ML20247L692

Electrical Distribution Sys Design Assessment

ML20247L692
Person / Time
Site:	Palo Verde
Issue date:	05/08/1989
From:	ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR
To:
Shared Package
ML20247L671	List: ML20247L666 ML20247L685 ML20247L692 ML20247L727
References
TAC-73246, NUDOCS 8906020236
Download: ML20247L692 (47)

Text

_ _ _ _ - _ . -

.4 " '.

7 ELECTRICAL DISTRIBUTION SYSTEM DESIGN ASSESSMENT May 8,1989 i-(

Preimred by:

g /g /p g, Reviewed be: M' '%, tl %L.'li%[tQ . { , Sk/gf

{}{.,, / * .,  :) ' '

/ / * , I dygr.cred r by: c4- ! . ( . .

{/ 1 - I , / ; ,l.1 'it i

Cg scurrred by:

\./

59 l

8906020234 890516 PDR ADOCh 05000313 P PNU L

I

?

e.

g.

C ELECTRICAL DISTRIBUTION SYSTEM DESIGN ASSESSMENT TABLE OF CONTENTS SECTION PAGE TA B L E O F C O NTE NTS ........................ ...... .... ...... .. .. .................. .... 2 EX E C UTI V E S U M M A R Y ................................. ..... .............. .... ................ 3 A N A LYSIS A N D EV A L U A TI O N . ........ .......... ........... .................... .............. 6 P VN GS R ea ct or Trip IIis tory ......... .... .............. .. .............. ...... .. ........ 7 -

Electrical Distribution Aligninent & Design Alternatives . .......... ..........13 Subsynchronous Oscillation (SSO) Relay Scheine......................................17 Evaluation of tbe Reliability of of her Relays at PVNGS............................ 22 Lo n g-Te rin E v al u a t i o n ................ .............. .... .. .................... ........ ...... 2 6 Electrical Distribu tion Sysicin Maintenance............... .............. .. ....... . 30 CONCLUSION AN D ACTION PLA NS............ .. .... .......................... .... .... 35 ATTA C I I M E N T .. ........ .......... ...... .. .. ........ ........................ .. .......... .. .... .. .. .... .. .. .. 4 0 Ta ble I -P VN G S R eacto r Trip E v e n (s .......................................................... 41 2

7 u a S 'f 6 e e

4 EXECUTIVE

SUMMARY

(

sur L

3

1

(

EXECUTIVE

SUMMARY

In r'esponse to NRC concerns reganting PVNGS electrical distribution system design ade-quacy and reliability, particularly as it relates to involvement in coinplicated reactor trips and/or natural circulation cooklown, Arizona Public Service Company (APS) has completed an evaluation of the historical performance of this system, as documented herein, with rec-commendations for enhancement modifications.

A review of PVNGS reactor trip history reveals that the Fast Bus Transfer (FBT) scheme has functioned as designed during plant trips initiated at power levels up to 100% reactor power. On four occasions natural circulation entry has occurred when FBT did not take place; however, three of these events involved circumstances under which FBT operation would not be reasonably expected. The fourth resulted in a ' design modification to enhance FBT perfonnance. Coastdown of the turbine-generator and Reactor Coolant Pumps (RCPs),

with subsequent lilockinr. of I'll'I'. has venulted in natural circulation ronhinwns an<l compli-cated accovesy actions.

An examination of electrical distribution system alignment and/or design altematives was

( perfortned to identify potential enhancements, and has resulted in the following three-part action plan:

1) Operate with one RCP bus powered from the Unit Auxiliary Transfonner and with the other RCP bus powered from the Startup Transformer.

2) Revise the FBT scheme to initiae bus transfer upon the receipt of signals from the Subsynchronous Oscillation (SSO) relays and the generator back-up distance relays.

3) Provide direct tripping of'the reactor in the event of any turbine-generator trip, i as well as evaluate the feasibility of disabling Steam Bypass Control System (SBCS) and Reactor Power Cutback System (RPCS) operation above some reactor power level to be established.

APS is still evaluating the SSO relay operation experienced at PVNGS Unit 3 on March 3, 1989; however, interim actions pending completion of this evaluation have been established.

They include adding these relays to the FitT scheme, and using digital fault recordets to as-sist in future diagnostic activities.

APS has determined that other electiical distribution system relays have not unreasonably contributed to historical reactor tiips or thei complexity.

L 4

"4 $

APS has evaluated the vendor-recommended maintenance for electrical distribution system equipment and has identified six items which are considered mandatory precursors to unit re-

{ stan. These items are in addition to other required maintenance, such as Equipment Qualifi-cation-related activities.

Furthennore, APS has initiated a long-tenn electric'ai distribution system evaluation project.

This multi-phase pmject incorporates,a historical, industry-wide nuclear plant and electrical system reliability evaluation; studies on specifie- current issues such as 'Statiini 111ackout*,

' Relay Coord'iation',

u and ' Lightning Protection'; and a detailed design bases review of the current electrical distribution system. All of the above are being conducted to enhance PVNGS design, reliability, availability, and safety.

i Od C

5

G 0 e 8

4 8 4 ANALYSIS AND EVALUATION

(

e 90

. L l

t f 6

p ANALYSIS AND EVALUATION PVNGS REACTOR TRIP IllSTORY Introduction APS reviewed unanticipated, automatic PVNGS reactor trips to detennine " lessons

. learned" from complicated trips; special emphasis was placed on . natural circulation cool-downs and events that involved operation of the Fast Bus Transfer (FBT) scheme. Reac-  !

tor trips occurring as a result of power ascensiod tests were included if the reactor' trip oc-

, curred upon receipt of an unanticipated protective signal. APS excluded planned and manual reactor trips from consideration (including planned natural circulation tests) to ensure that the " lessons learned" _ depend upon PVNGS equipment and personnel response to unfore-seen events.

Sununary Analysis

( Fifty-three -(53) PVNGS reactor trips met the criteria for the " lessons teamed" evaluation.

Table 1 (Attachment I) identifies these reactor trips in chronological onter, the Unit in-volved, associated Licensee Event Report (LER) number and title, operating mode, - reactor power at the time of the event, and whether the reactor trip involved natural circulation cooldown or FBT operation. The following bullets summarize the data contained in Table 1:

. Of the 53 reactor trips evaluated,35 involved Unit 1 (66 percent);

16 involved Unit 2 (30 percent); and 2 involved Unit 3 (4 Percent).

. Of the 53 reactor trips ev'aluated,47 (89 percent) occurred frcm op-erating Mode 1 (Power Operation); I (2 percent) occurred from Mode 2 (Startup); 3 (6 percent) occurred from Mode 3 (Hot Stand-by); and 2 (4 percent) occurred from Mode 5 (Cold Shutdown). The Mode 3 and Mode 5 events are reactor trip events because the Re-actor Trip Switchgear (RTSG) opened; reactor power was at zero percent in each case. These events were included not only for com-pleteness, but because one of the trips from Mode 3 also involved natural circulation couldown.

- Of the trips evaluated. 7 (13 peicent) involved entry into a natural circulation cochlown. As stated above, one of these occurred in l Mode 3; the remaining six occurred in Mode 1. APS has estimated 7

_ _ _ -- _ _- _ _ _--___ _ _ _ _i

1 a

the time spent in natural circulation based on oost-trip documenta-tion.. The natural circulation duration during these events has var-(. ied from approximately 43 minutes to approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> 25 minutes. The mean natural circulation duration predicted from these PVNGS trips is 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> 4 minutes, if the longest duration event is excluded from the calculation, or 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> 33 minutes if it is included.

An analysis of these natural circulation events is provided below.

- Of the 53 reactor trips, 30 (57 percent) occurred while the house loads for the affected unit were being fed from the Stanup Trans-fonners; therefore, no attempt was made to initiate an FBT. For 11 (21 percent) of the events, APS has not determined the electrical distribution system alignment at the time of the reactor trip. The unavailability of this data is not considered crucial to this evaluation because of the following:

a) Had failure of the FBT scheme occurred during these events, the failure would have been reported in the docu-mentation or the failure would have manifested itself as a natural circulation event.

b) If each of these iI events is assumed to have occurred while house loads were powered from the Startup Trans-fonners, the number of FBT attempts would be underes-C timated, and the reliability of FBT as estimated from these events would be conservative.

Of the 53 reactor trips, FBT was attempted but did not occur on 4 (8 percent) occasions; however, FBT was attempted and did occur during another 7 (13 percent) of the events. In an additional event, FBT becurred for one bus, but the feeder breaker for the other bus from the Unit Auxiliary Transformer (UAT) failed to trip. (The LER for this event does not provide a definitive root cause for the break-er failing to trip.) Of the 7 com'pletely successful FBTs,6 occurred during reactor trips from 99-100 percent reactor power and the sev-enth occurred during a trip from 50 percent reactor power. This data does not indicate low FBT reliability. Rather, it indicates that any l concems with FBT design should be limited in nature.

Natural Circulation Events The following provides a brief suunnary of each of the seven PVNGS natural circulation events that occuned during unanticipated, automatic reactor trips. The role of the FUT scheme during these events is also described.

L 8

l__ _ ___ _ _ ____.

d September 12,1985

( PVNGS Unit I was in Mode 1 at 53 percent reactor power when a main generator output breaker was opened to initiate a planned load rejection test. APS anticipated that the tur-bine would reduce speed and maintain house loads; however, the Electro Hydraulic Control (EHC) system did not maintain turbine control and main generator frequency decreased. The RCPs were being powered from the main generator along with other house loads, via the UAT. A reactor trip occurred when protective devices sensed the coastdown of the RCPs and projected an unacceptable RCS condition ( i.e., a low Departure from Nucleate Boiling Ratio (DNBR)). The reactor trip generated a turbine trip and, as generator speed continued to decrease, the RCP breakers opened as designed. Fast transfer of the RCPs to the offsite power source did not occur because of the low frequency on the RCP buses. The duration of natural circulation for this event was 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> 19 minutes. Restoration of forced RCS flow was delayed because the charging pumps which supply RCP seal injection had become gas-bound due to inaccurate Volume Control Tank (VCT) level indication and control. As a re-sult of the lessons teamed from this event, design changes were implemented in all 3 PVNGS Units to preclude a similar loss of VCT level and gas binding of the charging pumps.

October 3,1985 PVNGS Unit I was in Mode 1 at 52 percent reactor power when a reactor trip occuired due to a low DNBR condition projected by all 4 Core Protection Calculators (CPCs). At the time of this event the RCP buses were being powered from offsite via the NAN-S05 and NAN-( S06 buses, as required in preparation for a subsynchronous resonance test. An apparent malfunction of the Plant Multiplexer (PMUX) caused switchyard breakers to open and the re-sultant loss of power led to RCP coastdown and reactor trip. The duration of natural circula-tion for this event was 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> 34 minutes. To prevent recurrence, the switchyard breakers affected by the apparent PMUX malfunction were hardwired, bypassing the PMUX breaker control.

October 7,1985 PVNGS Unit I was in Mode 3, at zero percent reactor power, with the RCS at approximately 2250 psia and 565 degrees Fahrenheit, and with the part-length and shutdown Control Ele-ment Assemblies (CEAs) withdrawn in preparation for startup. Troubleshooting was being conducted on the PMUX to detennine the cause of the problem which led to the reactor trip on October 3,1985. Another apparent PMUX malfunction occurred, resulting in a loss of off-site power to the RCP buses and a reactor trip on the loss of forced RCS flow as sensed by steam generator differential pressure instnanentation. The duration of natural circulation for

~ this event was 44 minutes. To prevent secunence, the switchyard breakers affected by the apparent PMUX malfunction weie hardwiied bypassing the PMUX bienkei control.

Januaiy 9.1986 PVNGS Unit I was in Mode 1 at 100 peicent teactor power, with the Reactor Power Cutback

( System in " Auto-Actuate-Out-of-Service," when a turbine trip and subsequent reactor trip 9

were initiated as part of a scheduled power ascension program test. The turbine trip was ini-tiated by manual actuation of the unit differential genciator protection relay. The 525-kV

_( generator output bleakers opened as designed but, due to a sensed frequency mismatch be-tween the UAT and the offsite power source, a synchronization check relay blocked the antic-ipated FBT. Reactor trip occurred due to a CPC projected low DNBR condition, rather than on an anticipated high pressurizer pressure condition. The duration of natural circulation for this event is 43 minutes. Following this event and on an interim basis, the Unit was operat-ed with house loads aligned to the Startup Transformers. The design and operation of the synchronization check relay was reviewed by APS and an enhanced design was incorporated into the PVNGS FBT scheme.

July 12,1986 PVNGS Unit I was in Mode I at 100 percent reactor power, when a reactor trip ocemred up-on the Plant Protection System (PPS) sensing low RCS How through stetun generator #2.

Although this reactor trip was generated by 2-out-of-4 coincidence logic circuitry, subse-quent investigation revealed that an actual low RCS flow condition did not exist. Rather, the PPS serpoints were close to the safety analysis limits and did not provide sufficient operat-ing margin to pieclude this type of event. At that thne the undervoltage relays on the 13.8-kV NAN-S03 and NAN-SO4 buses weie sci at 95.65 percent of rated signal voltage and, im-mediately following the reactor trip, the undervohage relays sensed a grid perturbation. (The Transmission Control Center indicated that grid voltage can vary as much as 5 percent, plac-ing nominal grid voltage within the range of the trip setpoint.) RCP buses NAN-S01 and

( NAN-S02 were load shed from NAN-S03 and NAN-SO4, placing Unit 1 in a natural circula-tion cooldown. The duration of natural circulation for this event was I hour 23 minutes.

Corrective actions have included the establishment of new PPS and undervoltage relay set-points.

July 6,1988 PVNGS Unit I was in Mode I at 100 percent reactor power when phase "B" of the 13.8-kV NAN-S02 bus faulted to ground, immediately followed by grcund faults on the other 2 phas-es. The feeder breaker to the bus did not immediately trip because protection is afforded by a thne-overcurrent scheme. The time-overcurrerit protection was set to trip in 0.7 second (42 cycles) on a 3-phase fault; however, the UAT also experienced a fault and began to fail at 12 cycles. The UAT ruptured and caught fire. The RCPs were being powered from the UAT, and FBT could not be as.hieved because of frequency and voltage mismatches due to the ground fauhs. %e duration of natural circulation for this event was 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> 25 minutes. Re-covery of forced cliculation cooling for the RCS was dependent upon the actions necessary to

,, restore power to the RCPs safely, given the nature and extent of damage to the elect:ical dis- I I

tribution system.

March 3,1989 PVNGS Unit 3 was in Mode I at approximately 98 percent reactor power, when the main generator output breakers opened. A Reactor Power Cutback occurred as designed; howev-(

10 1

er, the control system for 4 of the 8 Steam Bypass Control System valves did not operate properly and the reactor tripped on steam generator #2 low pressure. Certain Engineered C Safety Features (ESF) actuations occurred (e.g., Safety Injection) so 2 RCPs were tripped in accordance with plant operating procedures. The RCPs were being powered from the Unit Auxiliary Transfonner at the time, and their power did not automatically fast transfer to the offsite power source because of the degrading frequency and voltage as main generator speed decreased. The two operating RCPs tripped on electrical protection. The duration of natural circulation for this event was 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> 42 minutes. Operation of subsynchronous pro-tective relaying resulted in the opening of the main generator breakers. The APS investiga-tion of the root cause for relay operation is ongoing (see Subsynchronous Oscillation (SSO)

Relay Scheme section for details).

Discussion Of the 7 natural circulation events,3 (43 percent) were initiated while the RCPs weie pow-ered from the Startup Transfonners; thetefore, FBT was not attempted. The remaining 4, (57 percent) natural circulation events involved an attempted FBT which did not succeed be-cause of m'ismatched frequency or voltage on the affected buses.

The iluce evems stiat vannett wtule powes wa3 suppliett to in uac lu,nl3 n oni stie :,i,m ni, Transformers include the October 3 and October 7,1985, and July 12, 1986, Unit I reactor trips. As indicated in the event summaries above, the Unit was aligned to offsite power on

( October 3,1985, in preparation for a subsynchronous resonance test; on October 7,1985, be-cause the Unit was in Mode 3 with the UAT out-of-service; and on July 12,1986, pending re-view of the FBT scheme following the Jaimary 9,1986, event. Since the Unit was poweied from offsite during these three events, it was vulnerable to natural circulation upon loss of off-site power.

The four natural circulation events which occurred while the RCPs were powered from the UAT include the September 12,1985, January 9,1986, July 6,1988, and March 3,1989, reac-tor trips. The July 6,1988, event involved more than one electrical ground fault and resulted in the UAT mpturing and catching fire. Clearly FBT would not be expected to occur during this event because of the extent of electrical distribution system problems. Additionally, the January 9,1986, event involved an attempted FBT which was blocked by a synchronization check relay. The design and operation of the synchronization check relay was reviewed by APS and an enhanced design was incorporated into the PVNGS FBT scheme.

The remaining two events (i.e., the September 12, 1985, and March 3,1989, events) are the

,, most significant in terms of identifying potential electrical distribution system enhance-ments. In both cases the main geneintm output breakers opened. disconnecting the main generator from the switchyard. Under such circumstances, the plant's control systems (e.g.,

Reactor Power Cutback System, Steam Bypass Control System, Tu bine Electro Hydraulic Control System, etc.) should reduce reactor and turbine power such that house loads are con-tinuously supplied f om the main geneiatoi via the UAT. The plant should arrive at a stable operating plateau with all systems in balance and with the RCPs maintaining forced RCS cir.

11

_ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ I

c culation.

( Although the plant control systems have operated properly on other occasions, even prevent-ing reactor trips, they did not achieve and maintain a stable condition during the September 12, 1985, and March 3,1989, events. On September 12, 1985, the turbine EHC system did not maintain turbine speed, causing main generator frequency to decay. The RCPs were still connected to the main generator at the time and the reactor tripped when the protection sys-tem sensed the RCPs slowing down. On March 3,1989, the control systems (principally the Steam Bypass Control System) could not maintain an adequate balance between primary and secondary systems, and the reactor tripped on low steam generator piessure. In both cases the main generator had coasted down as designed and had slowed sufficiently such that FBT would have been blocked.

The goal of this evaluation is to enhance the PVNGS ac electrical distribution system, over and above the current license basis, to prevent complicated trips. He identification of alter-natives and the selection of options is described in the Electrical Distribution Alignment &

Design Alternatives section of this repon.

)

• g L

12

• .* l ELECTRICAL DISTRIllUTION ALIGNMENT & DESIGN ALTERNATIVES

.(

Introduction -

APS examined the PVNGS electrical distribution system to identify conceivable, alternative distribution alignments and design changes (particularly protective relaying modifications).

The' purpose of this review was to minimize natural circulation cooldowns, by maintaining a -

reliable source of power to the RCP buses. The following ten alternatives were identified as - i a result of this activity:

Altemaiive 1 - Operate with both NAN-S01 and NAN-S02 aligned to the Unit Auxiliary Transionner.

Altemative 2 - Operate with both NAN-S01 and NAN-SO2 aligned to redundant Stastup Transfonners.

Altemative 3 - Operate with both NAN-Sul and NAN-S02 aligned to a single Startup Transfonner.

Altemative 4 - Operate with one RCP bus powered from a Stanup Transformer and the other RCP bus powered from the Unit Auxilia-ry Transfonner.

Altemative 5 - Modify the SSO relay to initiate FBT, and power the RCPs from the Unit Auxiliary Transfonner.

Altemative 6 . Disable the RPCS and SBCS functions, while power-ing the RCPs from the Unit Auxiliary Transformer or Startup Trans-fonner(s).

Altemative 7 - Combine Attemative 2 with the SSO relay modifica-tion of Altemative 5. -

Altemative 8 - Combine Altemative 3 with the SSO relay modifica-tion of Altemative 5.

Altemative 9 - Combine Alternative '4 with the SSO relay modifica-

. tion of Altemative 5.

Altemative 10 - Disable all non-diiect tuibine-generator trips.

L 13

Evaluation of Alternatives l Of the ten alternatives listed above, APS detennined that only three were viable options; the others either violated regulations, contained internal contradictions, or did not provide reli-ability improvement. The three viable Alternatives are Alternatives 1, 4, and 5. Alternative 5 was expanded to include two phases, "a" and "b".

Phake "a" will be comprised of a change to the initiation requirements of the turbine and reac-tor trip sequences. The tripping of the generator breakers will initiate a turbine trip which in-turn will initiate a reactor trip, when the plant is operating above an as-yet-to-be-detennined reactor power level. Phase "b" will be comprised of modifying the FBT circuitry to initiate FBT upon receipt of SSO iclay signals.

Avoidance of complicated trips was detennined to be at least as imponant as avoidance of reactor trips and/or natural circulation events. On this basis, Alternative 5 was detennined to be the preferred Alternative.

Alternative i exposes the plant to complicated trips due to its reliance on FBT, RPCS, and SBCS, on generator trip events. Altemative 4 requires additional operator monitoring of electrical distribution system interactions, as well as its reliance on FDT for the RCP hus aligned to the Unit Auxiliary Transfonner. This also leads to undue complexity, which is pter-erably avoided for a pennanent solution.

Innplementation Alternative 5 is to be implemented in two phases. Although it also relies on FBT, it minimiz-es natural circulation by initiating FBT at the first indication of a potential disturbance affect-ing the power supply to the RCPs. Figure 1 is provided to clarify the implementation schedule of Alternative 5 in its entirety.

96 L

14

. ,. i 2 . ..- ,

l: l

-s-

. (V l

r

, ]

. , ,I x FIGURE 1 Implementation of Alternative 5- 1

. i COMPENSATORY PHASEI PHASE 2 1

UNIT I TO BE IMPLEMENTED ' TO BE INSTALLED TO BE INSTALLED

- PRIOR TO RESTART BY THE FOURTH BY THE FIFTH RFFIIEl.ING OIITAGE REFilRI.ING 011TAGE UNIT 2 TO BEIMPLEMENTED TO BE INSTALLED TO BE INSTALLED j i

PRIOR TO RESTART BY THE THIRD BY THE FOURTH REFUELING OUTAGE REFUELING OUTAGE  ;

UNIT 3 TO BEIMPLEMENTED TO BE INSTALLED TO BE INSTALLED PRIOR TO RESTART BY THE SECOND BY THE THIRD REFUELING OUTAGE REFUELING OUTAGE i

I i

L iS 1

- _ _ - - _ - - - - - - - - - - - - - - - - - - - - i

-l* ..

>9 ..

m. ,. .

Until Altemative 5 can be implemented,' Alternative 4, to power two RCPs from the Unit Auxiliary Transfonner and two from a Stanup Transformer, will be employed. This alignment practically excludes natural circulation entry for initiating events that do not involve the Start-up Transfonner. It also prevents tripping two PVNGS Units upon. an initiating event involving one Startup Transfonner, as would happen if RCPs for all three Units' RCPs were powered from the preferred (offsite) power source. 'Ihe current design and license bases for -

PVNGS are not affected for restart.

Tlic first phase to be implemented is Alternative 5(b), that is, initiate FBT upon receipt of signals from . the SSO. relays and generato: back-up distance relays. This modification is currently being designed as a Design Change Package (DCP) with a target issue date of .

June 1, 1989. It will then be implemented either at the next Unit 2 refueling outage (estimated Fall 1989) or sooner, if possible, as Shon Notice Outage Work (SNOW). As noted in Figure 1, this change will be implemented in PVNGS Units I and 3 during their current refueling outages. Operation of the Units without this DCP does not present a challenge to any design bases, since Altemative 4 'vinually precludes natural circulation as well as other complicated trips.

• The second phase to be implemented is Alternative 5(a), the use of RPCS and SBCS up to some predetennined, but as yet undetennined (e.g., 50-60 percent reactor power) power level. The exact level will be determined upon completion of a detailed engineering study of the RPCS, SBCS, Feedwater Control System (FWCS) and the Plant Protection System ,

(PPS). Among the factors to be considered in the study are system interactions and human I responses to plant transients (e.g., a main feedwater pump trip). Once the exact level is detennined, the RPCS and SBCS would be kept out of service whenever the reactor is at or above this level. In order to achieve this the turbine tripping scheme would be' modified to operate on the tripping of the generator breakers and the PPS would be modified to initiate a direct reactor trip upon a turbine trip. Operation of the Units without this portion of Altemative 5 does not present a challenge to any design bases, since Alternatives 4 and 5(b) virtually preclude natural circulation and other complicated trips.

. i L

16 i

SUllSYNCIIRONOUS OSCILLATION (SSO) RELAY SCIIEME Design Ilasis Review APS included the SSO relays in the o iginal plant design to protect the turbine generators from the adverse affects of subsynchronous resonance (SSR). SSR has caused catastrophic failures of turbine-generators. The original design for the SSO relay tripping scheme was to sense a SSR event, isolate the turbine-generator, and permit continued operation of the reac-tor, turbine, and generator to supply the house loads. March 3,1989, in Unit 3, SSO relay op-eration initiated a chain of events that led to coastdown of the RCPs, resulting in a reactor trip and natural circulation cooldown. This event is discussed below with ahernatives to en-luuice the SSO ielay scheme.

All thtee PVNGS Units are equipped with two SSO relays. Operation of either or both of the relays will initiate the opening of the 525-kV breakers. Opening these breakers isolates the main generator from the 525-kV transmission system. The application and setting of SSO relays at PVNGS was a result of extensive modeling, testing, and analysis.

The SSO relay settings for each PVNGS Unit is different to prevent simultaneous tripping of all three linits. During the occmrence of one SSR event, the isolation of one tinit will detune the electrical system; thesefore, the other two Units would not be expected to uip. At PVNGS, the Unit 3 SSO relays are set to be the most sensitive, then Unit 2 and Unit 1, re-spectively. Unit I trips last, since its control room contains the switchyard mimic hus, the use of which may be necessaiy following an SSR trip event to prevent a joint Unit /switchyaid C transient control room response requirement.

PVNGS has experienced two Unit tiips involving operation of the SSO relays. The first trip occurred oa January 10, 1987, in Unit 1. The "l A" SSO relay had an internal design problem in its phase-lock-loop circuitry, that generated a false trip action which isolated the Unit I turbine-generator from the 525-kV transmission system. APS worked with the SSO relay manufacturer, Westinghouse, to resolve the phase-lock-loop circuitry problem..

The second trip occurred on March 3,1989, in Unit 3. Printouts from the Plant Monitoring System indicate that an SSO relay-initiated trip signal opened the main generator 525-kV breakers. Initial investigation and analysis of a simulation of the condition on the 525-kV transmission system, at the time of the event, indicated that the SSO relay should not have operated. Functional tests performed on the SSO relays at PVNGS showed no apparent fail, ure of either relay. Based on these findings, " bench" tests were performed on the SSO relay circuit boards. The results of these tests did not indicate component malfunction or failure.

Since test results of the Unit 3 SSO velay.s themselves have not identified the canse of the SSO relay / relays operation, APS investigated possible sources of enoneous signals to the SSO relays. The turbine-generatoi Powei System Stabilizer (PSS) was reviewed as a pos-sible sontce of erroneous input signals lo the SSO relay. A malfunction of the PSS could ge"-

erate a signal which could cause SSO ielay operation; however, test results indicate tha' PSS is not a likely source for the pioblem. Additionally, APS is testing the noisy genermor 17

current transfonner circuits as a potential source of an erroneous signal. APS is also moni-( toring the Unit 3 SSO relay performance in its plant environment with a synthetic sthnulus (i.e., a stimulus that simulates the input to the relays at the time the Unit 3 SSO relay opera-tion occurred) being applied during the refueling outage. Upon discovery of the cause of hn-proper SSO relay operation, APS will take action to reduce or climinate recurrence Based on experience and analysis, an SSO relay operation will occur in the future. The pos-sibility of going into complicated trip situations (such as natund circulation) for these events must be minimized. This possibility can be minimized by realigning the power supplies to the reactor coolant pumps as discessed in the Electrical Distribution Alignment Alternative sections. In addition,5 alternative actions have been identified as long term means to mini-mize the negative hupacts of a SSO relay operation, to enhance the capability to identify the cause of a SSO relay operation, and to huprove SSO relay security. These alternatives are described in the following sections. A summary of these attematives is contained on Figure 2.

O 09 L

18

N 3t 8 d i pt 9d i p R d i

n

/

n s /

1 n s 0 n U

1

/2 6

m1 e -

/2 1

mi e /9 2 1

/

1 F o F o F.

N 2 9R t i r t t r0 R i t

t r

t r R t 8d pota9d / p o a0 O i n

/

1

/3 r nis1 r e/3 r

mis r /

t 9 d r I

c p r/ e1 3 r

U T 6 c pr1 1

E L -  :

.F F F F F PE1 R 9R 0R R

- 8 t

i 9 i t

0 R pd /9 h MTi t

/h pd mr / h m3 r t O AnU 1 t 1 t 1

/ / / 4 4 e 3 4 .

6 1 c C D 1 s s y y l

a l a

g g e e s d k d k R R e e o p o e p v g l e i g

a eg O O i

gn b g g h S S k S S k ha t

a y kpan y c n n p c d r d r a a / /

r e i o a r e c h I o ar ph c i s i s

n s s e g h o g h o e n ng g n y y

~

t nh g yp y p n g l l a

rar me a l

a r ai A h s i

a m r e h s a s i n n n c e c de h e A A e n d oT pt n oT pt c d

_ i s

N i gt n mn i gt n mn nt g ne e e

e e

e O e s e e e s e e e e is S- S-2D I d me Tme d m T me em E E T el el el e

el d el N N Ev a C up up up u p e p O O Rl A s

s m s s m s s m s s

m um s

N N UeR I I I I I I I I I

s I A

G I O o o o o o FS t e

t e e t

e t

'e t

e S g r gb g e g g e b f

o nf a s no at r s ni a u t

o n

a n

a t t h n h s e h c h h c n c ar c ed c r egn c c

c

'n ni orc i

o nt i

t n c r o ni n i

t N g s gt g e at t

i a

g g o

g s

O i i a i s u s e is c i s

u s e a r e ne a er t e

l e

deb n I g o d uql d r h p d n d T

l t t

a e o nya u t si t t t t i t v P nat nf e er y y n p n E I ef a r e af l a e p e R mf e o e mlel m e nl mi r m C

p e r a - e g o re e t e po no it l l pO g l

p a we l

p ya l pO S mi y s t

E i t

a a i mSS id i ml oloh vl a y t

i ml e r i m

D dt e i l d r n d d d ni nr n o o n al b l d

nO n af d ad id a aiO e e ta d S e e e t dd t e de we ey t t aS e S e t e

ai r aihv aci a p l a s a i luS i v co i t i i c r i

E t t o i t

oi h ini s it v t i s V innfo I i inr I pr e I n r p wid I n er ni I d y I TR AE NB RI EB L 1 2 3 4 5 T

L N A

p g

l

l, l

j 1

Analysis The following subsections describe potential SSO relay enhancement alternative. APS de-termined that the following alternatives are viable. These subsections describe the logic be-hind the decision to implement or not.

)

Alternative 1 t Initiate and implement design changes to include initiation of fast bus transfer for SSO relay action.

This is not a restart item since the attemate alignment of the reactor coolant pumps (described in Identification of Electrical Distribution Aligrunent Altematives section of this report) addresses the complicated reactor trip / natural circulation concerns. This enhance-ment will be completed during the next refueling outages for the three units. I Alternative 2 Initiate and implement design changes to provide for SSO relay quantities to be recorded on digital fault recorders. This change will be implemented in Unit 2 prior to restart and in Units 1 and 3 during the current refueling outages via a temporary modification. The temporary modification will provide identical infonnation on the SSO relay as will the design change package. The scope of the design change package includes many other areas; therefore, re-quiring additional evaluation prior to completion.

Altemative 3 Initiate and implement design change to modify the existing voltage reference circuit so it will allow only that target initiated by the relay operation to be displayed. Th s change will be implemented in Unit 2 prior to restart, and in Units I and 3 during the current refueling outage on a temporary mod. This modification provides immediate indication of the section of the SSO relay that operated. The modification providing for digital fault recording capability for the SSO relays, attemative #2, will provide the same information; however, this informa-tion is not immediately available. The design change package completion will be given ap-propriate priority, scheduled and worked as indicated in Figure 2.

Alternative 4 '

i Initiate and implement a design change to revise the SSO relay tripping logic. This alterna-tive is not considered as a desirable option at this time. This p oposed modification will not necessarily reduce significantly the pmhability an event similar to the Unit 3 trip in March

' 3,1989.. The actual cause of the Unit 3 relay operation is, and may continue to be. unknown.

APS believes the cause could either be a iesult of an intennittent component failure in the relay or a noise problem in the system. For a noise problem in the system a logic modifica-tion would have no positive effect In addition, a logic modification could reduce the pioba- '

bility of tripping for an actual SSR event since the scheme would have to be more complex (i.e. more than one relay would have to ope ate to protect the units from the hazards of a SSR event).

20

Alternative 5

, Initiate and iniplement a design change to distble the SSO relay tripping scheme. This would make all three units vulnerable to an SSR event, which could result in catastrophic fail-ures of the turbine-generators. Detailed studies, shnulations, and tests have justified, based on the probability of a SSR event occurring, that SSO relay tripping schemes are re-(luired on each unit as second contingency protection. First contingency protection is not re-quired.

M L

21 2

( EVALIIATION OF THE RELIATHLITY OF OTHER RELAYS AT PVNGS Introduction Ai)S reviewed the list of PVNGS unanticipated reactor trips, especially those that re in an entry into Natural Circulation, to identify trips in which relay prob / cms were deter-mined to be the root cause of, or a contributor to, the event. The purpose of this evaluation was to confinn that the logic configuration of the ot/ser power system a/ays did not unreason-ably contribute to those trips or trip complexity.

Summarv Analysis of Relav Contribution to Reactor Trio A total of 53 PVNGS reactor trips met the criteria for this evaluation; of these,10 reactor trips or 9 events (one event involved the tripping of two units) were considesed to be relay failure /misoperation related. Two of the Unit trips resulted from SSO relay problems and are addressed in the Subsynchronous Oscillation (SSO) Relay Scheme section of this repon.

This leaves a total of seven events due, in pan, to other relay problems. The following sec-tions summarize those seven events:

C December 4,1985 Unit I was in mode 1 at 54 percent power when a reactor trip occurred due to Low Depar-ture from Nucleate Boiling Ratio (DNBR). All four channels of Core Protection Calculators actuated; the trip was attributed to a high penalty factor being inserted by the Control Ele-ment Assembly Controller due to the dropping of sub-group 12 (part length control element assemblies). All equipment functioned as designed and no safety system actuated. No Tech-nical Specification [ Tech. Spec.] limits were exceeded.

During the resultant turbine-generator trip, bus IENAN-S02 experienced a momentary loss of voltage.

This was subsequently attributed to a dip of the supply voltage, caused by the loss of Main Generatcr output from the system in combination with a less than optimal ta setting of the Startup Transformer (SUT) supplying IENAN-S06, SO4 and S02. No distur-bance was encountered on buses IENAN-S05, 503 and S01. APS reset the (SUT) tap set-tings to the same value as the other (SUT) feeding the unaffected

,, buses.

This problem was setting ielated anil did not involve relay failme. All equipment and protec-tive devices functioned as designed. Naimal Circulation entry was avoided.

22 l

I l

L_____---___--_-_ - - - _ - -

January 9,1986 Unit I was in mode 1 at 100 percent power when a turbine trip and subsequent reactor trip were initiated as pan of a scheduled power ascension test program test. The turbine trip was initiated by manual actuation of the unit generator differential protection relay. Fast bus transfer (FTB) from the Unit Auxiliary Transfonner (UAT) to offsite power was blocked by a frequency mismatch between onsite and offsite power. Upon the loss of RCP power, Natural Circulation was initiated.

The device which inhibited the transfer block was the Syne Check relay which performed as designed during a subsequent investigation; APS found no indication of malfunction of FBT devices. Sync Check reset time (1.5 see) may have been the prbnary contributor to this anomaly..

An engineering investigation detennined that the original Sync Check relay, an electro-me-chanical induction type, was not suitable for high-speed, bus transfer applications because of slow perfonnance characteristics.

Consequently, APS replaced the Sync Check relay with a high-spud, solid-state device.

May 31,1986 Unit 2 was increasing in power when the Turbine Generator (TG) set tripped at approximate-C ly 35 percent power. Approximately 12 seconds after the TG trip, the reactor tripped on high pressurizer pressure (all 4 PPS channels tiipped). RPS response times were within seguired values. )

Safety functions were maintained and the event was classified as an uncomplicated i seactor trip. The SBCS did not prevent the reactor trip, due to master controller not being in REMOTE AUTO, but it did support the subsequent cool-down. Natural circulation was not entered.

The event initiator was the TG trip on negative sequence current. The actual cause has not been detennined, but it is believed to be false operation of the phase current imbalance pro-tection relay. The re!ays perfonned as design' ed during post trip diagnostics; however, pre-trip alanns had been received during the week prior to the event and PR&C had been involved in troubleshooting.

July 12,1986 PVNGS unit I was in Mode I at 100. percent power when a reactm trip occuired on how RCS flow as indicated by Steam Generator #2 differential pressure. APS detennined that an actu-al RCS low flow condition did not exist; rather the PPS serpoints did not provide sufficient operating snargin to preclude this event. The undervoltage relays on the 13.8KV NAN-SO3 and NAN-SO4 buses weie set at 95.65 percent of rated system voltage and immediately fol-

{ lowing the reactor trip, the undervoltage selays sensed a dip in the grid voltage. RCP buses 23 1

m' e 1

NAN-S01 and NAN-S02 were shed on undervoltage placing unit 1 in a Natural Circulation cooldown.

( The Systems Operations Department indicated that the grid voltage can normally vary as much as five percent, which places the nominal grid voltage within the range of the UV trip setpoint. As a result, APS lowered the relay settings temporarily 2.5 percent to 93.2 per-cent of rated voltage and initiated an investigation of the system grid and plant distribution models to determine a pemianent relay setting.

August 6,1986 Unit 2 was operating at seven percent reactor power when water from a leaky valve resulted in a fault / failure of load center NAN-S02. Subsequently, the SUT, NAN-XO3, feeding both units 2 & 1 tripped on differential current resulting in the loss of two RCPs and consequent trip of each unit.. A current transfonner (CT) was later found to have failed, (the project's failure history for CT is consistent with nuclear industry history, as given in IEEE std. 500-1984). All other devices operated acceptably. Natural Circulation entry was avoided during the course of the event in both units.

September 2.1986 Unit i experienced a teactor trip on Steam Generator (SG) low flow (RPS channels B and

( D). Following the reactor trip, the TG tripped and a generator coastdown trip cccurred shed-ding 13.8KV bus NAN-S02. A fast bus transfer occurred maintaining power to the PsCPs.

The loss of NAN-S02 was attributed to an undervoltage relay (type 227-1) trip under volt-age transient conditions. Natural Circulation cooldown entry was avoided.

The problem has been corrected; relay setting of the undervoltage (UV) relay was changed to 90 percent of rated voltage froin the previous 95.6 / 93.2 percent values to inhibit under-voltage trips during grid disturbances. As a result of the actions taken here, reactor trips from this cause have not recurred.

April 16,1987 Unit 2 was operating at 100 peicent power when a reactor trip and turbine trip occuned due to interaction between the RPS system, CPCs and troubleshooting of a suspected ground

- fault in the B train 120V AC (class IE) system. Natural Circulation entry was avoided.

A relay problem was encounicied dming the recovery actions. This pioblem involved an overcurrent trip of loadcenter breaker 2ENGNLOSB2, when no faults (i.e.: overcunent) were present. APS found that the shunt trip (SST) device in the bleaker protection circuit was de-fective due to silicon controlled rectifier (SCR) leakage. Replacement of the bleaker /SST de-vice compkted action on this item.

24 u____._ _ _ _ _ . _ _ _ _ _ _ . _ - _ _ J

,o-

~ '

Conclusion and Action Plan 1 1

Two of the eight events (25 percent of relay-related trips or 4 percent.of all trips) considered to'have been affected by other relay problems resuhed in unanticipated entry into Natural Circulation.

These two events involved Fast Bus Transfer and Undervoltage (UV) Trip  !

problems. Both of these trips occurred in Unit 1. APS has taken corrective actions which provide a reasonable assurance that these problems have been resolved.

Of the other Reactor trips, where Natural Circulation was not entered one trip was complicat-ed by Undervoltage relay actuation under transient conditions (corrective activities conduct-ed as a result of this event also resolved the problems identified in the UV trip-Natural Circulation event addressed above). Another trip involved a possible maintenance induced j

scram during troubleshooting. Another trip occurred during postcore power ascension test-ing, and involved SUT tap settings. Three events, involving four reactor trips, were the re-sult of random component failures / spurious actuations. A Phase Current Imbalance relay in the Main Generator Protection System, a CT in SUT X03 and a SST overcurrent device in a nonIE Load Center Based on the historical data presented above, APS has determined that modifications to

( Power System relaying design are not warranted. The corrective actions taken to date have minhnized the potential for a recurrence of those problems deemed preventable. The relay-related problems, that resulted in entry into Natural Circulation have been essentially elimi-nated.

Random failures cannot be absolutely prevented; the u:;e of a multiple (redundant) relay logic schemes would add significantly to the cost and complexity of the relay systems with little (if any) gain in reliability. Given the minhnal consequences of the random relay failures to date, changes to the existing Generator, SUT, or Load Center Protection systems are not justified.

As discussed in the preceding section, the investigation regarding upgrades /bettennents to the SSO relay system is continuing.

L l 25

______D

c w.

LONG TERM EVALUATION .

_Ilistory

+ APS initiated this evaluation February 1989 to coordinate a number of PVNGS projects and ensure compatibility of recommended enhancements. The evaluations and results document-ed in this Electrical. Distribution Design Assessments report are one part of phase 1 of the -

long tenu evaluation.

- Overview Currently, APS is coordinating a number PVNGS studies and evaluations to enhance power system reliability. This effort is comprised of the fo!!owing four Phases:

Phase I - Evaluation of PVNGS performance on its own base and against the industty, including a thorough review of possible distri- '

bution aligmnent and protection configurations .

Phase 2 - Incorporation of the results of the Electrical Reliability Stud-les Phase 3 - Incorporation of the results of the Engineering Excellence Program IU System Reviews

- Phase 4 '- Incorporation of the results of the Engineering Excellence Progsam non-lE System Reviews Sununary PHASE I - PVNGS Unit Trio Studv APS, under Phase 1, is reviewing all PVNGS plant trips and evaluating Electrical Distribu-tion System irwolvement. This data will be epmpared with statistics from other US nuclear plants. The PVNGS unit trips will be evaluated to detennine severity of electrical system contributing factors. The contribution of the Electrical Distribution system will be grouped in-to one of the following four categories:

I) Electrical Root Cause - Root cause detennined to be entirely elec-trical

2) Other Root Cause - Non-electrical root cause resulting in a unit trip as a result of an electiical contiihuting cause

3) Electrical Problem-not a cause - Electrical problems ielated to plant trips

4) No Electrical involvement 26

l I l ., .

i l

l j

The PVNGS trip data will be comp:ued with industry trip data to determine if PVNGS is ex-l

{ -

periencing a higher number of trip incidences involving the Electrical Distribution system (

I than the industry. The following four comparisons will be made:

1) PVNGS Electrical Trips vs. PVNGS Tothl Unit Trips

2) PVNGS Trips vs. other Bechtel-GE Plants (licensed since 1980)

3) PVNGS Trips vs. all Region 5 Plants

4) PVNGS Trips vs. US Plants (licensed since 1980)

The PVNGS Electrical Distribution system will be evaluated to deteimine any unique sys-tem design characteristics. If unique characteristics are identified they will be reviewed for any impacts they may have had on seactor trips. The PVNGS design will be compared with other Bechtel plants using similar designs.

As noted above, the results of this Electrical Distribution System Design Assessment re-port is also included bi the long team evaluation phase 1.

PilASE 2 - PVNGS Electrical Reliability Studies As a result of recent events at PVNGS, the following additional studies have been initiated:

. Electrical Power Sysiems Reliability Sttuk As a result of failures ex-perienced in the non-clasr IE power systems in the Power Block C and the 525 kV Switchyard, APS initiated a reliability analysis.

APS is coordinating this effont between APS and SRP to perfonn this reliability analysis and tecommend any enhancements to im-prove the reliability of the power systems.

- Station Blackout Study in accordance with the requirements delin-eated in 10CFR 50.63 Station Blackout, APS evaluated PVNGS us-ing guidance from NUMARC 87-00 document, " Guidelines and Teuinical bases for NUMARC Initiatives Addressing Station Blackout at Light Water Reactors", except where Regulatory Guide 1.155 takes precedence.

- Relav Settinn/ Coordination Evaluation in response to IE Notice 88-45, " Problems in Protective Relay and Dreaker Coordination", APS is reviewing the relay setting and coordination efforts at PVNGS.

s

- Lichty_ing livaluation line to ilie .lanuary 1989 Linit 2 lisle inmsham-er failures, APS performed an evah.rtion to determine il lightning was the root cause or a conuibuting factor. APS completed a light-ning protection design evaluation July 28,1986. This evaluation was initiated in response to INPO Significant Event Repon (SER)

No.84-76.

(

27

r .

l

- Fast Bus Transfer Desien Review Following the March 3,1989, Unit 3 trip during which a Fast Bus Transfer did not occur, APS ini-tiated an evaluation of the Fast Bus Transfer design.

PHASE 3 - O Electrical Systems j The Engineering Excellence Program" (EEP) system review working groups for the Q class electrical systems will factor the findings of the Phase 2 report into the evaluation of the de-sign basis review and their overall review findings into the Phase 3 report. The Q class elec-trical systems are as follows:

PB - Class 1 E 4.16-kV Power System PE - Class 1 E Standby Generation System

- PG - Class f li 480-V Power Switchgear System Pil - Class IE 480-V Power MCC System PK - Class l E 125-v DC Power System ,

PN - Class !E Instrument AC Power Syt, tem Pil ASE 4 - Non-O Electrical Systems The Engineering Excellence Program system review working groups for the Non-Q electrical systems will factor the findings of the Phase 2 report into the evaluation of the design basis

( review and their overall review findings into the Phase 4 report. The Non-Q class electrical systems are as follows:

- M A - Main Generation System MB - Excitation and Voltage Regulation System NA - Non-Class 1E 13.8-kV Power System

- NB - Non-Class IE 4.16-kV Power System NG - Non-Class 1 E 480-V Power Switchgear NH - Non-Class 1E 480-V Power MCC System NK - Non-Class 1E 125-V DC Power System NN - Non-Class IE 120-V Instrument AC Power System NQ - Non-Class IE 120-V 'Uninterruptible AC Power System Completion APS will isst.e a repon at the completion of each Phase. The schedule completion dates for these reports are subject to change based on completion of the individoal activities included in that Phase .The current scheduled completion dates and Repon muputs me listed below.

06/01/89 Phase 1 - Lessons Leamed

- Preliminary Recommendations for Consideration 28

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ a

} ,

~

( 07/14/89 Phase 2 - Lessons Leamed

- Recommendations for implementation 03/12/91 Phase 3 - Refinement of Phase 2 Recommendations i

08/05/93 Phase 4 - Refinement of Phase 2 Recommendations Phases 3 and 4 are considered to be confirmatory and additional lessons-learned informa-tion. The results of phases 3 and 4 will provide in-depth review of the plant changes required or recommended during Phases 1 and 2.

Op L

29

_ _ _ _ - - _ _ - - )

1 ELECTRICAL DISTRIBUTION SYSTEM

! i MAINTENANCE Introduction Arizona Public Setvice Company (Ah>S) conducted. a thorongh investigation of the PVNGS Electrical Distribution System maintenance requirements. This investigation consisted of re-view of the vendor instruction and maintenance manuals to identify pertinent functional crite-ria and maintenance recommendations for system components. Functional criteria are the tasks required to be perfonned (whether performed recently or not) prior to the restart of a unit. APS detennined the functional criteria based on the operational history of specific criti-cal electrical system components or engineering recommendations (IEEE and other estab-

'lished standards). These functional criteria may exceed the normal preventive maintemmce requirements and scope.

APS compated the maintenance recommendations and functional criteria with the current 3 electrical preventive maintenance program. The implementation schedules of these tasks were reviewed to detennine the overall technical adequacy of the cunent PVNGS Electrical Distribution System Maintenance Program.

APS reviewed the PVNGS restart activities related to Electrical Distribution System main-( tenance recommendations and functional criteria, including immediate and future enhance-ments. This review included safety analyses ar applicable to the maintenance program.

The following repoit desetibes the scope, methods used, and results of the investigations described above.

Discussion APS reviewed PVNGS Electrical Distribution System vendor instruction and maintenance manuals to identify vendor-recommended preventive maintenance. This review included the l fo!!owing Electrical Distribution Systems:

- 525-kV system (in-plant switchyard area only) kV system (system designator MA)

- 13.8 kV system (system designator N A)

+ - 4.16-kV system (system designators NB and PB)

- 480-V load centers (system designators NG and PG)

- 480-V motoi contiol cenicis tsystem designators Nll atul Pil)

- 120-V power and instrumentation (system designators NN and PN)

-125-VDC power amt instrumentation (system designators NK & PK)

- Miscelhuieous electiical components associated with the Fast Bus Transfer (FBT) and the Reactor Power Cutback System (RPCS).

30

( APS ieviewed the vendor-recommended maintenance and PVNGS operational history for critical electrical system equipment and components to determine the functional criteria.

APS identified the six restan requirements described in the following subsections (a through 0:

(a) fleanine of all exposed hich (525-kV ac) and medium (24 and 13.8 kV) voltane insulators, hushines and lichtnine arrestors.

APS identified cleaning of all high and medium voltage insulators, bushings, and lightning arrestors as a restart requirement based on the electrical flashover incidents experienced on the Unit 2 Engi-neered Safety Features (ESF) transformer, and on the Unit 3 main transfonner. Based on investigation and analysis results, APS de-termined that the above flashovers resulted from the accumulation of salt drift contaminants (from the cooling towers and evaporative ~

ponds) and the presence of a misting rain; lightning at the time of the incidents may also have contributed to the flashovers. Exposed 1*lant insulators. hushings, and lightning arrestors will be cleaned prior to rest;ut of any PVNGS unit.

(b) Lubrication and eveline of FBT associated breakers

( APS identified lubrication and cycling of Fast Bus Transfer breakers as a restart requirement, since the failure of one of the Unit 1 Auxil-iary Transformer Y-winding, feeder breakers (13.8 kV breaker IENANS01) to trip after a reactor / turbine trip caused an incomplete fast bus transfer sequence. The failure of the breaker to trip was caused by coil failure, or failure of the mechanical linkages in the breaker. APS reviewed the maintenance history for the above breaker and found no previous failures had been experienced; how-ever, scheduled maintenance had been waived twice consecutively just prior to the failure-due to, operational considerations. APS has not yet determined if waiving the scheduled maintenance contribut-ed to the failure of the breaker to trip; however, perfonnance of the scheduled maintenance may have identified potential or existing problems. The FBT breakers E-NAN-SOI A, E-NAN-S02A, E-NAN-S03B & E-NAN-SO4B will be inspected, lubricated, and cy-c'ed prior to restait of any PVNGS unit.

(c) Oil sample evaluation from Iarce oil-filled transformers l

APS identified sampling and evaluating oil from large, oil-filled trans-

! fonners as a restan iequiicment based on manufacturers recom-mendations. Oil samples will be evaluated to detennine the current k oil characteristics and possible degradation which could inhibit per-31 E_- -- -- - }

l-t e

n fonnance and lead to transfonner failure. Oil samples will be taken from the large, oil-filled transfonners (Main, Stan-up, Auxiliary,

( Normal Service & ESF) and tested prior to restart of any PVNGS unit.

(d) Inspect Lartre Oil-Filled Transformers and the Salt River Project (SR P) Casende Potential Transformer for Oil Leaks APS identified inspecting large, oil-filled transfonners (including the SRP cascade potential transformer) for oil letis as a restan re-quirement since oil leaks from large, oil-filled transfonnets may re- ,'

flect or lead to transformer operational problems. Transfonner oil leaks can lead to low transfonner oil level and failure, including fire and explosion. Large, oil-filled transfonners, including the SRP cas-cade potential transformer will be inspected for oil leaks prior to re-stan of any PVNGS unit. , ,

(c) Verify Proper Operation of the Dehydrating' Breather filter on the Main Transformer APS identified verification of proper operation of the dehydrating breather filter on the main transfonner as a restan requirement, since filter saturation can cause degraded transformer perfonnance

( or failure. The dehydrating breather filter of the main transfonner removes moisture from transformer intake air. When saturated with moisture, the dehydrating material in the breather should be replaced. Over-saturation may eventually lead to transfonner fail-ures or inadequate performance. P oper operation of the dehydiat-ing breather filter on the mam transformer will be verified prior to re-start of any PVNGS unit.

(f) Performance of a Service Test on the Non-lE Batterv System APS identified perfonnance of a service test on the Non-lE Battery System as a restart requirement to verify the ability of the system to satisfy design requirements (battery duty cycle). The non-lE batteries provide control power to a multitude of breakers and pro-tective relays. Improper operation of these breakers or relays could affect plant perfonnance and reliability. The service test is one of the recommendations of IEEE standard 450 (IEEE Recommended Practice for Maintenance. Testing and Replacement of Large Lead Storage Batteries for Geneinting Stations and Substations). A ser-vice test of the Non-lE Battery System will be conducted prior to restart of any PVNGS unit.

1' L

32

_ _ _ _ . -_ _ - 2

i.

Review of PVNGS PM Tasks and Scheduled implementation APS compared the maintenance recommendations described above with the existing Electri-cal System Preventive Maintenance Program. This comparison indicated that the existing Preventive Maintenance Program for the Electrical System equipment reviewed tecimically meets or exceeds vendor recommendations.

APS has developed a Preventive Maintenance schedule to ensure that the restart require-ments listed ht the preceding paragraphs are completed prior to restart of any PVNGS unit.

The identified PM tasks and functional criteria do not provide relief from the Equipment Qual-ification (EQ) related special maintenance requirements. These Special Maintenance Re-quirements are provided in the PVNGS Equipment Qualification List (EQL), a controlled document which provides required maintenance directions for electric equipment within the scope of 10 CFR 50.49.

Review of PVNGS Desien Channes Impactine PM Tasks APS reviewed current PVNGS design changes that affect the PM tasks. This review indicat-ed that due to the flashover occurrences (subsec: ion (a)) a design change package (DCP #

1,2 & 3FE-NA-041) has been initiated to install creepage extenders on the ESF trans-fonners. This change will help minimizing the flashover incidents and will extend the mainte-( nance intervals for the clean-up of the bushings. The ESF transfonner bushing creepage extenders installation is scheduled prior to a unit restart.

A design clarification currently being implemented at PVNGS is the installation of a drip loop on conductors connected to the ESF transfonner bushings. This drip loop will prevent water carrying salt drift contaminants from mnning down the condnctors onto the bushings; thereby, helping to minimize flashover incidents. Installation of the drip loops on the ESF Transfonn-er is scheduled prior to unit restart.

In addition to the immediate enhancements described above, a long-tenn study is being initi-ated to investigate the possibility of adding s'pecial insulator coatings which could also have a measurable effect on minimizing the flashover incidents and prolong the maintenance inter-vals.

i CONCLUSION l n The review of vendor-recommended pieventive maintenance combined with the teview of the existing Electrical Distribution System Pieventive Maintenance Pmgram indicates that the existing program technically meets os exceeds the vendor-recommended requirements. The maintenance requirements and functional criteria identified in the preceding sections will be performed prior to unit start-up to assme seliable operation of the Electrical Distribution sys-tem. In the future, the existing Preventive Maintenance Program in combination with the pro-

{

33 L- ---- _ - - - - _ - - - - - - -

posed design and proceduint changes will be sufficient to assure enhanced operational reli-(. ablity of the Electrical Distribution systein.

6 1

1. dP L

34

m-,.---, .

e

1. 0 g k( ,

e e C

CONCLUSIONS AND ACTION PLAN

.e e

i i

l C l

35

i .-

e .- ,

' ekI

.s -

'(

i

(

-i

. CONCLUSIONS AND ACTION PLAN Conclusions APS review and evaluation of the reactor trip history on PVNGS-(specifically complicated -

Ltrips and natural circulation events) with electrical systems scheme involvement, has result-

- ed in a recommended enhancement action plan which, when implemented. will help in the minimizing of complicated reactor trips and the preclusion of natural circulation events. This review and evaluation focused primarily on electrical control and protection schemes in partic-uhtr, with direct or indirect reactor trip involvement, and on the electrical distribution system o aligmnent and reliability in general. The specific areas of the above study included; identifi-cation of electrical distribution aligiunent -altematives, subsynch~ronous oscillation (SSO) re-lays, evaluation of the reliability of other relays at PVNGS and the electrical distribution system maintenance. The followings summary includes the teview and evaluation conclu- -

sions for each of the above specific areas:

Identification of Electrical Distribution Alignment Alternatives Of the ten identified electrical distribution alignment alternatives, APS detennined that one alternative is most feasible for minimizing' natural circulation mid complex trips. This alter-native _ entails a change to the initiation requirements of a turbine trip and a reactor trip when the plant is operating above a predetermined reactor power level. The turbine trip will be ini-tiated by the tripping of the generator breakers and the turbine trip will intiate a reactor trip. It will also include the initiation of a FBT upon the receipt of SSO relays signals. As a compensatory measure, in all tlue'e unit, one of the unit's reactor coolant pump' buses will be' supplied by the UAT and the other reactor coolant pump bus will be supplied from the SUT.

Subsynchronous Oscillation (SS0) relays l Of the five identified SSO relays operation enhancement alternatives, APS has determined that only three of the alternatives are presently desirable means to minimize the negative impact of an SSO relay operation and to improve SSO relay security, thus minhuizing compli-cated teactor trip situations. The three proposed recommendations are:

~

Initiate and implement design change to include initiation of FitT for SSO re lay operation; also add diveise ieactor trip in a second phase.

- Initiate and innplement design change to provide for SSO ielay quantities to be be secorded on digital fault secorders

- Initiate and implement design change to provide a voltage reference circuit which will only allow the target initiated by the relay operation

{

36 j

to be displayed Evaluation of The Reliability of Other Relays at PVNGS Based on this evaluation, APS has determined that further modifications to PVNGS power system relaying design are not warranted. The corrective actions taken to-date have mini-mized the potential for a recurrence of those problems deemed preventable. The relay-relat-ed problems that have resulted in entry into naturai circulation have been essentially ethninated.

Electrical Distribution System Maintenance APS review of vendor-recommended preventive maintenance combined with the review of the existing Electrical Distribution System Preventive Maintenance Program indicates that the existing program technically meets or exceeds the vendor-recommended requirements.

The maintenance requirements and functional criteria identified in the maintenance section will be performed prior to unit start-up to assure miiable operation of the Electrical Distribu-tion system. In the future, the existing Preventive Maintenance Program in combination with the proposed design and procedural changes will be sufficient to assure continued reliable op-eration of the Electrical Distribution System.

Based on the above conclusions, APS believes that this review and evaluation, with respect

( to PVNGS Electrical Distribution System contribution to the units complicated reactor trips and natural circulation events, has been adequately addressed. The implementation of the proposed recommendations (alternatives) should enhance PVNGS Electrical distribution system reliability and provide a major contribution to the minimization of complicated reactor trips and natural circulation events.

Action Plan .

The APS proposed action plan is to implement the recommended alternatives as detailed in their respective sections of this report and outlined in the above conclusion . The Action Plan implementation schedule is summarized in figure 3 (ACTION PLAN SCHEDULE).

L 37

ge ge ng ng c g n g ng W t r

a t

i r a ut d u o i

r u u d o a

t i a r t u u d o i

r a ut d u o ir a u u d o t

s rl r rl

• e r l oeu rl o e oeu ol e oe 3

r u yf yuf yf u o yf yf b er b e b er t

b er b er T r d d d r d e ts I o de dn ed r ed e ts N

l i

r l ' l l

l n l l 1 a

l a1 p la 2 l a2 U a 3 t s e t

s e s e t

s e e t t nh t s

nh t i nh t i nh t i nh t i I

~

ge ge ge ng ng ng ir a ir a ir a t t t t r

a uu t

u u t

u u r a a r

t d o d o d o t t s

• s rl cl s e rl oe oe e e 2 r oe u r r yf u yf u yf o o o

T I

t r

b er b er b er d

t r

t r

d d o o N i o ed r eh l ed r i r

i r

- U r p

l l

a3 .

l l

a4 t l a3 t

p p t

s e t

s e s e E i nh t i nh t i nh t L

U D

E 3H g e ge ge ge ge n g C i ng i n g a

ng ir a n g ir a ir a ES r a r t u u t

u u t

t r u t ut u u u

• a d u o d o d o d o d o ANA 1 t

s er r

oe l rl o e u rl oe yf u rl oe u rl oe u

GL T I o yf u yf b er yf b er yf b er I P b er b er F N t r d d d d d N U i o

r l eh t l eh t l

l eh t l

l ed r l l

ed r O p l

a4 l a 5 a4 a3 t t a3 s e I t s e t s e t

s e s e T i nh t i

nh t nh i t it nh it nh C

A .

e

- s i

y t

n yh t e n )

n r s f

i S o O e i ch y

a C i S

O D m e t l Bt, t a S S

S E n pi e S ne u R ig sw r r e n t

c o o I l oTe c o &me s a t

n t

U A t sAw ni t S c n

o O o n o

Q m e rUo a a t

c Cn i S

i t i p t i t E e se u n is i P a h a S d a R

t de oT di f Rn c n d c s v pod a i e sU if f S

yi u c . e i o r s i N

St a CPS o -

m n n d o T e y do me O

nn o r rCn pR o k

c i g

( is B dr la mt aT m o e I T

it e t go o bB me ed F

e c i t

ul n m e u C bA i

i t twtw n.F ed e t

sd yn t e

t a t r c r

A i

r af r s a s t l i

/

t s

r e o e Gt a y i n u a

c h d y S i e i

pnt f K D o o o di nt t ed a u y o l a

c n

S l e i t a r t a t t e A a s a g s l d i r T c i r v e O e e o ig e i

r e p S t n

R M D f

e t

Ro S R c I O

P l E

e S S

w I ll l

g

" a u a t

• t u u r

u t a 3 u o o o u t u

i o l l o o e l e

T I

l e f u e u u l

e u

l e

u f N f e

e r

f e e r

f e f u U r t s

r t

t s

r t

e r

t s I s 1 s t

_ i

. 1 i s i

t t r t r t t

r a r a a r t r

a t s t a a t

s s t t t

s e r e e s e s 2 e r r r e r

r o o o T o t t

t o o I t r r t t o r r N r o o o r o i r i i

r o U i r

p p r p p i

r p

i r

p E. ,

I

_. U D

E 3H e e e e g g e e C g a g a a g a

g ES a t

• a u t t

u t

u o u o t

u t u

RN 1 o l e l o l o o e T l e u e u

e u

l e l e l L I N u f e f f u u u

[UNA f e f f P e r e r e e d U r dr r

dr r r e dr dr d dr h 3 3 r c O 3 3 3 3 s I

T n o

C i t

_ A t a

n e

, m m e

t s e D y l p

E ,

r S R

I s

g s e y i m

n r U T t

k t a e s i

B c e Q hs F n h e e r

t t

a t E u f o c b B a R b o i

t a e g c S e g u g n E i d

gs r a i 1 N a o i n l a k t

a n n

O l t

t l c

v a r o i o s y e e d n

• I v e e l y T s r r c l p R h f o  ;

C iak r s hg a d l d

e g E A e ta id n m h n i

/ t h n a s

a F R k i T icr f a n X l t s O K Ci if o s o .l io h c F he e

S c g o r i s ta r d

e Xc r t

e N A

l a e inta c ke R l li c T n p nl i r a M f n i o s a u b e ialte v r

_ e s u r F il e n

i i

t c

n l

Ci Lb X

)

O

)

Mf

)

S

)

_ ) )

u a b c d e f

_ F O

g

ll i! [ l

3_ 8

, ; ;.= y3 . , , . o

N .. ,. 4 .

l(-

r l'.

t. .

. . , a i

' ATTACHMENT TABLE 1:

PVNGS Reactor Trio' Events e

i 40

e 3

, 2 r ~

1 2 2 2 3 1 1 1 2 1 r e u

1 o e e e e e e e e e T c e t t t t t t o

t o

t o

t t

o t

o t

o t

o o o o o o N B c N N N N 'N N N N N F O N N N

~ .

- i n n in n

. i o -

i m m im t 3 l

a a 9 4 4 4 3 4 r lu 1 tu a ir c h r

h r

h r h 2 3 O 0

_ N C

}

r

(

o r t

c e ~

4 2 0 a w O 0 9 3 0 3 2 O 1 4

4 5 5 5 8 5 e o 1 R P '

S e d

T ' o .

N 2 1

3 1 1

M 5 5 1 1 1 1 1 E

V -

n - r -

E n m -

o n y t e t e e e e e r r it a Sl u r t e e c S w P G u s

s y

t s w w e e d ed I

s S e o o j e s e co R m e T P P R a e nM a

1 T a P r n e e h F r k e n o P ec E t S

r e io it it s

it s

d a

y lo a LR iz t

c c

je O f

f f

f o e v

b Tb n n BO e O L d u w

op r

u t e f it c e f

o R o o f

AT L ir s r R o y e t a t T s P s b f ud TC f o or i'p e r

d a

s s

o s

o D e it in re Ona A n c t

T r P r o L o

L L d r ee o ipw ome E oa r o

h ig t

c g o o t

a o w t ior t

R it e aR t H a i n t e

t e in P it e u TP eD u

c e r u e a n R u u u I

% Dr d r oo w Den S t cr R

e o D D D p0 p a t cL ipib G Au s ip ip -

p ip ip i8r iC at r r N s s t n r r t

n ir T

r T

r Tm Tg r

ea TuT e ure e T T en T r r r o ri n Ry la r

on V lt i

T oP e

t e

r r o

r o

t et r io r

o o o t orF oz t i 1 m

t i P n o r v t c

t c va t c

t c

t c c c n ct R ot d a a d u a a a as t aor t io nn an E rr ra a e e at c e e e ee eh UA eio Rp L Ee I n R R InA R R R RT Rc 0 0 0 0 1 0 0 0 0 1 1 1 0 0- 0- 0- 0- 0-0- 0- 0- 0- 0- 0- -

3 8 6 1 8 0 0 2 9 9 3 9 7 8 9 8 1 0 1 4 4 6 5 7 0-0- 0- 0- 0 0- 0- 0- 0-r 0- 0- 0- -

e 5 5 5 5 5 5 5 5 5 5 5 5 8 6- 8 b 8- 8 8- 8- 8 8- 8 8- 8- - -

Rm 8 8 8 8 8 8

2 8

2 8

2 8

2 8

2 8

2 8

2 E u 2 2 2 2 2 5 5 5 5 5 5 5 L N 5 5 5 5 5 g

it n '

1 1 1 U 1 1 1 1 1  : 1 1 1 5 5 5 5 5 5 5 5 5 5 5 5 8 8 8 8 8 8 8 8 8 8 8 8

/ / / / / /

/ / / / /

t 2

/

1 4 1 7 2 3 7 4 4 6 0 n

e e t 1

/

2

/

1

/

0 1

/

1

/

0

/

0

/

2

/

0

/

1

/

2

/

2 v a 0 0 0 2 2 P 3 3 6 7 7 9 1 1

E D 0 0 0 0 0 0 1 1 1 1

[

d

- e r

1 1

1 r 1 1 e

e

- u e e s e s s s s e

t o

T c t o

t o e t

o t

o e e e N Y

B c N N Y N N Y Y Y

- F O -

- n

- - i o .

t l

a l a

r u u t c a i r N C

)

'ro (%

r t

c a e w 0 7 0 0 0 0 0 0

0 0

9 9 4 0

5 8 8 0 e o 0 1 1 1 R P -

1 -

S e T d o

N

^

1 1 1 1 1 1 M 1 1 1 1 E

) V de E - o - o a

- e n c c e c ad at at n i a a e r e e r e ib s e e R t o e nF e

uP nI F e u

Fe u n e

r u lt u R ,R dr P r e Gy i

R tyD tyD G T s w w a t e e e o o C r o

nT f a sn fa s n la n R L L t c

mi ax o Soit So u ia it u a pCR.TO s te r s i

t n M o u u c e

( de a u

dae u a ta io io r

R S.A r r M y r r iC ec t ec t b e r u u y wb o

1 eA eA y n p p n b C n . n igm b

d G e S Sl s a

o n

Ld EA igm n n d e op onig y o oi Et e Et e t

LE e t n t t it p it s s t ia ir vi a epi eS BR r i

d y d y ia it n eT u u i

t cT it um nS nS it g Dw in A a a n i p in Dw lo lo Ar o i Do .

TS I

nr nr ec t

p s

pi oe s pi oe ipi r ir ipF ipF ca ip G it r a o w ir t

a ow ip r

TrT r Wip r r t Tn a r t Tn a ne im Te r

T r

N l e tup tup t T

r t o t cr T

rl rl aR n r oys t

r o

o ar t

V t r cf r cf r op eo ooo o o o en P T i

oAo toAo t i cr t

ce rt c t c

t c t ns i t cS t

cL t

aC aC a en c s c s aT an r

oa a R aes aes er o er ee ce er o er it ai u o er E er o RtuL Rf o RG iRn Rt Rt o Ms Ri t Rt L RtuL .

0 0 0 0 0 0 0 0 0 0 0- 0- 0 0- 0 0- 0- 0- 0- 0- - -

3 3 4 3 7 9 3 3 5 6 5 4 0 4 2 3 3 4 1 0 0 0 0 0 0- 0 r 0- 0- 0- 0- - - - -

6 6 6 6 6 6 e 6 6 6 6 8 8- 8- 8- 8- 8-b 8- 8 8- 8- -

9 Rm 8 8 8

2 9

2 9

2 8

2 8

2 8

2 9

2 2 E u 2 2 5 5 5 5 5 5 5 5 5 L N 5 t

i n 2 2 1 1 1 2 2 U 1 2 1 6 6 6 6 6 6 6 6 6 6 8 8 8 8 8 8 8 8 8 / /

8 / / / / / / /

3 t /

6 5 5 8 0 2 1 1 6 2 n

e e p 0 0 1 2 2 3 1 1

/ /

0 / / / / / / /

9 9 t /

8 8 9 9 v a 8 8 0

8 0

8 0 0 0 0 0 0 0 E D 0 b

d e

r t 1 1

r 1 1 e

T c u e e -

s e e s s s , s e

t o

t t o e t

o t

o e e e

- B c o N N Y Y Y Y N F O N N Y -

n

- ' - i o e t

la l a

r tu cru a i N C

)

r

(

o r -

t c e 0 0 9 0 a

e o w 0 0

7 0 5

0 8

0 8

0 0

1 0

1 0

1 9 4 R P 1 S d e

T o -

N 1 1 1 1 1 1 1

M 1 1 1 E

) V - -

c de E at- o - o at e

a r

- e n h c a

c a R e e ad r

e e e e ib s e e t o

uP nI F e u

F e u n e

r u lt u R R dr P

r Gy nF e

i R tyD t yD G T s w w a t e e e o o C r fa s o ml nT a sn R f i n

l a n L L it t

c ax i ia So So u r s s u a eu (o

CR n e i

d it de a t

M o u u c S A t

n 1 O

T e

r tu ec eA a

r ec eA t

u M a

y b y

t a

r e

n io r

u p

i o

r u

p C

i r

n R

b y

wb o

C in gm n

igm b

d G

e S Sl s a o

n Ld EA n e n Et e d

e e op on ity p o t oi LE Et s s t ta i

in t

ir t

ig vi it ia epi BR d y d y nS ia it it n g eT u

eS u t icT r

it um A nS a- a n I

.p n Dwo Dw lo Ar o in Dc TS I

nr nr i r l t

s oe s oe ip ipir iWip ipF ipF ec ip ipc G pi ir a w t o pi it r w r Tr r T r r t r t Tn a ca ne r

Tem r

N le tup tup t a o T r t o t cr T Tn rl a rl aR n r oys t Tro o ar t

V t i r cf r cf r op eo oo ooo en P T oAo oAo t t i r

t ce r t r c tc o t c t ns i t cS t

cL aC t

c aC R

c aes s c aes s aT an ee oa ce er er iai t u

aen o a

er er o er o er o Rt E

L RtuL RtuL Rt RG InR Rf o Rf o Ms Ri t 0 0 0 0 0 0 0 0 0 0 0- 0- 0- 0- 0-0- 0- 0- 0- 0-5 6 3 3 4 3 7 9 3 3 3 4 5 1 4 0 0 4 2 3 0-0- 0- 0- 0- 0- 0- 0-r 0- 0-e 6 6 6 6 6 6 6 6 6 6 8- 8- 8- 8-b 8 8 8 8 8-8- - - -

9 Rm 8 8 8

2 9

2 9

2 8

2 8

2 8

2 9

2 2 E u 2 2 5 5 5 5 5 5 5 5 5 L N 5

+

it n 2 2 1 1 1 2 2 U 1 2 1 6 6 6 6 6 6 6 6 6 6 8 8 8

. D 8 8 8 8 8 /

8 / / / / /

t / /

5

/ /

5 8 0 2 1 1 3 6 6 2 n

e e 0 0 1 2 2 3

/

0

/

1

/

1

/ /

C / / /

t / /

8 8 9 9 9 9 v a 8 8 0

8 0

8 0 0 0 0 0 0 0 E D 0

m d e

r r

u 1

e 1

e s 1

e 1

e s s s s 1

t e

o T c t t o e t

o t

o e e e e B c o N Y Y Y Y. N F O N N Y -

N =

~

n i

o -

t la l a

r u t u c a ir N C

)

r

(

o r t

c a e w 0 7 0 0 0 0 0 0

0 0

9 9

0 4

5 8 8 0 e o 0 1 1 1 R P 1 S d e

T o '

N M 1 1 1 1 1 1 1 1 1 1 E

) V - - - - -

de E - o at

- o at e

a r

- e n i n c a a c

R e c e

ad r

e e e i

br s e e t o

uP F e F e u n lt R R dr r nF e

nI u e u u P Gy yD tyD T s a iR t

t e e G e w w C r o o o m nT a s fa s R f

o - n n la in a

L L it t

c ax Soit Soit u r s s u a te M CR d e a d e a n

a M t o

a u

io u c ir R e

S

/

(

O r u ec t r

ec t u M y r e

io r r u

C y w 1

T eA eA y b n u p p n b o C n igm n

igm b

d G e S Sl s a o n Ld EA n n Et e d te op ton y o oi Et e e i t t

p LE s s t a

a n t ir eS ig ivir t

ia ep BR i i d y d y nS it it n eT u u icT t

it n u A nS a a in i p

g n Dwo Dw lo Ar o i Do TS I

nr s nr ir t l

ec p pc G s oe ipi ipi oe w

ip ipir r T i.

Wip pF ir t pF irt ca im r iAr w ne ao r t

r a o t

r Tr r Tn a Tn a Te T N l e tup tup t T

r o t cr T

rl rl aR n roys t

r o

V t r cf t

r cf r op o ar t

eo o o ooo e i n c o i

P T oAo toAo t i t ce r t r c t t c t ns t

cS t

c cr aC a en t

R c

aes s c a s es aT an oa ce aC er er iaiu t o

a e

er o er o er ee E

L RtuL RtuL Rf o RG InR Rf o Rt o Ms Ri t Rt 0 0 0 0 0 0 0 0 0 0 0- 0- 0- 0 0- 0-0- 0- 0- 0- -

6 3 3 4 3 7 9 3 3 5 3 4 5 1 4 0 4 2 3 0-0 0 0- 0- 0- 0- 0-r 0- 0- 0 e

6 6 6 6 6 6 6 6 6 6 8 8- 8 8-b 8 8- 8 8 8- 8- - -

Rm 8

9 9 8 8 8 9 9 8 8 2 2 2 2 2 2 E u 2 2 2 2 5 5 5 5 5 5 5 5 L N 5 5 t

t in 2 2 1 1 1 2 2 U 1 2 1 6 6 6 6 6 6 6 6 6 6 8 8 8 8 8 8 8 8 8 8/ / / / / / /

t / / /

5 8 0 2 1 1 3 6 6 5 2 ne e t 0 0

/

1

/

2/

2/

3

/

0

/

1

/

9 1

/

9

/

9

/ 8 8 9 f v a E D 8

0 8

0 8

0 8

0 0 0 0 0 0 0 l

g r r

u 1

1 1 1 2 1 e

'1 e

1 e

1 e

T c e e e e e s t t t t B c t

o t

o t

o t

o .t o e o o o o F O N N N N N Y N N N N n

io t

- l a

r a l u u t c a i r N C

)

r

(

t o r c e a

e o w 4 2

0 0

0 0

0 4

0 0

0 0

0 0

8 7 0 5

1 1 1 R P 1 1

S e .

T d o

N M 1 1 1 1 1 1 1 1 1 1 E

) V de E r o

hc fol e b

r e n a

d n

r u

c n

n w

e u le t

w u o D E uP nI c

a e

e Tnr a

oep P o T r

o r n T h pt c f u

la d

Goit t

u p lo iR t R oio it r miw Mm h u t r

e tr s n PuSit f

nT t

d n o a Se g e in oni t a gm ate tad o o v s w nr dei m oy s a Su n C CR is s t yC r n a

s ome d

ep imf o t e S gk inr a o o W (

1 O

. T c

e x n Eg n

yi n o ev L

ys t FT eir

.r f

r n oI r e et a eL Fd nl e

uo Dt r l

n Dy uL e gB n -

ir u o f

t e

u C Ewo b im t la t r b y a i

S gt o Pu q iaa d o Dt u D EA de iL ueV nc M Fa eCr r sd r n s O LE t od lo n de n o

iar ue e de r u u u r x s r

eo w n Do t DR l

i a g

o c e co ue u BR iai h ca c t

o o t

c int cd c uC l it a

t i e a Wino r u e Owd O Be c n c A D m

lota 3. o i

l into r o n DuD p r OI O TS pt e Fc0 s ipP r ri r

p t

e ier pu is pp e

i G s Te i

pi 3I f Et ir u ip g r s ir a is r ic m r t T D r n F Ter r N e Ty r Te p na Tn lu Ti s r o he Th T S r s t S

o e r la oe r n rP r V r t

lt r r t rSi o

t oP oo o Te ct a oo oa l o

ei n i

T on t aS t

aR P t cl a t cl r n t

c e r t t i ct al a t

c ae t

ct ae t i cx t

c a

R ao aaei ah en en Rn eu aA e E eo ecpa e t pe eo o e Rt o RD Rt o R i

L RC RnOM i

Ois RIs Rb Ai t 0

0-0 0-1 0-0 0-0 0-0 0

L7 O

1 0

0 0-0 0-6 1 3 3 4 4 0 8 9 4 5 6 2 0 0 1 1 1 1 0 r 0- 0- 0- 0- 0- 0- 0- 0- 0- 0-e 6 6 6 7 7 7 7 7 7 7 Rm b 8- 8- 8- 8-8 8-9 8-8 79 8 8

8-9 8-0 8 8 9 2 2 2 E u 2 2 2 2 2 2 3 L N 5 5 5 5 5 5 5 5 5 5 t

in -

_ U 1 1 2 1 2 1 2 1 2 3

_ 6 6 6 7 7 7 7 7 7 7 8 8 8 8 8 8 8 8 8 8

/ / / / / / / / /

/

t 4 0 6 0 4 7 2 7 n

e e t 6

0 9

1

/

2

/

1

/

1

/

3

/

0

/

2

/

2

/

1

/

f

/

v a 0 1 2 1 4 5 6 8 1 2 E D 1 1 1 0 0 0 0 0 1 1

l l' l l  !

4 e r

r u -

2 e

1 e

3 e

2 e

1 e

1 e

2 e

t 3

t e

4 t

e T c s t o

t o

t o ~o t t o

t o o o o B c e N N N N N F O Y N N N N n n n

o im im i

l t 5 2 a

r la 2 4 u r tu c h h r

a ir 2 3 N C 1

)

(

r o r t

c e 0 8 0 a

e o w 0 0

1 9

0 0 0

5 7

2 1

0 1 0 9 0 1

1 1

R P 1 -

S d e

T o N M 1 1 3 1 1 1 1 1 1 1 E ~

) V -

de E d n

y lp i

ic

- er it r o r

e z

a r

a r

a e

r a

e r A s

a Fc e e uP p t i n a r u n n n n t nI r u A se s e e e e n o S eR G G G e i

t R r r n s s e G m r a nT E t udn a P r

m m m m le o ie w h

T Ca a a a a E n(CR 1

O T

- C W

n.

r e

P t

T o

s e E r

ie l

a r

sr ue Bmr Vo f h

iH g

o t

S L

e w

c S

L t

e w

o t

S L

e w

o S

t L

e o

w lo t

Cr r

n oe lu P k s EA oo n o in g

8 an t o o o t o toia LE t it t

w 3 r T

e t t t e e er F

BR en uu c e u lo 1 D

u e u

e u u u uo A Dl a f D lo Fty inyr D D D D Dt la TS ip pa i

l ltuilia ip p p ip ip pu ic G ipM r r icr a ux T r

ir T

i T

r T

r T

r Ta r l N e Tn t

T Ti r i t

F A C r re r ovle l

e r r le r le r le oy V t i

r om r

oer rC o dt nn ous ov caLe ov c e ov t l ce t t P T cp t t i

t c s t t c e cb am t

c cd aL aL R au al u a e uUp a e aL ee i

o er er er er o er E eq ea i e a Rp t

r gitn ir RP Rt o Rt o Rt Rt o Rs L RE RF

• 0 0 0 0 0 0 0 0 0 1 0- 0- 0- 0- 0-0- 0- 0- 0- 0-4 4 3 1 4 1 5 6 0 1 2 1 0 0 0 1 1 1 1 2 0- 0 0 r 0- 0 0- 0- 0 0- 0 -

e -

8 8 8 9 9 9 8 8 8 8 8- 8- 8-b 8 8- 8- 8- 8- 8- 8-Rm 8 8 8 8 2

8 2

8 2

9 2

9 2

0 3

8 2

E u 2 2 2 5 5 5 5 5 5 5 L N 5 5 5 t

t i

n 2 2 3 1 U 1 1 1 1 1 1 8 9 9 9 8 8 8 8 8 8 8 8 8 8 8/ 8 8 8 8/ /

8 / / / / /

3

/

5 t /

2 4 6 1 7 6 6 0 9

n e e t 1

/

1

/

1

/

0

/

2/

2

/

1

/

1

/

2 0

/

3 3

/

v a C 5 7 8 8 1 4 5 0 0 0 E D 0 0 0 0 0 0 1

lflI ;l 11

. ~; } -

1x- -

I ,

?'

.G "

ip 4

t r

t

- it n

o d s s a T e o

p mc u u l A v o

r U m dn re ic e e r

a f l h s a t d

n r

, oau tr p t r t a .

m o

r e r

h a

e e s n f u e R ub a r e .

t a

r F

' h t

t s k p n t a a tr i ep 0 h e s e 1 i t

E v a f d e r

b ot m 2 w t

e > r e r r r T o e ve p o e g d s de i n o l o d n ew er e lia L e c e n

e c o e f f d

T o U hr,db s re t

i k lo *F 'F* F > F 'F ' is n

f e a C 0 0 0 '

F 0 5 e o emo lu .

s er S 5 5 5 1 3 t w r w

s e u a b C 3 3 3 05 2 1 y e

e o h c R 2 g 23 $ $ u d

i v

e ns f

ne t c e e f S a r ar v t

a b h t n a

T m m r N t hg T e i s f e or f ht p u yr e m s

• s E m o h n a e s

r e

s e

)V t e h iax r

l i

t o,

e g t r

a u w l s

de E t a

r t

d u s a

lt o

t s

a c P o lt o

e A o t uP nI f

e s

n it d v f t

o o o l a b e

iR t

n a

i mnU o r t /

r o n e r m r

e r

u r e n d s nT t t e e n d iv h T o o s u d ht o p a id in t l c

.ACROT u y d b n c 2 if e d

( t s

e e ro e md n e

0 e S- d t

a  %%

a e

R h 1

EA C f a

o n

b t

o ed c n r n f l l

a f

u q

e r

N a A

N d e  %

5 5

s%%%% 0 0 0 0 l e

s s

LE s wh e e f o

v i

s e

)

a o wk l

u orp v BR (

s r h p e s

f s w o e A e t n n r u s e o t l

lo )

n ht TS m r e we ew o a c c c n f

e h G f o n b t b

e u s s e s

a k

(

i t

w N s n ig i d n s o d s o d y l e

V a r la a d e v

a d e n it v s P T ml o

id ie w0 0 i f it 9 9 9 9 9 9 99 99 99 t s

e p e e h 1 e c a v u s e r c 0 4 d a 0 0 0 0 0 0 e t

s u a. e r t

r a y s o s h na f

e S- 0 0 N 9-e r

a R 2 2 < < < shy co t t

S n b a c a A 8 s e re r

m o d a t t o N /8 e u H ubs dg he it d o n s 2 o r b u 5 -

f i d M n t d t r

n t

s lu b 05 n i d n e s o a o f l a o n i r

e id l

it a f f

) .

c r

o r R n o

i t

ar y n

ww o lc u

x u l

e w a n t t e e E b o d e ( ne I

it e E F o

p i cr f

s n

f s L n

a r p d dt n u h t

u "g e O umde v e

,, t a a a Sh S n "

d s

lec r t r p O

p r

p u S t i le e oc kr n t

a i

r e ld u w

o a o o t

s s t r t t l

o c d ow o w u u t S t o f e

e if w b b o n G e P S H H C R s

i c ie e ld t s

t s N d o

u e v a v o a a d V .

p e h co o F F d i

P M 1 2 3

. 4 5 6 H S r 3 4 5 S 2 p E T

O 1

t e

o N

t e

o t

N e

o t

N e

o t

N e

o N N

-