ML20207N880
| ML20207N880 | |
| Person / Time | |
|---|---|
| Site: | Byron  |
| Issue date: | 10/12/1988 |
| From: | Hinds J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20207N875 | List: |
| References | |
| 50-454-88-19, NUDOCS 8810190437 | |
| Download: ML20207N880 (9) | |
See also: IR 05000454/1988019
Text
_ _ _ _ _ _ _ _
. _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .
'
.
,
'
'
.
.
.
U. S. fiUCLEAR REGULATORY C0fYtISS10fi
REGION III
Report No. 50-454/88019(DRP)
Docket No. 50-454
License No. NPF-37
Licensee:
Commonwealth Edison Company
Post Office Box 767
Chicago, IL 60690
Facility Name: Byron Station, Unit 1
Inspection At: Byron Station, Byron IL
Inspection Conducted: September 19 - 28, 1988
Inspectors:
P. G. Brochman
N. V. Gilles
-
.
Approved By:
J.
1. Hinds , Jr. ,
i
10 12 86
V ctor Projects Section 1A
Date
Inspection Suro,ary
inspection from September 19 - 28, 1988 (Report No. 50-454/88019(DRP))
Areas Inspected:
Special, reactive safety inspection by the resident
inspectors to review the events surrounding the loss of one train of the
residual heat removal (RHR) system while the water level in the reactor
coolant system was being lowered on September 19, 1988.
Results: Two examples of one apparent violation were identified (procedures
inadequately directed installation and use of tygon tubing for reactor
coolant system level indication - paragraph 7).
This apparent violation is
of safety significance due to the potential for the loss of both trains of
RHR when reactor vessel level is at the mid-plane of the hot leg nozzles and
both trains of RHR are in operation.
However, this event did not affect the
public's health and safety due to the prompt actions by the control room
operator to prevent the loss of the RHR system.
S810190437 891012
ADOCK 05000454
O
_ _ _ _ _ _ _ _ _ _ _ _ _ _
____.
'
'
.
.
.
.
DETAILS
1.
Persons Contacted
Conmonwealth Edison Company
R. Pleniewicz, Station Manager
- T. Joyce, Production Superintendent
- R. Waid, Services Superintendent
D. Winchester, Quality Assurance SJperintendent
T. Tulon, Assistant Superiatendent, Operating
- G. Schwartz, Assistant Superintendent, Maintenance
- L. Sues, Assistant Superintendent Technical Services
- 0. St. Clair, Assistant Superintendent, Work Planning
J. Schrock, Operating Fngineer, Unit 1
- D. Brindle Operating Engineer. Unit 2
- M. Snow, Regulatory Assurance Supervisor
- W. Pirnat, Regulatory Assurance Staff
- E. Zittle, Regulatory Assuranca Staff
- B. Jacobs, Acting Unit 1 Operating Engineer
- L. Bunner, Lead Operations Instructor
- D
Bump, Quality Assurance Inspector
- A. Chernick, Training Supervisor
i
- D. Berg, Nuclear Safety
- R. Polek, Regulatory Assurance
The inspector also contacted and interviewed other licensee and
'
contractor personnel during the course of this inspection.
- Denotes those present during the exit interview on September 28,
1988.
2.
Purpose (93702)
This inspection was conducted to review the circumstances surrounding
the inoperability of the 1A train of the residual heat removal (RHR)
system after air was ingested into the RHR system us a vortex formed
while operators were lowering the water level in the refueling cavity
to support maintenance activities on September 19, 1988.
3.
Description of the Event
Tne licensee entered the second refueling outage for Unit 1 on
September 3, 1988. On the morning of September 19, 1988, with the
reactor vessel (RV) head removed, repairs had been made to the refueling
cavity boot seal. The 1A RHR train was operating in the shutdown cooling
mode at a flowrate of aporoximately 3200 gpm. Operators had just
!
completed filling the refueling cavity to the 402 foot elevation to
check the boot seal repairs when they noticed that a vessel stud hole
plug had come free and was floating on the surface of the water. At
i
l
l
2
.
__
- - - - _ _ _-
-_
..
-- __
_ _ _ _ - .
- _ _ _ _ . - _ _ _ _ . _ _
. _ _ _ . _
_
_
. _ -_ _ __-_-
-
'
j
.
I
.
.
.
4
4:38 a.m., operators started the IB RHR pump in order to lower reactor
coolant system (RCS)/ refueling cavity water level to below the RV flange
so that maintenance personnel could replace the stud hole plug.
At 4:48 a.m., operators shut down the IB RHR pump with RCS level at
398 feet 8 inches, according to the temporary tygon hose level
instrumentation (the RV flange is at the 400 foot elevation). Between
approximately 6:00 a.m. and 7:00 a.m., a shif t change was occurring for
the operating shift. Atapproximately7:30a.m.,anequipmentattendant
(EA) assigned the duty of "tube watch' relieved the EA from the previous
'
shift and told the control room that the tygon hose was reading approxi-
mately 401 feet. The control room attributed the difference from the
earlier indication to the fact that the tygon hose had been acting
erratically on previous shifts. At 9:57 a.m., with level approximately
,
l
3 inches above the RV flange according to visual observation, the Unit 1
Reactor Operator (RO) started the IB RHR pump to lower RCS level further
to allow reinstallation of the stud hole plug. At this time the EA and
a mechanical maintenance foreman were at the refueling cavity and in
comunication with the control room. At 10:05 a.m. , the RO shut wwn
the IB RHR pump when told that level had dropped below the RV flange.
The EA had checked the tygon hose and told the RO that level had dropped
approximately 1 foot. At approximately 10:45 a.m., the Unit 1 R0 was
relieved by an extra RO on shift due to increased activity because of the
outage.
At about the same time, the mechanical maintenance foreman sent a man
into the refueling cavity to check the water level and subsequently
informed the control room that there was still water standing above the
RV flange / stud holes. The foreman then asked to have level in the
refueling cavity lowsred further to ellow dewatering of the stud hole and
replacement of the plug. At 10:52 a.m., the Unit 1 R0 started the IB RHR
pump again to lower refueling cavity level. At 10:59 a.m., just as the
maintenance foreman told the R0 that level had dropped far enough, the RO
noticed the 1A RHR pump ameter oscillating (30 - 50 amps) and letdown
flow fluctuating. The RO isolated letdown flow and shut down the IB RHR
pump, to stop the draining. At 11:00 a.m., the R0 shut down the 1A RHR
pump, with amps now fluctuating between 20 and 60 amps. The R0 then
entered Byron Abnormal Operating Procedure B0A Refuel-4, "Loss of RH
During Refueling," and action statement a. of Technical Specification 3.9.8.2 for the loss of one RHR pump in Mode 6 with watcr level less than
23 feet above the RV flange. At 11:02 a.m., the RO segan a gravity fill
of the RV/ refueling cavity from the refueling water storage tank (RWST)
using the 1B RHR train.
The 18 RHR pump was then started several minutes
later, after the R0 was certain that there was sufficient level in the RV
to prevent cavitation of the pump. By this time operators had checked
the containment radiation monitors and noted no changes in radiation
level. At 11:09 a.m., the R0 shut down the IB RHR pump with the refueling
cavity level approximately 6 inches above the RV flange. The R0 realigned
the IB RHR train for shutdcwn cooling and started the IB RHR pump at
11:14 a.m.
Procedure B0A Refuel-4 was exited. Operations department
management directed operators to fill and vent the 1A RHR train. An
air and air / water mixture was vented from a 3/4-inch sample line for
3
.
~
.
1
'
.
.
<
'
approximately 20 to 40 seconds, with its valyc 1/4 turn open.
At
1:39 p.m., operators began a gravity fill of the refueling cavity from
the RWST using the 1A RHR train and shortly thereafter started the 1A
RHR pump. An EA at the pump indicated that everything looked and sounded
normal. Control room indications appeared nonnal. Based on these
observations, the licensee declared the 1A RHR pump operable at 1:46 p.m.
,
and exited the Limiting Condition for Operation Action Requirement. At
!
3:33 p.m., the 1A RHR pump was shut down with the refueling cavity filled
j
to 24 feet 6 inches above the RV flange.
'
i
~
4.
Chrono'
of Events
All of the 6 ;nts described below occurred on September 19, 1988, and
utilize the 24-hour clock.
The times are approximate.
1
0430 Gravity fill of refueling cavity to 402 foot level to check the
effectiveness of the RV boot seal repairs.
0435 Boot repairs appeared satisfactory; however, an RV stud hole
plug came free,
j
0438 Started IB RHR pump to lower refueling cavity level to below RV
flange to replace the stud hole plug.
i
0448 Shut down IB RHR pump. RV level at 398 feet 8 inches per tygon
,
I
hose indication (corrected).
{
0730 Equipment Attendant (EA) checked level in the tygon hose, which
read approxiretely 401 feet. The control room attributed the
4
i
difference between this reading and the reading at 0448 to the
j
fact that the tygon hose indication had been acting erratically,
0957 Started IB RHR pump to lower RCS level further to allow
i
maintenance to reinstall the stud hole plug.
Level now
!
approximately 3 inches above RV flange by visual observation.
1005 Shut down IB RHR pump. Level lowered approximately 1 foot per
!
tygon level indication.
i
1045 Unit 1 R0 relieved by extra RO.
i
1052 Started IB RHR pump again. Maintenance requested slightly
!
lower level. Maintenance Foreman monitoring level at RV
I
cavity, and individual at flange.
1059 Isolated letdown and shut down IB RHR pump due to fluctuating
indicaticns of letdown flow and amps on 1A RHR purp (IA RHR
{
aligned for shutdown cooling at approximately 3200 gpm).
i
1100 Shut down 1A RHR pump. Amps still fluctuating fro'A 20 to 60
amps.
Entered Byron Abnormal Operating Procedure B0A Refuel-4,
"Loss of RH During Refueling." Entered action statement a, of
l
Technical Specification 3.9.8.2 on loss of one FAR pump during
l
refueling with water level less than 23 feet above RV flange,
t
1102 Aligned 18 RHR train to gravity fill refueling cavity from the
RWST.
'
)
1108 Started the IB RHR pump.
No change in containment low or high
range area radiation monitors.
i
1109 Shut down IB RHR pump.
RCS level approximately 6 inches above RV
{
'
1114 Started IB RHR pump to provide for shutdown cooling.
Exited BOA
l
Refuel-4
l
l
l
4
l
.
'
.
.
.
1153 Checked a 3/4" sample line at 1A RHR pump discharge.
Vented air
and air / water mixture for 20 to 40 seconds with valve 1/4 turn
open.
1339 Started gravity fill of refueling cavity from RWST using 1A RHR
train.
1342 Started 1A RHR pump. Pump observed locally. Control room
indicators checked and no problems indicated.
1346 Exited Limiting Condition for Operation Action Requirement on 1A
RHR pump.
1533 Filled refueling cavity to 24 feet 6 inches above reactoi flange.
Shutdown 1A RHR pump and aligned 1A RHR train for shutdown
cooling.
.
5.
Evaluation of the Event
Based on the inspectors' analyses of the event, it appears that water
level in the refueling cavity remained above the flange while the actual
level inside the RV dropped to approximately one inch below the center-
line of the hot leg norzles. This phenomenon was caused by the hydraulic
resistance of the small flow holes on the edge of the top hat, which
restricted the rate at whici; water could drain from the refueling cavity
into the RV while, simultaneous!y, the RHR pump was rapidly removing
water from inside the RV, as the control rod guide stub tubes provided
sufficient venting to allow for easy drainage.
The uncoupling of water
levels inside and outside of the RV was compounded by tne operators
believing that water level indication inside the RV could be obsarved
visually with the upper internals installed. As a result, RCS level was
unintentionally lowered to approximately one inch below the centerline
ef the RV hot leg nozzles.
This allowed a vortex to develop in the
suction line of the 1A RHR pump and air was entrained into the system,
thereby making it inoperable.
The procedure being ustJ to drain the re',eling cavity was BOP RH-9,
Revision 51A, "Pump Down of the Refueling Cavity to the RWST."
Prerequisite C.14 states, "ENSURE that the tygon hose connection, for
RCS level indication, is in place ... prior to lowering the water
level below the Reactor Vessel Flange (EL = 400'0")."
Subsequent to the event, the tygon hose was walked down by one of the
licensee's engineers.
The engine
removed several loon seals from the
hose and restructured the tygon hose to establish vertical installation
in all places. He also corrected a 2 foot level discrepancy with the
elevation markings for the tygon hose, which had existed since its
installation.
Based on interviews and discussions with licensee personnel and a review
of records, the inspectors have developed the following obs#:rvations
regarding this event:
There appeared to be o widely held, incorrect perception by the
licensed operators that water level inside the RV could be visually
cbserved with the upper internals installed,
ho training was
5
_
.
-
.
provided to the R0s or SR0s to either reinforce nor discredit this
idea. Licensed operator training discusses all of the approved
methods of RCS level indication however, no mention of visual level
indication is made.
The EAs had not received training which would allow them to relate
an elevation number to a physical point in the plant, e.g., the RV
flange is at the 400 foot elevation. Consequently, when the EA
reported that level was 398 feet 8 inches and then went up to the
refueling deck and observed level above the RV flange, he did not
recognize the disparity between these two observations and, conse-
quently, did not identify this fact to the control room.
Licensee management believes that the control room operators followed
l
procedure BOP RH-9 and that direct visual observation is an
l
acceptable method for determining refueling cavity water level.
l
Management believes that the BOP does not require that the tygon
i
hose be used and implies that visual observation is acceptable when
i
draining the refueling cavity to the RV flange. Management stated
j
that this had been done before, with no problems having occurred.
The inspectors have reviewed the BOP in detail and agree that there
is no explicit statement to utilize the tygon hose during evolutions
which lower water level below the RV flange.
In fact, there is no
explicit statement in the BOP to utilize any means of level
indication while lowering RCS level.
I
However, the inspectors note that prerequisite C.14 requires the
i
operators to ENSURE (emphasis original) that the tygon hose
l
connection is in place, prior to lowering water level below the RV
,
Consequently, the inspectors believe that it is the clear
I
intent of this prerequisite to require that the tygon hose be
l
utilized to detennine water level for any draining evolution which
will lower water level below the RV flange, as the licensee planned
to do.
The inspectors agree that visual observation can be useful,
but believe that it should be used in addition to, rather than
in place of the tygon hose.
The licensee did not intend to lower water level to mid-loop, only
several inches below the RV flange.
The licensee does not normally
use BOP RH-9 to drain down to mid-loop operation (below the RV
flange).
However, the 80P does not have any restriction concerning
lowering RCS water level below the RV flange, other than not to go
below mid-plane of the hot legs.
The inspectors believe that
several of the conservatisms utilized in BOP RC-4a, "Draining the
RCS" could be effectively utilized in B0P RH-9, such as reducing the
flow rate of the pump supplying shutdown cooling from 3000 gpm to
1000 gpm, draining via the letdown system (120 gpm) verses via RH8735
(RHR to Safety Injection pump isolation valve) to the RWST (750 -
1000 gpm), suspending draining operations if level indication is
unreliable, and resolving discrepancies in level indication before
resuming draining activities.
6
_ _ _ _ _
_ _ - _ - _ _ _ _ _ . - _ _ _ _ _ _ - - _ - _ -
.
.
.
.
The acceptar.ce by control room personnel that the tygon hose was
inaccurate (be it level error or slow response) while continuing
to perform draining operations, even if using another "acceptable"
method of level indication, does not appear to the inspectors to
be the most conservative method of operation.
The response of the control room operators, once they had received
i
indication of problems with the 1A RHR train, was prompt and
l
effective and prevented the potential for bo h trains to become
air-bound and inoperable.
The licensee has provided substantial training to the operators
on the hazards of mid-loop operation.
The licensee believes that
because the operators continuously monitor the performance of the
RHR pumps while draining the RCS they would not allow the level
to drop to the point where both trai'is would become air-bound
without the RO initiating action to stop the evolution.
The
licensee noted that even though the RO believed that RCS level
was in the vicinity of the flanca, the RO followed the conflicting
indications of RHR pump performance he was receiving and then
stopped the 1B RHR pump to secure the draining and initiated
corrective actions to recover level to a known elevation.
The licensee believes that the safety significance of this event is
mitigated by the ability of the RWST to gravity fill the refueling cavity
at 1500 gpm per train.
In addition, the high and intermediate head
emergency core cooling system pumps could have been returned to service
and then utilized to inject water into the RV had both RHR pumps becoce
Technical Specification 3.9.8.2 requires that two RHR loops shall be
operable and at least one loop shall be in operation with the unit in
Mode 6 and water level less than 23 feet above the RV flange. With less
than two loops operable, action a requires that corrective actions be
initiated immediately to re'. urn the affected loop to an operable status
or that wate level be established at greater than 23 feet above the RV
flange. The inspectors have reviewed the licensee's actions and believe
that they comply with Technical Specification 3.9.8.2.
6.
Corrective Actions Initiatcd by the Licensee
Subsequent to the event the following corrective actions were implemented
by the licensee:
a.
The licensee issued a Daily Order prohibiting draining of the
refueling cavity below the 'too hat" of the RV upper internals.
On September 22, 1988, after further evaluation of the event, the
licensee issued a second Daily Order prohibiting draining of the
refueling cavity below the tops of the control rod guide tube
assemblies whe.1 the upper internals are installed.
This is to
prevent the condition where flow into the RV plenum area is only
via the 32 small head bypass flow holes.
7
__
_
_ _ _ _ _ _ _ _ _ _ _
___- - _ _ _ _
. -
-
.
,
4
.
.
b.
The licensee conducted training sessions for all operating shifts
j
to discuss this event and the lessons learned,
i
!
c.
The licensee has revised B0P RH-9 which now requires that two
methods of level indication must be utilized and functional prior
to any draining operations more than 2 inches below the top of
the control rod drive shaft support housings. The procedure also
,
!
contains a caution that visual indication of RV level at or below
"
the "top hat" area of the upper internals is not reliable or
]
acceptable for one of the two methods for level indication while
draining the RV.
In addition, the procedure also requires that
,
draining below the 403 foot elevation (top of the control rod drive
shaf t support housing) be done at a slow draining rate, such as the
i
minimum indicated flow rate on the RHR pump, and in any case, less
than 1000 gpm. Draining below the RV flange (400 feet) is required
to be done using letdown, which allows a maximum flow rate of 120
,
gpm RHR to letdown, compared to a 500-1000 gpm flow rate when RHR
to RWST is used.
!
,i
>
l
d.
New instructions for the tygon hose level indicator installation
are being developed which will utilize aa installation checklist.
!
An SRO walkdown of the tygon hose after installation will be
.'
required prior to its use.
i
e.
The licensee has developed an operator aid which relates gallons of
water removed per foot of level change for both the reactor vessel
j
and the refueling cavity.
I
f.
The Westinghouse Owners Group is evaluating a Technical Specifica-
'
tion change which would utilize an RHR flow vs. time vs. level
j
concept for requirements on RHR in Mode 6.
i
j
The licensee has deferred development of further long tem corrective
actions until after the NRC issues the generic letter on mid-loop
>
operations, to ensure that no conflicts exist.
This generic letter
1
provides significant new requirements for licensees on mid-loop
l
operations.
7.
Conclusion
10 CFR 50, Appendix B, Criterion V, as iniplemented by Commonwealth Edison
Company's Quality Assurance Manual, Quality Requirement 5.0, requires
that activities affecting quality shall be prescribed by documented
procedures of a type appropriate to the circumstances. Byron Operating
Procedure B0P RH-9, Revision 51A, "Pump Down of the Resctor Cavity to the
RWST " paragraph C.14, requires that the operator "ENSURE that the tygon
hose connection, for RCS level indication, is in place or LI-RYO46
Reactor Vessel level Instrumentation operable prior to lowering the water
level below the Reactor Vessel Flange (EL-400'0")." However, nowhere in
the procedure does it specifically require that the tygon hose, or any
other means of RCS level indication, be utilized during the drain down
process.
The failure of B0P RH-9 to specify the use of an appropriat2
8
.-
.
'
-
.
means of RCS level indication while draining the refueling cavity to a
level below the RV flange is an apparent violation of 10 CFR 50, Appendix
B, Criterion V (454/S8019-01a(ORP)).
Additionally, BOP RC-4a, "Reactor Coolant System Orain," which contains
instructions for installation of the tygon hose level indication system,
provides no guidance for ensuring that the tygon hose was properly
installed to accurately reflect r<CS level.
For example, the procedure
did not require verification that the tygon hose was installed vertically,
that there were no loop seals, or that the elevation narkings were
accurate. The failure of B0P RC-4a to provide adequate guidance for
installing the tygon hose is an example of an apparent violation of 10 CFR 50, Appendix B, Criterion V (454/88019-Olb(ORP)).
The failure of BOP RH 9 to require the use of the tygon hose for RCS
level indication during draining of the refueling cavity led the control
room operators to utilize a method of level indication which was
ineffective when level was below the control rod driv: shaft support
housings. The consequence of this action was to unintentionally drain
the RCS to an estimated 1 inch below the center line of the reactor
vessel hot leg nozzles, thereby allowing a vortex to develop and air to
be entrained into the 1A RHR system, rendering one train of RHR
During his follow-up of this event, the inspector identified a concern
i
relating to the human factors design of the RHR flow indicators in
the main control room.
The flow meter scale is non-linear and very
-
compressed at the lower end, with only 1 graduation between 0 and 1000
!
gpm. Consequently, the licensee is requested to evaluate the
l
desirability of rescaling the RHR flow meter or other options such as
providing a second, lower-range meter, to improve the ability of the R0s
l
to accurately monitor low RHR flow rates. This concern will be tracked
!
as an open iten (454/88019-02(DRP)).
I
(
8.
Open Items
J
Open items are matters which have been discussed with the licensee, which
will be reviewed further by the inspector, and which involve some action
!
on the part of the NRC or licensee or both. An open item disclosed
during the inspection is discussed in paragraph 7.
9.
Exitinterview(30703)
i
The inspectors met with the licensee representatives denoted in paragraph
1 at the conclusion of the inspection on September 28, 1988.
The
inspectors sumarized the purpose and scope of the inspection and the
!
findings.
The inspectors also discussed the likely informational content
f
of the inspection report, with regard to documents or processes reviewed
by the inspectors during the inspection.
The licensee did not identify
any such docunents or processes as proprietary.
l
t
i
l
9
[
.