ML20056F162
| ML20056F162 | |
| Person / Time | |
|---|---|
| Issue date: | 06/10/1993 |
| From: | Congel F Office of Nuclear Reactor Regulation |
| To: | Tipton T NUCLEAR ENERGY INSTITUTE (FORMERLY NUCLEAR MGMT & |
| References | |
| NUDOCS 9308260131 | |
| Download: ML20056F162 (12) | |
Text
r o
O June 10, 1993 Thomas E. Tipton DISTRIBUTION Vice President & Director Central Files RErickson ANelson Operations, Management and NRC & Local PDRs FKantor Support Services Division TMurley RPedersen Nuclear Management and Resources Council FMiraglia AMohseni 1776 Eye Street, N.W.
WRussell SBoynton Suite 300 FCongel MGCrutchley (YT 0930097)
N hinoton, D.C.
20006-3706 EButcher NRR Mailroom (YT 0930097)
LJCunningham EJordan, AE0D
Dear Mr. Tipton:
PMcKee KBrockman, AE0D Thank you for your letter of May 12, 1993, in which you requested the U.S.
Nuclear Regulatory Commission (NRC) to review the " questions and answers" (Q&As) stemming from the NUMARC workshop on implementation of NUMARC/NESP-007,
" Methodology for Development of Emergency Action Levels." We have completed our review and have enclosed our comments for your consideration.
The NRC endorsed the NUMARC guidance in Regulatory Guide 1.101, Revision 3,
" Emergency Planning and Preparedness for Nuclear Power Reactors." Our review of the Q&As focused on evaluating the impact of each response on the implementation of Regulatory Guide 1.101. Specifically, the Emergency Preparedness Branch assessed each " answer" to determine whether or not it modifies the original intent of the associated initiating condition in NUMARC/NESP-007.
The comments provided concentrate on those " answers" that may require clarification to ensure appropriate implementation of the NUMARC guidance.
The enclosed technical comments were coordinated and prepared by Mr. Scott Boynton of the Emergency Preparedness Branch.
Mr. Boynton can be reached at 301-504-2926 to answer any questions in that regard.
Sincerely, Original Signed by Frank J. Congel Frank J. Congel, Director Division of Radiation Safety and Safeguards Office of Nuclear Reactor Regulation
Enclosure:
Comments on NUMARC "Q&A"s
/
I C: pef /hhR D:DRSS:NRR [\\/
0FC EPS:PEPB:NRR EPS:PEPB:NRR SCE:PEPB:NRR NAME SABoynton ASMohseniM FKM RAE/iNon FJCongel hk V
DATE (c //O/93
[ //C/93
[/10/93 (s / b3
[o//0/93 0FFICIAL RECORD COPY DISK / DOCUMENT NAME:
G:\\0930097.YT
((:
v). ~/ (k, c
- a
- )
I M%M M%c (gy mn>
_~
i ENCLOSURE STAFF COMMENTS ON QUESTIONS AND ANSWERS CONCERNING IMPLEMENTATION OF NUMARC/NESP-007:
. " METHODOLOGY FOR DEVELOPMENT OF EMERGENCY ACTION LEVELS" I.
BACKGROUND In September 1992, NUMARC conducted a workshop on the implementation of NUMARC/NESP-007, " Methodology for Development of Emergency Action r
Levels." NUMARC/NESP-007 was endorsed by the Nuclear Regulatory j
Commission (NRC) in Regulatory Guide-1.101, Revision 3, dated August 1992..
It is an alternative methodology to that described in NUREG-0654, l
Appendix 1, by which licensees may develop site-specific emergency action level schemes. The workshop was designed to present the NUMARC l
methodology to the industry and to respond to questions _ regarding its implementation.
4 NUMARC committed to provide the industry with written responses to generic questions to clarify potential inconsistencies and implementation concerns identified at the workshop.
The Emergency Preparedness (EP) staff of the Office of Nuclear Reactor. Regulation (NRR) has worked closely with NUMARC on the development of the l
responses. A public meeting was held on April 7,1993, in which members of NUMARC discussed the proposed responses with the EP staff. As a result of that meeting, NUMARC' agreed to revise several of their responses based upon verbal input from the staff.
By letter, dated May 12, 1993, Thomas Tipton, Vice President & Director, Operations Management and Support Services Division, NUMARC, provided the revised-responses to the staff for formal comment.
II.
DISCUSSION The staff's review of the " questions and answers" (Q&As) was conducted 1
to ensure that the responses do not alter the intent of NUMARC/NESP-007 and that the Initiating Conditions (ICs), as clarified, continue to meet the requirements of 10 CFR 50.47(b)(4) and Appendix E to 10 CFR Part 50.
j III. COMMENTS A.
GENERAL 1.
Question #6 - A majority of the EAls list several exampies for each Initiating Condition. Are utilities required to address each example or select only those examples that are appropriate for the site?
If it is the latter, is it required to address in the basis document or NRC submittal why all the other alternatives were not selected?
O O
l l
f NUMARC Response:
Licensees should address each example that applies to thei.r particular sites. While not required, it is recommended that they also explain why those examples not selected are not i
applicabl e.
Disposition of all of the example EAls will facilitate the staff review of licensee st,bmittals.
j Comment: The staff agrees with the proposed response; however, it I
should be noted in the response that if an EAL does not' apply because of its wording (e.g., valid reading on perimeter radiation monitoring system greater than 10 mR/hr sustained for 15 minutes or longer), the licensee is expected to use other means, if available, l
for ei.iry into the IC.
In other words, for the example given above, it may not be enough to state that this EAL does not apply because the licensee does not have a perimeter radiation monitoring system.
l The intent is to use all available data to determine whether the IC l
I should be entered.
2.
Question 89 - If, after the fact, it is discovered that an event has occurred that caused an EAL to be reached without adverse consequences, should a classification declaration be made?
NUMARC Response:
If an emergency condition nn longer exists, there is no reason to declare an emergency.
The NRC shall be notified i
after discovery within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, meeting 10 CFR 50.72 reporting
]
criteria.
Licensees may wish to make arrangements to inform other off-site authorities of the circumstances of the situation by non-emergency means.
Comment: The last sentence of the proposed response should be 3
revised to say: " State and local authorities should also be notified as soon as practical, or in accordance with arrangements made in i
advance".
4.
Question #12 - Questions indicated a desire to allow normal i
Technical Specification Action times associated with RCS leakage to identify and/or stop a leak that is above the Unusuai Event threshold of SUS.
For example, under NUREG-0654 guidelines, with >
10 gpm leakage, we currently have the 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> allowed by tech specs to identify and reduce the leakage before the event must be classified.
NUMARC guidance requires immediate classification of the event without allowing the operators time to identify and reduce the leakage.
NUMARC Response: The threshold for declaration has been significantly raised from typically 1 gpm to 10 gpm Unidentified Leakage and 10 gpm to 25 gpm identified leakage. With a leak of such magnitude, it is not reasonable to expect isolation and an j
Unusual Event should be declared immediately.
However, if the plant specific technical specifications allow 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> before the event is declared, the tech specs take precedence.
o O
i Comment:
The response appears to contradict itself by first stating that an Unusual Event should be declared immediately upon exceeding the threshold and then giving credit for technical specification action statement time to recover prior to declaration.
It is unclear as to what is meant by tech specs taking precedence.
Credit for the action statement time in deferring an emergency declaration should only be given when leakage exceeds technical specification limits but has not yet exceeded the Unusual Event threshold.
The response should be revised to reflect this.
B.
SYSTEM MALFUNCTION i
1.
Question #1 - Does the EAL of SUI apply to one unit whose essential busses can be energized from another (unaffected) unit at l
a multi-unit site?
NUMARC Response:
No, provided that the cross-tie is made within 15 minutes and no load is being carried by an emergency generator.
l Comment:
Sul does apply to this situation.
Plants that have the capability to cross-tie power from a companion unit may take credit for the redundant power source in the associated EAL for this IC.
Inability to affect that cross-tie within 15 minutes is grounds for i
declaring the Unusual Event. The response should be revised to reflect this.
2.
Question #7 - Must all control rods be inserted for a scram to be considered successful? If not, what is the threshold for determining a partial scram to be adequate per SS2?
NUMARC Response:
For PWRs, the scram is considered successful when enough control rods are inserted to cause the reactor power to fall I
below that percent power associated with the ability of the safety systems to remove heat. This is typically between 3 percent and 6 i
percent rated thermal power. The wording of the EAL should include the concept that reactor power is below [X] percent and decreasing under existing plant conditions.
Subsequent actions necessary for the reactor to be prepared for a cooldown (normal boration to a i
xenon-free cold shutdown boron concentration) and depressurization i
are not to be considered.
In SA2, the discriminator is whether the i
automatic reactor protection system functioned.
If it did not or if rector power was not decreased to below X percent and decreasing due j
to insufficient rod insertion, an Alert is declared.
Note this i
assumes that the manual reactor trip function immediately available to the operators was effectively used to complete a successful scram (power < [X] percent and decreasing).
Failure of the manual reactor trip function from the control room is the discriminator for the Site Area Emergency per SS2.
In this situation, wording should be chosen to include-the concept of " power not less than [X] percent I
i i
O O
~
i
. )
and decreasing and manual reactor trip function has not been effective."
If emergency boration is needed to bring the reactor to below X percent and decreasing, the scram was unsuccessful.
For BWRs, all control rods need not be inserted.
A scram is considered successful if it results in achieving a state in which the reactor will remain shut down under all conditions without boron injection.
For SA2, if manual actions result in the reactor being shut down under all conditions without boron injection, an Alert is declared.
Escalation of a Site Area Emergency (SS2) is not required.
If sufficient control rods are not inserted to reduce reactor power to below the APRM downscale setpoints, an immediate Site Area Emergency (SS2) is declared.
If the APRM downscale setpoint is achieved, but suppression pool temperature is greater than Boron Injection Temperature, a precursor exists for a threat to containment and thus a Site Area Emergency is warranted.
Comment:
EALs are considered more effective when a negative statement is used to define a threshold, therefore, the response should emphasize the idea of an unsuccessful scram.
In that light the scram can be considered unsuccessful when the rod insertion fails to reduce power below [X]% and decreasing. The first two sentences should be rephrased in the negative to articulate this threshold.
Similarly, the first two sentences of the second paragraph should also be rephrased in the negative.
C.
HAZARDS AND OTHER CONDITIONS AFFECTING PLANT SAFETY 1.
Question #9 - HU4/EAL 1 refers to a bomb device discovered within the Protected Area and outside the Vital Area. No EAL or IC specifically addresses a bomb discovered within the Vital Area.
NUMARC Response: A bomb device discovered within a Vital Area is a i
Security Event in accordance with the plant Safeguards Contingency Plan.
The event therefore exceeds HSI, " Security Event in a Plant Vital Area." The basis of HU4 does state that bomb devices found in the Vital Area would result in EAL escalation.
Coment :
The staff agrees with the proposed response. A confirmed
[
explosive device in a vital area is a direct threat to vital equipment designed to protect the public. A Site Area Emergency is, therefore, warranted.
2.
Question #13 - If a fire affects only a component of a safety system such that the entire system function is not disabled, then why declare an Alert as required in HA2? For example, technical specifications allows one RHR pump to be inoperable.
From a plant safety perspective, how is having the pump damaged by fire any different from tagging it out of service for maintenance?
t
o O
~.
NUMARC Response:
It is important to note that this EAL addresse.s a i
fire and not the degradation in performance of affected systems.
System degradation is addressed in the System Malfunction EALs. The
[
reference to damaag of systems is used to identify the magnitus of the fire and to discriminate against minor fires.
The reference to safetV systems is included to discriminate against fires in areas having a low probability of affecting safe operation. The significance here is not that a safety system was degraded but the i
fact that the fire was large enough to cause damage to these systems. Thus, the designation of a single train was intentional i
and is appropriate when the fire is large enough to affect more than one compo ~ t.
This could involve the loss of a function necessary for the prt.. action of the public -- the definition of a Site Area l
Emergency.
i A similar situation exists at the Unusual Event.
The intent of the l
15 minute duration of extinguishing efforts is to size the fire and to discriminate against small fires that are readily extinguished (e.g., a smoldering wastepaper basket).
i Removal of equipment for maintenance is a planned activity controlled in accordance with procedures and, as such, does not i
constitute a substantial degradation in the level of safety of the plant -- the definition of an Alert.
Comment: The staff agrees with the proposed response. The purpose of this 1C is to help ensure timely emergency response organization (ERO) activation when propagation and consequences of a fire cannot be readily determined.
If plant conditions then degrade, appropriate ERO posture will already have been established.
t The last two sentences of paragraph I are not required to answer the l
question and may raise further questions.
It is recommended that they be deleted from the response.
3.
Question #14 - The design for shutdown outside the control room is based upon the capabilities of the operating crew only. Why, then, does the basis for the EAL in HA5 state that assistance from the TSC or other center is necessary?
NUMARC Response: Declaration of an Alert is appropriate to notify onsite and offsite emergency organizations that a control room l
evacuation is taking place and that the possibility exists, hwever small, that control cannot be established outside of the contrcl
}
room.
Comment:
It is recommnded that the following be added to the response:
t t
g a
v
. As stated in NUREG-0654, the rationale for the alert class is to provide prompt notification of minor events which could lead to more serious consequences given operator error or equipment failure or which might be indicative of more serious conditions which are not yet f ully realized.
When an Alert is declared, based upon control room evacuation or any other EAL, the Technical Support Center is staffed.
4.
Question #15 - The basis of HS2 states that the time to establish plant control should not exceed 15 minutes.
Our plant has an analysis which shows RCIC injection is not necessary until 23 minutes after control room evacuation.
Can we take credit for this analysis?
NUMARC Response: This information will be helpful in determining the appropriate time frame for your facility, but should not be used exclusively. The intent of the EAL is to establish control of important plant equipment and knowledge of important plant parameters in a timely manner.
Primary emphasis should be placed on those components and instruments that supply protection for and information about safety functions.
Typically, these safety functions are reactivity control (ability to shut down the reactor and maintain it shut down), reactor water level (ability to cool the core), and decay heat removal (ability to maintain a heat sink) for a BWR. The equivalent functions for a PWR are. reactivity control, RCS inventory, and secondary heat removal.
The intent of the 15 minutes is to ensure timely recognition of the inability to control or monitor these functions.
If, for example, the reactor is shutdown upon control room evacuation and there is early evidence of the inability to start a RCIC pump (BWR) or HHSI pump (PWR), this will affect the ability to cool the core and a declaration of a Site Area Emergency is warranted.
In some instances, it may be desired to re-evaluate the order in which specific equipment is transferred to the remote shutdown panel.
In these cases, it is appropriate to include failures of both trains of specific equiptrent within a 15-minute time frame.
In other instances, all equipment control is transferred simultaneously after the control switches are properly aligned.
In these cases, the time frame for establishing indication should be specified.
In both cases, indications must be monitored and other EALs (Fission Product Barrier, System Malfunction, etc.) applied as appropriate.
Coment: The response should be revised as follows:
The intent of this IC is to capture those events where control of the plant cannot be reestablished in a timely manner.
Each licensee is expected to determine a maximum time to establish control of the plant from the remote :,hutdown panel.
The guidance in NUMARC allows the maximum time to be 15 minutes without additional justification.
e O
. Justification should be provided when a licensee desires a maximum time exceeding 15 minutes.
The determination of whether or not control is established at the remute shutdown panel is based upon the judgement of the Emergency Director (ED). The ED is expected to ma'Ke a reasonable, informed judgement within the maximum time, that the licensee has control of the plant from the remote shutdown panel.
D.
ABNORMAL RAD LEVELS Ouestion #8 - The radiation effluent example initiating conditions call for an assessment to be performed. What are examples of adequate assessments? Can RASCAL or site-specific models be used?
What should happen if the assessment is not completed in a timely manner?
NUMARC Response: The minimum assessment varies with classification.
The same assessment method may be used for all classes if the method meets the most restrictive assessment requirement.
For the Unusual Event, the assessment should be based on the method described in IE Notice 84-15. The intent here is to correlate the valid monitor reading with the fraction of T/S activity released.
This may be done with graphs or computer software. At the higher classifica-tions, the assessment method must include the actual meteorology as an analysis input.
The intent is to ensure that the X/Q based on actual meteorology at the time of the event is not significantly more adverse than the annual average X/Q used to establish the monitor EALs.
If this was to be the case, a higher classification might be warranted.
in the absence of a more site-specific dose assessment capability, RASCAL could be appropriate for sites that can be adequately modeled by simple Gaussian models if site-specific inputs and actual meteorology were used.
If the assessment is not completed in a timely mant er, the event should be declared on the basis of the existence of valid radiation monitor readings that cannot be readily discounted. However, the assessments should continue in order to identify whether or not an escalation to a higher classification might be warranted.
Comment: The NUMARC response is acceptable provided the sixth sentence is revised to read: "The assessment method should include the actual meteorology as an analysis input."
E.
FISSION PRODUCT BARRIERS - PWR Ouestion 43 - NESP-007, Rev. 2, Pages 5-32: S/G secondary side release with pri/sec leakage.
If condenser air ejector off-gas is
O O
. filtered and monitored prior to release to atmosphere, is this still considered a release to atmosphere in the same light that a direct release to atmosphere from atmospheric steam dump or main steam relief valve would be?
NUMARC Response: The primary intent of this indicator is to address steam generator tube ruptures that constitute a loss of both the RCS and the containment barriers. This indicator should be used in conjunction with the SGTR indicators under the RCS barrier. The threshold for establishing the bypass of containment was intended to be a prolonged (6 to 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />), significant release of radioactivity from the steam generator directly to the environment. Such an event could occur, for example, on a SGTR with a concurrent loss of offsite power if the ruptured steam generator must be used to cool down the plant.
If the ruptured steam generator is isolated in accordance with plant procedures in a timely manner, this threshold has not been exceeded. There are other possible prolonged secondary releases that would be addressed by this indicator such as a failure, outside of containment, of a secondary line on the ruptured steam generator that prevents isolation, or a stuck open relief valve on the ruptured steam generator.
The wording of this indicator sizes the primary-to-secondary leak at rates greater than technical specifications (typically 1 gpm).
The threshold on the RCS barrier indicators of SGTRs are set at leak rates comparable with the capacity of charging pumps.
If a prolonged release occurs from a steam generator during a leak, only an Unusual Event would be declared.
If the steam generator was ruptured without a prolonged release occurring, an Alert would be declared. At the point that the steam generatcr is ruptured and has a prolonged release, an SAE is declared.
Further escalation would occur if fuel damage were noted.
While a radioactivity release via the main condenser air ejector is physically a path for a release to the environment, it is generally a controlled and/or monitored path as the question notes. As such, it does not represent the level of degradation of plant safety as do the conditions discussed above.
Typical air ejectors remove non-condensable gases at rates of tens of lbm per hour, whereas a stuck open relief valve can pass hundreds of thousands of lbm per hour.
Also, iodine plateout and partitioning in the steam generator, steam lines, and condenser hotwell are efficient in reducing the iodine available for release. The noble gas release consequences would be negligible in the absence of fuel damage.
Comment: The proposed response should be revised as follows:
The primary intent of this indicator is to address steam generator tube ruptures (SGTRs) that constitute a loss of both the RCS and the containment barriers. This indicator should be used in conjunction
e O
~
_g_
\\
wit' he SGTR indicators under the RCS barrier.
The threshold for contoinment bypass was intended to be a prolonged release of radioactivity from the ruptured steam aenerator directly to the environ ent.
This can be expected to occur when the main condenser is unan t li bl t tu accept the contaminated steam (ie. SGTR with concurrent loss of off-site power and ruptured steam generator is required for plant cooldown).
When the main condenser is available, and no other pathways exist for release, the main condenser air ejector is, by itself, a path for release of radioactivity to the environment; however, it is generally a controlled and/or monitored path as the question notes.
As such, a SGTR, with release via the main condenser, does not meet the intent of a prolonged release directly to the environment and does not require the escalation of the event to a SAE.
Other examples of crolonaed releases are (1) unisolable failure, outside of containment, of a secondary line on the ruptured steam generator, or (2) a stuck open relief valve on the ruptured steam generator.
The wording of this indicator sizes the primary-to-secondary leak at rates greater than technical specifications (typically 1 gpm).
The threshold for potential loss of the RCS barrier during a SGTR is set at leak rates comparable with the capacity of a single charging pump.
If a prolonged release occurs from a steam generator during a leak, an Unusual Event would be declared based upon the loss of the containment barrier.
If the steam generator was ruptured without a prolonged release occurring, an Alert would be declared based upon the potential loss of the RCS barrier.
If a prolonged release is occurring or expected in conjunction with the SGTR, a Site Area Emergency is declared based upon the potential loss of the RCS barrier and the loss of the containment barrier.
Escalation may also be required if fuel damage were noted.
F.
FISSION PRODUCT BARRIERS - BWR 1.
Question *3 - How is the TW sequence (loss of decay heat removal in PRA usage) addressed for loss of containment since it requires containment venting?
NUMARC Response:
Intentional venting of containment in accordance with the E0Ps is considered a loss of containment under " Primary Containment Barrier Example EALs," number 2.
Comment:
The staff agrees with the proposed response. Moreover, off-site officials should be notified as soon as possible, preferably prior to the initiation of venting.
2.
Question #7 - If an SRV is stuck open or cycling, should an emergency declaration be made?
O O
NUMARC Response:
The answer depends upon other events.
If an SRV is stuck open or cycling and no other emergency conditions exist, an emergency declaration is not appropriate. However, if the fuel is damaged and the relief valve is allowing the fission products to-escape into containment, a loss of RCS has occurred.
It must be determined if an accident is in progress.
If not, the FPB matrix is not applicable.
Comment:
The staff agrees with the proposed response. Abnormal operating procedures (A0Ps) or technical specifications may require the plant to shutdown when an SRV is stuck open or cycling.
For cycling SRVs, technical specifications for suppression pool temperature and/or level, and SRV operability, may drive operators to shutdown the plant.
This is within the analyzed operating envelope of the plant and does not represent a degradation in the level of safety.
This also does not represent a significant l
precursor to further plant degradation.
For a stuck open relief valve, A0Ps would dictate a shutdown when all actions directed in the procedure failed to reseat the SRV. This may be a normal or rapid shutdown depending on the plant.
10 CFR 50.72 would direct reporting of this event and technical specifications would require an indepth analysis if the stuck open SRV caused an uncontrolled cooldown.
However, this, in and of itself, does not represent an emergency condition.
f 3.
Question #10 - On the FPB table under containment barrier, the maximum core uncovery time limit curve is referenced in conjunction I
with 2/3 core height.
It seems this is the improper use of this chart considering the input assumptions for the curve.
j NUMARC Response: What is being presented here is that substantial core damage has taken place (-20 percent clad damage) and that this inventory is now in the containment.
For various times after shutdown, it will take longer to reach this amount of damage due to decreased decay heat. The utility must generate a site specific curve, or select a single time, that will reflect 20 percent clad damage due to decreased water inventory.
i Comment:
The response should be revised as follows:
The questioner's assertion is correct.
The Maximum Core Uncovery Time Limit (MCUTL) defines the maximum amount of time that the core can remain completely uncovered without resulting in a Peak Clad Temperature of 1500 *F.
The curve is based upon the core power history and the time since shutdown when the uncovery occurs. The i
MCUTL curve should, therefore, be applied to the EAls when the core is completely uncovered or RPV Level is unknown.
This is consistent with GE Emergency Procedure Guidelines, Revision 4.
i l
s
~
l l Due to the large uncertainties in severe accident progression, it should be assumed that severe core melt is imminent when the Maximum Core Uncovery Time Limit (MCUTL) is exceeded.
It would not be acceptable to delay the declaration of a General Emergency and issuance of protective action recommendations beyond this point.
It should be noted that RPV Level is also an indicator of the status of the RCS and fuel Clad Barriers in the FPB matrix.
t I
i l
J 1
?
b I
I I
_ _, __ _.