ML16341F480
| ML16341F480 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 12/13/1989 |
| From: | Mendonca M NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V) |
| To: | |
| Shared Package | |
| ML16341F479 | List: |
| References | |
| 50-275-89-31, 50-323-89-31, EA-89-241, NUDOCS 9001030202 | |
| Download: ML16341F480 (32) | |
See also: IR 05000275/1989031
Text
U. S.
NUCLEAR REGULATORY COMMISSION
REGION
V
Report Nos:
50-275/89-31
and 50-323/89-31
EA No.:
89-241
Docket Nos:
50-275
and 50-323
License
Nos:
DPR-80 and
Licensee:
Pacific
Gas
and Electric Company
77 Beale Street,
Room
1451
San Francisco,
California 94106
Facility Name:
Diablo Canyon Units
1 and
2
Inspection at:
Diablo Canyon Site,
San Luis Obispo County, California
Inspection
Conducted:
October
17 through
December
11,
1989
Inspector:
Approved by:
P.
P. Narbut, Senior Resident
Inspector
Yi. N, Nendonca,
Chief, Reactor Projects
Section
1
Date Signed
Summary:
Ins ection fror, October
17 throu
h December
11,
1989
(Re ort Nos. 50-275/89-31
and 50-323/89-31
Areas
I*us ected:
The inspection
was
a special
inspection
conducted
pursuant to the discovery of
debris in the Unit I containment recirculation
by the
NRC inspector.
Inspection
procedures
30703,
37700,
37828,
40500,
61726,
62703,
and
71710 were
used
as guidance
during the conduct of this inspection.
Results of Ins ection:
This inspection identified several. potential violations of regulatory
requirements
with respect
to the containment recirculation
( sump).
The
potential violations identified include:
Failure to assure
that there
were
no gaps
in the Unit
1
sump screen
configuration.
Failure to assure
sump screen
configurations
consistent
with design
requirements.
Failure to assure
that the Units
1 and
2 sumps
were free of debris.
Failure to assure
that the Units
1 and
2 sumps
were not opened during
power operations.
Failure to update
the
FSAR.
Failure to report adverse
conditions
found during operation.
Additional concerns
which warrant licensee
management
evaluation:
The need to assure
consistently conservative
engineering
decisions.
Failure to identify sufficient acceptance
criteria for assuring
design
change construction
met design
requirements.
Failure to update
design drawings.
The
need to improve plant staff awareness
of the importance of
verification signatures.
DETAILS
Persons
Contacted
- J.
D.
Townsend,
Plant Manager
"D.
M.
Miklush, Assistant Plant Manager,
Operations
Services
"M. J.
Angus, Assistant Plant Manager,
Technical
Services
~B.
M. Giffin, Assistant Plant Manager,
Maintenance
Services
- M. G. Crockett, Assistant Plant Manager,
Support Services
~M.
D. Barkhuff, Acting equality Control Manager
"D. A. Taggert, Director equality Support
"M. J. Kelly, Regulatory Compliance Engineer
H. J. Phillips, Work Planning Manager
R. Washington,
Acting Instrumentation
and Controls
Manager
"J.
A. Shoulders,
Onsite Project Engineering
Group Manager
M.
E.
Leppke,
Engineering
Manager
S.
R. Fridley, Operations
Manager
~E.
C. Connell, Project Engineer
"M. T.
Rapp,
Chairman,
Ons1te
Safety
Review Group
"P. J. Roller,
System
Engineer
The inspectors
interviewed several
other licensee
employees
including
shift foremen
(SFM), reactor
and auxiliary operators,
maintenance
personnel,
plant technicians
and engineers,
quality assurance
personnel
and general
construction/startup
personnel.
"Denotes
those attending the exit interview on December 8,
1989.
2.
General
Overview
On October
17,
1989,
the inspector
inspected
the Diablo Canyon Unit 1
containment recirculation
sump.
The Unit was
shutdown for refueling and
in Mode 6.
The containment
had been temporarily
covered with
plastic sheeting
as
a cleanliness
control
measure for the outage.
Consequently,
an inspection of the outer surface
was not possible
at the
time.
The inspector's
examination of the internals of the
sump revealed
debris (including a hacksaw blade
and
a
2 foot by 2 foot cloth) and
an
as-built condition different from that described
by the applicable
licensee
drawing and from that described
in the Final Safety Analysis
Report
(FSAR).
Specifically, the drawing and
FSAR show
a
sump with an
inner 3/16 inch mesh screen
in a "pup tent" arrangement
over each of the
two Emergency
Core Cooling System
(ECCS) suction pipes.
The inner screen
was found not to be installed in Unit 1.
The entire
sump is enclosed
in
steel
grating which is covered with an outer screen of 3/16 inch mesh.
On November
26,
1989,
the outer screen
was found to have openings,
tears
and gaps
(up to
1 inch wide by 5 foot long),
when the temporary
cleanliness
coverings
were
removed
and the outer
screen
and structure
were inspected
by licensee
personnel.
The debris
found in the
and the missing inner screen
coupled with
gaps in the outer screen
for Unit 1 degraded
both trains of the emergency
2
core cooling system for the containment
sump recirculation
mode function.
The licensee
corrected
the configuration deficiencies of the Unit 1 sump,
cleaned
the debri-s
from the
and verified by remote visual
examination (television)
and radiography that the emergency
core cooling
system piping (potentially affected
by the
sump debris)
was in fact
clean.
The licensee
also examined the Unit 2 sump
and found its as-built
configuration
as specified in the
FSAR and applicable
drawings.
The
licensee
found the Unit 2 sump to be clean but subsequent
inspection
by
the resident
found a minor amount of debris
(such
as
a one-half inch
tubing end cap) which was evaluated
as acceptable
by the licensee
due
to'ts
location, size,
and the presence
of the Unit 2 inner screen.
The
inspec'tor considered
the licensee's
evaluation 'to be acceptable.
The
debris
was not removed
because
the Unit was in an operational
mode which
precluded
opening the
sump door.
The Unit 2 emergency
core cooling pipe
potentially affected
by debris
was
examined
by radiography:
One nut was
found in the pipe and allowed to remain
when its size
was determined to
be small
enough to pass
through the residual
heat
removal
pump without
causing
damage,
and to be large
enough to be captured in the residual
heat
removal
(RHR) heat exchanger
and therefore
not detrimental
to
downstream
components.
The licensee
has instituted corrective actions to
improve and expand the existing procedural
instructions for verification
of the containment
sump cleanliness.
In addition,
as
a result of examinations
of the work done in the
sump, it
was determined that the Unit 1 containment
sump,
which has
an access
door
approximately
3 foot by 3 foot,
was
opened twice during the last
operating cycle while in operating
modes requiring emergency
core cooling
system operability.
Unit 2 likewise had its
sump access
door opened at
power at least
one time.
The
had
been
opened
on these
occasions
for approximately
12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> periods to perform corrective maintenance
on
sump level indicators,
and there is
no documentation
that personnel
were
continuously stationed at the
sump location.
The breach of the
barrier defeated
the debris protection function of the containment
making the two trains of the emergency
core cooling system potentially
inoperable for greater
than the technical specification
maximum allowed
time of one hour.
Se
uence of Events
o
June
1974,
the Nuclear Regulatory
Commission
(NRC) issues
Regulatory
Guide 1.82.
The guide
recommends
redundant
Oiablo Canyon's
design
was
90X complete at the time.
A licensee
analysis of the
Regulatory Guide, Pacific Gas
and Electric
(PG&E) file 140. 150 dated
July 25,
1974, states
the
sump is
a single structure divided at
center line. It states
further that if one half of. the fine inner
screen
develops
a hole and thus
becomes
ineffective the redundant
system is still operable.
o
September
18,
1975,
Supplemental
Safety Evaluation Report
(SSER)
No.
3 was issued
by the
NRC.
In it, the
NRC staff concluded that the
design of the containment
was acceptable.
The
NRC staff had
studied
the design at that time which consisted
of a coarse
outer
screen
(1/2 inch) and fine mesh inner screen
(3/16 inch) with
separate
cages
over each
pump suction.
In September
1980, Mestern
Canada Hydraulic Laboratories
Ltd.
performed
a study for
PGRE to assess
the probability of vortexing in
the containment
sump.
The configuration studied
was that of a
coarse
outer screen
(1/2 inch) and
a fine inner screen
(3/16 inch)
as originally designed.
The study concluded that the Diablo design
was satisfactory,
provided the inner screens
did not become heavily
blocked by debris'he
report
recommended
that the fine screens
be
moved to the outer structure.
o
February
1981,
design
change
DCO-EC-908 was issued to remove the
inner 3/16 inch mesh
screen
from both units and to add
a 3/16 inch
mesh screen
to the outside trash rack.
This was accomplished
on
Units
1 and
2 and represented
a departure
from the
NRC reviewed
configuration of the
sump.
o
On September
9,
1981,
a license for 5X power was issued for Unit l.
On November 19,
1981, the license
was
suspended
due to the
identification of improper seismic
design inputs.
o
Per the licensee,
in September
1984
an
FSAR update
was submitted
which erroneously left indication of an inner screen
on the
contrary to the
1981 design
change.
o
November 2, 1984,
NRC issued
a Full Power License for Unit l.
o
November
7,
1984, Unit 1 achieved initial criticality.
o
April 23,
1985,
NRC issued
a
Low Power License for Unit 2.
o
April 24,
1985,
a Diablo problem control sheet
(DP-2-1193-C)
was
written noting gaps in the Unit 2 recirculation
sump.
o
On June
20,
1985,
the design
change for Unit 2 (DC2-SC-32446
Revision 1) issued to repair
the gaps
was signed off as complete
by
the field engineer.
The design
change
completion also reinstalled
the inner screen
assembly
due to the earlier fai lure to revise the
drawing specification for the inner screen.
Unit 2 now was in a
configuration
as
shown in the
FSAR.
August 20,
1985, Unit 2 achieved initial criticality.
On August 2,
1985,
an action request
(AR A3520) was issued to
inspect
the Unit 1 containment
sump for gaps greater
than 3/16 inch,
as
a result of the findings in Unit 2.
The inspection
was limited
to the problem areas
found -in Unit 2 and did not identify the gaps
later found in 1989.
The inspection resulted in the identification
and repair of two spots
on the
sump ends.
The personnel
did not
identify the absence
of the inner 3/16 screen
assembly
on Unit 1,
which had just been reinstalled
in Unit 2.
Unit 1 remained
different than the design
drawing and the
FSAR,
and apparently
had
outer screen
gaps at this time.
o
-May 1986,
SSER
No.
33 was issued evaluating allegations.
Allegation
No.
100 dealt with the lack of quality control for painting in the
containment.
The
NRC staff calculated
that the paints inside
containment
could generate
28.2 cubic feet of debris
in Unit 1.
The
NRC staff found the potential
debris acceptable
due to the
likelihood that paint debris
would not enter the
sump nor cause
loss
of "suction if it collected at the bottom of the outer
sump screen.
The
FSAR description of the
sump at that time showed
a containment
sump with an inner and outer fine screen of 3/16 inch and
a
construction
requirement of a maximum gap of 3/16 inch anywhere
on
the surface of the
sump.
o
June 30,
1987, temporary procedure
TP 8706, Revision 0,
was reviewed
by the Plant Staff Review Committee
(PSRC).and
accepted.
The
procedure
allowed the opening of the unit sumps
and adding of
borated water to check the accuracy of sump level transmitters'.
The
procedure
was approved for use
on October 6, 1987, after
a
10 CFR 50.59 evaluation.
The 50.59 evaluation concentrated
on the effects
of adding borated water to containment
including such
items
as the
effect of reduction of containment
volume
by the addition of 5000
gallons of water to the, sump.
Neither the 50.59 nor the
PSRC
apparently
recognized
the potential effects
on
sump operability with
the
sump hatch
open.
o
October
12,
1987, while the reactor
was in Mode 1, the Unit 2
containment
sump hatch
was
opened for approximately
12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> to
calibrate
the
sump level dete'ctor using
hoses
and
pumps in the
open
8706
was used;
work order
C0022698.
June
16,
1988, while in Mode 5, the Unit
1 containment
was
verified clean, free of structural distress
and corrosion at the end
of the second Unit
1 refueling outage,
in a'ccordance
'with the
technical specification
and the licensee's
procedure
STP-M45
Revision 6, Containment Inspection.
The sign-off for the inspection
was completed
by
a licensed Senior Reactor Operator
(SRO)
and
mechanical
maintenance.
The
was
inspected
from the outside of
the outer screen.
o
July 5,
1988, while in Mode 3, the Unit
1
was
inspected
again
from the outside
and verified clean
by
a different licensed
individual.
o
September
7,
1988, the Unit
1 containment
was
opened for about
12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> while in Mode
1 to perform
a calibration of sump level
transmitter
LT940, in accordance
with work order
C 0038090.
PSRC
approved
temporary procedure
TP 8706 was
used to fill and empty the
sump with borated water.
During this time,
hoses
and
a
pump with
temporary lines were in the
and
passed
through the
sump door.
On September
8, 1988,
the
was verified clean.
The shift foreman
had authorized
contairment entry at 1:55 p.m.
on
September
7,
1989.
The pumping equipment
was
logged
as
removed at
1:30 a.m.
on September
8,
1989,
Licensee
personnel
at the exit
stated all such fill and
pump down operation would-be expected
to
take about
12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> with the
sump hatch open.
o
Hay 11,
1989, while the reactor
was at 100K power operations,
the
Unit 1 containment
sump was again
opened for about
a 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> period
to calibrate
sump level transmitter
LT940 per work order
C 0053302.
A verification of sump cleanliness
from the outside
was performed
by
maintenance
personnel
in this case.
The unit logs
show the level transmitter declared
inoperable at 5:42
p.m.
Based
on the conditions
found by the inspector
on October 17,
1989,
the
sump must have
had debris in it at the time of the
May 11,
1989
licensee verification of cleanliness
discussed
above.
o
October
17,
1989, the
HRC inspector
identified the absence
of the
Unit 1 inner
sump screen
and the presence
of debris inside the outer
sump screen consisting of:
one hacksaw blade,
12 x 1/2 x 1/32 inches,
one wiping cloth, yellow,
2 x 2 feet,
one piece of duct tape,
3 1/2 x 3 1/2 inches,
and
one piece of metal
banding material
12 x 3/4 x 1/64 inches.
o
October 19,
1989,
the Unit 2 contai'nment
was examined
by the
licensee's
system engineer
from outside
the
sump.
The
was
noted to have the inner screen installed.
The
was found to be
free of debris.
o
October
19,
1989, the licensee
found that in February
1981,
design
change
OCO-EC-908
removed the inner screen
from both units.
The
design
drawing was not properly updated
and the inner screen
was
still shown in one View.
Consequently
as previously discussed
the
screen
was reinstalled
in Unit 2 in 1985.
o
October 21,
1989, the licensee
issued
a justification for continued
operation for Unit 2, recognizing the presence
of the inner screen.
Although this screen
was in accordance
with the
FSAR, the presence
of the screen
was not in accordance
with the 1981 design
change,
DCO-EC-908.
o
October
22,
1989,
the system engineer
performed
a walkdown of the
Unit 2 general
containment
area
looking for general. debris.
The
unit was at full power.
Although the areas
were generally clean,
small articles
were found at various locations
throughout the
containment.
Items included gasket material,
tape,
broken seal
material,
a paper towel, paint chips,
and other potential
flotsam.
The condition of Unit 2 at power is typical
and can
be expected
to
represent
that of Unit 1 as well.
Unit 1 was in a refueling outage,
therefore
the normal state of cleanliness
could be inferred from
that found in Unit 2.
o
November 1, 1989,
the
NRC Resident
Inspector in conjunction with a
licensee
health physics technician
examined the Unit 2 containment
sump which had been declared
clean
by the
PGEE staff.
The inspector
and the technician
found
a utility knife,
a tubing support bracket,
and
a 1/2 inch tubing cap.
All observed
items were larger than 3ll6
inch and.were
judged to be such that they ~ould be stopped
by the
inner
screen
which was installed
on Unit 2.
o
November 2, 1989,
licensee
personnel
inspected
the Unit 1 sump
suction piping for debris using
a television (T.V.) camera.
No
debris
was found.
o
November 4, 1989, Unit 2 recirculation
sump suction piping was
examined
by radiography.
One item was found in the horizontal
run
of piping.
The unit was in Mode 1 power operation
and the item
could not be removed without making
a sump entry, which would
potentially make both trains of the emergency
core cooling system
The item was. determined to be
a low density nut
(aluminum or plastic) of sufficient size to be caught in the
heat exchanger
and not pass
through.
The items presence
was
evaluated
to be acceptable
for continued operation
by the licensee.
Subsequently,
on November
5, the licensee
issued
a justification for
continued operation for Unit 2.
o
November 16,
1989,
Mestinghouse
provided to
PG8E
a containment
s'ump
debris evaluation (.letter
PGE 89-811) for the hacksaw blade,
wipe
cloth, 'duct tape,
and banding material
found in Unit 1.
The study
assumes
the entrance of the debris cannot
be precluded,
that only
one train of RHR is available
due to a single failure.
The study
concludes that the
RHR pump would likely not be affected.
However,
the wiping cloth and tape
(588 sq.
inches)
could significantly block
flow in the
RHR heat
exchanger
whose inlet tube sheet is 400 sq.
inches.
The cloth has
some porosity but the beneficial effect of
this was not quantified.
The study also indicated that the
RHR heat
exchanger
may have tube degradation
caused
by a piece of hacksaw
blade.
Mestinghouse
determined that other debris'pieces
could pass
through the 0.527 inch diameter
RHR heat
exchanger
tubes.
Mestinghouse
determined that this debris would not cause failure or
binding of the downstream safety injection (SI) or charging
pumps;
however, restart capability might be affected.
Mestinghouse
considered
small valves
(2 inch and smaller)
may become
blocked.
Mestinghouse's
engineering
judgement regarding the fuel was that any
debris trapped would not cause significant core flow blockage.
The study concludes
that the major system
impacts identified were
degraded
RHR flow due to heat exchanger
blockage,
degraded
ECCS flow
due to blocked valves, 'and degraded
hot leg recirculation flow due
to the inability of the SI pump to restart.
The subsequent
consequences
were evaluated
also, indicating long term containment
pressure
could increase significantly and increased
containment
temperatures
could affect environmental
qualifications.
Additionally, flow and
NPSH may be inadequate
causing catastrophic
pump
damage.
o
On November 17,
1989,
a Nuclear Operations
Support
memorandum to
Oiablo Canyon
Power Plant
(DCPP) Support Services
was issued.
The
memorandum
states
that an analysis of the risk significance of the
unavailability of the containment
sump during power operation
was
performed
by probabi listic risk methods.
The results
stated
the
increase
in the annual
core
damage
frequency
was approximately 9.2
E-8 for each
hour the
sump is unavailable
compared to"the total core
damage
frequency of 2.0 E-4 per year.
o
On November 18,
1989,
licensee
personnel
concluded
and told the
NRC
inspector that the expected
post
Loss of Coolant Accident (LOCA)
water level in containment
would be
6 feet
5 3/4 inches.
The height
of the outer
sump screen
is
5 feet
2 inches
per the as-built
drawing.
Therefore,
the top 'of the
sump would be under water and
an
open
sump screen
door could expose
the
RHR pump suction pipes to
debris,
should
a
LOCA occur.
o
November
21,
1989,
a letter was issued
from the
PG8E Project
Engineer to the Assistant Plant Manager for Technical
Services
stating that it was unlikely that the debris
found in the Unit 1
sump would have
a significant impact on safety,
based
on a review of
the Mestinghouse
assumptions
which apparently
were considered
by the
licensee 'to be overly conservative.
The letter indicated that the occurrence
of an accident
requi ring
the recirculation
was
low.
The letter also indicated the
ingestion of the wiping cloth and tape
were not likely, even with
the missing inner 3/16,inch
mesh screen
due to the existing inner
grating.
The grating is typical walkway grating of egg crate design
made of 1', by 3/16 inch steel
spaced
1-3/16 inch center to center
with cross
bars
spaced
4 inches center to center.
The letter also
states
the probability of the metal pieces
being carried to the
piping inlet are
low.
The letter states
the rag will probably block
a small
area of the heat exchanger.
The letter further indicates
, that if a single failure is not assumed,
then the loss of one train
(due to debris) is tolerable.
The letter also indicates that for
the large break
LOCA the fai lure of the SI pumps to restart will not
be
a problem due to the ability of the
RHR system to provide flow to
a depressurized
RCS.
The
PGIIE letter concludes
that there
was not
a
significant risk.
o
November 21,
1989,
the licensee
made
a four hour
non-emergency Title
10 Code of Federal
Regulations
(CFR) Part 50.72 report regarding the
Unit 1 containment
sump:
The licensee
reported the presence
of the
debris
and reported their assessment
that it would be unlikely that
the debris would have
a significant impact on safety.
They also
reported the Mestinghouse
analysis results
except that, they took
cr edit for the availability of a second
RHR heat exchanger.
o
November 26,
1989,
the licensee
performed
an as-built inspection of
the Unit 1 containment
sump.
Gaps
were found which were present
since construction.
One gap,
about
1 inch wide by 5 feet long,
and
two gaps,
2 1/2 x 3 inches,
were the biggest identified.
Several
rips,
as large
as
1/2 x
2 1/2 inches,
in the outer screen
were
noted.
Other defects
such
as missing bolts were noted.
The specific results
are recorded
on licensee as-built sketches
SKC-43642-1
through SKC-43642-5.
Additionally, a section of structural
grating (which is used to
support the 3/16 inch screen)
was found missing.
The licensee
has
not yet assessed
the consequence
of the missing grating.
o
On November 28,
1989, the licensee
proposed
that the Unit
1
configuration for restart of the unit should
be without an inner
3/16 inch screen.
Subsequent
to discussions
with NRC, the licensee
decided
to install the inner screen
and issued
design
change
DCI-EC-43770 to accomplish that task.
o
On December
1, 1989, the Unit
1
had outer gaps
repaired
and the
inner screen reinstalled
by design
change.
On December
2, the
cleanliness
of the
was personally verified by the Assistant
Plant Manager for Technical
Services
and quality control personnel.
o
On December
11,
1989 the licensee
provided
a preliminary assessment
of the consequences
of the as-built gaps
found on November
26 in the
Unit
1 containment
sump outer screens.
The licensee
concluded that
there
had
been
no significant risk of system
damage
due to debris
larger than 3/16 inch being able to pass
through the outer screen
gaps.
The licensee's
conclusior
was
based
on their procedural
requirement to maintain
a clean containment
and
an assessment
of
accident related debris
such
as paint and insulation.
The licensee
subjectively judged that paint ingestion
would not have noticeable
effects,
nor would increased
ingestion of calcium silicate
insulation.
The licensee
did not plan to have Westinghouse
perform an analysis
of the potential
consequences
of the as-found outer screen
gaps.
The licensee
planned to finalize their own analysis
in the
nonconformance
report written for the containment
sump concerns.
o
On December
14,
1989 the licensee
provided
a copy of an emergency
procedure
(ECA-1. 1) which provides contingencies
for operator action
to mitigate the consequences
of a
LOCA with the "Loss of Emergency
Coolant Recirculation".
This procedure
provides actions
to restore
emergency
coolant recirculation capability, to delay depletion of
the
RWST by adding
makeup
and reducing outflow, and to depressurize
the
RCS to minimize break flow.
Summar
of Concerns
Overall this inspection
has
demonstrated
weaknesses
in the implementation
and maintenance
of design
bases,
and the conduct of safety related
surveillances.
The licensee
should give consideration
to assessing.
the
causes
of the following problem areas,
including an assessment
of the
potential
breadth of the errors
which occurred in their respective
timeframes.
a
~
Potentiall
Ino erable
Emer enc
Core Coolin
S stems
Diablo Canyon technical specification 3.5.2 requires
two emergency
. core cooling sub-systems
be operable
including an operable
flow path
from the containment
sump during the recirculation
phase of
operation.
The. two emergency
core cooling systems
in Unit
1 appear to have
been
degraded
since initial star tup on November 7,
1984 because
of the
gaps in the outer
sump screen.
This is
a potential violation.
b.
Potential
Sum
Inner Screen
Discre ancies
c
~
The Unit
1 inner screen configuration
was different from that in the
FSAR and that of Unit 2.
The acceptability of the various
screen
configurations
should
be assessed
and specified
by the licensee
so,
that there is
no confusion
on the
sump design
bases,
function,
and
operability.
This is
a potential violation.
Potentiall
Unacce table Verification of Sum
Cleanliness
.
Technical specification
4 .5.2.c requires
a visual inspection
be
performed to verify that
no debris is present
which could cause
restrictions of the emergency
core cooling system
(ECCS) suctions
during
a loss cf coolant accident
(LOCA).
Loose debris
including
a hacksaw blade
and wiping cloth were found
in the Unit
1 containment
by the
NRC inspector
on October
17,
'989.
The debris
could cause restrictions of the
ECCS suctions
durino
a
LOCA.
The
had
been
inspected
arid verified cleari by
licensee
personnel
on May'l, 1989.
This is
a potential violation.
d.
Potential
Failure to Declare
Emer enc
Core Coolin
S stem
Ino erabi it
Technical specification 3.5.2 requires
two emergency
core cooling
subsystems
be operable
in Miodes 1,
2 and 3, including an operable
flow path from the containment
sump during the recirculation
phase
of operation.
If both trains of the Emergency
Core Cooling System
(ECCS) are inoperable,
technical specification 3.0.3 applies,
and
requires that within
1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> action
be initiated to place the unit in
a mode in which the action statement
doesn't
apply.
The tm
ECCS subsystems
were potentially not operable
while in Mode
1 for about
12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />
each
on October
12,
1987 for Unit 2 and
on
Sept. ember 7, '1988 an) May 11,
1989 for, Unit
1 and action
had not
been initiated to place
the Units in a mode where the action
statement
does
not apply.
Th'is condition
was
due to the opening of
the containment recirculation
sump screened
access
hatch to allow
the addition of and subsequent
pump
down of borated water with hoses
for the calibration of the
sump level detectors.
This is
a
potential violation.
e.
Potentiall
Non-Conservative
En ineerin
Decisions
'0
The inspector
concluded that the licensee's
engineering
decision to
remove inner
sump screens
in 1981 appear
to have
been
a
non-conservative
engineering
decision which did not fully consider
basic
defense
in depth strategy defined
by
NRC requirements
and
guidance.
Further,
the inspector
concluded that the licensee's
preliminary engineering
decision to return Unit
1 to
power after
refueling without an inner
sump screen
in place
may have likewise
been
a non-conservative
engineering
decision in regard to thoroughly
evaluating
the need for defense
in depth.
Potential
Failure to
S ecif
Construction
Desi
n Chan
e Acce tance
Criteria
To implement
a design
change
in 1981, prior to lice'nsing,
the method
used to maintain configuration control was:
o
To issue
a design
change.
o
Construction
would formulate
a work package
based
on applicable
specifications
and general
construction
procedures.
o
After construction,
inspection
would occur against
the
specifications
and procedures.
o
Construction
would generate
detailed as-builts of the
modification.
o
Engineering
would receive,
review,
and accept as-built drawings
and incorporate
them into the applicable
design
drawing.
This was the process
used for the relocation of the
sump screens
done in 1981.
The design
drawing was not properly updated
in that
reference
to the inner 3/16 inch mesh
screen
was not removed in one
detail
o< the design
drawing 443259,
Revision 4, issued
t1arch
16,
1981,
The
1981 desion
change
process
reflected
a practice
not totally
uncommon in the 'early 1980s of field design followed by engineering
analysis
and approval or modification if required.
ln the case of the containment
sump,
the critical construction
parameter
("The
max gap opening
anywhere
on the surface of the
shall
be 3/16.")
was not defined
by engineering for the field unti 1
December
15,
1986
when Revision
8 of the containment
sump drawing
443259
was issued.
The revision
was issued at that time to reflect
the Unit 2 design
changes
of 1985
made to close
the gaps.
As
explained
in the sequence
of events,
the licensee
missed
the
opportunity to discover
and correct the Unit
1 gaps
in 1985.
The apparent failure to specify adequate
construction
acceptance
criteria for the containment
sump screen
gaps
led to the as-built
condition discovered
in 1989 of large
gaps
in the outer screen.
E
g.
Potential
Failure to Pro erl
U date
Desi
n Documents
The failure to properly update
the design drawing for the
containment
sump led to the reinstallation of inner screens
in Unit
2 in 1985.
h.
Potential
Failure to
U date the
The failure to update
the
FSAR description
and drawings of the
containment
sump pursuant
to 1981 design
changes
led to faulted
assessments
of the
design
bases
on the part of the licensee's
engineering
personnel
and
NRC reviewers.
This is
a potential
violation.
l!
Potential
Failure to
Re ort Adverse Conditions
Found Durin
~0ereti on
The April 25,
1985 problem report for Unit 2 and the August 2,
1985
action request for Unit
1 were both issued after licenses
were
issued for the units.
The problems identified reflected
on the
operability of the
ECCS system but were not reported to the
NRC.
This is
a potential violation.
j.
Potential
Failure to Perform Proper
Ins ections
and Ins ection
~Si
no, fe
A number of relatively recent occurrences,
in addition to the
1985
failure to 'identify sump screen
g'aps,
indicate that plant personnel
performing inspections
and sign-offs in procedures
were not
performing their duties in a precise
manner
and were'aking
broad
and non-conservative
interpretations
of the verification statements
of procedures.
For example
on June
16,
1988 the containment
sump and
RHR suction
piping was verified clean
and free of structural
distress
and
corrosion
by mechanical
maintenance
and senior operator
licensed
personnel.
The procedure
annotations
indicate the inspection
was
performed
from outside
the
sump by flashlight.
The inspector
found
it difficult to 'assure
an adequate
inspection of the
RHP, piping in
this case
given the screening
and grating visual interferences
and
the distances
involved.
Although no debris
was found in the pipe in
1989 inspections
by remote television devices,
the inspection
performed in 1988
was not suitable to attest to piping cleanliness.
The
same
method of inspection
was again
used
and the
same
inadequate
verification was
made
by
a different licensed individual
on July 5,
1988,
On Nay 11,
1989 the cleanliness
of the
was attested
to by
maintenance
personnel
after level indicator calibration.
Licensee
evaluation indicates that the debris later found in October
1989
was
in the
sump at this time.
Previous
inspector findings such
as the violation issued for a
inspector attesting
to
a valve cleanliness
without visually
inspectino
the valve body (50-275/88-03)
and current problems with
misaligned valves which had
been verified and signed for by
12
operators, 'indicate that the proper instincts to precisely perform
. verifications
have not consistently
been demonstrated
by personnel-
at Diablo Canyon.
Descri tion of Safet
Functions
The Diablo Canyon
FSAR section 6.2.3.2.2:1
describes
the containment
recirculation
sump.
It states
that the
sump is
a large collecting
reservoir designed
to provide
an adequate
supply of water with a minimum
amount of particulate matter to the safety injection system,
the charging
system,
the residual
heat
removal
system,
and the containment
spray
system during the recirculation
mode of emergency
core cooling system
operation following a loss of coolant
accident."'he
FSAR states
the-sump is approximately
50 feet from components
that
could become
sources
of debris.
It further states
a baffle arrangement
surrounds
the
sump to prevent floating debris,
high density particles
or
anything larger than 3/16 inch from entering the
sump.
It states
that
the two )4 inch suction pipes are located
on opposite
sides of the
to prevent
any unforeseen
blockage of one from affecting both containment
spray trains.
The
FSAR indicates
that the fluid from .the
passes
into the
14 inch
residual
heat
removal'(RHR)
pump suction piping.
The flow passes
through
the
RHR pumps to the heat exchangers.
The flowpath continues
to the
suction of the safety injection system
(SIS)
pumps.
The SIS
pumps
discharge
to the hot legs of the reactor coolant system
(RCS), irito the
core
arid out through the ruptured
RCS loop onto the containment floor and
to the containment
sump.
The
FSAR describes
the flowpath to the centrifugal charging
pumps
similarly from the
RHR heat exchangers
to the suction of the charging
pumps to the cold legs of the
RCS.
The
FSAR describes
the
RHR flow path similar to the flow path from
the'HR
heat exchanger
to the cold or'hot legs.
The
FSAR describes
the flowpath to the containment
spray system
similarly.
The flow continues
'from the
RHR heat
exchangers
to the
containment
spray
and out the 3/8 inch spray nozzles.
In
recirculation
mode,
the
RHR pumps instead of the Containment
Spray
(CS)
pumps are
used for motive force.. The screen
mesh size of 3/16 inch for
the
sump screens
was sized to preclude plugging of the 3/8 inch spray
nozzles.
Redundanc
of the
Sum
The
FSAR section 6.2,3.3.7
in discussing reliability states:
"To ensure
that the failure of a portion of the protective screen
a'ssembly
(Figure 6.2-11) will not negate
the effectiveness
of the
ensnare
assembly,
separate
outer and inner assemiblies
are provided.
In addition,
the inner assembly contains
a steel divider that
prevents
a hole in the screen
assembly
over one
RHR pump suction
13
pipe from influencing the effectiveness
of the screen
over the
redundant
pipe."
Figure 6.2-11 of the
FSAR shows 3/16 inch wire mesh installed
on the
outside
sump structure
(refe'rence
Section
A, Elevation
B, Section
E and
Detail 2)
and 3/16 inch wire mesh installed
on the inner wedge grating
structure
(reference
Detail 1).
Note
6 of the drawing states,
"The max
gap opening
anywhere
on the surface of the
sump shall
be 3/16."
Section
G of the drawing
shows
1/4 inch steel
plate dividing the inner
assembly
into two separate
assemblies.
Consequently,
the inspector
concluded that
( 1) the
FSAR describes
functionally redundant
design features
of the containment
sump,
and (2)
this appears
to be,a desirable
feature which was
an important
consideration
in the
NRC review of containment
sump design.
The
NRC standard
review plan
(SRP) for Safety Analysis Report
(SAR)
review states
that the containment
system
branch
(CSB) will:
"CSB reviews the system provided to allow drainage of containment
spray water
and emergency
core cooling water to the recirculation
suction points
(sumps).
CSB reviews
the design of the protective
screen
assemblies
around
the- suction points.
CSB reviews plan
and
elevation
drawings of the protective screen
assemblies,
showing the
relative positions
and orientations of the trash
bars or grating
a'nd
the stages
of screening,
to determine that the potential for debris
clogging the screening
is minimized.
CSB also
reviewst the drawings
to determine that suction
pints
do not share
the
same
screened
enclosure.
The effectiveness
of t e protective screen
assem
y will
be determined
by comparing
the smallest
mesh size of screening
provided to the clogging potential of pumps,
heat exchangers,
valves,
and spray nozzles.
The methods of attachment of the trash
bars or grating
and the screening
to the protective
screen
assembly
structure
should
be discussed
in the
SAR and
shown
on drawings.
A
discussion of the adequacy
of the surface
area of screening
with
respect
to assuring
a low velocity of approach of the water to
minimize the potential, for debris
in the water being
sucked against
the screening
should
be presented.
Regulatory
Guide 1.82 (Ref.8)
represents
guidelines for the acceptability of the design of
containment
sumps."
(Underlining added)
Regulatory
Guide 1.82
"Sumps for Emergency
Core Cooling and Containment
Spray Systems
issued
in June
1974 states
as part of its guidance:
"Redundant coolant
sump screens
and
pump suction pipes
should
be
separated
as uuc
as prac~tica
to reduce
the possihi'lity that
a
partially c ogged
screen
or missile
damage
to one screen
could
adversely affect other
pump circuits."
(Underlining added)
In addition,
10 CFR 50 Appendix
A General
Design Criteria, Criterion 35,
Emergency
Core Cooling requires
a system to provide "abundant
emergency
core cooling,"
The criterion requires:
14
"Suitable redundancy
in components
and features,
and suitable
interconnections,
leak detection, isolation,
and containment
capabilities
shall
be provided to assure
that for onsite electric
power system operation
(assuming offsite power is not available)
and
for offsi te electric
power system operation
(assuming
onsite
power
is not available)
the system safety function can
be accomplished,
assuming
a single failure."
Likewise, Criterion 38, Containment
Heat Removal, requires
a system
be
provided to remove heat
from the containment
and also specifies that
same
redundancy
and protection against single failure requirements.
The definition of single failure provided in 10 CFR 50 Appendix
A as it
deals with passive
components
like the containment
sump states:
"A single failure means
an occurrence
which results, in the loss of
capability of a component to perform its intended safety functions.
Fluid and electric
systems
are considered
to be designed
against
an
assumed
single failure if neither
( I) a single failure of
any active component....
...nor (2)
a single. failure of a passive
component
(assuming active components
function properly), results
in
a loss of the capability of the system to perform its safety
functions."
A referenced
footnote states
however:
"The conditions
under which
a single failure of a passive
component
in
a fluid system
should
be considered
in designing
the system
against
a single failure are under development."
It was clear to the inspector that the single failure conditions
applicable to the containment
sump were develo
ed
and defined in
Regulatory
Guide 1.82 specifying "redundant
coo ant
screens,
separated
as
much
as practical."
Since the Diablo
FSAR described
"redundant
sump screens,
separated
as
much
as practical," the information available
to
NRC reviewers
indicated
that Diablo Canyon
was to implement
NRC requirements
and guidance for
redundancy
and separation
of containment
sump screens.
Further,
the
indicated that the licensee
was to provide. defense
in depth against
possible
screen failure and the incapacitation of two trains of emergency
core cooling.
On December 8,
1989
an exit meeting,was
held with the licensee's
representative
identified in paragraph
1.
The inspector
summarized
the
scope
and findings of the inspection
as described
in this report.