ML20202G294
| ML20202G294 | |
| Person / Time | |
|---|---|
| Issue date: | 05/01/1986 |
| From: | Catton I Advisory Committee on Reactor Safeguards |
| To: | Boehnert P NRC |
| References | |
| ACRS-CT-1844, NUDOCS 8607150360 | |
| Download: ML20202G294 (4) | |
Text
.__ _ _ _
i g-/gn TO:
Paul Boehneet i May 1986 f
FROM:
Ivan Catton
SUBJECT:
Thermal Hydraulic Subcommittee Meeting j
RE: Research Plan
~ /])
' 1 May 1986
' Street, Washington D.C.
l I.
MIST Follow On Program--
i The memo from Beckner to Shotkin, dtd 2 April 1986, contains reasons one might consider a MIST follow on study using the B&W facility at Alliance, Ohio. Beckner concludes that the follow on should not be j
pursued. His reasons are similar to those I expressed at the Palo Alto subcommittee meeting last January. The reasons Beckner gives as to why B&W plants should recieve more attention are certainly valid:
1.
high proportion of serious events, 2.
experimental data base alim, 3.
identified code deficiencies, and
- 4. inadequate modeling of AFW spray heat transfer.
j To this list I would add ; tot leg flow regime modeling as well as several other weaknesses in our modeling capabilities. NRR gives arguments for continuation that are similar to those above but are
)
i in terms of their licensing needs. NRR notes that they must 1.
confirm safety margin,
- 2. respond to internal and external requests for analysis of events that have occurred or might occur, 3.
examine potential problems and design changes, 4.
examine adequacy of operator guidelines, with code predictions.
The case for studying B&W reactors is strongly made. The question is what one should do to address the stated needs. Here I will only consider whether or not the proposed ' MIST Follow On Program" will do so.
The inadequacies of the Alliance facility have been the subject of numerous subcommittee meetings.
Briefly, the following deficiencies have been noted:
j
- 1. the downcomer experimental model only grossly resembles a plant e
I downcomer, 2.
the vent valves are not well modeled, 3.
the hot leg will not see the correct flow regimes, d
- 4. the steam generator AFW spray cooling may not be modeled well 1,
l enough to give the correct thermal center, and 3
q j
- 5. the downcomer configuration may preclude the loop to loop interactions that could occur in a full scale plant.
To address the needs of NRR, one must be able to scale the integral results to a full size plant. This can only.be done after one has addressed known weaknesses in the code (or codes) to be used. It is l
my opinion that the present facility will not do this. Further, the j
full power upgrade will not help. Separate effects studies are needed DESICITATED ORIGI1!AL 8607150360 860501 1
R/
PDR ACRS Cortified ny f ff CT-1844 PDR M'
W-1
instead of more inappropriate integral testing. A concern here is the tendency for a code to be calibrated against existing data without too much concern for its biases.
II. B&W Methods Development Program--
The politics of this aspect of the research program are more interesting than the program itself. NRR did not ask for the program and is holding up concurrence on the funding to force RES to carry out the " MIST Follow On Program". It seems to me that NRC management look at how business is done within the Commission.
needs to take a In a time of decreasing budgets, one does not have the luxury of petty politics.
B&W Methods Development Program will not get into what should be The done to remedy problems. Rather, it is a study to develop methodology.
There HRR is then supposed to use the methodology to make decisions.
are several programs a11 ready addressing B&W plants; 1.
A45 (AND-1)
- 2. HRR B&W Reassesment Program (Crystal River, Oconee, and AND-1)
HRR denies responsibility for the program. NRR has concerns about The HRR concerns have been communicated to RES. We have the program.
It seems to not seen the communication expressing the NRR concerns.
me that we should be privy to what is going on if we are to make a sensible recommendation.
It is my view that there has been enough methodology study. I believe we know what should and needs to be done. It is time to do it instead of studying it.
I III. Technical Integration Centers--
Before criticizing the Technical Integration Concept (TIC) as presented, I would like to note that a single focal point where NRC licensing research needs are addressed is needed. The TIC as described seems, however, to have the wrong focus. Presently it seems to be focused on the codes rather than a plant's devices whose l
behavior we need to understand. I would recommend that there be enveral seperate TICS, each with a supporting university program; 1." Thermal Hydraulic Research Center" where studies of the engineering aspects of the important fluid systems and components can be carried out. Such a facility should have the capability for large scale testing. A companion university research center might be nimilar to that presently ~,
at the University of Maryland.
2." Operations Research Center" where human factors type studies f
ranging from maintenance to operator training techniques and computer aided operations are studied. Here a companion university research center might be the Ur.iversity of Michigan I
Industrial Engineering School.
2 i
,s I !
3.' Electrical System Testing Center' where the various systems in use can be tested for reliability and usefulness. The center should have the capability for developmental work as well as confirmatory testing. One might choose either the University of California or Stanford electrical engineering departments as companion research centers.
a select group of universities should be associated As noted above, More basic research could be carried out at the with the centers.
universities in support of the work at the cent,ers.
The Other research centers might focus on materials and controls.
several research centers should be coordinated by a TIC whose focus The TIC proposed by RES seems is the scenario or possible accident.
to be the coordinating TIC without the support which at this time seems to be more important.
1 a
IV. Codes The first is the The code program appears to have three parts. the second is the NPA International Code Assessment Program (ICAP),
development progragm and the third is the documentation effort along with what might be a major new code develop [ ment program.
The ICAP is yielding a great d'eal of information about the codes It is too bad that the and their computational capabilities.
The main experience with their use is being gained by others.I have is on how the c criticism The models used in the codes are frozen whereas assessment process.
l the nodalization and time step used are at the user's descretion.
Adjustment of the nodalization and time step will result in different As a result nodalization and time step adjustment can answers.
overcome poor modeling good comparison with experimental data.
Questions about ill-posedness and nodalization.should be a part of the research program and deserve a prominent place in the promised code documentation.
The NPAs are essentially highly user oriented versions of TRAC and 1
It As such great economies of use should not be expected.
RELAP5.
is my view that an NPA should be a BNL type code that has effecient
]
One should write a code for a particular operation as its target.
an existing code. If you want a purpose rather than try to changefast running code then you should give so one.
if produced, will go a long way The planned code documentation, Such towards responding to the many criticisms made in the past.
commitments have been made in the past and we have seen -nothing.
the code QT program will entail would I
A clear statement about what have been very helpful.
There was discussion about efforts to daevelop a 3-D two-step version This could well be a major code development of TRAC in early 1987.
-...z... ~..
u.....
3
.-.:- -...: N :.. i.::....:.-.:.. :..:*.:
L It seems to me that the earlier code programs should be program.
completed before a new program is undertaken. Further, it is not clear that the approach planned for the next generation code is the most appropriate given the needs of licensing. Future plans such as those hinted at are a major commitment of funds and as such deserve an in-depth review before they are undertaken.
V.
Seperate Effects Program The Seperate Effects Program (SEP) is too lean. Of the four or five programs mentioned, only one is underway. None'of those mentioned addresses code problem areas. RES should be encouraged to enhance this area in a responsive way.
i
=
l I.
i
?
e 4
,