ML20084D906
| ML20084D906 | |
| Person / Time | |
|---|---|
| Site: | Ginna  |
| Issue date: | 03/01/1978 |
| From: | Daniels G ROCHESTER GAS & ELECTRIC CORP. |
| To: | Raymond W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE) |
| Shared Package | |
| ML20084D901 | List: |
| References | |
| NUDOCS 8304120669 | |
| Download: ML20084D906 (2) | |
Text
w.----
M...
o o
E.:
ROCHESTER GAS AND ELECTRIC CORPORATION = 89 EAST AVENUE. ROCtIESTER, N.Y.14649 n e..c..e
.... c.c.... o c 2 7on March 1, 1978 l Mr. William Raymond l Office of Inspection and l Enforcement, USNRC
SUBJECT:
Ginna Station Safety Injection and Undervoltage Schemes 1
1 l
An engineering review,has been performed to evaluate the l
l ' adequacy of the load she' ding and sequencing circuits at Ginna Nuclear Plant considerin'g'the potential for relay reset drift.
The engineered safety features include two SI accumulators, three safety injection pumps, two residual heat removal pumps, and two contain-ment spray pumps along with other equipment.
These components and their actuation circuits are divided between two redundant, indepen-
- dent " trains", A and B,
each of which is designed to mitigate the consequences of accidents.
Analyses of transients requiring ESP operation, including a spectrum of loss of coolant pipe breaks and main steam line breaks, has shown that a failure of any component, i or even an entire train, does not result in unacceptable conse-quences.
In fact, when analyzing these transients the worst single f ailure is assumed to have occurred.
Thus complete failure of a relay does not disable the entire ESF system or cause the transient conseguences to be more severe than previously analyzed and found acceptable.
If offsite power is available the undervoltage relays will not b required to function and a plant transient that generates a safety injection signal will sequence ESF pumps onto the 480 volt busses unaf fected by undervoltage relay reset times.
A potential problem could exist only when there is a safety injection signal concurrent with a loss of of f site power and a dif ference in certain under-voltage relay reset times exists which is large enough to cause both a close and trip signal to be present for ESF pumps.
The difference in relay reset times must be greater than the sequence time delay in order to cause both a close and trip signal to be present.
Except for safety injection pumps IA and IB and the containment spray
, pumps, ESF pumps are sequenced into operation with time delays of S to 32 seconds.
These time delays are larger than the difference be-tween reset times which have existed for calibrated relays with com-
_plete loss of offsite power.
The containment spray pumps are not i
?
8304120669 780403 PDR ADOCK 05000244 S
PDR I
c-weer no.
p.c:.rcu cas suo trect(~ conr.
p>TE March 1, 197&
2 To Mr. William Raymond l tripped by undervo!tage.
Therefore, only safety injection pumps
' 1A and IB could be affected under these circumstances-safety injection pump IC would be unaffected.
A full train of ESP equip-ment would remain even if pump 1 A or IB failed to start as a result of relay timing problems.
The consequences of events requiring l safety injection have' been analyzed, and found acceptable assuming the f ailure of a safety injection pump to start.
- The probability of the combination of events of ESF actuation, loss of offsite power and failure of a safety injection pump as the result of undervoltage relay timing failure is small.
The proba-events has been bility of pipe breaks initigting logs of coolant estimated to range from 10 to 10 per year.
The probability of breaks in pipes 2 inches in diameter and larger which initiate loss of coolant events 1s estimated in HASLI 1400* Table III 6-9 to be _
approximately 10 per year.
The probability of breaks in main 4
s team or feedwater lines should be comparable.
The probability of loss of of fsite power during3the time immediately following the break is estimated to ho 10
/yr (
Reference:
WASH 1400 Section 6.4).
The probability of relay timing difficulties is unknown but is less than 1.
There f ore, the probability of a large pipe break concurrent with loss of offsite power and relay timing difficugty
-3 reg.ultigg in loss of a safety injection pump is less than (10
)(10 or 10 per year.
The loss of 2 or more safety injection pumps is even less likely.
The NRC does not typically requirg analysis
~
of events which have a probability of occurrence of 10 or less.
Modifications to existing equipment which can reduce the proba-bility of f ailure of ESP components are being considered.
Our present system, however meets the design requirements of Commission and the General Design Criteria the Nuclear Regulatory of Appendix A to 10 CPR Part 50.
A substantial safety hazard as defined by 10 CPR Part 21 does not exist.
Y-George Daniels Manager, Electrical Engineering,
GWD:cem l
l l
l
- The WASI! 1400 report was prepared to represent " typical" nuclear I
power planto.
It is therefore not specific to Ginna but is appli-cable to the types of systems used at Ginna.
I