IR 05000413/1986028

Insp Repts 50-413/86-28 & 50-414/86-31 on 860707-11.No Violations or Deviations Noted.Major Areas Inspected:Unit 2 Loss of Control Room Tests,Completed Unit 2 Startup Tests & Diesel Generator Troubleshooting Experience

ML20214M274
Person / Time
Site:	Catawba
Issue date:	08/21/1986
From:	Burnett P, Jape F, Larry Nicholson, Schnebli G, Matt Thomas NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:
Shared Package
ML20214M246	List: IR 05000413/1986028 ML20214M243
References
50-413-86-28, 50-414-86-31, NUDOCS 8609110034
Download: ML20214M274 (9)

Text

• '

. x an (fro ' UNITED STATES f 'o NUCLEAR REGULATORY COMMISSION

[' , REGION 11 g'

• -

j , 101 MARIETTA STREET. e ATLANTA, GEORGI A 30323 o

'+4 . . . . . .p

Report Nos.: 50-413/86-28(a'nd50-414/86-31 -s Licensee: Duke Power Company 422 South Church Street

'

Charlotte, NC 28242 Docket Nos.: 50-413 and 50-414 ' License Nos.: NPF-35 and NPF-52 Facility Name: Catawba 1 and 2 Inspection Conducted: July.7-11, 1986 Inspectors: e/m 'c MWs 8/2//86 P. T. Burnetf Date Signed

$ W $ // 6 L. E. Nicholson Date Signed df G. dcM'

G. A. Schnebli 2/zt/ P4 Date Sfgned h6 M. Thomas

& m les MMM - 2/ h Vate Signed Approved by: m /w F. Jape, Sbcti p Ch ef f /[$

Dhte 56gned Engineering Branch Division of Reactor Safety SUMMARY Scope: This routine, unannounced inspection addressed the areas of witnessing the Unit 2 loss of control room test, review of completed Unit 2 startup tests review of Unit 2 steam generator level control problems, and review of diesel generator trouble-shooting experienc Results: No violations or deviations were ideni;ifie PDR ADOCK 05000413 G PDR

.

REPORT DETAILS Persons Contacted Licensee Employees

• J. W. Hampton, Station Manager H. B. Barron, Superintendent of Operations

• J. W. Cox, Superintendent of Technical Services R. Blessing, Associate Engineer S. Brown, Reactor Engineer T. Crawford, Integrated Schedules S. Cooper, Procedure Supervisor C. Gregory, Test Coordinator

• C. L. Hartzell, Compliance Engineer

• D. Kimball, Operations Engineer

• J. H. Knuti, Operations Engineer P. G. LeRoy, Licensing Engineer

• C. Muse, Operations Engineer

• R. Scarborough, Test Engineer

• R. O. Sharpe, Nuclear Engineer

• F. P. Schiffley, Licensing Engineer

• D. Tower, Operating Engineer R. White, Catawba Safety Review Group (CSRG) Group Supervisor Other licensee employees contacted included construction craftsmen, engineers, technicians, operators, mechanics, security force members, and office personne Other Organizations R. Wolfgang, Westinghouse, Electrical Engineer E. Zimmerman, Westinghouse, Electrical Engineer NRC Resident Inspectors

• P. H. Skinner, Senior Resident Inspector

• P. K. VanDoorn, Senior Resident Inspector

• M. Lesser, Resident Inspector

'

NRR Personnel

. *K. N. Jabbour, NRC/NRR, Project Manager i * Attended exit interview

-. -- -

.

2. Exit Interview The inspection scope and findings were summarized on July 11, 1986, with those persons indicated in paragraph 1 above. The inspector described the areas inspected and discussed in detail the inspection finding No dissenting comments were received from the license The licensee did not identify as proprietary any of the materials provided to or reviewed by the inspectors during this inspectio . Licensee Action on Previous Enforcement Matters This subject was not addressed in the inspectio . Unresolved Items Unresolved items were not identified during the inspectio . Review of Completed Startup Tests (72600)

The following startup test procedures (TP) and periodic test (PT) used to accomplish startup tests were reviewed: TP/2/A/2100/01, Controlling Procedure for Power Escalation, was up-to-date for all completed activities when reviewed on July 10, 198 All zero percent and 20 percent power-plateau prerequisites were signed off as well as the 30 percent trip point change. Test performance was signed off through step 12.3.3.1 of the 30 percent power-plateau test The last procedure change entered was change 10, which deleted performance of the doppler-power coefficient test at the 50 and 90 percent power plateau Pursuant to License NPF-52, Condition 2.C(3),

this chanae in the test arooram as described in FSAR Chapter 14. must be reported to NRC/NRR w'ithin 30 days (July 26,1986) of' the ch'ang This requirement will be addressed during a later inspection, PT/2/A/4150/05, Core Power Distribution, was performed on June 26, 1986, to obtain full core map (FCM) 2/01/003 at 28 percent power. Both the hot spot and hot channel factors were within the limits of technical specifications. Some planar power distribution limits were exceeded for rated-thermal power conditions, but not for the prevailing condi-tion No violations or deviations were identifie . Loss of Control Room Test (72302, 72583)

The first attempt to perform this test on Unit 2 on June 27, 1986, led to an unplanned depressurization of the reactor coolant system (RCS). That event is discussed at length in inspection report 50-414/86-2 *

.

The revised test procedure was reviewed prior to reinitiating the tes TP/2/A/2650/03- (Change 6), Loss of Control Room Functional Test, is an

~a daptation of abnormal procedure (AP) 2/A/5500/17, Loss of Control Room, to account for special test conditions, such as leaving the reactor coolant pumps running to simulate decay heat and requirements, tripping the reactor from outside the control room, and performing all actions through stabilizing the plant in hot shutdown with the minimum shift cre Enclosure 13.6, Additional Actions to be Walked Through During the Performance of the Test, was added in change The enclosure simulated the remote manual initiation of all safety injection systems except upper head injectio The enclosure was responsive to inspector followup item (IFI)

414/86-27-0 The procedure also included test termination criteria to assure control was returned to the control room if necessar The test was performed the morning of July 11, 1986. All inspectors who witnessed the test attended the pre-test briefing. Upon initiation of the test two inspectors remained in the control room to observe activities there. One inspector accompanied the operator responsible for tripping the reactor at the breaker panels and from there to the auxiliary feedwater pump turbine control panel (AFWPTCP).

Two inspectors went directly to the two auxiliary shutdown panels (ASP).

Once the panels were manned by the minimum shift crew, a non-licensed operator from the crew was dispatched to perform enclosure 1 One of the senior resident inspectors accompanied the operator and confirmed the feasibility of remote manual initiation of safety injection had been demonstrate <

Close observation of the operators at the remote panels confirmed that the plant could be held stable and cooled down 50*F from those locations without difficult No deficiencies were observed. The completed test procedure and licensee's review for possible deficiencies will be inspected during a later inspectio No violations or deviations were identifie . Followup on Steam Generator Water Level Control Problems (92700)

A review was conducted into the recurring level control problems experienced with the Westinghouse Model D-5 steam generatoM at Catawba Unit 2. The difficulty in maintaining program level at low power levels became evident from the large number of Licensee Event Reports (LER) submitted involving steam generator level induced transient The inspectors reviewed documentation associated with 28 transients in which: (1) a main feedwater isolation occurred as a result of reaching the Hi-Hi steam generator level setpoint, and/or (2) a reactor trip occurred on steam generator Lo-Lo level. A further breakdown of the data revealed ten

.

.

transients in which the initiating event' appeared to be an anomaly within the level control system, with the remaining 18 transients initiated by a known cause (i.e., valve failure, malfunctioning controller card,' etc.)_ that resulted in an uncontrollable steam generator leve The inspection conducted by Region II personnel concerning this issue involveddiscussionswithresponsiblelicenseeandcontractpersonnelinthe/,h(

engineering, operations, and instrumentation departments. The licensee considers the major contributor to the steam generator water level control

[

problem in Unit 2 to be the design differences between the D-5 model steam generators installed in Unit 2 and the older D-3_ model steam generators installed in Unit 1. The. difference is in the location of the narrow range level taps and is believed by the licensee to be the predominant cause due to the following (See Attachment I for a comparison between D-5 and D-3 steam generators):

Total narrow range indication for the D-5 instrumentation is approximately one-half that of the D-3, 128" versus 234". This causes a much tighter band of control for both the operator and safety-related trip functions. The range for 30% power operation between the Hi-Hi level trip (which causes an Engineered Safety Features (ESF) actuation to isolate main feedwater and trip the turbine) and Lo-Lo level . trip (which causes a reactor trip) is approximately 78".for the D-5 versus 152" for the D-3 mode The level control system for the D-3 utilizes a variable program level between 38%-66% of the narrow range level indication (89"-154") from 0%

to'100% power. The D-5 system maintains a constant 50% program level (64") for all power levels making the compensation for shrink and swell during power level changes more difficul *

Both upper and lower narrow range level taps for the D-5 are located in the riser section of the steam generator which contains the primary moisture separators versus the D-3 which has the lower tap located much-lower in the steam generator, just above the top of the tubes. The problem with the tap locations in the upper section of the steam generator is that this portion is an extremely volatile steam-water mixture and thus any variation in temperature, pressure, or flow, could have an effect on the indicated level. In addition, the auxiliary feed water nozzle enters the steam generator in this same area. Feedflow to the steam generator enters through this nozzle up to about 15% power and then feedflow is shifted to the main feedwater nozzle with a small portion of feedflow being directed through the auxiliary feedwater nozzle (13% of total feedflow at 100% power). The variable auxiliary feedwater flow into this area could also effect the indicated leve The licensee has been actively pursuing the level control problem in an effort to find and correct the root cause. The licensee considers the level control system to be in the preoperational test phase and that they are in the process of the final system grooming. The following actions have been cecomplished to correct the problem:

.

• Removed the feed forward controller which obtained its input signal from nuclear power indication to anticipate a change in required feedwater flow based on a change in reactor power. This signal was provided for the feed regulating bypass valve but proved to be too sensitive. At present, the bypass valve receives its signal from steam generator leve The licensee is in the process of fine tuning the level control system by varying the gain on the various controllers in the system. This will be an on-going evolution until the optimum setpoints are achieve The Controlling Procedure for unit startup, OP/2/A/6100/01, was modified with Westinghouse input to shift from the auxiliary feedwater nozzle first, then shift from the feedwater bypass valve to the main feedwater regulating valve. Prior to the procedure change the valve shift was accomplished before the nozzle shift. This change was made to reduce flow and level oscillations and has worked successfully on three attempts. The inspectors were present in the control room to witness the nozzle shift during the last startup subsequent to the Unit 2 trip on July 8,1986. The evolution required the undivided attention of one control room operator to properly balance the system prior to the shif Discussions with operations personnel indicated the same evolution could be conducted on Unit 1 in a matter of minutes, whereas, on Unit 2 it would take several hour Startup procedures have also been modified to ensure all feedwater heaters are in service prior to nozzle or valve shift This causes the feedwater to be hotter and consequently, nozzle and valve shifts will not have as much of an effect as that of colder feedwate Additionally, the procedures were revised to increase main feed pump discharge pressure to aid in system stability prior to the nozzle trans fer. The increased feed pressure also allows a higher reactor power level while still feeding the steam generators on the bypass valve which in turn causes feedwater temperature to increas *

While troubleshooting the level control system on July 9,1986, four out of eight steam flow detectors were found to be connected incorrectly on the piping side. These detectors are D/P cells and the HP and LP piping were connected to the wrong ports of the cell. The detectors affected were two on B steam line and two on C steam lin Discussions with licensee and Westinghouse instrumentation personnel indicated that this would not have caused the level control problems experienced to date. They stated that these detectors provide the steam flow input to the water level control system which is then compared to the feed flow to generate a steam flow-feed flow mismatch signal. At low power levels these signals provide essentially zero input to the system due to the low flow rates. The dominant level control signal at low power is provived by indicated steam generator leve *

.

In addition to the ongoing troubleshooting and fine tuning of the control system, the licensee has been conducting discussions with Westinghouse on the possibility of moving the lower level tap further down in the steam generator. The proposed new location for the tap would be approximately the same as that of the D-3 model. Westinghouse stated the design of the D-5 model includes a higher recirculation ratio due to more moisture being returned to the steam generator from the separators. The increased recirculation ratio causes additional flow in the area of the proposed tap location. This turbulent flow may cause an instrument level error of approximately 22". Westinghouse stated that the relocation of the lower tap would require a reanalysis of the steam generato Westinghouse also considers that troubleshooting and tuning of the system will correct the problem Region II inspectors will continue to monitor the efforts to correct this problem as the startup phase of the test program progresse Within the areas examined, no violations or deviations were identifie . Unit 2 Diesel Generator Surveillance Testing During the inspection, licensee personnel stated that each emergency diesel generator (EDG) had experienced discrepancies during surveillance testing on July 8, 198 During performance of periodic test PT/2/A/4350/028 for EDG 28 four trip annunciators alarmed. The annunciators which alarmed were high jacket water temperature; 10-10 lube oil pressure; high bearing temperature; and high lube oil outlet temperature. Nearly two minutes passed between the time the first alarm was received and when the EDG tripped. The licensee stated that operations personnel in the EDG room during the test monitored the various parameters and they all appeared to be within their normal operating rang While investigating the problem under work request 33489 OPS, the licensee found the lube oil pressure transmitters corroded, so the transmitters were replace The EDG was tested and no problems were encountered. At the conclusion of this inspection, the licensee had not determined the cause of the delayed response of the EDG to the alarms and the reason for the corrosion in the lube oil pressure transmitter The licensee was still evaluating these questions. In addition, the licensee was evaluating the test to determined whether it should be considered a successful tes During EDG 2A testing, the diesel started but did not reach the required speed and frequency within the time specified in the Catawba Technical Specifications. The licensee investigated the problem under work request 33516 OPS and found that the left bank fuel rack assembly was binding slightly. The left bank fuel rack assembly had experienced binding problems prior to this occurrenc The previous binding problems are discussed in the licensee's letter to NRC dated June 17, 198 The fuel rack assembly bracket was ground and shims were piaced under it in order to achieve proper alignment. Licensee personnel stated that the fuel rack assembly will be monitored in order to detect any signs if binding starts to reoccu *

.

Actions taken by the licensee to resolve the concerns for both EDGs will be reviewed during a later inspectio No violations or deviations were identified in the areas inspecte . Followup of Regional Request (92705)

On July 8, 1986, Unit 2 tripped following an RCS average temperature increase, which was a consequence of an unplanned dilution. At the request of _ Region II, an inspector analyzed the event using plant data, information from the unit curve book (OP/2/A/6700/01), and formulas and curves from the vendor supplied plant operations package (P0P). Considering isothermal and power defect effects, the net reactivity change was about 50 pcm, which was indicative of a decrease in boron concentration of about 4 ppmb from the pre-event concentration of 824 ppm This information was reported to the Region by telephone on July 9, 1986, 1 InspectorFollowupItem(92701)

(Closed) Inspector Followup Item 414/86-27-03: Demonstrate feasibility of manually initiating safety injection from outside the control room. This demonstration was performed successfully on July 11, 1986. See paragraph 6 for detail _ ._..________,a_ -

,.A a _ ,.m . - - _ . - . - _ . . _ _ . _

.

-'

[

.

]~~ ? #[ l> $

.

_ . . , _

L

._

i I y,

o

,

= S

.

mLi y k

h f ,,

N

-

Alg

$4 hl M; k,h  :

2 N .

s

'

-

H -

_

i b 'i

- t c..-. : ' .=.1

-

,-- ,

\ .

_

-

Q

--

'

eks '

a 4 -

h a 9 4 ..

N <

,

e ,, h- .'

j 34, ld;'gi .

'o I

.

s

.

I

!

_,