View source for IA-86-678, Forwards Comments Re Seabrook Station Risk Mgt & Emergency Planning Study. Methods of Bypassing Containment Not Fully Explored & Potential Risk Impact of Sabotage Ignored

{{Adams
| number = ML20215D045
| issue date = 10/06/1986
| title = Forwards Comments Re Seabrook Station Risk Mgt & Emergency Planning Study. Methods of Bypassing Containment Not Fully Explored & Potential Risk Impact of Sabotage Ignored
| author name = Lyon W
| author affiliation = NRC OFFICE OF NUCLEAR REACTOR REGULATION (NRR)
| addressee name = Berlinger C
| addressee affiliation = NRC OFFICE OF NUCLEAR REACTOR REGULATION (NRR)
| docket = 05000443, 05000444
| license number = 
| contact person = 
| case reference number = FOIA-86-678
| document report number = NUDOCS 8610100598
| package number = ML20214K191
| document type = INTERNAL OR EXTERNAL MEMORANDUM, MEMORANDUMS-CORRESPONDENCE
| page count = 17
}}

=Text=
{{#Wiki_filter:_                                                                        _     ._     _ _ _ _ _
W 9( 1!%t                         '
.                                                                                                 p '. L c) A 6 f/l .' /J f/f C MEMORANDUM FOR:      Carl Berlinger, Chief                                     84J # f 8 d 8 " /' /
Reactor Systems Branch Division of PWR Licensing-A                                Np KN       /_ ,
r w dt THROUGH:              Richard Lobel, Section Leader                                     -
Reactor Systems Branch Division of PWR Licensing-A FROM:
Warren C. Lyon, Sr. Nuclear Engineer Reactor Systems Branch Division of PWR Licensing-A

==SUBJECT:==

REVIEW OF SEABROOK DOCUMENTS PERTAINING TO CHANGE IN EMERGENCY PLANNING ZONE SIZE

==REFERENCES:==
: 1.   "Seabrook Station Risk Management and Emergency
       -                             Planning Study", Pickard, Lowe and Garrick, Inc.,
PLG-0432. December 1985.
     ?
: 2.   "Seabrook Station Emergency Planning Sensitivity Study", Pickard, Lowe and Garrick, Inc., PLG-0465, April 1986.
I have read the reference documents and have complied a number of observations and questions which may be useful in our continuing review. These are documented in Enclosures 1 and 2. I suggest these be transmitted to BNL for their consideration during the formal review.
I find this work to represent a considered investigation of Seabrook Station response to severe accident conditions. It is a logical extension of the g
     -        Probabilistic Risk Assessment work which I helped to review, and it describes the most detailed investigation of LOCA outside of containment that I have encountered.
In general, I believe the work to be well founded, although there are exceptions which are identified below. It is not clear if the exceptions would seriously perturb the overall conclusions, and this is an area which should be investigated. If the exceptions art overlooked (for the moment), the reports present reasonable conclusions in the areas in which I am qualified to formulate a professional opinion.
ToIA-86418 e8, in                                                                 '
IOI0059R          Xt1   17ff                                      _ _ _ _ __ _ _ _ _

l My perception is that we are investigating this area from the viewpoint of a comparison between what was done several years ago and what we know now, with the objective of reaching a conclusion with respect to the rational.e used in fonnulating the existing emergency planning criteria.
A concern is that this i
approach may result in the exclusion of items that were not included by the original planners, but which impact risk in areas close to the plant when the overly conservative original basis is corrected.
Clearly, Seabrook Station has a large volume, strong containment which will mitigate almost any realistically postulated severe accident. (I do not believe the staff should expend extensive resources investigating this, although we should satisfy ourselves of the correctness of the conclusion.)
Risk problems will not be found by studying the containment.
found by rehashing products,      and the like. chemical reaction phenomena, vaporization of fissionNor will they The problem with respect to plant response, if there is one, will be in containment bypass, and it is here that I believe we should
             -    concentrate our resources. It is also here that I believe the Pickard, Lowe and Garrick (PLG work is potentially                  deficient. They have not fully explored ways to bypass co)ntainment.
         -      be valid are:                                Examples which are not obvious which I believe to
: 1.        _ Steam Generator [S_Gl Tube _ Rupture.
g This is the rupture of multiple tubes in response uncovery.         to ~hTTtemperature          which  in turn is a result of core This acciuent sequence should be of concern any time there is a core melt with the Reactor Coolant System at more than a few hundred psi pressure, with no water in the SG secondary side. These conditions lead to a potential for natural circulation transport phenomena to significantly heat the tubes prior to breach of the reactor vessel. The resulting loss of tube strength can lead to tube rupture. Reactor
         %                Coolant Pump operation, as outlined in many plant emergency procedures, almost assures this to be a concern.
the secondary side valves are open, the secondary side is breachedIf tu outside containment, or the reactor coolant system pressure is above the SG relief valve setpoints, containment is bypassed. This has not been adequately PLG   work. investigated, and is not recognized as a release path in the' 2.
LOCAseen TTa've      Outsidein thisContainment.
area. In manyPLG             has done some of the most          innova didn't go far enough.                             respects,   it is excellent. But they Comon cause failures appear to be weakly investigated. Some failure paths are ignored (were they investigated and foundnegligible?).
I also question the data base and its application.
3.
Other. What have we missed? I believe a careful investigation is in order to assess if there are unidentified bypass paths. This will be a difficult things thattask are to accomplish difficult to properly find.      since one is looking for. the 16

My final concern with the reported work, and the risk assessment approach, is that we are ignoring the potential' rfsk impact of sabotage. This is just as real as an earthquake or any of the equipment failures which can initiate a severe accident, and it will impact risk.
I do not believe the rational of planning radius zones.
is sufficient when one is potentially considering on We are already violating this rational with the large number of potential accident paths considered in the Seabrook PRA and with seismic considerations.
Of importance here, there are items that were negligible with the old, highly conservative assumptions that become significant with removal of conservatisms based upon recent knowledge. These may impact risk assessments when a reduced zone size is considered. Steam generator tube rupture due to overheating is one such item. Sabotage is another, gylg.inal s1Gud tys Warren C. Lyon Senior Nuclear Engineer e                                                                            Reactor Systems Branch, DPL-A

==Enclosures:==
As Stated cc:    T. Novak C. Rossi B. Sheron R. Ballard V. Benaroya J. Milhoan G. Bagchi n*                    B. Doolittle V. Nerses I          ff
:PWR-A f.Lo85)-
3 jrWM-5 55                8.
1 '
L ,4 ww              C          1     \,
86                                                                                              -
9/f]86            /
17

I i
ENCLOSURE 1           -
REVIEW COMMENTS ON "SEABROOK STATION RISK MANAGEMENT AND EMERGENCY PLANNING STUDY" (PICKARD, LOWE AND GARRICK, INC., PLG-0432, DECEMBER 1985)
SECTION 3 Page 2         The assumption is made that a loss of instrument air will allow the purge valves to close prior to core damage or uncovery. What is the justification for this assumption? (How long does it take for the instrument air pressure to decrease to the point that that valves close?)
What likelihood is assigned to mechanical failure which results in the purge valves failing to close?
What is the likelihood that human error, such as failure to properly close the personnel hati:h, will ;;rovide containment bypass?
      ,                            Deliberate attempts to create a release have not been addressed. Obvious 6
reasons for this omission are the difficulty in assigning a likelihood of occurrence and the need not to publicly draw a map for potential saboteurs. However, this is a real consideration, and one which the            -
l                                  I believe should be addressed before reaching a favorable finding on any request for a decrease in the emergency planning zone radius.
Operation at conditions other than full power and accident conditions while at shutdown have not been addressed. Altihough accidents under l                                 these conditions are generally considered to represent less risk than accidents at full power, I believe they should be considered before one reaches a favorable finding in regard to decrease of the emergency
                                               ~
planning zone radius.                             ,
i i
As far as I know, no PRA ha considered rupture of steam generator tubes during the approach to or the progression of core melt accidents. This is of concern due to the high temperatures in the reactor vessel, the possibility that the approach to melt occurs at high reactor coolant system (RCS) pressure, and the presence of mechanisms which may exist for
\         - . - - - - -          .. .        . - - .               -.- . - - - _ _

transporting hot fluid into the steam generator tubes, thus significantly reducing their strength.
(Mechanisms of concern involve both natural circulation and the use of reactor coolant pumps as a "last ditch" effort to prevent or delay core melt.) I believe we should consider such                                            ;
containment bypass paths before reaching a favorable finding in regard to decrease of the emergency planning zone radius.
                                                                        ! consider it necessary that all reasonable release paths be considered                                       l either directly or via a suitable allowance (for unknowns) in order for                                       ;
PRA associated work to be used as part of the basis for a reduction in the radius of the emergency planning zone. In light of the above, this
                                                  ,                    has not been accomplished. Therefore, I reconinend a careful evaluation of possible containment bypass paths and a search for areas which have
                                                ,                   ' not been considered in the PRA associated work that is being used as a basis for a potential request to reduce the radius of the planning zone.
5      Reference is made to Fig. 4-8.                            This figure shows the Boron Injection Tank (BIT) as part of the Engineered Safety Features (ESF) System. The BIT is no longer a part of the Seabrook ESF. This should be corrected.                                         '
In general, we should be assured that the information used in the report g                      is up to date.
I                                                              -
A minimal number of paths are presented as leading to pressurization of the low pressure portions of the RHR system. Figure 4-8 shows many more potential paths. I suggest that all potential paths be listed, and that we be provided with the reason for the rejection of each (such as a likelihood of the path beira open and the contributors to that likelihood). Consideration should be given to operating procedures, the        ,
likelihood that the procedures will not be followed, possible errors (such as the recently discovered difficulty with insufficient head to j
supply the SI and charging pumps in the recirculation mode and the impact upon LOCA outside of contairment, if any), and interlock behaNor.
2 1

Other risk contributors that are often overlooked in PRAs are design and construction errors. The prior consnent in regard to insufficient head is a good example.
Please address how these are contained in the data base which is being provided to the staff to justify the decrease in emergency planning zone radius.
5 Seabrook has recently proposed changes in the operating procedure for initiation of the recirculation mode of SI operation. Do these changes have any influence on the probability of satisfactory switchover from the injection to the recirculation mode?
        . 6 The FSAR gives the relief rate as 900 gpm with a set pressure of 450 psi.
The flow rate does not agree with the value used here. Please explain.
      ?
What is the mechanism for assuring that plant changes and new knowledge are promptly factored into the technical considerations which form a part of the foundation for staff consideration of a reduced emergency planning zone radius?
I note that FSAR Figure 6.3-2 appears to be inconsistent between the
{           actual figure and the text which describes valve positions. This also impacts upon proper operation of the SI system. See, for example, Valves 14, 21 and 22.
Is this correct? If so, what are the imp 1tcations with respect to the issue under consideration here?
7 What is a sump low level alarm? (Fifth paragraph) 7 What consideration has been given to relatively.small breaks and interaction with the fusible links in the ventilation system? (Is the potential impact worth consideration?) The concern is that enough hot water may be released to activate the links, thereby tenninating ventilation and indirectly causing failure of the pumps due to               {
overheating of the pump motors, and that this could occur at a time earlier than might occur due to flooding.
                                                                                              ]
l 3

l g
l                         I do not understand the conclusion that presence of water in the reactor  ,
cavity will decrease (significantly?) the revaporization of fission products from RCS and perhaps RHR surfaces. I anticipate that a significant quantity of heat producing radioisotopes will remain in the wreckage of the reactor vessel, and this may be effective in heating whatever gases or vapor are flowing toward the break. Has this been         '
investigated?
10     What is the justification for the statement that the first sign of trouble will be pressurizer low level or low pressure alams? I anticipate a number of other indicators may be first, such as abnomal indications from the PRT or even a smoke alam.
        ,        11 ~There have been a number of indications that there is a good chance of containment spray being actuated due to a high RHR relief valve release rate into containment. What is the justification for this conclusion?
Include the effect of containment heat sinks and containment cooler operation in the respons,e.
i 11    The statement "As soon as the pumps begin to produce flow to the RCS,        '
s             valves in the miniflow lines close and all RHR pump flow is injected into the reactor vessel via the RHR cold leg injection Ifnes" is not correct.
The sensors are not located at the RCS to detect flow at that location.       '
Further, one is postulating a break in the RHR system, and a significan't portion of the pump flow may never reach the RCS (as is stated in the following paragraph).
)               11   The last paragraph contains a number of timing of event statements.           '
Please provide justification of each. Plots of plant behavior showing         .
suitable parameters and indicating the event points are sufficient for most. Operator response infomation, in addition to RCS parameter              '
infomation, is necessary for the statement that RCPs will be ' tripped within about 21 seconds of break initiation. (I personally observed a simulator run in which RCPs were not tripped for several minutes in a 4

e I
large break simulation.
The operator missed the step, and I didn't say anything because I wanted to see how long it would take before someone caught the mistake.)
12 Failure to diagnose a LOCA outside containment is identified in the third paragraph. What actions are under way with Seabrook Emergency Procedures to correct this situation? What notifications have been given by Seabrook to,the NRC, Westinghouse, INP0, or others in regard to this deficiency? (See also ti,e discussion and recommendations later in the report, such as pages 3-34 and 3-35) 15 This discussion is not clear with respect to what is in the present study
              ~
and what was not considered in the SSPSA but is considered in the present study. A number of LOCA outside containment events are missing. For example, the following do not appear:
: a.      Inadvertent opening of the two hot leg suction valves due to common cause failure such.as improper maintenance, malfunction of the interlock system, design error, or unidentified means.
j         b.
Failure of the stem or other internal connections in valves equipped with limit, switches or failure of a limit switch (including improper maintenance such as reversing indication).
c.
Malfunction due to fire or other electrical short circuits.
(Includes testing operations of all types, such as testing that involves jumpers which could be incorrectly connected to cause a LOCA outside containment. The testing is not necessarily limited to testingofvalves.)
: d.      Common cause failure associated with improper saintenance such as installation of improper components (gaskets, seats, or va'lve disks) which may fail almost inmediately or at a later time.
l l
5

            ~
0 15 The coment is made "The disc. rupture failure mode has not been reported in the nuclear industry data base." The staff notes a number of disc valve failures have been reported in the nuclear industry for non-SI or
                            -RHR systems, including some where there were comon cause failures.
Were these considered in the work being reported here? (Note several valve failures at San Onofre Unit I may be too recent.)
15 Are the individual check valves leak tested after each RCS depressurization or are they checked at once in series? If they are checked individually, please describe how this is accomplished. Include checking of the hot leg injection valves in the response. (Note the hot leg injection path was excluded from consideration because there are three valves in series.)
         ?
15 Please describe the valve inspections that are promised each time the plant goes to cold shutdown or is refueled. (The implication is that the valves are opened up for inspection.)
16 The largest leak rate in Figure 3-3 is of the order cf 200 gpm, whereas the arena of interest ranges to 65,000 gpm. Please justify this 5
extrapolation. In addition, include consideration of the valve designs, operation modes, and sizes used in the data base as contrasted to Seabrook.
16 Several references have been provided to not failing the RHR system at pressures of roughly 2250 psi. What calculations substantiate these statements or assumptions? What temperatures were involved while the RHR was considered to be exposed to 2250 psi?         .
17 What likelihood of failing to test after cold shutdown or failure to test after refueling was included in this work? What likelihood was assigned to an incorrectly conducted test that would impact the conclusions?
What is the likelihood that a 1.0CA outside containment can be teminated if initiated? Include consideration of all possible paths and address the dynamic interaction of closing the various valves against water j                       flowing from the RCS at high pressure.
1 l                                                            6

i 17 In the discussion of the BWR events, would it be accurate to state that none of the events were included in Table 3-8 and Figure 3-37 17 Inclusion of accumulator check valve leakage is viewed as a conservatism according to the report authors. Were accumulator check valves also incorporated in the counting to obtain the total number of valves in the RCS and ECCS systems? If so, how was this a conservatism?
17 There are large differences between check valves. What is the impact of this on Seabrook's estimated results?
An item under consideration for advanced nuclear power plants is the ability to monitor pressure on the low pressure side of check valves.
This could provide early warning of check valve leaks and would provide monitoring capability to help assure check valves were operating properly. The same monitoring capability with respect to RHR suction line valves could identify if individual valves were mispositioned or malfunctioning. Would such a system for Seabrook be of significant benefit in reducing risk in a reduced size emergency planning zone?
21 The first POV in the RHR suction line is identified as not having a stem mounted limit switch. What is the impact of this on plant risk and what would be the cost of adding appropriate instrumentation so that valve ,
position would be indicated?
21 Why is it conservative to assume that MOV valve leakage and failure upon j
demand due to a sudden pressure loading are the same as for check valves, particularly if the check valves are already closed when exposed to the sudden pressure loading?
21 An item.that concerns me is dependent failure of the RHR suction valves.
This is not mentioned. Why?
7

22 The split fraction in which the flow is greater than the relief capacity is given as 0.09. Please provide justification for this with consideration given to the previously identified concerns regarding RHR suction line valve failure modes.
23 This is a cursory list of actions an operator can take to mitigate the accident.
23 What is the frequency of failures in the pipe tunnel that led the authors to conclude they are very low?
24 As identified earlier, little thought appears to have been given to the real world of mitigation. For example, this discussion of potential
             ' actions presumes the charging pumps still work. What procedures exist to turn them off prior to damage.
24    The meaning of the following is not clear: "The second consideration is contingent on the interfacing LOCA being located in the TRR vault at an elevation higher than that required to get significant scrubbing due to flooding. If such a leak has occurred, the configuration of the vault is 6
such that the leaking primary coolant itself will flood the vault.
External sources for flooding the vault could also be employed." If the LOCA is located high enough that flooding will not provide release mitigation, what does all of this mean?                                  '
l 24     With respect to modeling all of the important failure modes I have already commented above.
25 Are the emergency procedures at Seabrook updated as identified in " Top Event 01"? See also the middle of page 3-26 for the same issue.
27 There are a number of cases where the combined sump pump capacity is sufficient to remove leaks and keep the vaults from flooding. In these cases, the RHR, Si, and CS pumps are assumed not to be impacted by flooding. No consideration is given to failure of one (or both) sump pumps.
8 l
l

    .                                                                                                        1 l
l i
30                "To our knowledge, the IDCOR evaluations discussed above have not been refuted by the NRC." This statement adds little to the report if a knowledgeable reader is involved. There are many things that .have not been refuted by NRC which NRC has not reviewed, or which, for whatever reason, nothing has been said.
30              The text indicates a pipe failure probability of about 6 x 10-3 if the low pressure piping is exposed to 2250 psi. This is based on an assumed probability of failure at the material yield strength of 0.01 and a probability of failure at the ultimate strength of 0.99. As previously identified, temperature is not specified.Is it the 350'F design temperature? Please provide substantiating information. Cover all components which are exposed to a pressure or temperature which is above
                         ' the design value. This should also include items such as the RHR pump seals and seal material response at RCS temperature.
What is the maximum flow rate that can be injected into the RCP pump seals? (Of potential interest since it may be an alternate path for injection into the RCS.)
g        35 Shutting an RHR system crosstie valve is identified as an action to help isolate a LOCA outside containment involving the RHR/SI systems. Has a careful evaluation of these systems been performed to assess isolation            ,
strategy?                                                                   _
35-36 Relative water levels in the RHR vaults and the RCS are mentioned.
What are the water volumes in these regions as a function of elevation?
(Of particular interest is the level at the top, of the core and at the elevation of the hot leg connections to the RHR.)
36 What is the justification for the statement that the water lev,el in the vaults will be approximately the same as that in the RCS? (I'donot agree.)                           -
37             What is the likelihood that motor operated valves can be used in a flooded RHR vault?
9

(
37 "End state DLOC contains sequences in which the interfacing LOCA has been teminated, and the ECCS has been degraded (D) (RHR or SI puh.ps have failed). ... The point estimate frequency of DLOC is 4.0 x 10-7 per year.
The additional failures required to achieve core melt would lower this frequence by at least one order of magnitude." What is the justification for this conclusion? (We have already lost a portion or all of the ability to inject water into the RCS via the usual paths.)
37 (Bottomofpage) Why does failure of one charging pump lead to core melt?
The perception is that sufficient flow could be provided by alternate means to keep the core covered (two other charging pumps, perhaps the reactor makeup water pumps).
e         39
                                 " Containment recovery is assumed successful once the conteinment spray and recirculation functions have been accomplished." Please address the likelihood that initiation of containment spray could reduce the water content in the containment atmosphere, thereby making possible a deflagration involving hydrogen and oxygen. In addition, address the behavior of core melt in the reactor cavity and whether late addition of water prevents containment meltthrough.
      \
45 What is to be the status of the " temporary" 34.5 kV power lines?
46                                                                                                                .
What is to be the status of the mobile power supplies?
46 What capability has been provided to connect external pumps as identified in the second and third paragraphs? (The brief mention on page 3-48 may imply that little has been accomplished.) Use of a pump to simply inject water into containment via the sprays on a short term basis (no recirculation) does not appear to be identified. Has this been l                       considered?
46 This page identifies a numbe~r of possibilities. What are the specific plans?
10

                                                                        ~
48    There have been several references to purchase of a mobile electric generator by pooled resources. What is the likelihood that such a generator would be needed by several plants at the same time?
l SECTION 4 l
Chapter 4 contains little elaboration on the accident sequences which I have identified earlier as receiving inadequate consideration. Indeed, the material in this chapter appears to substantiate my concern that potentially important items have been missed in the technical evaluation intheSeabrook(PLG) document.
12   A generalization is drawn to the effect that not-through-the-insulation
    .          heat losses are large as compared to the heat generation rate. Thus the conclusion is drawn that "...the primary system heat losses are sufficiently great that the potential for long-tem revaporization within the primary system for a PWR with a large, dry containment is negligible. Therefore, this issue does not influence the Seabrook Emergency Planning Study." This is incorrect. Lets start with core melt. If we apply this argument to the core rather than to the upper a          plenum, we can argue that heat loss from the primary system is sufficiently great that all generated heat is balanced by heat loss.
Therefore, THE CORE WILL NOT ATTAIN HIGH TEMPERATURES AND WILL NOT MELT!
l              (Thus, we don't have to worry about any of these phenomena, and all of ^
l             the work is unnecessary!)
The fallacy in the argument pertains to the location of energy generation and energy transport. One must be able to transport the heat to the location of loss without attaining a high tempe'rature. This is clearly i             not the case in the core. It similarly is not the case for the upper j             plenum,' but the situation is not as clear. There are two contributors to l             the problem:                                                     -
1 I
l 11

J l
: 1. Most calculations of upper plenum behavior involve one dimensional     j modeling of flow, and any fluid (liquid, vapor, or gas) that passes through the core is assumed to flow through the upper plenum and out the hot leg. This mode: ling is inccrrect for upper plenum behavir,*
under severe accident conditions where a major portion of the core has-been uncovered or the core is being vapor or gas cooled. Strong recirculation patterns will develop which link the core and upper plenum temperatures. At pressures in the range of nonnal operation
;                          (typically 2250 psi), the linkage is strong, and some of the upper plenum component temperatures can be expected to closely follow core temperature. The strength of the linkage diminishes with decreasing pressure. Although some calculations have been perfonned which indicate a qualitative coupling, I AM NOT AWARE OF SUBSTANTIATED
      ,                    QUANTITATIVE INFORMATION WHICH CAN BE APPLIED TO THIS ISSUE.
: 2.      A significant quantity of heat producing radioisotopes probably has left the core under the conditions of interest, and substantial deposits can be expected in the upper plenum structure. Most calculations incorrectly .Jo not include thus effect. The influence of the incorrect modeling is the same as in item 1. THE s                    UPPER PLENUM TEMPERATURES ARE SUBSTANTIALLY UNDERCALCULATED.
THIS ISSUE CANN T BE DISMISSED AS READILY AS IS DONE IN THE SEABROOK (PLG) DOCUMENT.
i 12-13        The report authors att arguing that that little material will be released with the exception of noble gases, even in the case of eventual high pressure containment failure. I suspect the containment atmosphere will be depleted of oxygen early in the accident. I wonder what happens if the containment is suddenly opened with all of that hot stuff around inside containment that would love to get hold of some oxygen. I am particularly interested in regions where there are large quantities of material, such as under the vessel and within the vessel.
12

15     The question is not the temperature at which zircaloy can liquefy the fuel, but the temperature of a significant quantity of the melt at the time of release. What is this temperature and what are the vapor pressures of the constituents at that temperature? (See,forexample, Figure 4-16.)
19    The discussion of the feedwater pipe penetration indicates a probability of 17% that the penetration fails and 83% that the pipe fails. Pipe failure is stated not to be a problem because the external feedwater pipe and valve will maintain containment integrity.. We shculd check to see what valves are involved (checks?) to assess the situation if the applicant is taking credit for this behavior.
36    What is the status of the decontamination factor of 1000?
36     For the case where there is no suppression pool, what is the likelihood that the filtration system can hold the fission product inventory from a thennal aspect?
43    "The containment atmospheric purge line is the only penetration that can
{                    be opened during normal operation and provide a direct release path."
What is the likelihood that one of the large purge lines will have failed to seat following a refueling shutdown, and that the failure will not have been discovered?
70      (I've only glanced at portions of this table.) Is the basemat thickn'ss              e j
the overall thickness, or the thickness under the reactor cavity sump?
(These represent a significant depression 1'n both Zion and Seabrook.)
13

ENCLOSURE 2 OBSERVATIONS AND QUESTIONS PERTAINING TO "SEABROOK STATION EMERGENCY PLANNING SENSITIVITY STUDY",
PICKARD, LOWE AND GARRICK, INC., PLG-0465, APRIL 1986 l
I have few comments that were not made in Enclosure 1. Such coments are not duplicated.                                                                    '
SECTION 4                             COMMENTS 1
The second paragraph does not provide the background for the conclusion that the release via basemat meltthrough at Seabrook does not provide attenuation as would be the case with a soil based plant. Seabrook's containment rests on bedrock, but is not bonded to the bedrock.
Therefore, if there is a basemat meltthrough, the contents of the containment are assumed to vent directly to the atmosphere via a gap between the bedrock and the containment. This may be a conservative assessment.
1 The last paragraph contains "For consistency with NUREG-0396, only releases during the acute accident time of 24 hours were considered."
Although I have no trouble with a consistent comparison, as I have comented previously, I don't believe we should limit the assessment to a
  !          completely consistent basis from the viewpoint of public risk. (Further, thereareinconsistenciesanyway.) A public risk assessment should be achieved, and items such as the above should be justified in order to b.e excluded.
O 14}}