ML19308C410
| ML19308C410 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 12/13/1979 |
| From: | Cornell E NRC - NRC THREE MILE ISLAND TASK FORCE |
| To: | Dorfan D CALIFORNIA, UNIV. OF, SANTA CRUZ, CA |
| Shared Package | |
| ML19308C407 | List: |
| References | |
| TASK-TF, TASK-TMR NUDOCS 8001230386 | |
| Download: ML19308C410 (1) | |
Text
-
,q t
1 V'
. :.a a rc o
-1, c,
UN!TED STATES E'
7 %
NUCLE AR REGULATORY COMMISSION t
C V. ASH f 4GToN, D. C. 20555
..;. l. i $
D,.
Deeember 13, 1979
^
In Reply Refer to:
NTFThi 791213-02 5tr. Dave Dorfan Department of Physics University of California Santa Cruz, California 95064
Dear 5!r. Dorfan:
Enclosed is a second draft of the NRC's Special Inquiry Group staff report on human factors which will form most of the human factors section of Voltae II of the report to the Commission. Two sections; evaluation of selection and training, and findings and recommendations, are stilI being worked on.
Please try to provide any comm.ents you may have by December 20, 1979.
If you have enough time, send them in writing; otherwise call in your comments to either me (301/492-S902), or Gordon Chipman (301/492-S924).
Sincerely, i
- s. % k L. -sLA-%
u.
c'M E. Kevin Cornell, Staff Director NRC/T5!I Special Inquiry Group
Enclosure:
Second draft: Human Factors l
l N
8 0 01
//3/77 I
EVALUATION OF PROCEDURES Introduction The actions of TMI operators during the accident as well as their own state af te rwards (Reference) suggest that emergency procedures were of little use either for diagnosing the problem being faced or for arriving at the appro-priate corrective actions.
This is not surprising since, as the analysis in section (refer to analysis of human factors and the accident) suggests, the written procedure of TMI had serious deficiencies. We did not perform a detailed analysis of all the TMI-2 emergency procedures. We did, through our contract with the Essex Corporation, perform a detailed evaluation of one procedure, 2202-1.3 " Loss of Reactor Coolant / Reactor Coolant System Pressure."
In addition Essex also performed an assessment of the-impact procedures had on the accident and on assessment of Metropolitan's procedures for developing and updating procedures.
The discussion which follows draws substantially from the Essex review.
In order to be of use procedures should be written in such a manner as to allow easy identification of which procedure should follow at the time. Essex found that the emergency procedures for TMI were not so written and thus 7
operator use relied heavily on memory. While this approach may be acceptable during normal operations or for single fault situations, Essex maintains and we agree, it fails miserably in multiple failure conditions, as was the case at TMI-2 on March 28.
l I
j
We feel t. hat the emergency procedure which would be most relevant to the situation at TMI-2 was 2203-1.3 " Loss of Reactor Coolant / Reactor Coolant System Pressure."
The Essex evaluation of this emergency procedure from a human factors engineering standpoint revealed a number of deficiencies including:
The procedure was not complete in several regards; o
It failed to define a leak or rupture which is within the capability of system operation; It lists symptoms but does not address diagnostic procedures and tests.
dicates that the CR0 should monitor liquid leveis, reactor building parameters, and safety feature flow rates, but does not indicate acceptable and non-acceptable values.
The procedure has several content coverage problems, notably; o
-- Step 2.2.2 under A, "close MU-V376 letdown isolation valye and start the backup MU pump if required" -- does not discuss how to determine if required.
Section 3.2.5 (A) states that continued operation depends on the capability to maintain pressurizer level and RCS pressure above the 1640 psig safety injection actuation setpoint.
The procedure completely ignores the situation where level is maintained well above its low level alarm point while pressure minutesaftertheaccide[ntinitiationthroughthe150-minute is below 1640 psig, the tuation that was present from 2 point.
Problems with procedure clarity and conciseness; o
-- Too many subjective statements are used in symptoms, such as
... becoming stable after short period of time."
-- It is not clear if all symptoms must be present, or only some l
subset, or only one of the symptoms, in order to diagnose the problem.
-- Section 2.2.2.1 of Section B states that the CR0 dedicated to recognizing a LOCA must accomplish four steps within 2 minutes.
Step four states that MUP di> charge cross connect valves must be opened within 5 minutes of the LOCA.
It is not clear how a l
step taking 5 minutes must be accomplished within 2 minutes.
l g