ML17212B575
| ML17212B575 | |
| Person / Time | |
|---|---|
| Site: | Saint Lucie  |
| Issue date: | 04/26/1982 |
| From: | Clark R Office of Nuclear Reactor Regulation |
| To: | Robert E. Uhrig FLORIDA POWER & LIGHT CO. |
| References | |
| TAC-08377, TAC-8377, NUDOCS 8205070042 | |
| Download: ML17212B575 (18) | |
Text
~,.
~ e Docket No. 50-335 APR 2g QQ APR 86 1982 Dr. Robert E. Uhrig Vice President Advanced Systems 5 Technology
. Florida Power 5 Light Company P. 0. Box 529100 IHami, Florida 33152
Dear Dr. Uhrig:
DISTRIBUTION Docket File Local PDR ORB Rdg CO D.Eisenhut JHeltemes RAClark
@~CD >~@
PKreutzer E'ELD
~pgg g)982~
NSIC untUgg g',Qgu.l 9
(I )
Q
@~PPy,rg lg~e~~<gg g ACRS (10) iO CNelson Gray File Ti 0 QPI 5 During our review of your proposal to operate St. Lucie.Unit. 1 at 2700 Mwt; we requested and received fL=S1-477 dated NoveIIIher. 13,-3.~81) your comitmejt to install start up flgx channel alarms for the detection of boron dilution events /by'he next (Cycle 6) refueling outage.
Our review of this issue was documented in section D.7.1.1 of the Safety Evaluation accompanying our approval to operate St. Lucie Unit 1 at 2700 f4vt (Amendment No. 48 issued November 23, 1981).
At that time we indicated that we were evaluating the capability of operating PMRs to provide adequate protection against uncontrolled boron dilution events.
That evaluation has proceeded to the point where we no longer require that licensed operating reactors install the subject alarms.
Our basis for this revised position is contained in Enclosure 1.
Therefore we no longer require that you install the alarms as committed to in your letter of November 13, 1981.
Further evaluation of this issue or future events at nuclear power plants could affect this position,'owever, we do not expect any evaluations we have planned to restore the requirement for alarms at St. Lucie Unit 1.
In light of the above you are requested to inform us, in writing and by Hay 28, 1982 of your intent with respect to installation of the subject alarms.
Sincerely, Originai signed bg Robert A. C'lark Robert A. Clark, Chief Operating Reactors Branch P3 Division of Licensing
Enclosure:
As stated R
DL:0 C 6 n/dgd 4/
/82
.D.L:.0 PKre zer 4/
/82 OFFICE P SURNAME/
DATEf NRC FORM 318 (10-80) NRCM 0240 cc:
See next Paq
'205070042 820426
'DR,'ADQCK,05000335 gP,-,
-,PD RSB: D
...4..J8 R.
..at:.........
4432.8P...........
OFFICIAL RECORD COPY USGPO: 1981-335.960
~ ~
t
~
~
pP e
Florida Power 8 Light Company CC:
Harold F. Reis, Esquire Lowenstein, Newman, Reis 8 Alexrad 1025 Connecticut Avenue, N.W.
Washington, D. C.
20036 Norman A. Coll, Esquire McCarthy, Steel, Hector 5 Davis 14th Floor, First National Bank Building Miami Florida 33131 Indian River Junior College Library 3209 Virginia Avenue Fort Pierce, Florida 33450 Administrator Department of Environmental Regulation Power Plant Siting Section State of Florida 2600 Blair Stone Road Tallahassee, Florida 32301 Mr. Weldon B. Lewis County Administrator't.
Lucie County 2300 Virginia Avenue, Room 104 Fort Pierce, Florida 33450 U.S. Envfronmental Protection Agency Region IV Office ATTN:
Regional Radiation Representative 345 Courtland Street, N.E.
Atlanta, Georgia 30308 Mr. Charles B. Brinkman Manager - Washington Nuclear Operations C-E Power Systems Combustion Engineering, Inc.
4853 Cordell Avenue, Suite A-1
- Bethesda, Maryland 20014 Mr. Jack Schreve Office of the Public Counsel Room 4, Holland Building Tallahassee, Florida 32304 Resident Inspector/St.
Lucie Nuclear Power Station c/o U.S.N.R.C.
P. 0.
Box 400 Jensen Beach, Florida 33457 Bureau of Intergovernmental Relations 660 Apalachee Parkway Tallahassee, Florida 32304 Regional Administrator Nuclear Regulatory Commission, Region II Office of Executive Director for Operations 101 Marietta Street, Suite 3100 Atlanta, Georgia 30303
ENCLOSURE
'INADVERTENT BORON DILUTION l.
Introduction and background Boron dilution eve'nts are routinely analysed in every PWR's FSAR. (l)
The analyses cover two rather different circumstances:
Inadvertent boron dilution with the reactor at power and inadver.ent.boron dilution with the reactor subcritical (i.e, while in shutdown or refueling modes).
It is (2) the latter that has been questioned.
There 'have been 25 reported instances of inadvertent boron dilution during maintenance and refueling.
Although none has yet occurred, the (2) safety concern is the possibility of an inadvertent criticality. If the boron is sufficiently diluted and the reactor core is near beginning of cycle,, it is possible to bring the reactor to criticality with all. of the control rods inserted into the core; The only way to shut the core down again in such a circumstance would be to re-borate the moderator, which I
could take considerable time.
The events have occurred with sufficient frequency to raise the question whether, considering their possible consequences, the degree of protection is'appropriate.
Several branches have engaged.in a dialogue (2,3,4) on this matter.
2.
Safety significance 2.1 Estimated frequency of inadvertent criticality.
Boron dil 'tion events during a shutdown or refueling have usually been causeC either by human error or by failures of special, non-process equipment such as inflatable seals.
Therefore,'vent frequencies cannot be easily calculated by fault tree analysis.
- Moreover, because no event has yet resulted in criticality, it is not possible to simply add up the r
number of events in operating history.
The fact that no inadvertent criticalities have happened in 337 PWR-years allows us to estimate an upper bound to the frequency.
By assuming a Poisson distribution and using the conventional 95Ã confidence level,.it is straightforward to demonstrate that the frequency o, inadvertent. criti-
-3 calities is, at most, 9 x 10 events per PWR-year.
- However, an upper limit is not sufficient'to gauge the significance of boron dilution events; a "best estimate" (in some sense) is needed.
The only information available is contained in the frequency of boron dilution events which have happened but which did not result in criticality'.
Host of these events can be considered "precursor" events to an actual inadvertent criticality.
'he severity of a precursor event is defined here in terms of the shutdown margin remaining at the end of the event.
That is, an event which'was halted with 2% shutdown remaining is considered more severe than event which was halted with 10$ remaining shutdown margin.
~n the figure at. ached, we have plo.ted the number'f events of a given severity vs. the severity using the information in the ORNL/NSIC report.
Final shutdown margins were calculated from final boron concentrations (where available) using "typical" values of boron worth taken from RESAR and Crom the Yiidland SAR.
1nitial boron concentra"..ions were used to estimate the point i'n the fuel cycle when the event occurred.
3 w It was assumed that all control rods were inserted: for, events which occurred during shutdown mode (vessel head in place).
However,'e assumed that one rod was removed for events which took place during
~ I refueling--which is, of course, a realistic assumption for fuel handling operations.
Not suprisingly, the number of events goes down as the severity class-ification increases.
To estimate an expectation value for the number of critical events, a two-parameter exponential distribution was fitted to the data.
Extrapolation of this distribution to the point of zero shut-down margin gives a value of 0.67 events in a'ime interval of (currently) 337 PWR-years.
- Thus, we expect the frequency of inadvertent criticalities
-3 to be on the order of 2 x 10 events per PWR-year.
This calculation, although rough, gives an answer that is reasonable.
With 46 PWRs presently operating, we would expect an inadvertent criticality roughly every ll years, if nothing were done.
0 However, this number does not take into account the effect of the neutron monitoring instrumentation.
As a reactor core approaches criticality, neutron flux does not rise linearly.
Instead; the reciprocal of the flux drops linearly as'shutdown margin decreases.
The net effec.-is that neutron flux rises slowly as the reactor core goes from 10% to 9'4 shutdown, but rises very dramatically as shutdown margin drops below 0.5~.
None of the events tabulated in Reference 5 came close enough to criticality for the neutron monitoring channels to trigger alarms. 'hus, to realistically estimate the frequency of aq event that continues in dilution to criticality, we must gi've some credit for the neutron flux channel
- alarms, which are usually set one hal f to one decade above background;
4 Since the control rods are already fully inse'rted into the core in this event, the only actions which will prevent criticality are stopping the dilution or reborating the moderator..Both are done by the opera'tor.
~ f Thus, the credit to be given for neutron flux alarms is governed by the t
reliance which can be placed on the operator.
Me will assume
{based purely on judgment) that the operator will be able to correctly diagnose the problem and successfully prevent criticality 90 percent of the time.
This drops the frequen'cy of a'riticality by one order of magnitude, to
-A 2 x 10 events. per PWR-year.
2.2 Consequences of an inadvertent cr iticality An inadvertent criticality event, whatever its implications concerning plant operations and procedures, is not hazardous because of the fact of criticality.
In actual fact, the achievement of criticality is a rather subtle and unspectacular event.
If the event were terminated by scram
{as has actually happened in BMRs), there would be no significant safety hazard associated with the event at all.
In the PWR case under consideration here, all rods are either already in the core or are disconnected from their drives.
'Ei her way, there is no scram reactivity available.
Shutdown by emergency boration will take much more time than shutdown via scram.
The impor.ant parameter is the peak power level achieved by the core.
Once the core becomes critical, it will heat: up with a positive period governed by the rate af dilution and by moderator temperature and Doppler feedback.
Eventually the coolant may boil and the peak power level will b limited by void generation in the moderator.
Prel:iminary calculations indicate that, assuming BOG parameters (worst case),
a power level of about 3% of rated would be reached.
(These calculations are limited (e)
~ I in their ability to model the multidimensional aspects of void feedback.)
Two aspects of safety significance result from such an event.
- First, the neutrons and fission gammas (and possible airborne activity from any leaking fuel pins) are a hazard to workers in the vicinity of the reactor.
- Second, should reactor thermal power become high enough to fail fuel,
.there is a possibility that activity will be released to the environment, espec'.ally if the event occurred when the reactor vessel head was removed.
We have not yet been able to quantify the hazard to workers in containment.
However, it is doubtful that the hazard is serious since evacuation alarms connected to. the source range excore neutron flux monitors will in most cases give workers warning before conditions become hazardous.
- moreover, the water and shieldsng located around the core are probably enough to shield wor kers from neutrons and gammas with a core power of on the order of 35 of rated.
Similarly, a core power of 3~ of rated is not likely to fail fuel that must withstand decay heat rates of.his same order.
The enly likely con-sequence is'he. release of gap activity from any leaks already present.
If we make the standard assumption of users of the GALE codes that 0.16$ of the fuel leaks, the total activity released to the coolant would be roughly 69,000 ci.
This is not enough activity to be significant unless the vessel head is removed.
(In-Reference 5, one sixth of the cold shutdown/refueling events took place in the refueling mode, and thus presumably took place with the vessel head removed.)
If the vessel head were not in place, about 10K of. this activity, or 6900 ci, would escape from containment, based on analyses of dropped fuel assembly events.
3.
Possible fixes Since these everits are caused by a wide spectrum of causes,'it is not practical to reduce the frequency of boron dilution events other than by bringing the matter to the attention of plant operations personnel and having them upgrade their procedures (if and where appropriate).
It has been proposed to install a microprocessor-based monitor on the source range neutron flux instrumentation.
Such a monitor, if connected to a
display panel such as the Sa,ety Parameter Display System (SPDS), could give earlier warning of loss of shutdown margin than is possible with the present instrumentation, and thus would reduce the probability of a boron dilution event leading to criticality.
We have evaluated the cost of such a system The results are:
(7) control grade instrument, alarm only............$ 50,000.
safety grade instrument, alarm plus automa.ic initiation of emergency boration 0
~
~
~
~
~
~
~
~
~
~
~
~
~
$300,00o 4.
Priority score With these numbers in hand, it is relatively straightforward to. estimate a priority score for this issue.
(see Reference 8 for a description of the basis and significance.of these scores.)
~ ~
Fr equency:
Me expect inadvertent critical ities to occur at a rate of 2 x 10 events per PMR-year.
Of these, roughly one sixth will take place with the vessel head removed.
Thus, the frequency of radioactivity-releasing events is 3 x 10'PMR-Y.
The upper limit (95% confidence) on inadvertent criticality frequency without credit for neutron flux alarms was a factor of.5 over the "best" estimate.
If we assume a symmetrical distribution and also assume a factor of 5 error in the credit for the neutron flux alarms
~
and a factor of 3 error in the chance of the head being off the vessel, the estimated error in the frequency of radioactive release is plus or h
minus a factor of 8.
Consequences:
The release is expected to be on the.order of 6900 curies, primarily noble gases.
Me will use an estimated error of a factor of 5, again I
based on judgment.
Costs:
.Me have estimated a cost of $50,000 per plant for the cheapest hardware (7) fix.
Roughly one half of this figure represents the cost of paperwork, and is thus relatively insensitive to the exact nature of the fix.
The cost to the NRC is estimated to be two staff months plus one staf, week for each of 46 operating PWRs.
This 'corresponds to an NRC cost of $64,000.
The uncertainty in the costs, which ar e dominated by the
$50,000 per plant, is at most a factor of two.;
Score.
Using the numbers
- above, the results are:
Score 4 x 10 Ci/Y/10 $
Range
~
~ I 4x10 to 4x10 These scores are in the low priority range when compared to other issues competing for NRR's a tention.
5.
Withdrawn but lrippable control rods The issue of withdrawn but trippable control rods has also been raised (g) in connection with boron dilution events.
The essence of the issue Ls that the definition of shutdown margin in the Standard Technica'1 Specifications allows a plant to take credit for control rods which are not actually in-serted in the core, provided they are trippable.
In such a'ase, if a boron dilution event were to occur with the plant in shutdown or refueling modes, the absence of one or more control rods would allow criticality to occur sooner than assumed in the analyses described in Section 1 above.
'his issue is somewhat peripheral to the basic issue of inadvertent boron dilution.
Nevertheless, several comments can be made:
First, the issue is not very realistic., It is physically impossible to remove a control rod but still have it trippable when the head is. removed.
L When the vessel head is in place, it does not appear likely that pignificant consequences will arise from an inadvertent criticality, even if the missing rod.were not trippable.
- Second, the fact that the withdrawn rods are trippable means that the reactor will be scrammed by the neutron flux channels before the power level
-9>>
could become significant.
(This is true even if:the highest-worth rod sticks.)
The analagous event has actually happened in BWRs, with no safety consequences.
Third, the Standard Review Plan calls for refueling mode analysis (1) which assume all rods are removed from the'core.
Thus, the validity of analyses which conform to SRP 15.4.6 is not alter ed by withdrawn rods, at least during refueling mode.
Fourth, the concept of hiving some negative reactivity "cocked"- and ready to shut the reactor down is not necessarily a
bad one.
We have allowed LACBWR to do this during core alterations, for example.
The reason for this is that shutdown margin is not easy to measure directly even with sophisticated laboratory-type equipment such as pulsed neutron sources.
(Note that a microprocessor was necessary to give early warning in ghe hardware fixes described in Section 3.)
Having negative reactivity ready to insert rapidly allows one to terminate an inadvertent criticality early in the event, allowing time for worker evacuation, event diagnosis
- and, if necessary, emergency boration.
This "cocked rod" concept, although virtually never used during PWR core alterations, is in effect what is done during certain physics startup.tests.
Therefore, we conclude that there is no basis for not allowing credit for withdrawn but trippable rods.
~
4
-1 0-6.
ConClusions ard Recommendations 6.1 Occurrence of non-severe boron dilution events 25 boron dilution events in 337 PWR-years, coupled with the fact that 46
~.4 PWRs are now operating, imply that we must expect a boron dilution event every three months or so.
This is not a cause for concern, since the like-lihood of the dilution resulting in criticality is very low.
The real significance of the event is its implications regarding the plant's pro-cedur es during shutdown and refueling.
6.2 Occurrence of inadvertent criticality An estimated frequency of 2 x 10 (one event'very 5000 PWR-years) is 4
not high enough to justify emergency action.
Even with 100 PWRs'operating, events would occur only every 50 years or so.
Nevertheless, the frequency is Sigh enough that an inadvertent criti-cality would be predicted to take place before the last.PWR now in the CP stage is decommissioned.
6.3 Significance for reload reviews Based on the low estimated frequency and low estimated consequences of an inadvertent criticality, we conclude that boron dilution events do not constitute a significant risk to the public.
The licensing process need not wait while this mat.er is resolved.
Reference 10 discusses the legal and procedural aspects of such situations.
7.
Future work The future work discussed here is nest intended to imply that the'on:lusions
reached in Section 6 above will change.
Instead;-,it-.is intended to confirm some figures and bring the task to completion.
The only outstanding issue is the significance of an inadvertent criti-cality to wor kers within containment.
This is being invdstigated by AEB.
Even though the hazard is expected to be sma')1 a confirming calculation should be completed.
When this is done, RSB and PTRB should prepare needed revisions to SRP 15.4.6.
iz Severity of Boron Dilution Events Number of events vs.
[12'~ - SDM]
References:
1.
Standard Reveiw Plan, NUREG-0800, Section 15.4.6, issued July 31, 1981.
2.
- Memo, R. J. Hattson to T. E. Hurley, dated September,15, 1981.
3.
- Memo, R. A. Clark to T.
P. Speis, dated October 6, 1981.
4.
- Memo, R. J. Mattson to D.
G. Eisenhut,.dated October 23, 1981.
r I 5.
E.
W. Hasen, "Evaluation of Events Involving Unplanned Boron Dilutions
'in Nuclear Power Plants,"
prepared under contract W-7405-emg-26 (NRC FIN BO755), in press.
6.
- Letter, N. S.
DeMuth (Los Alamos National Laboratory) to R. T. Curtis (NRC, dated November 18, 1981.
7.
Final Report, "De.ermination of the Cost of Modifications Needed to Mitigate Boron Dil'ution Events in PWRs during Cold Shutdown, "DOE Work Order 20-81-297 (NRC FIN A6452) dated December 14, 1981.
8.
Enclosure 3 of "Plan for Early Resolution, of Safety Issues,"
~
dated August 25, 1981.
9.
- Hemo, R. J. Mattson to S.
H. Hanauer, dated January 26, 1982.
10.
- Note, S.
F. Scimto to W. Butler, dated March 10, 1977.
0
~ ~
I f
1 A