ML20004F028
| ML20004F028 | |
| Person / Time | |
|---|---|
| Site: | Purdue University |
| Issue date: | 06/08/1981 |
| From: | Stansberry E PURDUE UNIV., WEST LAFAYETTE, IN |
| To: | James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| Shared Package | |
| ML20004F024 | List: |
| References | |
| NUDOCS 8106160409 | |
| Download: ML20004F028 (2) | |
Text
s e PURDUE SCHOOL OF NUCLEAR ENGINEERING June 8, 1981 Mr. James G. Keppler, Director Region III United States Nuclear Regulatory Comission 799 Roosevelt Road Glen Ellyn, IL 60137 RE: Report of vio ition of Tecnnical Specifications Cear Mr. Keppler:
This letter is in response to the telephone call and telegram dated June 2, 1981 in which we reported that tests made on June 2, 1981 indicated that reactor scram times may be in possible violation of the Technical Specifications.
Problem Procedures developed at the time of the installation of the reactor in 1962 measured the rod drop times, the time from interruption of the magnet current to the full insertion of the safety control rods. This time initially was specified as less than 500 milliseconds but was found to be too short and the specification was subsequently changed to be less than 600 milliseconds. Measured rod drop times, as published in the annual reports, fall between 593 and 563 milliseconds. This parameter, the rod _
drop time, was considered important for the safe operation of the reactor, since any increases in the drop times would indicate a possible binding of a control rod in the guide channel.
During the development of the Technical Specifications, filed in 1975, the rod drop times were inadvertantly used as "The time from initiation of a scram condition in the scram circuit until the shim-safety l
rod reaches the rod lower limit switch." (Sec 3.2.c). The Purdue Univer-l sity Reactor has two scram systems. The " fast" scram anticipates either an over power condition or a short period and reduces the magnet current l in the control rod drive mechanisms until the magnet can no lor.ger support the rod and it drops. Since this is an electronic circuit, response times are short and new procedures developed to measure the scram time ("the
shim safety rod reaches the rod lower limit switch") give the same results
! as the rod drop time, within the reproducability of the experiment.
l l
l l
ne%
% Q j West Lafayette, Indiana 47907
, O 8106166400
Mr. James G. Keppler June 8, 1981 Page 2 The " slow scram, initiated by any one of a variety of conditions, interrupts the 110 volt power to the regulated power supplies for the control rod magnets. Thus, the energy stored in the regulated power supply must drain off before the magnet current is reduced below the holding current and the cetrol rods are released. According to measurements made for the first time on June 2,1981, the time delay for the regulated power supplies to reduce their current to the point at which the control rods will drop averages about 60 milliseconds. This, coupled with the rod drop times gave " scram times" between 663 and 629 milliseconds, which exceeds the limits in the Technical Specification.
Hazard Analysis The only accident analysis perfonned for the preliminary Hazards Report that involved the reactor scram time, was a postulated rar reactivity increase by withdrawing all control rods simultaneously with a failure of the period interlocks, control rod interlocks to allow withdrawal of all
, control rods simultaneously, set back trips, and period scrams Assuming a maximum allowable excess reactivity in the reactor of 0.006 ,gg- the reactor period at the time the Limiting Safety System setting N,f o 1.20 kw is reached is 3 seconds. Assuming the high level fast scram is inoperable, and the slow scram would allow the power level to increase for an additional 60 milliseconds, before the shim safety rods would drop. This would correspond to a trip point of 1.22 kw rather than 1.20 kw. Even if the reactor should b placed on a 1 second period (corresponding to a reactivity --
change of 0.006 gg), the maximum power would be 1.27 kw, well below the Safety Limit of 56 kw. Therefore, no additional hazard is considered to i exist.
Action l Since no additional hazard is considered to exist and since the reactor l has operated safely since its startup in 1962 with this condition, it is proposed to continue to operate the reactor with no changes and to request a change in the Technical Specifications to remove the ambiguity between
! measured rod drop times and reactor scram times.
If we may be of any further assistance in evaluating this problem or can supply additional information, please contact me.
Sincerely,
, - i. .
E. R. Stansberry Reactor Supervisor ERS:sab
--_,y--.w__ y , . , , , . - pm%f, , - - . . .,, - . ,,9 * - ~ _ . _ . . _ - -