ML20216D681
| ML20216D681 | |
| Person / Time | |
|---|---|
| Site: | Maine Yankee |
| Issue date: | 09/03/1997 |
| From: | Dan Dorman NRC (Affiliation Not Assigned) |
| To: | Sellman M Maine Yankee |
| References | |
| NUDOCS 9709090390 | |
| Download: ML20216D681 (17) | |
See also: IR 05000309/1997004
Text
__
-_-_------_- - - --- - - - -
- %
..
t
p
%
UNITED STATES
g
j
NUCLEAR REGULATORY COMMISSION
e
WA'tHINGTON, D.C. 3066H001
\\
September 3, 1997
Mr. Michael B Sellman, President
Maine Yankee Atomic Power Company
329 Bath Road
Brunswick, ME 04011
SUBJECT:
REVIEW OF PRELIMINARY ACCIDENT SEQUENCE PRECURSOR ANALYSIS OF
OPERATIONAL CONDITION AT MAINE YANKEE ATOMIC POWER STATION
Dear Mr. Sellman:
Enclosed for your review and comment is a copy of the preliminary Accident
Sequence Precursor (ASP) analysis of an operational condition which was
discovered at the Maine Yankee Atomic Power Station on January 22, 1997
(Enclosure 1), and reported in Licensee Event Report (LER) No. 50-309/97-004.
This analysis was pre)ared by our contractor at the Oak Ridge National
Laboratory (0RNL).
T1e results of this
{
this event may be a precursor for 1997. preliminary analysis indicate that
i
In assessing operational events, an
effort was made to make the ASP models as realistic as possible regarding the
j
specific features and res)onse of a given plant to various accident sequence
<
initiators.
We realize tlat licensees may have additional systems and
emergency procedures, or other features at their plants that might affect the
analysis.
Therefore, we are providing you an opportunity to review and
comment on the technical adequacy of the preliminary ASP analysis, including
the depiction of plant equipment and equipment capabilities.
Upon receipt and,
evaluation of your comments, we will revise the conditional core damage
probability calculations where necessary to consider the specific information
you have provided.
The object of the review process is to provide as
realistic an analysis of the significance of the event as possible.
In order for us to incorporate your comments, perform any required reanalysis,
and prepare the final report of our analysis of this event in a timely manner,
you are requested to complete your review and to provide any comments within
30 days of receipt of this letter. We have streamlined the ASP Program with
the objective of significantly improving the time after an event in which the
final precursor analysis of the event is made publicly available.
As soon as
.
Our final analysis of the event has been completed, we will provide for your
zu
information the final precursor analysis of the event and the resolution of
,
your comments.
In previous years, licensees have had to wait until
publication of the Annual Precursor Report (in some cases, up to 23 months
after an event) for the final precursor analysis of an event and the
resolution of their comments.
- I
We have also enclosed several items to facilitate your review.
Enclosure 2
contains s)ecific guidance for performing the requested review, identifies the
criteria w1ich we will apply to determine whether any credit should be given
in the analysis for the use of licensee-identified additional equipment or
specific actions in recovering from the event, and describes the specific
information that you should provide to support such a claim.
Enclosure 3 is a
copy of LER No. 50-309/97-004, which documented the event.
ado $ ob000009
NRC FILE CENTER COPY
D
S
.
_ _ _ _ _ _ _ _ _
._-
- a
3
Mr. Michael B. Sellman
-2-
Please contact me at (301) 415 1429 if you have any questions regarding this
request.
This request is covered by the existing OMB clearance number (3150-
0104) for NRC staff followup review of events documented in LERs.
Your
response to this request is voluntary and does not constitute a licensing
requirement.
Sincerely.
Original signed by
Daniel H. Dorman. Project Manager
Project Directorate I-3
Division of Reactor Projects - 1/11
Office of Nuclear Reactor Regulation
Docket No. 50-309
Enclosures: As stated
)lSTRIBUT'ON
Jocket Fi
e
'
,
PUBLIC
PDI-3 RF
B. Boger
R. Eaton
D. Dorman
E. Peyton
0GC
C. Hehl
P. O'Reilly. AEOD T-4-A9
S. Mays. AE00 T-4-A-9
DOCUMENT NAME:
G:\\DORMAN\\MYS97STD.PRL
,
To rec 16ve a copy of this document. Indicate in the bon: *C' = Copy without attachment / enclosure */f' je Chy
h attachment! enclosure
- N* = No copy
OFFICE
PDI-3/PM,
_
l l PDIV 2/LA
l l (A)D:Pp k3 U
l l
l !
NAME
DDorman Mf>(
EPeyton , Q @
RE_atof i'
!
DATE
$/ 3 /9F V
08/ % /97
I
C4 d /97
I
(
OfflCIAL RECORD C@Y
_
-.
_
.
. - . -
-
. _ _
-.
..
$
.
Maine Yankee Atomic Power Station
Maine Yankee Atomic Power Company
cc w/ encl:
Mr. Charles B. Brinkman
Mr. Robert W. Blackmore
Manager - Washington Nuclear
Plant Manager
Operations
Maine Yankee Atomic Power Station
ABB Combustion Engineering
P.O. Box 408
12300 Twinbrook Parkway, Suite 330
Wiscasset, ME 04578
.
Rockville, MD 20852
'
Mr. Michael J. Meisner
Thomas G. Dignan, Jr., Esquire
Vice-President
Ropes & Gray
Licensing and Regulatory Compliance
One International Place
Maine Yankee Atomic Power Company
Boston, MA 02110-2624
329 Bath Road
Brunswick, ME 04011
Mr. Uldis Vanags
State Nuclear Safety Advisor
Mr. Bruce E. Hinkley, Acting
State Planning Office
Vice-President Engineering
State House Station #38
Maine Yankee Atomic Power company
Augusta, ME 04333
329 Bath Road
Brunswick, ME 04011
Mr. P. L. Anderson, Project Manager
Yankee Atomic Electric Company
Mr. Patrick J. Dostie
580 Main Street
State of Maine Nuclear Safety
Bolton, MA 01740-1398
Inspector
Maine Yankee Atomic Power Station
-
Regional Administrator, Region I
P.O. Box 408
U.S. Nuclekr Regulatory Commission
Wiscasset, ME 04578
475 Allendale Road
King of Prussia, PA 19406
Mr. Graham M. Leitch
Vice President, Operations
First Selectman of Wiscasset
Maine Yankee Atomic Power Station
Municipal Building
P.O. Box 408
U.S. Route 1
Wiscasset, ME 04578
Wiscasset, ME 04578
Mary Ann Lynch, Esquire
Mr. J. T. Yerokun
Maine Yankee Atomic Power Company
Senior Resident Inspector
329 Bath Road
Maine Yankee Atomic Power Station
Brunswick, ME 04011
U.S. Nuclear Regulatory Commission
P.O. Box E
Mr. Jonathan M. Block
Wiscasset, ME 04578
Attorney at Law
P.O. Box 566
Mr. James R. Hebert, Manager
Putney, VT 05346-0566
Nuclear Engineering and Licensing
Maine Yankee Atomic Power Company
329 Bath Roao
Brunswick, ME 04011
Friends of the Coast
P.O. Box 98
Edgecomb, ME 04556
-
.
S
l
...
r
LER No. 309/97-004
.
LER No. 309/97-004
Event Description: Reactor coolant system hot leg recircu'ation valves subject to pressure
locking because of post LOCA thermal expansion of the trapped water
Date of Event: January 22,1997
Plant: Maine Yankee
Event Summary
Maine Yankee was shut down for refuchng when engineers noted a plant design deficiency while conducting
a piping system review in response to Nuclear Regulatory Commission (NRC) Generic Letter (GL) 96 06.
Personnel determined that the coolant trapped between the containment integrity check valve and the loop fill
motor-operated valves (MOVs) could cause the loop fill MOVs to become pressure locked following a loss-
of-coolant accident (LOCA). The LOCA could cause the fluid that is normally trapped between these valves
to expand. This, in tum, could cause the pressure between the valves to exceed the torque available to open
the MOVs (i c., the valves become thermally pressure locked). Hence, without sufficient leakage past the
loop fill MOVs, the valves would be rendered inoperable. The loop-fill MOVs are used for hot leg
recirculation in order to prevent the boron from precipitating in the core. Boron precipitation could lead to
core damage by obstructing coolant flow through the core, thereby reducing the removal of decay heat
(Ref.1). The estimated increase in the core damage probability (CDP) over a 1-year period for this event
(i.e., the importance) is 1.3 = 10
The base probability of core damage (the CDP) for the same period
(i.e.,1 year)is 6.9 * 10~5 Uncertainty in the freluency of a large-break LOCA (none have occurred) and the
likelihood of a MOV failing under post LOCA conditions contributes to the uncertainty in this estimate.
Event Description
On January 22,1997, while Maine Yankee was shut down for refueling, engineers were conducting a piping
system design review in response to a deficiency identified in licensee event report (LER) 309/96 022 (Ref. 2)
and NRC GL 96-06 (Ref. 3). These engineers determined that the section of loop fill piping between the
containment integrity check valve (CH 72) and the loop fill MOVs (RC M 15,25,35) was susceptible to
thermal pressure locking tollowing a LOCA (Fig.1). Emergency Operating Procedures require this section
of piping to be used for hot leg recirculation 19 h aller a large cold-leg break. This is because a large cold-leg
LOCA allows a significant portion of safety injection flow from the cold leg injection paths to bypass the core
by flowing directly out the break. The result is boiling in the core, which causes the boron concentration to
increase. At Maine Yankee, if the boiling persists for more than 19 h, the concentration of boron in the core
can reach the saturation point allowing precipitation to occur. Boron precipitation would further reduce or
obstruct flow through the core and could lead to core damage because of insufficient decay heat removal.
Boron precipitation is prevented by the timely switch to hot leg recirculation. This insures that injection flow
will go through the core before flowing out the break in the cold leg.
Pressure locking occurs when the fluid in the valve bonnet is at a higher pressure than the adjacent piping at
the time of the valve opening. The two most likely scenarios for elevating the pressure in the valve bonnet
relative to the pressure in the valve system are
1
Enclosure 1
_
.
..
.
>
,..
v
-
LER No. 309/97-004
1.
Thermal pressure locking (or bonnet heat up) can occur when an incompressible fluid is trapped
in the vah e bonnet (e g., during valve closure) or a section of piping bounded by a check valve
(as in this case), followed by heating up the volume in the bonnet / piping The bonnet heat up
scenarios include heating the s alve bonnet by an increase in the temperature of the environment
l
during an accident, heat up because of an increase in the temperature of the process fluid on
either side of the vahe, etc. (Normal ambient temperature variation is not considered because
it occurs over a long time and pressure changes tend to be alleviated through extremely small
amounts ofleakage. Further, operating experience shows that normal temperature variations are
l
not a source of pressure locking es ents.)
1
2.
Hydraulic pressure locking (or pressure-trapping) can occur when an incompressible fluid is
I
trapped in the valve bonnet, followed by depressurization of the adjacent piping before valve
opening. Examples of hydraulic pressure locking scenarios include back leakage past check
valves, and system operating pressures that are higher than the system pressure when the valve
is required to open. This did not occur in this event.
Pressure locking is of concern because the pressure in the space between the two discs of a gate valve can
become pressurized abos e the pressure assumed when sizing the vah e's motor operator. This could prevent
the valve operator from opening the vah e when required.
Additional Event-Related information
During the injection phase, Maine Yankee uses two charging pumps for high pressure safety injection (HPSI)
of water from the refueling water storage tank (RWST) into each of the three reactor coolant system (RCS)
cold legs via two cold leg injection trains. A third charging pump is available as a spare, but it must be
manually aligned (Fig.1). During hot leg recirculation, the charging pumps inject water through the loop-fill
,
'
MOVs. A safety injection tank (accumulator)is available to inject water into each cold leg automatically if
RCS pressure decreases below the pressure in its accumulator (about 200 psig). Low pressure safety injection
(LPSI) is provided by two residual heat removal (RHR) pumps, which also pump water from the RWST into
each RCS cold leg. The two RHR pumps are also used for shutdown decay heat removal (Ref. 4).
The containment spray system provides for recirculation and containment cooling. On a containment spray
actuation signal (CSAS), the spray system uses two spray pumps to del"cer water through the RHR heat
exchangers to two separate spray rings located inside containment. The water is supplied from the RWST
or the containment sump following a recirculation actuation signal (RAS). A third spray pump is available
as a spare, but it must be manually aligned. On a RAS, the containment spray pumps provide water from the
containment sump, through the RHR heat exchangers, to the charging pump suction (and the spray rings if
a CSAS exists).
Modeling Assumptions
Boron precipitation was only considered to be a mechanism for core damage in a break of a RCS cold leg
because this allows a significant portion of safety injection water to bypass the core and flow directly out the
break. Large. and medium-break LOCAs currently are not addressed by the models used in the Accident
2
_ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ - _
.
'
...
e
.
LER No. 309/97 004
Sequence Precursor (ASP) Program. Therefore, it was necessary to construct a model specifically for this
analysis. A large break LOCA esent tree was created for the Integrated Reliability and Risk / nalysis System
(IRRAS) model based on the success trees in Maine Yankee's Indwidual Plant Eramination (IPE). The
Maine Yankee IPE estimates the frequency of a large break LOCA to be 2.7 = 10' per year (Ref. 5). The hot
leg fill valves were considered impacted in response to a LOCA during this period (i c., the water is trapped
{
between valves and the increased temperature resulting from the LOCA hests the water) A l year condition
l
assessment was cot.Jucted since this is typically the longest period analyzed by the Accident Sequence
Precursor (ASP) Program llence, a 70% plant availability provides a duration time of approximately 6,130 h
for this event.
Successful response to a large-break LOCA includes either of the following:
l
1.
one HPSI train, one LPSI train, one accumulator, and successful switch-over to cold leg
recirculation that requires at least one containment spray pump. Hot leg recirculation is not
addressed by the IPE; however, the LER for this event indicated that the switch over to hot-leg
recirculation is required 19 h after a large cold leg break.
2.
no HPSI trains, two LPSI trains, one accumulator, and successful switch over to cold-leg
recirculation. Again, switch-over to hot leg recirculation is required 19 h following a large cold-
leg break.
Based on these combinations, a large-break LOCA event tree was constructed (Fig. 2).
l
The event tree for the large break LOCA es ent tree includes the following branches:
IE-LBLOCA. The initiating event is a large-break LOCA. The frequency of a large break LOCA is estimated
to be 2.7 = 10"/p. This value is consistent with a suncy oflarge and medium-break LOCA frequencies
provided in the 1994 ASP report (see Appendix H.6 in Ref. 6 for additional information). A reactor trip is
not a prerequisite for preventing core damage following a large break LOCA because void formations
resulting from boiling terminates the fission process. Therefore, a reactor trip is not included in the event tree.
HPSI. The existing IRRAS fault tree for the HPSI system is used for this event tree branch. A system failure,
by itself, does not lead to core damage. However, most HPSI system components (e.g., pumps, valves,
piping) are required for successful recirculation.
LPSI l The fault tree for this event was constructed assuming that the break occurs in one of the three RCS
cold legs. Therefore, one of three injection paths is assumed to be unavailable with all injection flow from
that path flowing out the break. Success for this branch requires one of the two remaining injection paths to
be available and one of the two available RHR pumps to operate.
IJ'SI-2. The fault tree for this event is the same as the LPSI l fault tree except that both RHR pumps and both
injection trains are required for success.
3
.
__.
'
.
S
.. ,
.o
LER No. 309/97-004
Al. The fault tree for this event was also constructed assuming the break is in one of the RCS cold legs.
Therefore, the contents of one accumulator are assumed to flow directly out the break. Success for this
branch occurs if one of the two intact loop accumulators injects borated water properly. An accumulator is
assumed to operate correctly if the associated accumulator hiOV remains open and the corresponding check
valve opens as designed.
Cold Leg Recirculation. Success for this branch implies that at least one contairunent spray train operates
to supply water to the suction of the HPSI system from the containment sump. At least one HPSI train must
also operate and inject water into one of the two intact RCS cold legs.
ColdLcg Brcok. Success for this branch implies the LOCA occurred in one of the RCS hot legs, because hot
leg recirculation would not be required to prevent core damage following a hot leg break. This analysis
assumes an equal likelihood that the large-break LOCA will occur in a cold leg or hot leg (i e., LEAK IN-
COLD LEG - 0.5).
Hot Leg Recirculation. Success requires at least one of the three loop fill MOVs to open, and one of the two
cold leg injection paths be isolated Because the required components for successful hot leg and cold leg
recirculation are essentially the same (such as HPSI pumps, RWST isolation valves), it is assumed that if cold
leg recirculation failed, then hot leg recirculation will not be successful, leading to core damage.
Predicting with certainty the containment environmcoi 19 h following a large-break LOCA is difficult, The
location of the loop fill valves and the temperature of the c.antainment heat sink will greatly influence the
status of the loop fill MOVs. Additionally, due to the piping :onfiguration, if any one of the loop fill valves
opens, the pressure will be relieved on all three valves. As a result, the failure probabilities for the loop-fill
MOVs (HPR MOV CC HOTA, HOTB, HOTC) were not adjusted (the nominal failure probability for each
4
valve is 3.0 x 10 ). However, the common cause failure probability (HPR-MOV-CF HOT) was adjusted
from its base probability to 0.1 because of the potential for thermal pressure locking at the point in time that
they are required to operate. It was assumed that the operator cannot recover the hot-leg recirculation
configuration if the loop-fill valves failed because of thermal pressure locking. Therefore, the operator non-
recovery probability (HPR XHE NOREC HL) was set to TRUE (1.0).
Analysis Results
The increase in the CDP over a 1 year period for this event is estimated to be 1.3 x 10
The nominal CDP
4
over the same 1 year period for all sequences is 6.9 x 10
There is substantial uncertainty in this estimate
4
because of the uncertainty in the frequency of a large-break LOCA (none have occurred) and the likelihood
of MOVs failing under post LOCA conditions. The dominant core damage sequence for this event (Sequence
3 on Fig. 2) involves the following events:
a postulated large-break LOCA in one cold leg,
-
success of the HPSI system,
+
success of the LPSI system,
+
success of the accumulator system,
-
success of the cold leg recirculation configuration, and
+
a failure of the hot leg recirculation configuration.
+
4
_..
.
e
,,
v
.
LER No. 309/97 004
This sequence accounts for almost 100% of the total contribution to the increase in the CDP. No other
initiating events are affected by a failure of the hot leg recirculation configuration.
-Acronyms
accident sequence precursor
conditional core damage probability
core damage probability
containment spray actuation signal
GL
generic letter
,
high pressure safetyinjection
independent plant examination
IRRAS
Integrated Reliability and Risk Analysis System
LER
licensee event report
loss-of-coolant accident
low pressure safety injection
motor operated s alve
NRC
Nuclear Regulatcay Conunission
recirculation actuation signal
refueling water storage tank
References
1.
LER 309/97-004, Rev. O,"RCS Loop Fill Header MOV Overpressure" February 24,1997.
2. LER 309/96 022, Rev. O," Containment Primary Component Cooling Piping Design inadequacy Due to
Lack of Thermal Relief Valves," August 19,1996.
3. NRC Generic Letter 96 06, " Assurance of Equipment Operability and Containment Integrity During
Design Basis Accident Conditions," September 30,1996.
4.
FinalSafety Analysis Report, Maine Yankee Atomic Power Company
5. Maine Yankee Atomic Power Company,IndividualPlant Examination.
6.- Precursors to PotentialSevere Core Damage Accidents: 1994, A Status Report, NUREG/CR-4674, Vol.
21, December 1995.
i
5
,
- - - - - -
!
,
L
,
t
.
e
' , .
.
-
e
.
yQ
oS3I
%
\\/ )
-
r
hTl R
k
s
n
R
yy
r
' o,
r
1 e
3e
w
e
H
r e
e
pd
pd
- H
r
R
g p
oe
r
o
o n
d
o
uS
oe
.
o e
m
LH
LH
r
me
r
o
oZ
oM
l
T F
oM
o
rF
T P
T F
F
FC
.
-
4b C
m
5
5
5
o
2
3
\\/
1
e
4
t
r
s
5
s
C
re
C
y
.
g
eg
S
t
n
a n
R
R
R
S
v& h s
-
u
g
Wha
n
i
a
len
v
h
H
i
C
eE
t
I
,
ek
St
I
b
1d
o
a
T
o
e
2
Te
cN
7
kn
H
-H
ni
ay
$
Yt
'
_.
1I
-
e
efa
n
iS
Vg
A
V
a
V
~qM
Mreu
.
s
.
s
se
Qh9
rP
g
l
h
,
g
i
$v -
x
igr
1
g
@
H
n
r
r
i
a e
h d
-
e
}
Ce
~
mH
X
AQ
F
o
m_ P
'
!
r
8
S
4
4
4
'
1
1
1
-
-
-
P
P
'
u
-
3
D*'
m
H
.
i
.'
f<>oy9<g
Q!
M
V
i
D
-
-
_
~
~
-
-
M
%
gn
,o
,
li -
ts
ts
-
.o
,
e c
c.
.i
e
loj
3
n
n
ci
/(
2m
y?.R
am
gy ma v . E3
1
n :r : j.oY5
o
e
o
_.
.-
-
-- -
..
.
. . .
-
...
. _ _ -
- *
,.
,
LER No.309/97 004
.
8,I
5
5
8
8
8
8
g
g
g
g
g
g
j
-
a
a
.
.
.
~
.
.
,
,
-
5
ili l
II l
l3
ll11
_i
$lr
a
I! It
5-f
!!
!!! i
a
-
-
ll
!! I
i!! I
11
al
w
Fig. 2 Dominant core damage sequence for LER No. 309/97 004.
7
.
-
- -
-
-
-
.
_-
. -
- _ - . . .. .
.- -
- - - - .
. . . .
-
- - - . . .
.
.
- .
.. .
e
.
LER No. 309/97 004
Table 1. Definitions and Probabilities for Selected Basic Events for LER No. 309/97-004
Modified
Event
Base
Current
for this
name
Description
probability
probability
Type
event
Initiating Event-Large-break
4.4 E 008
4.4 E 008
NEW
No
a
Failure of MOV RC M-l$ (Loop-
3.0 E 003
3.0 E-003
NEW
No
Fill) to Open for flot Leg
Recirculation
'
IIPR MOV.CC-IlOT13
Failure of MOV RC M 25 (Loop-
3.0 E 003
30E@3
NEW
No
Fill) to Open for 110: Leg
1
Recirculation
Failure of MOV RC M 35 (loop.
3.0 E 003
3 0 E-003
NEW
No
Fill)to Open for llot Leg
Recirculation
Common-Cause Failure ofIlot
7.9E 00$
1.0E 001
NEW
Yes
Leg Recirculation MOVs
llPR XilE NOREC IIL
Operator Fails to Recoser 110:
8 0 E 001
1.0 E6
NEW
Yes
Leg Recirculation
TRUE
LEAK-!N-COLD-LEG
Large-Dreat LOCA Occurs in
5.0 E 001
3.0 E-001
NEW
No
One of the Three RCS Cold Legs
8
..
--
_
__
_ . _ _ _ _
.
.
. _ _ .
_ _ _ .
__
__._
.. .
, ,
,
.
L
i
.
LER No.309/97 004
Table 2. Sequence Conditional Probabilities for LER No.309/97 004
h
Conditional
Event tree
Sequence
core damage
Core damage
Importance
Percent
name
number
probability
probability
(CCDP-CDP)
contribution *
(CCDP)
(CDP)
03
1.3 E-005
1.4 E-007
1.3 E-005
99.9
- Total (all sequences)
8.2 E-005
6.9 E-005
1.3 E-005
' Percent contnbution to the total importance.
i
.
Table 3. Sequence Logic for Dominant Sequences for LER No. 309/97-004
l
)
Event tree name
Sequence
Logic
number
03
/HPl, /LPSI 1, /A1, /COLDLEG,
RCSCOLD, HOTLEG
4
Table 4. System Names for LER No.309/97-004
System name
Logic
Al
Failure of I of 2 Accumulators (Assumes a Cold Leg
Break)
COLDLEG
Failure of Cold Leg Recirculation
HOTLEG
Failure of Hot Leg Recirculation
No or Insufficient Flow From the HPSI System
LPSIl
LPSI Fails When 1 of 2 Pumps are Required
RCSCOLD
Large Break LOCA Occurs in One of the Three RCS
Cold Legs
.
9
4
__
_______--_-- .
.
.
..
.
. _ .
..
.*:
,..
, +
-
LER No. 309/97-004
1
Table 5. Conditional Cut Sets for Higher Probability Sequences for LER No. 309/97-004
)
Cut set
Percent
Change in
number
contribution
Cut sets *
(Importanice)*
LBLOCA Sequence 03
1.3 E 005
- o
j
i
97.4
1.3 E 005
LEAK-IN COLD-LEO tIPR MOV CF ilOT. IlPR-X11E-NOREC-IIL
Total (all sequences)
1.3 E-005
m
^
__
'The change in conditional probabihty (importance) is determined by calculating, the conditional probabihty for the penod in which the
condition existed, and subtracting the conditional probabihty for the same penod but with plant equipment assumed to be operstmg
i
nominally. The conditional probabihty for each cut set within a sequence is determined by multiplying the probabihty that the portion
of the sequence that makes the precursor sisible (e g , the system with a failr:is demanded) will occur dunng the duration of the esent
l
by the probabilities of the remaining basic esents in the minimal cut set, This can be approximated by I . e#, wiscre p is determined by
i
multiplying the expected number of initiators that occur dunng the durat4on of the esent by the probabihties of the basic events in that
minimal cut set. The expected number of initiators is given by At, where A is the frequency of the initiating event (gisen on a per-hour
basis), and I is the duration time of the esent. This approximation is senservative for precursors made visible by the initiating event,
De frequency of interest for this event is 6 - 4 4 x 10*1 The daration time for this event is 6,130 h (8,760 h = 0.7)
"flasic esent ilPR XilE-NOREC-Illis a type TRUE event. This type of esent is not normally included in the output of the fault tree
reduction process, but has been added to aid in understanding the requences to potential core damage associated with the event.
10
1
-
. _
-
.
--
.
.
..
1
.
. .
,
,
e.
n
.
'
1
GUIDANCE FOR LICENSEE REVIEW 0F
i
PRELIMINARY ASP *NALYSIS
'
i
Background
The preliminary precursor analysis of an operational event that occurred at
your plant has been provided for your review. This analysis was performed as
a part of the NRC's Accident Sequence Precursor (ASP) Program.
The ASP
Program uses probabilistic risk assessment techniques to provide estimates of
4
operating event significance in terms of the potential for core damage. The
types of events evaluated include actual initiating events, such as a-loss of
off site power (LOOP) or loss-of-coolant accident (LOCA), degradation of plant
conditions, and safety equipment failures or unavailabilities that could
increase the probability of core damage from postulated accident sequences.
-
This preliminary analysis was conducted using the information contained in the
plant-specific final safety analysis report (FSAR), individual plant
examination (IPE), and the licensee event report (LER) for this event.
Modeling Techniques
.
The models used for the analysis of 1995 and 1996 events were developed by the
Idaho National Engineering Laboratory (INEL).
The models were developed using
the Systems Analysis Programs for Hands on Integrated Reliability Evaluations
(SAPHIRE) software.
The models are based on linked fault trees.
Four types
of initiating events are considered: (1) transients, (2) loss-of-coolant
4
accidents (LOCAs), (3) losses of offsite power (LOOPS), and (4) steam
generator tube ruptures (PWR only).
Fault trees were developed for each top
event on the event trees to a supercomponent level of detail.
The only
i
support system currently modeled is the electric power system.
,
The models may be modified to include additional detail for the systems /
components of interest for a particular event. This may include additional
'
equipment or mitigation strategies as outlined in the FSAR or IPE.
Probabilities are modified to reflect the particular circumstances of the
,
event being analyzed.
l
Guidance for Peer Review
Comments regarding the analysis should address:
Does the " Event Description" section accurately describe the event as it
e
occurred?
-
Does the " Additional Event-Related Information" section provide accurate
e
additional information concerning the configuration of the plant and the
'
operation of and procedures associated with relevant systems?
Does the "Modeling Assumptions" section accurately describe the modeling
e
done for the event?
Is the modeling of the event appropriate for the
events that occurred or that had the potential to occur under the event
conditions? This also includes assumptions regarding the likelihood of
equipment recovery.
Enclosure 2
.
.. - ,
.
, -
s
e
Appendix H of Reference 1 provides examples of comments and responses for
previous ASP analyses.
Criteria for Evaluating Consents
Modifications to the event analysis may be made based on the comments that you
'
provide.
Specific documentation will be required to consider modifications to
the event analysis.
References should be made to portions of the LER, AIT, or
other event documentation concerning the sequence of events.
System and
component capabilities should be supported by references to the FSAR, IPE,
plant procedures, or analyses. Comments related to operator response times
and capabilities should reference plant procedures, the FSAR, the IPE, or
applicable operator response models. Assumptions used in determining failure
probabilities should be clearly stated.
Criteria-for Evaluating Additional Recovery Neasures
Additional systems, equipment, or specific recovery actions may be considered
for incorporation into the analysis. However, to assess the viability and
effectiveness of the equipment and methods, the appropriate documentation must
be included in your response.
This includes:
normal or emergency operating procedures.'
-
piping and instrumentation diagrams (P& ids),'
-
electrical one-line diagrams,'
-
results of thermal-hydraulic analyses, and
-
operator training (both procedures and simulator),' etc.
-
Systems, equipment, or specific recovery actions that were not in place at the
time of the event will not be considered. Also, the documentation should
address the. impact (both positive and negative) of the use of the specific
recovery measure on:
,
the sequence of events,
-
the timing of events,
-
the probability of operator error in using the system or
-
equipment, and
other systems / processes already mod 61ed in the analysis (including
-
operator actions).
For example, Plant A (a PWR) experiences a reactor trip, and during the
subsequent recovery, it is discovered that ond train of the auxiliary
feedwater (AFW) system-is unavailable. Absent any further information
regrading this event, the ASP Program would analyze it as a reactor trip
with one train of AFW unavailable. The AFW modeling would be patterned
after information gathered either from the plant FSAR or the IPE.-
However, if information is received about the use of an additional
system (such as a standby steam generator feedwater system) in
recovering from.this event, the transient would be modeled as a reactor
trip with one train of AFW unavailable, but this unavailability would be
- Revision or practices at the time the event occurred.
.
!
.
l
.
e
,
v.
e
.
mitigated by the use of the standby feedwater system.
The mitigation
effect for the standby feedwater system would be credited in the
analysis provided that the following material was available:
standby feedwater system characteristics are documented in the
-
FSAR or accounted for in the IPE,
procedures for using the system during recovery existed at the
-
time of the event,
the plant operators had been trained in the use of the system
-
prior to the event,
a clear diagram of the system is available (either in the FSAR,
-
IPE, or supplied by the licensee).
previous analyses have indicated that there would be sufficient
-
time available to implement the procedure successfully under the
circumstances of the event under analysis,
the effects of using the standby feedwater system on the operation
-
and recovery of systems or procedures that are already included in
the event modeling.
In this case, use of the standby feedwater
system may reduce the likelihood of recovering failed AFW
equipment or initiating feed-and-bleed due to time and personnel
constraints.
Materials Provided for Review
The following materials have been provided in the package to facilitate your
review of the preliminary analysis of the operational event.
The specific LER, augmented inspection team (AIT) report, or other
e
pertinent reports,
o
A summary of the calculation results. An event tree with the dominant
sequence (s) highlighted. Four tables in the analysis indicate:
(1) a
summary of the relevant basic events, including modifications to the
probabilities to reflect the circumstances of the event, (2) the
dominant core damage sequences, (3) the system names for the systems
cited in the dominant core damage sequences, and (4) cut sets for the
dominant core damage sequences.
Schedule
Please refer to the transmittal letter for schedules and procedures for
submitting your comments.
References
1.
L. N. Vanden Heuvel et al., Precursors to Potential Severe Core Damage
Accidents: 1994, A Status Report, USNRC Report NUREG/CR 4674 (ORNL/N0AC-
232) Volumes 21 and 22, Martin Marietta Energy Systems, Inc., Oak Ridge
National Laboratory and Science Applications International Corp.,
December 1995.
4'
. . . . . . . . . . . . . . -
_ _
.- ._
.
.
.-
.
a$
,
e.
.*
%
4
i
MaineYankee
PE LI ABL E E LE CTRICliv SINCE 1972
e
329 BATH ROAD * BRUNSWICK, MAINE 04011 * (207) 7984100
February 24,1997
MN-97-36
JRH-97-45
.
UNITED STATES NUCLEAR REGULATORY COMMISSION
.f
Attention:
Document Control Desk
Washington, D. C.
20555
,
Reference:
(a)
License No. DPR-36 ( Docket No. 50-309 )
Subject:
Maine Yankee Licensee Event Report 97-004, RCS Loop Fill Header MOV
Overpressure
Gentlemen:
Please find enclosed Maine Yankee Licensee Event Report 97-004. This report is submitted
in accordance with 10 CFR 50.73(a)(2)(ii).
Please contact us should you have any questions regarding this matter.
Very truly yours,
% q. w'
d
'
James R. Hebert, Manager
'
Licensing & Engineering Support Department
mwf
..
Enclosure
c:
Mr. Hubert Miller
i
p 2 h ji
Mr. J. T. Yerokun
Mr. D. H. Dorman
Mr. Patrick J. Dostle
Mr. Uldis Vanaes
."J
N
6M
[
-
Enclosure 3
@.IlE55555.I!
. . . _ _ .
_ _ _ _ _ _ _ _ . _ _ _ _ . - . . . _ . _
. . _ _ _ . . _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _
1;_,,=to
'
- "
'
.
.
b
'
U.S. NWLEAR REGt'LATORY COM 415510N
APPROVED 8Y OMS too. 31504104
+
l
84 4 s)
ExmREs 04t30res
,
.
!,'*'.l"JE'"."*.".%."#8'nai ,*,eW.W ,'=
.
!
LICENSEE EVFNT REPORT (LER)
Ec2*eIS AIE!."! "^"b a"c'. Nr.v E,',!.3
"'" ^
comuswts neoanoMou
J
estemare To twe e,ponesavica
is - 'a r.***d =*= a'
3%."i*caraf.alM."'." " van /httelrai
'
digli % erectors for sech idocM
to rue Paremv0nu neoucTi0ss en0JacT isiso4toet Oppice Os
i
eAcany maae its
oocatimuussa as
pam ian
Maine Yankee Atomic Power Company
50 309
1OF3
,
l
mu .
RCS Loop Fill Header MOV Overpressure
eve = =n .e.
u. =u
e. .e.
om un n,
or. .
n.. .,v.v.. .e.
_
te
M
,
es0 NTH
DAY
VSAA
VSAR
teoNTH
DAY
YEAR
t
I
"*" "
- * * " " " " "
01
22
97
97 -- 004
00
02
24
97
--
!
oPERAT9fo
7 4 s REP RT is susMiTTE ) m insUANT TO THE REQUIRERIENTs oF 10 CFR 0; (Check ene er eneral till
3
- .
IAoOE ist
20.2201(b)
20.2203(eH2Hvl
50.73(eH2Hil
50.734eH2Hval
powg'
- #**"
- " "'""
"
'U7*#""
- * ' ' ' ' ' " ' " ' '
l --
i
LEVEL h'01
0
20.'2203teH3Hul
60.73teH2Hel
73.71
20.2203(eH2Hil
!
20.2203(aH2HE)
20.2203(aH4)
50 73faH2Hivl
OTHER
,
{-
(
YkV '
20.2203teH2Het
50.3SicH1)
50.73teH2Hvl
g
gw
e
20.2203teH2Hivl
50.36te H2)
50.73(eH2Hvsl
m-e
i
I
LICENSEE CONTACT FOR THis LEPl 112)
j
mane
nua.o.m Nuraan
A c .
George N. Stowers, Senior Nuclear Safety Specialist
(207) 882-5749
Conspla15 oNE LINE Fo. EACM COMpofeE8f7 744Lud. DESC4WD us THIS SpostY #13
- gk
M',',',",,aeu
7n,,"g"g8
-
caust
sveTem
coes.osewt
aanssupacTunsa
CAuee
systges
conspomewT
tammupACTumn
y
gg
..
Afy
_
SUPPLEMENT AL liEPoRT EXPECTED (191
"*""
"
"
EXPECTED
8 $Y[gIh"
'
YES
X NO
0
AssTRACT lumet to 1400 spaces,I,s.. eppro.6metely 15 eingle spaced typewritten uneel (10)
'
On January 22,1997 Maine Yankee was in a Cold Shutdown condition.
While conducting a review of plant design in response to a design deficisacy identified in LER 96-
022, Containment Primary Component Cooling Piping Design inadequacy Due to Lack of Thermal
Relief Valves, and subsequent Nuclear Regulatory Commission Generic Letter 96 06, Assurance
Of Equipment Operability And Containment integrity During Design basis Accident Conditions,
engineers noteC that under post LOCA conditions, sections of piping associated with the Reactor
Coolant Loop Fill Header could become pressurized in excess of design' pressure due to thermal
expansion of fluid trapped in the header.
As the conditions postulated to cause over pressurization only exist post LOCA, and the plant is
currently in a Refueling shutdown condition, no immediate corrective / compensatory action is
rsquired. Relief valve protection will be installed in the header prior to plant startup.
anc conu ase i44si
-
. - -
. . - - - - - - - - - - . ~ . ~ . -
. ~ - - . ~ . .
- - - -
. - . ~
- - - - -
'
.
ee
,
4'
""
.
.'
U,s, NUCLEAR REGULATORY CoMMessioW
.
1446)
+
!
'
LICENSEE EVENT REPORT (LER)
-
TEXT CONTINUATION
FACILITY NAME M)
DoCIE1
LER MU"R I l)
PAGE(3)
_,
vsA
seoUen rat
Arve
.
Maine Yankee Atomic Power Company
50 309
2
OF
3
l
97
004
0
-
-
,
TEXT (# aieve space k W. use addeleant eerdee et W4c Fe,m .966N (171
!
INJTIAL PLANT CONDITIONS:
,-
On January 22,1997 Meine Yankee was in a Cold Shutdown condition.
!
,
'
j
EVENT DESCRIPTION:
'
l
While performing a review of containment penetrations sa a follow up to LER 96 022,
j
Containment Primary Component Cooling Piping Design insdequacy Due to Lack of Thermal Relist
Valves, and subsequent Nuclear Regulatory Commission Generic Letter 96 06, engineers noted
that certain sections of the Reactor Coolant (AB) Loop Fill piping are potentially susceptible to
j
over pressurization due to thermal expansion of trapped fluid post LOCA (Loss of Coolant
Accident). At Maine Yankee, Emergency Operations Procedures call for initiating hot leg
i
l
r: circulation after 19 hours2.199074e-4 days <br />0.00528 hours <br />3.141534e-5 weeks <br />7.2295e-6 months <br /> after a postulated large cold leg break to prevent the likelihood of
I
!
boron precipitation in the core.
!
Boron precipitation results when a cold leg LOCA allows safety injection flow to bypass the core
!
r:sulting in boiling in the core. If this condition is allowed to persist for greater than 19 hours2.199074e-4 days <br />0.00528 hours <br />3.141534e-5 weeks <br />7.2295e-6 months <br />,
j
the concentration of boron in the core can increase to the point where the concentration of boron
l
reaches the saturation point for existing conditions and precipitation occurs. The resulting further
r: duction in flow through the core caused by the buildup of boron precipitate can lead to
,
insufficient decay heat removal and eventually core damage. At Maine Yankee, boron
precipitation is prevented, by opening loop fill motor operated valves (MOVs) (RC M 15, 25 & 35)
!
which are located inside containment. Opening these valves creates a flow path that forces water
j
through the core thereby cooling the core, and diluting the baron concentration.
!
As a result of preliminary evaluations it has been concluded that the loop fill MOVs could be
-
!
subjected to increased internal prescure, during post LOCA, due to thermal expansion of fluid
l
tr:pped between the loop fill MOVs and containment integrity check valve (V) CH 72. Based on
i
cngineering judgement it has been determined that this pressure increase could render the loop fill
v lves inoperable either by physically damaging the valves or hydrostatically locking the valve
!
disks in place. The actual maximum pressure would likely be lower due to leakage past the
!
boundary valves.
<
j
SAFETY SIGNIFICANCE:
The conditions postulated to cause over pressurization in the loop fill header only exist post
!
LOCA. As the plant is currently in a Refueling Shutdown this condition does not represent an
Immediate threat to nuclear safety.
i
NRC POPM 366A 14 SGI
J
,. .
.
_ _ - - - - . - -
- - . - . . . . - . . - - -
- _ - . - _ -
_ = - - . .
_ - - - - - . - .
_ - -
.
.l.?.
~ '
4
.
Nnc FonM 34sA
d4-961
U.S. NUCLEAR REQULAToRY CotMAtsStoN
LICENSEE EVENT REPORT (LER)
{j
~
TEXT CONTINUATION
,
FACILITY MAnat til
DOCKET
LER Numansaga)
pAgg (3)
,
i
YEA
stoOWiTIAL
MEYtst
Maine Yankee Atomic Power Company
50 309
3
0F
3
4
_
_
TEKT Wme,e space k oonmaket une annesonet cooine of NRC form 366A) l118
Maine Yankee's existing design basis requirer the ability to perform hot leg injection. In the event
l
cf a large cold leg LOCA, hot leg injection could have potentially been rendered inoperable due to
over pressurization of the fill header. Under such conditions there is the potential that ECCS
,
]
cooling of the core could have been partially degraded by precipitation of boron in the core.
.
4
CAUSAL FACTORS:
>
i
The cause of this condition appears to be inadequate consideration of the potential for over
!
pressurization caused by thermal expansion of trapped fluid under post LOCA conditions when
Maine Yankee was originally designed in the late 1960s.
'
,
j
CORRECTIVE ACTIONS:
P
ll
As the conditions postulated to cause over pressurization only exist post LOCA, and the plant is
- ,
currently in a Refueling shutdown condition, no immediate corrective / compensatory action is
r: quired. Pressure relief valve protection will be installed prior to plant startup.
!'
I.
PREVIOUS SIMILAR EVENTS:
l
i
The following two LERs describe previous instances of inadequate over pressure protection.
i
LER 96-022, Containment Primary Component Cooling Piping Design inadequacy Due to Lack of
Thermal Relief Valves
!
LER 95 012-01, RHR Spring Reliefs inadequate for Low Temperature Over Pressure Protection.
,
,
i
i
b
I.
3
i
j
-
NRC PORM 344A (4 95)
_.
.
_
.
. - .
-.