ML20040E523
| ML20040E523 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 02/03/1982 |
| From: | Smith I Atomic Safety and Licensing Board Panel |
| To: | AAMODTS |
| References | |
| NUDOCS 8202050032 | |
| Download: ML20040E523 (14) | |
Text
- _ _ _ _ - _ _ _ _ _ _ _ _ - _ _ _ - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _
Bd 2/.3/82 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION 00tKETED U9;T ATOMIC SAFETY AND LICENSING BOARD Before Administrative Judges:
'82 FEB -3 P 4 :09 Ivan W. Smith, Chairman Dr. Walter H. Jordan Dr. Linda W. Little W5Z',XC[
- 3;jc;;
n s.
In the Matter of
)
gg g ggh
)
I METROPOLITAN EDISON COMPANY )
Docket No. 50-289
)
(Three Mile Island Nuclear
)
(Restart)
Station, Unit No. 1)
)
g%/
February 3,1)(d g
GCCI; g
};
s FEB 4 7gg Jf MEMORANDUM AND ORDER p #fg+grn ON HPI EXAMINATION QUESTION
/7 wf2 a 3
- 4 g
Some of the TMI-l control room operator candidates f ailed the po of the October 1981 NRC operators' license examinations which required the i
candidates to list all conditions to be satisfied prior to terminating or I
throttling high pressure injection (HPI).
The Board considered whether to reopen the evidentiary record to evaluate the effect of these failures but has decided not to do so sua sponte.
Related notions to reopen are denied.
Background
In the Partial Initial Decision (PID) of August 27, 1981, in view of disclosures of cheating on NRC licensing examinations, we retained juris-d'ction to consider relevance of cheating to operator competence and to g503 s
8202050032 820203
~
PDR ADOCK 05000289 G
2-consider the adequacy of the operator training and testing programs.
E.g., P ID '1 45. We also retained jurisdiction following the PID of Decem-ber 14, 1981. PID 11 2012-23, particularly 11 2016-17.
It is widely perceived that premature HPI termination was a major contributor to the severity of the TMI-2 accident. Therefore we believed that if the TMI-l operator candidates were tc be found competent by-the NRC operator examinations, they should at least be well versed on the criteria for terminating or throttling HPI. Accordingly, somewhat as a spot probe, we requested and were provided the results of testing on that question. Our concern was directly related to the substantive quality of operator training and competence.
At the same time the Special Master, Judge Gary L. Milhollin, decided to consider the HPI answers in the context of the issues in the reopened proceeding. Although he expressed doubts about the relevance of the HPI matter to the issues before him, his tentative concern was that poor performance on the HPI question could be an indicator of the defeatability of the Licensee's training and testing program, the latter being an issue squarely within the scope of his jurisdiction.
Judge Milhollin has informed the Board that he will defer his determination until after the Board rules on the substantive safety significance of the HPI test resu'lts so that he can consider the Board's reasoning in his disposition.
I i
O By our order of December 16, 1981 we provided for party com-1/
ments.
We also have before us the relevant emergency operating procedures for TMI-1, the answer keys to the HPI question, and respective examiners' notes on the NRC oral tests on HPI procedures.
In this order we address only the substantive safety significance of the HPI question, leaving the possible cheating relevance of the issue to Judge Milhollin. The Licensee and the NRC Staff oppose reopening the record.
The Commonwealth of Pennsylvania sees no purpose in reopening.
The Aamodts and Union of Concerned Scientists (UCS) seek to have the record reopened.
-1/
The material briefs and comments are:
Aamodts' December 21, 1981 Motion that Hearing be Reopened... regarding the HPI Question; Li-censee's December 24, 1981 Comments on HPI Questions and Answers in NRC October Examinations; NRC Staff's December 24, 1981 Comments Re-garding the HPI Documents; UCS' December 24, 1981 Response to Board Order on HPI Questions and Answers; Commonwealth of Pennsylvania's December 24, 1981 Comments on the Significance and Relevance of the Answers to the HPI Question; Licensee's December 30, 1981 Response to the Aamodt Motion to reopen record; Licensee's December 30, 1981 Reply to UCS' Response to the 8oard Order on HPI Questions and An-swers; NRC Staff's December 30, 1981 Response to Parties' Comments on HPI Documents; Aamodts' December 30, 1981 Response to Licensee's Com-ments on the HPI Questions and Answers; UCS' January 4, 1982 Response to Licensee and Staff Comments on HPI Questions and Answers, i
l Discussion The Board has decided not to reopend the record sua sponte for rea-sons subsumed by the.easons for denying the parties' motions to reopen.
Appeal Boards on seseral occasions have spoken to the standards to be applied in considering whether the evidentiari record should be reopened to receive additional evidence.
Most frequently cited is Kansas Gas and Electric Company (Wolf Creek Unit No.1), ALAB-462, 7 NRC 320, 338 (1978),
where the Appeal Board noted that the proponent of a motion to reopen the record has a heavy burden, that motions must be both timely and addressed to a significant issue, and, in the case where the initial decision has already issued, the movant must establish that "... a different result would have been reached initially had [the proffered information] been considered." See also Pacific Gas and Electric Company (Diablo Canyon Units 1 and 2), ALAB-598, 11 NRC 876, 879 (1980).
Our partial initial de-cisions on the safety significance of operators' training have issued.
Although we have, as noted above, retained jurisdiction over the issues of l
training and operator competence, that action goes to jurisdiction itself, not to the standards for reopening an evidentiary record.
Accordingly, we l
l have applied the legal criteria for reopening set out in Wolf Creek and i
Diablo Canyon, supra.
l The Aamodts' pleadings, insof ar as they are addressed to the sub-l stantive safety significance of the HPI test results, rest entirely on the assumption that any answer differing from the Staff's answer keys is a failure which raises a significant safety concern. The Aamodts do not l
address the factual significance of the test answers. We deny their l
l l
L motion to the extent it is addressed to substantive safety issues for that reason independently, and for the additional reasons set out below in the discussion of UCS' motion.
UCS' motion requests that the test results and relevant papers on the April 1981 NRC examinations as well as the October 1981 papers be 2/
received on a reopened evidentiary record.-
Nowhere in UCS' pleadings do they even allude to, let alone explain the patent tardiness of the request pertaining to the April 1981 examination.
The request as
-[
it relates to the April 1981 examination is unjustifiably late.
Moreover UCS f ails even to discuss the significance of the April 1981 examination.
Therefore that aspect of UCS' motion is denied out-of-hand.
d The request as it relates to the October 1981 examination is timely and it relates to a significant safety issue in that the criteria for ter-minating or throttling HPI is a matter of safety significance.
UCS asserts that the matter is relevant to its Contention 10 which alleges that a lesson learned from the TMI-2 accident is that safety systems at TMI-l should be designed so that the operator cannot terminate the opera-tion of a safety system until its safety function has been completed.
No party disputes, and we agree that the matter is relevant to issues proper-ly before'the Board.
2/
UCS is vague as to the ultimate relief it seeks. Apparently UCS would have the Board, without any further ado, accept the proffered evidence, then decide UCS Contention 10 in its favor.
The Licensee, Staff and UCS all discuss the factual significance of the candidates' answers to the October 1981 HPI questions. The Staff op-poses reopening because all but one of the 32 candidates listed both the basic throttling criteria addressed by the questions and because this in-j formation coupled with the availability of written emergency procedures l
provides adequate assurance that operators will be able to take the proper corrective action. Response at 6.
The Staff therefore concludes that the i
relevant documents raise no significant safety issue.
Id,.
The Licensee and UCS are more detailed in their dispute over the safety significance of the issue and we address their positions below.
Immediate vs. Follow-up Actions As noted above, the written examination question discussed here is
" list all the conditions to be satisfied prior to terminating or throt-tling HPI."
Licensee observes that the NRC answer key for this question is based on Emergency Procedure 1202-68 (Loss of Reactor Coolant / Reactor Coolant Pressure) and goes on to explain that though some of the examinees l
l confined their answers to the material contained in the answer key, other examinees included throttling provisions permitted in other emergency pro-cedures such as: EP 1202-6C; EP 1202-68, Attachment 2; EP 1202-5; EP 1202-2; and EP 1202-4. Licensee comments at 1-4.
l Licensee explains that since the answer key is based on a section of the emergency procedure which involves a follow-up action as contrasted to l
an immediate action, operators would have available to them the written emergency orocedure. Licensee believes that "... the distinction i
i between immediate and follow-up actions is a major element in determining the level of detail which operators are required to commit to memory and thus can be expected to recite in a closed-book examination."
Id. at 4, 5.
Licensee further comments that, with respect to the evaluation and grading of the examination answers:
1.
All but one of the 32 examinees referenced both of the basic throttling criteria in EP 1202-68, i.e., 50* F subcooling and LPI flow.
2.
The majority of penalties assessed were for failure to include all of the elements of the two basic criteria.
3.
Grading penalties for not stating that it was only necessary to meet the 50* F subcooling or, rather than and, the LPI flow criteria were not justified.
4.
Grading penalties for using the wording "to keep from going solid" rather than "to prevent pressurizer !evel from going off-scale high" similarly were not justified.
Id. at 5-7.
UCS' comments give three arguments why it believes Licensee's argu-ments on immediate versus follow-up actions fail.
First, according to UCS, the conditions under which it is permissible to throttle HPI appear in the "Immediate Action" section in two emergency procedures and, thus, the conditions for terminating or throttling HPI in these emergency proce-dures should have been committed to memory.
Response at 3-4.
Second, according to UCS, the Licensee's distinction between immediate and follow-up actions is not valid because of the aroitrary and inconsistent placement of similar. statements in either the immediate or follow-up sections of the emergency procedures.
Examples are given.
4 Also, UCS states that criteria for throttling or terminating HPI should be i
placed in the "Immediate Action" sections because it has not been shown that for all accidents "... the conditifons under which it is P
permissible (or required) to throttle or terminate HPI will not occur shortly (e.g., sooner than 10 minutes) af ter the accident begins." UCS uses the TMI-2 accident as an example of HPI throttling and termination in less than 10 minutes of accident initiation.
Thus, argues UCS, it is appropriate to have the criteria in the "Immediate Action" section and to require operators to memorize the details.
Id. at 4-5.
Third, argues UCS, the criteria for throttling or terminating HPI are a special case, and premature termination of HPI was a major factor in the TMI-2 accident.
Operator training has been conducted on the criteria.
Examination on the criteria was conducted in April and October 1981.
At this point the criteria "... should be as rote as a nursery rhyme or the Pledge of Allegiance.
Licensee's argument that the examinees can be ex-cused for still not knowing the criteria because they are in the follow-up section of one emergency procedure must be rejected."
Id. at 5.
i The Board has reviewed the emergency procedures provided by Licensee on December 17, 1981 with particular attention to the immediate and fol-low-up HPI actions referenced in these procedures.
We observe first that
.~.
l
_g.
since the written examination does not seem to identify either the emergency procedure involved or the emergency conditions which cause HPI to be operating, it is not surprising that examinees interpreted and answered the question differently.
In fact, we think the examinees did a good job extrapolating what the examiners had in mind inasmuch is only one of the examinees did not list both of the two basic throttling criteria given in the answer key based on EP 1202-6B.
Commenting on UCS' first reason, we have examined the citations pre-sented by them, namely, Attachment 2 to EP 1202-6B and Section II.2.C.5 of EP 1202-5.
The former is referenced at Immediate Action step 7.d which refers the reader to the non-LOCA overcooling section (Attachment 2) if such overcooling is indicated.
At this point the operator presumably would follow the check list prov1ded in Attachment 2.
So, even though these overcooling recovery steps are referenced in the Immediate Action section of the procedure, the actual steps to be followed are presented in a check list, a form which is not demanding on the memory of the operator.
The latter citation does indeed incorporate the Follow-up Action section in Immediate Action section as UCS states.
Response at 4.
It seems to us here that since again a check list is provided, the demands on operator memory need not be great.
In fact, it appears to the Board in this latter example, that if Licensee had simply labeled the Follow-up Action as Sec-tion 3.0 rather than 2.0.C, there could be no quarrel about what type of action is involved.
UCS' second reason for the failure of the Licensee's arguments on this subject is unpersuasive.
The Board does not find inconsistent the l
location in different sections of different procedures for different emergencies the steps for throttling or terminating HPI.
This seems entirely logical to us.
Pointing to the TMI-2 accident sequence, with its premature termination of HPI, as a reason for requiring all such HPI throttling and termination steps in "Immediate Action" sections, seems to be without justification.
Regarding UCS' third reason for failure of the Licensee's arguments, the Board agrees that premature termination of HPI was a major factor in the TMI-2 accident and that considerable operator training and examination has been conducted on this subject.
However, our view is that, given the manner in which the question was asked on the written examination, the sexaminees did a reasonable job of answering the question; only one of 32 examinees did not list both basic criteria given in the answer key.
Detail Required in the Answers Licensee notes that incomplete answers to correctly identified basic criteria reflect primarily on the examinees' inability to recall from memory all of the elements of written criteria which would be available for guidance in follow-up actions.
Comments at 6.
UCS states that the saturation meter was added to TMI-l and the in-core thermocouple readout was modified for the purpose of avoiding incorrect operator action.
Therefore, "[i]f the operators do not know how
to determine whether the 50* F subcooling criterion is met, their training has been inadequate." Response at 5.
Similarly, states UCS, there is a
"... significant difference between specifying > 1000 gpm LPI flow and specifying > 1000 gpm LPI flow in each of the four LPI lines and that the flow has been stable for at least 20 minutes."
Id. at 6.
The Board rules for the Licensee on this point because we agree that operators will have access to the written emergency procedures when necessary, and because the question asked on the examination did not clearly ask for all the subcriteria nor did the question identify the specific emergency procedure or emergency condition.
Conditions in the Alternative?
Some of the examinees were penalized for failure to indicate that the two basic throttling criteria, 50* F subcooling and LPI flow, were alternative possibilities, that is, that only one of the criteria need be met before throttling.
Licensee states that only one of the 32 examinees read the question as calling for a statement of the criteria in the alternative.
Comments at 6.
UCS however feels that Licensee's argument fails in the case of the "B" version of the examination because in this version, the word "all" was not included in the question.
Response at 6.
Even though the Board was previously unaware of this difference in the questions on the two examinations, "A"
and "B", we believe that for
either question, the examinees did not make a serious omission in their answers by leaving out the word "or".
One does not need to be a highly trained and qualified operator to understand that if the low pressure injection system is operating during a LOCA, coolant leaving the core i
region may not be subcooled.
I Penalties for Different Wording Licensee commented that some of the examinees were penalized for characterizing one of the elements of the subcooling criterion as a re-l quirement that HPI throttling be necessary to keep from going solid or to prevent filling the pressurizer. They seem to have been penalized for not using the precise wording of the emergency procedure "... to prevent 4
pressurizer level from going off-scale high." Comments at 6-7; EP 1202-6B at 6.8.3.11.
Licensee adds that since the function of the criterion is to allow the operator to keep from going " solid" once the subcooling cri-terion has been met, the answers should have been accepted as correct.
Id. at 7.
UCS disagrees, arguing that different operator action might result from following the procedure given in the examinees' answer.
UCS gives as an example a rising pressurizer level following HPI initiation during a LOCA.
Under these conditions, says UCS, an operator might decide to throttle HPI "to prevent filling the pressurizer", even though such action would not be necessary" to prevent pressurizer level from going off-scale high.
UCS goes on to explain that the break size might be such that flow
out the break will balance the incoming HPI flow, thus making throttling unnecessary.
Respsonse at 7.
The Board notes that the scenario described above by UCS is precisely the one which resulted in the core damage at TMI-2 and it is precisely this scenario which has resulted in the additional instrumentation to mea-sure subcooling for these accidents and in the additional operator train-ing to assure that proper operator actions are taken in these situations.
We disagree with what appears to be UCS' generalization that for some break sizes, the operators might be taking the wrong action.
We are un-convinced from our review of the examination answers that operators are poised and ready to terminate HPI prematurely.
Furthermore, we agree with Licensee that grading of the examination in this area was " irregular". Comments at 7.
Conclusion Accordingly, the Board concludes that the answers to the HPI ques-tions do not raise a significant safety concern, and that receiving the relevant papers into evidence would not bring about a different result.
The motions to reopen by the Aamodts and UC5 are denied.
We add the ob-servation that the operators who failed portions of the October 1981 l
examination have been rescheduled for NRC retesting in early February 1982.-3/Successful completion of these tests is a requirement for renewal of the operator licenses involved.
ORDER OF THE BOARP:
FOR THE ATOMIC SAFETY AND LICENSING BOARD d/
Chairman Ivan W. 5dith
~
ADMINISTRATIVE LAW JUDGE Bethesda, Maryland February 3, 1982 3/
TMIA Exhibit 90 at 1.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _