IR 05000289/1998009
| ML20207H768 | |
| Person / Time | |
|---|---|
| Site: | Crane |
| Issue date: | 03/05/1999 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20207H745 | List: |
| References | |
| 50-289-98-09, 50-289-98-9, NUDOCS 9903160094 | |
| Download: ML20207H768 (20) | |
Text
~-
..,
,
.
U.S. NUCLEAR REGULATORY COMMISSION REGION 1 Docket No.
50-289 License No.
DPR-50 Report No.
98-09 Licensee:
GPU Nuclear,Inc. (GPUN)
Facility:
Three Mile Island Station, Unit 1 Location:
P. O. Box 480 Middletown, PA 17057 Dates:
December 27,1998 through January 30,1999
. Inspectors:
Wayne L. Schmidt, Senior Resident inspector Craig W. Smith, Resident inspector
i Approved by:
Wayne L. Schmidt, Acting Chief Projects Branch No. 7 Division of Reactor Projects i
i
.
9903160094 990305 PDR ADOCK 05000289 PDR O
-. -
.,. -. - - - -. -.
. - - _ - -.. _ - -. - - - - - - - - -
,
.
EXECUTIVE SUMMARY Three Mile Island Nuclear Power Station Report No. 50-289/98-09 This inspection included aspects of licensee operations, engineering, maintenance, and plant support. The report covers a six-week period of resident inspection.
GPU Nuclear (GPUN) operated Three Mile Island Unit 1 (TMI) safely, at essentially 100 percent power throughout the inspection period.
Operationg The control room staff operated the unit safely, properly responding to several equipment
.
failures. Shift supervision appropriately evaluated operability for the failed equipment.-
{
(Section O.1)
'
The control room operators responded well to two separate failures of the integrated
-
control system (ICS), taking effective actions to retum the unit to normal steady state l conditions. (Section 01.1)
--'
GPUN did not implement adequate controls over the emergency boration path from the boric acid mix tank since removal of the technical specification (TS) requirement and incorporation of the design basis into the updated final safety analysis report (UFSAR).
'
No procedure existed to verify the temperature of the boric acid piping from the boric acid mix tank (BAMT). Further, in two examples GPUN did not meet proceduralized guidance on the availability of the BAMT as an emergency boration path. GPUN reported the condition as outside the design basis in accordance with 10 CFR 50.72 and i
planned to submit a licensee event report (LER). This issue was an apparent violation pending review of the LER. (eel 98-09-01) (Section O2.1)
Maintenance GPUN conducted observed maintenance and surveillance activities well, in accordance
.
. with approved plant procedures. (Section M1)
GPUN used the corrective action process (CAP) to properly document equipment
-
i failures and problems as part of the maintenance rule. (Section M1.1 )
j l
l The quarterly maintenance rule report for the last quarter 1998 provides good
.
information on the status of systems that required enhanced monitoring. No risk
- significant maintenance preventable functional failures (MPFFs) occurred during that
period. (Section M1.1)
Dai!y trending of the core damage frequency (CDF) appeared useful in planning and
.
identifying areas where the cumulative effects of out of service equipment can be minimized. (Section M1.1)
ii i
- -_
. - - -
_ - -
.
.,
_
_
-.
-
-
. -.
. -.
-.. _ _. -
.
..--.. -
-.-.- - -. -. _...---
,
,
i
!
GPUN responded well to the clogging of a decay heat river stainer. However, this was l
-
considered the second instance of a repeat MPFF on this strainer within one year. As
-
such GPUN followed their CAP for a significant failure. Root cause determination and corrective actions were underway and will be reviewed as GPUN completes the process i
as an unresolved item. (URI 98-09-02) (Section M2.1)
GPUN met their approved commitments, described in NRC safety evaluation report for
.
,
the TS Amendment 122 for the testing the of the fuel handling building (FHB) emergency l
safety feature ventilation system and for ensuring FHB operability prior to refueling
r operations. (Section M2.2)
+
Ennineering
Site engineering personnel responded well to several equipment failures including the
-
ICS module and the DR strainer clogging. (Section E1)-
.
The methodology and detail of the updated voltage regulation study demonstrated i
conformance of the electrical distribution system with the UFSAR design basis.
'
'
However, outstanding issues identified as a result of the calculation required additional review of corrective actions documented in the CAP. (Section E1.1)
The EFW system remained operable, assuming that in October 1997 GPUN used
!
,
'
appropriate methods, although different from the original method, to calculate the l
required loss of feedwater accident (LOFW) analysis flowrate. This analysis resulted in GPUN lowering the required flow to both OTSGs following a LOFW. However, GPUN
- failed to complete a safety evaluation before implementing this analysis. More recently, between September 1998 and January 1999, GPUN also failed to recognize that calculations showed that the installed pumps had capacities lower than previously assumed, specifically the TD pump could not supply the full accident analysis flowrate, by itself. The NRC staff is concemed that these changes may have introduced an unreviewed safety question, which required NRC staff review prior to implementation, in accordance with 10 CFR 50.59. (eel 98-09-04) (Section E4.1)
GPUN engineering did not perform adequate reviews of details in preparation of the safety evaluations completed after the EFW change had been implemented.
Specifically, engineering did not identify the needed change to the TS bases and UFSAR Section 10.2 and subsequently, completed incorrect changes to the UFSAR and TS bases regarding the capability of the turbine driven pump (i.e., it was not full j
capacity.) (Section E4.1 )
Plant Suonort
!
!
I Plant housekeeping remained good. No negative radiological conditions were identified
-
during routine plant tours. (Section R1)
!-
iii
.
.
-
,
2.-.y er
--
e-
+
- -, - -
- - -
- *,
-
-n
--- - -. -
_ - _..
..
_._.. _ _ _ _. _ _ _
_.
_
-
_
_._ -
. _.
.
_
_ _.
,
,
I
lr TABLE OF CONTENTS EXEC UTIVE SUMMARY....................................................... il l
TABLE OF CONTENTS........................................................ iv l
. l. Operations............................................................... 1
Conduct of 0perations............................................ 1 O1.1 - Failures of Integrated Control System Resulting in Plant Transients c
(717 0 7).................................................. 1 O2 Operational Status of Facilities and Equipment........................ 2 O2.1 Emergency Boration Review: (Open) Apparent Violation 58-09-01; Failure to Follow Soluble Boron Control Procedure (71707).............. 2 I I. M aintena nce............................................................. 5 M1 Conduct of Maintenance (61726, 62707)............................. 5 M1.1 Maintenance Rule Review (62707)............................ 5 M2 Maintenance and Material Condition of Facilities and Equipment..........
!
M2.1 Decay Heat River Strainer Clogging: (Open) Unresolved item 98-09-02; Review Root Cause Analysis and Corrective Actions for Repeated Decay Heat River Strainer Clogging (62707)......................... 6 M2.2 (Closed) Unresolved item 98-02-01 Fuel Handling Building Engineered Safety Feature Building Ventilation System Testing (92902)........
111. Enginee ring........................................................... 8 E1 Conduct of Engineering (37551).................................... 8
E1.1 Alternating Current Voltage Regulation Study (37551)............. 8 i
E1.2 (Closed) Licensee Event Report 98-010-00; Potential Violation of Design
Criteria During Single Auxiliary Transformer Operation (92700)...,. 10 E4 Engineering Staff Knowledge and Performance....................... 11 E4.1 (Closed) Unresolved item 98-03-02; Emergency Feedwater Flow i
Requirements Safety Evaluation Adequacy; (Open) Apparent Violation i
98-09-03; Introduction of an Emergency Feedwater Unreviewed Safety
,
Q uestion (92903)......................................... 11 IV. Plant S u pport........................................................... 13 R1 Radio logical Protection and Chemistry (RP&C) Controls (71750).........
V. ~ M anagement Meetings.................................................... 13 X1 Exit Meeting Summary........................................... 13 INSPECTION PROCEDURES USED........................................... 14
'
ITEMS OPENED, CLOSED AND DISCUSSED................................... 14 LI ST OF ACRONYM S U S E D................................................. 15
.
-.
..
.. -
_
.
_ _ _ _. _ _. _ _ -. _ _.. _.
_ _. _. _ _. _.. _ _ _ _ _
r e
!
o I'
)
,-
Rooort Details Summary of Plant Status l
L GPU Nuclear Inc. (GPUN) operated Three Mile Isiand Unit 1 (TMI) at 100 percent power L
. throughout the inspection period.
_
l.
)
I.. Operations j
!
Conduct of Operations (71707)
The control room staff operated the unit safely, properly responding to several equipment
,
failures. Shift supervision appropriately evaluated operability for the failed equipment.
01.1 Failures of Inteurid Control System Resultina in Plant Transients (71707)
a.
Inspection Scope I
.
On January 13 and 14, the inspectors observed control room operator response to two separate failures of the integrated control system (ICS) that resulted in plant transients.
The ICS normally operates to provide fully automatic control of: reactor power (core thermal power) by moving control rods; once through steam generator feedflow by positioning feed control valves; and generated electrical load by positioning the steam valves supplying the turbine generator. While in automatic the reactor power, feedflow, and generator output are all controlled by the unit load demand (ULD) setpoint, which the operator sets for a given generator megawatts electric (MWe) output. Neutron error is the difference between the reactor power demand signal, from the ULD, and the
' indicated reactor power from the power range nuclear instruments.
b.
Observations and Findinas
- On January 13, control room operators planned to place the ICS in manual to support scheduled system maintenance. While shiftingto manual, an erroneous neutron error signal developed and resulted in the outward movement of the reactor control rods.
The control room operators took manual actions to retum the plant to a stable condition, by initially placing the rod control system in manual, stopping the outward motion.
Reactor power, as measured by the power range nuclear instruments, momentarily increased to 101.5 percent before the operator could respond to drive the control rods in and reduce reactor power to less than 100 percent. The momentary increase in reactor power caused an increase in the reactor coolant system (RCS) pressure and the pressurizer spray valve opened, as designed, at 2205 psig to lower pressure to the normal steady state value of 2155 psig. Two faulty analog memory modules caused the erroneous neutron error signal.
On January 14, a control room operator observed core thermal power slightly greater j
l than the 100 percent rated value of 2568 megawatts thermal (MWt) and properly lowered j
L the ICS ULD setpoint to achieve an approximate 6 MWt decrease in core thermal j
l
'
,
.,
.
. - - -.
..-..
.
.
. -.
.
..
-
.
-
-
-
-
.
.
.. -
-
.
,
,
power. The ICS responded by reducing core thermal power by 29 MWt at which point core thermal power began to slowly increase with no corresponding adjustment to the ULD signal. In response to these observations, the control room operator took manual control of the ICS, which removed the signal that caused the increase in core power.
Troubleshooting determined that a failed circuit between the ULD input and output signals caused the increasing demand signal.
In both cases instrument and control (l&C) technicians replaced the faulty components and the operators returned ICS to automatic control. The operators properly documented both issues in the GPUN corrective action process (CAP).
c.
Conclusion i
,
The control room operators responded well to two separate failures of the ICS, taking effective actions to retum the unit to normal steady state conditions.
Operational Status of Facilities and Equipment O2.1 Emeraency Boration Review: (Open) Accarent Violation 98-09-01: Failure to Follow l
Soluble Boron Control Procedure (71707)
a.
Inspection Scooe The inspectors reviewed the emergency boration requirements and completed a system walkdown of the applicable portions of the chemical addition and sampling (CA) system.
TMI emergency operating procedure (EOP) guideline procedure 1210.10 referenced three approved emergency boration paths: from the borated water storage tank (BWST),
l from the boric acid mix tank (BAMT) in the CA system, or from the liquid waste system
'
reclaimed boric acid tank (RCBAT). GPUN no longer uses the RCBAT.
l b.
Observations and Findinas in review of the emergency boration path the inspectors found that GPUN received Technical Spedfication (TS) Amendment 196 in 1995, which removed the requirements for having the BWST and either the BAMT or RCBAT available for emergency boration operable. The BWST remained as a TS required emergency boration source, as part of the emergency core cooling TS. However, the NRC's basis for allowing the removal of the BAMT and RCBAT from the TS included that the pertinent design requirements and
,
l bases would be incorporated into the updated final safety analysis report (UFSAR). The
'
TS bases, prior to the amendment, stated that the emergency boration was needed to take the unit to the cold shutdown boron concentration, based on the most reactive control rod withdrawn at any time in core life. The TS bases also specified the required boric acid volumes and concentration required in the BAMT or RCBAT along with the temperature requiremente for the tanks and associated flow path,... _. _ _ _. _
. -. _ - - _ _ _ _
_ _ _ _. _ _. - _ _. _ _ _ _ _.. _. _. _ _. _. _.
.
'.
.
I
In review of the UFSAR the inspectors found:
? Section g.2 for the CA system clearly discussed the BAMT volume,
=
concentration, flow path and temperature / heat tracing requirements to achieve a cold shutdown RCS boron concentration at any time in. core life with the most reactive control rod withdrawn. These requirements had been properly transferred from the TS to the UFSAR. This section also stated that the minimum
~
boron solution volume and concentration would be specified in the core operating
'
'
' limits report (COLR) for the given plant operating cycle.
Section g.1 for the makeup system was less clear as far as the design
-
_
requirement for the BWST. The section stated that the BWST was an altemate boration bath to the BAMT.
In review of GPUN procedures the inspectors found:
l The EOP guidelines list the BWST, the BAMT or the RCBAT as emergency
.
boration sources and provided direction to procedure 1103-4, " Soluble Poison Concentration Control," for additionalinformation.
Procedure 1103-4, " Soluble Poison Concentration Control":
Did not provide any instruction for using the BWST as an emergency
-
boration source. The inspector found that operators understood the l
emergency boration path from the BWST; however, this path was not clearly proceduralized. There were no supporting calculations to determine that amount of BWST water that needed to be injected to achieve the equivalent to the COLR cold shutdown requirement, with the (
most reactive control rod fully withdrawn.
Stated in the administrative section that the reactor shall not be critical
-
unless either the BAMT or the RCBAT had the COLR amount of boric acid and had a operable flow path to the make-up tank (MUT), including meeting the temperature requirements, with one operable transfer pump.
GPUN did not have a specific surveillance to ensure that the boric acid piping
temperature met the UFSAR specif; cation. Procedures were in place to monitor the BAMT fluid boron concentration and temperature, along with the pump capabilities through inservice testing (IST) and refueling outage testing.
The inspector identified two instances where GPUN failed to comply with the Soluble l
Poison Concentration Control procedure which required that the reactor shall not be l
critical unless the BAMT, and one transfer pump are operable with the associated l
piping 10*F above the crystallization temperature for the sampled boric acid L
concentration:
'
<
._. _
_ _.
_ _.. -
_
.
..-
. _.
__
.
,
,
l l
l In walkdown of the system the inspectors determined that the piping near the
,
,
I discharge of both of the two positive displacement transfer pumps (from the i
BAMT to the MUT) was not above 127*F, which was approximately the temperature requirement for the 17,000 ppm boron concentration in the tank.
i The inspectors brought this observation to the shift supervisor on December 31, 1998. During a plant tour with the site vice-president on January 15,1999, the inspectors pointed out that the discharge piping was not at the required temperature as stated in the UFSAR. While it appeared that GPUN was taking some actions based on prior discussions, the final outcome of determining that the piping temperature did not meet the UFSAR requirement was not reached until January 20, when the system engineer, after taking some confirmatory.
temperature measurements, reported the issue via a CAP. GPUN determined through a plant review group (PRG) meeting that this was a condition outside the
design basis as described in the UFSAR and reported it in accordance with 10 CFR 50.72 on January 22,1999.
As corrective actions GPUN ran the quarterly IST on the transfer pumps to prove l
that the line had not been blocked and took other actions to improve the heat j
tracing and insulation in the area of the pump discharges.
The inspectors also observed that GPUN took both boric acid transfer pumps out
.
of service to allow downstream valve work on January 20. The operators logged that they were switching the emergency boration path to the BWST in accordance with an on-line maintenance risk document. In review of this document and procedures, the inspector found that the contingency action in the risk document directed that the valves be open to align the makeup pump to the BWST in the event that emergency boration was necessary. This did not appeat to be the proper use for a contingency plan since, while simple, these steps were not in an approved procedure nor were they reviewed as a change to a procedure would have been.
These two instances represent an apparent violation (EEI) of TS procedural usage requirements and will remain open pending receipt and analysis of the GPUN licensee event report (LER) addressing the piping sections that had temperatures below the UFSAR requirements. (eel 98-09-01)
c.
Conclusion GPUN did not implement adequate controls over the emergency boration path from the boric acid mix tank since removal of the TS requirement and incorporation of the design basis into the UFSAR. No procedure existed to verify the temperature of the boric acid piping from the boric acid mix tank. Further, in two examples GPUN did not meet proceduralized guidance on the availability of the boric acid mix tank as an emergency boration path. GPUN reported the condition as outside the design basis in accordance with 10 CFR 50.72 and planned to submit a licensee event report (LER). This issue was an apparent violation pending review of the LER.
i.er--w-*-
e
=
-*r+*
-4
=*
7T--~
'-- --
-*~
l
.. -. _. _.. _ _. _
,
,.,
e l
'
11. Maintenance (
M1 Conduct of Maintenance (61726,62707)
!
j GPUN conducted observed maintenance and surveillance activities well, in accordance with approved plant procedures.
l M1.1 Maintenance Rule Review (62707)
a.
Inspection Scope
!
The inspectors reviewed the equipment failures during the period and the Maintenance Rule report for the last quarter 1998. Also the inspectors reviewed the use of core damage frequency (CDF) tracking as it related to on-line maintenance, testing, and equipment problems.
]
l b.
Observations and Findinas GPUN personnel used the CAP to properly identify the ICS failures (Section 01.1) the BAMT heat tracing issue (Section O2.1) and the clogged decay heat river (DR) strainer (Section M2.1). The initial CAPS provided good documentation of the issues. All three
'
systems were in the maintenance rule scope. Prior to these failures GPUN considered the ICS and DR systems as structure, systems, or components (SSCs) in the (a)(1)
'
status due to previous reliability issues.
For ICS, GPUN believed that the past and current failures were random and that
,
'
preventive maintenance (PM) items would not be effective at identifying the failuras prior to occurrence. The system will remain in the enhanced monitoring status until the end of the Fall 1999 outage.
For the DR system the recent strainer clogging was similar to the January 1998
issue that caused the system to be placed in the enhanced monitoring mode.
The maintenance department and engineering prepared the quarterly maintenance rule report for the last quarter 1998 which provided appropriate information on the status and plans to trend and correct issues on the systems currently in the (a)(1) status. No significant maintenance rule preventable functional failures (MPFFs) occurred over the quarter. Of the four MPFFs that did occur, two were due to emergency light batteries that would not hold a charge and two were minor issues on instrument air valves.
,
GPUN used a monthly plot to plan for and trend the effect of system outages, testing, or component failures on the CDF. The graphic provides a daily and a monthly average CDF and indicated, as a planning tool, where the most significant benefit would be on reducing out-of-service time or where moving planned work could be effective at reducing CDF.
l l
..
-. _ _
- - -.
... -.. -
-.
. - -.- - - - --- --
--
- -.>.-
-
i.
'
c.
Conclusion GPUN used the CAP to properly document equipment failures and problems as prA of the maintenance rule.
The quarterly maintenance rule report for the last quarter 1998 provided copropriale -
)
l good information on the status of systems that required enhanced monitoring. No risk
'
(
significant MPFFs occurred during that period.
e f
Daily trending of the CDF appeared useful in planning and identifying areas where the cumulative effects of out of service equipment can be minimized.
M2 Maintenance and Material Condition of Facilities and Equipment M2.1. Decay Heat River Strainer Cloaaina: (Open) Unresolved item 08-09-02: Review Root Cause Annivsis and Corrective Actions for Reoeated Decav Heat River Strainer
!
Cloaaina (62707)
j i
a.
Insoection Scooe.
The inspectors reviewed the GPUN identified high DR strainer differential pressure, which occurred while running the 1 A DR pump during high river level and flow
]
conditions.
b.
Observations and Findinas
- FoH0 wing the identification by an auxiliary operator that the strainer appeared clogged, the control room staff secured the pump. The pump had been running to increate the discharge flow due to environmental discharge temperature considerations (i.e., it was not being used to perform its function of removing heat from the decay heat closed cooling (DC) system heat exchanger). The operators declared the system inoperable, i
. entered the appropriate limiting condition of operation (LCO), realigned the cooling water to the A make-up (MU) pump and wrote a CAP.
River flow and level conditions at the time of the clogging were high, with large amounts of material being removed from the intake flow by the traveling screens.
Maintenance personnel cleaned the strainer and returned it to service in approximately 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br />.' Operators completed the quarterly IST on the system prior to declaring it operable.
In review of strainer performance history, the inspectors found that this same strainer l-had clogged in January 1998. GPUN had written a CAP on that condition and had taken corrective actions to open the backwash valve on any stainer when the associated pump started.~ The backwash valve was designed to cycle open on routine intervals and to open on a high differential pressure. GPUN recognized, after the clogging that this y
corrective action had not been effective.
r
!
l
,
_
,
-
.
- -
_
'
,
,,
l
GPUN considered this a repeat significant failure and pulled together a multi-disciplined team including engineering, operations, and maintenance to begin a review. The team appointed a root cause leader to conduct the overall assessment of the situation in accordance with the CAP. Engineering organized this meeting well, including presentation of information from the strainer manufacturer and the strainer clogging history. Initial GPUN assessment was that the 1 A DR pump had a lower discharge head than the 1B DR pump and therefore did not have as much pressure backwashing the strainer. GPUN planned to rebuild the 1 A DR pump during the upcoming outage or possibly to replace it with a higher head pump, based on recommendations of the root cause team. ' Further, the manufacturer recommended decreasing clearances in the rotating assembly to allow less bypass flow and more backwash flow. GPUN was in the process of developing the paper work needed to make this change.
The inspectors considered the strainer clogging an unresolved item (URI) pending review of the GPUN root cause assessment and final corrective actions to prevent
{
recurrence. (URI 98-09-02).
.
c.
Conclusion
,
i GPUN responded well to the clogging of a decay heat river stainer. However, this was j
considered the second instance of a repeat MPFF on this strainer within one year. As j
such GPUN followed their CAP for a significant failure. Root cause determination and i
corrective actions were underway and will be reviewed as GPUN completes the process j
as an unresolved item.
M2.2 (Closed) Unresolved item 98-02-01 Fuel Handlina Buildina Enaineered Safety Feature
Buildino Ventilation System Testina (92902)
}
a.
Inspection Scope a
The inspectors opened this item to allow review of the required testing on the fuel handling building engineered safety feature (FHBESF) ventilation system, used during i
refueling to mitigate the consequences of a refueling accident on the fuel handling i
building (FHB) fuel handling floor.
i
The inspectors reviewed the testing specified in the NRC's safety evaluation report l
(SER) for TS Amendment 122, dated December 12,1986, which implemented the TS on
' the FHBESF. The inspector also reviewed the GPUN testing conducted during system
,
preoperational testing and the procedures used to verify FHB integrity prior to refueling operations.
b.
Observations and Findinas
Based on the review of the SER the inspectors concluded that GPUN met their commitments to ensure that the system would function. Specifically, the NRC staff in the i
,
- SER required only a preoperational negative pressure test and procedural commitments
'
<
!
I y-m r
,
y - -- - _ - -
-
--
.
-- ---
_,
o
.
.
-
.
,
- ,
,,
to ensure that any significant or unevaluated changes to the FHB physical arrangement have not taken place pris to refueling.
The inspector reviewed the preoperational test 141/19 completed in January 1987 and the fuel handling procedure,1505-1, and radiation monitor test procedures, finding them adequate to meet the commitments as outlined in the SER.
The inspectors did note that GPUN was looking at enhancing the instrumentation of the FHB to allow monitoring the ability of the system to draw a negative pressure, however this instrumentation and testing would be strictly voluntary.
c.
Conclusion GPUN met their approved commitments, described in NRC SER for the TS Amendment 122 for the testing the of the FHBESF ventilation system and for ensuring FHB operability prior to refueling operations.
lll. Enaineerina E1 Conduct of Engineering (37561)
Site engineering personnel responded well to several equipment failures including the ICS module and the DR strainer clogging.
E1.1 Altematina Current Voltaae Reaulation Study (37551)
a.
Inspection Scone The inspectors reviewed GPUN Calculation C-1101-700-E510-010, TMI-1 Altemating Current (AC) Voltage Regulation Study, to verify the adequacy of the calculation and the conformana of the electrical distribution system with the design basis as described in
,
the UFSAR.
The UFSAR described the design basis capacity and capability of the offsite power system and onsite distribets 1 system, in part, as being adequate to automatically start he well as operate all required safety loads within their required voltage ratings in the event of an anticipated transient, or an accident, without manual shedding of any electrical loads. The UFSAR further described the onsite distribution system as consisting of two full-size auxiliary transformers connected to different 230 kilovolt (Kv)
offsite buses that provide a source of power for startup, operations, shutdown, and after hutdown requirements. Each auxiliary transformer has two isolated secondary s
windings, one at 6900 volts (v) and one at 4160 v. The 6900 v auxiliary system was designed solely for the reactor coolant pump motors. The 4160 v auxiliary system has five bus sections.' Ruses 1A,18, and 1C supply non safety-related BOP loads. Buses 1D and 1E are the.wo redundant Class 1E safety-related electrical systems with the preferred power sources from the 230 Kv offsite grid through the two auxiliary i..
.
..
_ _ _ _ _ _ _ _ _
,
_
,
.,...
.
,
.
transformers. The 4160 v buses in tum feed the 480 v auxiliary system through 4160/480 y trarrdormers.
.
b.
Observations and Findinas The inspectors found the calculation methodology acceptable to demonstrate the adequacy of the 41SO v voltage regulation when supplied by offsite power sources.
l Engineering performed calculation C-1101-700-E510-010, TMl-1 AC Voltage P.egulation j
Study, to support updating the existing degraded grid analysis study and to demonstrate j
the adequacy of voltage regulation of the auxiliary power system when supplied by
'
offsite power sources. The calculation required the performance of several tasks:
demonstration of the acceptability of the auxiliary transformers for the most
limiting combinations of: one or two transformer alignments; minimum expected
grid voltage; normal, accdont, and shutdowr, loading; steady state and accident j
block loading sequence, demonstration of the acceptability of the safety related 4160 v bus degraded a
voltage relay settings, determination of the maximum BOP and safety related 4160 v bus loading with a
minimum expected offsite grid voltage, single auxiliary transformer, under normal operation, and demonstration of adequate current carrying capacity of the power distribution a
system under worst case degraded grid conditions.
The calculation listed acceptance criteria that were to be met based on the tasks outlined
'
above and applying them to the limiting conditions of system alignment. Among these acceptance criteria were:
adequate voltage was available at the terminals of all safety related equipment,
separation from offsite power does not occur during the conditions of minimum a
expected offsite system grid voltage and worst case combinations of one or two auxiliary transformer operation, maximum BOP loading, and block loading of safeguards equipmer? Ato the safety-related 4160 v buses during accident conditions, and loading under worst case degraded offsite system grid voltage and maximum a
' BOP load conditions does not exceed established bus ratings.
GPUN immediately instituted temporary procedure changes to address the identifed system design deficiencies as they arose during the conduct of the calculation and con idered the electrical power distribution system operable. The procedure changes directed operator action to iestore bus voltage by securing unnecessary electrical loads
_.
_
_
.
_
.
,, -.
prior to exceeding the system design parameters assumed in the updated voltage regulation study. The outstanding items discussed below will address permanent corrective actions for long term design enhancements to the onsite electrical power distribution system.
GPUN identified through this calculation that under certain limiting conditions these acceptance criteria could not be met without operator action to reduce BOP electrical loading to prevent separation from the offsite grid. GPUN entered the following items on CAP No. T1998-1143 to track resolution of these and other outstanding issues identified as a result of the updated voltage regulation study:
permanently implement procedure changes directing operators to reduce BOP
loading by tripping a circulating water pump when offsite voltage reaches 232.4Kv, revise the system design description for TMI (SDD-T1-000) to reflect the BOP a
loading assumed in the opdated voltage regulation study, conduct an impact evaluation on the 480 v motor control center and 120 v control a
circuit voltage drop calculation (C-1101-733-E510-021) using the revised inputs from the updated voltage regulation study, evaluate the need for procedural or design enhancements for those safety-a related loads where the required terminal voltages were not available under all operating conditions, revise motor operated valve voltage drop calculations using the revised inputs a
from the updated voltage regulation study, and prepare a safety evaluation and technical specification change request (TSCR) to a
revise the as-left tolerances for the safety related 4160 v bus degraded voltage relay setpoints assumed in the updated voltage regulation study. Administrative controls were put into place to control the relay setpoints within the more restrictive tolerances until the TSCR can be processed.
c.
Conclusion The methodology and detail of the updated voltage regulation study demonstrated conformance of the electrical distribution system with the UFSAR design basis.
However, outstanding issues identified as a result of the calculation required additional review of corrective actions documented in the CAP.
E1.2 (Closed) Licensee Event Report 98-010-00: Potential Violation of Depian Criteria Durina Sinale Auxiliary Transformer Operation (92700)
The inspectors reviewed LER 98-010-00, dated October 9,1998, to verify GPUN completed a comprehensive evaluation and took adequate corrective actions in
.,,
l response to finding deficiencies in the calculation supporting the design basis of the onsite electrical distribution system. The updated voltage regulation study discussed in section E1.1 of this inspection report was completed by GPUN in part as a corrective action in response to this LER. The inspector identified no issues not previously addressed by the updated voltage regulation study. The inspectors note no violations of
NRC requirements and closed the LER.
E4 Engineering Staff Knowledge and Performance E4.1 (Closed) Unresolved item 98-03-02: Emeroency Feedwater Flow Reauirements Safety J
Evaluation Adaaum: (Open) Apoarent Violation 98-09-03: Introduction of an Emeroency Feedwater Unreviewed Safety Question (92903)
a.
Inspection Scope The inspectors first discussed this emergency feedwater (EFW) issue in NRC Inspection Report 98-03, Section E2.1, and left it as an unresolved item. The TS bases specifically listed the full capacity turbine driven (TD),920 gpm, and the two half capacity,480 gpm, motor driven (MD) EFW pumps for a loss of feedwater accident (LOFW). GPUN lowered the EFW flowrate requirements in October 1997 to 240 gpm to each OTSG,480 gpm total from any two of the three EFW pumps, based en a new design basis calculation.
The initial GPUN safety evaluation was incomplete in that it did not recognize the lowering of the EFW flow requirements as a change to the UFSAR or the TS bases.
The inspectors reviewed the GPUN safety evaluation (SE) completed in September 1998 and a subsequent TS bases change completed in November 1998.
b.
Qhservations and Findinas As stated previously in IR e8-03 the inspectors did not have any operabiiity concerns regarding the EFW system, assuming that GPUN used acceptable methods to calculate the design accident analysis flowrate requirements. However, the NRC staff did have concerns regarding the suitability of the safety analysis used to justify the changes made to the EFW system in October 1997.
GPUN prepared a SE, dated September 1,1998, for the change in EFW flowrates and prepared an update to the UFSAR.
The UFSAR update appeared to properly address the new calculation
requirements for EFW flow of 240 gpm to each OTSG.
GPUN did not identify this change as an unreviewed safety question (USQ)
=
since, as stated in SE Section 4.2, the initial UFSAR acceptance criteria (reactor power not exceeding 112 percent, RCS pressure not exceeding 2750 psig, and the pressurizer not going solid) were still achievable based on the new calculation.
_
.-.
- _
_.
.
-
-- -
. _ _.- -
.
..
.
-..
-
,
..
SE Section 4.4, stated that the TS bases did not need to be changed since it only
-
required 460 gpm. This was again in error, sinu., the 460 gpm, as stated in the TS bases, was for the half capacity rnotor-driven pump. The inspectors identified that the SE was wrong and that the TS bases did need to be changed.
GPUN engineering prepared a TS bases change, dated November 11,1998, and an
.
associated SE to correct the error made in the September 1998 SE. This evaluation removed the 920 and 460 gpm capacities and stated that this was not a USQ since the change was necessary to clarify the TS bases confusion, noting that the design basis flowrate was as described in the UFSAR as updated by the September 1998 SE.
Based on review of the November SE and the flowrate acceptance test criteria
-
calculation, the inspectors found that GPUN once again did not complete a thorough review. GPUN did not identify that their calculation indicated that the
)
turbine drive pump was not able to supply the full accident analysis capacity, by
]
itself (i.e., it cannot supply the required 240 gpm to each OTSG.) Therefore, the l
TS bases and UFSAR Section 10.2 needed to be changed to state that the TD pump is not full capacity and that any two of the pumps (the TD plus a MD or both MD) are required to supply the 480 gpm totar.
The inspector had previously reviewed the calculation that developed the minimum flow acceptance limits for the rdJeling cycle iriection test to the depressurized OTSG. This calculation was based on the system requirements of the new (October 1997) LOFW analysis and system / pump characteristics. This calculation also developed the flowrates that each pump should be able to achieve to the pressurized OTSGs, assuming that the acceptance criteria for depressurized OTSG flow were met.
Pump Combination Net flow achievable to both OTSGs -
from GPUN calculation with OTSGs at 1050 psig (480 gpm required to be full capacity)
TD alone 391 gpm MDA+MDB 517 gpm MD A + TD 485 gpm MD B + TD 494 gpm TD + MD A + MD B 552 gpm
The TD EFW pump would meet the criteria of 350 gpm to both OTSGs for a
LOFW with a loss of offsite power (LOOP) or a station blackout (SBO). This
'
number is lower than the 480 gpm, since approximately 130 gpm is needed to remove the heat added by the RCPs, which is not a factor following a LOOP, since the RCPs trip.
-
.
-.
. - -. -
_ _ -. -
...-_.
~--
.-------
t
,,.
i i
j
i
'
in the review of these changes, the NRC staff considered that GPUN may have introduced an unreviewed safety question when they completed a new calculation, using j
a new methodology, which resulted ir lowered EFW flow requirements to the OTSGs.
Further, GPUN did not identify that their calculation determined that the TD emergency feedwater pump was not capable of supplying 100 percent of the accident analysis required flowrate as described in the bases for TS 3.4.1.1 and in the UFSAR, Section
10.2. The NRC staff is concemed that GPUN failed to initially complete a safety evaluation before implementing the analysis and that these changes represented a substantial difference between the original analysis discussed in the UFSAR and as such reduced the margin of safety for the EFW system as defined in the bases for TS 3.4.1.1.
i The inspectors considered this an apparent violation pending further NRC review and
closed URI 98-03-02. (eel 98-09 03)
i
,
c.
Conclusion i
The EFW system remained operable, assuming that in October 1997 GPUN used appropriate methods, although different from the original method, to calculate the
required LOFW accident analysis flowrate. This analysis resulted in GPUN lowering the
'
required flow to both OTSGs following a LOFW. However, GPUN failed to complete a safety evaluation before implementing this analysis. More recently, between September
'
1998 and January 1999, GPUN also failed to recognize that calculations showed that the installed pumps had capacities lower than previously assumed, specifically the TD pump could not supply the full accident analysis flowrate, by itself. The NRC staff is concemed
,
that these changes may have introduced an unreviewed safety question, which required NRC staff review prior to implementation, in accordance with 10 CFR 50.59.
GPUN engineering did not perform adequate reviews of details in preparation of the safety evaluations completed after the EFW change had been implemented.
Specifically, engineering did not identify the needed change to the TS bases and UFSAR Sechon 10.2 and subsequently, completed incorrect changes to the UFSAR and TS bases regarding the capability of the turbine driven pump'(i.e., it was not tull l
capacity).
,
l IV. Plant Suonort
!
R1 Radiological Protection and Chemistry (RP&C) Controls (71750)
Plant housekeeping remained good. No negative radiological conditions were identified during routine plant tours.
V Manaaement Meetinas X1 Exit Meeting Summary
{
Following the conclusion of the inspection period the inspectors conducted an exit meeting with
!
GPUN managers on February 12,1999.
i l
GPUN staff comments conceming the issues in this report were documented in the applicable report sections. No proprietary information was included.
-
l
.
.
-
.-
,
,
.
, _ _,
-
,
. _.. _. _ _ _. _ _.. _. _ _ _ _... _. _
e -,, -.
l
INSPECTION PROCEDURES USED IP37551 Onsite Engineering IP61726-Surveillance Observations IP62707 Maintenance Observation IP71707-Plant Operabons IP71750 Plant Support Activities l
lP92700 Onsite LER review i
IP92902 Maintenance Follow-Up IP92903 Engineering Follow-Up.
ITEMS OPENED, CLOSED AND DISCUSSED i
Opened.
98-09-01 eel Emergency Boration Review - f allure to Follow Soluble Boron Control
. Procedure 98-09-02 URI Review Root Cause Analysis and Corrective Actions for Repeated Decay Heat River Strainer Clogging
- 98-09-03 eel Introduction of an Emergency Feedwater Unreviewed Safety Question Closed'
98-02-01 URI Fuel Handling Building Engineered Safety Feature Building Ventilation System Testing j
'
98-03-02 URI Emergency Feedwater Flow Requirements Safety Evaluation Adequacy; I
98-010-00 LER Potential Violation of Design Criteria During Single Auxiliary Transformer i
Operation Discussed-None
,
o,m
/
4.
.-,p._
s,.J3 a.
-M.-4 4s44-
-
- -+,=
=+
-
-.
-4 a
-
Stuaea4-
.--4-d AJ
..
i
l
!
LIST OF ACRONYMS USED j
AC Alternating Current
,
BAMT Boric Acid Mix Tank BWST Borated WaterStorage Tank
>
BOP Balance of Plant i
CA Chemical Addition and Sampling
'
CAP Corrective Action Process CDF Core Damage Frequency
CFR-Code of Federal Regulations
,
COLR Core Operating Limits Report i
DC Decay Heat Closed Cooling I
DR Decay Heat River (ECCS)
i eel Escalated Enforcement item j
EOP Emergency Operating Procedure
!
FHB Fuel Handling Building FHBESF Fuel Handling Building Engineered Safety Feature GPM Gallons Per Minute GPUN GPU Nuclear, Inc.-
I&C instrumentation and Controls ICS Integrated Control System IFl Inspection Follow-Up item IR inspection Report IST Inservice Test Kv Kilovolt LCO Limiting Condition of Operation LER Licensee Event Report LOFW Loss of Feedwater Accident.
LOOP Loss of Offsite Power MD Motor Driven I
MPFF Maintenance Preventable Functional Failure MWe Megawatt Electric MWt Megawatts Thermal MU Make-Up MUT Make-Up Tank i
NRC Nuclear Regulatory Commission j
OTSG Once Through Steam Generator i
PDR Public Document Room PM Preventive Maintenance PRG
_
Plant Review Group RCBAT Reclaimed Boric Acid Tank RCP Reactor Coolant Pump
<
RP&C Radiation Protection and Chemistry Controls i
SBO-Station Blackout SDD System Design Description
,
i-
.
.
.. _
. _ _
. =. -
._
..
-
..
.
. - -
..
-.
..
e...,.
.
SE Safety Evaluation SER Safety Evaluation Report
'
l SSCs Structure, Systems, or Components TD Turbine Driven TMI Three Mile Island-Unit 1
,
TS Technical Speedication TSCR Technical Specification Change Request UFSAR'
Updated Final Safety Analysis Report l
ULD Unit Load Demand t
URI Unresolved item USQ Unreviewed Safety Question l
v Volts j
r h
,
!
!
l I
i f
i
'
!
l I