ML20210F342

Forwards,For Preliminary Informal Review,Ga Technologies,Inc Re Fuel Element Temps in Triga Reactors During Operation & Loca.Proposed Words for Inclusion in Final Draft Tech Specs Sections 2 & 3 Also Encl.Comments Requested

ML20210F342
Person / Time
Site:	Reed College
Issue date:	02/05/1987
From:	Joseph E Pollock REED COLLEGE, PORTLAND, OR
To:	Dosa J Office of Nuclear Reactor Regulation
References
NUDOCS 8702110057
Download: ML20210F342 (6)

Text

_ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _

54-268 REED C O L L E O'E 'Pordand, Oregon 97202 CEACTOR FACILITY February 5,1987 Mr. John Dosa Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D.C.

Dear Mr. Dosa:

I tried unsuccessfully to reach you earlier this week as I had indicated I would do. Here is a report on the status of the Technical Specifications.

l At our meeting on Monday, February 2, the Reed Reactor Facility Safety and Operations Subcommittee on the Technical Specificatons reviewed the last of the unresolved questions which you raised in your informal comments on the Draft Technical Specifications. We believe that we have now l resolved all but about 10 of those comments. These have been divided ,

j between the subcommittee members for resolution. A couple are involving l discussions with General Atomic and volunteer experts from Trojan who l

are assisting us. I anticipate submission of the final draft to the full committees on February 26 with their approval at their next meeting which should be scheduled about one month from then.

!' I am enclosing copies of two documents for your preliminary informal i

review. The first is a letter from GA Technologies relating to fuel

element temperatures in TRIGA Reactors during normal operation and loss of coolant accidents; the second is a set of proposed words for inclusion in our Technical Specifications (Sections 2 and 3) on this topic. These were presented to and are under review by the internal subcommittee and will be included in the final draft in roughly this form unless you have l suggestions before then. Informal comments, would be appreciated.

Sincerely,

,[_' /~

J. Michael Pollock cc. J. Frewing

M. Cronyn L. Ruby, with attachments 0 O

! s702110057 870205 I PDR ADOCK 05o00288 P PDR

,b GATechnologies GA Technologies Inc.

Po. Box 85608 SAN OiEGo, CAUFCANIA 92138 (619) 455-3000 TREX: November 19, 1986 695065 GA TECH SOG Mr. John Frewing Portland GE 121 Southwest Salmon St.

Portland, Oregon 97204

Dear Mr. Frewing:

Subsequent to our telephone conversation last week I was able to find a file copy of the GA contribution to the Reed College SAR, which was written in early 1967 I have enclosed a copy of Appendix B of that document, which is the analysis for the loss of coolant accident. For your additional information, I have also enclosed a copy of the introduction, fuel temperature and accident analysis sections of a much more recent SAR for a system similar to that at Reed College. The main difference is that for about the last 15 years, the standard TRIGA fuel element has been 15 inches of U-ZrH 1 .6 clad with stainless steel vs the original TRIGA fuel at Reed College which is 14 inches of U-ZrH1,o clad ,

with aluminum. The more current SAR will give you an overview of our present analyses and results for accidents generally included in TRIGA SAR contributions from GA. In a quick review and comparison of the safety sections of each document, I note that the much greater data base of information available about the operational characteristics of TRIGA fuel and greatly improved analysis tools, have apparently combined to produce significantly different results in the newer SAR compared to the original one. The overall conclusion wouldn't change, in that fuel clads would not rupture resulting frem the loss-of-coolant accident. However, it appears that significantly higher temperatures exist during normal operation and as a result of a loss-of-coolant accident than are shown in Appendix B of the original analysis.

From Appendix B (text and Fig. 1) it appears that the maximum operating fuel temperature at 250 kW is calculated to be just over 1000C and after loss of coolant the maximum fuel temperature rises from this temperature to about 15000. These temperatures are markedly different from those resulting from the more recent analysis where a maximum fuel temperature at 250 kW is shown to be about 2700C and the maximum temperature after loss of coolant is also about 2700C. In this case the fuel temperature was at a shutdown temperature of 2000 when the air cooling began. Note on page 8-20, it is stated that the maximum fuel temperature following a loss-of-coolant is not increased by much if the fuel starts from 10955 JOHN JAY HOPKINS OR . CAN DIEGo. CAurORNIA 92121

\

\

., u Mr. John Frewing Portland GE November 19, 1986 operating temperature vs a shutdown temperature. This statement is addressed more at the higher power levels (1 to 2 MW) where temperatures after loss-of-coolant can be about twice the normal operating temperatures. The initially stored energy is relatively small compared ,

to the energy generation over a long period to maximum temperature after a loss-of-coolant. The influence of the initial temperature (stored energy) is greater at lower power levels.

It appears that the reason for at least a major part of the difference in temperature between the old and new analysis is the absence of accounting for the effect of heat transfer across the fuel-clad gap in the old

, analysis. That was a time when a great deal of information about the operating characteristics of TRIGA fuel was just being learned and #

apparently no account of this effect was incorporated in the original analysis.

As I understand the NRC question, they are asking how or why the temperature in an accident condition (loss-of-coolant) can be less than for normal operation. The answer is that, while the cooling medium has changed from water to air-providing much less cooling ability, the power being generated has dropped drastically (initially to only 55 of the operating power and steadily decreasing from that point) and thus the heat removal requirements are drastically reduced. This is true because the reactor is shutdown due to the large loss in reactivity (regardless of whether the rods have dropped in or not) when the water is drained from the core. Thus, the operating core at 250 kW produces maximum fuel temperatures of about 27000, but when the reactor is shutdown and air . ,

cooled, the combination of much lower power generation and air vs water cooling produces a maximum fuel temperature also of about 2700C. This maximum temperature af ter loss-of-coolant is, however, dependent upon how long it takes to lose the core water af ter the the reactor is shutdown, as shown on page 8-13 of the more recent SAR. In any event, it appears that the 250 kW power level is about the dividing line between loss-of-i coolant temperatures being greater or smaller than normal operating temperature when the loss-of-coolant temperature rise starts from a shutdown value of about 200C.

l I hope this information helps in forming your answers for the NRC l questions. If you have further questions, don't hesitate to call.

Very truly yours, i Gordon B. West Project Manager i

TRIGA Reactor Division GBW/1vg i Enclosures Excerpts from Doc. Ell 7-478 & GA-7860 l

0,, Y k $C h W{%~f&

wjf*

4.

'2.1 Safety Limit

. . -a Applicability:

This specification applies.to reactor core parameters which constitute a safety limit which in turn avoid incurring any significant off-site impacts as a result of an accident at the Reed Reactor Facility.

Objective:

The objective is to define conditions that can be permitted with confidence that no damage to fuel element and/or cladding will result.

Specification:

The temperature in a water-cooled Standard TRIGA fuel element shall not exceed 1150*C under any operating condition. For the Reed Reactor Facility, this is equivalent to saying that-the maximum reactor power shall not exceed 250 kW.

Basis:

The safety limit for the Standard TRIGA fuel element is based on calculations and experimental evidence. The results indi-cate that the stress in the cladding due to hydrogen pressure '

from the dissociation of zirconium hydride will remain below the ultimate stress provided that the temperature of the fuel does not exceed 1150*C and the cladding does not exceed 500*C.

Appendix E of the SAR documents that 225*C is the approximate maximum fuel temperature for continuous operation at 250 kW.

This temperature is lower than the condition stated earlier and assures a cladding temperature of less than 500*C under all design basis accident conditions (step insertion of all f available excess reactivity or-instantaneous loss of cooling l

i accident). Section 2.1 of the SAR states that under an instan-taneous loss of cooling accident condition, after operation at 250 kW for an infinite time, the maximum fuel temperature l

. reached during the transient is 150*C. The equilibrium pres-

[ sure resulting from fission gases, trapped air, and hydrogen at p

150*C is less than 30 psi. Section 7.1 of the SAR documents l

that a maximum measured fuel temperature less than 500*C results from a 2.25 percent Ak/k ($3.00) step reactivity l

insertion.

n Other thermal and hydraulic calculations indicate that Standard l TRIGA fuel elements may be safely operated at power levels in excess of 1,500 kW with natural convection cooling. Details on the performance of TRIGA fuel are provided, as stated previ-ously, in the SAR and in papers available as " Fuel Elements for l Pulsed TRIGA Research Reactors", Simnad, et al, Nuclear l Technology, H , 31 (January 1976), and "The U-ZrH g Alloy:

Its Properties and Use in TRIGA Fuel", GA Project No. 4314, i

i ,

t

s k

Report E-117-833, General Atomic Company, PO Box 81608, San Diego, California 92138, 1980.

2.2 Limiting Safety System Setting Applicability:

This specification applies to the settings that prevent the safety limit from being reached.

Objective:

The objective is to prevent the safety limit from being reached.

Specification:

Linear power and percent power channels shall initiate a scram at 110 percent of 250 kW (275 kW).

Basis:

The basis for 275 kW is that this safety system setting will prevent the limit of Section 2.1 being surpassed. The 275 kW setting is well below the 1500 kW safe natural connection cool-ing limit discussed in the Basis for Section 2.1.

3.0 LIMITING CONDITION FOR OPERATION .

3,1 Reactor Core parameters 3.1.1 Power Level Applicability:

This specification applies to the energy generated in the reactor during normal operation.

Objective:

The objective is to ensure that the fuel temperature safety limit and associated power level limits will not be exceeded during normal operation.

Specification:

The reactor power level shall not deliberately be raised above 250 kW as measured by the linear or percent power channels under any normal conditions of operation, including calori-metric calibration of instrumentation.

'e ,

d .

'A Basis:

Maintaining indicated reactor power levels below 250 kW will ensure the power level associated with the fuel temperature safety limit will not be exceeded. For the purpose of testing the 110 percent full power safety scrams, calorimetric calibra-tion can be performed at a power level of 225 kW and then 110 percent full power safety scram setpoints can be set electronically.

3.1.2

(

I i

l 4

6-2650B l

l

- - - . - - . , . - - . - . . - . - , . - . . . . - - . - . - . _ ---. . - - - - . . - , - - - --