ML19340A836
| ML19340A836 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 04/19/1977 |
| From: | Bolger R COMMONWEALTH EDISON CO. |
| To: | James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| Shared Package | |
| ML19340A835 | List: |
| References | |
| NUDOCS 8009040659 | |
| Download: ML19340A836 (4) | |
Text
~
C:mmonwssith Edis:n one F.rst Nabonal Pt.va. Chicago, lihnois q
Address Reply to: Post Othee Box 767
- Q Chicago, finnois 60690 p
April 19, 1977 4
Mr. James G. Keppler, Director Directorate of Inspection and Enfor.:ement - Region III U,J. Nuclear Regulatory Commission 799 Roosevelt Road Glen Ellyn, IL 60137
Subject:
Dresden Station Units 1, 2, and 3 Response to IE Inspection Report Nos. 50-10/77-04, 50-237/77-04, and 50-249/77-04 NRC Docket Nos. 50-10/237/249 Reference (a):
C. Fiorelli letter and Inspection Report to Byron Lee, dated March 29, 1977.
Dear Mr. Keppler:
Attached to this letter is Commonwealth Edison's response to the two items of noncompliance noted under Enforce-ment Items in the Summary of Findings section of Referenco (a).
Please direct any additional questions concerning this matter to this office.
Very truly yours, ll d
R. L. Bolger Assistant Vice President Attachmenc i
5 800904.of5.7 APR 211977
Comm:nw :lth Edis:n NRC Docket Nos. 50-10 50-237
~
Attachment 50-249
~
s Infraction Unit 2:
Contrary to Technical Specification 6.2.A.5., a detailed procedure was not prepared, approved, and adhered to for calibra-tion of LPRM's which had experienced seal failures.
Discussion:
Failure of the cermnic seal separating an LPRM chamtar from its cable has occurred at Dresden and at other plants employing this type of General Electric neutron flux detector.
The mechanism involved is well understood, and results in an increase in the sensitivity of the detector.
Under no circumstances will a seal failure cause the output
_of an LPRM to not increase when the reactor power in the vicinity of the detector increases.
For rapid power increases (for which the APRM scrmn is the primary limitation), the behavior of a seal-failed LPRM is indistinguishable from that of a normal LPRM.
The primary function of an LPRM, namely, serving as one of about twenty inputs to an APRM channel, is thus not adversely affected by the seal failure.
The increased nonlinearity exhibited by seal-failed LPRM's is an effect that is realized only during longer term power changes.
This nonlinearity results in the need to calibrate the APRM system on a slightly greater frequency than would otherw,e be necessary, 3
in order to maintain the effective scram settings a. or below those specified in the license.
Dresden checks and calibrates the APRM system at a greater frequency than that specified in the license, which has proven entirely adequate to climinate any calibration problems due to seal-failure-induced nonlinearity on the APRM system.
All operable LPRM's are calibrated on an approximately monthly basis, using a detailed approved procedure, DTS 8232.
This procedure, however, does not accommodate the calibration of an individual LPRM independently of a whole core calibration, an operation which has becomo desirable with increasing numbers of seal failures.
Treatment of scal-failed LPRM's has been a topic of concern at Dresden.
Correspondence from the LPRM vendor (General Electric),
including the original and first revision of SIL 146, outlined a recommended treatment program.
Dresden chose not to implement the 4
l
)
NRC Docket nos. 50-10 Comm:nwacith Edis:n
~50-237
., 50-249
('
GE program and to alternatively implement a modified program, because it was determined that the GE program would result in greatly increased wear and detector burnup for the TIP system.
Dresden's program was implemented by the Nuclear Engineering group, without detailed written and approved procedures, using experience and engineering judgment.
The underlying rule in Dresden's
-program has been that an LPRM whose sensitivity was discernably increasing should be bypassed until the sensitivity becomes stable.
Following stabilization, the reading of a seal-failed LPRM was then appropriately adjusted through use of the associated TIP reading.
Corrective Action:
Dresden Station management committed to write and
-implement procedures for identification of LPRM seal failures, returning of stabilized seal-failed LPRM's to service, and cali-bration of individual seal-failed LPRM's.
Although the investi-gation is still in progress, it was agreed that sufficient experience in the treatment of seal failures had been gained to write the above LPRM procedures.
The procedures were to be in effect within one month.
Corrective Action to Avoid Recurrence:
Procedure DTP 8354, a comprehensive guideline for the identification of specific LPRM abnormalities (including seal failure) and proper immediate corrective actions, and procedure DTP 8353, a detailed description of criteria for LPRn recovery from any outage and proper method for individual LPRM calibration, were written, approved, and implemented by April 5,1977.
In addition, a system of handling instances of LPRM abnormal behavior, in a smooth fashion consistent with good record-keeping, was established between the Nuclear Engineering group, the Instrument Maintenance Department, and the Operating Department.
The system integrates utilization of these and other new or revised procedures, which in turn, provide references to both existing and newly-created record files.
Primary responsibility for the assurance of proper LPRM detector performance and calibration was shifted to the Nuclear Engineering group.
Date of Full Compliance:
Full compliance was achieved on April 5, 1977.