ML19270J140
| ML19270J140 | |
| Person / Time | |
|---|---|
| Issue date: | 12/26/1979 |
| From: | Ahearne J NRC COMMISSION (OCM) |
| To: | Udall M HOUSE OF REP., INTERIOR & INSULAR AFFAIRS |
| References | |
| NUDOCS 8001160306 | |
| Download: ML19270J140 (3) | |
Text
.
- j{#* " %'o, LVGossick UNITED STATES TARehm y",9~~j
- NUCLEAR REGULATORY COMMISSION RGSmith
- i
-l W^sHincTon, o. c. 20sss EDO 7399 T.
/
HShapar December 26, 1979 h*t CHAIRMAN gg Gleanne Cook HDenton Millie Broff WMinners a
son The Honorable Morris Udall, Chairman c
Subcommittee on Energy and the Environment SECY-79-2630 Committee on Interior and Insular Affairs United States House of Representatives Washington, D.C. 20515
Dear Mr. Chairman:
- This letter is in reply to your letter-of September 13, 1979, in which l you requested the Comission's comments on a proposal that future nuclear power plants be limited to one design developed and approved by the Government.
We agree that a standard nuclear power plant designed to incorporate the be t safety features of current plants would have advantages, if it could be developed without any pressure to emphasize cost reduction at the possible expense of safety.
New ideas and concepts could also be adopted.
If significant numbers of new plants are built, such a standard design would reduce the effort needed to both license them and regulate their operation.
Safety research, including experimentation and analysis, could be less diverse and focused on the one design.
In fact, if a standard government design were to be built, we would recommend that a confirmatory safety research program, possibly including a prototype, be undertaken to address the issues specifically related to the design.
Such a standard design would also tend to improve operations.
Plant operating procedures and operator training including simulators could be more standardized.
An obvious disadvantage to any standard design is that all plants are likely to be affected if a defect or error is present.
Since the probability of an accident resulting from an undetected design fault increases in proportion to the number of plants containing the fault, a safety hazard as well as an economic loss might result.
Even if there were no accident, all plants of all single design might have to be shut 2243 a18 Cricinated NRR:WMinners S00 0
The Honorable Mor'ris Udall, Chairman down to correct the common fault.
Even current designs, although diverse to varying degrees, have common elements and are subject to common faults as illustrated by the number of generic unresolved safety issues.
If relying on a single design were considered to create an unacceptable risk, then the lack of diversity could be overcome by developing two or three standard designs. A small number of designs would not defeat the purpose of the original concept, but would offset this disadvantage.
The concentration of design and associated technical expertise as a result of eliminating diverse design organizations might, by itself, be deleterious.
Our experience is that the investigation, analysis and evaluation of plants by separate vendor and architect-engineer groups is beneficial, not only because this results in a redundant review of the conmon elements of design (as well as the competitive features), but also because the different perspectives and methods of the design teams provide greater assurance that deficiencies are not overlooked.
A disadvantage of any plant based on a new design is the lower reliability generally experienced during the initial phases of operation: Although the reliability and safety may be significantly better after an initial learning period than plants based on previous designs, design errors, improper implementation of the design and unanticipated problems are all to be expected in any new design.
If a standard design is to be practical, it must be capable of being built and safely operated at almost any site.
Thus, the standard design would have to satisfy the limiting design condition for each site-related natural phenomenon (earthquakes, flood, wind, etc.).
For any specific site, some features would be over-designed, i.e.
designed to withstand natural phenomena more severe than would be expected at the site.
This is probably more of an economic disadvan-tage than a safety disadvantage.
A final disadvantage would be that using a government endorsed design might dampen the initiative of all parties -- vendors, architect, engineers, constructors, operators -- for seeking more effective ways to increase safety margins.
In my view, nuclear power plants are too complex to be designed solely on the basis of the regulator's minimum safety requirements.
Your proposal would tend to concentrate responsiblity for design in the jovernment's hands, thus perhaps decreasing the motivation, willingness and capability of all other involved parties to evaluate the safety of the design, and to assure that it is properly integrated with the operating staff and procedures.
We should note, however, no matter how standard the design, individual plants may have components (pumps, valves, etc.) supplied by different vendors, may be built by different construction companies during different climate and weather conditions,'and will be operated by different individuals.
Even with the same vendor providing 2243 319
The Honorable Morris Udall, Chairman a given component, that vendor will have different models as his own design evolution progresses.
If one were to standardize down to the component level, this would entail procuring the component from a single source of supply with all the disadvantages that such lack of competition entails. This is further aggravated with time (e.g.,
it probably will be impossible to buy the exact valve 20 years later if it needs to be replaced - the manufacturer probably doesn't make it anymore).
Thus, no matter how standardized the design, there would still be a need for a continuous program of qualification of new component designs as these designs change from vendor to vendor and as they evolve from the same vendor.
With regard to your second question, such a standard design would not affect the number of items on the current list of unresolved safety issues.
These issues apply to plants of current design.
Of course, resolution of these issues would be incorporated into the standard government design, just as it could in standard commercial designs.
While it is possible that a standard design would result in the identification of fewer safety issues in the future, the resolution of safety issues and implementation of necessary changes would be simpler with a standard design.
The NRC has no experience from which to derive an answer to your third question concerning the time that would be required to produce such a design.
If the design were only a small departure from or evolution of current designs of light water reactors and described in equivalent detail, we would estimate that a safety review would require a minimum of two years, and more likely three, to complete.
Instead of one discrete review period, it would probably be advantageous to involve the regulatory perspective starting early in the design development and continuing coordination through to design finalization.
A different concept, such as a gas-cooled reactor, might require an additional year.
If you have any further questions, we would be pleased to answer them.
Sincerely, b' [ L hn F. Ahearne 2243 a20
r.rr :. - - -.
- c.. s n. uo.u.m c. aw.~
......,-..ov..
.... u..... u u..
,,,,,,,.g,,.,,,,,,c.;
........m.....
.u.
...,.6..,.,.
COMMITTEE ON INTERIOR AND INSULAR ATFAl._.s.
gg,,.,,.,,.,
,u.
....vo.......
n...................
...o%.
u.o
,,,,,...,,,g e...*
...s...3 U.S, HOUSE OF REf"RE5ENTATIVES w......
....so...
....2*........
..... e. (,.. ),.......s.a.
.u.
,7,g.,g,... g g
..*s.a.:.....,.,,..
..... 4 u..s...... ca u r.
WAS HINGTON. D.C.
2*315 o.....
,,gg.,Lg.,,,gL
.e......*s,.,u.
. w e.......
vv..
<......=.r.
F:* Lt 2 s eTsoas
....u.
- o..< c. c. :
r-m. u....
............,..ou..
September 13, 1979
,e my eoo..a
,.u...............
....n........
.:....,.....n......
...,......u......
a..
u........u.
..........c.
.........,,oou,.
..,.......m.,.
en.g ke. e. 3, p.v.
The' Honorable Joseph Hendrie Chairman, Nuclear Regulatory Commission Washington, D.C.
20555
~
Dear Chairman Hendrie:
The accident at Three Mile Island raises questions as to whether the diver:.ity in nuclear power plant. design is so great as to overwhelm the ability of those involved in the regulatory process to sufficiently analyze each plant.
In addition, it appears that each plant has its own peculiar strengths and weaknesses with.no one design as safe as it
~'
would be vere it to include certain features from other designs.
Also, in some circumstances the desire to keep costs down seems to have led to decisions to omit certain features that would increase the safety of a p, articular reactor.
In order to reduce safety hazards resulting from diversity.
in design and failure to incorporate in each plant features that maximize safety, I am considering recommending that the Atomic Energy Act be amended to required that after some date (e. g.
January 1, 1985) no construction permit be issued except for plants designed in accordance with a general design develo.ed by the DOE and approved by the NRC.
Such a design might be a hybrid of existing light' water reactor designs (or it might be a gas reactor) but would in any event be based on existing technology, and would not. require significant research and development. An essential element of the basic design would be that changes could be incorporated into it as necessary.
2243
- 21
~
1 The Honorable Joseph Hendrie page 2 I would appreciate the Commission's views on this subject.
In particular:
(1)
What would be the advantages and disadvantages of a regulatory system in which the only plants eligible for licensing would be those built in accordance with the DOE design?
(2)
To what extent would such a design reduce the number of items on the NRC's list of high priority unresolved ger.aric safety issues?
(3)
How much time would be required to specify an optimum design consisting of a hybrid of current designs?
1 Tn order that the Subcommittes might have the benefit of j
the Commission's comments when we consider amendments to j
the Atomic Energy Act, I would appreciate receiving your response by October 31.
Thank you for your assistance.
Sincerely,
~
~
ORRIS K. UDALL Chairman 2243 322 0
O e
I
.