ML20126G175
| ML20126G175 | |
| Person / Time | |
|---|---|
| Issue date: | 12/31/1992 |
| From: | Cintula T NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| To: | |
| Shared Package | |
| ML20126G173 | List: |
| References | |
| TASK-AE, TASK-T92-10 AEOD-T92-10, NUDOCS 9301040098 | |
| Download: ML20126G175 (7) | |
Text
. .
UNIT: N/A Technical Review Report: ABOD 'I9210 DOCKET: N/A Date: December 1992 LICENSEE: N/A EVALUATOR / CONTACT: T. Cintula SUlUECT: PROSPECTIVE TREND OF LOW RELIAlllLITY EMERGENCY DIESEL GENERATORS
SUMMARY
The emergency diesel generators (EDGs) with less than a 95 percent reliability to start and load-run were identified. They account for approximately seven percent of the EDGs yearly.
The low reliability EDGs improved to the industry average, but were succeeded by a new group of low reliab;11ty EDGs yearly. No overall improving EDG trends were identified from 1988 through 1991.
DISCUSSION AEOD 'I92-08 used data from the Nuclear Management and Resource Council (NUMARC)
Initiative SA to show the great variance in the number of times the EI)Gs were started by different licensees. This report uses the same data to identify EDGs with less than average industry performance and to trend their performance in succeeding years. NUMARC claims that the industry average for the EDGs is better than 99 percent reliability to start and load-run. For this study, a start and load-nm reliability of 95 percent or less in any year was selected to identify a sub-par or " troubled" EDG. This threshold was chosen because Unresolved Safety hsue'A-44, " Station Blackout" (SBO) established a need for an EDG reliability program that had the capability to achieve and maintain EDG reliability levels in the mnge of 0.95 per demand or better to cope with a SBO.
The NUMARC data includes 195 EDGs at 63 stations (84 percent of the 232 EDGs at 72 stations) that was partitioned into four yearly periods from 1988 through 1991. The data is
~ identified by number only, so it is not possible, without further identification, to determine the plant name of the nuclear unit. Tables 1 through 4 identify the " troubled" EDGs and their subsequent reliability for the calendar year 1988,1989,1990, and 1991, respectively. -
As the NUMARC data ends with calendar year 1991, the subsequent reliability aggregate in Table 1. includes 3 years of data, Table 2.; 2 years of data, etc.
- Tables 1 through 4 follow:
I 9301040098 921216-PDR ORG NEXD
Table 1. " Troubled' EDGs -1983 l
Year of 1988 Years 1989,1990 and 1991 1
Number of: Number of: ,
EDG Reliability Re!! ability R Number :'dentification Starts & Failures Percent Starts & Loads Fallares Percent Loads
- 1. 2B 10 1 90.0 53 0 100.0
- 2. I8A 63 5 92.6 6S 0 100.0 H
3, 21A 18 2 88.9 1 40 1 973 l 4. 2iB _
17 2 88.2 46 2 95.6
- 5. l 25B l 19 2 89.5 a 55 0 100.0 l
- 6. 33B 15 I 93 3 51 1 93.0
- 7. 36B 13 1 92.3 59 1 98.3
- 8. 37D 19 1 94.7 70 1 98.6 g l 9. 48E Ii 1' . 9G.9 53 2 l 96.2
-h. g ll 10.. 55B- 25 2 92.0 -l 43 0 100.0 I1. 57A 19 I 94.7 66 0 100.0 !
~
g Totals 234 19 91.9 a 609 8 l 93.7 j 2
9 9
- w. . , . _ .
.,m. . _ . m._ m _
--, .m _~ . ., x:.
~ '
Table.2. 7roubled* EDGs -19S9
'If T .,
I' r- *
' e Year of 1939 - Years 1990 and 1991 P Number of: Number of:
EDG _
Reliability g Reliability Number Identification Starts & Leads Fallera - . Percent Percent
, Starts & Leads l Fallures
~
! !. 6B 19 I 94.7 ! 26 0 100.0 h
- s
-g ..
. l~ ~1 EA 15 i 93.3 56 I 98.2
-. a! 3. 178- 16 1 l 93.8 26 0 100.0 hu- -j -4. 21D 18 94.4 43' 100.0 1 O h.
- ' 5. 33A - 15 2 36.7 31 I O 100.0 6.' 33B 19 1 94.7 32 0 100.0 _
. 7. 39C 13 1 92.3 4 30 0 100.0 8.~ !' 43A 22 2 90.9 28 0 100.0
- 9. 45B- 20 1 95.0 37 0 100.0 l
- g 10, 52A 20 I' -t 95.0 'l 34 0 200_G i!
! II. 59B 20 1 95.fi l- 116 0 100.0 f Totsis - 197 13 93.4 9 459- 1 99.8 l 3
f i. fi ,{;! <[!l lilt t ;' ?l;j[ ft !i[: ;i!I- h! ': !tfi {
- a
- e s
E
,l~ i e
y t t i
l n 6 7 0 0 8 0 0 0 0 0 0 0 0 5 7 barce i
6 5 0 0 3 0 0 0 0 0 0 0 0 5 7 ie 9 8 0 1
0 1 9 0 1
0 1
0 1
0 1
0 1
0 1
0 1
0 1 9 9 ,
l eP R
r 1
9 9 s 1 e r
f u 1 3 0 0 1 0 C 0 0 0 0 0 0 1 6 o l i
r a s
e :
F f
Y o r ,
e b s d
m a u o N L 9 1 7 9 6 5 6 5 6 6 6 7 5 2 0 1 2 2 3 1 1 1 1 1 1 1 1 1 1 2 7 s 2 t
r t
a 0
S 9 -.f '
=
9 1 -
- y t
s t G
i l n i
b e 6 7 7 4 5 9 5 1 9 4 7 8 3 7 7 D ar c 1 6 4 4 3 0 9 4 2 4 6 3 2 4 2 E ie l
9 8 9 9 9 9 8 9 9 9 8 9 9 9 9
- eP d
e R l 4 b
u o
r 0 9 s T 9 1
e r
f u 2 2 I 2 2 1 2 1 1 I 2 3 2 1 3 3 o l
l 2 e r e t
a F h e f:
s Y o T r e
b s d
m a u o -
N L +
7 5 9 6 1 9 7 4 S 5 3 6 9 5
& 2 1 I 3 1
3 I 1 1 1 I ! 4 2 1 1
3 t
s r
t a
S I - l n
i o
- t a
Gc E B A A C B A A C A B B E B Df i 3 4 8 1 1 9 6 7 7 2 3 7 3 O Ei t I 1 2 3 3 3 4 4 4 4 . S n s e l a --
d t I o T
1 i '
r e
b .
m I 2
3 4
5 6 7 8
9 0 1 2 3 4 1 I 1 1 1 u _
N
- l1~
h .[
ei l[.
1 >-
' , ' ' j , .' <' i ; : .
. . ~ . ~ . -
. _ . . . . ~ . . . . -.. . .- - - _ . . . - . ..
i
-9 Table 4. ' Troubled
- EDGs - 1991 Year of 1991 -
Number of:
rug Reliability Number '
Identification Starts & leads Failures - Percent
- 1. 4B 21 3 85.7
- 2. 6C 13 -1 , 923
- 3. 11C 16 1 -93.8 4, 13 8 14 1 92,9
- 5. ' 17C 17 2- 88.2 ,
- 6. 18C. 36 2 94.2
- 7. 19C 18 1 94.4 8, 21A 14 1 92.9
- 9. 21B 18 2 88.9
- 10. 318 - 13 1 : 923 _. ,
i i
- 11. MA 18 o '2 ' S8.9
- 12. 36B 19 1 94,7 -
, 13.
40A - -15 __
1 933
- 14. 40E 25 :~
2 92.0 .
- 15. 41B .16 l' 93.8 I
- 16. 43C 17 - 1 94:1
- 17. 4SB 15- 1' I 933
- 18. 48D ' 37- 3: 91.9
- 19. -'49A- -
40 2 . 95.0
- 20. 51A 14 1- .. -92.9
^). --j 62B .
27 -2 l 92.6 1 j Totals .
423- .32 92.4 .
mam-- m amsa- - a mmmaa . mass.u se--_- . m.
r 5
1
- - } C.
i , & s
-
- E $ 1"J 9 td- ' /f
4 l
- FINDINGS The summation data for each year was:
No.of
" Troubled" Subsequent Yeat ERGS Reliability Reliability 1988 11 91.9 % 98.7 % ,
1989 11 93.4 % 99.8 %
1990 14 92.7 % 97.8 %
1921 21 92.4 %
Totals 57 92.6 % 98.9% -
The number of low reliability EDGs ranged from 5.6 to .10.8 percent with an average of 7.3 percent yearly. The number of low reliability EDGs is increasing yearly. -
The composite reliability of the identified " troubled" EDGs in the above table was only 92.6 percent. After identification of a " troubled" diesel, the subsequent reliability of the EDG improved dramatically - to that of the high reliability observed throughout the industry.
The changing identification numbers of the EDGs in each succeeding table show that existing
" bad" EDGs are replaced with new " bad" EDGs yearly.
Eight of the 195 EDGs (4.1 percent) had 2 years of below 95 percent reliability in the 4-year period. They were:
EDG. Initial Repent he h Xvat Xtat
- 1. 4B 1990 1991
- 2. 8A 1989 1990
- 3. 1IC '1990 1991 4, 21A 1988 199.1
- 5. 21B 1988 1991
- 6. 33B 1988 1989
- 7. 36B- '1988 1991
- 8. 48E 1988 1990
. No EDGs exceeded the unreliability threshold for more than 2 years in the 4 year period.
Three EDG3 (L5 percent) did not achieve an average reliability of 95 percent in the 4 year period from 1988 through 1991. They were EDGs Nos. 4B,21 A, and 21B. - The least .
favorable reliability was 93.7 percent for EDG 21B.
6
- .9;
~
A 4 > . _t . -> 5 _ m- -- , m . ., r
. ~.
d t dne station (station 36) was the only station where all of their EDGs were below the 95 Lpercent threshold. The "A" EDG exceeded the unreliability threshold in'1990 and the "B" EDG in 1988 and in 1991, i
CONCLUSIONS - q Approximately seven percent of the EDGs have a start and load-run reliability of less than 95 - ;
percent in any given year. These " troubled" EDGs start and load run reliability improves :
dramatically in subsequent years to high reliability commensurate with the industry average. -!
IIowever, yearly EDG rotation does occur; one years " troubled" EDGs are replaced with a new set of low reliability EDGs. 'me trend in the number of low reliability EDGs is-increasing yearly, and no yearly improvement is apparent in their composite reliability.
The study did not identify any permanently " troubled" EDGs and no station appears to have .
an inordinate number of low reliability EDGs onsite. 1 J
7 W
- t , i
- . ,
(c .