IR 05000344/1986029

Insp Rept 50-344/86-29 on 860728-0801.No Noncompliance or Deviation Noted.Major Areas Inspected:Tmi NUREG-0737 Open Items

ML20212P496
Person / Time
Site:	Trojan File:Portland General Electric icon.png
Issue date:	08/15/1986
From:	Richards S, Willett D NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To:
Shared Package
ML20212P486	List: IR 05000344/1986029 ML20212P483
References
RTR-NUREG-0737, RTR-NUREG-737 50-344-86-29, TAC-45175, TAC-45416, TAC-46486, TAC-51292, NUDOCS 8609030157
Download: ML20212P496 (7)

Text

7, ,

-- .

.

,

<

,

U. S. NUCLEAR REGULATORY COMMISSION

.

-

REGION V

.

'

Report No.. 50-344/86-29

. Docket No. 50-344 License No. NPF-1 Licensee: Portland General Electric Company 121 S. W. Salmon Street Portland, Oregon 97204-Facility Name: Trojan Nuclear Plant Inspection at: Rainier, Oregon

Inspection Conducted: July 28 to August 1,1986

.,- Inspector: YMM 9.J.WillettfReactorInspector 8'/[db Date Signed

~

Approwd by: h M 7'I'N S.Ripards, Chief,EgineeringSection Date Signed Summary:

Inspection During the Period of July 28 to August 1, 1986 (Report No. 50-344/

86-29)  ! $

Areas Inspected: ' Routine, unannounced inspection of TMI NUREG 0737 Open_-

Item '

Results: No items.of noncompliance or deviations were identifie i

, t F

f a

~

860903o13786gg),$44

,

PDR ADOCK PDR G

- - -_ , --_- - , - , .- , ,

. ,

n DETAILS

. Persons Contacted

• D. Keuter, Manager, Technical Services ~ m -

,

• J. D. Reid, Manager, Plant Services .

.

e

• S. Bauer, Engineer, Nuclear Safety and Regulation Department' x

• H..Dawson, Site SPDS System Engineer

~

'

L. Wildfoag, Plant Training -

M. Peery, Corporate SPDS Engineer -

D. Modeen, Corporate Staff

^

J. Thale, Corporate Staff *

• C. Brown, QA Operations Manager

• H. Moomey, Oregon DOE .'

• Denotes those present-during the Exit Interview'on August 1, 198 '

The inspector also talked with other licensee personnel during'the inspection. These included plant staff engineers, technicians,: licensed

'

operators, administrative assistants, and document control personne . Followup of TMI NUREG-0737 Open Items TMI Item I.D.2. (0 pen) Safety Parameters Display System (SPDS)

The inspector held discussions with managers and personnel responsible for the installation, modification, troubleshooting, program and input verification and validation. In addition the inspector also talked with supervisors and operators who use the SPDS and with the training department personnel responsible for training SPDS user This inspection attempted to address those areas of concern that were outlined in the NRC pilot audit in August 1985 as well as the requirements of NUREG- 0737 TMI Item I.D.2. -The inspector utilized the checklist provided by NRR which was specifically for a inspection of the SPDS at Troja The four concerns identified in the Science Applications

. International Corp. (SAIC) report of the August 1985 pilot audit, were:

1) system reliability 2) inadequate training 3) system testing incomplete 4) lack of user acceptance It appears that in the past the great unreliability of the system, which was due mainly to internal computer terminal wiring problems and in part to system size and an incomplete verification of software and input source, caused potential users to be skeptical of the system and dat . 2

,

'. '

Many improvements have been made to the system since the pilot audit to improve its reliability, verify its programming and data, and

~

enhance its usability as well as increase the options and information available to the use The bulk of the systems problems cleared up with the repair d approximately 75 bad or marginal connections , internal to,the - '

computer. This problem was due to a design change by Westinghouse, ,

-

the vendor, switching to 24 from 22 gauge wire but still using a wire crimper set for 22 gauge wire. Additional problems with

'

, software were overcome when the vendor obtained their.own P2500 -

computer in-house so that licensees' could troubleshoot software problems with the vendors help. Prior to this the vendor was running the SPDS software on a different. type of compute'r which would not model the-problems that the licensees! were encounterin The vendor also added a fourth CPU to the' system and the original drum memory has been replaced with a" static memory, which has increased the response and reliability of th e system to an acceptable leve '

The system is comprised of two redundant trains that are interconnected and rely on each other to function. Three of the CPU'S are required for operation, and CPU's 1 and 2 are required because they each can control the rebooting of the computer which has a functioning automatic rebooting feature. The computer, located in the TSC, is powered by the UPS security battery inverte The displays in the control room are powered by interruptible AC which till be picked up by the diesel generators during load sequencing, i

The System Verification, Validation and Testing Program was

. completed in several phases. First the vendor did the verification and validation (all of the software). Second, under licensee direction and witness, the vendor completed the acceptance testin Third the licensee completed the input verification hardware test

! (TPT-145) in May and June 1986. This test covered approximately 5000 inputs to the system. PGE committed (July 3, 1985 letter PGE to NRR)

to complete this input verification during the April-June 1986 outage.

I The four inputs remaining to be completed are : Two process

'

radiation monitor points for the Reactor Coolant Drain Tank Discharge, and Containment Sump discharge, the containment sump level, and C-9 the condenser available permissive. These remaining inputs are awaiting hardware for implementation. The licensee is currently evaluating a more comprehensive and useful containment isolation parameter display for the SPDS, one which would provide these inputs to other SPDS function During the testing program, the licensee identified numerous software problems such as incorrect vendor algorithms, and analog calibration constants. Supplement-1 to NUREG-0737 identifies that the Detailed Control Room Design Review (DCRDR) task analysis should be used to verify SPDS parameter selection, display, and functio ~

.

---

r - * - , e d - - my- -7--- - -

_

_

. 3

.-

The licensee completed a human factors review of the SPDS in November-1985 (General ~ Physics did this review and they were also the one who did the DCRDR). The licensee also implemented many of ,

the General Physics recommendations which are identified.in PGE 1041 DCRDR Vol. 2 Item 449 through 472.- A summary of this review contained in a' memo D. Modeen to J. Lentsch December 19,]985 (791-85). *

The~SPDS has been; turned over to operations by letter dated July 25, 1986.. The licensee estimates that system reliability /

availability during the past year has been greater than 99%. PGE

~

Corporate Engineering has had an engineer dedicated full time to the SPDS over.the past'two years who devoted 85% of his time to the

-

~ system The site has airesponsibleisystem engineer who has devoted in excess of '70% of his' time to the system over the past year. The -

licensee has and is continuing to improve trie capabilities and usefulness of the system now that the' inputs, software, availability

'and reliability of.the system have been demonstrated to be satisfactor The current formal _ procedure ~to control software modifications to the SPDS, Administrative Order A0-5-6 " Controlled Computer Software Changes" does not provide.for a review by Human Factors or by all organizations which may be affect'ed. The draft Nuclear Division Procedure NDP 200-5, due to be issued in August 1986, for control of computers that do' quality affecting work excludes the plant and TCS (SPDS) computer The system now has many more features and displays than the original packag The system is user friendly and this simplicity of operation does not intimidate potential user The SPDS alerts the operators to invalid or questionable data by a magenta color and a subscript following the data of P for poor or B for bad data. Additionally the SPDS alerts the operator of SPDS malfunctions by stopping the screen clock and the flashing of the cursor. The system will clear the screen as it reboots, leaving behind a message that the system has reboote The SPDS does not depend on a processing system which serves other users. There are no calculations by the plant computer (P250) for the SPD The only necessary input from the plant computer, to the SPDS, is the plant computer power supply for the digital _ data points accessed by the SPDS. This power supply is the same source which powers. the SPDS multiplexer and control room SPDS terminal screens.

, A Trojan specific SPDS users manual is in final draft and out for commen It is due to be issued sometime in August 1986. Until its issue the Westinghouse generic manual is in use. The training department will complete their lesson pian for the SPDS, after the users manual is completed, in time to teach the next class which is

scheduled for Sept 4, 198 t

, - - - , , .

Nw

.D

% 4

. ,e

? % ..

There is still'no direct relationship between the Emergency

~

<

Operating Procedures (EOP'S) and the SPDS. Trojan Phil_csophy is that the SPDS'is= intended to be just another indicator and that all actions are taken based on board instrumentation. Additionally, supplemental information for decision making is not available at the

~^ % SPDS. For example: If you were trying to determine if there was voiding in a RCP hot leg, you could look to see if pump amperage

'

'

,

, (not available on SPDS) had decreased or if you were trying to

'

verify flow, pump power (not available on SPDS) could aid you in your determinatio ,

Some of the E0P's identify that board information is also addressable on the SPDS. The licensee has a new program available on

-

the SPDS called Functional Restoration. This program is an exact duplicate of the flow paths in the plant procedures which are used for symptom / event diagnosis. The licensee intents to scrap the

. Bypass Inoperable Status Indicator program (BISI) as an

' unserviceable, unusable feature, but many of the inputs and data fields are intended to be modified and used for valve status indication and tracking. Currently there is no hardware or requirements for the Nuclear Data Link function / capability of the SPD Control operators still do not like the the SPDS and all persons contacted think that the iconics display is useless. 0ne person commented that how would you like to try and' drive your car using an iconic display instead of the warning lights and speedometer ? (The iconic display does not display change as a quantifiable magnitude and additional analysis / aids are needed to interpret changes.)

The operators cannot visualize a use for the SPDS during casualty or transient conditions. There is no control function at the SPDS but on the panel there is both information, supplemental information and control as well as istdiate response to contro The Operators

,

.like-the.new Functional Restoration Program which takes plant parameters and' inputs them'into the logic paths of the symptom / event L diagnostics contained in the emergency procedures. This provides for i a automated fault tree analysis. The operators like the trending l

capabilitie's of the SPDS and have used this feature to identify problems, such as identifyin~g a increase in VCT pressure coincident l with a waste gas PRM alarm.in the Auxiliary Building.

t While the operators do not like and reject the concept of being forced to use the SPDS during transient or casualty conditions they are using select features of the system and think it is useful for steady state controlled evolutions. The licensee has changed their l simulator training course from Zion, Illinois to Pittsburg, Pennsylvania because the Pittsburg simulator' has a Trojan type SPDS

. integral to the' simulator / training. The operators still harbor distrust of the system because of past deficiencies and there are still complaints about the availability of the system because the l system is brought down intentionally by systems engineering on the l

day shift.

l l

l

__

,  :  ;~ ..

• - - -

, a,, :

.

ha y ..y - - -

.

'

,

~ ;

'

, g >~ U ' '^ ' '  !

'*

. . ;; j' - a rk * '

.

'

'

In summary, it would ' appear tihat the SPDSJ ha's tsregain its respect bP' through reliability and availability, andl user: personnel still'need

~

c - additional training on the modified featurescand new capabilitie ~ .

,"

..

'

L O TMI Item II.F.2 (OPEN)-Instrumentation for Detecting Inadequate s

. Core Cooling a -

-(' .The. licensee is currently in the process of, installing cabling for the core exit thermocouples. The old cable was not environmentally-

,~

qualified. The cabling will.be connected to the existing thermocouples during the next; outag New1sub-cooling margin monitors will=be installed during the.next-

,' outage. The current monitors are not seismically qualified and PGE analysis has concluded that replacement'is more desirable than upgrad ,

. TMI Item II.K.3.5-(0 PEN) Auto Trip of Reactor The licensee has submitted criteria and supporting analysis for

'~

manual trip. under specific conditions' in: lieu of automatic . trip actuation. PGE is, awaiting a SER from NRR

-

' TMI Item III.D.3 d (OPEN) Control Room Habitability The licensee has committed to perform a new design review / analysis and:resubmi.t this' analysis by September 1986. This new analysis,

.

which was performed by Bechtel, will include proposed TSAR and Technical Specification. changes. The following are some of the changes'and proposals 'which will be in the submittal. Design makeup V flow has been increased to 500 cfm and pressure in the control room will be maintained at a positive 1/8 inch water gauge relative to -

the' atmosphere. The control room is approximately positive. 08 inch water gauge relative to-the turbine builoing.-(the turbine' building has supply and not' exhaust fans). The licensee automatically turns off the controlsbuilding HVAC (CB-5,l6 and 70), in the event CB-2 fan trips, to maintain a positive pressure in the control roo Additionally, the Ammonia and Sulfur-Dioxide detection system is t .

be deleted from the plan I The licensee has still to submit a supplemental follow-up of LE descr.ibing the results 'of there investigations.and analysis of NRC findings identified during an inspection of-the control' room ventilation system during February and April.1986. This4 Supplement will. include the licensee's discovery that there was an.approximat'e

'

0; 6 by 12-foot opening in'the overhead of the vestibule (leading.to the control room that existed after the double doors were removed (in November, 1985). Additionally, there was;a ventilation tiuct the emptied into this vestibule area' from the CB'2 ventilation, system.

F, The remaining open TMI items : III.A ; III..A.2.4, 5, 6 are awaiting

' *

the Emergency Response Facility (ERF) final inspection projected for l 1987, a I3

. . .

.

}

l t

i

. . . . , . - , , , , - - - . . . . . ~ - - , , . _ - . . , - - , . . . , - , . , - . _ . - . _ - - - . . . - _ , . . . - , _ , . _

. - .. . . . . . .

.

_

.. 6

. .

'

, Exit Interview-

'-

The inspector. met with licensee representatives (denoted.in paragraph 1)

on August 1, 1986. The' scope'and findings of the inspection, which were discussed during the exit interview, are set forth in paragraphs 2 of this report.

..

.5 w

I

. 1, I ' F ,

o

.'.t ' '

q .c= $

t

=

+ $.

I o-

, s

. ,

b

'"4

'

h

i

.

,

i

,

9

?

- .- . ,_ , . . _ . . _ . . _ _ . - ~ . . _, . _ . , . . . . _ , ~ . . . - _ . _ . -.. _,,- _..._.._.. . . . . . . , , , . .__ _