ML20246F841
| ML20246F841 | |
| Person / Time | |
|---|---|
| Issue date: | 09/12/1988 |
| From: | Ree G Advisory Committee on Reactor Safeguards |
| To: | Michelson C Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-CT-1941, NUDOCS 8905150118 | |
| Download: ML20246F841 (4) | |
Text
-_. -__ ._ _ _ .
7f
[%
/ %
UNITED STATES 'fpM 6//N8 J~ n NUCLEAR REGULATORY COMtUSSION
, U E ADVis0RY COMMITTEE ON REACTOR SAFEGUARDS
- W ASHINGTON, D. C. 20555
} September 12, 1988 F
MEMORANDUM FOR: C. Michelson, Chairman Subcommittee on Maintenance Practices and Procedures FROM: G. A. CRS Consultant
SUBJECT:
COMMENTS ON THE MAINTENANCE PROPOSED RUI.E. THE I DUSTRY STANDARD, AND ON THE SUBCOMMITTEE MEETING OPTUL 7. -
I'
, 4 h) e Since these comments will likely follow by several days the full committee c'ecision and letter, and since I expressed my points orally at the meeting.
I will be brief for the record as follows:
- 1. The " expectations" and "shoulds" for an Industry Standard as listed on pages 6 and 7 of Enclosure 1, are quite general, and the NRC staff people (including Resident Inspectors) can interpret to create moun-tains of paper to negatively impact the industry costs and safety.
The NUMARC representative indicated they expected to use their exist-ing guideline documents for the most part for the Industry Standard, O
c4 There should be a prompt meeting of minds to resolve NRC "expec-gO tations" vs. NUMARC-INPO. Personally, I feel INPO is burdening the o plant people niore than enough right now; progress is being made, and
$ the NRC should not change or magnify present directions and in-e clusions.
w -
86- 2. The NRC staff showed slides and furnished documents that clearly
$ addressed nuclear utilities with " poor maintenance programs" as the la$ - target of the rule. Upon questioning, the staff stated the word R$d intended was " performance" not programs. This might be followed to "O see if future paper 9errects wording to " performance" since if the
" programs" wording continues, and is so interpreted, mountains of paper can result even where a utility's performance excels. I h 3. At one point in the NRC presentations, a speaker remarked that if Gg M
industry comes up with a good standard then the NRC staff would write a "one page guide." This seems appropriate even if the present INPO m'
guidelines documents are used.
$5o/
@ \j i t M 4. The loose use of t5e words " BOP" and the issue by some that BOP should
$ come under NRC regulations should be clarified. The failure to g jf include aux 111ary icedwater systems as safety-related equipment many A
g years ago should not be used as a " door opener" to clamor for B0P
- encroachment. Where auxiliary feedwater, instrument air, and certain D check valves are necessary to provide for certain decay heat removal
! $ service (or other vital safety need) the equipment should be
? o f
L_ _ _ _ _ - - - - - _ - - _ _ _ - _ - - - - - - - - . - _ -- - - - - - - _ _ - - - _
t.
4 classified as safety-related, not B0P. If a Maintenance Rule, Stan-dard, or whatever, envisions implementation into such as turbine maintenance then mamoth amounts of paper will result and all the cost benefit results in the enclosures to the V. Stello memorandum to the Commissioners (undated) concerning a proposed maintenance rule will become grossly incorrect.
- 5. I endorse the Option 3 route and believe the maintenance issue has been stirred up so much that some sort of rule has to be promulgated for more or less political reasons. Technical justification for a rule doesn't seem to exist at this point in time. I would hope the rule would be carefully worded so as not to impact negatively on the efforts under way, or to provide amunition for paper and procedural mountains vs. real craftsmanship application.
- 6. In my opinion, about one-half of all maintenance activity thus far in the budding nuclear field has been caused or related to flawed design.
(Even the latest AE00 sumary report on incidents stated design was '
responsibleorinvolvedinabout50%.) We all know that designers have escaped their appropriate responsibility by the " sole utility licensee" dump off, but we should continue in formulating new rules, guides, etc., to search for avenues to bring designers (vendors, component manufacturers, etc.) to accountability. Apparently in the study and creation process of this proposed rule, the issue' of " des- i ignated representatives" surfaced, but not in very good focus, and not j too much directed toward vendor-manufacturer designers. Regrettably, q utility owners and their maintenance people have struggled long, and l will continue to struggle to try to correct flawed design in the 1 at'sence of forthright involvement of real " designated representatives" I from the vendor-manufacturer designers.
The XRS subcommittee should note that not once does the words " des-ignated representative" appear in the meeting review package or the i meeting handouts, j
- 7. Another word missing from the paper is the word " selection." Lots is said about program indicators, training, etc., but the basic elementa-ry key to achieving fine maintenance is the selection of personnel who have the appropriate mechanical comprehension and manual dexterity.
This is just absolutely so for I&C maintenance. I can never forget the early days of a nuclear project where selection was ' e practiced ;
and where one worker broke enough equipment to keep Eb u w o dozen !
busy repairing same. The ACRS subcommittee might also reflect on TVA performance and the fact that TVA had not used selection techniques, and at least one TVA management person stated "give me warm bodies and I can train them to do the best."
- 8. The use of " reliability centered maintenance" techniques is somewhat prombted in the staff documents. In my opinion, it is much too early in time to consider applying reliability centered maintenance to nuclear. There are only about 100 nuclear plants in the U.S.A.,
nothing is standardized, and the variety of components and manufactur-ers are in the thousands. Without standardization, repetitive use of i eiactly the same equipment in like situations, a data bank cannot be l
! i i ;9-yy, <
,1 A , ,
- meaningful toward any useful reliability centered maintenance. The a fact that the. French may be using RCM only supports its nonapplicabil-ity here, since their nuclear is at least somewhat standardized.
9..
~
It would be unfortunate if maintenance gets so regimented and "verba-tim": that shif t operators cannot tighten the packings tf a manual, .
valve.-- vs. being regimented into a Maintenance Request route that J takes 10 days before a. maintenance worker might approach the valve --
af ter, of course, the MR has gone through Approval, Investigation.
Prioritization, Planning, Scheduling, CHAMPS, etc., and finally to physical action. -By this time under regimented and verbatim compli- i ance,- the valve' packing would have'been blown and the unit tripped.
i
- 10. Regrettably, " craftsmanship " and motivation are being lost or de-5 stroyed in the U.S. A. Since their destruction has proceeded so far, particularly in large companies and in certain geographical areas, there is probably no other way than to implement more paper regimen-tation for maintenance at many nuclear utilities. Hopefully, this implementation will not be forced across the board and at utilities j
where " craftsmanship," motivation, and performance exist. If such forcing occurs, the good performer will be torn down to the level of the poor, and I doubt that the poor performer will improve drastically
".t with his implemented paper mountains, unless other more astute steps are taken. -
- 11. In preparing for this meeting I visited a' plant and had telephone n communication with a second plant., The first plant has two' units and a total T/0 of about 300 people, of which 68 are assigned to (Mech.
Elec. and I&C), maintenance supervision, and physical work. This plant has an outstanding performance-record (nationally recognized --
not necessarily NRC) with -a unit forced outage of about 0-1 per year.
This facility practices " craftsmanship" and is structured to do so.
The second facility of one unit has about 1400 total people with 157 in maintenance and now practices maintenance paper much in line with
- what the NRC staff is proposing. The performance record at this second plant is~ poor by comparison to the first.
l It should be noted (and reflected upon) that in fir. Walter Scott's presentation (and, handout) in the " Summary :of Recommendation" that the German maintenance qualification and motivation emphasis is on
" craftsmanship." Unfortunately, the name of the nuclear improvement game in the'U.S.A. since the THI-2 incident, has been " throw numbers of bodies" at the job. " Numbers" of people without " quality" of people serves to decouple the striving for excellence in nuclear plant performance.
- 12. At the meeting, an NRC staffer stated that if a good performance was beingimpactednegatively(referringtocosts)becauseofforced implemented ' programs from the Rule or Guide, then he could avail himself of protection under the Backfit Rule. I would like to believe
~
l this is so, but doubt such opportunity would exist or be used once the avalanche starts, f
I
l; 2. -
^l J
9
- 13. In my opinion, the NRC staff is indulging in too many team inspections 1 i
and has too many planned for maintenance, and in other areas. When one considers all the INPO inspections, there is hardly any time left at the plants for the key personnel to focus on what they should be focusing upon. This is a generic problem of U.S.A. nuclear and is negatively impacting plant personnel attention motivation, and performance. i Also it is contributing to an unfortunate key and-experienced labor turnover, and the seeking of early retirement.
14 C'est la guerre!!
cc: R. F. Fraley C. Wylie J. Carro11
}
9 0
i