ML20129E340
| ML20129E340 | |
| Person / Time | |
|---|---|
| Site: | Zion File:ZionSolutions icon.png |
| Issue date: | 10/22/1996 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20129E314 | List: |
| References | |
| 50-295-96-11, 50-304-96-11, NUDOCS 9610280030 | |
| Download: ML20129E340 (44) | |
See also: IR 05000295/1996011
Text
i .
U.S. NUCLEAR REGULATORY COMMISSION
l
REGION lli !
Docket Nos: 50-295, 50-304
Report No: 50-295/96011(DRS); 50-304/96011(DRS)
Licensee: Commonwealth Edison Company
Facility: Zion Nuclear Generating Station, Units 1 and 2
Location: 105 Shiloh Boulevard
Zion, IL 60099
Dates: July 22 - August 22,1996
Inspectors: Zelig Falevits, Team Leader, Region ill
Joel Guzman, Reactor Inspector, Region lll
Bob Winter, Reactor Inspector, Region 111
Daniel Rich, Reactor Engineer, Region ill
Robert Stakenborghs, Contractor, Parameter, Inc.
John Heller, Contractor, Parameter,Inc.
Approved by: Mark A. Ring, Chief, Lead Engineers Branch
Division of Reactor Safety
insoection Summarv
A team inspection of engineering and technical support was performed using NRC
inspection procedures (IPs), including IPs 37001, 37550, 37551, 37700, and 37828.
Results: Five apparent violations were identified. The first apparent violation involved an
inadequate 10 CFR 50.59 safety evaluation process, which resulted in a lack of safety
evaluations or inadequate safety evaluations. The second apparent violation involved a
failure to ensure that field-installeci (4esign change modifications had been properly
evaluated, tested and signed off prior to p!9cing them in service. The third apparent
violation concerned failure to indicate the operai21ity status of systems structures and
components that had been modified and placed in use, but not yet fully tested. The fourth
apparent violation concerned failure to take timely corrective actions for an extended
period of time to address equipment nonconforming conditions. The fifth apparent
violation concerned inadequate procedures and failure to follow procedures, which
contributed to Technical Specification Interpretation deficiencies and the failure to generate
Problem Identification Forms on Updated Final Safety Analysis Report conformance review
discrepancies.
9610280030 961022
PDR ADOCK 05000295
G PDR
1
8 *
EXECUTIVE SUMMARY
Zion Nuclear Generating Station, Units 1 and 2
Report No. 50-295/96011(DRS); 50-304/96011(DRS)
This inspection included aspects of Zion's engineering and technical support and was
conducted from July 22 through August 22,1996. The team determined that the Zion
engineering organization was staffed with mostly experienced engineers. However,
engineering performance was considered weak in several technical areas and processes.
A summary of findings identified by the team follows:
o The licensee's process for controlling modifications to ensure adequate
post-modification testing package closure appeared to have broken down. Nine {
i
safety related modifications and 19 non-safety related modifications had been
physically installed and placed in service, even though the modification packages
were not signed as completed and were not authorized for use by operations.
These modification installations ranged from the mid-1980s through 1996. Most of
these modifications also contained inadequate documentation to conclude that
post-modification testing had been accomplished,
o Selected operability assessments reviewed by the team (3 of the 12 reviewed in
'
detail) were found to be weak or inadequate. For example, the operability
assessment for charging pump degradation noted during emergency core cooling
system full flow testing was inadequate; the assessment failed to consider the
pumps' ability to deliver the required flows at design conditions.
e The team identified numerous deficiencies in the sample of 10 CFR 50.59 safety
evaluations reviewed during the inspection. The concerns identified indicated a
continuing lack of quality, completeness and thoroughness relative to 50.59 safety
evaluations. Of note was the licensee's superficial approach towards the safety
evaluation of scaffolding erected and left inside the containment during plant
operation. In view of the repetitive nature of these deficiencies and of the
licensee's failure to implement effective corrective action for previous violation and
self assessment findings related to 50.59, the team's findings demonstrate
significant deficiencies in engineering activities and management oversight of the
50.59 safety evaluation process,
e The licensee failed to maintain adequate management controls over the Technical
Specification Interpretation process, thereby allowing inappropriate Technical
Specification interpretations.
e A recent Site Quality Verification audit that examined some engineering activities
wcs considered to be good, in that it identified some deficiencies in the 50.59
process; however, Zion's self assessment program had not identified some of the
significant problems identified by the team, such as the specific 50.59 problems
noted in this report, modification testina and closecut problems, and corrective
action issues.
2
. ,
j 4
e
i
i
i
i
!~ e The mechanical modification related calculations reviewed by the team were ,
l generally well organized, complete and technically adequate. The electrical !
j calculations reviewed were considered to be better than the industry average and !
j were found to be appropriate to support the plant's modifications. !
I
j. e The recciitly initiated instrumentation upgrade program established to resolve )
!
'
potential discrepancies in the safety related and/or seismic classification of l
instrumentation was considered a good initiative to improve the design bases at
Zion; however, it was not yet fully implemented.
'
. e Engineering activities relative to lube oil analysis, trending of equipment problems
i
and resolution of recurring deficiencies were considered weak, in that recurring
l tertiner failures, out of tolerance instrumentation, and unsatisfactory oil samples
l were not trended or investigated promptly.
!
. e- The licensee's Updated Final Safety Analysis Report (UFSAR) conformance review
l was viewed as a thorough, in-depth technical effort that identified many meaningful
! discrepancies between the UFSAR and the as-built and as-operated plant. This was
! considered a strong, positive licensee initiative. However, the subsequent delay in
l initiating prompt resolution to these discrepancies was considered inadequate.
l
l
e inadequate procedures or failure to follow procedures contributed to Technical ;
- . Specification Interpretation deficiencies and to the failure to generate Problem j
identification Forms on deficiencies identified during the UFSAR conformance
'
review.
e Longstanding problems existed in the various radiation monitoring systems. Many
radiation monitors had failed frequently and were in poor materiel condition.
Operators were required to take extensive compensatory actions to deel with this l
condition. The licensee had developed a comprehensive plan to correct the
problems, but little progress had been made to date. However, the licensee's
program to assess dose to the public from radioactive effluents was still effectively
implemented.
e Based on interviews with engineers and managers, the team concluded that
management expectations were not always clearly defined, nor understood by the
engineering staff. For example, management expected that system engineers
evaluate lube oil analysis sample results, even though the engineers had not been
trained nor qualified to perform this evaluation. Also, management expectation
relative to breaker testing following maintenance activities was not clearly
understood by the engineers. Consequently, breakers were not always tested after
they were racked into the cubicle.
3
, .
I
TABLE OF CONTENTS
P_a_ste
E1 CON DU CT OF EN GIN EERING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
E1.1 Modifications, Design Changes and Post-Modification Testing ......... 6
E1.2 Parts Replacement Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
E1.3 Modification Related Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
E1.4 Temporary Alterations ..................................... 15
E1.5 10 CFR 50.59 Safety Evaluations ............................. 16
E1.6 Review of Operability Assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
E1.7 General Conclusion for Conduct of Engineering . . . . . . . . . . . . . . . . . . . . 22
E3 ENGINEERING PROCEDURES AND DOCUMENTATION . . . . . . . . . . . . . . . 22
E3.1 UFSAR Conformance Review ................................ 22 l
E3.2 Vendor Equipment Technical Information Program . . . . . . . . . . . . . . . . . . 24
E3.3 Miscellaneous Engineering Tasks (Procedures and Documents) ......... 25
E4 ENGINEERING STAFF KNOWLEDGE AND PERFORMANCE ............ 28 l
E4.1 Root Cause Determination and Corrective Action Process . . . . . . . . . . . . . 28
l
E4.2 Trending and Materiel Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 i
E4.3 Lube Oil Analysis Process and Packing Leaks ..................... 31
E4.4 Auxiliary Power System Automatic Bus Transfer . . . . . . . . . . . . . . . . . . . 32
E4.5 Containment Spray System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
E4.6 G eneral Conclusione . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 l
l
'
E5 ENGINEERING STAFF TRAINING AND QUALIFICATION .............. 34
E5.1 Engineering Staff Training and Specialized Training ................. 34 )
E6 ENGINEERING ORGANIZATION AND ADMINISTRATION . . . . . . . . . . . . . . 34
4
a .
1
E6.1 Zion System, Design and Maintenance Engineering ................. 34
E7- QUALITY ASSURANCE IN ENGINEERING ACTIVITIES ............... 35
E7.1 Licensee Self Assessment Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
E8 MISCELLANEOUS ENGINEERING TASKS ........................ 36
E8.1 Technical Specification interpretations . . . . . . . . . . . . . . . . . . . . . . . . . . 36
E8.2 Radiological Protection Controls for Liquid, Gaseous Radwaste
and Area Radiation Monitoring Systems . . . . . . . . . . . . . . . . . . . . . . . . . 40
E8.3 Instrument Classification Upgrade Program . . . . . . . . . . . . . . . . . . . . . . . 41
X1 Exit Meeting S um mary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5
_
1
-
s
l
REPORT DETAILS
The. NRC performed a team inspection of engineering and technical caoport (E&TS).
During this inspection, the team reviewed the adequacy of the design, installation and
testing of selected permanent and temporary modifications,10 CFR 50.59 Safety
Evaluations (50.59 SEs), operability assessments, root cause analysis, trending and
corrective action programs and their implementation. In addition, the team performed
system walkdowns with system engineers, conducted interviews and reviewed
engineering related audits, surveillance and self assessments. The team also examined the
interface between engineering and other plant support groups, including corporate. Five i
apparent violations were identified. !
111. ENGINEERING .
l
E1 CONDUCT OF ENGINEERING j
l
E1.1 Modifications, Desian Chanaes and Post-Modification Testina
a. Insoection Scooe l
)
The team reviewed design documents, calculations,50.59 SEs, operability
assessments and procedures related to the performance of design modifications to j
the facility. The review included approximately 35 mechanical, electrical and
instrumentation modification packages in various stages of completion. The
modification packages were reviewed for technical adequacy and completeness.
The team focused on the Auxiliary Power (AP) system, the Area Radiation (AR) and
Process Radiation (PR) systems, the Containment Spray (CS) system and the Safety
injection (SI) system during this inspection,
b. General Observations and Findinas
1. Inadeauate Post-Modification Testina and Incomotete Closeout of
Modification Packaaes
The team determined that 9 safety related (SR) and 19 non-safety related (NSR)
design change modification packages had been physically implemented in the field
and placed in use since installation, even though the packages were not signed off
as completed and were not authorized by operations. These modifications had
inadequate documentation to confirm that all modification test requirements had
been successfully accomplished. The operating / shift engineer, engineering
supervisor, quality assurance supervisor and training supervisor did not approve / sign
the modification packages to indicate that the modifications had been completed
and could be declared operable.
6
I *
The SR modifications included:
M22-0-84-OO1: OP1 A-PR29 (process radiation monitor) signal to control room
isolation and safety injection.
M22-0-87-042-B: Provide redundant bubble tight damper isolation to the control
room heating, ventilation and air conditioning.
P22-0-90-551: S.S.O.M.I. (Safety System Outage Modification Inspection)
item area 15 cable / cable tray.
M22-1-84-052: Receptacle on battery charger and test switch.
M22-1-90-0168: NIS source and intermediate range replacement with accident
monitoring qualified instrumentation.
E22-1-93-248: Change steam supply to auxiliary feed pump steam traps to l
choke traps (Unit 1).
l
E22-2-93-248: Change steam supply to auxiliary feed pump steam traps to
choke traps (Unit 2).
M22-2-84-080: Replace vent valve on CC surge tank.
l
E22-2-93-272-B: Remove the MOV motor brakes from 2MOV-Sl8804A&B and 1
2MOV-Sl8809A&B. l
The status of each modification package could be placed into one of the following
three categories: (1) testing incomplete or unacceptable; (2) configuration
management items required prior to operability not completed; and (3) required
signatures not obtained for package closecut. Some modification packages were
installed in the field as early as 1986 and as late as 1996. Many of the packages
were misplaced and could not be found when requested by the NRC. It was not
clear during the inspection if all specified/ required testing was completed. The
licensee issued Problem Identification Forms (PlFs) 96-2024 and 96-2046 to
document this problem and determine root cause.
Until questioned by the team on August 6,1996, the licensee had not established a
list of the misplaced modification packages that had been field-installed, not
declared operable, but placed in use Nor was the licensee cognizant of the status
of testing on the misplaced modifications. In addition, the licensee failed to indicate
the operability status of the Systems, Structures, and Components (SSCs) that had
been modified, placed in use, but not yet fully tested.
Examples of incomplete or inadequate post-modification testing for a system which
has been placed in use included:
7
- . . - . - . . - . - . . -- .- .-.
,
l
l
(a) SR Modification M22-0-84-001
This modification was installed in 1989 to provide another control room isolation
upon receipt of an SI signal. The SI signal was added to shift the control room
Heating, Ventilation and Air Conditioning (HVAC) to recirculation (a commitment to
the NRC, since the existing radiation monitoring circuit was NSR). The modification
binder was described to the team as missing, but eventually found on the desk of
an engineer who had not worked at Zion for years. The PR system engineer was
unaware of this open modification in his system.
The team determined that not all post-modification required testing was completed.
For example, the modification approval letter required testing be performed to
assure independence of the modified system from all other systems upon electrical
failure of the installed modification. This was not accomplished. Also following
inspector questioning, the licensee had to rely on other periodic operatioral tests to
verify that the original testing requirements were met. It was not clear if all post-
modification testing requirements were addressed. When questioned by the team,
the licensee performed an operability assessment during the inspection to ensure '
that Technical Specification (TS) requirements were not being violated as a result of
the incomplete testing.
To compound this problem, a second modification, MED0-87-042-B, was
subsequently installed in 1990. This modification charged port (ons of circuit logic
that was installed by modification M22-0-84-001. The modification provided
bubble tight damper isolation to the control room HVAC. After modification
M22-87-042-B was implemented in the field, the licensee could not locate portions
of the original test data and had to rely on subsequent operational tests. During the
processing of modification M22-0-87-042-B, the licensee again failed to evaluate
and complete the test requirements specified in original modification M22-0-84-
001. In addition, several Field Change Requests (FCRs) were issuec against this
modification, but were not in the modification package, and it was not clear if they
were approved and field implemented. It was also not clear if all applicable
procedures had been revised. These modifications were not declared operable at
the end of the inspection.
(b) SR Modification M22-1(2)-84-052
This modification installed and wired test switch blocks and test Jacks on direct
current (DC) distribution center busses 011,111,112,211 and 212. The test
switches were installed to allow for testing of the DC bus relays without
disconnecting any electricalleads. The test jacks were installed to allow voltage
and current reading from outside the panel.
In 1986, test switches were installed on DC busses 111 and 112. Construction
test ECTP #19 was performed in 1987. Modification testing on the test switches
was not performed until August 1996, in August 1993, test jacks were installed
on busses 111 and 112 and ECTP #19 was performed. No modification test was
performed.
8
.__.___ __- _ _. ._ _ _ _ _. _ . _ . - - _ _ _ _ _ _ _ _ _ _ . . . _ _ _ _ _ _ _ .
j. ' *
J
,
J
-
In March 1994, test switches and jacks were installed on bus 011. In November-
]-
- 1995, ECTP #19 was performed. The test switches and Jacks on bus 011 were
1 placed in use but not modification tested nor declared operable until August 1996
) following questioning by the NRC.
,
4
,
in March 1992, the Nuclear Engineering Department (NED) reviewed this
1 modification and determined that the modification was partially installed, but was
j never completed. There were six modification addendum letters to this modification
'
since the initial approval letter in 1984. The following post-modification testing :
l requirements were identified by NED, using ENC QE06.4, Exhibit A:
!
} e Electrical construction test procedure #19.
I
e Verify the DC bus relays are functioning properly by test.
j e Test all three meter jacks with a multimeter and observe proper readings
l- with no circuit loading.
!
j e Verify that no DC bus grounds were created by the new wiring.
!
i
in October 1995, post-modification test procedure TSSP 128-93 was written to <
verify that the test jacks were installed properly during modification M22-1-84-052.
- In July 1996, modification testing TSSP-128-93 was performed. During the review
, of the test results, the team noted that not all post-modification test requirements !
{ were accomplished by this test or documented as completed. The team identified. l
l the~following concerns: :
- * The test results for bus 111, as documented in TSSP 128-93, failed to meet
-
the acceptance criteria set in Section 7 of the test, which required that the
l panel meter reading and the DVM should be within 4.5% of each other to
include allinstrument tolerances. In addition, step 8.0.11 of the test I
{
l
procedure for bus 011 and 112 documented a reading of 0 for the battery
- ammeter reading and a O for the DVM reading. This step required that the
- readings on both instruments be compared. The reading of 0 was obtained
. because the scale on the instruments used was not appropriate to read the
'
low values expected.
I
e Test switches were purchased NSR and field installed, but no documentation
was available in the package to determine that NSR components in SR ,
4 circuits were acceptable.
4 * Test requirements specified in the modification approval letter were not i
- documented as accomplished in the modification package. The Operational.
- Analysis Department protective relay sheets, OADMP-184, were apparently
j. used to indicate successful post-modification test results.
- :
- o Modification ground testing requirement was not documented in the package
! as performed.
J.
,
,
d
$
, .-4 - , , - - , - ,, , - , ,
. - . . .-- -. _. -- - - - . - . . _ _ . -
'
l
1- The team noted during the inspection and interviews that the current DC system
, engineer exhibited a good questioning attitude for present-day activities. The
system engineer questioned the data results for the August 13,1996, monthly l
surveillance test on battery 112. As a result of his inquiry, the licensee later
determined that the wrong battery was tested by the electrical maintenance
technicians (112 instead of 111) and that the monthly surveillance on battery 111
was missed.
1
(c) SR Modification E22-1(2)-93-248
This modification was installed in February 1994. The modification changed the
,
auxiliary feedwater (AFW) pump steam supply steam traps. The system was )
j operating, but had not been returned to operability in accordance with the j
modification procedures. Additionally, the team determined that no 50.59 SE had j
been performed to support the operation of the system in its installed configuration, j
The installed configuration differed from the as-designed configuration in that the
,
steam traps, which would normally have steam flow, were placed in the bypass l
position continuously. This resulted in larger steam flows to the condenser than
i
would have been present through the modified traps. The team questioned if this
- could invalidate any of the assumptions made for the operation of the AFW pumps.
j The licensee reviewed this circumstance and determined that the loss of steam j
through the open bypass would not impair the function of the pumps. The !
,
following concerns were noted:
.
- This modification was in service even though testing was incomplete.
Modification testing was not satisfactorily completed. Periodic operational
tests were performed.
.
e The modification was operating in a mode which was not adequately I
-
analyzed (continuous bypass of the orifices). No design change or 50.59 SE
was performed to address the new operating mode prior to the inspection.
This is considered en example of an apparent violation of 10 CFR 50.59
(295/96011-01 A(DRS); 304/96011-01 A(DRS)).
i
(d) SR Modification M22-1-90-0168
i
! This modification was installed in January 1994. The modification replaced existing
NIS source and intermediate range instrumentation with accident monitoring
environmentally qualified instrumentation to satisfy R.G.1.97.
- The modification package resided with the operating engineer since September
3
1995 to complete additional testing during power operation.
The licensee could not locate test documentation for four NIS drawers and loop
calibrations for four NIS channels. The calibrations were performed under WR
Z28666 which could not be located by the end of the inspection. Subsequently,
the licensee located the test documentation. The final closure of the modification
was dependent on additional testing to be performed during power ascension
10
t
- - . - - . .- -
e =
activities following the outage in 1994. This additional testing was finally
, completed in 1995, but the package was not signed by operations until September
i 1996. Other Operations Department priorities combined with tuinover of the
modification upon retirement of the design engineer contributed to the delay in
declaring the modification operable in September 1996.
4
(e) NSR Modification M22-0-88-20, " Install Time Delav Relav and Rewire Circuit" i
i
Water treatment facility improvements: it was not clear from the modification
, documents what was installed in the plant. Field as-built walkdown was needed to
i determine field conf;guration.
- (f) NSR Modification M22-0-84-014, " Pressure Gauaes/ Surae TanksNatues to Alum
Feed"
This modification was installed in 1989. The licensee stated that equipment was
'
apparently out of service. An FCR would be issued to abandon installed
<
components.
2. Corrective Action and Root Causes
When first identified by the team, the licensee stated that only four such
modifications existed. Following NRC interviews with system and design engineers,
additional packages were found by the system and design engineers. At the end of
j the inspection, the licensee identified 24 additional packages in this category.
.
_ The licensee has reviewed and evaluated the post-modification testing
j accomplished to determine the status of testing of the modifications. An operability
assessment was performed on two modifications. The modified systems were
found to be operable in addition, in some instances, the licensee relied on periodic
operational testing as a substitute for the post-modification testing specified in the
4
modification letter.
The majority of the SR modifications were declared operable by the end of this
.
inspection. The team reviewed portions of the licensee's efforts to as'sure that
modification testing was adequate to declare the modifications operable and
j identified the concerns noted in previous sections 1(a), (b), (c), and (d). As of the
end of this inspection, the licensee had not evaluated the status of the 19 NSR
~
- modifications.
) The team determined the apparent contributing causes for inadequate post-
modification testing and incomplete closeouts to be:
e inadequate control cf the modification closeout and testing process.
- Lack of management expectations such that installed modifications were
allowed to be in use prior to successful testing and package closecut.
!
11
1
0
. .
,
- Lack of management oversight and involvement in the modification testing
and closecut process. ;
- Lack of configuration management control.
1
'
- Lack of questioning attitude by the engineers that possessed these
modifications for an extended period of time.
1
'
- Inadequate modification turnover program (from engineer to engineer).
- Inadequate process for prioritization of activities (conflicting priorities).
- Large backlog of engineering work activities. .
l
Also, the modification process relied on the lead engineer to verify that all testing,
documentation changes, and any other requirements listed in the modification
package were accomplished prior to closing the modification. No other tracking
mechanism was in place to assist the lead engineer or support the process in the
lead engineer's absence.
The team informed the licenses that:
(a) Failure to ensure that SR modification c!eseout requirements have been successfully
accomplished in accordance with portions of the Modification, Installation and
Testing procedure, Sections G.6.d, " Quality Control;" G.6.e, " System Engineer;"
G.7.c, " Modification Coordinator;" G.7.d, " System Engineer Supervisor;" and
Appendix B, Section 3, " Mod Test Results Reviewed" of procedure ZAP 510-02,
" Plant Modification Program," prior to declaring the modified SSCs as operable and
placing them in use is an apparent violation of 10 CFR 50, Appendix B, Criterion V
(295/96011-02A(DRS); 304/96011-02A(DRS)).
(b) Operating the plant with installed SR modifications that have not been completely
tested to demonstrate that modified SSCs will perform satisfactorily on demand is
an apparent violation of 10 CFR 50, Appendix B, Criterion XI
(295/96011-03A(DRS); 304/96011-03A(DRS)).
(c) Failure to indicate by use of suitable marking or tracking systems the operability
status of SR SSCs modified by plant modifications that had been installed as early
as 1986, placed in use, but not declared operable or signed as completed, is an
apparent violation of 10 CFR 50, Appendix B, Criterion XIV (295/96011-04(DRS);
304/96011-04(DRS)).
c. Conclusions
The team concluded that the licensee had operated the plant, in some cases for up
to ten years, with multiple installed modifications that were placed in use by plant
staff, but had not been completely tested (construction, modification and operation
tests) or authorized by operations. At least one modification was being used in a
12
- . - - _. -.
. .
mode which was not adequately analyzed (i.e., continuous bypass of the orifices).
In one case, a second modification was installed and changed the design / logic
configuration of the first installed modification; testing was not completed.
However, the team did not identify any immediate safety issues with the
incomplete modifications. Four apparent violations with multiple examples involving
procedures, 50.59 SEs, test control and operability status were identified.
E1.2 Parts Reolacement Proaram
a. Insoection Scooe
1
The team reviewed parts evaluation No. 2-95-xx-00989-00, dated January 24,
1995, and the related replacement package to determine its technical adequacy and ,
its adherence to site procedures. The parts replacement package allowed the j
replacement of a valve disk and stem with similar parts.
b. Observations and Findinas
The intent of the valve disk replacement was to increase the thrust margin by ;
increasing the thrust to fail the disk. The team's review of the parts replacement
package indicated that the only change to the disk was a slight material change .
(from CF8 to CF8M). Per the package, there was no increase in the new material I
allowable stress or ultimate stress values. Therefore, the team concluded that
some other change had been made to the valve disk to accommodate the increased
thrust value for failure. The team determined that this particular design activity
lacked the required interface between the parts replacement engineers and design
engineers. Also, the technical expertise of the engineers performing and reviewing
this parts replacement was determined to be lacking in that there was no apparent
knowledge of the reason behind the change. Subsequently, the licensee reviewed
the documentation provided by the disk manufacturer and verified that strength
properties of the disk had been changed. The licensee agreed that the original parts
replacement package was in error and agreed to revise the package to address the
change. Although the distinction between the disks was minor, and the safety
significance appeared minimal, this issue high lighted inadequate controls in the
cesign process,
c. Corduyions
The ter m concluded that an inadequate design review was performed on this
package. Also, a lack of engineering knowledge was noted in performing the
eva!uation in that the engineers failed to recognize that a design change to the disk
had to exist in order for the thrust margin to increase. in addition, there was poor
interdiscipline coordination between design and parts replacement engineering.
13
- . - . - - - . .- - - _ - - - - . .. - ... _ _ - _ -
. -
1-
,
E1.3 Modification Related Calculations
1. Mechanical Modification Calculations
q a. Insoection Scoce
$ The team reviewed mechanical calculations associated with plant modifications.
Three calculations were reviewed during the inspection. The calculations were
.i reviewed for technical adequacy and completeness.
b. Observations and Findinas
.
The calculations reviewed were generally well organized and complete. The
j assumptions were clearly stated and references listed. The calculations were
technically adequate in all instances but one.
The team reviewed calculation No. 22S-B-006M-0051, dated February 13,1996, I
! which determined the thrust required to open a pressure locked gate valve. The l
!
team determined that this calculation had an apparent error in the text portion of l
the calculation. The bonnet pressure was reported in the text to be acting on an
incorrect area of the valve disk. However, upon review by the licensee, it was l
determined that the calculated thrust was correct. The thrust was calculated by a
- computer program generated by the licensee corporate staff. The team agreed that !
this program did indeed use the correct disk area. It appeared that the licensee
.
plant personnel were unsure of how the program arrived at its answer. Further
.
review by licensee personnel found two additional instances in similar calculations
l where the text was incorrect and did not match the computer calculation. The
licensee agreed to revise these calculations to remove the error from the text in
'
these calculations.
1
j c. Conclusions
- The calculations reviewed were found to be adequate and technically sound, with
'
the exception of the error discovered in one calculation. However, based on that
error and the additional examples found by the licensee, the team determined that
the technical knowledge and expertise of the mechanical and structural engineering
staff appeared lacking in certain instances.
2. Electrical Modification Calculations
a. Insoection Scoce
The inspection team reviewed 11 electrical engineering calculations, four of which
related to the plant modifications that were reviewed. The remaining, though not
reviewed in close detail, related to the 4 kV bus second level undervoltage or
degraded voltage relay setpoint changes that were to be completed during the
Unit 2 outage, which was scheduled to start in September 1996.
14
_ _ ._ _ _ _ _ _ _ _ _ _ _ . . _ . _
,
. .
1
!
j' b. Observations and Findinas
! The team determined that the electrical calculations reviewed, which were
- performed in the last two to three years, were of good quality, well documented
! with regard to assumptions, input, methodology, conclusions, etc., and served their
j purpose well. The team concluded that the electrical calculations reviewed were
l better than the industry average based on what had been observed by the team
i during inspections of other nuclear stations.
l
l c. Conclusion .
!
! The inspection team found the electrical calculations that were reviewed adequate
and appropriate to suppoit the plant's modifications for which they were intended.
{'
4
{ E1.4 Temoorarv Alte gtions (TAs)
I a. Insoection Scone
The team reviewed selected TAs for technical adequacy and completeness.
1 TA 96-013 was performed to install heat tracing and insulation on a portion of the
l Si recirculation piping to prevent freezing of the water. Another TA was performed
j to provide a temporary method of filling the CS diesel fuel oil day tank.
b. Observations and Findinas ;
l
The team reviewed TA 96-013, dated March 18,1996, which installed temporary .
heat tracing and insulation for the SI recirculation piping. This TA was determined
to be inadequate in several instances. The TA lacked sufficient detail in describing
the actual modification. There was little description of a thermocouple which was
installed on the piping to verify temperature. The team reviewed the SE screening
which was performed to support this alteration. This screening determined that a
50.59 SE was not required. The basis fer this conclusion was that the modification
did not alter the method of performing the function of the recirculation piping.
However, the heat tracing and insulation appeared to be necessary to prevent the
piping from freezing, and this was a change to the system as described in the l
UFSAR. The screening also addressed the impact of the modification on another !
safety system. Both of these points should have prompted a 50.59 SE per site I
procedure ZAP-100. A review of the installation indicated that there were several
failure modes associated with the heat tracing and insulation which should have
been reviewed and documented in accordance with the requirements of 10 CFR
50.59. !
The licensee stated that a 50.59_ SE was not required, since the modification was i
'
not required to ensure that the piping did not freeze and it did not constitute a
change to the design of the Si recirculation system. However, the team disagreed
with the licensee's conclusion and noted that the licensee event report which was i
performed to report the freezing condition specifically discussed the use of the heat
l
15
. -
. .
tracing and insulation as a means of protecting the piping from freezing and
ensuring the safety function of the SI system.
The team informed the licensee that failure to perform a 50.59 SE and document
the results for a temporary modification that appeared necessary to prevent
freezing, was different from the UFSAR, could potentially impact another system,
and could introduce new failure modes is considerud an example of an apparent
violation of 10 CFR 50.59 (295/96011-01B(DRS); 304/96011-01B(DRS)).
E1.5 10 CFR 50.59 Safetv Evaluations
1. General Review of Safetv Evaluations
a. Insoection Scoco
The team reviewed selected 50.59 SEs. The review included technical adequacy,
completeness, and thoroughness of the review for any Unreviewed Safety Question
(USQ), particularly with respect to potential failure modes. The review involved
various SR modification evaluations and safety evaluations associated with other
activities. The other activities consisted of normal plant evolutions, such as
scaffolding erection, a setpoint change for a radiation monitor alarm, and a diesel
fuel oil tank out of service,
b. Observations and Findinas
The review identified several weaknesses in specific 50.59 SEs as well as in the
process. Details of the specific instances are provided below.
(1) Safetv Evaluation 50.59/0166/95
1
The team reviewed safety evaluation 50.59/0166/95, dated November 16,1995 l
(revised on April 12,1996), performed to allow scaffolding around the Si l
accumulators to remain in place inside containment during plant operation. The
accidents reviewed in the original 50.59 SE to support the installation were
incorrect. Also, there was little technical basis to support the USQ determination. 1
This 50.59 SE was also reviewed by the licensee's off-site review committee four I
months after scaffolding installation and was determined to be unacceptable. As a
result, the safety evaluation was revised in April 1996. However, because the
licensee had placed the units in operation during the four month period, this review
process was inadequate to prevent a modification or other activity with a potential
USQ from occurring.
A subsequent NRC walkdown identified that the scaffolding was installed near
instrumentation and piping. The revised 50.59 SE still did not discuss any potential
impact on thN piping and instrumentation as a result of the scaffolding installation. i
The 50.59 SE also did not address other interactions, such as thermal growth of
piping and pipe break effects. Further, one of the scaffolds was installed around
the pressurizer relief tank (PRT) instead of an accumulator. However, other than
16
,
_ . _ __ _ _. _ _ __ _ _.
- . -
1
4
k *
l noting this fact, the safety evaluation made no specific reference to the PRT or its
} associated piping and instrumentation. All of the descriptive rationale included in
)
4 the safety evaluation was based on installation around accumulators. The failure to
j. perform an adequate safety evaluation is considered an example of an apparent
- violation of 10 CFR 50.59 (295/96011-01C(DRS); 304/96011-01C(DRS)).
l -(2) SR Modification E22-1-95-218
The team noted that modification E22-1-95-218, "MOVs SI-8803A/B Bypass Line
-
- Addition," and the associated 50.59 SE credited a containment isolation valve seal
water system function for the volume control (VC) system that was not credited in
. the Updated Final Safety Analysis Report (UFSAR). In this case, seal water system
3 function means that the charging pumps in the VC system would stay running
I
following an accident.- The pressure from the charging pumps would be applied to
j the outside containment side of the valve, thus " sealing" any radiation inside
. containment and maintaining containment integrity. The licensee subsequently
j- identified two additional instances where a plant system was not credited as a seal
i system in the UFSAR, but had been assumed to be a seal system in the licensee's
i containment isolation testing program. The safety evaluation was inadequate in
j that it failed to identify that UFSAR Table 6.2.4 states that there is no seal system 3
! for containment penetration P-4. The modification took credit for the seal system. l
l This was considered by the NRC to be a change to the facility as described in the
l- UFSAR. As of August 22,1996, this UFSAR change had not received a written
l safety evaluation. This is considered an example of an apparent violation of ]
t 10 CFR 50.59 (295/96011-01D(DRS); 304/96011-01D(DRS)). '
i
) In addition, the team determined that the containment isolation system engineer
! was not consulted by design engineering and had no input to this particular
i modification.
!
! 2. 50.59 Safetv Evaluation Process Deficiencies
i
! a. Inspection Scope
!
l The team reviewed 50.59 SEs and procedures used by Zion for generating safety
i evaluations. These included ZAP 100-06, Rev. 7, " Safety Review and Approval,"
l and NEP 04-03, Rev.1, "10 CFR 50.59 Safety Evaluations."
I
b. Observations and Findinas
.
The team identified numerous concerns in the sample of safety evaluations
l reviewed during the inspection as well as with the 50.59 SE process. Deficiencies
- in the 50.59 SE process were also identified during a recent Site Quality
- Verification (SOV) audit completed in August 1996.
i ~
i
The team determined that the numerous examples of 50.59 SE violations identified
j during the inspection indicated a continuing lack of quality, completeness and
,
thoroughness relative to 50.59 SEs, as follows:
)
j 17
,
_ _ - _ _ . - - .
_ _ . _ _ . _ . . _ _ . _ _ _ _ _ _ _ _ _. _ _ _ _ . _ _ . _ _ _ ._. _ _ . _ ___.m_
. -
.
!
J
. (1) TA 96-14 installed heat tracing and insulation in March 1996 to preclude freezing
l (along with administrative controls) of the Si pump mini-flow recirculation piping.
l As of August 22,1996, the change did not have a written safety evaluation to
k
provide the bases for the determination that the change does not involve a USQ or
. TS change (see Section E1.4 for details).
4
(2) Modification E-22-2-95-218 credited the use of the charging system as a seal
, system for purposes of containment isolation. The safety evaluation for this
modification was inadequate in that it did not identify that UFSAR Table 6.2.4
i states that there is no seal system for this containment penetration (see
i Section E1.5.1.b(2) for details).
J
j (3) Technical Specification (TS) 4.3.1.A.4.b and TS 3.7-1 were incorrect and had been
l changed via TS Interpretations (TSis) 89-03 and 94-02, respectively. The TSis did
j not receive a 50.59 SE and the licensee did not identify that the TSI would result in i
! a change to the TS. Subsequently, the licensee identified TSI 95-01 and TSI 95-03
i as also requiring amendmtnts (see Section E8.1.b for details).
i i
- (4) SE 50.59/0166/95, performed to leave scaffolding installed in containment (four !
'
scaffolds per unit) during operation, was inadequate. Failure effects of the !
- scaffolding were not appropriately analyzed (see Section E1.5.1.b for details).
] (5) Modification E22-1(2)-93-248 changed the AFW pump steam supply steam traps.
- The modification was operating in a mode which was not adequately analyzed. No
j design change or 50.59 SE was performed to address the new operating mode prior
- to this inspection (see Section E1.1.b.1.(c) for details). i
}- Additionally, the NRC considered that corrective actions to four 50.59 SE related
j violations issued by the NRC over the last three years were ineffective in correcting
j 50.59 related problems (Reference NRC Inspection Report Nos. 50-
' 295/304-93023; 50-295/304-94020; 50-295/304-94021; 50-295/304-95014).
i
i
! ' A recent SOV audit at Zion identified numerous 50.59 SE procedure
j noncompliances across the three engineering groups and among alllevels of the
engineering organization. This finding became a Level 11 Corrective Action Record
,
(CAR). The issues identified were similar to 1993 SQV findings that had also
j on completed training and perceived improvements in 50.59 SE quality,
f Deficiencies in 50.59 SEs identified during the 1996 SQV audit included:
- -
'
e inconect safety evaluation answers.
e missing safety evaluation information.
! e deficiencies in quality, completeness and thoroughness.
! e incorrect determination of UFSAR effects.
e use of wrong revision of safety evaluation forms.
i
i
a
The team also determined that there was no mechanism in place to track and
j address commitments made in 50.59 SEs.
l
t
18
4
1
+
.-r , _ ,,- - ,
.- - - - . - . = . - -- - _. . - .
. *
4
Based on interviews and numerous discussions with design and modification
i engineering staff who were qualified reviewers and preparers, the team concluded
that the 50.59 SE process was not always clearly understood, appreciated or
rigorously applied. For example, the shortcomings in the safety evaluations for the
j scaffolding in the containment and for the Si piping heat tracing were considered by
Zion staff to be of minor significance and were not rigorously re-evaluated. It
appeared to the inspectors that the 50.59 SE reviews were viewed by some
licensee staff as "just another piece of paper to complete" versus understanding the
purpose behind 50.59 and conscientiously applying the tests.
3
,
c. Conclusions
4
,
The 50.59 SEs reviewed by the team were, in some instances, technically weak
j and did not provide sufficient detail to support the determination of no USQ. In at
4
least one instance, a system described in the UFSAR was modified without a 50.59
SE being performed. The expectations for quality standards of SEs were unclear,
and the training in response to earlier violations appeared to be inadequate for
preparers and reviewers in alllevels of the organization.
.
, in view of the repetitive nature of these violations and of the licensee's failure to
implement effective corrective action for previous violation and self assessment
-
findings relative to 50.59, the team's findings demonstrate significant deficiencies
in engineering activities and management oversight of the 50.59 SE process.
The examples of a violation of 10 CFR 50.59 collected in this section are discussed
in more detail in other sections of this report. However, these examples are being
considered collectively as a single apparent violation, indicative of a breakdown in
the licensee's process for complying with 10 CFR 50.59.
E1.6 Review of Ooerability Assessments
The team reviewed selected operability assessments performed by the licensee.
The operability assessments were reviewed for technical adequacy as well as
thoroughness of the arguments used to establish operability.
1. Review of Ooerability Assessment ER95004439
a. Insoection Scoce
The team reviewed operability assessment ER95004439, dated November 8,1995,
which verified that a residual heat removal to refueling water storage tank suction
valve was operable in Mode 4. The assessment addressed the potential pressure
locking concern for that valve.
b. Observations and Findinas
The team determined that the assessment did not conclude that the valve was
operable, and recommended a procedural change to lock the valve open. This
19
,._ _. _ . _.._ _ _ .. _ _ _. _ _ _ _ _ _ _ _
. . t
,
,
change was rejected by operations and the operability assessment was revised only
to reflect this fact. No additional evidence to support operability was provided until
the licensee was questioned by the team. The additional information, that the valve
does not have a design basis function to open during Mode 4, was sufficient for the
team to concur with the operability assessment.
The team reviewed several assessments that were written for the same systems by *
different authors. These assessments differed in the list of system safety
functions. For example, in one instance, the penetration seal function was not ,
listed for the VC system. The licensee informed the team that the operability
assessment would be revised to add the required penetration seal function.
2. Review of Ooerability Assessment ER9604259
a. Insoection Scope
The team reviewed operability assessment ER9604259, dated August 13,1996,
which supported unit operability with scaffolding installed in the containment. .
b. Observations and Findinas ,
,
The operability assessment was determined to be ineffective in that evidence was
not provided to support the operability determination that the scaffolding was
acceptable for long-term use in the containment. The team was concerned that the '
determination was made based solely on engineering judgment. The team '
performed an independent walkdown of the scaffold installation in Unit 1 to verify
the assertions made in the operability assessment.- The team noted that
interactions between scaffolding, equipment and instrumentation were not fully -
addressed. ;
3. Review of Ooerability Evaluation ER9603282 (Charaina Pumo Performance
Dearadation)
a. Inspection Scone
The team reviewed operability assessment and engineering request ER9603282
(PIF 96-1218), portions of the Emergency Core Cooling System (ECCS) Small Break
Loss of Coctaat Accident (SBLOCA) analysis, ECCS test procedures TSS 15.6.4,
and in Service Testing (IST) documents,
b. Observations and Findinos
The licensee generated an operability assessment as a result of a UFSAR
conformance review finding related to charging pump degradation. The team
determined that the initial operability assessment was inadequate. The assessment i
failed to address the affects of pump degradation on design flow requirements until I
prompted by the NRC. !
l
20
i
. _ __ __ _ _ _ __ -
_ . _ _ _ . _ _ _ _ - . _ _ _ _ _ _ _ __
, .
..
By plotting the full flow (approximately 500 gpm) pump test data onto the IST
pump curves, the licensee's conformance review determined that all four charging
pumps had total developed head of greater than 10% difference from the
corresponding IST pump curve values. However, Pump 1B had the greatest
performance degradation with (-)24%.
In response to the identified degradation, the licensee determined in its initial i
operability assessment that no operability issue existed because: (1) the pump had
consistently met the IST test acceptance criteria conducted at low flow; and (2) the
pump had met the ECCS full flow test acceptance criteria of less than pump runout.
Since the licensee determined that an operability issue did not exist, no further
corrective action was taken or required. The NRC found the operability assessment
to be inadequate because the pumps' ability to deliver the required flows at design
conditions was ignored in the operability assessment. Neither of these tests
verified minimum flow at SBLOCA design conditions.
In view of the degradation data, the NRC was also concerned with the validity of
the IST 10%-degradation curve used for these pumps. Because pump testing at
design conditions (approximately 325-450 gpm at 1100-1500 psig for SBLOCA)
was not possible, pump degradation testing under IST was conducted at lower than
design condition flows. IST testing for these pumps was conducted using the
pump minimum flow test line at a reference point of approximately 100 gpm and
recent testing indicated that the pumps' performance was within 3% of the
reference curve at that point. A basic assumption of the IST program was that the-
10% required action degradation curve stayed constant along the entire pump
curve. But if evidence of greater than 10% degradation is noted, even if it's at a
.
point different from the IST testing point, the validity of the 10% degradation curve
is questionable and its applicability to design conditions cannot be ignored.
ECCS full flow test TSS 15.6.84, " Charging and Safety injection Flow Test," dated
October 1995, contained the noted pump degradation data. The stated purpose of
the test was to verify that the flow rates were within the bounds of those assumed
in the accident analysis for Zion station. The acceptance criteria, which was
recommended by Nuclear Fuel Services and Site Engineering to ensure minimum
acceptable ECCS performance, verified that runout flows were not exceeded and
that large break loss of coolant accident minimum flows were satisfied. However,
the criteria failed to address minimum SBLOCA flow requirements. This inadequate
test evaluation is considered an example of an apparent violation of 10 CFR 50,
Appendix B, Criterion XI (295/96011-03B(DRS); 304/96011-03B(DRS)). The
licensee's conformance review actually identified this problem and provided it to the
station in an exit on June 28,1996. However, as described earlier, the licensee's
response to the issue exacerbated the situation by failing to recognize the problem
and thus failing to take action.
After prompting by the NRC, the licensee reviewed the effects of the worst case
pump degradation flowrates at design conditions. This was done by assuming a
degradation trend curve drawn through the two available data points (IST and full !
flow test data) and by comparing these flows against those assumed in the ECCS
21
. -
SBLOCA accident analysis. The comparison demonstrated that the flowrates
available with the worst case pump degradation were still greater than the flowrate
values used as input into the accident analyses. Although the margin in the
analysis had prevented the unnoticed degradation from impacting the pump's
specified function, the NRC viewed this as fortuitous. Had the unnoticed
degradation been slightly higher, the analytical conservatism may not have been
sufficient. The charging pumps may not have been able to deliver the SBLOCA
design flows, but would have still met the full flow test and the IST test acceptance ;
criteria. )
4. Conclusion for Section E1.6
The team concluded that in general, the operability assessments reviewed were
weak and some were inadequate. For example, the operability assessment for
charging pump degradation noted during ECCS full flow testing was inadequate.
The operability assessment did not evaluate operability at design flow conditions
and the licensee failed to fully address this until prompted by the NRC. The team
also noted that the amount of detail provided in the assessments reviewed was
inconsistent. Several assessments contained sufficient detail to verify the
operability determination, while others provided the bare minimum of information.
There were also discrepancies in the system safety function descriptions.
E1.7 General Conclusion for Conduct of Enaineerina (Sections E1.1 throuah E1.6)
The team determined that in general, many of the modification packages reviewed
were technically correct and complete. The modifications, in most cases,
appropriately addressed the problems they were to resolve. The calculations
reviewed were found to be generally adequate and technically sound. The testing
portions of many of the modifications reviewed contained the tests required to
verify successful installation. However, the team found that the engineering
modification process was deficient in areas associated with modification completion
and post-modification testing, engineering knowledge, design reviews, 50.59 SEs,
operability assessments and interdiscipline coordination between engineering j
'
groups. The deficiencies in the licensee's control of modifications and compliance
with 10 CFR 50.59 were widespread and repetitive, such that licensee programs in
these areas appeared to have broken down. Four apparent violations involving test
control, safety evaluations, procedures and operability status were identified.
E3 ENGINEERING PROCEDURES AND DOCUMENTATION !
E3.1 UFSAR Conformance Review
i
a. Insoection Scope ;
!
The team reviewed Zion's UFSAR conformance review findings, plans for resolution j
of UFSAR discrepancies, PlFs and operability assessments generated from the
findings. The team also reviewed Zion procedure 700-08, Rev.10, " Problem
identification Process."
22 i
._ _ . _ . . _ . _ _ . _ _ _ . _ . . _ _ _ _ . _._ _._.- . _ ._. _ _ _
,.
. .
!
<
!
j' b. Observations and Findinas
!
! During June 1996, Zion performed a vertical slice sample review on the Chemical .
1
and Volume Control System (CVCS), the radiation monitor (RM) system, and the
4
steam generator tube rupture UFSAR sections in response to NRC Information
i' Notice 96-17. The UFSAR conformance review focused on operating practices and
j- procedures to ensure conformance with the UFSAR. At the conclusion of the v
review on June 28,1996, Zion's review team had identified 175 UFSAR
'
l
- . discrepancies. Upon further review and validation, the number of discrepancies
{ was reduced to 115. ,
! The NRC determined that the UFSAR conformance review was a positive
j management effort, and noted that the team consisted of experienced technical
F reviewers that identified important issues. However, after identification of the
! discrepancies, the licensee failed to promptly document the identified deficiencies in '
l the corrective action process.
!
Zion administrative procedure ZAP 700-08,0, " Problem identification Process," l
Rev.1, Appendix A, item 16, required generation of PlFs for events or conditions ;
identified by an assessment group. Although the conformance review was _
completed by the end of June, only one discrepancy had a PlF generated as of the
week of July 22,1996, when the NRC team arrived on site. Twenty-four PlFs l
were subsequently generated for the remaining discrepancies.
The failure to generate the PlFs for identified self assessment issues as prescribed -
by ZAP 700-08 is considered an example of an apparent violation of 10 CFR 50,
Appendix B, Criterion V (295/96011-02B(DRS); 304/96011-028(DRS)).
Since PlFs were not promptly generated for the conformance review discrepancies,
the licensee had not decided whether to fix the discrepancies or change the FSAR.
As a' result, the team was concerned that the licensee had not promptly addressed
the identified discrepancies under 10 CFR 50.59, either. If the licensee chooses not
to fix the discrepancies under an Appendix B corrective action process, then
changes in the facility or in procedures as described in the Safety Analysis Report ;
(SAR) must have a written SE which provides the bases for determining that the j
'
change does not involve a USO. Instead of formally pursuing either of these
courses of action, Zion relied on a group of station engineering and operations
personnel to effect daily reviews of these discrepancies for safety significance and j
operability. The NRC was concerned that these reviews were not documented until 1
after the NRC team questioned the review process.
c. Conclusions ;
Zion's UFSAR conformance review was considered a positive management effort.
The NRC noted that the review team consisted of experience,d technical reviewers.
This review eppeared to be thorough and rigorous, identifying several meaningful
issues. However, after identification of the discrepancies, the licensee failed to
initiate prompt corrective action to address the identified deficiencies or determine if .
23
_ _ _ _ __ _ _ ..
. .
a 10 CFR 50.59 concern existed. One example of an apparent violation concerning
procedures was identified.
E3.2 Vendor Eauioment Technical Information Proaram (VETIP) ,
a. Inspection Scope
The team reviewed the licensee's vendor document control program including
i
etaluations of vendor information and the vendor document backlog. Documents
,
reviewed included procedure ZAP 510-01, " Vendor Equipment Technical ;
i information Program," Rev. 4 and procedure NEP 07-04, "VETIP Process Control,"
Rev.O.
i
l
'
b. Observations and Findinas
!
The licensee had recently implemented a program to improve the VETIP process by l
'
reducing the backlog of unreviewed VETIP binders and by streamlining the process,
j However, evidence of progress in reducing the ^cecklog was minimal; also, the NRC
3
was concerned with the potential use of non-controlled VETIP binders for activities ;
affecting quality. ;
- I
!
As of this inspection, a backlog of cpproximately 260 VETIPs were pending review
, of revision / update documents. The team could not evaluate the effectiveness of
I the licensee's recent initirdves. No cata was yet available to demonstrate an .
'
improving trend in backf ag reduction.
I l
l The team noted that the VETIP binders were uncontrolled and considered reference
! oocuments only; however, the team determined that unrevised VETIPs may j
, potentially be used for activities afbeting quality. For example, NTS items i
,
304-201-96 CAT 4-0426, 0488, and 0610 detail caces where VETIPs were required I
i to repair or troubleshoot instrumentation after failed surveillance tests. This
indicated that the maintenance procedures had insufficient detail for the
'
maintenance personnel and that the uncontrolled VETIPs may have been required
for activities affecting quality. The team noted that lack of detail in maintenance )
procedures led to reliance on non-controlled VETIPs for completion of activities
affecting quality. This issue is considered an Unresolved item (URI) pending review
of the licensee's actions, including review of maintenance procedure adequacy
(295/96011-05(DRS); 304/96011-05(DRS)).
c. Conclusions
The team concluded that weaknesses in maintenance procedures and in the VETIP
program may allow use of uncontrolled VETIPs for weg activities affecting quality.
Some changes to the VETIP program had been initia1M but evidence of backlog
reductions was not apparent.
,
24
\
l
. , - -- -
_ -- _ . _ . , _ - . - -~ -
-.
~
,
l
j
j b.3 Miscellaneous Enaineerina Tasks (Procedures and Documents)
- -
l a. InsDection Scone !
!
t i
l The team reviewed various engineering refred p.rs.:edures and documents. These i
j included PlFs and other procedurally drivem activities such as surveillance testing ;
j and scaffolding.
i b. Observations and Findinas
i
l 1. Scaffoldina Erection inside Containment
l The team reviewed the details of the installation of scaffolding inside the
i containment (portions of this issue are also discussed in Section E.1.5.1.b).
l This scaffolding was installed and left in place during unit operation. . Four
'
i scaffolds were erected in the Unit 1 and Unit 2 containments (eight total),
i- one near each Sl accumulator tank, with one exception, where the
! scaffolding was built near the pressurizer relief tank. The scaffolds were
i erected under the site scaffolding procedure ZAP 920, Rev.1.
L Procedure ZAP 920 requires that scaffolding be removed following work
I completion. The scaffolds were erected to allow operations access to valves
- manipulated during start-up and shutdown. Since the scaffolds were located
! inside the containment and the units were buth operating, no work was
{ being performed. Therefore, the scaffolds should have been removed.
j Essentially, the scaffoldings were being used as continuous and psrmanent
a changes. The team determined that, contrary to the requirements of
l ZAP 920, the scaffolds had not been inspected every month since
installation. The team also determined that the scaffolds were installed in
j direct contact with SR equipment. This was also contrary to the
j requirements of ZAP 920. The team also noted that the licensee corporate !
level scaffold procedure (TID-MS-01) contained greater detail than the site
procedure. This was determined to be a weakness in the site procedure. !
! The licensee agreed to perform a modification to permanently install the
scaffolds.
!
- The team informed the licensee that failure to follow procedure requirements, as
l discussed in the paragraph above, is considered an example of an apparent violation
l of 10 CFR 50, Appendix B, Criterion V (295/96011-02C(DRS);
304/96011-02C(DRS)).
2. System Performance Monitorina information
The team noted that in the r anth of June 1996, at least eight system engineers i
failed to update the materiel .:ondition monitoring information used for trending of I
component and system performance on 13 systems. This information is used for
the maintenance rule trending data and materiel condition windows color
determination. Procedure ZAP-500-13A requires a monthly update of the window
25
.- -
.
.
data. This is considered an example of an apparent violation of 10 CFR 50,
Appendix B, Criterion V (295/96011-02D(DRS); 304/96011-02D(DRS)).
l
3. Safety Iniection Flow Balancina and Full Flow Testina
The team reviewed the procedures associated with the performance of the Unit 1 Sl
flow balancing and full flow testing (TSSP 139-92, dated June 22,1992, and TSS
15.6.84, dated October 25,1995). The team reviewed the full flow injection
procedure and noted an apparent discrepancy in the differential pressures obtained
in the four chaiging injection paths. Three of the four differential pressures were
identical and one was approximately half of the other three under the assumed
same flow condition. This indicated that the flow in this leg could be less than
anticipated. The team reviewed the original procedure, which established the
differential pressure for flow balance. This procedure had the same apparent
discrepancy. The licensee performed a walkdown of the piping near the differential
pressure taps to determine if differences in piping configuration could be the root
cause of the discrepancy. The results of the walkdown indicated that the piping
configuration could not account for the discrepancy, but the orientation of the flow
measurement piping may account for the discrepancy. The licensee reviewed the
Unit 2 procedure and determined that a similar condition could exist. The licensee
then performed an operability assessment to determine the effects of the potentially
degraded flow on unit operability. The licensee also committed to verify flow
independently of the differential pressure method during the next refueling. This
item is considered unresolved pending licensee test results from both units
(295/96011-06(DRS); 304/96011-06(DRS)).
4. Protective Relav Settina Chances
The team found that protective relay setting changes for the SR 4 kV system were
the responsibility of the corporate System Protection Department. This was said by
the licensee to be a non-nuclear expertise group of protective relay specialists who
were responsible for protective relaj settings,4 kV and above, throughout the
Commonwealth Edison system. Any 4 kV relay setting changes were to be
developed by the System Protection Department and issued using Relay Setting
Orders to Zion Station for implementation. The team was concerned that such
relay setting changes could be implemented without the formal review, concurrence
and approval of the electrical design group (Electrical /l&C Engineering) at Zion
Station regarding impact and compatibility with the Class 1E 4 kV system's design
and safety functions. In response to this concern the licensee stated that the need
to make specific requirements for Electrical /l&C Engineering review prior to
implementation will be evaluated. This was acceptable to the team.
5. Desian Basis Document and Diaaram Revision
The team observed that the design basis document relating to the station's
electrical system, including the Auxiliary Power System Design Basis Document
Manual, DBD-Zl-024, was well written and adequate in scope; however, it was out-
of-date in that it was issued in 1992 and had not since been revised to resolve
26
i
_ _ _ _ _ _ . . . _.. _ _ _ .. _ .. _ _ __ _ _, _ _ _ _ _ _ _ .
.
, .
! <
1
]
- i
l i
<
several open items and recent plant modifications. The licensee recognized that the l
j document required updating.
i
- During the review of the degraded voltage relay setpoint changes, the team found
,
'
that six SR Key Diagrams, 22E-1-4000H, 22E-1-4000J, 22E-2-4000H,
22E-2-4000J, 22E-1-4000C, and 22E-2-4000C, had not been revised to show the
l addition of the SR degraded voltage relays which had been installed in mid-1980.
1
The licensee issued PlF 96-1883 to address the issue. The team informed the
licensee that failure to revise the six diagrams to show the SR degraded voltage
j . relays is an example of an apparent violation of 10 CFR 50, Appendix B, Criterion V
1
3 6. 50.59 SE Procedures
- The NRC identified that weaknesses in the safety evaluation procedure wording (for
j both ZAP 100-06 and NEP 04-03) may allow a USQ to go unnoticed, because
facility or procedure changes that "may increase the probability and consequences
y of malfunction of equipment important to safety" are specifically reviewed only if
- the change affects an analyzed accident. In response, Commonwealth Edison
l Company committed to change and clarify the procedures to separate the
" malfunction of equipment" question so that it is also addressed with changes not
- associated with analyzed accidents. Further, a sample of previously completed
I 50.59 SEs was being reviewed to ensure that past facility changes appropriately
i- addressed thir 'ssue.
l
! 7. Inadeauate Emeroency Ooeratina Procedure I
i
l The team reviewed the Station Emergency Operating Procedures to determine if the
- VC system's required safety function as a penetration seal water system was !
j. incorporated 'in the procedures. This was needed to ensure that the VC pumps
f were not turned off when the seal water system was needed.
!
l The team noted that Emergency Procedure ES 1.3, " Transfer to Cold Leg
.
Recirculation," Rev.18, failed to include guidance and instructions for the VC
- system, which was used as a penetration seal water system. This could have
i resulted in system operation being terminated and loss of the penetration seal. This
! is considered an example of an apparent violation of 10 CFR 50 Appendix B,
Criterion V (295/96011-02F(DRS); 304/96011-02F(DRS)).
'
! The licensee agreed to review this situation and update the procedures as required.
!
'l c. Conclusion
- - Tha team identified examples of failure to follow procedures and inadequate
'
procedures. Also, electrical diagrams had not been revised to reflect as-built
conditions.
3
,i
j 27
4
i
, .. ,- ,,, -
- - - . . . - - , , . - . , -
. -
The team concluded that contributing factors to the procedural problems noted
included: (1) lack of clear management expectations; (2) lack of questionin0
attitude and attention to detail by engineering personnel in performing tasks; and
(3) failure to follow procedures and inadequato quality of procedures.
E4 ENGINEERING STAFF KNOWLEDGE AND PERFORMANCE
E4.1 Root Cause Determination and Corrective Action Process
a. Insoection Scope
I
l
The team reviewed the licensee's root cause investigation process, which included
'
reviews of PlFs, root cause investigations, and corrective actions taken on identified
problems.
b. Observations and Findinas
1
(1) Recent Root Cause Initiatives l
I
The licensee recently implemented root cause investigation training initiatives aimed
at event reduction by focusing on prevention, detection and correction.
Implementation of these methods included formation of a dedicated multi discipline i
root cause investigation team with its goal being to improve the quality of I
investigations, of trend analyses and to propose effective corrective action. The
team concluded that Zion's attention to improving root cause investigations was
necessary and was viewed as a good management initiative. However, the team
noted that engineering corrective action efforts have failed to identify the root
cause of problems noted. Technical documents and evaluations reviewed have
been incomplete or narrowly focused. The following examples were identified.
(2) Inadeauate Root Cause of Recurrina Occurrences
The team reviewed several PIFs associated with instrument out of tolerance
conditions for the Containment Spray (CS) system sodium hydroxide (NaOH) Spray
Additive Tank level indication (PlFs 295-201-95-CAT 4-1227,
295-201-96-CAT 4-0010, 295-201-96-CAT 4-0011, 295-201-96-CAT 4-0916, and
295-201-96-CAT 4-1044). The review determined that there were numerous
instances of out of tolerance over the past two years. The PlFs associated with
these instances were Category 4 PIFs, which meant that no root cause would be
performed. The number of consecutive PlFs should have triggered a soot cause PIF
to identify the cause of the problem and rectify it. The licensee informed the team
that there was no procedural number of consecutive Category 4 PlFs which would
trigger a root cause evaluation. The team determined that this was a weakness in
the PIF procedure. The team did note that the system engineer had initiated an
Engineering Request to replace these level instruments prior to the inspection. No
implementation date for the modification has been established.
28
, . l
The team identified one instance when both trains of the redundant level lnstrument
were out of tolerance simultaneously (PlF 304-201-95-CAT 4-2001, dated July 27,
1995). This constituted entry into a TS action statement requiring an alternate
means of level indication within eight hours of the discovery. Contrary to this, the -
TS action statement was not entered until after eight hours had expired. The
licensee initiated a PIF to investigate this incide.nt and concluded that the out of
service condition was not identified until some time after the initial event.
Therefore, the licensee concluded that alternate level indication was established
within the eight hour limit.
Failure to identify the root cause of the repetitive out of tolerance condition of the
level instrumentation and implement appropriate corrective action to preclude
repetition is considered an example of an apparent violation of 10 CFR 50,
Appendix B, Criterion XVI (295/96011-07A(DRS); 304/96011-07A(DRS)).
(3) Reoeat Failures of 4 kV Breaker Not Trended for Root Cause
The team reviewed recent problems associated with 4 kV breakers. The team
noted repeat failures cf 4 kV Breaker No.1412 to close on demand following
corrective maintenance activities. Although this breaker was classified by the
licensee as NSR, it is the reserve feed breaker for Unit 2 Engineering Safeguards
Feature (ESF) busses and could challenge the SR function of SR loads and busses.
'
Failure of this breaker will require o :ergency diesel generator startup to feed the
ESF busses. This issue was consiaered a potentially significant condition adverse
to quality and measures were not established to assure that the cause of the
condition was determined, nor was corrective action taken to preclude repetition.
The breaker failed to close five times since September 1995. The failures were not
being evaluated or trended for root cause determination and to prevent recurrence.
A post-modification test was not always specified nor performed after the breaker
was repaired and racked back into the cubicle. There was no procedural
requirement to test breakers after they are racked back into the cubicle.
The No.1412 breaker failed to close on September 22,1995, during preventive
maintenance activities. A control relay contact was found out of position.
Following repair, the breaker remained outside bus 141 until December 1995, when
it was installed in Main Feedwater Pump 1 A, Cubicle 2 (swapped with the breaker
in Cubicle 2). The breaker was not tested in the cubicle and failed to close again on
January 3 and 4,1996. On at least two instances, on January 3 and 4,1996,
operations attempted to reclose the breaker after replacement of the control relay
contact block and before the system engineer completed his review of the work
request packages. Post-modification testing was not specified by the engineer prior
to closing the breaker. The breaker failed to close on both attempts. On April 25,
1996, the breaker failed to close again. A work request was written, but repair and
root cause determination were not performed. The breaker failed again on July 21,
1996, when operations attempted to close it after it was swapped back into its
original cubicle.
29
I
!
_ _ _ _
_ . _ _ __
., .
The team followed the licensee's troubleshooting on the 1412 breaker after it failed
to close on July 21,1996. Disassembly of the electrical contactor unit revealed
that the support for the control relay contacts was broken. Further investigation
showed that the striker kver, which was connected to a cam, was forcefully
activated by the chargir., springs striking the contact activating arm with excessive
force. The activating arm then struck the control relay contact support. This
apparently resulted in the control relay contacts being misaligned and the broken
support. The broken relay contact device was replaced and relative positions of the
striker lever and activating arm were then adjusted to minimize the force. This
appeared to be the cause of past breaker failures to close. It appeared that l
previous failures of the breaker to close had been corrected by replacing the
damaged control relay contact block unit without determining the root cause.
The team also noted that the AP system engineer did not receive all of the PlFs i
issued on his system. This could be a contributing factor to the inadequate root :
cause determination on Breaker 1412. l
Failure to identify the root cause of nonconforming conditions and implement ;
appropriate corrective action to preclude repetition is considered an example of an ;
apparent violation of 10 CFR 50, Appendix B, Criterion XVI (295/96011-07B(DRS);
304/96011-07B(DRS)).
The licensee subsequently issued PlF 96-2047 to address concerns noted in
Sections E4.1.b(2) and (3) above.
c. Conclusions i
The team evaluated the breaker problem and concluded that the apparent causes
for the repeat breaker failures were: (1) incomplete root cause identification sad
corrective action; (2) inconsistent trending and prioritization; (3) a large engineering
backlog; and (4) many equipment deficiency workarounds. Additionally, corrective
action taken with the CS system NaOH Spray Additive Tank levelindication was
inadequate, no trending or root cause was initiated. The team considered this an
example of lack of engineering staff and management attention to recurrent
problems. Recent implementation of a root cause investigation team and event
reduction training were considered positive management initiatives, but
improvements were not yet clearly evident.
E4.2 Trendina and Materiel Condition
a. Inspection Scope
The team reviewed the licensee's equipment trending program for the Si
system, CS system, and RM system. The responsible engineers were
interviewed.
30
, =
b. Observations and Findinos
The licensee had systems in place to assist department managers trend various
indicators, such as overall numbers of problems, work backlog, and work
completion. However, equipment performance trending by system engineers was-
inconsistent. Interviews with system engineers indicated that trending of system
performance and materiel condition was not very effective. The team observed
that RM system maintenance rework and monitor out of service time was trended
monthly. Other systems had no trend analysis beyond what was performed by
programs such as Vibration Monitoring and IST. Data from some programs, such
as Lube Oil Analysis, were not consistently trended. In addition, the team's revitsw
of prioritization of engineering work indicated that due to the large load of
engineering work, little prioritization was done,
c. Conclusions
The team concluded that the licensee's trending program was weak. In some
-
cases, engineers did not perform trending analyses on their systems, such as out of
service time, failure rates, and system performance indicators. Also, some
important engineering problems appeared to not have received a high priority for
resolution.
E4.3 Lube Oil Analvsis Process and Packino Leaks
a. Insoection Scoce
The team reviewed the licensee's Lube Oil Sampling and Analysis program for the
Si and CS systems. The responsible engineers were interviewed and available
sample results for motor and pump bearings were reviewed for the past three
years.
b. Observations and Findinos
The team noted that the responsibility for the Lube Oil Sampling and Analysis
program was recently transferred from one lube oil coordinator to the various
system engineers, each accountable for his own equipment. The engineers
interviewed relied on the preventive maintenance engineer to notify them that
samples were due and thus did not systematically track sample performance or
receipt of results. The engineers did not maintain organized records, and in the
case of the CS system, not all sample results requested by the team were available
for review. The responsible system engineers did not have the expertise to analyze
sample results, and in cases where the performing lab noted unacceptable results or
made recommendations for further analyses, PlFs were not written and corrective
action was not aiways taken.
The 2B SI pump bearing oil samples had high amounts of wacer and sediment at
least since 1993 with no root cause analysis performed and no corrective action
taken beyond changing the oil at the standard frequency. Review of the bearing
31
l
l
'
.y,:
'
-
_ , , _ . . .
. .
.
_ _
=
l
,
,
vibration measurements performed by the station on the 28 Si pump showed no I
degradation. Several samples in the CS system showed elevated levels of copper, ,
iron and water, in one case, the lab's recommendation of ferrographic analysis was i
apparently not evaluated or followed. A PlF was not written for any of these {
problems. Some samples had been analyzed by the licensee's own lab, which
provided a history with each sample, making trend analysis easy. Potentially
contaminated samples were sent outside the company for analysis and results did i
not include any history. In this situation, trend analysis was not usually performed. ,
!
The team noted that all CS pump packing glands leaked excessively and repair _
efforts have not been effective. in one case, a packing replacement work package
called for a new style of packing to be installed. However, due to oversight, the old
style was installed instead. After adjustments, the old style packing was left
installed.
c. Conclusions
The team concluded that the Lube Oil Sampling and Analysis program was weak.
-The responsible individuals did not track sample performance or receipt of analyses
and sample results were not retained in a useable manner. The purpose of the
system was defeated when no action was taken for unsatisfactory sample results.
At the end of the inspection, the licensee informed the team that the lube oil ,
coordinator position will be re-established. ,
-E4.4 Auxiliarv Power System Automatic Bus Transfer
a. Inspection Scope
The team reviewed the auxiliary power system automatic bus transfer logic to
assess the potential impact on the SR loads connected to the 4 kV system busses,
which were subjected to bus transfer.
b. Observations and Findinas
The team was concerned that exposure to out-of-phase 4 kV bus transfer could
potentially damage motors. During the review of the auxiliary power system's
design bases document, DBD-El-024, the team observed that no mention was made
that the scheme for the automatic bus transfer (ABT) of 4 kV busses, including the
4 kV SR busses from their normal to alternate sources of power, was superv..ad by
synchro-check logic to assure SR motors and their driven loads were not
overstressed by transient torques resulting from exposure to out-of-phase bus
transfer. Severe transients could cause immediate mechanical failure, while less
severe transients could have a cumulative affect which, in turn, could lead to
accelerated loss of qualified equipment life expectancy. ABT is reported by the i
licensee to occur during plant startup and planned or unplanned plant shutdown.
As noted in Section 15 of ANSI C50.41-1977, " Polyphase induction Motors for
Power Generating Stations," such transient torques could range from approximately
two to twenty times rated torque. To limit the possibility of damage to a motor or
32 i
-- -_- . . _ , . . . , , - ~ ,m - - , -.- .- - , _ ---
,
,- *
driven equipment, or both, ANSI C50.41-1977 recommended "...that the power
supply system be designed so that the resultant vectorial volts per hertz between
the motor residual volts per hertz and the incoming source volts per hertz at the
instant the transfer...is completed does not exceed 1.33 per unit volts per hertz on
the motor rated voltage and frequency bases." When queried by the team, the
licensee could not identify a study or calculation which demonstrated that the
"1.33 per unit volts per hertz" criterion would not be exceeded during ABT at Zion
and the criterion was not part of the .;tation's design bases. Further, the licensee
stated that the "1.33 per unit volts per hertz" criterion was not widely accepted
and was under scrutiny for acceptability. The team disagreed, since the IEEE
Paper 93 WM 033-1 PWRD, Motor Bus Transfer, which the licensee referenced in
their response, states "...the ANSI C50.41 standard was the most common
criterion for analyzing a bus transfer." Further, while the criterion may have been
under scrutiny, resolution may prove that it is not tight enough for some drives,
such as those with lorig shafts between the motors and driven loads, as is the case
with some service water pumps.
c. Conclusion
.
The team was concerned that exposure to out-of-phase 4 kV bus transfer could
potentially damage motors. Based on the team's concern, the licensee committed
to evaluate the need to perform additional analysis and take corrective action if the
g evaluation concludes there was a potential to demage motors. This was acceptable
4 to the team.
E4.5 Containment Sorav Svstem
The team determined that the design basis of the CS system was under review by-
the licensee and the NRC. Some outstanding issues and concerns include: (1) pH
-
limit of spray; (2) total amount of NaOH required in sump; (3) presently no required
surveillance to verify NaOH add flow; (4) environmental qualification of containment
equipment; and (5) need for nitrogen blanket on spray add tank.
These issues are discussed in a separate NRC inspection report and will be followed
and documented in URI 50-295/96006-08(DRP); 50-304/96006-08(DRP).
E4.6 General Conclusions (Sections E4.1 to E4.5)
The team concluded that the licensee's trending program was weak. Also, some
important engineering problems appeared to not have received a high priority for
resolution.
33
_ _ _ _ _ -
_ - - _ _ . _ _ . _ _ _ _ _ ~ _ _ . _ . _ _ _ _ _ . m._ _ . -
'
, .
I
j E5 ENGINEERING STAFF TRAINING AND QUALIFICATION
- E5.1 Enaineerina Staff Trainina and Soecialized Trainina
i
j a. Insoection Scone
f The team reviewed the licensee's engineering training program, including TPG-201,
! Rev.7, " Engineering Support Training," as well as qualification guides for
- engineering groups and individuals.
!
l b. Observations and Findinas
The engineering support training program, which applied to all engineering
departments at Zion, consisted of four phases: (1) initial
(orientations / applications / intro to systems training); (2) Secondary (power plant
operations / simulator); (3) Position Specific (selected topics / qualification guides j
based on department and position); and (4) Continuing Training (general / position -
specific).
The efficacy of Position Specific training specifically dealing with 50.59 SEs was of
concern based on the findings in the 50.59 SE area. As part of continuous training,
Zion has recently completed special training on the subject and has implemented !
mandatory continuous training requirements in order to maintain 50.59 SE preparer -
and reviewer qualification. The team reviewed 50.59 SE training course material
and training records. A sample of system / design / maintenance engineering j
personnel records indicated that 50.59 SE training requirements were satisfactorily
fulfilled. Nevertheless, based on the team's findings, the quality and thoroughness
of safety evaluations was considered problematic, warranting further attention.
c. Conclusions
Overall, Zion had a well organized engineering training program; however, based on
inspection findings, weaknesses were evident in the effectiveness of position
specific (or specialized) engineering knowledge training, such as 50.59 SE training.
E6 ENGINEERING ORGANIZATION AND ADMINISTRATION
E6.1 Zion System. Desian and Maintenance Enaineerina
a. Insoection Scone
The team verified whether System, Design and Maintenance engineering activities
that support the plant were properly prioritized, coordinated and effectively
implemented. The team examined engineering interface with other support groups
and determined effectiveness of management's involvement in the engineering
process. This was accomplished through review of engineering processes and
interviews with engineers and management staff.
34
__ __ _ - -
._ __- _ _ . . - - _ . _ _
__ _ . _ _ _ _ _ _ _ _ _ . _ _ _ . . _ _ _ _ _ _ _ . _ _. _
a
i
b. Observations and Findinas ;
)
The Engineering Department at Zion consisted of three separate departments.
l
Design Engineering, Maintenance Engineering and System Engineering. Design
Engineering was primarily responsible for allissues associated with the design basis
of the plant, including design changes, modifications and configuration
management. Maintenance engineering was primarily responsible for allissues
associated with technical support for maintenance activities, component failure
troubleshooting and trending and testing programs (i.e., IST, in Service inspection
(ISI), Motor Operated Valve (MOV), etc.). System engineering was primarily
responsible for all issues associated with technical support for operating activities,
system performance trending, system problem troubleshooting and on-site review
functions.
I
The Zion engineering organization appeared to be staffed with mostly experienced ]
and knowledgeable engineers. The team observed that most engineers interviewed i
'
appeared well qualified for their work, were cognizant of the station's design bases
and aware of nuclear plant design change practices. The team determined,
however, that the technical knowledge and expertise of the mechanical and
structural engineering staff appeared lacking in severalinstances. Also, i
management first line supervision and engineering involvement in the modification
process, including 50.59 SEs, post-modification testing, modification closure and
the resolution of engineering related issues, were considered weak. In addition, the .
corrective action program, the lube oil analysis program and the trending program i
exhibited weaknesses. The team also noted severalinstances where procedures
were not being followed or were inadequate. Examples were noted where the
required interface between design, system, and maintenance engineering was
ineffective or nonexistent.
c. Conclusions
The team concluded that the Zion engineering organization appeared to be staffed
with mostly experienced and knowledgeable engineers. However, increased
management involvement was warranted to address the deficiencies in other
sections of this inspection report.
!
E7 QUALITY ASSURANCE IN ENGINEERING ACTIVITIES
l
E7.1 Licensee Self Assessment Activities !
a. Insoection Scooe l
The team reviewed the licensee's self assessment program to determine it's
effectiveness. Various SOV and independent Safety Engineering Group audits and
surveillance reports were reviewed. in addition, corporate engineering self
assessments were also reviewed.
35
.
.
.
I
l
b. Observations and Findinas
I
'
Although the NRC's sample review of recent corporate and Zion engineering self
assessments noted primarily compliance based findings, a positive trend towards
more findings related to technicalissues was also noted. For example, the findings
in the corporate-sponsored Chief Engineer's reviews and in the assessments of ISI
and MOV programs indicated that experienced auditors with good technical I
backgrounds were utilized. However, the team noted that corrective actions on I
findings and timeliness of closeout of findings were not consistent. For example,
three of twelve engineering related findings from the 1994 audit were still open, yet
a sample review of corrective actions on some closed items indicated that
comprehensive follow-up action was taken. 1
A recent SQV audit, QAA-22-96-09, was considered to be good. The audit was
thorough and self critical and identified deficiencies with maintenance rule
implementation, operability assessments, UFSAR discrepancies, adherence to
procedures, and 50.59 SEs. However, Zion's self assessment program had not
identified many of the problems identified by the team.
c. Conclusions
Findings from audits and surveillance conducted on Zion engineering were generally
compliance based, but increased technical findings were noted. Timeliness and
comprehensiveness of corrective action taken was inconsistent. A recent SQV
audit, QAA-22-96-09, was considered to be good; however, Zion's self assessment
program had not identified many of the problems identified by the team.
E8 MISCELLANEOUS ENGINEERING TASKS
E8.1 Technical Specification interoretations (TSil
a. Insoection Scone
The team examined the licensee's TSI process. This included a review of existing
TSis.
b. Observations and Findinas
The licensee had used TSis as an addendum to TSs since 1979. Thirty of the
thirty-six TSis which had been issued were still active when the NRC began this
inspection.
Procedure ZAP 130-02, " Technical Specification interpretations," dated
October 18,1995, stated, in part, that, "A TSI shall not change the TS requirement
or the intent of the applicable TS. The purpose of a TSIis to clarify not modify
TSs." The team and the licensee determined that Zion had generated and could
have used at least thirty TS!s that had potentially changed the TS requirements or
intent. Additionally, ZAP 130-02 required that a 50.59 SE be included in and
36
_ .. _ . _ _ _ ._ -
. _ _ _ _ _ _ . . _ _ . . _ _ _ . _ _ _ _ _ _ _ _ .___
, a
i
!
4
i
i distributed with the on-site review package. No provisions were inc!uded in
- ZAP 130-02 to review previously issued TSis for conformance to the procedure
3
'
requirements. For example, no 50.59 SEs existed for previous TSis, nor were
earlier, still active TSis reviewed to verify that TS intent was not charged. The
- licensee stated that these TSis will either be deleted or revised, 50.59 SEs will be
i performed on all existing TSis, and prompt corrective action will ba taken on all
j identified problems.
!
Subsequent to NRC raised concerns, the licensee conducted a thorough review of
'
i all TSis. This resulted in identification of the following three categories of TSI l
j deficiencies:
1
5'
1. Inaoorooriate Use of TSis to Amend TSs
i.
)' In the first category, a change in the TS was warranted; however, Zion failed to
j change the TSs via an amendment, and instead, h: appropriately attempted to
4 " clarify" the TSs via a TSI. Four TSis were used to interpret TSs that required
j amendments. Three TSis appeared to be more conservative than the TS 1
4 (TSis 89-03, 94-02 and 95-01). One appeared to be less conservative (TSI 95-03).
Zion was pursuing TS amendments for the four affected TS sections.
l
I (a) TS 4.3.1.A.4 b, involving steam generator operability for Modes 1 and 2, was
] determined by the licensee to be incorrect, because it directed use of secondary
i side wide range steam generator water level instead of narrow range in determining
l- reactor coolant loop operability. Use of wide range level was nonconservative, as it
- could result in insufficient inventory in the steam generator to ensure natural :
! circulation, in lieu of correcting the discrepancy via TS amendment, Zion had
2 -
addressed the discrepancy via TSI 89-03 (which was written for TS 3.3.1.A.5).
- Subsequent to the inspection, the licensee was pursuing a TS amendment. The ;
j team also noted that TS 3.3.1.A.5 was insufficient in that it failed to define
! " natural circulation." The licensee planned to revise TSI 89-03 to define " natural
! circulation;" however, it appeared to the team that a TS amendment was needed.
l
!
(b) TS 3.7-1 requires that 20 Main Steam Safety Valves (all) shall be operable
!
whenever the reactor is heated above 350*F, unless the Power Range Neutron Flux
High Trip setpoint is reset for the most restrictive loop in accordance with TS
- Table 3.7-1. TS Table 3.7-1 values for maximum Power Range Neutron Flux High :
- Trip setpoints (when Main Steam Safety Valves are inoperable) were determined to
[ be nonconservative based on Westinghouse and licensee analyses. The existing
i setpoints may allow a secondary overpressurization condition to exist. In lieu of
l amending the TS via TS amendment, Zion had provided the new setpoints via a
j revised table in TSI 94-02.
l
- (c) TS 3.15.2.C footnote allows the common "O" diesel generator to be inoperable for
j up to seven days for maintenance or testing provided that one unit is in Mode 5
) or_6 and three remaining diesel generators are operable. TSI 95-01 extended the
Modes of applicability of the TS from Modes 5 or 6 to Modes 5 or 6 or defueled.
4
i
j 37
.
l
.
d
, , _ _ , ,- - --
_ _ _ _ . . _ ___ ___ _ __ _.__._.____._ . _ _ _ _ _ _ . _ _
- . o
! '
,
i
! (d) TS 3.13.1.A.4 requires that at least one residual heat removal pump and heat
j - exchanger shall be in operation during core alteration operations. The TSI was i
, inconsistent with the TS in that TSI 95-03 allowed suspension of residual heat . l
removal operation for up to one hour in an eight hour period during core alterations
!- with water levels of greater than 22 feet above the top of the reactor pressure
vessel flange. This TSI was less conservative than the TS.
j 2. TSis To Be Revised
! TSis 89-02 and 91-05 changed the intent of the TSs and appeared to be less
l conservative than the TSs.
(a) TS sections 3.7.2 and 4.7.2 involve AFW flowrate requirements for all AFW pump
. and valve lineups while in Modes 1,2, and 3. TSI 89-02 changed the TS intent by
allowing verification of valve position in lieu of testing the "C" motor driven AFW
pump to verify header alignment. This test would be required when AFW piping
headers were divided. Subsequent to NRC queries, the licensee determined that
this portion of the TSI was not needed since existing operating procedures required
that this temt be conducted whenever the AFW headers were manipulated.
(b) TSI 91-05 affected TSs 3.8.7.A, B, and C, which involved service water pump
operability in various combinations, and allowed five hours to be in hot shutdown
for a specific combination of operable / inoperable service water pumps. However,
the TSI was in error, and changed the intent of TS, because it failed to note that
TS 3.8.7.D only allowed four hours to hot shutdown for the same situation. The
TSI was less conservative than the TS and the licensee stated it would be revised.
The licensee's subsequent review determined that the following TSis also needed
revision: TSis 79-04, 89-03, 91-03, 91-04, 93-02, and 95-05. Zion planned to
revise the TSis to provide the needed clarification to the TSs.
3. TSis To Be Deleted
TSis 84-01,86-03 and 92-02 appeared to be less conservative and changed the
intent of the TSs. The licensee stated these will be deleted.
(a) TSI 84-01 was incorrect in that it allowed moving into modes beyond cold
shutdown with an inoperable, but shut, main steam isolation valve. This differed
from the intent and wording of TS 3.9.4, which required all main steam isolation
valves operable unless the reactor was in cold shutdown, and was less
conservative than the TS in that it would allow the station to operate outside the
existing TS.
(b) TSI 86-03, interpreted TSs 3.22.1.A, 3.22.2.A, 4.22.1.A.3 and 4.22.2.A.3, and
changed the TS intent by allowing post snubber failure engineering evaluations to
be fulfilled by visualinspection only. Visualinspection alone may not be
appropriate to determine if the system or component were affected by a snubber
failure. Other evaluations such as pipe stress analysis may be required. This TSI
38
__ __ _ _ __ _ _ __ _. _ _
i o
was considered to be less conservative than the TS, because it allowed potential
circumvention of ASME Section XI requirements.
(c) TSI 92-02 allowed the licensee not to verbatim comply with TS surveillance
4.15.2.E when unit batteries were cross-tied. TS 3.15.1.E required that all 125 !
Vdc batteries and chargers in an operating unit be charged and in service in
Modes 1 and 2. If one battery is inoperable, the proper actions are located in TS 3.15.2.E, which required either to shut down the affected unit (be in cold shutdown )
within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) or cross-tie one battery to a 125 Vdc operable battery of a unit I
that is already in shutdown or refueling mode. TS surveillance 4.15.2.E required '
that charger output voltage and current shall be checked and recorded immediately
and once per shift thereafter until the cold shutdown condition is attained.
However, TSI 92-02 allowed the licensee not to comply with TS surveillance
4.15.2.E, but rather to follow the requirements of TS surveillance 4.15.1.E, which
was less restrictive and required weekly and quarterly battery checks.
Three TSis to be deleted did not appear to affect the TSs conservatism (TSis
79-05, 92-03, and 93-01).
The team informed the licensee that the issues documented in Sections 1(a)
through 1(d), 2(a), (b), and 3(a), (b) and (c) above had modified the TS requirement
or the intent and therefore, the licensee failed to follow the requirements of j
procedure ZAP 130-02 relative to use of TSis. These are considered examples of i
an apparent violation of 10 CFR 50, Appendix B, Criterion V
(295/96011-02G(DRS); 304/96011-02G(DRS)).
4. Concerns With TSI Administrative Controls
!
l The team identified the following concerns. The methodology of controlling the
! copies of TSis was ineffective, since the shift engineer's copy of TSs contained a
- signed TSI which was not listed on the index and was not in other copies. The
} extra TSI of TS 3.17.2.1 defined diesel generator ventilation system operability,
! and thus, diesel generator operability. A cross reference method of red stamping a
'
reference to an applicable TSI on the appropriate TS pages was not complete and
{ several pages were not stamped where the TSI was expected to be consulted.
t
c. Conclusions
k The team determined that the licensee failed to follow the requirements of Zion's
j TSI procedures and failed to maintain sufficient management controls over the TSI
l process. The team concluded that the licensee had issued and used TSis that were
i potentially inconsistent with the TSs.
I
!
!
l
!
1
3
i 39
!
!
'
- . - - . -
_ _ . _ _
,. .
E8.2 Radioloaical Protection Controls for Liauid. Gaseous Radwaste and Area Radiation
Monitorina Systems
a. Insoection Scone
The licensee identified UFSAR discrepancies involving the Radiation Monitoring
(RM) system and longstanding problems with certain of the Process Radiation (PR) -
monitors. As a result, the team performed an inspection to determine if these
problems impacted on the licensee's ability to satisfactorily implement the liquid
and gaseous effluent monitoring and dose assessment program. The team *
reviewed selected PR and Area Radiation (AR) monitor calibrations, portions of the
licensee's liquid and gaseous radwaste program, instrument setpoint methodology
and procedures, TS and offsite dose calculation methodology requirements, PlFs
generated as a result of the identified UFSAR discrepancies, and walked down and
observed PR monitors and skids.
b. Observations and Findinas
Selected calibration and functional tests of some of the TS effluent monitors
indicated they were performed in accordance with requirements and approved
procedures. . However, long term inoperability problems associated with the RM
system continued to persist, causing reliability concerns with the system and
continued need for compensatory actions. Over the past two years, there have r
been several TS and non-TS effluent monitors that were out of service for greater
than 30 days. For example, one of the noble gas decay tank monitors, which has
an isolation function, had been inoperable since January 1996 and had been
routinely inoperable since installation in 1994. The nature of the problems
associated with the RM system included detector, electronic circuit board, flow
switch / indicator and sediment build up problems.
In addition, the team observed that the RMs had failed frequently and were in poor
materiel condition. Operators were required to take extensive compensatory
actions which added additional burden during routine operations, and complicated
transient responses. The licensee had previously recognized the issue, developed a
comprehensive program to correct it, but had made little progress fixing the
problems. )
l
The licensee recently identified over 100 UFSAR discrepancies with the RM system
program. The majority of the discrepancies fellinto the following categories:
e Discrepancies between the UFSAR and plant configuration - 30%
e Discrepancies between the UFSAR and plant procedures -15% ;
e Discrepancies between different sections of the UFSAR -32%
e Other plant documentation errors (e.g., drawing errors) - 10%
40
.
.. .
At the time of the inspection, corrective actions addressing these discrepancies
were being developed,
c. Conclusions
Although there have been longstanding operability problems with certain of the
effluent monitors and re ;ently identified UFSAR discrepancies with the RM system,
the licensee's program to assess dose to the public from radioactive effluents was
effectively implemented. However, a comprehensive plan, which was developed
about two years ago by system engineering to upgrade the RM system had not
been fully implemented.
E8.3 Instrument Classification Uoarade Proaram
The team examined the recently initiated instrumentation classification upgrade
program established to resolve potential discrepancies identified in the safety and
seismic classification of instruments. This effort was considered to be a good
initiative to improve the design bases at Zion. However, it was too early to
determine the effectiveness of its implementation.
V. Manaaement Meetinas
X1 Exit Meetina Summary
The inspection results were presented to members of licensee management at the
exit meeting on August 22,1996. The licensee acknowledged the findings
presented.
The team asked the licensee whether any materials examined during the inspection
should be considered proprietary. No proprietary information was identified.
41
____ ___ ___
_ - - - . -.- .- - -....-=.-~ - - . . . - - _ . _ - _ _ . . - - - . - - - - . . . . .
, . . . l
.
.
l i
!
i
1
l
PARTIAL LIST OF PERSONS CONTACTED
{
! Licensee
J. Mueller, Vice President, Zion Station
. J. Hosmer, Engineering Vice President
! K. Schwartz, Station Manager 1
j B. Giffin, Engineering Manager i
4
J. Reiss,' System Engineering Director
i B. Fitzpatrick, Operations Manager
j G. Vanderheyden, Training Manager
! D. Farrar, Reg. Assurance Manager
M. Wiesneth, Regulatory Assurance Engineer
F. Gogliotti, Design Engineering Supervisor
D. Galanis, Design Engineering Supervisor
M. Burns, System Engineering, Supervisor l
G. Wagner, Engineering Training Coordinator
D. St. Clair, Work Control Manager
L. Peterson, Design Engineer
P. Gazda, Design Engineer
B. Mcdonald, System Engineer
NRC
G. Grant, Director, Division of Reactor Safety, Rlli
M. Ring, Chief, Lead Engineers Branch
L. Miller, Chief, Reactor Projects Branch 6
R. Westberg, Acting Senior Resident
D. Calhoun, Resident
Illinois Department of Nuclear Safetv
J. Yesinowski, Illinois Resident Engineer ,
<
42
. . ---
. . _ . _ _ _ _ _ _ _ . _
- _. _ . _ _ _ _ _ . _. . _ - _ . _ _ _ _ . _ _ . _ _ _ _
,. . o l
1
,
l
,
INSPECTION PROCEDURES USED
IP 37001 10 CFR 50.59 Safety Evaluation Program 1
IP 37550 Engineering
, IP 37551 Onsite Engineering
IP 37700 Design Changes and Modifications
IP 37828 Installation and Testing of Modifications
ITEMS OPENED
50-295/304-96011-01 A thru D VIO 10 CFR 50.59 Safety Evaluations
50-295/304-96011-02 A thru G VIO Procedures,10 CFR 50, Appendix B, Criterion V
50-295/304-96011-03 A and B VIO Testing,10 CFR 50, Appendix B, Criterion XI
50-295/304-96011-04 VIO Operability Status,10 CFR 50, Appendix B,
Criterion XIV l
'
50-295/304-96011-05 URI VETIP Program
50-295-304-96011-06 URI Safety injection Balancing / Testing
50-295-304-96011-07 A and B VIO Corrective Action,10 CFR 50, Appendix B,
Criterion XVI 1
l
DEFINITIONS
Unresolved items (URis)- URis are matters about which more information is required in
order to ascertain whether they are acceptable items, violations, or deviations. (URis i
disclosed during the inspection are discussed in paragraphs E3.2.b and E3.3.b.3.) j
i
l
l
43
).o ,
l
LIST OF ACRONYMS USED
50.59 SE 10 CFR 50.59 Safety Evaluation
ABT Automatic Bus Transfer
AP Auxiliary Power
AR Area Radiation
CVCS Chemical and Volume Control System
DC Direct Current
E&TS Engineering and Technical Support
ECCS Emergency Core Cooling System
ESF Engineering Safeguard Feature
FCR Field Change Request
HVAC Heating, Ventilation and Air Conditioning
ISI in Service inspection
IST in Service Testing
MOV Motor Operated Valve
NED Nuclear Engineering Department
NSR Non-Safety Related
PiF Problem identification Form
PR Process Radiation
RM Radiation Monitoring
SAR Safety Analysis Report
SBLOCA Small Break Loss of Coolant Accident
SE Safety Evaluation
SI Safety injection
SQV Site Quality Verification
SR Safety Related
SSC Systems Structures and Components
TA Temporary Alteration
TS Technical Specification
TSI Technical Specification Interpretation
UFSAR Updated Final Safety Analysis Report
URI Unresolved item
USQ Unreviewed Safety Question
VC Volume Control
VETIP Vendor Equipment Technical Information Program
44