ML20115D730

Forwards Preliminary Rept Re TMI Item II.A.2.2 Concerning Emergency Preparedness Dose Assessment Capability,Per Request

ML20115D730
Person / Time
Site:	Calvert Cliffs
Issue date:	04/15/1985
From:	Denton R BALTIMORE GAS & ELECTRIC CO.
To:	John Miller Office of Nuclear Reactor Regulation
References
TASK-2.A.2, TASK-TM NUDOCS 8504190101
Download: ML20115D730 (9)

Text

C' s

BALTIMORE G AS AND ELECTRIC COMPANY P.O. BOX 147 5 B ALTIMOR E. M ARYL AN D 212o3 NUCLEAR POWER DEPARTMENT CALVERT CLIFFS NUCLEAR POWER PLANT LUSBY, MARYLAND 20657 April 15,1985 U. S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation Washington, D. C. 20555 ATTENTION: Mr. James R. Miller, Chief Operating Reactors Branch #3 Division of Licensing

SUBJECT:

Calvert Cliffs Nuclear Power Plant Unit Nos.1 & 2; Docket Nos. 50-317 & 50-318 TMI Action Item III.A.2.2; Emergency Preparedness, Dose Assessment Capability Gentlemen:

A preliminary report on the Reliability of the Dose Assessment Computer Used in Initial Dose Estimate is provided in enclosure (1). This report is provided in response to a verbal request by Mr. D. H. Jaffe, NRC Project Manager.

~

Should you have further questions regarding this reply, we will be pleased to discuss them with you.

Very truly yours, c >~

R. E. Denton General Supervisor -

Training & Technical Services RED /MTF/GCR/sjb Enclosure cc: D. A. Brune, Esquire [

G. F. Trowbridge, Esquire D. H. Jaffe, NRC /!Qh T. Foley, NRC f 'h 8504190101 850415 PDR ADOCK 05000317; F PDR

• 4 ENCLOSi3RE (1)

PRELIMINARY REPORT

.i.

RELIABILITY OF THE DOSE ASSESSMENT COMPUTER USED IN INITIAL DOSE ESTIMATE

PRELIMINARY RFJORT RELIABILITY OF THE DOSE ASSESSMENT COMPUTER USED IN INITIAL DOSE ESTIMATE Subsequent to observations made in CALVEX 84 (September 1984 Emergency Response Exercise), the Emergency Planning Unit (EPU) began' operational checks of the Dose Assessment Conputer System (EPU Open Item 84-213). Checks began in November 1984 and consisted of Unit personnel accessing the DAC daily during weekdays only. Prior to this, DAC surveillance was based on vendor notification that the system was down.

,In January 1985 EPU formalized the daily checks into a reliability study (EPU Open Item 85-09). The original study with daily checks on weekdays was revised in March to include 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> coverage with data taken once each eight hour shift.

EQUIPMENT The study considers:

1. Digital Graphics Incorporated's i.ainframe computer, a VAX 11/750.

2. Digital Graphics Incorporated's back-up conputer, a Sperry Univac V77.

3. Calvert Cliffs' Data Recording Terminal computer, a Sperry Univac V77.

4. Calvert Cliffs' CRT terminals, Techtronix 4014 terminals.

5. Calvert Cliffs' meteorological sensors, FRI 60 meter windspeed, 60 meter wind direction, 60 meter /10 meter temperature differential.

6. Chesapeake and Potomac's phone lines and Calvert Cliffs' modems.

Software being tested is Pickard, Iowe & Garrick's MIDAS dose assessment system and the Data Recording Terminal meteorological data collection system.

PROCEDURE The DAC system is available when it can produce a display of the 10 mile EPZ map and EAL using Emergency Response Plan Inplementation Procedures. The cal-culations are based on real time meteorological data from the primary neteo-rological system. Stability class is determined by tenperature differential.

Radiological inputs are real time hardwired inputs from the main vent effluent monitors. The system must work with no manual inputs and without delay.

The system may fail to produce an EPZ map and EAL with automatic inputs, but still be available for use with manual input of primary meteorological data. Manual input into the calculation menu can be made in five minutes by an experienced operator using readings from the main vent monitor and the V77 Data Recording Terminal or back-up meteorological analog strip charts on control room panels.

1

~

y }

-Availability; test _resultsare'loggedonceeachshift. If data is

!not taken-(work load problems, etc.), "no test" is marked on the Availability

-Ing Sheet.

Radiation Safety performs night shift tests. Emergency Planning I performs day shift tests.-

sAll data' collected are saved in EPU files for the duration of the study. Hard copies of the CRT display are maintained in a' log book. The

- time-data _ reduction -is maintained on the Dose Assessment Conputer Avail-ability log Sheet.

Monthly'and annual percentage availability is tracked on the Dose Assessment Conputer Availability Sheet and maintained in a log book.

Emergency Planning Unit action _during' downtime is to notify groups responsible for maintenance and repair, and to assure repairs are made.

The system is unavailable from the time a test failed until the time

a successful test is performed. Conversely, the system is available from the time a successful test is performed until a failure is experienced or

- notification of downtime is given.

Units of time assessed are hours per. year. The study considers a-base'of 8760 hours0.101 days <br />2.433 hours <br />0.0145 weeks <br />0.00333 months <br /> per year; the time when both reactors are below cold shutdown is deducted f::cm the base. The remainder of time is equal to

. 100 petrent of time that the' system should be available.

The total nunber of hours unavailable is deducted from the time equal to 100 percent, divided again by the time equal to 100 percent. The result is percentage availability per year. Availability is tracked monthly until 1 the year is conpleted.

.A back-up MIDAS program is maintained on a back-up computer (Digital Graphic's V77); time-for maintenance is not deducted from availability un-less a test of the back-up system fails.

RESULTS Data included:

January availability. 61 percent February availability 93 percent March availability 79 percent

. Year to Date availability 77 percent 2

4

- ,c -

P

-DISCUSSION

~

Two1 January outagesLwere a result of DRT circuit card malfunction and.

a teletype; interface problem.

c

-February downtime resulted from a VAX 11/750 - DRT interface problem.

The VAX was unable to obtain data from the DRT for 50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br />.

March downtime was due to operator error,'three days of MIDAS program problems and an instance of calculations not performed with' automatic data

- ings. - The calculations cannot be perfomed with automatic inputs _during sc aduled DRT. calls for historical data.- A test of the system directly after the call would have passed.

Maintenance 'of the DRT has -inproved availability in April - no lost

. time.due to DRT problems to date. Time lost in April to date was 26 hours3.009259e-4 days <br />0.00722 hours <br />4.298942e-5 weeks <br />9.893e-6 months <br /> due to MIDAS program problems.

This study.was developed to test the availability of the Dose Assessment Conputer model when used in the control room for a quick accident assessment.

- It requires that all automatic data collection and conputer interfaces be

-functioning. . The study tests.the longest chain of interfaces possible.

- Higher. availability is expected when manual input of data is used. This

-is the case when the dose assessment. personnel are assembled.

NOIE: The following attachments are provided for-.information:

ATTACH E NT 1~ Dose Assessment Conputer Availability ' l Irg Sheet ATTACHENT 2 Dose Assessment Conputer Availability Statistics ATTACHENT 3 - Procedural Guide ATTACHENT 4 MIDAS System Diagram I

f

' .g m

3

ATTACHENT 1 DOSE AN (X3fUIER AVAIIABIII1Y IDG SEE!Er JAIUM 198510 DECDEER 1985 Surveillance Test Results for (date) 0000-0800 Available Down No Test 0800-1600 Available Down ?b Test 1600-2400 Available Down No Test Time available: From to , from to Hours available this date:

Hours available month to date:

Hours available year to date:

Time down: From to , from to Hours down this date:

Hours down month to date:

Hours down year to date:

DOWrDE CAUSE:

Hardware Software DGI VAX11750 MIDAS DGI V77 DRT CCNPP V77 CCNPP 4014 CRT OIHER CCNPP SENSORS Explained C&P LINES />0DEM C0bitWIS:

ATTAONENT 2 .

4 DOSE ASSIX9fNr GMURR AVAIIABILI'IY JAPRIARY 1985 'ID DFIDefR 1985 12 POil11 AVBUE: 617. 767. 777.

JAN. FER. MARDI APRII, MAY . RIPE .R H.Y MK2ST SFfr. M'IDRER i MNFMER IE.

~

I 937.

901 -

80%

791 70% _ _

i i

b 60L 617.

501 40L _._

i 307.

I 201 101 _

OL Percent Availability = _C VI total hours -(165 ) hours unavailable = (79)% for Month of Pturh 1985 (74's) total hours Percent Availabiltief (212%) total hours - 608 hours0.00704 days <br />0.169 hours <br />0.00101 weeks <br />2.31344e-4 months <br /> unavailable = (77)7. for Year to Date (2184) total hours

r

- a ATTAODENT 3 CUIII SQERIK TGM: ET-D-18 GHIE Rev. 1 l, ACM M: Sheet 1 of 1

]

1. Perform Dose Assessment Conputer (DAC) operability check. Use ERPIP  !

i 3.0, " Radioactivity Release - Quick Dose Estimate" as procedure for access.

2. Obtain hardcopy of display on CRT and file in log book.

3. Log results of surveillance test on Dose Assessment Conputer Availability l Log Sheet (Attachment 1). Time is to the nearest whole hour. Round off I any hourly fractions. l

4. If conputer is down contact:

Mark Stachnik X4061 and  :

Mark Abrams (202) 296-8633

{

5. Perform a verification test when notified that conputer is available again. I

6. Irg results of verification test on Dose Assessment Conputer Availability log Sheet.

(

7. Track monthly and annual availability on Attachment 2 at end of each mtnth.

l l- .

1 I -

. n i

((

e u l 99 .

l

i g h ATTACBENr 4 MIDAS SYSIEM l

l BMX-UP Er SD60RS

FY ANALOG MIDAS SIRIP O MRIS ROCKVIllE y VAX 11/750 MAllE1WE mmm 8034 PRI!!ARY Er SENSORS ANAIDG DATA RE00RDIfC SIRIP OMRIS TDMINAL ANAIDG PRIVATE LIlE DISPIAY V77 .g7g g Er Er -- --

TOER TRAllfR CENIER IOCAL LI}E Er DATA 5 N N I / I / I f 4014 4014 4014 4014 CRT CRT CRI CRT EECIRIC 00tGROL 2Sr - mm =

l l

\

L