IR 05000346/1989201
| ML19354D674 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 12/19/1989 |
| From: | Isom J, Konklin J, Lanning W Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19354D671 | List: |
| References | |
| 50-346-89-201, GL-87-06, GL-87-6, NUDOCS 8912290045 | |
| Download: ML19354D674 (36) | |
Text
_.
y.
is U.S. NUCLEAR REGULATORY COMMISSION
,
OFFICE OF NUCLEAR REACTOR REGULATION Y
Division of Reactor Inspection and Safeguards Report No:
50-346/89 201 Docket No:
50-346 Licensee:
Toledo Edison Corporation Edison Plaza - Stop 712 300 Madison Avenue
.
'
Toledo, Ohio, 43652 Facility:
Davis-Besse Nuclear Power Station
!
!
Inspection Conducted:
October 30 through November 9, 1989 Team Leader:
_s 5,.,.. Wm u r ~.
abs /PS n
J. ~ At Isom, Operat ns Engineer Dat6 Signed Special Inspect Branch, NRR Team Members:
R. Mendez, Reactor Inspector, Region III D. L. Smith, Human Factors Engineer, NRR A. L. A11aly, Consultant D.-R. Eike, Consultant 0:-l. Gertman, Consultant D. L. Jew, Consultant P. Kohut, Consultant O. R. Meyer, Consultant Other NRC personnel attending the exit meeting:
J. Konklin, R. Cooper, K. Campe Reviewed by:
w [.
/289/99
- TeamInspectionSetionC Date Signed ies E. Konklin, Chief Approved by:
/kdf//f
/pecialInspectionBranch,DRIS,NRR hyne D. Lanning, Ch'ief Date Signed S
l l
I 8912290043 891222 PDR ADOCK 05000346 Q
m
,
,
u
,
.
'
EXECUTIVE SUMMARY The NRC team conducted an interfacing system loss-of coelant (ISLOCA) inspec.
tion at Davis Besse Nuclear Power Station during the period from October 30 through hovember 9, 1989. The objective of the inspection was to collect data and information about plant conditions, including design features, systeros, equipment, procedures and operations which could affect the ISLOCA event. The inspection also included review of aspects of human factors related to the plant design and operations which could affect operator detection, diagnosis and response to an ISLOCA event.
In addition, the team collected information related to hurtan reliability analysis (HRA) for a study being conducted by hkC's Office of Research.
ISLOCA refers to a closs of postulated events in which the high. and low pressure boundary between the reactor coolant system (RCS) and a low-pressure system is breached resulting in a loss of primary coolant outside containment. TheteamIocusedonthedecayheatremoval(DHR)andthe high pressure injection (HFI) systems because of the potential consequences of component failures or operator errors in those systems. The team also reviewed, to a limited extent, the makeup and letdown system and the PCS drains, vents, and sampling systems.
In general, the team did not identify any significant inadequacies in the man-machine interface at Davis-Besse Nuclear Power Station with regard to minimizing the pFobability of an ISLOCA event. The team did not identify any instance in which a single operator error could directly or indirectly initi.
ate an ISLOCA.
The following paragraphs summarize the team's findings and conclusions:
o The team found that the operators were not aware of potential ISLOCA scenarios or indications of ISLOCAs and ISLOCA precursors during conduct of normal plant evolutions.
Interviews with operators and technicians, review of operator training, and review of various plant procedures by the team found that the potential for ISLOCAs had not been addressed by the training department nor incorporated into procedures used by operations, o
The team found that, since there were no formal procedures for responding to computer alarms which could be used to detect ISLOCA events, the computer alarms coulo be ignored for some time before an ISLOCA event was noticed, o
The team's review of the licensee's maintenance history indicated that the licensee had repeated problems in seating one of the four pressure isolation check valves in the DHR system and that there was no preventive maintenance program for any check valves, which is contrary to recom-l mended industry practice.
l
!
=
,
I n
u t
.
.
,
o The team's review of maintenance work orders (MW05) found thet two of the four pressure isolation valve (PlV) cittk v61ses in the DFk system could experience higher than normal failure rates because they were installtd vertictily in the Dhk piping, anc because they were both subject to upstream flow disturbances, lhe team noted that these check valves were recommended by the vendor's manual for horizontal instellation only, o
The team found that the licensee's acceptance criteria in their surveil-lance test for deternining back-leakage rates for two of their DHk interfacing valves was nonconservative in that the acceptance criteria did not incorporate the effect of instrument inaccuracy. Consequently,
the tees.1 concluded thet the f ailure to incorporate instrurent inaccuracy into the test acceptance criteria could le6d to test results greater than the niaximum leakije rate al10wed by the pl6nt lechnical Specifications, o
The team's review of the PlVs found that there were two aeditional valves which should have been classified as PlVs, but were not. Also, it was not clear whether two additional valves referenced in a note in the licensee's response to Generic Letter 87-00 were being considered as Plvs by the licensee, o
lhe team found that none of the motor-operated one manually-operated gate anc globe valves which were classified as PlVs were leek tested to verify their pressure isolation capability. These types of valves constituted approximatelytto percent of all PlVs at Davis Besse.
-.
..
_-_- _______-_ -______
.....
.
..
.
....
......
..
-
.
- %
> x
,e
. >,
<
L TABLE OF CONTENTS
,
INTERFACING SYSTEM LOCA INSPECTION AT DAVIS-BESSE NUCLEAR POWER STATION (Inspection Report 50-346/89-201)
PAGE 1.0 Background..........................................
..........
2.0. Inspection Approach............................................
3.0 System Design..................................................
3.1 Decay Heat Removal and Low-Pressure Injection Systems..........
3.1.1 System Description and Interfacing Configurations..........
.2 3.1.1.2 DHR Discharge Check Isolation Valves...................
3.1.1.2 DHR Suction Gate Isolation Va1ves......................
3.1.2 Conclusions................................................
3.2 High-Pressure Injection System.................................
3. 2-1 System Description and Interfacin
Conclusions......................g Configurations..........
.
- 3. 2. 2
..........................
3.3 Makeup and Purification System..........................-.......
3.3.1 Conclusions..ption and Interfacing Configurations..........
System Descri
3.3.2
.............................................
3.4 RCS Drain, Vent, and Sampling System...........................
3.4.1 System Description and Interfacin
3.4.2, Conclusions......................g Configurations..........
..........................
4.0 Maintenance, Surveillance and Testing..........................
4.1 Maict. ance...................................................
4.1.1 iO;k Valve Maintenance History............................
4.1.2 Installation of DHR Discharge Swing Check Valves...........
- 4.1. 3 Maintenance Procedures.....................................
4.1. 4 Conclusions................................................
4.2 Surveillance and Testing.......................................
4.2.1 Pressure Isolation Valve Identification and Classification.............................................
4.2.2 PIV Surveillance Testing...................................
4.2.3 DHR Suction Valve Pressure Interlocks......................
4.2.4 Conclusions................................................
5.0 Human Factors and Human Reliability............................
5.1 Human Factors..................................................
5.1.1 ISLOCA Scenarios...........................................
5.1.2 Man-Machine Interface......................................
5.1.3 Operating Procedures.......................................
5.1.4 Training...................................................
i
____..
o c.;
.-
'.it 4. --
t
..
.
'
5.1. 5 Control of Plant Configuration.............................
5.1.6 Conclusions.................................................
5.2 Human Reliability..............................................
5.2.1 Operator Task 0emands......................................
5.2.2 Procedures.................................................
5.2.3 Operator Training..........................................
5.2.4 Performance Shaping Factors (PSF)..........................
5.2.4.1 Positive P5Fs..........................................
5.2.4.2 Negative PSFs..........................................
23.
5.2.5 Conclusions................................................
APPENDIX A:
PERSONNEL CONTACTED.................................
A-1
-
APPENDIX B:
DOCUMENTS REVIEWE0..................................
B-1
-
.-.~
ii
'
{
.
,,;
<,
-.
.
1. 0 BACKGROUND
-
The interf acing system LOCA (ISLOCA) refers to a class of postulated loss-of-coolant accidents in which an interf ace between the high pressure bounde.ry of the RCS and connecting low pressure piping is breached. This type of accident is of special concern because the overpressurization of low preswre systems could result in a rupture outside containment, thereby discharging reactor coolant to the environment.
Furthermore, LOCA mitigation
<
systems could be adversely affected by an ISLOCA.
2. 0 INSPECTION APPROACH The primary objective of this inspection was to evaluate specific plant design features, systems, equipment, procedures, operations activities and human actions which could af fect the initiation or progress of an 15LOCA.
This included identification of generic events or system features related to ISLOCAs, potential initiating or precursor events, and human performance and potential errors.
Within the scope of the primary objective, the team also assessed licensee programs relevant to the ISLOCA event and reviewed various licensee records to determine the effectiveness of preventive, corrective, and mitigative mea-The team considered pressure isolation valves (PIVs) to be valves sures.
which isolated the higher pressure RCS from the lower pressure interfacing
'
systems.
The team focused its review on the decay heat removal (DHR) system, and the high presTure injection (HPI) system because of the importance of these two systems to the ISLOCA event scenarios and potential consequences.
- The team also reviewed the makeup and letdown system and the RCS drains, vents, and sampling system.
The ISLOCA accident initiated as a result of the failures of two check valves in series between high pressure and low pressure systems was referred to as Event V in the Reactor Safety Study (WASH-1400).
Event V and other events of this type are heavily dependent on plant features, break locations and miti-gating actions, and are subject to substantial uncertainties.
Thus, inspec-tions by Nuclear Reactor Regulation (NRR), as well as a series of related audits by NRR Of fice of Research, were initiated to collect information on plant features which could affect the frequency and severity of an ISLOCA event.
The first of these ISLOCA inspection was conducted at the Haddam Neck Nuclear Power Station.
The team used the results of that inspection to prepare for the Davis-Besse ISLOCA inspection.
Davis-Besse was the first Babcock & Wilcox (B&W) plant inspected for the ISLOCA event and was the first plant at which the audit for the Office of Research was incorporated into the NRR ISLOCA inspection.
3.0 SYSTEM DESIGN The inspection team (1) reviewed the Final Safety Analysis Report (FSAR),
piping and instrument diagrams (P& ids), and operational procedures; (2) interviewed operators; and (3) walked down systems to collect data and information about the interfacing systems, in order to assess their design
--
_ _......
.
.
.
..
.
---
..
strengths and weaknesses.
The team selected the following systems for inspec-l tion based on their importance and potential consequences to the team's
.
postulated ISLOCA scenarios:
o decay heat removal (DHR) system o
low pressure injection (LPI) system o
high pressure injection (HPI) system o
makeup and letdown purification o
RCS drains, vents and sampling system 3.1 Decay Heat Removal and low-Pressure injection Systems 3.1.1 System Descriptions and Interfacing Configurations The DHR system in the emergency core cooling system (ECCS) mode of operation provides the RCS with heat removal capability in the recircu-lation phase of the postulated LOCA accident.
Following a large LOCA, water is injected from the borated water storage tank (BWST) to the reactor ve.vsel by the DHR pumps functioning in the LPI mode.
In the recirculation phase of the accident recovery, water is pumped from the containment sump by the DHR pump through the DHR coolers and back to the reactor vessel.
An important feature of an ISLOCA event in the DHR system is that it may disable the essential system required to mitigate the consequences of a LOCA, and the water inventory would not be collect-ed in the containment sump.
Depending on the severity of the accident either the injection capability or the recirculation capability may be lost for.atM east one train of the ECCS.
The operation of the other train could also be affected directly (common pipe break location) or indirectly (through environmental conditions) during a sequence of events which could result in eventual core damage.
The DHR system also provides heat removal capability during normal shutdown operations and functions as the auxiliary spray to the pressur-izer for complete depressurization when the reactor coolant pumps are not available.
The DHR system consists of two pumps and coolers and a number of isolation and control valves on the suction and discharge lines which function to isolate the system from the RCS, DHR system suction piping and discharge piping are designed for much lower pressure and temperature.than the RCS, which is designed for 300 psig at 350 F and 450 psig at 350 F, respectively.
During normal power operations, the DHR suction line, which is connected to the hot leg, is isolated from the RCS by two motor-operated gate isolation valves.
The DHR discharge line, which connects to the two core flood lines, is isolated from the RCS by two check valves in each loop.
As a result of an NRC Order issued in 1980, these two pairs of check valves were included in the Davis-Besse's Technical specifications (TS) and have since been tested periodically.
Also, the DHR suction piping is provided with an alternate suction line bypass through two manually-closed gate valves which are used for maintenance purposes, l
f
,,
v
..
3.1.1.1 DHR Discharge Check Isolation Valves
"
The DHR systen is isolated at its interface with the RCS by a pair of normally closed check valves on each of the two DHR injection lines.
Tht;re are no indications, interlocks, or limited control f eatures associ-ated with these interfacing valves because of their self-actuating design. Each pair of valves incluots one swing check, and one stop check valve that could be shut using its manual operator for maintenance purposes. Manual operators for these stop check valves are normally locked open so thdt the valves will function as check valves.
l Although the team considereo the use of check valves as pressure isola-tion valves (PlVs) in the DHR system made this system more vulnerable to an ISLOCA event, the team found that the licensee had implemented other controls to ensure that these Plys were in their proper positions and j
were functioning adequately. The team found that these PlVs were
!
individually leak tested by the licensee af ter each cold shutdown to ensure their pressure isolation integrity. The swing check valves were tested using RCS pressure as the pressure source and core flood tank level change to determine the amount of back-leakage. The stop check valves were checked using the core flood tank (600 psig) as the pressure source, with the back-leakage collected in a bucket. The leak test results of the stop check volves were extrapolated to the RCS normal operating presy re to account for the difference between test pressure and RCS pressure.
The team founo that there was a pressure alarm on the computer in the main control room to warn the operators of any back-leak 6ge across these pressure isolation check valves. The pressure increase in the DHR piping was locally aisplayed at pressure instruments which were monitored every four hours by the auxiliary operators. Also, in the event that leakage
occurred in either the DHR cooler or pump rooms, there were other instru-ments available for leak detection, such as sump level monitors that dlarm on the main Control boersi (MCB) and a number of area radiation monitors in the reactor auxiliary building which would alarm upon radio-activity release, The Davis-Besse DHR system discharge piping was built with small-capacity t
relief valves which relieve to the reactor coolant drain tank (RCDT).
'
The setpoint for the relief valves is 450 psig and, although these relief valves are not designed specifically to eliminate all of the pressure l
increase caused by an ISLOCA event, they will reduce the rate of pressure
l increase. Also, there are MOVs in the DHR systen located on the
injection line outside containment. These valves are always kept oper.
during normal operation and the control power is removed to prevent inadvertent closure beceuse of emergency core cooling considerations, if
l the RCS leakage is sm611, these valves could be closed to mitigate an L
3.1.1.2 DHR Suction Isolation (Gate) Valves
!
The DHR system is isolated on its suction side from the RCS by two l
12-inch motor-operated gate valves. There are two additional 8-inch manually-operated valves which bypess the 12-inch gate volves during
!
l i
i
l
.
-
.
- -
- -
- -
-
-
- -
-
-
-
-
-
-
- -
.
..:
i, -
-.
.
.
maintenance evolutions.
The positions of these MOVs are indicated on the
..-
main control board (MCB) with the standard red and green position lights.
The valve position lights are powered separately from the valve control circuitry and, therefore, positive valve position indications are avail-able to the operators even when the motor operator circuit breakers are opened.
The team found that there were both design and administrative controls to prevent opening the suction valves at power.
For example, valve opera-tion was interlocked with RCS pressure, preventing the inadvertent operation of the isolation valves unless the RCS pressure was less than 305 psig.
Although this interlock was normally bypassed by jumpering the contacts during normal cooldown operation in order to place the DHR system 19 service, the " Plant Shutdown and Cooldown" procedure clearly in.tructed the operator to remove the jumper and to restore the function of the interlock prior to power operations.
In addition to the pressure interlock, the motor control circuit breakers for these valves were kept open during pressurized conditions and a danger tag was placed on the breaker cubicle to prevent any inadvertent local operations or mainte-nance actions.
Also, the control power switches were placed in the "off" position to remove control power from the MOV control circuit.
On the other hand, because the motor operators for the suction valves were sized to open with a large differential pressure across the valves, it appeared to the team that, in the hypothetical event that all inter-locks and c9etrols were defeated or bypassed, it would be possible to open the DHR suction valves during power operations.
The team considered the ability to bypass the suction isolation (gate)
valves by way of the 8-inch suction bypass line to be a design weakness from an 15LOCA perspective because it provided another possible ISLOCA path.
The team also found that neither the gate valves, nor the 8-inch bypass valves, were leak tested to verify their pressure isolation integrity.
However, these valves were locked with chain and padlock devices and included in the licensee's procedures for " Operation and Control of Locked Valves," to ensure closure and to prevent inadvertent operation.
In addition, during plant startup, these valve positions were verified by the " Locked Valve Verification" test procedure.
The team found that the DHR suction piping had a large capacity relief valve downstream of the suction isolation MOVs which discharged to the containment sump.
It was also equipped with several instruments to detect leakages and to warn the operator about the conditions of the system.
The outlet line of the large capacity relief valve was instru-mented with a temperature element that provided a high temperature computer alarm on the MCB computer display.
In addition, there were temperature indicators and an alarm monitoring the DHR pump suction piping temperatere.
3.1.2 Conclusions Although the inspection team identified some design weaknesses, such as use of check valves as PIVs and lack of leak testing on PIVs; in general, the team found that the DHR system was appropriately designed with regard
?
-,
,,
,,
to the prevention and mitigation of the ISLOCA event.
The team found
'
that instrumentation was available at both discharge and suction piping
.
of the DHR system so that the operators could detect leaks past the PIVs in the system, and that relief valves in the system would mitigate the pressure increase within the DHR piping to reduce the probability of a pipe break in the event of an ISLOCA.
3.2 High-Pressure injection System 3.2.1 System Description and Interfacing Configurations
The high pressure injection (HPI) system is designed to inject borated
)
water into the RCS following a LOCA.
It is also used in conjunction with the DHR pumps to provide high head recirculation flow from the contain-ment sump to the reactor core.
The design of the HP1 system during power operation is configured to be in standby with the pumps secured, motor-operated injection stop valves closed, and the suction of the pumps connected to the BWST with a path for minimum flow recirculation provided back to the BWST.
There are four injection lines with two check valves and one closed MOV per injection line.
These two check valves consist of one swing check valve and one stop check valve, with a manual operator to allow for closure of the stop i
In addition, there is a check valve at the discharge of each of the two HPI pumps.
Asafety-inj5ctionsignalstartstheHPIpumpsandopensthe motor-operated injection stop valves in order to provide high head l
injection into the reactor vessel.
Because one of the injections is also used by the makeup and purification system to return purified reactor coolant back into the RCS, the injection check valves are always open and this portion of the piping from the injection check valves to the
,
motor-operated injection stop valve is normally pressurized to approxi-mately RCS pressure.
The HPI pump discharge piping is designed for high pressure service and the licensee's analysis indicated that the discharge piping up to the HPI i
pump was capable of operating at the RCS design pressure of 2500 psig at 650 degrees F.
The injection line check valves and the HPI pump dis-charge check valve function as the two PIVs which isolate the lower HPI j
suction piping from the RCS.
The injection line check valves are consid-ered as one PIV by the licensee because they cannot be individually tested.
Also, the motor-operated injection stop valves are not consid-
'
ered as PIVs because they are cycled at power to verify their stroking times.
Since the HPI pump discharge piping was designed for high pressure i
service, the HPI discharge piping has no relief valve; however, there is a relief valve on the pump suction piping.
Also, there is a pressure
sensor located between the isolation MOV and the HPI pump discharge check valve which operates on a high pressure alarm on a computer to detect i
leakages from the RCS.
Additionally, the ECCS pump rooms are equipped with drain sumps which collect leakage and alarm on high sump level in the control room.
,
.
-
- - -
-
-.
-
.
- -
-
.
-
-
..
..
..
.
- ,-
3.2.2 Conclusions
'
In general, the team identified no significant inadequacies in the HPI system with regard to the ISLOCA accident. The team considered that an ISLOCA accident would be less likely in the HPI system than in the DHR system, because the piping from the RCS to the HPI pumps was designed to the RCS pressure and temperature requirements, and because numerous valves between the RCS and the lower pressure HPl pump suction piping would have to fail before an ISLOCA event occurred in the HPI system.
3.3 Makeup and Purification System 3.3.1 System Description and Interfacing Configurations The primary purpose of the makeup and purification system is to maintain the proper water inventory in the RCS.
It also provides purification of the primary coolant to reduce corrosion and radioactive products.
The letdown flow rate from the RCS is controlled by either a pressure reduc-ing orifice or by an alternate bypass line.
At Davis-Besse, the alter-nate bypass line, which uses a flow control valve to reduce the RCS pressure, is normally in service to allow improved automatic control of the letdown system.
The makeup flow is provided by two high pressure rated pumps, with the discharge flow routed into one of the high pressure injection lines.
The makeup system is separated from the RCS by toe two HPI injection check valves and by two makeup system discharge check valves. - 1:
The makeup pump discharge piping is designed to RCS pressure, and the letdown piping downstream of the pressure reducing orifice is designed for low pressure service of 150 psig.
The low pressure piping is pro-tected against overpressurization by a large-capacity relief valve which opens at the low pressure piping design pressure and which relieves to the reactor coolant drain tank (RCDT).
The flow orifice or the flow control valve reduces the RCS pressure to prevent the low pressure piping in the letdown portion of the makeup and purification system from becoming overpressurized.
The orifice can also be isolated by an MOV.
In addition, there are four other isolation valves installed on the high pressure section of the letdown system which shut automatically with a safety features actuation signal (SFAS). These four isolation valves have indications on the MCB and are remotely controlled from the MCB.
The team concluded that the operators would have ample warning of a potential ISLOCA event that results from failure of the pressure-reducing elements.
The operators would be alerted to high pressure conditions through temperature, pressure and flow indications.
High pressure in the letdown system would immediately be noticed because of the alarm on the annunciator panel.
3.3.2 Conclusions In general, the team identified no significant inadequacies in the design of the makeup and purification system with regard to the ISLOCA event.
The team considered that ISLOCA scenarios for the makeup and purification
- _ - _ _ _
__- -
-
-
- _ _ _ _ _ _ _
..
~,..
..
.
.
k
'
system were both less likely and less severe in terms of consequences than the scenarios for the OHR or the HPI systems.
In the event of an 15LOCA in the letdown section of the makeup and purification system, the team considered that the licensee's isolation capability outside of containment could be impaired, but that there were four isolation valves which automatically shut on SFAS, of which three were located inside containment so that the operators could terminate the 15LOCA event.
3.4 RCS Drain, Vent and Sampling System 3.4.1 System Description and Interface Configurations The RCS interfaces with low pressure piping through a number of small lines equipped with isolation valves.
These lines are primarily used for draining the RCS and its components, venting the system during the RCS filling, providing nitrogen supply during cooldown operations, and taking samples out of the pressurizer.
Most of the isolation valves in these lines are either 1-inch or 2.5-inches in diameter.
The piping from these various interfacing lines is all routed into common headers, some of which are located outside containment.
The team found that, in general, the potential break locations for the low pressure piping for this system were inside containment or, for the lines in the drain system, outside containment near the RCDT.
The team found that all of these~.syO. ems were isolated from the RCS with at least two valves and that, becauke of the piping size, the break would result in a small LOCA.
In addition, the team concluded that operation of the ECCS equipment should not be af fected severely by such a break, and that the use of the HPI and/or makeup pumps should be sufficient to mitigate the small LOCA.
The team also considered that the break outside containment could be detected through the radiation monitors in the auxiliary building.
3.4.2 Conclusions The team identified no significant inadequecies in the design of the RCS drain, vent and sampling systems with regard to the 15LOCA event.
The team considered that a break in the RCS drain, vent and sampling system would result in such a small LOCA that the postulated loss of coolant would be within the capacity of the ECCS system.
4.0 MAINTENANCE, SURVEILLANCE AND TESTING The team reviewed the licensee's maintenance and surveillance programs to determine the present and past performance of the pressure isolation valves.
In the maintenance area, the team reviewed maintenance procedures, completed work packages, and vendor manuals in order to determine the effectiveness of the licensee's corrective and preventive maintenance being performed on the PlVs.
In the surveillance and test area, the team reviewed the licensee's surveillance program to ensure that all valves which functioned as PlVs were identified and tested adequately.
Also, the team reviewed completed surveil-lance test results to ensure teat the tests were performed at the required intervals and that the results were satisfactory.
The operation of interlocks associated with MOVs in the DHR system was examined to determine the adequacy of the interlocks to prevent an ISLOCA event.
l
e
-
-.e
.e
,,
.
k
..
4.1 Maintenance
,
4.1.1 Check Valve Maintenance History Check valve degradation and failures have been identified and documented by numerous incustry and NRC reports 6Ld generic communications. A recent industry study indicates that the major causes of check valve failures are misapplication of the valves and inadequate preventive maintenance. Because check valves constitute approximately 70 percent of all the PlVs in the DHR and HPl systems at Davis-Besse, the inspection team reviewed numerous maintenance work orders and the valve maintenance histories, to determine whether the licensee had experienced significant check valve problems.
The review by the team indicated that Davis-Besse had numerous check valve failures in both the DHR and HPl systems over the course of its operating history.
For example, valve CF-30, a swing check valve in one of the two DHR injection lines, had been found in 1980 to have excessive back-leakage.
The licensee found that the excessive back-leakage was the result of gross check valve failure. The licensee's internal examination of the check valve found that the check valve disc had fallen into the pipe; the disc's mounting bolts had come loose through excessive vibration which had occurred during extended DHR operation. To prevent this type of problem from occurring again, the licensee installed locking brackets on the disc mopeting bolts. The team's review of the maintenance history of CF-30 indicated that since the locking brackets were installed, this problem has not happened again.
However,-in December of 1986, and again in March and May of 1987, the licensee experienced additional problems with swing check valve CF-30.
As a result, the licensee issued four maintenance work orders to seat this check valve. The team's review of the licensee's )rintout of equipment maintenance history showed that some 584 man-1ours were expended during this time period in an effort to seat check valve CF-30 by various methods, including use of pneumatically-operated mechanical vibrators, dead blow hammers, and a tamper made from an oak, block.
The team found that, even though the licensee had a difficult time seating CF-30, they were eventually successful.
Following are some of the descriptions of the work performed on CF-30, as documented in the
-
maintenance work orders:
o
"12/05/86 Vibrated the valve, CF-30, with dead blow hammer attempting to seat the disc. We were not successful."
o
"12/06/86 The body of the valve was covered with 1/2 inch sheets of lead. A brass plate was affixed to the body of the valve with a
+
come-along and straps. A tamper with a brass plate affixed to the end of it was used to vibrate the valve in an effort to get it to seat.
We were not successful."
,
_ _ _ _ _ _ _ _ - _ _ - _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ __
_-
F
1
-
..
..
<w
.
,
'
o
"12/7/86 Brackets were made up for seven mechanical vibrators to be attached to the piping at CF-30. An air manifold and lines were set up.
Installed the brackets on the CF-30 piping and moved associated equipment in place."
o
"12/07/86 Under the direction of eng. dept. we used air tamper and vibrators to seat the valve. After repeated attempts, we were successful'. There is no leakage on the valve at this time."
o
"12/20/86 Connected hard piping to drain valve behind CF-30 for CF-30's tell tail. Hit CF-30 with dead blow hansners to try and seat the valve.
No luck."
o
"12/21/86 Hooked up vibrators to CF-30 and shook the tar out of it.
Used tamper with oak block to massage valve CF-30.
Finally, seated at 0500."
o
"03/17/87 There are 6 vibrators hooked up to the valve. We went down and ran the vibrators and used the thumper and was unsuccessful in closing CF-30. ops is performing S.T. - Started at 6:00 AM."
o
"03/17/87 Operator opened air line to the six vibrators at approxi-mately 0655. ops then tried to close CF-30. After numerous attempts the air was turned off and CF-30 was vibrated manually. The valve attempgd to close but always returned to the open position."
o
"03/17/87 Started all 6 vibrators full throttle.
Also started using ~
the thumper and hammers.
Valve closed and ops started to perform their ST. Valve closed approximately 9:15 AM."
Finally, the licensee disassembled and repaired check valve CF-30 as a result of a surveillance test performed in May of 1987,-which found that the check valve had a leakage rate in excess of 5.0 gallons per minute (gpm). The licensee found that check valve CF-30 had significant wear both at the disc-to-hanger arm joint and on its anti-rotation lugs.
The valve internals were replaced with a new disc, hinge pins and larger anti-rotation lugs; these modifications were completed on June 12, 1987.
The team noted that the licensee has not experienced leakage problems with check valve CF-30 since its modification in 1987.
Because of the problems experienced with check valve CF-30, the licensee had made repairs and modifications to check valve CF-31, a similar check valve in the other DHR injection loop during the fifth refueling outage.
The team found that check valve CF-31 had not experienced the same leakage rate problems that valve CF-30 had demonstrated.
The team also found that similar mechanical agitation technique used for the DHR injection loop stop-check valves. These stop-check valves were the second PIVs, upstream of swing check valves CF-30 and 31 in the DHR injection loops. The team was informed by the licensee that the mechani-cal agitation method referenced in the plant startup procedure for the stop-check valves was the use of a tamper and an oak block.
i
._.
.
. -
.
..
-
.-.. - -
.
.
. _.
- -.
...
,,
.
.
The team found that the licensee had experienced problems with the HPI
in.lection swing check valves, anti-rotation lugs in 1985.
As a result of IE Information Notice 83-06,-issued February 24, 1983, concerning failure of Velen check valves to seat because of binding between the anti-rotation lugs on the disc and the swing arm, and because of other leakage problems with Velan swing check valves the licensee modified the anti-rotation lugs on the smaller (21-inch and 3-inch) swing check valves in the HPl system.
The nodifications to the HPl injection swing check valves and the HPl pump discharge check valves were completed in January 1985. The licensee did not modify the Velan check valves (CF-30 and CF-31) in the DHR system because they believed that the larger anti-rotation lugs found in the DHR check valves would make this type of failure highly unlikely for the DHR check valves.
The team also found that seating failures had been experienced with one of the stop-check valves, DH-77. On November 23, 1988, when the nunter one core flood isolation valve opened for a surveillance test of the core flood tank discharge check valve, about 500 gallons from the core flood tank were lost because the stop-check valve, DH-77, was lea king. During this incident, decay heat piping relief valve PSV 1509 lif ted to prevent overpressurization of the DHR system.
As a result of the problems with check valves identified by the team's review of MW0s and maintenance history, the team reviewed the licensee's preventive maintenance (PM) program for check valves which were classi-fied as PlV6 The team found that, contrary to recommended industry practice, the licensee did not have an established PM program for Piv
-
.
check valves, or for check valves in general.
Discussion with the licensee's staff indicated that a PM program for check valves was being developed and had not been implemented because the licensee was still in the process of reviewing past check valve failures in order to determine
.
which check valves should be included in the PM program.
!
4.1.2 Installation of Check Valves CF-30 and 31 in the DHR System The team's review of the licensee's check valve reliability study entitled " Check Valve Reliability Program" found that check valves CF-30, CF-31, and HP-59 were installed vertically in the piping although these
,
l valves were reconsnended by the vendor's manual for horizontal installation only.
The check valve reliability program study was
'
completed as a response to an industry program for improving check valve reliability.
The industr of valves in the system (y efforts had identified incorrect orientation e.g., vertical installation versus horizontal L
installation) as a frequent cause for check valve failure.
The team also
'
noted that one additional pressure isolation swing check valve, HP-59, was identified as being vertically mounted.
l The team's review of the licensee's check valve reliability study found that both CF-30 and CF-31 were subject to upstream flow disturbances.
CF-30 was one pipe diameter and CF-31 was six pipe diameters from their respective piping elbows, which caused the flow disturbances. The l
i j
l
.
n
,
. ~. '
...
<
e
- a
.,
industry-study had also'found that check valves which were located in
areas of high turbulence were more susceptible to failures, The reli-ability study also recommended that check valves HP-59 and CF-31 be disassembled for a detailed inspection during the licensee's fifth refueling outage in 1987, and that a long-term inspection plan be devel-oped following issuance of the check valve application guide by the check valve task force in the fall of 1987.
Although the check valve applice-tion guide had not been completed at the time of this inspection, the j
team found that the licensee had performed inspections of CF-31 and HP-59
-
as recommended by the reliability study.
]
Although the licensee believed that the root cause failure of CF-30 was the prolonged DHR operation during a period of 14 months in 1985 and 1986, the team concluded, through interviews with the plant engineering i
. staff,-that it vould be possible for CF-30 to undergo a similar type of wear and failure after a number of shorter shutdown periods which would be equivalent to approximately one 14 month outage.
4.1.3 Maintenance Procedures The team reviewed the licensee's corrective maintenance procedures for check valves and identified no significant deficiencies.
Two procedures,
'
l MP 1411.08, "Velan Bolted Bonnet Check Valves Disassembly Repair and i
Reassembly," Revision 00, and MP 1701.99, "Velan Forged Glove and Stop i
Check Valve Maintenance," Revision 00, provided instructions for the'
l disassemblyr inspection and reassembly of Velan swing and stop-check
'
valves.
These procedures not only implemented the vendor's recommenda -
tions for repair of'the valves, but also specified requirements for measuring and test equipment, special tools, and replacement parts.
The I
procedures detailed steps for disassembly and reassembly of the bonnet cover, valve disc, hanger brackets and bushings, as well as requirements for inspection of the check valve internals.
In addition, the procedures clearly specified post maintenance testing acceptance criteria.
l The team also reviewed the preventive maintenance procedures for l
motor-operated valves (MOVs) and identified no significant deficiencies in these procedures.
Tne team considered that the licensee's PM proce-dures, DB-ME-09301, "Prevontative Maintenance for Type SMB and SMC Limitorque Valve Operators," and DB-MM-09587.01, " Maintenance and Repair l
of Limitorque Valve Operators Types SMB-00 and SMB-000," were detailed, l
technically correct and generally well written.
Those procedures re-
~
quired internal and external visual inspections for loose, missing or broken hardware, and moisture or dirt on the actuator.
The procedures also required cleaning and lubrication of the valve stem, and inspection of the wiring terminals, contacts, torque switch block, and rotor assem-l blies for cracks and signs of excessive wear.
The team also noted that the licensee had a program for analyzing the valve grease and for sched-
'
uling MOVATS testing.
l-l 4.1.4 Conclusions The team considered the licensee's corrective maintenance procedures for check valves and the preventive maintenance procedures for MOVs to be a strength.
The team considered the procedures to be well written, de-tailed and comprehensive.
However, the team was concerned with the
- -
- -
-
-
--
-
- - -
-
- -
.. _ _ _ _ _ _ _
g
..
..
-4
-.
l repeated maintenance problems experienced with check valve CF-30 and the
'
lack of a PM program for PlV check valves at Davis-Besse.
In addition, the team considered that, based on industry experience, the vertical orientation and location in turbulent flow areas of some check valves may have contributed to their failures.
4.2 Surveillance and Testing 4.2.1 Pressure Isolation Valve Identification and Classificatior The team reviewed the licensee's response to Generic Letter (GL) 87-06 and the piping and instrumentation diagrams (P&lDs) fcir the RCS, DHR, HPl ano biakeup and purification systems to ensure that the licensee had identified two valves in each system as PIVs at all high-pressure to low-pressure interfaces.
The valves which were identified by the licensee as PlVs are listed in Table 4.2.1.
Because PlVs should be
<
individually tested, the team considered the back-to-back check valves in the HPl injection lines, which were welded together, to be only one PIV.
Therefore, the team considered the high-pressure to low-pressure boundary to be located further up the HPl injection piping. As a result, the team considered that the HPl discharge / recirculation cross-connect valves, which isolated the HPl recirculation line from the borated water storage tank (BWST), and the HPl pump discharge check valves, to be PlVs.
In addition, the team's review of the licensee's response to GL 87-06 found that the two HPl discharge check valves were not included with the other PlVs. AlthdITgh these valves were tested to determine their back-le6kage and were 'influded in a note at the end of Table 11 in GL 87-06, since these valves were considered to be PlVs by both the team and the licensee, the team concluded that they should be placed in the table and removed from the note. This is considered to be an open item, pending further NRC review (50-346/89-201-01).
The team also found that none of the motor-operated and manually operated gate and globe valves that were classified as PlVs were leak tested to verify their pressure isolation capability. These type of valves consti-tute approximately 60 percent of all PlVs at Davis-Besse.
4.2.2 PlV Surveillance Testing The team's review of Davis-Besse surveillance procedures associated with leak testing of PIV check valves found that, in general, the procedures were well written, concise, and technically adequate in determining the leakoge rates for the PlVs. The team's review included verification of proper valve lineup for the tests, use of appropriate instrumentation to measure test results, and the establishment of proper test conditions and acceptance values. However, the team found that the inaccuracy associated with instrument use in the "RCS isolation Check Valve Leak Test" procedure was high in that it was slightly greater than half of the surveillance test acceptance value.
Leakage rates for the DHR swing check valves (CF-30 and CF-31) were calculated by measuring the change in the core flood tank (CFT) level.
Using the maintenance test line and the RCS as the pressure medium, the CFT level increase was recorded off the computer in the control room.
.
I-
-
.,
..
.
.
Af ter the quantity of leakage was known, it was divided by the test
,
duration of one-half hcur to obtain the leakage rate. The inspection team requested that the licensee provide information regarding the accuracy of the instrument used in the test because the CFT level dis-played in the plant computer implied that instrument accuracy was one one hundredth of a foot (.01 foot). However, the team's review of the licensee's calculation showed that the accuracy was plus or minus
.0336 feet. The instrument accuracy of.0336 feet implied that the flow rate calculation was only accurate to plus or minus.568 gpm for the test duration. Therefore, the team concluded that, in the event the licensee obtained a test result greater than.432 gpm, the actual value could potentially be above the surveillance acceptance limit of 1 gpm. Simi-larly, the team noted that the inaccuracy associated with this test could lead to test results greater than 5 gpm.
Five gpm is the maximum rate allowed by Technical Specifications (TS) before the licensee must enter a limiting condition for operation, (LCO) as specified by the NRC Order of April 20,1981.(EventVorder). This is considered to be an open item, pending further NRC review (50-346/89-201-02).
4.2.3 DHR Suction Yalves' Pressure Interlocks The team reviewed calibration procedures DB-MI-03134 R0, " Channel Cali-bration of 64-B-ISPRC02A3 Reactor Coolant Loop 2 Hot Leg Wide Range Pressure SFAS CH 4," and IC-2701.42 R1, " Calibration of Static-0-Ring Pressure Switch," used to calibrate interlocks associated with DH-11 and 12, respect 6ely,andfoundtheprocedurestobeadequate. DH-11 and 12 are gate valves which are in-series in the DHR suction piping. The pressure switches for the DHR suction valves provide the pressure inter-lock which prevents inadvertent opening of these valves when the reactor coolant pressure is above 301 psig for DH-11 and 266 psig for DH-12.
The team found the calibration procedures to be detailed, comprehensive and generally well-written.
In addition, the calibration records for both DHR valves showed the as-found pressure interlock values to be close to the desired setpoint values.
The team also examined the interlocks which would prevent opening the DHR suction valves and identified no significant inade Normally, these valves cannot be opened at power because (1)quacies.
their circuit breakers off, and (3) pen with a red "Do Not Operate" tag, (2) control power is are tagged o pressure interlocks would not be satisfied.
4.2.4 Conclusions Although the team identified some specific problems with regard to identification of PIVs and the measurement of leakage rates for DHR swing check valves, the team generally concluded that the surveillances per-formed on PlVs were adequate.
The team's review of the calibration procedures used for the calibration of pressure instruments used to prevent inadvertent opening of DHR suction valves found that the proce-dures were comprehensive and well-written, and that proper operation of DHR suction valves' interlocks minimized the potential for ISLOCA in the DHR system.
~
,.
.,,
..
9 TABLE 4.2.1
,
Valve Description Valve IST Leak Frequency Notes Type Q Tested CF-28 CF Tank 1-2 to RX Check Check A/C yes 2 years
CF-30 CF Tank 1-2 to RX Check Check A/C yes CS>9 mos 1,2 DH-76
- 2 DH Line Stop Check Check A/C yes CS>9 mos 1,2 CF-29 CF Tank 1-1 tc RX Check Check A/C yes 2 years
CF-31 CF Tank 1-1 to RX Check Check A/C yes CS>9 mos 1,2 DH-77
- 1 DH Line Stop Check Check A/C yes CS>9 mos 1,2 HP-49 HPI Line 1-2 Stop Check Check A/C yes CS>3 mos 2 HP-51 HPI Line 1-2 Check Check A/C yes CS>3 mos 2 HP-48 HPI Line 1-1 Stop Check Check A/C yes CS>3 mos 2 HP-50 HPI Line 1-1 Check Check A/C yes CS>3 mos 2 HP-56 HPI Line 2-2 Stop Check Check A/C yes CS>3 mos 2 HP-58 HPI Line 2-2 Check Check A/C yes CS>3 mos 2 HP-57 HPI Line 2-1 Stop Check Check A/C yes CS>3 mos 2 i
,
HP-59 HPI Line 2-1 Check Check A/C yes CS>3 mos 2
'
DH-11 RC System to DH Isol.
MOV B
no 3,4
DH-12 RC System to DH Isol.
MOV B
no 3,4 i
DH-21 RC System to DH Bypass Manual B
no
3,4 j
DH-23 RC System to DH Bypass Manual A
no 2 years 3,4,5 DH-49 RCS to Thermal Expan.
Check C
yes 2 years
'
RC-51 DH Auxt Spray Check Check C
yes 2 years
i DH-2735 DH Aux. Spray Stop MOV A no 2 years 3,5 RC-33 Loop 2-1 CL Dr. Stop Manual 3,4
-
no
RC-29 Loop 2-1 CL Dr. Isol.
Manual no
3,4
-
RC-30 SG 1-2 Drain 1501.
Manual no 3,4
-
RC-31 SG 1-2 Drain Stop Manual no
3,4 i
-
RC-34 Loop 2-2 CL Dr. Isol.
Manual 3,4 no
-
RC-35 Loop 2-2 CL Dr. Stop Manual no 3,4
-
RC-36 SG 1-1 Drain Isol.
Manual no
3,4
-
RC-37 SG 1-1 Drain Stop Manual no
3,4
-
RC-38 Loop 1-1 CL Dr. Isol.
Manual no 3,4
-
RC-39 Loop 1-1 CL Dr. Stop Manual no 3,4
-
RC-40 Loop 1-2 CL Dr. Isol.
Manual no
3,4
-
RC-41 Loop 1-2 CL Dr. Stop Manual no 3,4
-
RC-55 Surge Line Dr. Isol.
Manual no
3,4
-
RC-56 Surge Line Dr. Stop Manual no 3,4
-
RC-5004 Loop 1 HL Vent Isol.
Manual
,
3,4
-
no
I RC-43 Loop 1 HL Vent Stop Manual no 3,4
-
NN-69 Loop 1 HL N2 Isol.
Manual no 3,4
-
RC-5003 Loop 1 HL N2 Stop Manual 3,4 no
-
RC-5001 Loop 2 HL Vent Isol.
Manual no 3,4
-
'
RC-45 Loop 2 HL Vent Stop Manual no 3,4
-
NN-68 Loop 2 HL N2 Isol.
Manual
-
no 3,4
RC-5000 Loop 2 HL N2 Stop Manual no 3,4
-
RC-239A Press. Vapor Sample MOV no
3,4
-
RC-239B Press. Liquid Sample MOV no
3,4
-
RC-240A Press. Sample Isol.
MOV A
no 2 years 3,4,5 RC-200 Press. Vent Stop MOV no 3,4
-
NN-64 N2 to Press. Isol.
Manual 3,4 no
-
!
-
-
-
- -
--
-
_ _ _ _
- -... -....
.
.. " -
.0
.
-
Notes:
1)-
These valves are considered PlVs per Technical Specifications
.
paragraph 4.4.6. 2.1, Table 3.4-2.
2)
Valves which were listed in Table 1 of the Licensee's response to Generic Letter 87-06, dated June 30, 1987.
These were either Technical Specification PlVs or were identified as PIVs
,
in the SER for the revised valve IST Program.
.
3)
Valves which were listed in Table 11 of the Licensee's response to Generic Letter 87-06, dated June 30, 1987.
These were identified as P]Vs by the Licensee per the definition in Generic Letter 87-06.
4)
Motor-operated and manually-operated' gate and globe valves which were classified as PIVs per the Generic Letter 87-06 response, and which were not subjected to any formal and documented leakage test.
5)
Classified as a Containment Isolation Valve and subjected to an Appendix J, type C, leakage test.
This test utilizes contain-ment atmospheric conditions at 38 psig.
?
!
._
,
.;, "
n a
..
5. 0 HllMAN FACTORS / HUMAN REllABILITY
,
With regard to the human factors area, the team focused on the man-machine interface (MM1), procedures and training, and the potential for operator error, for several specific 15LOCA scenarios.
The team's primary focus in the MMI assessment involved various aspects of the main control room, such as control panel layout, control-display integration,
!
,
and control and display design, which could influence operator performance during an 15LOCA event.
Within the MMI assessment, special emphasis was
!
placed on the design of controls to minimize greater error and of an 15LOCA.
i In addition to the main control room, the MMI assessment examined local
!
control of selected manual valves outside containment.
The methodology employed during this portion of the inspection involved direct observation of j
equipment and facilities and interviews and discussions with operations i
personnel.
l The inspection team also reviewed surveillance test and operating procedures relative to postulated 15LOCA scenarios.
The procedures were reviewed to determine the extent to which the procedures could &ffect the initiation of an 15LOCA event.
The team reviewed selected surveillance test and operating j
procedures with operations personnel.
!
5.1 Human Factors i
5.1.3 ISLOCA Scenarios The team evaluated the following 15LOCA scenarios in order to identify potential operator error as an 15LOCA initiator.
The objective in developing these scenarios was to describe a credible series of plant
,
evolutions which involved specific operator decision-action sequences
.
'
'
that could contribute to the course and outcome of an ISLOCA event.
The first scenario assumed that the in-series MOVs on the DHR suction
'ine would be inadvertently opened, with a subsequent loss of the RCS to
.
the DHR/LPI pressure boundary and eventual overpressurization of the low pressure suction piping upstream of the DHR pump suction valves.
The scenario postulated that the pressure within the DHR system would in-crease until the setpoint (320 psig) for the relief valve (PSV 4849) was reached, at which time the relief valve would lift and begin venting to the containment emergency sump.
The second scenario involved mispositioned manual valves in the HPI
!
recirculation line with a back-to-back check valve failure in the HPI injection line when'the HPI discharge MOV was opened.
The team did not
'
consider opening of the HPI discharge MOV to be an operator error because this MOV was periodically opened for the licensee's quarterly surveil-lance test to verify the MOV stroke time values.
The team also performed a very limited review of a scenario in which one of the two sets of in-series pressure isolation valves in the DHR/LPI system (CF-30/DH-76 cr CF-31/DH-77) failed with an subsequent loss of the RCS to the DHR/LPI pressure boundary.
- .
V, m - w -
..
.
.
5.1.2 Man-Machine Interface in generol, the team identifled no significant inadequacies with the MMI at Davis-Besse with regard to minimiring the prcbability of an operator error initiating an ISLOCA event. The principal determinant of the probability was the fact that most of the PlVs associated with the ISLOCA scenarios were check valves and, therefore, required no operator action and were not subject to mispositioning from the control room.
Although the licensee had alarms in the DHR/LPl and HPl discharge lints tu warn operators of a potential ISLOCA event, the team found-that the alarms were not safety-grade e d thet the operators did not appear to be sensitive to individual computer alarms because of the nuniber of computer alarms typically received during each shif t.
The team found that there were no procecures for responding to computer alarms. Operetors stated that they routinely monitored computer alarms and were cognizant of response requirements; however, the team's review of computer alarm printouts maintained by records management revealed that, in at least one inste.nce, the HF1 system was in 6 high-pressure alarm condition for more than two hours before being cleared. The team considered that, based on that occurrence, the computer alarms 1or detection of au ISLOCA event might be ignored for some time before the ISLOCA event wub noticed.
The team reviewed the design features associated with the DHR suction valves to mintsize the likelihood of an inadvertent ISLOCA initiation by an operator. 'Under normal power operation, these valves are closed with their breakers opened and the control power removed. The team found that control power was removed by pushbuttons on the main control board as indicated by blue "CNTL PWR OFF" legeno lights integrated with the pushbuttons.
The team found that, even though there were no formal procedures for responding to computer alarms, some information regarding critical plant parameters which could warn the operators of an DHk/LPI ISLOCA was available in the control room. The team noted that overpressurization of the LHR system could be detected by a computer alarm from a thermocouple at the outlet of the DHR relief valve and, similarly, there were pressure sensors on all four of the HP1 discharge lines which would initiate a computer alarm for a high-pressure condition in the HPI discharge piping.
In addition, the controls ano indicator lights for the sump pumps were located on the main control board, such that sump pump startup would probably be noticed quickly by an operator. The team found that the controls for the HPl MOVs, which were integrated into the HPI mimic board, were clearly labelled and that operators were aware of the normally closea position of the valves.
5.1.3 Operoting Procedures The team's review of operating procedures reveolto that there was no procedure which directly addressed the ISLOCA event. Although the operations and training personnel stoted that they felt that the "Small
-..
..m>,,_
.
,
RCS Leak" procedure would accommodate the conditions of the scenarios,
'
i the team found that this procedure did not contain any specific references to ISLOCA. Also, the licensee's emergency procedure titled,
"RFS,. SFA, SFRCS Trip, or SG Tube Rupture," did not contain any referentcs to ISLOCA.
The team identified two errors in the licensee's procedures which could potentially have an impact on the ISLOCA scenarios.
For example, the team found that valve HP 1556, which was one of three valves required to be open in order to vent the HPl piping that had become pressurized during the surveillances, was on.itted f rom the procedure. Consequently, the HPI discharge high-pressure alarm could be in the alarmed condition for a prolonged period of time resulting in reduced operator sensitivity to an actual ISLOCA event.
Secondly, the team noted that the plant startup procedure did not Specify removal of control power from the DHR suction valves. The team consid-ered this omission from the startup procedure to be important because, in adoition to the MOV breakers which were opened in the procedure, removal of control power provided a second layer of protection against an ISLOCA by preventing inadvertent operation of the DHR suction valves. This is considered to be an open item, pending further NRC review (50-346/89-201-03).
5.1.4 Training r
The team found that the licensee's training program did not directly address the ISLOCA event, in that there were no training objectives related to ISLOCA, and operators did not receive specific training on detecting, diagnosing or mitigating an ISLOCA event. The team's interviews with operations personnel verified that the operators had not received simulator training on ISLOCA events or training on the RCS small leak procedure as it might relate to managing and recovering from an ISLOCA.
One example of the lack of attention to the ISLOCA event was the licensee's procedure titled "ECCS Valves Train 2 Quarterly Test," which appeared to dismiss the possibility of an ISLOCA event when the operators received a HPI discharge high-pressure alarm.
In addition, the team found, through interviews with the operctors, that the operators were not fully aware of the potential for an ISLOCA event at Davis-Besse.
5.1.5 Control of Plant Configuration The team identified two cases in which the as-found conditions of the plant were not consistent with the controlled P&lDs and with normal valve lineups contained in procedures. Although none of the identified valve inconsistencies involved PlVs, the fact that prevention of an ISLOCA relies partly on properly positioned and controlled manual valves made these inconsistencies a concern.
_
_ _.....,...
-
i,..,n
,
The team found inconsistent references to the position of HP 1556, which is one of the three valves used to vent the HPI piping.
During review of
-
plant documentation and discussions with operations personnel, it was noted that the position of this' valve, and the type of valve, was incon-sistently described.
The team's repeated attempts to determine the position for the valve were unsuccessful.
Direct observation of the local handwheel and valve stem were necessary to determine the actual position of the va'Ive.
The team found that various licensee documents referred to this valve as either a normally closed air-operated valve; or closed, but not locked; or locked closed.
The team eventually determined that HP 1556 was a normally closed, manually-operated valve with its air-operator removed.
The team's interviews with the operators found that their knowledge of the valve was just as inconsistent.
At various points in the inspec-tion, operations and maintenance personnel reported that the HP 1556 valve status was mechanically " gutted," with internal parts removed (i.e., normally open); normally closed, cir-operated valve; f ailed open, with air removed; locked closed, with air removed; or normally closed with air power removed (the as-found condition).
The team also found that the local control station for this valve was
- intact, with no tags or labels to indicate that it was inoperable.
Secondly, the team found that the plant P&lDs and valve lineups contained in the plan Pstartup procedure (DB-0P-06901, Revision 0, October 10,72988) indicate that the reactor drain valve, RC 1773A, and reactor vent valve, RC 1719A, were normally open.
The as-found position for these valves was closed, and therefore did not appear to increase the probability of an 15LOCA event.
Operators were unable to provide a rationale for these valves being in an position other than what was indicated in the P&lD.
5.1.6 Conclusions Although the team did not identify any instance in which a single opera-tor error could directly or indirectly initiate an 15LOCA, the team did identify problems with procedural inconsistency, operator insensitivity to compuer alarms, and surveillance and operations procedures which appe. red to have omitted required valve and control power switch opera-tion, lhe team also found that there was little operator training for the 15LOCA event, and that operators had limited knowledge regarding the.
initiation and mitigation of the 15LOCA event.
Also, the team found a lack of consistency between plant reference documents and the as-found conditions, and an inability by the operator to explain the condition.
5.2 Human Reliability The team's human reliability evaluation at Davis-Besse focused on obtain-ing information on operator tasks and performance shaping factors (PSFs)
from which quantitative human error estimates could be derived later.
The team's review was restricted primarily to human actions which could be considered as either precursors or initiating events.
The evaluation of PSFs focustd on conditions surrounding operations, maintenance, surveillance and inservice test actions carried out by plant personnel.
l
-
- _
_
_ - _ _ - -
- -
- - - - - - -
f v
!
q., w he
.
Human reliability analysis (HRA) was instituted in order to place human
performance for a number of actions in perspective from a probabilistic viewpoint.
These human actions may be categorized as either (1) precursors. (2) initiating events, (3)-detections, (4) riitigation/ intervention actions, or (5) final recovery actions.
Complete HRA can be used to determine upper and lower confidence levels for human error probabilities for the human action categories.
Typical-ly, these probabilities are placed on event trees, which are then used in conjunction with hardware component failure rates by PRA analysts to determine plant-specific and sequence-specific coremelt probabilities.
The team reviewed (1) a series of generic events related to ISLOCA, (2) principal systems (DHR, LPI, HPI, Make-Up/Let Down) which would have a bearing on ISLOCA events, (3) the interaction of these systems, and (4) potential initiating and precursor events.
The team conducted structured interviews with a number of plant personnel and also reviewed various licensee documents, such as Potential Condition Adverse to Quality (PCAQ) reports, LERs, and operations, surveillance, test and maintenance procedures.
5.2.1 Operator Task Demands The team identified a task burden, related to HPI surveillance testing, which could potentially irepact an ISLOCA initiation event.
This task required surveillance testing of the HPI system during power operation and, therefere, required cycling open, at various times, all valves between therRCS and the lower Dressure HPI piping.
Although the team found that no planned operation created a potential for an ISLOCA through a single equipment failure or operator error, these operations did indicate the need for an awareness of ISLOCA precursors by the operators who were responsible for the final, on-line control of value lineups and operations.
The team also identified several concerns with regard to monitoring for an ISLOCA event at the local stations.
Accessibility for local operation of valves was cumbersome, and the lighting level was low and non-uniform with deep shadows even with normal lighting (lighting with emergency lights was not inspected), and operators carried flashlights as normal equipment.
5.2.2 Procedures The team found the licensee's procedure for operation and control of locked valves to be weak in that the procedure did not recognize the fact that the proper positions of certain locked valves provided protection against an ISLOCA event as well as, in some cases, preventing safety-related system or system relief paths from becoming inoperable.
The " Operation and Control of Locked Valves" procedure defined the administrative controls and documentation for the locked valve control program.
The procedure's major objective was to provide protection against the " operation of valves which could potentially render a Safety-Related System or System relief flow path inoperable." Addition-ally, the team's review of the operating procedures, abnormal operating
-
- -
.<
,
i..,,,
-.
.
procedures, and energency operating procedures (EOPs) found that these
-
procedures did not provide warnings and precautions regarding the consequences of operator errors or equipment failures which could initiate an ISLOCA event. However, procedural content was technically correct in providing the necessary instructions to avoid overpressuri-
'
zation or initiation of reverse flow.
5.2.3 Operator Training The team reviewed the training lesson plans for the DHR/LPI, HPl/BWST,
'
makeup and purification, and core flood systems, and discussed ISLOCA training with the Manager of Operator Training, to determine whether the operator training provided operators with an understanding of the ISLOCA phenomenon and its prevention and mitigation.
The team reviewed the lesson plan for the the reactor technology course, and found that, although it was well constituted to follow the systematic approach to training principles, ISLOCA precautions were not in the lesson plan content or in the lesson objectives. Although the team was aware of no requirements for conducting training on ISLOCA scenarios and events, the team noted that some training in this area would improve the operators'
ability to detect, diagnose and mitigate'an ISLOCA event. The team found that the lesson plan objectives and content provided for operation of the systems within their pressure and temperature limits, but did not contain discussions with specific warnings and precautions to prevent an ISLOCA event, such,as:
P-o The consequences of mispositioning the manual valves used to vent the HPI lines to the BWST.
o An awareness of potential challenges to the HPI injection PlV check
valves during surveillance testing of the HPl discharge MOVs, if the makeup pump tripped or was secured.
o An awareness that the computer alarm high-pressure on the HPl line could be caused by either leaking check valves or by the makeup and purification pump.
5.2.4 Performance-Shaping Factors (PSF)
L The team collected PSF information to support the current human
'
reliability analysis data collection effort which is part of the ongoing efforts by the Office of Research to employ PRA methods to assess the risks of ISLOCAs.
Human performance in complex systems is subject to influence from such sources as (1) the nature of the task, (2) the adequacy and design characteristics of the MMI, (3) training, (4) experience of the crew or crew member performing the task, (5) procedures, and (6) stress.
Normally, the team finds both positive as well as negative PSFs at any given plant. System feedback, the pacing of the task, and the operator's image of the system are also sources of variation in task performance and
O-f....
s c
.:
'
are used to assist the HRA analyst in determining human error probabilities.
The team's HRA evaluation performed at Davis-Lesse was not intended to calculete the overall plant risk, since thet would not be possible without applying an appropriate numerical estimation method. Addition-ally, the HRA analysis performed did not include an in-depth review of possible restoration actions which might be taken by operations staff anc which might thereby limit the risk of core melt should an ISLOCA occur.
The team made the following PSF and human reliability observations during this assessment.
5.2.4.1 Positive PSFs 1.
The team could find no single action in the operation, maintenance or surveillance test areas which could, in and of itself, result in an ISLOCA.
2.
The team found no indications that the current workload of personnel would unduly introduce either initiating events or precursors for ISLOCA.
3.
The team considered the use of operating schematics by the licensee to be.a pesitive factor. An operating schematic is a non-cesign drawing which is a graphical representation of the operational and
.
functional characteristics of a system. The schematic allows plant personnel to see and understand each system within the plant from a single document versus numerous P&lbs conventionally used.
Information contained includes written control logics, setpoints for control of relief valves, computer points with accresses, annunciators and alarm setpoints, and all inoications and recorders.
4.
The licensee's human performance engineering system (HPES), when fully implemented, should prove to be valuable in human performance monitoring at Davis-Besse.
5.
The practice of repeating verbal directions, for example, echoing back procedural steps between shift supervisor or SRO and equipment operators was an effective and beneficial communication requirement.
6.
Operations and training personnel were observed to be generally aware of reactor safety principles, although they were not specifi-cally knowledgeable of potential ISLOCA scenarios.
7.
Ergonomic factors, such as labeling or accessibility in the control room, should not directly contribute to an ISLOCA initiation.
8.
Personnel were generally well motivated. The mechanics, pipe fitters and electronics technicians interviewed by the team appeared to be conscientious and highly motivated.
9.
The team considered the use of the Gatronics System to be a positive form of communication.
_ _.
_
f..",,
,
i:.
-.
...
10.
Personnel were observed to be aware of safety principles including
-
knowledge of radiation areas and of the requirements for anti-contamination suits and respirator packs for differtnt jobs.
5.2.4.2 Negative PSFs 1.
Some inconsistencies were noted in lighting levels inside the makeup pump room. Accessibility for some valves was observed to be difficult because of their-location.
2.
Minor procedure problems in format and consistency were identified.
3.
No main control board (MCB) pressure indication or MCB pressure alarm was observed in the DHR system.
4.
ho alarm acknowledgentent procedure was available for computerized alarms to control room operators.
5.
There were no training plans or learning objectives which discussed ISLOCA scenarios.
6.
Ladders which were required to operate sonie manual valves were not kept near-the immediate work area where they might be used.
7.
Although5he team considered the use of laninate tags for valve identification to be a good practice, the team found several valves which did not have 16minate tags. However, the team noted that these valves had permanent snetal valve identification tags on them.
5.2.5 Conclusions The teau did not identify any instance in which a single operator error could directly or indirectly initiate an ISLOCA. Most of the possible ISLOCA scenarios involvea failure of two in-series PIVs, which were relatively unaffected by personnel error because of their design, or which had other features such as interlocks to minimize operator error.
The team did find a lack of awareness of the ISLOCA events generally in the plant operations and training staff as indicated through procedural reviews and personnel interviews. Although plant operators, as well as other plant personnel, appeared to be motivated and safety conscious,-the team identified concerns regarding timely operator response to computer alarms, and procedure errors which indicated that the operators were not aware of the potential for an ISLOCA event.
t,..,
,,
c
,
- -
APPENDIX A
.
i PERSONNEL CONTACTED:
i Baker, D.
TEC Procedures Engineer i
-
Black, J.
TEC Nuclear Engineer, Performance Engineering l
-
Bladel, L.
TEC Zone Engineer, HPI Operations
-
Bonnett, W.
TEC Equipment Operator Ill
-
Brandt, D.
TEC Manager of Operations
-
Burton, B.
TEC Order Planning
'j
-
- Byron, P.
NRC Senior Resident
-
Caba, Erdal
TEC Manager, Performance Engineering
-
'
- Campe, K.
NRC Section Chief, NRR'MEP/RAB
-
- Cooper, R.
NRC Eng. Branch Chief, bnS, R-Ill
-
Elfstrom, R.
TEC Station Performance Technologist
-
Evans, L.
TEC Engineer, I&C
-
- Gaston, Ron TEC Engineer, Licensing
-
Hayes, J.
TEC Performance Engineer
-;
-
Hoffman, M.
TEC Maintenance Trainer
-
'
- Holiday, Roy T.
TEC Engineer, Licensing
-
Jacobsen, P.
TEC Supervisor, Electrical Engineering
-
- Jain, S.
TEC Engineering Director
.
-
Kasper, J.
TEC Operator
!
-
- Khazrai, Mort TEC Nuclear Engineer, Performance Engineering
{
-
- Konklin, J.
NRC Section Chief, NRR/DRIS/RSIB
!
-
- Kosloff, D.
?-
NRC Resident Inspector I
- Kuhtenia, D.
TEC Nuclear Engineer
-
- Lash, J.
TEC ISE Manager
-
'
Lightfoot, D.
TEC Procedures Engineer
-
Marley, J.
TEC Sr. Engineer, Systems Engineering
-
Martin, S.
TEC Shift Supervisor
-
L
- Matranga, E.
TEC Systems Engineer
-
Mizik, Robert TEC Shift Supervisor
-
Myers, L.
TEC Reactor Operator
-
Napolitan, P.
TEC Engineer, I&C
-
'
Nixon, J.
TEC Records, Management
-
- 0'Neil, J.
TEC Supervisor, Mechanical Maintenance Engineering
-
Parker, M.
TEC Supervisor, Station Performance
-
- Prasad, K.
TEC Manager, Nuclear Engineering (Acting)
-
Prasoon, G.
TEC Senior Engineer, Mechanical
-
Roberts, R.
TEC Assistant Shift Supervisor
-
Schreiner, D.
TEC Manager, Performance Engineer
-
Schuck, R.
TEC QC Supervisor, Mechanical
-
- Simon, L.
Sr. Performance Technologist
-
- Simpkins, R.
TEC Operation Training
-
Staudt, D.
TEC Assistant Shift Supervisor
-
- Storr, L.
TEC Davis-Besse Plant Manager
-
Szanyi, Frank TEC Nuclear Engineer, Performance Engineering
-
- Timms, D.
TEC Systems Engineer
-
Wise, S.
TEC Operations Training (ReQual)
-
- Wuokko, D.
TEC Regulatory Affairs Supervisor
-
Wilson, A.
TEC Maintenance Technician
-
Young, B.
-
TEC Assistant Shift Supervisor
Attended exit meeting on November 9, 1989.
A-1 J
.
.
..
f....,,
.
-
APPENDIX B
.
DOCUMENTS REVIEWE0 1.
Maintenance Procedures:
Procedure No.
Title Rev. Effective Date AD-1844.01.0B Preventive Maintenance
06/09/89 DB-ME-09301 Preventative Maintenance for Type SMB
and SMC Limitorque Valve Operators DB-MM-09331.00 HP1 Stop Check Valve Maintenance
DB-MM-095B7.01 Maintenance and Repair of Limitorque
Valve Operators Types SMB-00 and SMB-000 DB-MN-00001 Conduct of Maintenance
03/31/89 DB-MV-00002 Preventative Maintenance
DB-PN-00007 Control of Work
DB-PN-00025 Prs-Maintenance and Post-Maintenance
09/15/89 Tbsting Requireme0ts
-
EN-DP-01070 Preventive Maintenance
07/28/89 EN-DP-01074 Predictive Maintenance
07/23/89 EN-DP-01072 Post-Modification Tests
07/16/89 MP 1411.04 Maintenance and Repair of Limitorque
05/03/88 (DB-MM-Valve Operators Types SMB-00 Through 095B7-01)
.5MB-000 MP 1411.05 Maintenance and Repair of Limitorque
05/03/88 (DB-MM-Valve Operators Types SMB-0 Through 095B7.02)
SMB-4 MP 1411.07 Maintenance and Repair of Limitorque
10/28/86 Valve Operator SMC-04 MP 1411.08 Velan Bolted Connet Check Valves
Disassembly, Repair and Reassembly MP 1701.99 Velan Forged Glove and Stop-Check
Valve Maintenance MP-1700.04.01 Limitorque Manual Actuators, 09/19/89
--
Disassembly, Repair, and Reassembly MP 1702.94 HPI Stop-Check Valve Maintenance
12/29/87 B-1
e..
.r.....,
f ' *?
-
2.
Operatino Procedures
'
Procedure No.
Title Rev. Effective Date AB 1203.35 Loss of Heat Removal System
04/06/87 Emergency Operating Procedure DB-0P-00008 Operation and Control of Locked
03/10/89 Valves DB-0P v 016 Removal and Restoration of Station 03/15/89
--
-
Equipment DB-0P-00100 Shift Turnover
09/27/89 DB-0P-02500 Turbine Trip
10/20/89 DB-0P-02531 Steam Generator Tube Leak
11/28/88 DB-0P-04004 Locked Valve Verification
07/05/89 DB-0P-06000 Filling and Venting the RCS 10/13/88
--
DB-0P-06002 Draining and Nitrogen Blanketing of
09/25/89 the RCS DB-0P-06011 H@hPressureInjectionSystem 21-11/21/86 Procedure DB-0P-06014 Core Flooding System Procedure
02/06/89 DB-0P-06015 Borated Water Storage Tank
09/12/89 Operating Procedure DB-0P-06901 Plant Startup
07/11/89 DB-0P-06903 Plant Shutdown and Cooldown
06/22/89 DB-PF-02000 RPS, SFAS, SFRCS Trip, or SG Tube
11/18/88 Rupture DB-PN-06900 Pre-Startup Checklist
07/10/89 NG-IM-00115 Preparation and Control of Nuclear
09/11/89 Group Department and Section/ Unit Procedures PP-1101.01 Nuclear Steam Supply System
11/01/86 Limit and Precautions SP 1104.02 Makeup and Purification System
05/02/87 SP 1104.04 Decay Heat and Low Pressure
05/05/87 Injection Operating Procedure SP1104-02 Makeup and Purification System
09/28/89 Operating Procedure B-2
-
-
to...,3
-
3.
Surveillance Procedures
Procedure No.
Title Rev. Effective Date AD 1823.01 Setpoint Control
09/06/87 DB DP-00013 Surveillance and Periodic Test
06/23/89 Program DB-ME 09300 Testing of Motor Operated Valves
10/25/89 Using Movats DB-MI-03134 Channel Calibration of 64B-ISPRC02A3
10/10/88 Reactor Coolant Loop 2 Hot leg Wide Range Pressure SFAS CH. 4 DB M1 03716 Channel Calibration of $2-ISFHP03B,
6/13/88 High Pressure Injection 2-2 Flow DB-PF-03065 ASME Section XI Pressure Test
7/27/89 DB-PF-03935 HPI System Pressure Isolation
7/25/88 Integrity Test HP-22 and MU-800 DB-PF-03936 H&1 System Pressure Isolation
7/25/88 lhtegrity Test HP-23 and MU-169 DB-PF-03937 HPI System Pressure Isolation
9/30/88 Integrity itst Back-To Back Check Valves PB-PN-06900 Pre-Startup Checklist Da 11f05/88
^
,
DB-SP-03130 Decay Heat Removal System Isciation
12/05/88 Test DB-SD-03300 RCS !solatior Velte Leak Test
10/26/67 (CF-30, CF-31, DH-76, DH-77)
DB-SP-03301.01 CF-30, DH 76, DH 1A Combined Leak Test DB-SP-03302 CF-28 and CF-29 Leak Test
10/26/87 i
IC 2701.42 Calibration of Static-0-Ring Pressure
4/17/89
!
(DB-MI-05152) Switches IC 2701.86 String Check of 51-ISLCF 3Al Core
8/21/88 Flooding.'ank 2 Level IC 2701.87 String Check of 51-15LCF 3A2 Core
9/26/88 Flooding Tank 2 Level IC 2701.88 String Check of 51-ISLCF 3B1 Core
9/26/88 Flooding Tank 1 Level B-3
_ -_ __
. _ _ _ _ _ _ _
- ....,
,
it IC 2701.89 String Check of 51-15LCF 3B2 Core
10/1/88 Flooding Tank I Level i
SP-1204-04 Decay Heat and Low Pressure Injection
08/15/89 Operating Procedure l
ST $099.11 Operational Checks of Passive Valves
11/04/86 4.
Pipino and Instrumentation Diaorams
!
Drawing Number Title Rev.
!
M-019 Nitrogen Supply System
!
M-030A Reactor Coolant System
M-030B Reactor Coolant System Instrumentation
M-031A Make Up and Purification System
M-031B Make Up and Purification System
M-0310 Make Up and Purification System
M-033A High Pressure Injection
i M-033B Decay Heat Train 1
M-033C Decay Heat Train 2
M-034 ECC5 & CTMT Spray & Core Flooding Systems
M-040A Reactor Coolant System Details
5, Electrical Drawings 7-
.
Drawing No.
Title Sheet Rev.
E-30B General Guides - Elementary Diagrams
20 Contro1 Switch and Pushbutton D? tails t
E-30b Gencrcl GJ des Elementary Miscella-S 6 Diagrams
neout Switch Cevelopment E-52B Rear. tor cooling System - DH Normal 24A
Suction Valve F 32B Reactor CooMeg $ystea DH Suetinn 24B
Section Valve E-575 Peactor Cooling System - DH hormai 2aC
Suction Valva i
E-52B Reactor Cooling System - DH Normal 24D
Suction Valve
,
,
E-52B Reactor Cooling System - HP Injection 26A
Line Valves E-52B Reactor Cooling System - HP Injection 26B
Line Valves
,
E-528 Reactor Cooling System - RC DT Liquid
8
& Gas Isolation VLV E-528 Reactor Cooling System - RC DT Liquid
8
& Gas Valve B-4
. -
e
,
.N....s
,
s'
6.
Other Documents
-
A.
Davis-Besse Training Inf ormation Manuals
!
High Pressure Injection System, Revision 1, 10-13-88.
!
Makeup and Purification System, Revision 1, 10-07-88.
B.
System Review and Test Program Reports Core Flood System, Revision C, September 29, 1986.
Decay Heat Removal System, Revision D, October 27, 1986.
High Pressure Injection, Revision D, September 30, 1986.
Mateup and Purification, Revision D, 09-29-86 C.
Maintenance Manuals
'
Instruction Manual f or Installation and Maintenance of Velan Manual
!
Operated Bolted Bonnet Gate, Globe, Stop Check and Check Valves, Manual No. VEL-HO-1.
Maintenance Manual for Bolted Bonnet Gate and Globe Valves and Bolted Cover Check Valves, Manual VEL-CSVM.
D.
Miscellaneous Davis-Besse,Muclear Power Station, Unit No. 1, Instant Service Reactor Operator Qualification Manual, Revision 2, June 14, 1989.
Human Performance Evaluation System, "DB-PN-00013 ', Revision 00, February 2, 1989.
02-02-89 8.eiter 6.C. Wr);h;, NRC fegion 'll to Telede t.dison, AriN:
Donala shelten, f-ecemret 12, 15B8 '.Special Inspection of the Davir-Ber.se frergency Operating Prnrecaret),
IF #:
OLC-PWR-1043, Revision 3, 03-29'89, Decay Heat Removci/ Low Pres-sure injection, LP #:
OLC-PVR-042.03, Revision 3 Hich Pressure injection /8957.
,
LP #:
OLC-PWR-1044.1+visien 3, L3-L1-eb Makeup and Purification.
LP #:
OLC-P.;R-1041, Revision 2, Core Flood System.
Nuclear Group Procedure NG-lM-00115 Preparation and Control of Nuclear Group Department and Section/ Unit Procedures Policy No. P-0PS-6,10-12-89, Incorporating Operating Experiences into the Operations Training Program.
Potential Condition Adverse to Quality Report No. 89-0187, Hold Points Missed MWO 1-88-192-05.
Potential Condition Adverse to Quality Report No. 89-0158, CV5011D Would Not Stroke.
B-5
ws
-
-
,
,
!
.dassaso
'
>
o-o
., '
Potential Condition Adverse to Quality Report No. 88-0620,- Procedure Cross Comment Review, Potential Condition Adverse to Quality Report No. 88-06B0, Procedure Requirement on Operators Without Training.
Potential Condition Adverse to Quality Report No. 89-0255, ST5016.09 Fire l
System Valve Positions-Definiencies.
'
Procedures Writers-Guide PWG SEC 3.0 R01.
!
Quality Assurance Department Procedure QA-0P-00250 Conduct of the Quality Control Section (10.1).
j -
L
'
-Quality Assurance Department Procedure QA-DP-00650 Inspection of Mainte-
nance and Modification Activities, w
,
!.
l
!
>
b r-t
,
,r-Is
l
.,'
.
!
b
.
I i
B-6
.
.
..
..
_-.
.-
__
_
. -.
..
.
F"
.,
.'s
<> a O
%
Mr. D. C. Shelton-2-
No formal response to this report is required.
In accordance with 10 CFR 2.790(a), a copy of this letter and the enclosure will be placed in the NRC Public Document Room.
Should you have any questions concerning this inspection, please contact Mr. J. E. Konklin (301-492-0953).
.
Sincerely, Gary M. Holahan, Acting Director Division ot~ Reactor Projects !!!/IV/V and Special Projects
,
l Office of Nuclear Reactor Regulation
!
I Enclosures:
1.
Executive Summary 2.
NRC Inspection Report 50-346/89-201
.
cc w/ enclosures: See page 3
pistribution: See page 4
'
.
.
{
,
f C
B f
RJBarr tt 12/ &/8
,
- SEE PREVIOUS CONCURRENCES
-
)
GMHo]ah[tn
,
AD: RSP r'
- RSIB:DRIS
- SC:RSIB:DRIS *C:RSIB:DRIS
- TECH EDITIOR,
RIE JAlsom:bt JEKonklin WDLanning RSanders
'BJ!GHmes 12/19/89 12/19/89 12/19/89 11/29/89 12/g/89 12/7,1,/89