ML19290C407
| ML19290C407 | |
| Person / Time | |
|---|---|
| Issue date: | 11/16/1979 |
| From: | Angelo J Office of Nuclear Reactor Regulation |
| To: | Hanauer S NRC - TMI-2 UNRESOLVED SAFETY ISSUES TASK FORCE |
| References | |
| REF-GTECI-A-17, REF-GTECI-SY, TASK-A-17, TASK-OR NUDOCS 8001110294 | |
| Download: ML19290C407 (4) | |
Text
%
~,
ff f7
'o, UNITED STATES 7
,y j > 'r,(f(,g NUCLEAR REGULATORY COMMISSION g ( y : '/j[f
,ag WASrtlNG TO N. D. C. 20555 o,
a
%,'%d lid \\ I6
- +
fl0TE T0:
S. H. Hanauer FR0ti:
J. Angelo
SUBJECT:
FOLLOW-Ori WORK TO PHASE I 0F TASK A-17 Attached are draft documents which are intended to start work at Sandia Laboratories on or about January 1,1980 as follow-on work to Phase I of Task A-17.
One of the subtasks for this follow-on work was originally meant to be done in Phase II, and one of the sub-tasks is clean-up kind of work, more or less, from Phase I.
The other sub-tasks are really an exten-sion of work that was not originally intended to be done but which recent events suggest should be done.
These subtasks are described briefly in (A Proposal for Phase I Follow-on Work).
It is proposed to have Sandia Laboratories prepare a Schedule 1S9 (Progran and Budget Proposal) along the lines of our proposal in the Enclosure 1.
You are requested to review the proposal and indicate your concurrence or comments.
7 J. Angelo
Enclosure:
Proposal for Follow-on Work to Task A-17 cc:
M. Aycock 1733 252 M111oem
EilCLOSURE 1 l
TASK A-17 SYSTEMS IflTERACTION IflliUCLEAR POWER PLANTS A PROPOSAL FOR PHASE I F0LLOW-0N WORK The original intention of Task A-17 was to respond to the concerns expressed by the ACRS in its letter dated June 17, 1977, to Mr. Lee V. Gossick, Executive Director for Operations.
In that letter the ACRS offered two examples of possible studies.
The examples are excerpted from that letter as follows:
1.
Examine the physical configuration of safety systems (a) in relation to their presumed " redundant" divisions or channels, (b) in relation to their supportive sub-functions, and (c) in relation to non-safety systems and features, for actions or consequences in one that have a direct or indirect deleterious effect on another.
Such confi;ura-tions might allow failures or local hostile conditions to unduly inter-fere with the minimum functions required to remove decay heat after shutdown.
Particular attention should be given to the potential for
" cascading" failures leading to a terminal event which interferes with some aspect of the shutdown functions.
2.
Examine interrelated functions and actions as they relate to operating practices, such as the recent action which caused the burn-out o'f a diesel-generator during a loading test.
This would include reevaluation of Technical Specifications to ascertain whether undue degradation of minimum shutdown heat removal capability may be occurring because of un-recognized indirect connections between systema.
Another example would be the mechanical maintenance of a valve or pump in one train concurrent with electrical maintenance of switchboards or relay panels in the "redun-dant" train which is inactive but supposed to be imediately available.
1733 253
Phase I of Task A-17 will be responsive to some of the concerns noted above.
The NRC purposely abridged the task in order to get some useful results as soon as possible and in order to demonstrate a practicable and workable nethod.
We are confident that a practicable method has been developed and we believe that useful results will be demonstrated upon completion of Phase I.
Now we prcpose to do some " touch-up" work on Phase I that will make the results more useable.
This " touch-up" or follow-on work will be concentrated in these areas:
1.
Simplify the fault trees so as to make them more truly generic rather than specific to Watts Bar.
Another way to describe this simplification L
is to make the trees more granular. The purpose is to render the trees easier to use by others for specific plant analyses.
2.
Run additional SETS code analyses to identify cut sets that are more
~
fi
, i ',
,s peculiar to operator errors. The specifics of how to do this requires dc,d bp a
t some more thought.
Also, the fault trees may have to be modified p
hy [y p'y W t p
g slightly. One suggestion is to develop three categories of operator actions:
(a) actions that can be performed at the control consoles in the control room, (b) actions that can only be performed locally and manual-ly; and (c) actions that can only occur because of prior maintenance, sur-veillance and testing, installation and calibration.
3.,' Develop additional branches to the fault trees for mitigating systems fo*
pipe breaks or other ruptures of the pressure boundary.
1733 254
4.
Develop a " transfer" package of information that will assist a potential user. This package will explain how the systems interaction analysis t
was performed, including the methods and procedures for data input to the analysis so as to reduce the cut sets to manageable numbers.
It is estimated that these tasks outlined above will require nearly six months to complete, assuming that Sandia Laboratories will work on these tasks at a reduced level of manning, say from an average of four plus persons to about two plus.
Simultaneously with this follow-on work, there will be a nominal effort needed to complete the final report of Phase I, including some minor changes to the results of Phase I as they may appear in the draft report of Pnase I which is due by December 31, 1979.
The estimated funding for all of the above work, including the won k on the final report of Phase I is about $80,000.
1733 255 9