ML18064A875
| ML18064A875 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 08/22/1995 |
| From: | Gardner R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML18064A874 | List: |
| References | |
| 50-255-95-10, NUDOCS 9509050021 | |
| Download: ML18064A875 (5) | |
See also: IR 05000255/1995010
Text
U. S. NUCLEAR REGULATORY COMMISSION
REGION I II
REPORT-NO. 50-255/95010(DRS)
_ FACILITY
Palisades Nuclear Generating Station
LICENSEE
Palisades Nuclear Generating Plant
27780 Blue Star Memorial Highway
Covert, MI 49043-9530 *
_
DATES
August 1 through 4, 1995
INSPECTORS
R. Winter, DRS Inspector
D. Butler,*DRS Inspector
APPROVED BY
.£M~Jtck~
$/~~115
R. N .. Gardner, Chief
Plant Systems Section
AREAS INSPECTED
Date
A special, unannou~ced safety inspection to determine the ~oot causes that led
to.the installation of containment high pressure printed circuit board jumpers
in the reactor protection system (RPS) matrix trip logics. The areas
inspected primarily involved post-modification .testing activities associated
with the 1992 RPS Upgrade Project and current RPS surveillance testing
methodology.
RESULTS
Two apparent violations were identified involving:
1) failure of the post~
modification test for a 1992 facility change to identify that the RPS
containment high pressure matrix trip logics were bypassed; and 2) failure to
maintain the six RPS matrix logic channels for containment high pressure
operable in accordance with Technical Specification 3.17 ..
-.
9509050021 950823
ADOCK 05000255
Q
i . l
l l
t
{*
. '
'
~
- 1
1
-
~\\
I
l
t
i l
/.
INSPECTION DETAILS
1.0
Summary of Events
On July 28, 1995, while performing RPS circuit checkouts after connector
replacements, the licensee discovered that the RPS matrix trip logics * *
for containment high pressure were inoperable. Subsequent investigation
revealed that three jumpers had been created on each of the four matrix.
logic printed circuit boards. This occurred due to Pa}isad~s providing
RPS matrix logic wiring diagrams to Combustion Engineering that did not
reflect the as-b~ilt RPS configuration. Combustion Engine~ring *
developed wiring lists to manufacture new matrix logic modules from
these drawings during the 1992 RPS Upgrade Project.
The jumpers
bypassed the two of four (2/4) matrix trip logic signals for containment
high pressure.
The RPS rack bins were designed to accept 12 trip modules.
Palisades.
original RPS design had 10 RPS functional trips and 2 unused m6dules._
Containment high pressure was added during original construction as ari
eleventh RPS trip function. *The six 2/4 logic trip matrixes were.
designed as* a ladder network that receives two inputs from each RPS tri~
Opening of coincident parallel
contacts will initiate a reactor trip in any ladder network.
A trip
module logic input was hard wired bypassed (original construction) or
printed circuit jumpered (current design) when unused.
The original
construction matrix module was. wired correctly with only the unused
twelfth logic input having a bypass jumper.
However, the original
construction wiring diagram was never updated and reflected 2 unused
modules that were bypassed by jumpers.
Channel eleven {containment high pressure) was last tested and
demonstrated operable on December 3, 199lj prior to* the outage that
upgraded the RPS system.
During and following the 1992 outage, the RPS
containment high pressure logic inputs were not tested from the matrix
trip logics through the four RPS initiation logics (de-energize~ the
clutch power supplies) either by a post-modification test (PMT) or by
subsequent surveillance tests.
2.0
Root Causes
The inspectors identified the following barriers that may have
contributed to this event:
o
The original construction wiring diagrams did not reflect the as-
built configuration.
The wiring d1agrams showed a 10 trip input
RPS {the original design) and had not been updated {by removing
jumper bypasses) to reflect the addition of the containment high
pressure trip function.
These wiring diagrams were sent to
Combustion Engineering to manufacture the new printed circuit
matrix logic modules.
0
The containment high pressure portion of the manufacturer's
factory acceptance test {FAT) did not identify the matrix logic
output anomaly during the matrix channel trip select test.
2
t
. t.
.~
~
- .i:
I
f
'i
- '.
. .
0
0
0
Without knowing the test setup, the inspectors were unable to
determine if the FAT could have identified the jumpers.
This
matter is under review by the licensee.
The licensee's PMT did not test all aspects of the modification.
The PMT used the factory acceptance test to verify that the analog*
- portions of the modification were in c.alibration. A surveillance
test was specified to complete the logic testing. Surveillance
test No. M0-3; "Reactor Protection Matrix Logic Tests,"
Revision 12, was the current procedure used to implement TS
Table 4.17.1, Functional Unit No. 13, "RPS Matrix Logic,"* and
No. 14, "RPS Initiation Logic."
The surveillance test did verify
proper' test overlap from the matrix trip logics through the
initiation logics. However, the procedure permitted the operator*
to select the functional RPS input, one of eleven Functional
Units, to test through the RPS initiation logics.
The PMT did not
specify that all *eleven RPS functional inputs should be. tested
through the RPS initiation logics. Per the survei.llance test, .the
operator randomly selected channel four.
If the operator had selected channel eleven (containment high
pressure}, the M0-3 procedure would have identified that the
channel eleven matrix tri.P logics were jumpered.
The licensee did not perform an adequate PMT review following the
completion of testing activities. The review. would.have
identified that only one RPS functional channel had been tested
through the initiation logics. For the scope of the 1992 RPS
Upgrade Project, the PMT should have demonstrated test ove.rl ap
from unmodified portions of the circuits through modified portions
of the circuits.
- The licensee recognized in early 1995 (prior to identifying the
jumpers during the outage) that they were randomly selecting the
RPS functional channel to test each month through the initiation
logics. The M0-3 procedure was revised to select the RPS channel
to be tested by using the test month.
Using this selection
method, channel one would be.selected in January.
However, the
licensee did not review past tests to determine if all channels.
had been tested through the initiation logics during an operating
cycle. This would have identified that channel eleven had not*
been fuliy tested though the initiation logics since 1991.
The containment high pressure printed circuit jumpers would have
been identified in November 1995 during the performance of
surveillance procedure No. M0-3.
Channel eleven would be the
~elected channel for matrix logic testin~ through the initiation,
logics that month.
3
i.
I
. i !
.,
J
r
,.
i
t'o
t
~ ..
..
3.0
Licensee Corrective Actions
3.1
3.2
0
0
The six RPS matrix channels have been modified by grinding away*
the circuit board trace material.
The gap provided in the circuit
board trace was approximately ~ inch.
-
.A PMT has been written to perform a comprehensive test {wire to
wire) following current RPS modifications.
o
A new refueling outage RPS logic tesf procedure has been
developed.
Procedure No. MI-3A, "Reactor Protection Matri~ Logic
Annual Test," ~erifies that all eleven RPS functional inputs will
remove power from the clutch coils. This procedure will b~
performed before declaring the RPS operable during the 1995
.refueling outage.
Testing
0
The license~ was reviewing 19 additional surveillance procedures
and 8 PMT procedures performed during the 1995 refueling outage.
The objective of this review was to verify that procedures used to
test safety related circuits completely tested the design
functions of the circuits.
4.0
Violations
The following apparent violations were id~ntified: *
o
10 CFR 50, Appendix B, Criterion Ill, "Design Control-," requires,
in part, that measures shall provide for verifying or checking the
adequacy of desig_n by performance of a .suitable tesUng program .
0
. Contrary to the above, during the 1992 refueling *outage, the post-
modification test for Facility Change {FC) No. FC-888 did not
identify that the RPS containment high pressure matrix trip logics
were each bypassed {jumpered) by a printed circuit board trace.
As a result, the RPS containment high pressure trip logic could
not perform its safety function until identified and corrected-
during the 1995 refueling outage.
-
Technical Specification {TS) 3.17, "Instrumentation Systems,"
requires that four Reactor Protection System {RPS) trip unit
_channels and the associated instrumentation for the functions
listed.in Table 3.17.l, and 6 matrix logic channels and 4
initiation logic channels shall be OPERABLE except as allowed by
the permissible operational bypasses.
Contrary to the above, from the 1992 refueling outage until the
1995 refueling outage, the six RPS matrix logic channels for RPS
c~ntairiment high pressure were not operable in accordance with
.
4
l
t
I I
1
l
I
'
\\
f
f 1
l
t
- ~ . -
<;
,..,
5.0
Safety Significance
The inoperability of the RPS containment high pressure trip since 1992
is considered significant in that a reactor trip function designed to
prevent or mitigate a serious safety event, such as a main steam line
break inside containment, could not perform its intended safety
function.
The licensee. was reviewing the safety analysis to determine
if a delayed reactor trip would exceed containment design pressure.
6.0
Management Debriefing
The inspectors met with licensee representatives (denoted in
Section 7.0) after the inspection on August 4, 1995, to discuss the
scope and findings of the inspection. During the exit meeting, the
inspectors discussed the likely*infrirmational content of the. inspection.
report and documents or processes reviewed by the inspectors during the
inspection:
Licensee representatives did not identify any such
documents or processes as proprietary.
7.0
Persons Contacted
r
Consumers Power Company
- T. J. Palmisano, Plant General Manager
- K. P. Powers, Plant Engineering and Modifications Manager_*
- G. B. Szczotka, RPS Project Team Leader
- R. A. Vincent, Licensing Administrator
U. S. Nuclear Regulatory Commission
- D. Passehl, Resident Inspector
- Denotes those pres_ent during the exit meeting on August 4, 1995.
5
t
\\ t
.r
!
f i Il
t
l
~ ; l
(
Il
. l
t
1I
i l
li