l
Conduit 2M-3240-A was installed less than 1 inch from Division B
uncovered Cable Tray MS-B.
Based upon the examination of field
installations, the inspectors expressed concern that the lack of
i
criteria for separation of safety-related conduits does not meet the
intent of Section 8.3 of the FSAR, in that protection and independ-
ence of redundant circuits may be compromised.
The concerns identified above regarding the potential for secondary
bridging, the potential use of Flamemastic as an alternative to solid
barriers, and the lack of criteria for separation of safety-related
conduits, will require further review by NRC (Unresolved Item
50-327/87-18-01;50-328/87-18-01).
-14-
'
a
l
TABLE I
SECONDARY BRIDGING CABLE SAMPLE
Control Room / Spreading Rooms
Panel 1-M-23B:
1M1597
1M1596
IV3521
IM1585
IV1595
1M1598
1M1599
ISR906
i
1A3690
1A3716
IM1587
ISR914
1M1586
1M1589
iM1588
l
Panel 1-M-23A:
1V3521
IV3522
IV3520
1A3688
IV3523
1M1590
PL3858
1M1584
1M1581
1M1591
IM1580
3
Panel 0-M-27B Riser 1:
2FE89
2FE74
SR210
2TCT1
'
PL3931
2FE97
2PL1994
2FE65
2FE73
2TCT2
2PL1894
2TCT3
2FE96
2FEP'
2PL1794
2FE66
PL3921
2FE90
2A4380
2FE81
,
Panel 0-M-278 Riser 2:
2M202
2M200
Panel 0-M-27B Riser 3:
2PL3646
1PM3191
1PL3646
1PM3227
2PL3661
2PM2411
1PM2256
2PM2271
2PM3122
1PM3217
2PM3257
2PM3182
2V5787
Panel 2-M-4 Riser B:
2A2581
2A2580
2SR694
2C222
2SR693
Panel 2-M-5 Riser B:
2SR695
Auxiliary Building
IV2944B
IV2963B
IV2943B
IPP664B
FE4968
FE4969
V1004
TI-1