ML20209G401

AEOD/T510, Xenon-Induced Power Oscillations at Catawba, Technical Review Rept.No Further Action by AEOD or Other NRC Ofcs Necessary

ML20209G401
Person / Time
Site:	Catawba
Issue date:	09/04/1985
From:	Freeman R NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
Shared Package
ML20209G379	List: ML20209G373 ML20209G401
References
TASK-AE, TASK-T510 AEOD-T510, NUDOCS 8509190262
Download: ML20209G401 (3)

Text

.

~

_1]

AE0D TECHNICAL REVIEW REPORT UNIT:

Catawba 1 TR REPORT NO.:

AE00/T510 DOCKET NO.: 50-413 DATE: September 4, 1985 LICENSEE:

Duke Power Company EVALUATOR / CONTACT:

R. Freeman NSSS/AE:

Westinghouse / Duke Power Company

SUBJECT:

XENON INDUCED POWER OSCILLATIONS AT CATAWBA EVENT DATE: March 18, 1985

REFERENCES:

(1)

U.S. Nuclear Regulatory Commission Inspection and Enforcement Inspection Report No. 50-413/85-12, Duke Power Company, Catawba Unit 1, Docket No. 50-413, dated April 26, 1985.

(2)

U.S. Nuclear Regulatory Commission, Regulatory Guide 1.68, Rev. 2, Initial Test Programs for Water-Cooled Nuclear Power Plants, dated August 1978.

SUMMARY

On March 18, 1985, with Catawba Unit 1 at approximately 50% power, during initial startup physics testing, severe core power tilts and oscillations occurred as a result of control rod insertion testing.

Because of unex-pected equipment problems and lack of procedural guidance, a control rod remained inserted in the core for an extended period of time. Upon withdrawal of the control rod to its fully withdrawn position, severe xenon induced power oscillations were experienced which resulted in a reduction in reactor power operation and the delay in the completion of startup physics testing. The corrective actions taken by the licensee were judged to be appropriate. Testing procedures which restrict the maximum allowable time of control rod misalignment could signicantly reduce the likelihood of inducing severe power tilts and oscillations. However, it may no longer be necessary for plants to perform some of the physics tests the licensee was conducting when the power oscillations occurred.

It may be appropriate to consider referencing the results of control rod insertion tests at plants with identical cores in place of the plant-specific tests.

Such referencing would reduce the potential for xenon induced power oscillations, which when combined with a power transient, could result in exceeding local departure from nucleate boiling (DNB) limits.

DISCUSSION Reference 1 describes the circumstances surrounding a xenon induced power oscillation event that occurred on March 18, 1985 at the Catawba Unit I nuclear facility. As part of Catawba's initial startup test program, control rod in-sertion testing was being performed from 50% reactor power in accordance with Regulatory Guide 1.68, Appendix A, Sections 5.F and 5.I (Ref. 2). This testing was designed to demonstrate:

(1) core thennal and nuclear parameters are in agreement with predictions with a single high worth control rod fully inserted during ano following return of the rod to its bank position and, (2) the 1/

This report supports ongoing AE00 and NRC activities and does not represent the nosition or recuirements of the responsible NRC program office.

i.

8509190262 850904 PDR ADOCK

. d _ _ __ 05000413 PM C.

- - =

{

capability for the. facility design to detect a control rod misalignment equal to or less than the maximum allowed by plant technical specifications.

The particular testing conducted by the licensee on March 18, 1985 was to J

measure the maximum flux perturbation expected from a dropped control rod.

This test requires having the reactor at approximately 50% power. A core flux map is initially performed using the incore movable flux detectors.

A high worth peripheral control rod is then fully inserted into the core, and then another flux map is conducted.

During this phase of startup testing, the high flux reactor trip setpoints are reduced from the normal 109% value to 70%.

l During the performance of the test, unexpected equipment problems were encountered causing the second flux map to be delayed. After 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> had elapsed, the plant staff decided to abort the test and return the control rod to its fully withdrawn position. The total time the control rod was inserted into the core was approximately 7.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />. The long period of control rod insertion compared to the xenon half-life of 9.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, induced a severe power tilt and oscillation. When the licensee recognized that the core was undergoing xenon induced power oscillations, core power i

was reduced and the reactor vendor (Westinghouse) was consulted and the situation reviewed by the licensee manacement.

It was determined that no limits were in danger of being exceeded and power operation could continue while attempting to dampen the power oscillations. The worst observed power tilt occurred with a reactor thermal power of 48.5%. Power in core quadrants 1 and 3 was indicated to be 45.5% and 46.8%, respectively, while quadrants 2 and 4 were 73% and 30.5%, respectively. After 5 days, the xenon induced power oscillations dampened out such that physics testing could resume.

ANALYSIS l

Operation with core power tilts is an undesirable condition, but is allowed by plant technical specifications for reactor power levels below 50% because the quadrant power tilt limits are only applicable for reactor power levels i

above 50%. Even though the high flux trip setpoints are reduced during the control rod insertion testing phase of the startup test program, automatic core protection may not be sufficient to prevent localized DNB to occur in i

the unlikely event of a power transient occurring during a peak power tilt.

1 This is because it requires two out of four power range channels to activate t

l the reactor protective system high neutron flux power range trip. DNB could occur during severe power tilts because the power in the core quadrant near a second detector channel would not be expected.to reach the trip setpoint prior.

i i

to the power in the core quadrant near the first tripped detector being eisove

.the local power leve) analyzed for DNB.-

Because the amount of power tilting is dependent on the length of time a control rod is misaligned, limiting the duration of misalignment woulii minimize the amount of power tilting. Testing procedures which address the maximum allowable time of control rod misalignment could significantly

-reduce the likelihood of inducing severe power tilts.and oscillations'. Such i

guidance could help in the preparation for conducting such tests and allow for early determination in aborting the tests when unforeseen ~ problems develop.

~

N

=-

In view of the potential for producing severe power tilts, performance of some physics tests at power may no longer be necessary, particularly if previous tests at reactor facilities of similar design can provide sufficient data to verify the adequacy of the physics analysis. Discussion with members in the Core Performance Branch, Office of Nuclear Reactor Regulation (NRR) have confirmed that the information obtained from the particular testing Catawba was performing on March 18, 1985 yielded little new information on the reactor nuclear physics analysis compared to existing information avail-able at similar reactor facilities. Since existing information available at other reactor facilities with identical cores may be applicable, the testing i

outlined in Regulatory Guide 1.68, Appendix A, Sections 5.F and 5.I may not have had to be performed.

Thus, allowing licensees to reference the results of tests conducted at plants with identical cores could be a viable option in lieu of performing the plant-specific tests. Such an option would reduce the potential for xenon induced power oscillations which would enhance reactor

]

safety as well as shorten the duration of low power physics testing.

CONCLUSION Because of unexpected equipment problems and lack of procedural guidance,.

a control rod was left inserted in the core for an extended period of time. The extended period of the control rod insertion induced a severe l

power tilt and power oscillations. The corrective actions taken by the licensee were judged to be appropriate. However, the need to perform some of the physics tests outlined in Regulatory Guide 1.68, Rev. 2, Appendix A, Sections 5.F and 5.I appear to be no longer necessary because the results of tests conducted at plants with identical cores could be referenced in place of performing the actual testing. Minimizing the testing would reduce the potential for xenon induced power oscillations and, thereby, l

minimize the likelihood of exceeding DNB limits.

l Discussions held with NRR representatives (Core Performance Branch) indicate that NRR would be willing to consider including in the next revision of Regulatory Guide 1.68 the option to allow referencing of test results conducted at plants with identical cores in lieu of performing the plant-specific tests. Allowing licensees this option should reduce the amount of testing, and, thereby, reduce the potential for xenon induced power oscilla-tions to occur.

No further action is deemed necessary by AE00 or any other NRC office.

i l

l h.

I bb +

,g

, - _, _.,