ML20202B941
| ML20202B941 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 06/02/1986 |
| From: | Debs B, Wilson B NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20202B938 | List: |
| References | |
| 50-369-85-38, 50-370-85-39, IEB-81-03, IEB-81-3, NUDOCS 8607100458 | |
| Download: ML20202B941 (45) | |
See also: IR 05000369/1985038
Text
_ . . . . . _
l
.
I
O UNITED STATES
[WCEG 'o,$ NUCLEAR REGULATORY COMMISSION
[ REGION 11
$ ., j 101 MARIETTA STREET.N.W.
- 2 ATLANTA. GEORGI A 30323
\...../
Report Nos.: 50-369/85-38 and 50-370/85-39
Licensee: Duke Power Company -
422 South Church Street
Charlotte, NC 28242
Docket Nos.: 50-369 and 50-370 License Nos.: NPF-9 and NPF-17
Facility Name: McGuire 1 and 2
Inspection Conducted: ,
Oct er 15-17, 1985 and January 27-31, 1986
Inspectors:
B. T. ' Debs
'
[!
Dite' Signed
M
F. McCoy
S. D. Stadler
W. Poertner
P. Moore
Accompanying Personnel: Gr y on L. Yoder, Ph.D. (ORNL)
Approved by hw '. M m b f?6
fson, Acting Section Chief Da'te Signed
B.W)ionofReactorSafety
Divis
SUMMARY
Scope: This routine, unannounced inspection was in the area of Nuclear Service
Water System Operability.
Results: Five violations were identified.
i
i
,
8607100458 860727
PDR ADOCK 05000369
G PDR
_
. _ , ___
_. ._ . _
REPORT DETAILS
1. Persons Contacted -
Licensee Employees
+G. Vaughn, General Manager, Nuclear Stations
- +T. L. McConnell, McGuire Nuclear Station Manager
- +R. L. Gill, McGuire Licensing
- +E. O. McCraw, Compliance Engineer
- +W. J. Kronenwetter, Design Engineer
- +W. M. Suslick, Associate Engineer
Other licensee employees contacted included construction craftsmen,
engineers, technicians, operators, mechanics, security force members, and
office personnel.
NRC Resident Inspectors
- +W. Orders, Senior Resident Inspector
R. Pierson, Resident Inspector
- Attended exit interview on 10/17/85
+ Attended exit interview on 01/31/86
2. Exit Interview
,
The inspection scope and findings were summarized on October 17, 1985, and
January 31, 1986, with those persons indicated in paragraph 1 above. The
inspector described the areas inspected and discussed in detail the inspec-
tion findings. No dissenting comments were received from the licensee.
The results of the inspection were discussed with utility management during
a meeting in Atlanta on March 14, 1986. The details of this meeting are
documented in Section 11 of this report.
During the exit interview the enforcement findings were presented as
preliminary and unresolved. Following NRC management review, the following
findings were determined:
369/85-38-01, 370/85-39-01 Violation - Failure to adequately perform
preoperational test on control room chiller - see paragraphs 7 and 8.
369/85-38-02, 370/85-39-02 Violation - Failure to implement and maintain
procedures - see paragraphs 7 and 8.
369/85-38-03, 370/85-39-03 Violation - Failure to meet Technical Specifica-
tion 3.7.4 for RN system operability - see paragraph 7.
369/85-38-04, 370/85-39-04 Violation - Failure to perform 10 CFR 50.59
evaluation on degraded equipment - see paragraph 8.
r
2
369/85-38-05, 370/85-39-05 Violation - Failure to identi fy and correct
conditions adverse to quality as * required by 10 CFR 50, Appendix B,
Criterion XVI - see paragraph 12.
369/85-38-06, 370/85-39-06 Unresolved Item - NRC followup of licensee
response of April 25, 1986 - see paragraph 11.
3. Licensee Action on Previous Enforcement Matters
This subject was not addressed in the inspection.
4. Unresolved Items
An Unresolved Item is a matter abcut which more information is required to
determine whether it is acceptable or may involve a violation or deviation.
A new unresolved item identified during this inspection is discussed in
Section 11.
5. Nuclear Service Water System Description
The McGuire Final Safety Analysis Report (FSAR) states that the Nuclear
Service Water (RN) System provides assured cooling water for various
Auxiliary Building and Reactor Building heat exchangers during all phases
of station operations. Each unit has two redundant " essential headers"
serving two trains of equipment necessary for safe station shutdown, and a
"non-essential header" serving equipment not required for safe shutdown. In
conjunction with the Ultimate Heat Sink, comprised of Lake Norman and the
Standby Nuclear Service Water Pond (SNSWP), the RN System is designed to
meet design flow rates and pressure heads for normal station operation and
also those flow rates and pressure heads required for safe station shutdown
normally or as the result of a postulated Loss of Coolant Accident (LOCA).
The system is further designed to tolerate a single failure following a
LOCA, and/or seismic event causing loss of Lake Norman, and/or loss of
station power plus offsite power (station blackout). Sufficient margin is
provided in the equipment design to accommodate anticipated corrosion and
fouling without degradation of system performance, j
6. Summary of NRC Findings
i
On October 4, 1985, the NRC Senior Resident Inspector reported to Region II 1
management that the 1A nuclear service water system, designated by the j
licensee as the RN system, had failed to meet the acceptance criteria of
its quarterly inservice test. Although the Technical Specificttion Action
Statement period of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> expired on October 7,1985, both units con-
tinued operation at full power based on the licensee's contention that
the 1A RN pump had been made operable by cross connecting it with the Unit 2
2A RN pump. On October 10, 1985, NRC informed Duke Power Company (DPC) that
operation in the unit shared mode was an unacceptable unanalyzed condition.
DPC restored unit separation and began justification for continuing opera-
tion with the apparently degraded pump.
1
7_
_ . _ _ _ _ _ _ _ _ _ _ _
.
. _ .
.. . . . . .. ..
.
.
3
Licensee representatives stated that they suspected that the pump was not
actually degraded, rather the pump discharge line flow orifice reading
was in error. One of the possible reasons stated was buildup of silt,
mud, or corrosion at the orifice. Licensee representatives subsequently
stated several months later that the flow indication was erroneous and the
pump was not actually degraded.
The NRC became concerned that if system fouling was that bad at the pump
discharge, what was the status of the downstream components, especially heat
exchangers. A reactive inspection was conducted October 15-17, 1985, to
review these matters. Numerous phone conferences and letters were exchanged
in ensuing months, and a followup inspection was conducted January 27-31,
1986. l
A summary of the major NRC findings presented in this report are as
follows:
a. Preop tests and subsequent surveillance tests performed in 1979 were
not adequate to ascertain operability of RN components.
Several test procedures did not contain acceptance criteria. For
example, a quarterly test of RN heat exchanger 1-A on October 7,
1985, indicated potential fouling but the test procedure contained
no acceptance criteria. The potential fouling was apparently
pursued only because of questions from the NRC and not addressed
by the licensee until October 14 when it was attributed to a
faulty flow instrument. The heat exchanger was assumed to be
- operable during this period of evaluation.
Flow was not measured through control room air conditioner heat
exchangers.
Test results were recorded in units of differential pressure when
acceptance criteria were in units of flow rate.
Heat transfer characteristics of heat exchangers were not normally
determined. Fouling factors or empirical tests could have been
used.
RN system was not originally preop tested in the most limiting
post-LOCA configuration in that both trains were not aligned to
simultaneously draw water from the Standby Nuclear Service Water
Pond.
l b. The positions of valves specified in preop test data were different
from the positions in operating procedures.
l
c.
'
The RN system had not been flow balanced since 1982 even though engi-
neering documents required it to be.
i
a
.-- - - - . . - . . . . _
4
2
d. The following heat exchanger fouling problems had occurred:
Containment spray heat exchanger 1A tested per IEB 81-03 showed
- increasing delta P from 20 psid- in 1983 to 29 psid in 1985.
In October 1982, a containment ventilation heat exchanger would
not function due to fouling.
Periodic cleaning of control room air conditioning heat exchangers
had been necessary since 1982 due to fouling.
RCP motor coolers required cleaning three times during the period
1984 - 1985.
Unit 1 component cooling water heat exchanger observed to be
fouled in September 1984.
- e. Inservice testing of the 1A RN pump indicated degraded flow on
i October 4, 1985. Instead of entering a Technical Specification Action
Statement which would have required the operating unit to be brought I
to the hot standby mode within six hours, the licensee inappropriately
cross-connected RN train A and train B and continued to operate.
f. A flow balance test on RN train IA conducted on December 17, 1985,
revealed flow rates through several safety-related heat exchangers
to be below FSAR values. At the request of the NRC in January 1986,
the licensee evaluated these test results pursuant to 10 CFR 50.59.
This evaluation, which was based upon heat transfer tests by DPC and
calculations by Westinghouse, was completed and justified continued
operation on January 14, 1986. .The licensee apparently assumed the
system to be operable between December 17 and January 14.
'
I Although it appears that RN heat exchangers were becoming progressively
more fouled with time, the licensee did not recognize the symptoms or place
priority consideration on the overall system operability and associated
safety concerns. Rather fouled components required for continued operation
were cleaned as needed but no regard shown for the status of dormant safety
equipment, such as the containment spray heat exchangers.
When the concern was raised by the NRC, the licensee devoted significant
resources toward correcting the problem. As a result, during the months
of investigation, there were several instances when individual components
! were found not to be capable of FSAR specified performance. On these
occasions, the licensee revised their accident analysis supporting calcula-
tions to justify continued power operation. This mode of operation complies
with regulatory requirements but does not appear to represent to the NRC the
most conservative safety philosophy.
,
4
_ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ . _ _ _ . . _ _
- - - _ . . . - - - - . - - - -
5
7. McGuire Nuclear Service Water System History
1979
Preoperational functional testing was completed by the licensee on July 25,
1979, for the Unit 1 RN system and on November 12, 1982, for the Unit 2 RN
- system. In January 1986, NRC Region II inspectors reviewed selected areas
,
of preoperational test packages for both Units 1 and 2 RN systems.
l It was noted that during the conduct of the Unit 1 preoperational tests
j of nuclear service water, the safety evaluation section (8) of the major
1 procedure form was marked as not applicable. Administrative Plant Manual,
- Section 4.2.4.1(e) requires that prior to procedure use, a safety evaluation
- of major changes to a procedure shall be performed. Examples of the major
changes made to the preoperational procedures included changes to the
minimum acceptable RN flow criteria, initial RN system configuration at
>
test initiation, and the methods utilized to determine component flows.
The use of "not applicable" for safety evaluations was allowed by a licensee
internal memorandum dated September 14, 1979. The memorandum deleted the
procedural requirement for a safety evaluation prior to fuel load.
,
The primary objective of the nuclear service water preoperational functional
test was to verify that the system could supply designed cooling water flow
j to variou., components and to set each component throttle valve to provide
the proper flow rate. Adequate system and component flow was to be verified
j for all modes of operation.
i
i One of the safety related RN loads during post-LOCA conditions is the
l control room air conditioner which requires a minimum flow of 789 GPM as
j stated in McGuire FSAR Table 9.2.2-1(8). During the RN preoperational
,
test for Unit 1, the flow to the control room air conditioning was unable
I
to be determined due to problems encountered with the installed instrumen-
l tation. Subsequently, a major change to the preoperational test procedure,
! TP/1/A/1400/01, was approved by the licensee to delete the requirement to
verify the minimum RN flow of 789 GPM. The change to the preoperational
test was justified by the licensee on the basis that the flow control valve
is air operated and fails open during accident conditions. This justifi-
cation assumed that there were no internal obstructions and that the wide
open valve flow would meet or exceed the FSAR~ required flow. Due to this
procedure revision, the subsequent RN preoperational test for Unit 2
also did not verify adequate flow to the control room air conditioning.
As stated later in this report, subsequent functional flow test data
- obtained in late 1985 and early 1986 indicated that the required 789 GPM
, was not being met. Failure to test the aforementioned component represents
1
a violation of 10 CFR 50, Appendix B, Criteria XI which requires a test
program to be established to assure that all testing required to demon-
a strate system components perform satisfactorily in service (369/85-38-01,
370/85-39-01).
'_
-. . . _ - _--
,
6 ,
The inspector noted that in several instances during the conduct of the
preoperational tests of the RN system, the measured flows were stated
as differential pressure (psid) rather than flow (GPM). The engineers who
performed the tests and the preoperational logs indicated this was due to [
problems experienced with the instaihd flow instrumentation. To continue !
the tests with the inoperable flow instrumentation, the licensee utilized i
temporary differential pressure instrumentation. The conversion from j
differential pressure to GPM was not made on test data enclosures. To
verify that the minimum FSAR flow results were achieved for the RN compo-
nents preoperationally tested, the inspector, in early 1986, requested that ;
the licensee convert the differential pressures to flows. In each case it I
was verified, based on the licensee's calculations, that the minimum accept- #
able flow rates had been achieved as stated in McGuire FSAR table 9.2.2-1.
The values from that table appear later in this report. ,
To assure minimum RN component flows, including adequate flow to the con- I
tainment spray heat exchangers during design LOCA conditions, the normally :
'
throttled valves associated with each RN component were required to be set
during p'reoperational testing of the RN system. These throttled positions ,
established during preoperational testing were to be incorporated into l
operating and surveillance procedures to protect these throttled settings !
'
during future operations. The inspectors noted that, in some cases and
particularly for Unit 1, the throttled valve positions listed in the ;
'
licensee's RN operating procedures and their locked valve verification
procedures were not consistent with earlier preoperational "as left" data.
It was noted, however, for those throttled valves reviewed, the operational
positions were further open than the "as left" preoperational test posi-
tions. The licensee acknowledged these discrepancies and committed to
revise the operational procedures to meet those valve settings established ;
during recent 1985 and 1986 RN flow testing.
The inspection team noted that since 1976 the licensee has had a functional r
system description for the RN system. Section 5 of this system description :
(MCSD-0138.00) states that annually each essential RN train must be checked !
for proper throttling. Also, after any throttle valve is repositioned, the
entire train must be checked for proper throttling. The system description. A
then presents a detailed procedure to verify that the minimum flow condi-
tions for operability of the safety related portion of the system are met. ,
4
The licensee had decided not to adopt the aforementioned recommendations. l
Consequently, no RN flow balances had been performed - beyond 1982 until
'
requested by the NRC in late 1985. Functional system descriptions are
2
not used as procedures by licensees and, consequently, failure to follow
'
MCSD-0138.00 is not considered to be a violation. However, compliance
with this document would have prevented the above violation. However, The
requirements to verify proper throttling position should have been in plant
procedures. ;
Failure to measure flow through components and failure to specify post-
tions of throttled valves in procedures represent examples of inadequate
- . _
_
. _. .- --. ,- . - ,
7
procedural controls and are, therefore, a violation of McGuire Technical Specification 6.8.1 and 10 CFR 50, Appendix B, Criteria V which requires
that adequate written procedures be implemented and maintained (369/85-
38-02,370/85-39-02).
In addition to adding procedural requirements for RN throttled valve posi-
tions as addressed above, the licensee has implemented several other posi-
tive methods to control these valves. Currently, these valves are verified
locked every six months under the Locked Valve Verification Procedure
4700/23. In addition, independent verification is utilized to ensure that
the valves are returned to the proper position following valve repositioning
for maintenance or other activities. Despite these positive controls, the
inspectors noted the following recent deficiencies in the licensee's control
over these throttle valves:
-
The Locked Valve Verification Procedure requires that the operator
verify the valve to be locked. No verification of the actual throttle
position is required.
-
The valve locks utilized for RN throttled valves are chain locks.
These chain locks work well for wide open valves, but the slack in the
chains cannot ensure that a valve remains open 1/4 turn. A valve that
is required to be open 1/4 to 3/8 turn could be locked in the full-
closed position without detection.
One potential solution identified by the licensee for better control of
these throttle valves include the use of locking collars which are used on
throttle valves in other systems. Since the locking collars can be sized
to ensure the exact valve opening desired, their use would provide positive
indication of valve position.
The licensee initiated a 10 CFR 50.72 notification to the NRC stating that
prior to January 27, 1986, the RN systems for Unit 1 and 2 had never been
tested under the requisite design basis accident configuration. Specifi-
cally, the system valves had never been positioned to supply the required
flow to essential headers for Units 1 and 2 with the system taking suction
solely from the Nuclear Service Water Pond. This issue is discussed in
Section 6. of this report.
1981
In response to IE Bulletin 81-03 which addressed the potential fouling of
safety related heat exchangers by clam and shell debris, the licensee com-
mitted to the NRC to monitor two RN supplied heat exchangers on a quarterly
basis. One of these heat exchangers is the 1A Containment Spray (NS) heat
exchanger. Additionally in the licensee's response, it was stated that
"if significant fouling is detected on these heat exchangers, other heat
exchangers in the RN system will be inspected." The licensee performed
their monitoring under procedure PT/1/A/4403/04. This procedure for the 1A
NS heat exchanger requires that the test be performed for a FSAR accident
. _ . - _ - = _ -.- -_
!
!
8
4
i
RN flow of 5000 GPM to the NS heat exchanger and that the heat exchanger
,
'
differential pressure (D/P) be recorded. In October 1985, the inspectors
reviewed the past test data which indicated the following:
DATE OF TEST D/P (PSID)
l 6/20/83 20
! 9/22/83 Not Available
10/2/83 23.5
! 1/18/84 25
4/11/84 23
7/18/84 25
29.5
'
11/9/84
2/28/85 23.5
6/27/85 25
- 10/7/85 29
- RN flow was 4600 GPM
The test procedure did not specify criteria for determining "significant
fouling" and, tnus, other components were not inspected as a result of
'
these tests. Further discussion of these findings appears later in this
report under the section titled 1985.
l
1982
On October 22, 1982, the licensee identified that fouling of the RN supplied ,
lower containment ventilation heat exchangers was a problem which was
causing unacceptable temperature increases in the lower containment areas.
This subsequently forced the units to operate at reduced reactor power
during certain seasonal conditions. In April 1983, the licensee attempted
to add a penetrant / dispersant to the RN system in an attempt to clean lower
- containment cooling units. The attempt was ineffectual. Eventually the ,
licensee modified the coolers with a self-cleaning mechanism which corrected
the problem.
As a result of a control room air conditioning trip due to fouling of the RN
supplied, safety related air conditioning chillers, the licensee established
a cleaning threshold based on increasing air conditioning condenser pres- i
sures. On the following dates, these chillers have been rodded out to
maintain operability.
TRAIN A TRAIN B
11/19/82 3/83
! 10/03/83 01/07/85
12/19/83 10/21/85
05/30/84 11/05/85 i
10/31/84 l
09/25/85 l
10/24/85 i
10/31/85
,
, - . - - - -
-- ----. ., , - - - . - - , ,, , - . . - -
.
9
1984
In March 1984, the licensee began development of a heat exchanger perform-
ance monitoring program. At the time of this inspection, Duke Power Company
had not fully implemented this program at their nuclear plant.
The inspector reviewed the section of the program which requires monitoring
the performance of heat exchangers such as those in the RN system. The
program appeared to be very comprehensive with provisions for monitoring
both flows and heat transfer capabilities, for increasing the frequency of
monitoring as warranted, and for initiating corrective actions as necessary.
Once fully implemented, this Performance Monitoring Program will be a major
improvement in the licensee's ability to monitor plant equipment perfor-
mance and to promptly identify degraded performance. A key to the relative
+ success of the program, however, will be the effectiveness and timeliness
of corrective actions taken in response to an identified deficiency. The
inspector noted that this corporate monitoring program was scheduled to be
implemented in stages at the various plants. The RN heat exchangers were
scheduled for performance monitoring implementation during the second phase
of the program which will be several months into 1986. As a result of
the fouling and degraded performance being experienced with the RN heat
exchangers and concerns expressed by the NRC, the licensee indicated this
phase of the program will be implemented on a priority basis.
Also in 1984, the licensee began to experience RN fouling problems in
their reactor coolant pump motor coolers. The licensee has performed the
following cleanings of these coolers on the dates indicated:
'
UNIT 1 UNIT 2
12/31/84 8/10/84
11/08/85
In September 1984, the licensee evaluated the Unit 1 Component Cooling (KC)
heat exchanger for fouling, although, according to the licensee, there was
no indication of reduced heat transfer or high differential pressure. As
part of the evaluation, DPC engineering calculated a fouling factor for the
KC heat exchangers. These calculations were based on informal test data
which appeared to the cognizant engineer as nonrepresentative. In November
1984 the Unit 1 KC heat exchanger was cleaned. In June and July 1985, the
,
Unit 2 KC heat exchanger was cleaned. Although visual inspection of the
heat exchanger by DPC engineering did not support the calculated fouling
factor (the calculated fouling factor appeared to be less conservative),
the licensee did not perform further evaluation of past operability of these
heat exchangers.
t
10
1985
On October 4, 1985, following in-service testing, the 1A Nuclear Service
Water pump performance was found to be degraded. The pump curve generated
from the test data deviated from the previously established base-line
curve. Delivered flow was estimated to be approximately 85 percent of that
required. Technical Specification (TS) 3.7.4 requires two loops of RN to
be operable. With only one loop operable, they must restore both loops to
operable within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> or be in hot standby within the next 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and
cold shutdown within the following 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />.
The licensee performed a 10 CFR 50.59 analysis to justify cross connecting
the 1A and the 2A RN trains in an attempt to boost IA RN flow. After
reviewing the 50.59 analysis and extensive interaction with the licensee,
the NRC Region II, on October 10, 1985, informed the licensee that the NRC
considered the licensee was not meeting the requirement of TS 3.7.4 which
requires two operable RN loops since the 1A train was inoperable due to
a degraded pump and that the cross connected configuration could not be
justified by a 50.59 analysis since it represented the possibility of an
unreviewed safety question and, in effect, changed the Technical Specifica-
tion.
The licensee's action to cross connect the 1A and 2A RN trains and to con-
tinue two unit operation for greater than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> was contrary to TS 3.7.4
and, therefore, represents a violation (369/85-38-03, 370/85-39-03).
During this time period, the licensee discovered that one of the cross
connect valves had an erroneous position indication. Thus, the valve
was actually closed when thought to be open. This matter was discussed
previously in Region II Inspection Report 50-369/85-35, 50-370/85-36.
As a result of the interactions with the NRC, the licensee split the RN
trains and took compensatory measures to continue operation of the 1A
train under reduced flow conditions. Further details of the apparent
degradation of the 1A RN pump are contained in NRC Region II Inspection
Report 50-369/85-37. As a result of the aforementioned event, during the
'
period of October 15-17, 1985, Region II inspectors reviewed the overall
RN system performance in light of the recent event.
The inspectors reviewed the licensee's quarterly performance, PT/1/A/
4403/04, data on the 1A NS heat exchanger which was tabulated earlier in
this report under 1981.
The following observations were made by the inspectors regarding PT/1/A/
4403/04:
The performance test lacked qualitative and quantitative acceptance
criteria.
- The test results suggest an increasing D/P across the 1A NS heat l
exchanger. ;
i
-_ _. -_.
__ _ . _ _ _ -
11
The pressure drop could not be measured at the required design basis
accident RN flow of 5000 GPM for the 1A NS heat exchanger because this
flow could not be achieved for the test performed on October 7,1985.
The measured flow was recorded as 4600 gpm.
- The 1A NS heat exchanger outlet throttle valve was closed to the
extent that the as found flow through this heat exhanger was 800 gpm.
It appears doubtful that the required accident flow of 5000 gpm could
have been achieved with this as found valve position.
The licensee indicated that, at that time, a qualitative or quantitative
acceptance criteria had not been determined but that work had begun to
provide such criteria. 10 CFR 50 Appendix B Criteria V states that
procedures shall include appropriate quantitative or qualitative acceptance
criteria for determining that important activities have been satisfactorily
accomplished. Contrary to this regulation, PT/1/A/4403/04 did not contain
an appropriate acceptance criteria. This represents another example of
violation (319/85-38-02, 50-370/85-39-02).
Regarding the aforementioned increasing D/P across the 1A NS heat exchanger,
the licensee indicated that, although the test results suggest an increasing
D/P, some mathematical analysis should be performed to prove the apparent
trend of an increasing D/P.
Regarding the low 1A NS heat exchanger RN flow recorded on October 7, 1985,
the licensee indicated that the low reading could have been a result of a
calibration problem. As a result of the inspector's questioning, the
licensee issued Work Request Number 65574 to check the calibration of the
flow instrument used to obtain the recorded 4600 GPM. On October 14, 1986,
the calibration results indicated that, at a flow of 5000 GPM, the instru-
ment indicated 4820 GPM. The licensee then took action to recalibrate the
instrument.
Based on the data reviewed and discussions with licensee personnel, the
inspector stated the following concerns:
Since the licensee did not have an acceptance criteria for the in-
creased D/P, could the apparently increasing D/P suggest heat exchanger
fouling which may have reduced heat exchange capacity to an unaccept-
able level? Could system flow reductions due to fouling affect other
RN system component performance? These concerns were discussed with
plant management on October 17, 1985. The inspector requested manage-
ment to consider the feasibility of performing a RN system integrated
flow test to provide confidence that all RN safety related loads could
be provided the requisite design basis flows. Additionally, the
inspector discussed the feasibility of measuring the heat transfer
capability of the 1A NS heat exchanger.
_
- _ _ _ .
12
After growing concern by NRC Region II regarding the current ability of
the Unit 1 RN system to perform its safety function under accident condi-
tions, the licensee was requested, on October 18, 1985, to provide the NRC
Region II Office with a statement of operability for the RN system. On
October 23, 1985, the operability statement was received from tne licensee.
This :tatement concluded that the RN system is operable and capable of
performing its intended safety function.
The statement of operability included an engineering evaluation by Duke
Power Company. The evaluation summarized the results of a Westinghouse
computer calculation which utilizes the LOTIC code. This code predicts
containment pressure response from inputs including the heat transfer
capability (UA) of the containment spray and component cooling water heat
exchangers. The Duke Power engineering calculations used to determine
the UA for the 1A NS heat exchanger assumed the same fouling factor which
was calculated for the Component Cooling Water (KC) heat exchanger in
early 1985. The inspectors expressed reservation over this assumption;
questioning the credibility of applying the existing fouling factor for a
single pass horizontal type heat exchanger (KC heat exchanger) to the NS
heat exchanger which is a vertical U-tube heat exchanger. Additionally,
RN flows through the tubes of the KC heat exchanger unlike the NS heat
exchanger where RN flow is on the shell side. It, however, was agreed by
the NRC that for lack of any other available data this approach was accept-
able until specific empirical data could be obtained.
Based on the aforementioned assumptions and calculation as utilized in
the LOTIC program (WCAP-8282), a maximum containment pressure of 13.3 psig
was predicted during a design basis accident. The McGuire containment
design pressure is 15.0 psig.
In response to NRC concerns over the potential fouling and degradation
of the 1A NS heat exchanger, the licensee developed a performance test
PT/0/A/4208/01, Containment Spray Heat Exchanger Performance Test. The
purpose of the test was to:
Determine if a high flow flush reduces the heat exchanger differential
pressure.
Assure the structural integrity of the heat exchanger tubes.
Determine the overall heat transfer coefficient and fouling factor of
the NS heat exchangers.
The McGuire FSAR analysis utilized a containment spray heat exchanger UA
of 2.J4 x 10' BTU-Hr-Deg. F. Empirical data from the aforementioned test
indicated that an actual UA of 7.35 x 105 BTU-Hr-Deg. F existed under
,
current plant conditions. This information was provided to Westinghouse
on November 27, 1985, to perform a LOTIC run utilizing this data. A con-
tainment response model which is less conservative than the one used in
the FSAR analysis was used by Westinghouse (WCAP-10325) for this run. Use
, _ . - ,
13
of this model was accepted by the NRC since this WCAP had been reviewed
and found technically sound by the NRC staff, although the NRC's Safety
Evaluation Report had not been issued at that time. This LOTIC run of
November 27, 1985, indicated that, for the aforementioned UA, a peak
containment pressure of 14.42 psig would be realized under design basis
accident conditions.
In addition to the heat transfer test, the licensee performed a heat
exchanger tube integrity test using the tritium activity of the Refueling
Water Storage tank (RWST) as a tracer source. The test results indicated
insignificant leakage.
Several cleaning attempts using various chemical and hydraulic techniques
were employed by the licensee to clean the 1A NS heat exchanger. The latest
performance test results (January 28, 1986) indicate that a UA of 2.03 x
10' BTV/Hr-Deg. F had been achieved.
The inspectors viewed video tapes of the licensee's fiber optic inspection
of the 1A NS heat exchanger. Approximately the first seven feet of the
upper portion of -the tube bundle could be viewed. The tape indicated that
a fairly uniform silica deposit completely covered the tubes, prior to
cleaning.
Confirmatory UA calculations were performed by the inspection team. These
calculations appear as Attachment 4 to this report. Those calculations
closely approximate those of the licensee.
'
8. Review Of Flow Balance Testing
The inspectors conducted a review of the RN flow balance testing conducted
on December 17, 1985, January 27, 1986, and January 28, 1986, for Train IA
of the Nuclear Service Water System. Additionally, flow balance testing
conducted on January 30, 1986, for Train 1B of the Nuclear Service Water
System was reviewed.
The Train 1A flow balance test conducted on December 17, 1985, was in
accordance with procedure TT/1/A/9100/105, Change 0 through Change 1. The
test provided for:
-
Isolation of Train IA and 1B essential header.
-
The low level intake providing Train 1A suction.
-
Isolation of the Unit I non-essential header from Train 1A.
-
Control Room and Equipment Room A Train Cooling Chillers being supplied
by Nuclear Service Water Train 1A.
1
,
'
. _ _ _ - , . .
14
-
Securing of Nuclear Service Water Train 2A due to a condition of
Nuclear Service Water operability resultant from prior degraded pump
performance in Train IA when supplying Control and Equipment room
cooling.
-
Alignment of service water valves in accordance with a lineup that
was consistent with actual Safety Injection and Containment Spray
conditions.
The Train 1A flow balance test conducted on January 27, 1986, was per-
formed in this same manner with the exception that Change 2 of _ procedure
TT/1/A/9100/105 was also in effect which changed the Train IA suction from
the low level intake to the service water pond in order to duplicate the
most restrictive condition of operation for testing.
Flow rates through those essential heat exchangers required to mitigate
accident consequences during Safety Injection and Containment Spray were
measured during these tests and compared to target values which were
specified in the FSAR. Measurement results and comparisons for Train 1A
tests are delineated in Table 1.
- The data for the December 17, 1985 test reflects that FSAR specified flow
rate values could not be attained for the containment spray heat exchanger
(4% degraded), control room chiller heat exchanger (10*s degraded), the
charging pump oil cooler (46% degraded), spent fuel pool pump room ai*
handling unit (27% degraded), and containment spray pump room air handlirg
unit (56% degraded).
Although the data from the December 17th test indicated multicomponent
degradation, the licensee performed an informal evaluation to support
continued operation. The results of this evaluation were not documented.
Not until requested by the NRC in January 1986, did the licensee perform
a detailed engineering evaluation as required by 10 CFR 50.59. Failure
to perform this requisite evaluation is considered a violation of the
aforementioned 10 CFR 50.59 (369/85-38-04, 370/85-39-04).
In an operability statement dated January 14, 1986, the licensee performed
an engineering evaluation to demonstrate the adequacy of the tested perform-
ance of the charging pump oil cooler, the containment spray pump room air
handling unit and the spent fuel pool cooling pump room air handling unit
with the observed reduced flow rates. In the operability statement the
licensee stated that the degraded containment spray heat exchanger flow
was adequate, and justified continued operation of Unit 1. This operability
statement was based on the actual tested values of the thermal efficiency
for this particular heat exchanger and a containment pressure calculation
performed by Westinghouse and forwarded to Duke Power Company by letter
DAP-86-513 dated January 16, 1986. The Westinghouse calculation was based
on assumptions which included the following:
An active sump volume of 90,000 cubic feet.
15
A thermal efficiency heat transfer coefficient of VA=7.35 x 105
BTU-Hr-Deg. F for the containment spray heat exchanger and UA=1.64 x
10' BTU-HT-Hr-Deg. F for the RHR heat exchanger. The licensee stated
that, for the containment spray heat exchanger, this represented a 75%
reduction in the UA coefficient. This value was a conservative selec-
tion by the licensee since the testing performed on December 17, 1985,
demonstrated the UA value to be nearly 58% degraded.
Under these assumptions the Westinghouse calculation demonstrated that
during a LOCA, containment pressure would remain below the containment
design pressure of 15 psig with service water flow through the containment
spray heat exchanger reduced to 4800 gpm. The licensee, therefore, con-
sidered that the results of their evaluations and calculations justified
continued operation of Unit 1. The basis for this conclusion was reviewed
and accepted by the NRC
Between December 17, 1985 and January 27, 1986, three cleaning cycles were
accomplished on the RN side of the 1A containment spray heat exchanger. The
licensee concluded that heat exchanger thermal ef ficiency increased from
42.1% to 74.7% as a result of these cleaning cycles. The affects of these
cleaning cycles is also demonstrated in the reduced RN header pressure
delineated in Table 1, for the flow balance test of January 27, 1986.
The data in Table 1 for the January 27, 1986 test reflects that, even af ter
the cleaning evolutions, FSAR specified flow rate values could again not be
attainec for the containment spray heat exchanger (2% degraded), control
room chiller heat exchanger (0.5% degraded), Spent Fuel Pool Pump Room Air
Handling Unit (30% degraded), containment spray pump room air handling
unit (56% degraded), diesel generator cooling water heat exchanger (8%
degraded), and safety injection pump motor air handling unit (15% degraded).
Degradation of the charging pump cooling flcws was attributed to faulty
flow indication which required instrument replacement.
The licensee stated that as a result of this test,' Train 1A of nuclear
service water v's declared inoperable pending resolution of the degraded
flow conditions and correction of the faulty flow indicator associated with
the charging pump oil cooler.
The inspectors noted that these flow balance tests were accomplished with
Unit 2 Train A secured which was not conservative with respect to the
design basis accident. Worst case conditions should assume Unit 2 Train A
providing unit coaldown loads during the operation of Unit 1 Train A to
mitigate accident conditions. This in effect would reduce the net positive
suction head for Unit 1 Train A. The inspectors considered that testing
should reflect this condition. The licensee stated that on January 28,
,
1986, another flow balarce would be performed and that Train 2A would
service necessary cooldown loads for Unit 2 during this test.
!
P
16
In conjunction with resolution of the degraded flow conditions reflected in
the service water Train 1A flow balance testing, the licensee had requested
that Westinghouse perform an analysis to determine new acceptable minimum
values of service water flow through containment spray and component cooling
water heat exchangers. The licensee was considering that a throttling back
of these two major heat exchangers would result in a higher RN header
pressure thus providing increased flow thru the smaller essential heat
exchangers. A Westinghouse calculation was forwarded to Duke Power Company
in January 1986 which demonstrated that, with service water flow through the
component cooling water heat exchanger reduced to 6000 gpm and service water
flow through the containment spray heat exchanger reduced to 3800 gpm, peak
containment pressure would remain below the containment design value of
15 psig during a LOCA.
On January 28, 1986, a third nuclear service water flow balance test was
accomplished on train IA. This test provided for reduced target flow values
of 6000 GPM through the component cooling water heat exchanger and 3800 gpm
through the containment spray heat exchanger which the licensae considered
to be acceptable target values based on the aforementioned Westinghouse
calculation. This flow balance test was performed under the same conditions
as the January 27, 1986 test with the exception that Train 2A was aligned
to provide a cooldown load of greater than or equal to 6000 gpm for Unit 2,
the flow instrument for the charging pump oil cooler had been replaced, and
the RN system took suction only from the SNSWP. The results of this test
are delineated in Table 2. The result of this test demonstrated that flow
values through all heat exchangers were within the new acceptable values
established by the licensee within the operability statement of January 14,
1986. On March 11, 1986, the licensee made a 10 CFR 50.72 notification
to the NRC stating that, prior to January 27, 1986, the RN system for
both units had never been tested under the requisite accident conditions
with all RN being supplied by the SNSWP. Apparently after addressing
both NRC and DPC engineering concerns regarding the desired RN flow test
system configuration, the licensee later realized that the preoperational
test configuration had not tested the system under the design basis
accident configuration. The aforementioned event represents another
example of a violation of 10 CFR 50, Appendix B, Criterion XI (369/85-38-01,
370/85-39-01).
The NRC later learned from the licensee that during the establishment of the
flow test system configuration on January 28, 1986, the RN system entered
a pressure transient. While base loading the RN pumps (gradually placing
requisite heat exchangers on the line), a significant decrease in RN header
pressure was experienced. This event was not allowed to go full term and
was terminated by throttling down on large component flows. The test was
repeated with the throttled valve positions and acceptable results were
obtained. On March 12, 1986, NRC Region II learned of the January 28 flow
transient shortly after DPC management had been informed of it. The NRC
expressed concern regarding the transient since it suggests that, under
actual accident conditions, the RN system's pumps could have lost net
positive suction head resulting in a loss of the ultimate heat sink for
both units. This concern is further addressed in Section 10 of this report.
- -
~
. . _
17
The Nuclear Service Water Train IB flow balance test was conducted on
, January 30, 1986, with the same test methodology utilized for the
January 28, 1986 flow balance test for Train 1A. The results of this test
are delineated in Table 3. The results of this test demonstrated that
- established operability values could not be obtained for the Spent Fuel Pool
Pump Room Air Handling Unit (5% degraded) and the Residual Heat Removal Pump
, Room Air Handling Unit (1% degraded). The licensee was advised by the NRC
that prior to establishing Train IB as being fully operable, these degraded
4
conditions would require further evaluation and resolution.
o
1
i
i
I
i
i
, -- . . . .
- -_ - ., -- _. , . , . . - . . _ ,
. _ - _ .
_ _ _ . _ _ . _ . . . _ _ . _ . _ _ _ . _ _ _ _ _ _ . - .
_ _m = = . _ _ . _ _ . _ . . . _ _ , . _ . . ._. _ . . __ -. _ . _ _ , _ _ , _ .
! ,
t
TABLE 1
Results of Heat Exchanger Flows and Comparison to FSAR Target Values During Nuclear Service Water
Train 1A Flow Balance Testing of December 17, 1985 and Janua ry 27, 1986.
December 17, 1985 Data Janua ry 27, 1986
Ta rget Flow flow Rate Header Flow Rate Header Pressure
Heat Exchanger Rate (CPM) (CPM) Pressure (psig) (CPM) (psig)
1. Component Cooling Water 8000 0000 67.5 8000 56
'
2. Conta inment Spray 5000 4800 67.5 4887 56
l 3. Diesel Generator Cooling 900 900 67.5 830 56
- Water i
4. Control Room Chiller 789 707 67.5 785 56
5. Cha rg i ng Pump Oi l Coo l e r 28 15 67.5 3 56
i 6. Safety injection Pump 20 21 67.5 17 56
Oil Cooler
c
'
7. Spent fuel Pool Pump 20 14.7 67.5 14 56 *
Air Handling Unit
8. Conta inment Spray Pump 45 20 67.5 20 56
Air Handling Unit
9. Residual Heat Removal 45 51 67.5 52 56
Pump Air Handling Unit ,
i
1
$
1
1
-_ - _ -
- ,, - , - -. , , - . - - .,
~ - .. - -- - . - - - . . . . ~ . . . - . . . . - . ~ . . . ~ - . - . . _ . - . . - . . _ . . - - - - - . - - _ _ . - - _ - - - - . - - - -
l
TABLE 2
Results of Heat Exchanger Flows and Comparison to FSAR Target Values and Licensee Established Operability
Values During Nuclear Service Water Train 1A Flow Balance Testing of January 28, 1986.
January 28, 1986 DATA
Licensee Established
Target flow Ope ra b i l i ty Va l ue Flow Rate Header Pressure
Heat Exchanger From FSAR (CPM) ' for Flow (GPM) (GPM) (psig)
,
1 Component Cooling Water 8000 6000+ 6000 62.5 ;
2. Containment Spray 5000 3800+ 3970 62.5
3. Diesel Generator Cooling 900 900 950 62.5
Water
4. Control Room Chiller 789 789 946 62.5
5. Cha rg i ng Pump Oi l Coo le r . 28 15 22 62.5
6. Safety injection Pump 20 20 23 62.5
Oil Cooler ,
7. Spent fuel Pool Pump 20 14.7 19 62.5
Air Handling Unit
8. Conta inment Spray Pump 45 20.0 23.5 62.5
Air Handling Unit
9. Residual Heat Remova l 45 45.0 64.5 62.5
Pump Air Handling Unit
+ Based on assumption that Containment Spray Heat Exchanger Thermal Efficiency is greater than or equal to 70%.
The rma l performance data reflects it is currently 74.7% and past history indicates degradation will inc rea se
due to fouling.
. - _ . - , - .- _ .. ._,____,_ _._ _._ _. . ,,_ ,_
- - - _ _ _ - - . - _ _ . _ - . . - - . . _ - . _ _ - . .. . - - _ . _ - _ . - - . - - - ~ . - _
. _
. - . . _ . - _ . . . . . -- -..
I
TABLE 3
Results of Heat Exchangor Flows and Comparison to FSAR Target Values and Licensee Established Operability
Values During Nuclear Service Water Train IB Flow Balance Testing of January 30, 1986.
January 28, 1986 DATA
I Licensee Established
j -
Ta rge t flow Ope ra b i l i ty Va l ue Flow Rate Header Pressure
Heat Exchanger From FSAR (GPM) for Flow (GPM) (GPM) (psig)
j
I
1 Component Cooling Water 8000 6000+ 6900 52
2. Conta inment Spray 5000 3800+ 5000 52
3. Diesel Generator Cooling 900 900 920 52
Water
j 4. Control Room Chiller *
789 789 912 52
i 5. Cha rg ing Pump Oi l Coo le r 28 15 20 52
, 6. Safety injection Pump 20 20 28.2 52
-
Oil Cooler
7. Spent ruel Pool Pump 20 14.7 14 52
Air Handling Unit
f
, 8. Containment Spray Pump 45 20.0 46 52
'
Air Handling Unit
9. Residual Heat Removal 45 45.0 44.5 52
Pump Ai r Handling Unit
+ Based on assumption that Containment Spray Heat Exchanger Thermal Erriciency is greater than or equal to 70%.
Thermal performance data reflects it is currently 74.7% and past history indicates dearadation will inc rea se
due to rouling.
I
l
, . - - .
7
18
Following performance of the nuclear service water train IA flow balance
test of January 28, 1986, the inspector observed a train IA Diesel Generator
operability test. During performance of this test, the inspectors noted
that the flow indicator for service water flow through the diesel generator
cooling water heat exchanger was off scale high (greater than 1000 gallons
per minute) rather than indicating an expected value of 900 gallons per
minute. Interviews with licensee personnel who had performed the earlier
Train 1A flow balance test reflected that, during test restoration, valve
IRN73A was left in the test position rather than being returned to the
normal position. The test position for this valve is " throttled to 900
gallons per minute in the test lineup configuration". The normal position
for this valve is " throttled to 900 gallons per minute in the normal lineup
configuration." Since the normal RN system lineup configuration isolates
the large engineered safety feature loads, RN header pressure was increased
which resulted in greater flow through those valves which were not throttled
back from the test position. Failure to restore valve IRN73A to its normal
position is contrary to step 12.8 of procedure TT/1/A/9100/105 and is a
third example of a violation for failure to properly implement procedures
(369/85-38-02,370/85-39-02).
The inspectors noted that restoration of the 1A train service water diesel
generator heat exchanger outlet isolation valve (IRN73A) and the 1A train
service water containment spray heat exchanger outlet isolation valve
(IRN137A) to their normal throttled positions could result in insufficient
nuclear service water flow being supplied to the diesel generator heat
exchanger and containment spray heat exchanger when the containment spray
heat exchanger is placed on line during transfer to cold leg recirculation
unless specific operator actions were taken to ensure proper flow through
these heat exchangers. A review of the emergency operating procedures for
safety injection (EP/1/A/5000/01, EP/2/A/5000/01) and for transfer to cold
leg recirculation (EP/1/A/5000/2.3, EP/2/A/5000/2.3) reflected that provi-
sions were not established to assure proper service flow through the diesel
generator cooling water heat exchanger and containment spray heat exchanger
when these component were required during accident conditions. These
inadequacies in the emergency operating procedures are considered a fourth
example of violation 369/85-38-02, 370/85-39-02, failure to properly
establish and implement procedures.
During the course of this inspection, test procedure TT/1/A/9100/105, RN
Train 1A Flow Verification, was revised, and test procedure TT/1/A/9100/107,
RN Train IB Flow Verification, was written to leave the service water outlet
isolation valve to the containment spray heat exchangers in the tested
throttle position. Additionally, licensee actions were initiated to
revise emergency operating procedures EP/1/A/5000/01, EP/2/A/5000/01,
EP/1/A/5000/2.3, EP/2/A/5000/2.3 in order to establish adequate service
water flow through the diesel generator heat exchanger and containment
spray heat exchanger, during safety injection and transfer to cold leg
recirculation.
- . _ _
!
. 19
9. Changes to the McGuire Containment Pressure Response Model
During the course of the licensee's engineering evaluations to justify
the apparent RN system degradation, many changes were made to the input
parameters used in the McGuire containment pressure response model.
The following parameters have significant effect on peak containment
pressure:
ice mass
NS and KC heat exchanger UAs
NS and KC heat exchanger tube and shell flows
mass and energy releases into containment
- auxiliary containment spray flow
auxiliary containment spray actuation time
active containment sump volume
Table 4 provides a chronology of these parameters and when each parameter
was changed by Duke. Some values such as active containment sump are based
on engineering judgement by Duke since calculations have not been completed
to justify the value.
TABLE 4
McGuire Containment Pressure Response Model Changes
Parameter 10/31 11/28 1/17 1/28
Ice Mass 2.220 2.220 2.220 2.220
(millions of LBM)
NS HX UA 1.86 0.735 0.735 2.03
(millions of BTU /HR- F)
(millions of BTU /HR- F)
NS/RN Flow (GPM) 5000 5000 4800 3800
KC/RN Flow (GPM) 8000 8000 8000 6000
Mass and Energy Release 1974 1979 1979 1979
Model (year)
ND Containment Spray 1623 1623 1841 1841 '
,
Flow (GPM)
NO Containment Spray 3000 3000 3000 3000
Actuation Time (SEC)
l
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
20
Active Containment 46,500 46,500 90,000 90,000
Sump Volume (FT3)
Peak Pressure (Psig) 13.3 14.42 14.45 12.7
10. RN System Walkdown
The inspectors conducted a detailed walkdown of portions of the Unit 1
Nuclear Service Water System. The inspectors reviewed the system operating
procedures, the valve checklist procedure and the system piping drawings.
The inspection was conducted to confirm that procedural valve lineups and
drawings matched the as-built configurations, to verify that equipment
conditions were satisfactory and items that might degrade performance were
identified and evaluated, to verify that valves were in proper positions and
locked if appropriate, and to verify that instrumentation was properly
valved in.
The inspectors made the following observations. Valves 1RN 893 and 1RN 894,
the inlets to the 1A1 and 1A2 Diesel generator Air Dryer and af ter dryer
respectively, were mislabeled. Valve 1RN894 was labeled as IRN893. The
Nuclear Service Water System valve checklist correctly described these
valves and the licensee made arrangements to correct the label plates on the
valves prior to the inspector leaving the site.
The inspector noted slight inaccuracies in the system piping diagrams, in
that relief valve 1RN-295 is located upstream of flow element 5360 as
opposed to downstream as indicated on DWG MC-1574-2.0 and vent valve IRN141
is located upstream of flow element 5930 as opposed to downstream of the
flow element as indicated on DWG MC-1574-2.0. The licensee made arrangements
to correct these inaccuracies prior to the inspectors leaving the site.
11. Details of NRC/DPC Management Meeting Held on March 14, 1986
a. Attendance at the Duke - NRC Management Conference on March 14, 1986,
held at DPC's request at the NRC's Region II Office included:
Duke power Company
G. Vaughn, General Manager, Nuclear Stations
T. L. McConnell, McGuire Nuclear Station Manager ,
W. A. Haller, Manager, Technical Services '
R. L. Gill, McGuire Licensing ,
B. H. Hamilton, McGuire Superintendent of Technical Services l
J. E. Snyder, Supervising Engineer !
E. O. McCraw, Compliance Engineer
W. J. Kronenwetter, Design Engineer
R. W. Revels, Design Engineer
W. M. Suslick, Associate Engineer
21
,
Nuclear Regulatory Commission
R. D. Walker, Deputy Regional Administrator
A. F. Gibson, Director, Division of Reactor Safety
C. A. Julian, Chief, Operations Branch
B. T. Debs, Acting, Chief, Operational Programs Section
i M. V. Sinkule, Chief, Reactor Projects Section
F. R. McCoy, Reactor Engineer
, W. T. Orders, Senior Resident Inspector, McGuire '
- C. W. Burger, Project Inspector
C. L. Vanderniet, Reactor Engineer
i b. Members of the Duke Power Company staff met with members of the NRC
Region II staff to discuss the status of the McGuire Units 1 and 2
Nuclear Service Water System. A copy of the meeting agenda and DPC
handouts appear as Attachments 1, 2, and 3 to this inspection report.
DPC representatives stated that, from the information available to the
. DPC staff, the Nuclear Service Water System had been and is currently
operable. The NRC staff acknowledged that, once the NRC had surfaced
concerns regarding the Nuclear Service Water System, the licensee has
placed extensive resources on solving the problem.
As a result of the aforementioned meeting, NRC representatives
contacted DPC staff on March 24, 1986, to request additional informa-
tion. DPC staff agreed to formally submit a response by April 25,
, 1986, regarding the following seven requested items.
-
Provide the as-found and as-left RN flow balance test results for
all RN trains.
-
Provide the as-found and as-left VA test results for all
containment spray heat exchangers.
-
Provide an RN operability determination for early October 1985
when RN flow was recorded as 800 GPM to the 1A containment spray
heat exchangers.
-
Provide safety evaluation of the January 28, 1986 RN header
,
pressure transient.
-
Provide an RN operability determination based on the 1A contain-
ment spray heat exchanger throttle valve setting which existed l
'
,
just prior to the first heat transfer test and based on expected
flow under accident conditions prior to heat exchanger cleaning )
cycles. )
1
-
Provide the final parameters for use in the LOTIC program and
their engineering basis.
l
-
Provide DPC plans to prevent a recurrence of these events.
-. -. -. . _ .
_- . - - -
,
22
By memo of April 25, 1986, Duke Power Company responded to these
requests. The responses contend that the RN system was continuously
operable. Inspectors will follow up on this information during a
-
future inspection.
The resolution of these matters represents unresolved item (369/85-38-06,
370/85-39-06).
12. General Conclusions
1 During the operating history of the McGuire plant, the licensee has experi-
enced an increasing degradation of the RN system. It is apparent that
the licensee has dealt with this situation on a case-by-case basis. Until
prompted by the NRC, the licensee had not determined the full extent of
the RN system degradation or taken adequate corrective action to preclude
i repetition. Although the licensee has recently dedicated significant
resources to addressing the problem, serious doubt exists regarding the
past operability of the RN system and those safety related systems, such as
containment spray, for which RN is an ancillary system. This doubt is
fostered as a result of aggregate observations of significantly reduced
- heat transfer capability of various safety related heat exchangers, reduced
, RN flows, improper throttle valve settings, increased corrosion, and lack of
'
adequate preoperational testing. This situation is contrary to 10 CFR 50,
Anpendix B, Criterion XVI which requires that measures shall be established
tr assure that conditions adverse to quality, such as failures, malfunc-
. tions, deficiencies, deviations, defective material and equipment, and
'
'
nonconformances are promptly identified and corrected. In the case of
significant conditions adverse to quality, the measures shall assure that
'i
the cause of the condition is determined and corrective action taken to
preclude repetition. The identification of the significant condition
adverse to quality, the cause of the condition, and the corrective action
taken shall be documented and reported to appropriate levels of management.
The licensee's failure to meet these requirements, in the case of the RN
,
system, is a violation (369/85-38-05, 370/85-39-05).
)
,
1 i
,
'
l
i .
!
. . _ - _ . .
'.'.
ATTACHMENT 1
!
DUKE POWER /NRC REGION 11
i
MEETING TO DISCUSS McGUIRE NUCLEAR STATION
NUCLEAR SERVICE WATER SYSTEM PERFORMANCE
MARCH 14, 1986
AGENDA
- OPENING REMARKS GERALD VAUGHN
- OVERVIEW OF NUCLEAR SERVICE WATER SYSTEM NEAL McCRAW
i
'
- NUCLEAR SERVICE WATER SYSTEM EXPERIENCE TONY McCONNELL
- RECENT OPERATIONAL EXPERIENCE BILL SUSLICK
(10/04/85 TO PRESENT)
- DESIGN CONSERVATISMS BILL KRONENWETTER
!
- CLOSING REMARKS GERALD VAUGHN
i
<
l
l
.
___
.
. -
.
Ln
I E
4 I OL
.* N O
< CO
t
RC
B
@) I Mo
l
Pn . .
I el NI " i
n, u
0
.
d J. Pm ) I
a 1
- L
ER
N 1 D tv
C .
X C( :
e
MNI dmo
Pn
Pw
e
$g
.
I
VI
RD
SS I
e
r
A
D
l
P
P
A
- 0
E
S
u
I Y
RS
I
@) a'm S s e
A
ER
LE
2
-
2
2
B
R
.c
i.
CT A
Fm UA
n
-
i
A
N NW -
.
@) Nl s :
v s N 1
_ S i W
.
- A
.-
c t
c I
D
.
_ i
- . 81
ER
i
a l
p
TL X
X d
0O
8
8 O
C c :, 1 i'
i
ns
ik
cE
nL nF
-
-
F
a
E
8
M X e.
A0
u0
er
X
- , le L
l
)i4
.
0
. E
s
a
DC wtB
E
EN
sF
C i
,# .L00
C
SR
S
,
rt I
C,
X
,
I
?
T
)$ h :lls[
yi6
.
I
- SX i!
m9
W
MM i
- X
1
.
_
.
-
.
S. NUO
EArP
UEO
tLI
R
" R ".
S
H
NN D g E. -
v
( t
cErC
I
Ul
La X M c= L"
=C
_
eLr/D S FDQ 2
X fBuAI .O S
f
I
=
M Aq
y8 ,
2 CElpC = 0' , >
>
>
o S ,
" S
_
$o
g
CI
2
nE
R
S
e
C
D
E,
R.E
U. ,
M' .
-
oL
3, s , g6 ACRO i
X
-
- J
-
I KoO
nC cL _
_
2 SIM
BO
_
_
_
- }lL a
E
1
X mE
R 4 - T.
U o.
i
o
j ;1
-
I
4l -
ACRO
l 3oO
uL
nC
n,
p.
E a.mw
r0
c $
o
6 J a.
S.
c
$g
1LLXmR
1
EMiE
A
S
X S
EN0AH
LTI
Y
S 2 X
E0W r
H 3 )IDCC V
R
t
s
@ ((
'
-
D
J
v
_ a & _
g ,6 ,M
e
X
_
_
_
_
_
]g I
y1 .
_
_
D
.
. 3 .
J * , o
X GL
I
, h
,
- -
[>
I
-
.
yo LYr t S
eNA
c=
S '
W t
S r D a Tm
i
.
S
u F J" t
_
S
XJu
U
A C
S .A
_
_
_
- -~
_
-
){\ ! '
i il ! ,
i
?
S S
S sR
sD * S SR N t'
a
. E EH EN> S"
EH aE
S
S
E
-
-
~ .
-
h- g 3
8
- A
0
3 3
1
A-
3 0 ' , 1
-
A
4 4 4 4 . - W Y
2 D L
- l N
E :
O
T 4 F
A 7 F E
D 5 A S
l
2 l U
. G
4 4 N
1 7 I
,, ? -
s 5 A I
l
a 1 N l
R . D A
- C R
( { S 1
3 : T
jd T I
I
g S W
g g
g g
C
- y
E
A
t
R
M s D
Q
A a g
6
1 g
~
4
2
1
H
2h A
4
)*5 .
3 3 5
%1
8
A
2
1
9
E
R
U
S
S
E
- A R
P _
A 5
^ A 2 4 T
_
- 1 0 l
0 3 l
E
, 3 I
3
,
S
I
I
F . A
W T .T
S U TI M
L N D U
CF I
S E DO
RF C E
U ,, A C1
S I
B lAH
aN
I
t
I
O R2
g , O
I , . T
S S .I
S N
2 2 U
2
R R R
a 2 OO R O
g I I
O I
I
T T T
4 s I I T II
0 1 N I I NI
5 5 UU I U
1 1 1 0
U 1
1 0 1
1
S S 00 1 0
R 0
E E E
CL
R
l S S N S
I O 0 DAE DL
I OC 0
C
I I
CC O C
P
T OO O O
T 'N T T T T
UN
VE
"E UU U U
A A A A
V V
i
LIT
VI
. . . .
T
d E "
V
I 2 3 4
' y T
AN N
VE
V
I S 2S
'VE
S S
I
I I
- II I
-
4 3
6 6 MID
I I
RID
I I u E y
)
I I C R
2
I (
I I I V I
,
H R
N SR CC
E D
CC S T
R SI
D C.
6
EI gi ,
, C
A
9
4
R S
I
D R
A
Y
S
S
C. 9 , 3 8 E
A A 2 , 8 2 L R
- - 2 C B
8 7
U T
4 4 N 4
8 1
W
S
S
E
.
! i
_.
'.: .
^*
4
NUCLEAR SERVICE WATER SYSTEM EXPERIENCE
l. ESTABLISHMENT OF BASIS FOR RN SYSTEM OPERABILITY
- RN SYSTEM PRE-OPERATION FUNCTIONAL TEST COMPLETION DATES
7/25/79 -
UNIT 1
i
11/12/82 -
UNIT 2
- NRC PRE-OPERATIONAL INSPECTION DATES COVERING RN SYSTEM
TESTING
11/03/78 -
UNIT 1 INSPECTION REPORT 369/78-33
8/16/83 -
UNIT 2 INSPECTION REPORT 370/82-19
- SURVEILLANCE TESTING IMPLEMENTATION DATES
a
1/06/80 -
UNIT 1, TRAIN A
2/06/80 -
UNIT 1, TRAIN B
, 2/22/83 -
UNIT 2, TRAIN B
2/23/83 -
UNIT 2, TRAIN A
IWV AND IWP TESTING WOULD HAVE BEEN IMPLEMENTED DURING
THESE TIME FRAMES.
- THE PRE-OPERATIONAL TESTS, lWP TESTS, IWV TESTS AND ESF
TESTS WERE OUR STANDARDS FOR ESTABLISHING AND
MAINTAINING RN SYSTEM OPERABILITY.
11. MAINTENANCE OF COMPONENTS BASED ON MONITORING OF OPERATIONAL
PARAMETERS
REFER TO LIST OF EQUlPMENT CLEANINGS
PERFORMANCE MONITORING PROGRAM BEGAN DEVELOPMENT IN
MARCH, 1984 l
l
l
________ _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _
'l.
t
.
Ill. BEGAN EVALUATING RN SYSTEM HX'S FOR FOULING EVEN THOUGH
THERE WERE NO INDICATIONS OF FOULING
DATE WHEN UNIT 1 COMPONENT COOLING (KC) HX'S WERE
EVALUATED FOR FOULING WITHOUT INDICATIONS OF A FOULING
PROBLEM
9/01/84
- DATE WHEN KC HX'S WERE CLEANED
11/84 - UNIT 1
6/85 - 7/85 - UNIT 2
- EVALUATION AND INSPECTION / CLEANING DID NOT DETERMINE
THAT KC HX'S WERE INOPERABLE
IV. RN SYSTEM OPERABILITY REEVALUATION BASED ON 1A RN PUMP TEST
RESULTS
DATE WHEN A FLOW MEASUREMENT PROBLEM ON 1A RN PUMP WAS
IDENTIFIED
10/04/85
A REEVALUATION OF OPERABILITY CRITERIA WAS BEGUN TO
REFOCUS OPERABILITY CONCERNS FROM THE RN PUMP TO THE RN
SYSTEM AS A WHOLE
V. ACTION ITEMS RESULTING FROM REEVALUATION OF RN SYSTEM
OPERABILITY CRITERIA
BEGAN THE PERFORMANCE MONITORING PROGRAM ON
l RN HX'S ON 11/01/85
'
THE RN SYSTEM TESTING PLAN WAS SUBMITTED TO
REGION 11 ON 12/01/85
THE UPDATED RN SYSTEM TESTING PLAN WAS SUBMITTED
TO REGION ll TO INCLUDE TESTING OF ALL 62 RN
HX'S AND RESOLVE 1A RN PUMP FLOW INDICATION PROBLEM
ON 12/18/85
! NOTE: IN ALL THE TESTING AND ANALYSIS DONE IN 1985, WE
DID NOT DETERMINE THAT ANY OF THE HX'S EVALUATED WERE
1
f
_ _ _ _
, .,
,
't
EQUIPMENT CLEANINGS j
!
LOWER CONTAINMENT VENTILATION HX FOULING WAS IDENTIFIED AS
ONE OF THE FACTORS IN THE LOWER CONTAINMENT COOLING PROBLEM
10/22/82
NOTE: (A) FOULING OCCURRED AT LAKE TURNOVER IN THE FALL.
ONLY TIME WE HAD TO CLEAN.
(B) BIOFOULING WAS EVIDENT DUE TO HOT AIR ON SHELL
SIDE.
- '
CONTROL ROOM VENTILATION (SHARED BETWEEN UNITS 1 AND 2)
TRAIN A TRAIN B
11/19/82 3/83 {
10/03/83 1/07/85
12/19/83 10/21/85
5/30/84 11/05/85
10/31/84
9/25/85
10/24/85
10/31/85
PENETRANT / DISPERSANT ADDED TO THE RN SYSTEM IN ATTEMPT TO
CLEAN LOWER CONTAINMENT COOLING UNITS
4/27/83
REACTOR COOLANT PUMP MOTOR COOLERS
UNIT 1 UNIT 2
12/31/84 8/10/84
11/08/85
.
ASSUMPTIONS
1. ALL SAFETY RELATED EQUIPMENT REQUIRE FLOWS CONCURRENTLY 5
"
THROUGHOUT DESIGN BASIS EVENT. g
5
s
4
m
2. HEAT EXCHANGERS DESIGNED FOR MAXIMUM POND TEMPERATURE OF 95 F.
4
FLOW AND FOULING DESIGN MARGIN
AFFECTS ON CONTAINMENT PEAK PRESSURE
(CONTAINMENT DESIGN = 14.9 PSIG)
(x10bfuHROF) (x10bIUHROF) tbs T0 OfE'
CLEAN
OfFFkbfEik 5.18 8.11 5000 8000 -
DESIGN
(FSAR) 2.94 5.00 5000 8000 12.36
75% NS
DEGRADED
FROM DESIGN 0.735 5.00 5000 8000 14.42
REDUCED FLOWS
DEGRADhDHXs 1.47 2.98 3800 6000 13.59
0 = U3 (LMTD)
U = (-00 LING, FLOW)
. ..
l-
.g
.
NUCLEAR SERVICE WATER
ESSENTIAL COMPONENT FLOW REQUIREMENTS ;
,
4
DESIGN F OWS !hkE
COMPONENT (FSAR AChl0WS
KD Hx 900 900
KC Hx 8000 6000
i
NS Hx 5000 3800
,
VC/YC CONDENSER 775 775
I
KF ES COOLER 20 15
NS ES COOLER 45 20
ND ES COOLER 45 20
NV PUMP COOLERS 28 15
- NI PUMP COOLER 20 20
i
i
!
<
, + , . _ _ . -- n - - - - - - -- ,, - - - - . ,. , ,-.----.,,--,.-,.--y.
- K ,*. ,,
ADDITIONAL DESIGN MARGINS
1. LOWER SNSW POND TEMPERATURE
2. HIGHER ICE WEIGHT
3. LOWER RWST TEMPERATURE
.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . . _ . - _ . _ _ _ _ _ . _ _ _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _____
w -
.- -
,.
s. ,
a
,, ATTACIGIENT 3
i
T. . <
PERFORMANCE MONITORING PROGRAM
- Reliability. Efficiency and
AvaiIabiiity
- Monitors the overall health of
equipment
- Development begun in March. 1984
- Tangible results already being
reaIized
.
. 3' .
l'
.
NUCl. EAR SERVICE WATER PUMP (RN) 1A
- RN Pump 1A did not meet its quarterly
IWP acceptance criteria (10/4/85)
- Replaced impeller (10/5 - 10/6/85)
- Performed new pump head curve /lWP
baseline test (10/7/85)
- Troubleshooting
- Evaluated the pump acceptance criteria
based on the actual system demand
- Conducted the pump head curve using
the 2A and 1A KC flow elements in
series with the 1A RN flow element
- Using the most conservative head curve
results 1A RN pump was declared
operable (10/11/85)
- Optimum replacement was a calibrated
84" flanged spool-piece with a 0.831
beta ratio orifice
- Installation (February 26-28,1986)
- 1A RN Pump head curve conducted with
new flow element (March 3, 1986)
- Summary - The pump was never
inoperable, fouling of the flow
element resulted in errors in the
conservative direction.
L
I
6
8
- * [
4 ;
-
3 "b- -
,h-
o C OoO
N
I ----
- - -
R
P
. e
R v
r l
N u
l
l
C 5 s.%-
0 68 '
1 d8 - q
e8
i 3 -
l- q \
f 0 p s'
i31 s
.
t s .\'
rff
eoo \
C , (' \
i
. g
.
t t .\
sss
.
,h.
n. ee r
.
- 're T T g , .\ a
ree g
4*
uvv
E
t rr
cuu \ w
aCC \ '
T f n
A s'
u
.
R udd l
W naa l
l
s
O aee
L
f
F I I
I l
's
F s
.
b
s
C S ,
'
U V i
!
Y' ,
D
A ,
N w
.
E
l
l
i
s
U.
i
t
C
I
M -
.i
,-, w
. AN !l i;
Y
D
!p! i !
, i
l
s's sr
A s
a
,
~
1 l L
!
P
f I
e
l
U .
P fI
N
d
R s
5 m
' s' 8 " p 6 " mmm
ppp i
n'
- j - 5 mmg 8 1
8 1 pp - 2 mmggg
-
3 pp
4 gg0 4 gg000
.
. l.
.
3
h
L
-
.
0
1
f
-
4000,
1 00
063
0
3
f
-
4
1
0
000
00 0 0,0,0, !
M
L o =281 o = 00024 i
h 681 1 1
.
l'
O s e@@@ s e J,
. r t r@0@@@
. hfl t
l o ... l o
.
. .
W u B t t t u
s
B . . . . .
t t t t t I.
.
.
.Y b s ff f
e efffff
e
.
- T e c
.
. 81 - R c1 7 5 R
.
. l
.
i1 02 i 37473
U t f t f .
.
s i 996 s i01 1 95
.
e r295 e r21 087
I
T O21 1 T O2221 1
e i
n
< ! ,i-
-
l
e.
o
n
!.j;{
It
li
(l
P
m.
lg
llI
I
_ I
-
l -
l
l
- _
.m
.
!
!
'
' h
p
g
.
. L
i:ln g J' .I
- f
.
? M 3.
.
g b, ,
g <"
.
3
l
?
l
!.
.
.
L
.
P d Nv h x Q $ H $. m y m @
-
.
'
t ',
. .
l'
.-
CONTAfNMENT SPRAY (NS) HEAT EXCHANGER
- 1A NS Heat Exchanger had a high
differential pressure
- Commission expressed concerns of
biological attack of s t a i ti l e s s steel
'
tubes
- Testing Performed:
1. Structural Integrity Test
2. Minute Leakage Test
3. Heat Balance Test
- Structural Integrity and Minute
Leakage Test indicated insignificant
leakage
- Visual Examination of the tubes
- Heat Balance Testing quantified the
extent fouling had occurred
"
- Peak Containment Accident Pressure
CLOTIC) calculations showed the heat
exchanger could still perform its
function
- Cleaning iterations
- Tested and cleaned the other NS heat
exchangers based on 1A experience
- * Summary:
1. NS Heat Exchangers are intact
2. The NS Heat Exchangers were
fouled; however reanalysis proved
operability
j
.
gc* - -
y e,- ---- --e- - m --pq a--n-- - me---w --e,--r 1 =ya---e- -~ ------e-- P
_ . . _ . _ . _ .
_ _ _ _ _ _. _ -_.
.*
. , , ,
._
Containment Spray lleat Exchanger Testing
i
t
i
.,
- lleat Exchanger
.
,
,
~
I
i
Temperature)
m i Y a
I i 'f
,
l
,
!
Containment
Spray Pump Flow
f
I
!
>
)/
FWST
3
(400,000 gal.)
'
i
l
.
",
,
"
. McGuire Nuclear Station
'
4/,w
80 -
74.7
!!!!
I;; p V g!f V/
... .... ....
f h$
d d d ild $/) N
As Found C!ng 1 Cing 2
40
36
.
[
- 7 27 27
! ::- Fy F/ 9
'
..
/ / /j/
-
/ p/)/ / '-9
- 1s s s s 4. . . . .
_ _ - . .-. . _ . - _ _ - _ _ - . _ . - , ,
___ _ ___ .
l*
'
s
P
e
OTHER HEAT EXCHANGERS and FLOW BALANCE
- Began evaluation and procedure
generation for testing essential heat
exchangers (10/24/85)
- 1A Train RN flow balance performed
aligned to low level intake (12/17/85)
- 1A NV Pump Speed Reducer Oil Cooler
cleaned (12/20/85)
- Conducted test using NSWP as suction
(1/27/86)
- Inadequate flow to some heat exchangers
- Reanalyzed the necessary flow rates to
the KC and NS heat exchangers
- Design review of alignment
configuration to properly conduct flow
balance to meet all design assumptions
- Conducted the 1A Train flow balance
throttling flow to the KC and NS heat
exchangers, aligned to NSWP with 6000
gpm supplied to other unit (1/28/86)
- Performed other flow balances
(1/30 - 2/28/86)
- Began extensive cleaning, testing and
inspections of all essential heat
exchangers (2/3/86)
Total Cleaned / Tested / Inspected: 54
Total Number of Hea t -Ex change r s : 62
- Summary:
Cleaning and testing of all essential
RN system components is on schedule
to meet March 31, 1986 completion
.- . _ - _ - . - . .- - __ - . - -- - . _
d
.
.
.
%
s
-
.
l
i
i ATTACHMENT 4
!
't
NRC Inspection Team Confirmatory VA Calculations
.
- Calculations were performed to evaluate the containment spray heat exchanger UA
j value used in containment pressure calculations performed by Westinghouse for
- Duke Power on November 28, 1985 and January 17, 1986. The following nomenclature
is used in the subsequent calculations:
.
Nomenclature
i
,
A -
heat exchanger area
Di -
tube inside diameter
i Do -
tube outside diameter
.
i
1 Fi -
tube inside fouling factor
Fo -
shell side fouling factor
F
oAPP
-
appropriate shellside fouling factor
i
i G -
mass flux
4
! hi -
tube inside heat transfer coefficient
ho -
shell side heat transfer coefficient i
! K -
water thermal conductivity ,
K -
stainless steel thermal conductivity
ss
- Pr -
water Prandtl number
1
Re -
Reynolds Number GD
D
l y
'
i oFo --
two standard deviation uncertainty in Fo
u
-
liquid viscosity
!
t
!
f
'!
l
i _ _ _ ..____._ _ _ _._. _. . _ _ _ . _ _ _ _ -_ ._ _ _ _ _ _ _ _ ._.. _ -_ _ _ ..- _ _ a
. _ _ . . __ _ . . ._ . . . . - . _ -.
- .
i-
!
1 -. Attachment 4 2
i
!
J
j The UA design value for this heat exchanger is 2.95 x 105 Btu /h/ F while
Duke provided Westinghouse with a degraded value of 7.35 x 10' Btu /h/ F,
24.9% of the design value. Experimentally determined UA (11/22/85) values
indicated that the actual degraded value was ~8.77 x 10' Btu /h/ F, 29.7% of
f
the design value.
Confirmatory UA calculations were performed by initially determining a
f design heat transfer coefficient for the shell side of the heat exchanger.
I This was done by using design value fouling factors, and assuming that the
I tube side heat transfer was correctly predicted by the McAdams equation at
j the design conditions. ,
'
hiDi
k =0.23 ReD.8Pr 1/3
. (1)
- The UA for the heat exchanger is
! UA = A (2)
- 1_ + pg 1_ + F4 po + F o +D g In p g
hg Di hi Di 2K
ss
UI
For the design condition, all values (including UA) are known except hg ,
[ which was determined to be 918 Btu /h/ F/ft2 using equation (2).
The information supplied to Westinghouse by Duke was acquired from experi-
- mental testing of heat exchanger 1A on 11/22/85. The data from this
e<periment was used to determine an appropriate value for the degraded UA by
'
j first determining the as-tested fouling factor. In order to do this,
i experimental flow rates, temperatures, etc. had to be used to determine both
tube side (h4 ) and shell side (ho) heat transfer coefficients appropriate
! for the test. Tube side heat transfer coefficients were determined using
i equation (1) evaluated at the test conditions. Shell side heat transfer
i coefficients were assumed to scale as:
Pr
13
f hg po - Re D
i K (3)
j This equation is used frequently in determining shell side heat transfer for
'
shell and tube heat exchangers. Equation (3) was evaluated at both design
and test conditions, and an hg for the test was calculated from the design
h, determined above. Equation (2) was then used to determine the fouling
factor appropriate for the shell side under as tested conditions assuming
! the tube side fouling factor is the design value of .0005 (this assumption
- actually has no impact on the final VA since the two fouling factors are not
1
'
a function of flow and fluid conditions). The shell side factor was
determined to be
F
g
= .00912 (4)
,
. - -
Y -. - _ - - - - - , . , , -
. . . , . ,
-_ - . - - . - . - ___ - - _ _ __ -
.
.
.
l
Attachment 4
'
. 3
,
l
l
for the experiment vs. the 0.001 design value. In addition to this
calculations, the experimental error associated with the testing equipment
and procedure was used to determine an uncertainty value for F . gThis
calculation was performed using propagation of errors (see for example
! Beers,1957, " Introduction to the Theory of Error") through the equation
- (5), the energy balance on the NS side of the heat exchanger (only the NS
flow was used to determine overall heat flow).
'
o
Q = mCp (T out -Tin) (5)
The uncertainty in temperature measurements were given to the NRC team
by licensee representatives as .4 F including both RTD, and signal
conditioning equipment error. These RTD's were apparently calibrated before
testing, which increases confidence in the temperature measurements.
Additionally, errors in the flow measurements were also included. Handbook
uncertainty values for uncalibrated orifice plates are typically 1%-2.5% of
measured flow. In addition to this, there are uncertainties associated with
the other instrumentation necessary to make the flow measurements (DP cells,
) readouts,etc.). The orifice plate was an uncalibrated process device so it
was estimated the overall uncertainty was ~5% of the measured value. Each
of the uncertainties stated above were treated as one standard deviation
.
(lo) uncertainties. It is believed that a two standard deviation (2o)
i uncertainty bound should be applied in order to insure conservatism (two
standard deviations give a 95% certainty of the measurement). The 2a value
4 for Q was found to be ~12%. Additionally, since design heat flow was based
i solely on calculations and not on tests. It was assumed that a 2.5% error
(lo value) was present in the design heat flow determination. It was also
assumed that equations (1) and (3) could be used to correctly scale with
temperature level and flow rate (0 uncertainty was assigned to this
- process). The two errors above, experimental and design, were used to
determine overall error in F by g
propagating errors through the calcula-
'
tions described above. The two-standard deviation uncertainty in F was
o
4
determined to be:
oF g = .00149 (6)
i for the uncleaned case of heat exchanger 1-A. An appropriate UA value for
the Westinghouse calculations was then determined by using:
FoAPP = F g + oF g (7)
i These values were determined for three cases: unit 1-A before cleaning,
unit 1-A as it existed after last cleaning, and unit 2-B. The table below
- summarizes these results (in all cases, RN flow was assumed to be 4800 gpm).
J
-_ - -. . - . - . - - _ _ . _ _ _ _ - - _ - , . , _ _ . - - _ _ - - - . _ - _ - . - - - _ . _ - , _ - . - _ , . , _ _ - -
.
'.
,
Attachment 4 4
Summary of Calculations
UNIT STATUS F
g oFo UA
1-A uncleaned (11/22/85) .009 .0015 8.18 X 105
1-A cleaned (01/16/86) .0033 .0007 1,63 X 105
2-8 uncleaned (01/24/86) .011 .0127 7.16 X 10'
Westinghouse input 7.35 X 105
The UA value calculated for the 2-B uncleaned case is slightly below that
given to Westinghouse on 11/28/85 and 01/17/86. However, if the containment
pressure calculations performed on 01/17/86 are used as a starting point,
and the containment pressure change with VA change is similar to that noted
in the 3 calculations performed on 11/28/85, the peak containment pressure
can be estimated for a UA value of 7.16 X 105 These calculations estimate
that the peak containment pressure for this UA value would be approximately
P = 14.56 psig, still below the 15 psig limiting value.
The calculational methods used to evaluate heat exchanger performance appear
to be reasonable. However, when calculations are being performed to deter-
mine heat exchanger performance at reduced flow, it is also necessary to
apply appropriate fouling factors to heat exchangers which are suspected of
being fouled. This has not been done in previous Duke calculations. As an
example, the inspection team looked at the charging pump speed reducer oil
cooler. Duke has found the oil inlet temperature to increase from 141'F to
166 F when RN flow to the heat exchanger is reduced from 20 gpm to 10.7. _In
addition to the reduced water flow, the effect of fouling should also be
considered. Confirmatory calculations were performed assuming both reduced
flow and a fouling factor of ~.008 on the RN side and .001 on the oil side
(design fouling factors were presented as a sum of Fg +F4 =.0025). The RN
fouling factor is an estimate based on findings in the uncleaned containment
spray heat exchanger (F, =.009) and recognizing that continuous water flow
through the oil cooler might reduce fouling somewhat. A summary of the
maximum oil temperatures is presented in the following table.
A calculation with the RN cooling water temperature reduced to 65 F is given
to demonstrate the cooling water temperature effect on heat exchanger
performance. As can be seen in the below table, the reduction in RN
temperature from 95 F to 65 F has a significant impact on oil temperature.
A similar effect would be seen in other heat exchangers in the train
(although not exactly the same magnitude).
Comparison of 011 Cooler Assumptions
Cooling Water Inlet Temp. Flow (gpm) F
9
T oj) ( F)
95 F (Design) 20 .0015 141
'95'F 10.7 .0015 166
95*F 10.7 .008 185
65'F 10.7 .008 155