ML19345E414
| ML19345E414 | |
| Person / Time | |
|---|---|
| Site: | Big Rock Point File:Consumers Energy icon.png |
| Issue date: | 04/14/1966 |
| From: | Boyd R US ATOMIC ENERGY COMMISSION (AEC) |
| To: | |
| Shared Package | |
| ML19345E413 | List: |
| References | |
| NUDOCS 8101160420 | |
| Download: ML19345E414 (8) | |
Text
~..
(
SAFETY EVALUATION BY THE RESEARCH AND POWER REACTOR SAFETY BRANCH DIVISION OF REACTOR LICENSING CONSUMERS POWER COMPANY PROPOSED CRANGE NO. 8 DOCKET No. 50-155 Introduction The Consumers Power Company of Michigan has proposed changes, by letters dated December 23, 1965 March 15, 1966 and supplement letter dated March 17, 1966, to the Technical Specifications of License DPR-6, Docket No. 50-155, issued to Consumers Power Company on May 1, 1964 for the Big Rock Point Nuclear Plant. These changes would permit (1) zircaloy clad reload fuel bundles containing cobalt targets (four corner rods of each reload fuel assembly removed to permit insertion of cobalt targets) to be substituted for the originally installed stainless steel clad fuel bundles, (2) a special test bundle with Zr-Cr alloy fuel cladding to be used as a reload bundle, and (3) removal of one control rod drive from the reactor vessel when the reactor is in the shutdown condition and the mode selector switch is locked in the " shutdown" position. The request for these changes has been designated Proposed Change No. 8.
iscussion Cobalt TarRets The cobalt irradiation program involves placing doubly-encapsulated cobalt target assemblies in the corner regions of the reload fuel bundles. As currently visualized by Consumers, four cobalt assemblies would be substituted for corner fuel rods in selected reload fuel bundles.
Initially, during the April 1966 refueling shutdown, 14 fuel bun (tes containing a total of 56 cobalt assemblies would be inserted into the Big Rock Poi t nuclear reactor. At subsequent refuelings, additional reload fuel bundles containing cobalt assemblies will be added so that of the full core loading of 84 fuel bundles, about 60 fuel bundles will ultimately contain cobalt assemblies.
These cobalt assemblies would remain in the reactor core for a period of from 2 to 3 years, af ter which it is expected that they will have attained an average specific activity of about 40 curies per gram. The equilibrium cobalt-60 production cycle will consist of removal and replacement of from one fourth to one third of the fuel bundles containing cobalt assemblies at each normal refueling.
The cobalt targets will be fabricated from 0.046-inch or 0.060-inch diameter, nickel-plated cobalt metal wire arranged as a helical coil within a 304L stainless steel capsule having an 0.011-inch thick wall. Targets will be fabricated in two lengths: The 0.060-inch diameter cobalt wire will be used for targets to be inserted into a 12-inch long capsule. The 0.046-inch diameter cobalt wire will be used for targets to be inserted into a 10-inch long capsule. Details of the cobalt target and target capsule design were shown in Consumers' March 15, 1966 proposal.
It is our understanding that before assembling the cobalt helical coil into the capsules, the targets and capsules will be quality-control tested for dimensional flo / /(o 0'/2 o
accuracy, cleanliness, weight uniformity and uniformity of the nickel plating on the targets. Fabrication of a target capsule will consist of inserting cobalt coils into a stainless steel capsule and welding the end plug. Welding will be done in a helium atmosphere. The end dimensions of the capsule will be certified by mounting each capsule (af ter welding) in a lathe chuck and machining if necessary. After welding and dimensional checking, the capsules will be 100-percent helium leak-tested.
Leaking capsules will either be repaired or discarded.
Each cobalt assembly will consist of five 12-inch long capsules and one 10-inch long capsule in a standard Zircaloy-2 corner fuel tube. Details of the cobalt assembly were also presented in the March 15, 1966 Consumers' proposal. Tolerances of the fuel pellet which capsules will be at least as stringent as the tolerances of the UO2 would normally be used in the fuel tube although in certain instances, tolerances may be more stringent in order to guarantee proper fit in the Zircaloy-2 tube. The normal fuel spacers and springs will be used to position the capsules within the Zircaloy-2 tube. To distinguish the corner cobalt-bearing Zircaloy-2 tubes from the corner fuel-bearing tubes, the top end-plug of each cobalt assembly will be made about one-quarter of an inch longer. However, even in the dimensionally worst case, there will still be at ; east 1/8-inch clearance between the top end-plug on the cobalt assembly and the top hold-down bar on the fuel bundle.
The 1/4-inch additional length of the cobalt assemblies also will serve as an aid in locating cobalt assemblies in a fuel bundle.
It is Consumers' intention to remove the cobalt assemblies from a fuel bundle after that fuel bundle has been irradiated for about two-thirds of its expected lifetime burnup. Fuel rods with selected reactivities will then be inserted in the fuel bundles wherever cobalt assemblies have been removed to help recover part of the reactivity lost when the fuel rods were removed originally. Physics calculations according to Consumers show no flux peaking problems even when the original fresh fuel rods are reinserted in the corner locations.
All subsequent batches of fuel containing cobalt assemblies (af ter the first 14-fuel bundle loading with cobalt assemblies in April 1966) will be designed expressly to compensate for the reactivity loss resulting from the addition of cobalt assemblies.
Reload, Fuel Bundles For reference purposes the design features of the reload fuel bundles have been described in the December 23, 1965 Consumers proposal and the principal calculated nuclear characteristics with and without cobalt targets in the reload fuel bundles are documented in the March 15, 1966 Consumers' proposal.
The principal design features of the reload fuel are very similar to the Zircaloy-2 clad research and development fuel now operating in the reactor. Each bundle utilizes two enrichments as well as reduced-size corner rods to minimize local power pe akin r,.
The higher enrichment fuel rods are located in the center of the bundle. Fiv<
wire-and-spring type spacers located along the length of the bundle hold the fuct rods in position and serve to minimize deflection and vibration of the fuel rods.
The reload bundles described in the Consumers' proposals and shown in Figure 5.3 will utilize the same hardware designed to permit removal of any or all fuel rods as the present Phase I and Phase II developmental bundles, except for the spacers.
Eight Phase I bundles utilizing this hardware have been in the reactor since April 1963, and have exposures of approximately 4000 Mwd /T. Fifteen Phase II bundles utilizing this same hardware have been in the reactor since May 1964, and have exposures of approximately 2800 Mwd /T.
The spacers are of the wire-and-spring type design currently being used by General Electric for its commercial fuel. The springs provide the necessary lateral pressure to minimize f retting wear of the zircaloy cladding.
- he dual enrichment utilized in these bundles is expected, according to Consumers, to provide a significant improvement in core peaking factors at a slight enrichment penalty. The effect upon other nuclear and thermal-hydraulic factors is not signif-icant.
The end-plugs on the fuel rods are shaped differently for each of the two enrichments to enhance identification during assembly and a separate quality control inspection vill confirm the fuel bundle loading pattern by means of easily visible end-plug markings. Further procedural controls are imposed during fabrication to route one type.of fuel rod to the assembly area for each loading step. Finally, the reload fuel as part of the normal quality control tests, is verified as to uniformity of nuclear properties by critical tests prior to shipment.
Zr-Cr Alloy Test Bundle The use of annealed Zr + 1.15 w/o Cr alloy, annealed special Zr-2, and cold worked standard Zr-2 as fuel cladding in one of the reload fuel bundles has been previously evaluated by Consumers and the Staff. There is no new information which would ef fect our conclusion as stated in Change No. 5.
Without further consideration the Staff agrees with Consumers that the proposed new Section 5.1.7 does not present significant hazards considerations not described or implicit in the hazards summary report and there is reasonable assurance that the health and safety of the public will not be endangered.
Control Rod Drive Removal The change to Section 7.5.7 is intended to clarify the restrictions on removing a control rod drive from the reactor vessel while the reactor vessel contains fuel as it may be desirabic on occasion to remove a drive for inspection or maintenance work.
To accomplish this the associated control rod must be fully withdrawn, after the primary coolant system has been depressurized and cooled below 125'F, to provide a water seal. The core shutdown margin of 0.3% likeff/kef f with the strongest rod out
_-_-__.- - --.. - - - ~
e i
of the' core will be met prior to control. rod removal. The requirement for one. control rod drive pump to be operating during removal, reinsertion, and during the time the control rod drive is outside the vessel provides added assurance that the 0.3%
dbKeff shutdown margin will be maintained. On this basis,'the staff considers theprofg,glchangetobeacceptablefromasafetystandpoint.
ose Safety Evaluation The precise loading pattern for the first partial core reloading is dependent on the results of core fuel examinations af ter reactor shutdown.' Because of this, the loading pattern will not be set until some time af ter that time, and detailed core calculations
- of heat flux, peaking factors, rod worth, etc. will not be available until then. For the same reason, the pattern of the second and third reload fuel batches will not be finalized until as late as possible in order to take advantage of the latest knowledge of fuel performance of the original core stainless steel clad fuel and the various developmental fuels.
A core typical of the next reloading and a core typical of the equilibrium "zircaloy" core yet several years away have been used by Consumers to illustrate the type of l
calculations which will be performed later, but prior to startup, when specific core i
j.
loadings have been determined..The data from these representative calculations were i
reported in Consumers's supplemental report March 17, 1966. The results of the sample calculations are witnin the tresent operating limits of the Technical Specifications.
With the expected ultimate loading of about 60 bundles containing cobalt assemblies 4
i (about 240 corner rods), the average core heat flux will increase by about 2 percent in order to generate the same power in the absence of the 240 fuel rods. This is compensated for, according to Consumers, by' the fact that the presence of cobalt assem-j blies in the corner locations depresses the local peaking factor, thus making the maximum heat flux in a feel bundle containing cobalt assemblies approximately equivalent to i
_the maximum heat flux in a normal reload fuel bundle. Depending on fuel bundle placement, however, the few fuel bundles without cobalt assemblies may encounter peak heat fluxes some 2 percent greater than otherwise. Since the normal overpower heat flux is approxi-2 mately 400,000 BTg/hr-ft for the 84-fuel bundle core, neither the license heat flux of 530,000 BTU /hr-ft nor the center melting limits are approached with this small potential increase in heat flux.
I The critical heat flux ratio will also be affected somewhat. Since power, flow and 4
~
-consequently the quality will be essentially unaffected, the critical heat flux will be i
unchanged from the no-cobalt situation. The critical heat flux ratio, therefore, will be decreased' proportionately to the indicated maximum heat flux increase.
i '
'The above applies.specifically to the ultimate loading of 60 fuel bundles containing cobalt assemblies. The thermal-hydraulics effects of any loading less than that (the ultimate 60 fuel bundles) are smaller than those indicated in the foregoing discussion.
{
' Since the fuel bundle hardware has been proven adequate by test bundles presently
. installed in the Big Rock Point nuclear core, and with the understanding that calculations r
, will be made prior to startup to assure that any mixture of initial SS clad fuel, R&D and reload fuel is within the present limits of heat flux, center fuel melting, fuel temperature, and critical heat flux ratio, the staff has concluded that there are no safety problems related to normal operation of the reactor with the new fuel loadings.
Consumers has investigated and analyzed the various methods of accidentally changing the configuration or location of the cobalt assemblies. The close clearances between turns of the cobalt coil, coupled with the spring action of the coil, will hold the target in position within the capsule, it concludes, and since the cobalt coils are fully annealed, it is unlikely that the coils would lose their spring action under reactor operating conditions.
In addition, since each cobalt assembly contains six independent cobalt capsules, it is extremely unlikely that the capsules could apprect-ably change their axial position in the core. Consumers further notes that the capsules have been autoclaved at 1500 psi and have demonstrated they can maintain their integrity outside of the Zircaloy-2 outer rod in a simulated reactor environment.
In general, all tolerances on the cobalt capsules are within the tolerances of the UO2 pellets used in normal fuel rods. Consumers states that calculations show that under reactor operating conditions of temperature, pressure and radiation flux, no problems are expected to exist from stress levels caused by the interaction of the stainless steel target capsule and the Zircaloy-2 outer tube.
Based on these considerations, the Staff concurs with Consumers that the performance of the cobalt bearing tubes should be as good, if not better, than the fuel bearing tubes.
Consumers has also calculated that if a high enrichment rod were to occupy a peak peripheral region in a peak bundle, it would experience a 43 percent increase in maximum heat flux. At licensed rated conditions, the maximum critical heat flux ratio (MCHFR) would drop 43 percent for the misplaced rod, which for the calculated 1.22 overpower condition would still give a value of 1.05 for the MCHFR.
Thus, burnout of the rod wouldnotbeexpegted. At conditions of the maximum licensed overpower heat flux of
$30,000 Btu /hr-ft (which is near the calculated threshold of fuel center melting),
this hot rod, being 43 percent hotter, would probably undergo some central melting.
This center melting could possibly result in fuel and clad swelling with potential cladding rupture. Consumers points out, however, that with the 84-bundle core, the present overpower peak heat flux of approximately 400,000 Btu /hr-ft2 is some 40 percent less than this licensed 530,000 Btu /hr-ft2 Hence, with any practical 84-bundle core configuration presently conceivable, including that with reload fuci, it is improbable that central melting would actually occur in the misplaced rod even at overpower.
In both the McHFR and central melting situations, the potential consequence of the rod misplacement is clad rupture. The ef fects of clad rupture are already Consumers feels, a matter of considerable experience in the case of stainless steel clad fuel, and are fully within the safe handling practices incorporated in regular plant operational procedures. Although inconvenience may result, no damage to core and internals is likely should the improbable fuel rupture actually occur due to rod misplacement.
In view of the very low probability of such an assembly loading error, wherein a high enrichment fuel rod occupies a peak power peripheral location in the bundle and the
1 bundle in turn is positioned in the peak bundle power region of thk 84 bundle core, the staff has concluded that this is not a significant safety problem.
Studies perforned by consumers indicate that the control rod and fuel bundle worths will be changed insignificant 1y.
For example, an equilibrium all-Zr clad core fuel bundle worth of.014 delta k corresponds to.012 delta k bundle worth bundle in the original stainless steel core. For the control rod drop-out accident, the difference between the maximum rod worth.046 delta k calculated for the all Zr-clad core and the original stainless steel core. 0.42 delta k, is insignificant. The consequences of accidental reactivity insertion from these sources are only marginally worse as a result of substituting zircaloy clad for stainless steel clad.
Based on this informa-tion and the very low probability of achieving the maximum reactivity worths, we have concluded reactor safety is not significantly affected by these considerations.
1 The applicant has reanalyzed the loss of coolant accident to take into consideration the possibility of a, zirconium-water reaction. We have reviewed the assumptions and calculational model and believe that they provide a conservative estimate of the time-rate of the zirconium-water reaction. The model is similar to that used in other reactors designed by the General Electric Company and considers radiative heat transfer within the subdivided regions of the core, although assuming that no heat was lost from the overall reactor core. The rate of metal-water reaction was calculated by an accepted rate law which gives the rate of the reaction as a function of clad temperature and local extent of reaction.
For the case in which the core spray functions as designed, less than 1% of the total zirconium clad and channels was calculated to react. This amount of reaction does not significantly affect the previously calculated containment pressure history.
The Staff believes that while all possible steps should be taken to prevent melting of the core, the basis of the containment design should be a complete core meltdown with associated metal-water reaction. The applicant has analyzed the case in which only the containment spray operates to illustrate that this criterion is satisfied with the proposed core loading.
The applicant has calculated that about 23% of the zirconium in the core v'Eul'd react before locally reaching the melting temperature and falling into a pool of water below the core. An additional 5% of the unreacted zirconium was assumed to react before quenching occurred in the water. This gave a total reaction extent of about 27% of the zirconium in the reactor clad and channels over a time period of about 30 minntes.
Despite the fact that all energy generated in the metal-water reaction was assumed to remain in the core to speed the reaction, all reaction energy was assumed to be added to the containment atmosphere over a conservative time period of 15 minutes. This results in a pressure increment of about 4 psi.
The hydrogen evolved in the reaction up to thy tlhe OE <eersinment spray initiation (about 18% metal-water reaction at 15 minutes) was assumed to be totally recombined, and the energy released directly added to the containment atmosphere. The applicant calculates,
m.
i 1
f and we agree, that this energy release would bring the containment pressure to approxi-mately the design value of 27 psig. The applicant has stated that.any further recom-bination energy could be removed by the actuation of the containment spray. After the containment spray is actuated (currently at 15 minutes) the containment pressure and temperature are rapidly decreased below the design values.
The automatically actuated containment spray system is fed from the same source as the core spray header. Both core and containment spray systems can he operated simultane-ously by the motor-driven fire pump. A manual backup containment spray, with separate spray header, is also fed by the fire system. The electrically driven fire pump is backed up by the diesel generator in the event of a loss of all other auxiliary power.
We believe that adequate redundancy has been provided so that reliance can be placed on the containment spray system to reduce containment pressures and temperatures.
However, the staff believes that some margin of safety between the calculated peak pressures and temperatures and the design values of the containment should be available.
This would allow for some uncertainties in calculations and insure that the effects of further research or improved methods of calculation would not result in a marginally designed facility at some future date. We believe that this margin of safety can be attained by reducing the delay time between receipt of cont &inment pressure signal and opening of the containment spray valve from the present time of 15 minutes to about 5 minutes.
In our opinion, this would still provide adequate time for detection and nullification of spurious actuation signals. However, since the peak temperatures and pressures attained are a function of the amount of zircenium available for reaction, we believe that the delay time need not be reduced until more than one-half the fuel bundles in t';c core are zirconium clad.
In view of the above, we believe that a full core loading of zirconium may be approved at this time contingent on proper adjustment of the delay in containment spray acuta-tion. This earlier actuation of the containment spray will provide assurance that the containment design temperature -is not exceeded, as well as preventing overpressurization of the containment atmosphere, and will also allow ample time for actuation of the manual backup system if necessary.
The staff has analyzed the doses at the site boundary and low population zone for the zirconium clad core and found that 10 CFR Part 100 guidelines are satisfied. The con-servative assumptions of TID-14844 were used for fission product release and the containment leakage rate was assumed to remain constant at 0.5% per day throughout the course of the accident.
Consumers has determined that the presence of cobalt in as many as 250 fuel bundle corner locations contributes negligibly to the total activity release. The chemical energy derived from a postu'eted cobalt water rsaction and the hydrogen derived from such a reaction is also negligible in relation to containment pressure and flammability level. Using 110 lb of cobalt, corresponding to 250 cobalt bearing rods, a very conser-vative assumption of a 100 percent reaction yields 192,000 Btu of chemical energy.
The Zr-H O reaction yields chemical energy of 9.23 x 106 2
Btu with a 26.8 percent reaction This postulated total reaction of cobalt with water, very unlikely because of the physica l
i l
l i
i l !
properties of cobalt and its double encapsulation, is still only 2 percent of that already considered in connection with operation of an all Zr core.'
t l
The calculations assume that each cobalt assembly contains 200 grams of cobalt, the i
final equilibrium core will contain 250 such rods, the discharge levels for cobalt assemblies will be as high as 40 curies per gram of cobalt (conservatively all cobalt assemblies will average 40 curies per gram in the equilibrium core) and the reactor power output with cobalt present will be unchanged, hence the fission product generation f rom the fuel will be unchanged. It is further assumed, as with nonvolatile solids in the fuel rods, that only one percent of such products are released from the encapuu-lation.
In view of the 304L stainless steel cladding on the cubalt, which in turn is housed in the normal helium-filled zircaloy cladding, this release is considered conse rvative. Release of 30 percent of this amount from the containment, therefore, is compatible with previous accident analyses for nonvolatile solids. A further comparison shows that the fission products released from just one rod and correspondingly from the containment for the postulated accident would be 10,000 curies for an average corner fuel rod and 12 curies for an average corner cobalt assembly. The fuel fission product activity is based on an exposure of 10,000 Mwd /T.
Based upon the calculations and considerations discussed above, we have concluded that the addition of cobalt to the Big Rock Point nuclear reactor does not present a significaat change in the hazards considerations described or implicit in the Final Hazards Summary Report.
Conclusion For the reasons statad above the staff has concluded that Proposed Change No. 8, does not present significant hazards considerations not described or implicit in the hazards summary report and there is reasonable asserance that the health and safety of the public will not be endangered.
Accordingly, with the added restriction that containment spray be initiated automatically within 5 minutes of a containment high pressure signal caused by coolant system rupture, we believe the Technical Specifications of License No. DPR-6 may be revised as indicated in Attachment A.
ondnd sonee Ly
[) Rogers.Boyd, Chief Dru e
f Research & Power Reactor Safety Branch Division of Reactor Licensing Date:
APR 14 566
-.