ML19323F288
| ML19323F288 | |
| Person / Time | |
|---|---|
| Site: | 05000054 |
| Issue date: | 05/15/1980 |
| From: | Ruzicka W UNION CARBIDE CORP. |
| To: | |
| References | |
| NUDOCS 8005280737 | |
| Download: ML19323F288 (4) | |
Text
.
f, a0\
pf/'-
UNION CARBIDE CORPORATION O~s MEDICAL PRODUCTS DIVISION P.O. BOX 324. TUXEDO, NEW YO A K 10987 TEL EPHONE: 914-351 2131 May 15, 1980 Director, Region I U. S. Nuclear Regulatory Commission 631 Park Aven ue King of Prussia, PA 19406
Dear Sir:
Mr. D. Caphton, of your office, was notified via telephone on May 5, 1980 of a separated control rod assembly in the Union Carbide Nuclear Reactor, License Number R-81. A confirming telecopy was also sent to your office on May 5, 1980. Tech-nical Specification 6.5.3.a.3.b requires this report to the Commission following any operation in violation of any Tech-nical Specification Limiting Condition For Operation. Sec-tion 3.6.C.2 of the Technical Specification states that all control rods shall be kept within + 10% of their mean position whan reactor power exceeds 500 KW. - Following the control rod separation the poison section remained in the core while the reactor power exceeded 500 KW.
This failure did not compromise the reactor shutdown margin as this type of separation will cause full rod insertion and will not cause a withdrawn rod to stick out of the core.
The~resulting flux tilt was insufficient to cause boiling and is not a cause for concern of reactor integrity. The rod failure is isolated to a single rod with all remaining rods of similar design inspected and showing no signs of deficien- i cies. Routine inspections of our present rods will continue with special attention given toward inspecting for similar type rod failures.
Details of Occurrence and Analysis In accordance with our routine operating cycle the reactor was shut down by a scheduled reactor scram and refueled on 1 May 1, 1980. At approximately 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br /> of the same day the reactor was started up. It is now assumed that this startup was done with one control rod remaining in the core. Prior to the startup all required prestart checks were completed.
Coarse and fine rod position indicators showed normal with-drawal and all rod seat and off magnet lights indicated nor-mal pick-up and attachment of the rods. The rod seat lights l indicated that the control rod pistons and end pieces and 0}k D '
presumably the control rods were withdrawing normally from t 8005280 7 p
Dircctor, Ragion I - Pega 2 -
U. S. Nuclear Regulatory Commission May 15, 1980 the core. The off magnet lights indicated that the control rod pistons remained attached to the scram electromagnet.
Prior to attaining 10% of full power reactivity checks were performed on the rods as per our startup operating procedure.
The reactor was then increased in power and via reactor heat balance taken to a 5 MW power level.
On May 3, 1980, at approximately 1400 hours0.0162 days <br />0.389 hours <br />0.00231 weeks <br />5.327e-4 months <br />, it was deter-mined by operating personnel that one rod was not attached to its rod drive and remained fully in the core. The reactor was then innediately shutdown. A determination was made that control rod number one had separated between the control rod end piece and its poison section. An investigation of the
- May 1st startup and the following 44 hours5.092593e-4 days <br />0.0122 hours <br />7.275132e-5 weeks <br />1.6742e-5 months <br /> of reactor opera-tion was conducted. All instrument recorder charts were ex-amined and it was determined that total power by heat balance did not exceed 5 megawatts, flow remained greater than 2200 gpm, and bulk pool temperature averaged 100.13aF. for the 44 hours5.092593e-4 days <br />0.0122 hours <br />7.275132e-5 weeks <br />1.6742e-5 months <br /> of operation.
The nuclear instrumentation recorder charts were analyzed and the time period of the startup reactivity checks determined.
The reactivity checks were conducted over a pericd of approx-imately eight minutes. Reactor power during these checks de-creased from 2.2 to 1.6% by Log N indication and Log Count i Rate indication decreased from 31 to 21 cps. The reactivity checks were performed during the May 1st startup by placing the reactor in automatic control and withdrawing each control 1 rod to a new higher gang height. While withdrawing each shim I rod the reg rod position was monitored. In automatic control -
the reg rod should ccmpensate for any withdrawal of a shim rod. During these reactivity checks any reg rod inward move-ment was interpreted as a compensation for the withdrawal of a rod. Other parameters were also changing during these re-activity checks which had an affect on the reg rod position.
- Xenon poison was decaying which drives the reg rod inward and the rod gang height was increasing with each rod checked.
Nuclear instrumentation rod shadowing was decreasing with the higher gang height, causing a higher indicated power level, which also drives the' reg rod inward when in automatic.
These two circumstances could have caused the indication er-roneously interpreted as reg rod compensation to shim rod withdrawal. The startup reactivity checks are being revised to place the reactor control in automatic while inserting and then withdrawing each shim rod a specified amount. By in-serting and then withdrawing each rod and monitoring for an opposite reg rod response in each direction the misinterpreta-tion of Xenon or shadowing effects should be eliminated. j
Dircctor, R:gion I - Pago 3 -
U. S. Nuclear Regulatory Commission May 15, 1980 Technical Specification 3. 6.C. 2 requiring all rods to be bal-anced when above 10% power is meant to minimize flux tilts
- ~
that could cause concentrations or shifts in core power' dis-
.. ~
tribution. Operation with one rod remaining in the core did cause larger than normal flux concentrations but these were insufficient to cause boiling in the core. Making the sim-plified conservative assumption that one sixth of the core was shutdown with the installed rod the maximum flux concen-tration achieved in the remaining regions of the core would be 120% power. This assumption is quite conservative because all the fuel surrounding the inserted rod would not be in a zero flux, zero power condition. Supplement No. 2 to the UCNR Final Hazard Summary Report uses conservative correla-tions to derive power levels for incipient boiling. At a bulk pool temperature of 100 0F. and reactor flow at 2100 gpm (our actural flow rate was 2200 gpm) 7.05 MW or 141% power is required to produce incipient boiling. Experimental compari-sons at ORNL have shown these numbers to be conservative.
The UCNR Technical Specifications require a Limiting Safety System Setting of 7.3 MW at 100 0 F. pool temperature. Limit-ing Safety System Settings are settings chosen to initiate dutomati'J protective action to insure safety limits are not exceeded. This power level of 7.3 MW or 146% power is well above the maximum 120% power concentration of the unrodded portions of the core.
The failed silver indium cadmium control rod and its mating end piece and piston underwent hot cell examination. This inspection showed that the poison section of the rod had failed in shear above the four flat head stake screws meant to hold the poison section to the end piece. The figure be-low illustrates the control rod and shows the area of metal failure. '
r" {
~
// . .
j jj L-
, _a G~
/( .:
I l
o . r *
-- I a
,'~
area of failure C! Y
{
m
. rod -
end piece 5.6" :Lpiston 3.1"+ -
. l .
Director, Region I - Page 4 -
U. S. Nuclear Regulatory Commission May 15, 1980 UCNR purchased five silver indium cadmium control rods from AMF Atomics of York Pennsylvania in late 1969. AMF Atanics no longer manufactures control rods. UCNR records are un-
- clear but purchase order dates strongly suggest that the j failed rod was manufactured by AMF. Only one of the original AMF rods was still in use at UCNR this year. All remaining silver indium cadmium rods were manufactured at a later date
, than the AMF rods and produced by a different supplier. Our records show the failed rod to be first installed in the re-actor in June of 1976.
The UCNR is now operating with four silver indium cadmium rods. These rods have undergone a hot cell inspection where the control rod, end piece, and piston section were visually inspected with the hot cell magnifying periscope. No deter-ioration of the silver indium cadmium alloy was indicated above the stake screws. All silver indium cadmium rods will be inspected at least annually to insure design integrity.
Special attention will be given toward inspecting these rods for similar end shear type failures.
This failure did not violate the required shutdown margin since the separated portion of the rod was confined to the core. Because the capability of the reactor to shut down and i remain safely ahutdown was not compromised, we have not re-4 ported this defect under Title 10 Code of Federal Regulations part 21.
Corrective Action -
All present silver indium cadmium rods have been inspected and show no abnormalities.
Annual inspections of all rods will be done to insure rod de-sign integrity.
Startup reactivity check procedures are being revised to pre-vent rod withdrawal misinterpretations.
Sincerely, (MM. 929uL~
WGR:ltm William G. Ruzicka Reactor Project Engineer cc: Mr. M. H. Voth Manager Nuclear Operations
__ _ , _