ML18058A372
| ML18058A372 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 04/21/1992 |
| From: | Slade G CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GL-91-18, NUDOCS 9204290005 | |
| Download: ML18058A372 (20) | |
Text
consumers Power POW ERi Nii MICHlliAN'S PROlillESS Palisades Nuclear Plant:
27780 Blue Star Memorial Highway, Covert, Ml 49043 April 21, 1992 Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 DOCKET 50-255 - LICENSE DPR PALISADES PLANT -
GB Slade General Manager UNREVIEWED SAFETY QUESTION - POTENTIAL FOR LEAKAGE OF CONTAINMENT SUMP WATER TO THE SIRW TANK DURING AN MHA - ADDITIONAL INFORMATION AND UPDATE OF JUSTIFICATION FOR CONTINUED OPERATION In a letter, dated June 14, 1991 and supplemented on July 17, 1991, Consumers Power Company (CPCo) notified the NRC that we had determined that a potential leak path existed whereby previously unaccounted for radioactive post accident primary coolant system water could leak to the Safety Injection and Refueling Water (SIRW) tank.
Our June 14, 1991 letter to the NRC, in accordance with 10 CFR 50.59 and pursuant to 10 CFR 50.90, requested that the Palisades Facility Operating License be amended by granting an exemption from the FSAR requirement to perform the Maximum Hypothetical Accident (MHA) analysis.
In a follow-up letter dated January 10, 1992, CPCo provided additional information relating to the scope of the discrepant condition and requested that the exemption for compliance to an MHA analysis be extended from our original request to the end of the 1994 refueling outage (beginning of cycle
- 12) when we could properly verify and modify our design as required.
CPCo met with the NRC staff on January 15, 1992 at the White Flint offices to discuss the ongoing MHA analysis and interim analysis.
As a result of NRC staff reviews, additional clarification and information has been requested by the NRC staff. Attachment 1 contains responses to each of the NRC questions.
~204290005 926421---~,
PDR ADOCK 05000255.
- I 2
Our January 10, 1992 submittal provided an evaluation of the acceptability of continued operation for the Palisades Plant. Attachment 2 contains a justification for continued operation which clarifies and expands the previous evaluation by including information contained in the responses to the NRC questions (from Attachment 1). The justification for continued operation follows the guidance given in Generic Letter 91-18, "Information to Licensees Regarding Two NRC Inspection Manual Sections On Resolution of Degraded and Non Conforming Conditions and On Operability." This justification for continued operation replaces in its entirety Section 7 (acceptability for continued plant operation) of our January 10, 1992 letter.
~A~
Gerald B Slade General Manager CC Administrator, Region III, USNRC NRC Resident Inspector - Palisades Attachment
ATTACHMENT 1 Consumers Power Company Pali sades Pl ant Docket 50-255 UNREVIEWED SAFETY QUESTION-POTENTIAL FOR LEAKAGE OF CONTAINMENT SUMP WATER TO THE SIRW TANK DURING AN MHA ANSWERS TO NRC QUESTIONS April 21, 1992 8 Pages
1
- 1.
Detailed figure and schematics with appropriate notes should be provided.
The figure should clearly identify all the valves that will be tested or..
have been tested to determine the total leakage into the SIRW tank (e.g.,
identify CV-3056, CV-3027, MV-3225).
Figure 1 shows equipment described in the amendment request.
The figure shows the approximate elevation of the valves along the piping systems and includes the elevations of the SIRW tank and containment sump.
- 2.
Other tank isolation valves (e.g., SIRW tank suction valves) through which no leakage to the tank is expected may not be tested. If so, justification for such exclusion should be provided.
As a result of a question that was raised in 1990, concerning leakage through the SIRW tank main discharge valves, an evaluation was completed that concludes that due to the elevation difference between the Palisades containment sump and the SIRW tank, no leakage of containment sump water into the SIRW tank would occur.
Our recent reviews show that the elevation difference between the minimum SIRW tank water level (1.6 feet) and the maximum containment sump level (6.4 feet) after recirculation begins is 54.03 feet.
The SIRW tank elevation is 643 feet 10 inch~s and the containment sump elevation is 585 feet. A calculation using a reference temperature of 120°F concluded that a containment pressure of 37.85 psia would be required to force flow from the sump toward the SIRW tank.
The peak containment pressure in the current FSAR analysis following RAS (no containment air coolers) is about 23 psig (37.7 psia). Therefore, the containment pressure will not push sump water into the SIRW tank.
The margin between the pressure needed to force water back to the SIRW tank and that calculated as a peak LOCA pressure appears to be small.
However, the containment pressure analysis contains the following conservatisms which drive the result of the analysis to this high peak pressure.
The containment analysis uses a bounding service water temperature, conservative service water and component cooling water system flow rates, a design heat exchanger fouling factor, and a conservative core decay heat correlation.
Each of these parameters affect the heat removal rate and containment pressure following RAS.
Also this containment peak pressure will exist for only a short period of time. Additionally, if the three air coolers are operating, the maximum containment pressure following RAS is less than 14 psig.
Considering the above results and these conservatisms, we conclude the dose consequence analysis need not include leakage of containment sump water into the SIRW tank, due to containment pressure.
During our reviews we have recently determined that other possible scenarios may exist where post accident containment sump water could migrate to the SIRW tank through the main discharge lines. These possible scenarios involve the assumption of a unique set of equipment failures including leakage through several valves in series. These valves are presently not leak tested.
For the interim we have calculated post accident doses using a plant specific LOCA analysis source term.- Administratively, we have set a 2
0.1 gpm total SIRW tank in-leakage limit.
We set this limit based on MHA source terms and the need to keep containment sump water out of the SIRW tank. Analysis has been completed using the plant specific LOCA analysis source term which shows that leakage back into the tank in excess of 20 gpm for 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> is acceptable.
Based on this analysis we are confident that higher interim leak rates could be justified if necessary.
Therefore, on an interim basis, by using the plant specific source term in the LOCA analysis the effect of leakage into the SIRW tank is not a significant contributor to control room operator or off-site doses.
As part of our long term solution, we will verify the validity of the other possible scenarios and include their impact in our final analysis and design as required.
- 3.
On Page 6 of the January 10, 1992 submittal, the bottom paragraph states, "The Palisades system configuration and procedures do not guarantee the desired pH can be achieved until several hours after a LOCA."
Identify clearly what steps you intend to take as short-term measures to guarantee achievement of desired pH (i.e., at least 7) in the containment sump as early as possible after recirculation begins. Evaluate at what time the desired pH will be achieved.
By our existing design, hydrazine is automatically added to the containment sprays and safety injection water that comes from our SIRW tank. Then, following RAS, our emergency procedures stated, until recently, that the results of our post accident containment sump sample would be evaluated and operations would be told when and how much sodium hydroxide to add to the containment sump to maintain a pH of 7 to 8.
Since no credit is taken in the present revision to the Standard Review Plan (SRP}, for iodine retention with the use of hydrazine, the following short term procedural controls assure that more immediate sump pH control is initiated following a LOCA.
Emergency Operating procedures EOP 4.0,"Loss of Coolant Accident Recovery," and EOP 9.0, "Functional Recovery Procedure," have been revised to direct the operators to add 200 gallons of sodium hydroxide upon recirculation actuation signal verification. This amount of sodium hydroxide will buffer the containment sump to a pH of 7 to 8.
The system operating procedure (SOP) 4.9, "Containment Spray and Iodine Removal System," for adding the sodium hydroxide is already in place and these other procedure changes assure that this action will be completed as soon as possible. Continued containment sump sampling and analysis will assure that containment sump pH is maintained.
3 Taking into consideration the containment pressure at the beginning of sump recirculation and the pressure head of the NaOH tank, the injection.
of 200 gallons of sodium hydroxide would take less than an hour.
The NaOH is injected into the containment spray and safety injection systems to ensure a thorough mixing and achieve a pH of 7 to 8.
- 4.
On Page 9, of the January 10, 1992 submittal Section 6, "Proposed Plan for Resolution of Issues," you say the fo71owing, "changes that might be incorporated, for example, include elimination of hydrazine, installation of passive pH control device in the containment sump, definition and inclusion of realistic valve leak rates into the SIRWT, possibte control room ventilation system modifications." Clarify which among the above will qualify as interim measures.
Clarify whether currently you add hydrazine or sodium hydroxide to the spray solution (it is not clear).
Currently, hydrazine is automatically injected into the containment spray and safety injection systems.
Sodium hydroxide is manually added for long term sump pH control. The elimination of hydrazine will be completed as a long term action.
We are evaluating the use of passive containment sump pH control as a long term modification.
For the interim we have revised our emergency operating procedures to add 200 gallons of sodium hydroxide as soon as the recirculation actuation signal is verified.
We have established 0.1 gpm as the total allowable leakage into the SIRW tank based on the MHA analysis. Testing during the 1992 refueling outage showed that the measured leakage through the HPSI recirculation valves (CV-3056 and CV-3027) was significantly less than (approximately 0.0 and 0.01 gpm) our established leak rate limit. This allowable leak rate (0.1 gpm) has been included in our interim calculation for the plant specific LOCA analysis.
The calculation shows that this leakage into the SIRW tank results in an acceptable control room and off-site dose.
A modification has been completed during the 1992 refueling outage which allows installation of a blank flange in the shutdown cooling connecting piping between the SIRW tank and MV-3225.
During power operation, this line will be blanked off to prevent leakage into the SIRW tank.
Verification that this blank flange is in place has been included in the shutdown cooling system start-up checklist CL 3.1, "Engineered Safety System Checklist (Shutdown Cooling in Service)."
We have no plans to modify the control room ventilation system on an interim basis.
4
- 5.
On Page 9, Section 6, states, "The previous SRP departures approved by the NRC's resolution of NUREG-0737, Topic III.D.3.4 for the Palisades plant, wi77 remain as is." The staff notes that these are {l) non-seismic and non-missile protected remote air intake, (2) no provision of radiation and smoke detectors in the intake, (3) no provision of automatic diesel power in the event of loss of off site power for the air conditioning compressors, and (4) no provision of some instrumentation in the control room that may be useful in confirming isolation of the normal intake by redundant dampers.
Provide justification as to why these deviations should be acceptable for the change in circumstances.
NUREG-0737, Item III.D.3.4, "Control Room Habitability," included a review of the Palisades plant control room design.
The NRC's safety evaluation in a letter dated April 29, 1983, concluded that plant control room habitability systems were acceptable.
It further discussed the evaluation results which concluded why the four areas of non-conformance to the standard review plan sections and regulatory guides referenced in NUREG 0737 were acceptable.
The discussions for the four areas of non-conformance, as they were stated in the safety evaluation, are listed below.
(1) A major improvement in the proposed modifications consists of placing an emergency air intake 95 meters from the outer wall of the containment.
To achieve this larger separation, the steel pipe connecting the intake to the control room habitability system must extend beyond those plant structures designed for seismic and missile protection. While the system active components are protected from missiles and seismic events, the emergency intake pipe is not, except for the inherent strengths of the steel pipe itself.
(2) In order to be protected, the radiation and smoke detectors are within the control building, rather than at the intakes as suggested by the standard review plan.
(3) While the fans and dampers are provided with automatic diesel power in the event of a loss of off-site power, the air conditioner compressors are not, and require either manual action or resumption of off-site power to be restarted, if needed.
(4) All instrumentation needed to monitor the operation of the emergency system is provided within the control room, however, some instrumentation useful in confirming the isolation of the normal intake by redundant dampers is not.
The NRC's SER furth~r stated that, "We have evaluated these departures from the Standard Review Plan and conclude that they are justified as being acceptable in order to achieve the advantages of the modifications within the constraints of existing structures and power supplies.
Successful operation of the system would not be prevented by lack of instrumented detection of a single failure or by failure of unprotected components."
5 An additional finding by the NRC staff is, 11 *** that the systems will provide safe, habitable conditions within the control room under both normal and accident radiation and toxic gas conditions, including loss-of-coolant accidents."
The NRC staff then concluded, 11 *** that occupancy can be maintained under accident conditions without personnel receiving radiation exposures in excess of 5 rem whole body or its equivalent to any part of the body for the duration of an accident."
System modifications may be required as a result of a new MHA analysis (see response to item 8) in order to meet the GDC-19 dose requirements.
Just as system acceptability in the NRC's safety evaluation was not dependent on the departures of the SRP and regulatory guides, meeting the GDC-19 dose requirements, in the future, is not dependent on meeting the previously approved departures of the SRP (above).
Therefore, the previous conclusions made by the NRC will remain valid when the GDC-19 requirements are met.
We conclude that circumstances have not changed such that the previous conclusions are affected and we do not intend to modify the control room habitability systems to fully comply with the criteria of the SRP and regulatory guides referenced in NUREG 0737; as they relate to these four areas of non-conformance.
- 6.
What is the rationale for referring to the control rod ejection accident on Page 12, of the January 10, 1992 letter, since it does not determine the limiting leakages to the SIRW tank?
(9 and 10 gpm for GDC 19 and 10 CFR 100 limits; MHA on the other hand will limit these to 0.5 and 2 gpm for GDC 19 and 10 CFR 100 limits as per page 7 of the submittal}.
The rational for referring to the control rod ejection accident was to compare the results of the worst analyzed accident (as compared to the MHA) and what the resultant dose impact it would have on the control room and off-site doses.
As noted in our amendment request and referenced in our FSAR, the MHA is postulated to release substantially more fission products and result in more severe consequences than any other accident analyzed in the FSAR.
We were showing what the calculated effects of a plant specific accident (i.e., control rod ejection) would have on the control room operators for consideration of continued plant operation on an interim basis.
As a result of our conversations with the NRC staff, we agreed to use a plant specific LOCA analysis to develop a plant specific source term for use on an interim basis to justify continued plant operation.
The result of that analysis is described in Attachment 2 and shows that when using the plant specific LOCA source term, the dose requirements for off-site and control room operator doses are well within allowable limits of GDC-19/ Standard Review Plan 6.4 and 10CFRlOO.
6 With the development of an interim plant specific LOCA analysis source term, the reference,to the control rod ejection accident is not needed to support continued operation of the plant and has been deleted from the revised justification for continued operation in Attachment 2.
- 7.
Justification Con Page 11 in the January 10, 1992 submittal, is not completely appropriate.
The ILRT acceptance criteria is 75 percent of the design basis leak rate to account for possible degradation and increased leakage during the long interval that separates two consecutive type A tests. Therefore, this should be reworded.
The point of the statement was that our latest ILRT results showed that we had a measured leakage rate of 0.04% per day by weight and that margin existed between this value and the maximum allowable leakage rate of 0.075% per day by weight. Although there may be increased leakage during the interval that separates the Type A tests, we were identifying our starting point as being lower than the acceptance criteria. Therefore, more margin existed between it and the maximum allowable leak rate.
Based on subsequent reviews the discussion has been removed from our revised justification for continued operation in Attachment 2.
- 8.
Justification for such a Jong time for continued operation i.e., until the end of 1994, should be provided.
Based on the results of the January 15, 1992 meeting with the NRC staff and their comments and questions on our previous submittal, we have clarified and revised our justification for continued operation.
In our January 10, 1992 revised amendment request we stated that we would submit a new MHA analysis to the NRC by April 30, 1992.
We also stated that if the NRC could review and approve the new analysis by November 1, 1992, CPCo would expect to complete modifications by the end of the 1994 refueling outage.
This schedule supports our intent to implement a 3-year modification process at Palisades (which has been outlined in separate discussions with the staff). While we can try to anticipate how calculation results will impact our present design, we will not be able to gage their full impact on our design until we can see the results and review what options we may have available to us to meet the revised analysis.
We anticipate that in late 1992, following NRC reviews of our revised MHA analysis, we will have decided on our final modifications.
These modifications could then be designed in late 1992 and into early 1993. Ordering of equipment could then begin in 1993 and with construction being completed in 1994.
Some of the questions we will be addressing are: {l) How will the proposed new source terms affect our analysis? (2) What will future NRC guidance in the area of control room ventilation have on our design? And (3) what is the possibility that ICRP 30 recommended dose limits will be approved for control room operator accident doses?
Any one of these variables could have an effect on proposed modifications.
We would like to include as much foresight in our design as possible to preclude further modifications at a later date.
7 As an example, following the guidance in Standard Review Plan, Section 6.4~ and ICRP 2, following an accident the control room operator can receive a total of 30 rem thyroid, 30 rem skin and 5 rem whole body dose.
Use of ICRP 30 would give different acceptable dose limits.
ICRP 30 has recently been accepted for occupational doses in the latest revision of 10CFR20, and it raises the occupational dose limit to an organ to 50 rem annually from the previous limit of 30 rem.
For our MHA calculations we may use the dose conversion factors in ICRP 30, but so far the limits in ICRP 30 have not yet been evaluated or considered as an acceptable bases for control room operator post accident doses.
If ICRP 30 is eventually approved as a basis for control room operator post accident dose limits, we believe that this could affect our decision to modify our control room HVAC system or consider other potential plant modifications.
SIRW TANK
_.644'-0' el 6"--
18 I 610'-0" el _____
TO~--
SPENT FUEL POOL TILT PIT CV-3056 572'-0..._e;,.;..I _.._.._
CV-3027 HPSI MINI-FLOW RECIRC LINE 584'-5" el 583'-4" el MV-ES108
.._..__578'-7' el
- r!~~(LEr iMV-3225 CROSS-CONNECT TO SHUT DOWN COOLING FIG9RE 1 - POTENTIAL SIRW TANK LEAK PATH CTMT SUMP 585'-0" el....... _
CV-3030 CV-3057 CV-3029 H PSI, LPSI, &
CTMT SPRAY PUMP SUCTION
ATTACHMENT 2 Consumers Power Company Pa 1 i sades Pl ant Docket 50-255 UNREVIEWED SAFETY QUESTION-POTENTIAL FOR LEAKAGE OF CONTAINMENT SUMP WATER TO THE SIRW TANK DURING AN MHA
- UPDATE OF JUSTIFICATION FOR CONTINUED OPERATION 8 Pages
1 JUSTIFICATION FOR CONTINUED OPERATION This justification for continued operation (JCO) provides the basis to conclude that plant operation prior to completion of an updated Maximum Hypothetical Accident (MHA) analyses and associated modifications is safe and acceptable, and would create no undue risk to the public or control room operators. This JCO provides the basis to operate until the end of the 1994 refueling outage when the modifications determined to be necessary to satisfy the requirements of 10CFRlOO and GDC-19 are committed to be installed. Five issues are addressed in this JCO.
These issues come from the JCO guidance in Generic Letter 91-18 and are:
- 1. Availability of redundant or backup equipment,
- 2.
Compensatory measures including limited administrative controls
- 3. Safety Function and events protected against
- 4.
Conservatism and margin, and
- 5.
Probability and PRA Reviews.
- 1. Availability of Redundant or Backup Equipment Component redundancy exists in the potential leak paths to the Safety Injection and Refueling Water (SIRW) tank.
The three potential leak paths are the high pressure safety injection (HPSI) recirculation line, shutdown cooling cross connect line, and the SIRW tank main discharge safety injection lines.
The HPSI recirculation line control valves CV-3056 and CV-3027 (Figure 1), are series gate valves. These valves are normally open and are seldom operated.
With the combination of non-corrosive valve materials and SIRW tank chemistry control, no significant mechanism exists for valve degradation and, therefore, the leak rate through these valves can be reasonably expected to be quite small.
Both valves were leak tested during the 1992 refueling outage as part of the valve inservice testing program and exhibited leakage well below the our administratively established acceptance criteria leakage rate of 0.1 gpm (approximately 0.0 and 0.01 gpm) to the SIRW tank, that we are using in our MHA calculation. The acceptance criteria leakage rate of 0.1 gpm was set as the total tank in-leakage limit when the MHA dose limit problem was identified. However, we have completed an interim analysis using a plant specific LOCA source term which justifies that much higher leakage is allowed.
We believe our long term MHA analysis will also allow more leakage if necessary.
A blank flange has been added to the shutdown cooling cross connect line to ensure that no containment sump leakage enters the tank through this path.
Finally, the tank main discharge lines are isolated with a control valve and a check valve in series to prevent leakage back into the tank. This leakage path has also been analyzed to show that due to the elevation differences between the containment sump and the tank, no leakage of containment sump water into the SIRW tank would occur.
As a result of a recent review we have determined that other possible scenarios may exist where post accident containment sump water might migrate to the tank through the main discharge lines. Valve redundancy exists for all these leak paths, however, leak rates have not been quantified for these paths.
- 2.
Compensatory Measures Including Limited Administrative Controls We have administratively set the SIRW tank in-leakage leak rate acceptance criteria at 0.1 gpm based on our original calculations using the FSAR MHA source term.
Testing was completed this refueling outage using R0-119, "Inservice Testing of Engineered Safeguards Valves CV-3027 and CV-3056."
Both valves exhibited leakage well below the acceptance criteria of 0.1 gpm (approximately 0.0 and 0.01 gpm), that we are using in our interim dose calculation.
2 A modification has been completed this refueling outage which will allow installation of a blank flange in the shutdown cooling connecting piping between the SIRW tank and MV-3225.
During power operation this line has been blanked off, eliminating this leakage path to the SIRW tank from the containment sump.
Verification that this blank flange is in place has been added to the shutdown cooling system start-up checklist CL 3.1, "Engineered Safety System Checklist (Shutdown Cooling In Service)."
In our June 14, 1991 submittal we stated that as an interim measure we would route the SIRW tank vent to a charcoal filter.
Our July 17, 1991 letter stated that instead of routing the SIRW tank vent to a charcoal filter, we would modify our procedures to implement a new valve line up that would route any HPSI recirculation valve leakage (CV-3056 and CV-3027), to the south tilt pit in the spent fuel building instead of to the SIRW tank.
Emergency Operating Procedures (EOP) 4.0, "Loss of Coolant Accident" and EOP 9.0, "Functional Recovery Procedure," were revised to require this valve line up to be made as soon as the recirculation actuation signal (RAS) is verified. This action is presently in place.
However, we have completed an interim analysis using a plant specific LOCA source term which shows that the release path in the spent fuel building may not have a significant dose advantage over allowing the release to go into the SIRW Tank.
The credit taken for the charcoal filters in the spent fuel building is less than what is allowed to be taken due to dilution and partitioning if the release were to occur in the SIRW tank.
We will delete this alternate interim leakage path to the spent fuel pool, to minimize post RAS operator actions, if this path offers no comparative dose reductions in our final MHA reanalysis.
By our existing design, hydrazine is automatically added to the
3 containment sprays and safety injection systems coming from the SIRW tank. Then, following RAS, our procedures stated that the results of our.
post accident containment sump sample would be evaluated and the operations department would be told when and how much sodium hydroxide to add to the containment sump to maintain a pH of 7 to 8.
However, since no credit is taken, in the present revision to the Standard Review Plan (SRP), for iodine retention with the use of hydrazine, we have initiated procedure controls to assure that more immediate sump pH control is initiated following a LOCA.
Our Emergency Operating Procedures, EOP 4.0, "Loss of Coolant Accident Recovery," and EDP 9.0, "Functional Recovery Procedure," have been revised to direct the operators to add 200 gallons of sodium hydroxide upon RAS verification. This sodium hydroxide will be injected by the containment sprays and HPSI system to buffer the containment sump to a pH of 7 to 8.
The System Operating Procedure, SOP 4.0, "Containment Spray and Iodine Removal System," for adding the sodium hydroxide is already in place and the other procedure changes assure that this action has been completed as soon as possible. Continued post accident containment sampling and analysis will assure that containment sump pH of 7 to 8 is maintained.
Taking into consideration the containment pressure at RAS and the pressure head of the NaOH tank, the injection of 200 gallons of sodium hydroxide would take less than one hour.
As the NaOH is added, it will be injected into the containment spray and core safety injection water.
This ensures a thorough mixing to achieve the required pH shortly after injection.
Dose reduction to the control room operator by approximately a factor of ten can be accomplished by using potassium iodine tablets to block the intake of radioactive iodine to the thyroid.
Guidance for the use of potassium iodine tablets is provided in the Emergency Implementation Procedure (EI) 9.0, "Off-Site Radiological monitoring." Also, the control room continuous air monitor is posted with an Operator Aid (OA-93) which directs the operator to consider the use of thyroid blocking agents (potassium iodine tablets) to reduce personnel uptake of radioiodine.
The operator aid refers to the emergency implementation procedure for additional guidance.
- 3. Safety Function and Events Protected Against The event to protect against is the Maximum Hypothetical Accident (MHA) release resulting i_n a dose at the site boundary or to the control room operators.
The SIRW tank provides the initial volume of borated water needed for core injection following a LOCA.
Following injection, the tank is isolated and the containment spray and safety injection pumps recirculate containment sump water into the containment atmosphere and through the core.
Three paths have been identified to the SIRW tank that could potentially
4 provide a leak path to put contaminated post accident containment sump water into the SIRW tank. This contamination of the tank could result in a previously unaccounted for source term that could contribute to both off-site and control room doses.
The safety function of not allowing unanalyzed amounts of water into the SIRW tank is met as follows:
A.
The SIRW tank main discharge lines provide the initial suction source for containment spray, high pressure and low pressure safety injection. Analysis has shown that, due to the difference in elevation between the containment sump and the SIRW tank, no post accident containment sump water would be forced back into the SIRW tank due to the post LOCA containment pressure.
The elevation difference between the minimum SIRW tank water level (1.6 feet) and the maximum containment sump level (6.4 feet) after recirculation begins is 54.03 feet.
The SIRW tank elevation is 643 feet 10 inches and the containment sump elevation is 585 feet.
Using a reference temperature of 120°F, we calculated that a containment pressure of 37.85 psia would be required to force flow from the sump toward the SIRW tank.
The peak containment pressure in the current FSAR analysis following RAS (no containment air coolers) is about 23 psig (37.7 psia). Therefore, the containment pressure will not push sump water into the SIRW tank.
The margin between the pressure needed to force water back to the SIRW tank and that calculated as a peak LOCA pressure appears to be small.
However, the containment pressure analysis contains the following conservatisms which contribute to this high peak pressure.
The containment analysis uses a bounding service water temperature, conservative service water and component cooling water system flow rates, a design heat exchanger fouling factor, and a conservative core decay heat correlation. Additionally, if the three containment air coolers are operating, the maximum containment pressure following RAS is less than 14 psig. Also, elevated containment pressure on the order of 37 psia will exist for only a short period of time (-30 minutes). Considering the above results and these conservatisms, we conclude the dose consequence analysis need not include leakage of containment sump water into the SIRW tank, due to containment pressure.
B.
Manual valve MV-3225 isolates the discharge of the shutdown cooling heat exchangers to the SIRW tank.
This discharge line can be used for transferring water between systems and has no post accident safety function. A modification was completed during the 1992 refueling outage to add a flanged connection between the SIRW tank and MV-3225 so that a blank flange can be installed prior to power operation. This assures no post accident contaminated containment sump water will enter the SIRW tank from this source.
C.
Control valves CV-3056 and CV-3027 provide isolation of the safety injection recirculation line to the SIRW tank. This line is needed for pump protection and it provides a path for pump discharge (recirculation) during periodic testing. Administrative leakage
limits have been established for these valves.
The individual valve leak rates were measured during the 1992 refueling outage (approximately 0.0 and 0.01 gpm), to assure that the leakage rate assumed in the revised MHA analysis (0.1 gpm) was not exceeded.
As a result of the above actions, we have verified that the safety function of not allowing an un-analyzed amount of containment sump water into the SIRW tank, is assured.
5 As a result of a recent review, we have determined that other possible scenarios may exist where post accident containment sump water might migrate to the SIRW tank through the main discharge lines. These other possible scenarios involve the assumption of a unique set of equipment failures including leakage through several valves in series. These valves are presently not leak tested.
For the interim we have calculated post accident doses using a plant specific LOCA analysis source term.
Administratively we have set a 0.1 gpm total SIRW tank in-leakage limit.
We set this limit based on our MHA FSAR source terms and the need to keep containment sump water out of the SIRW tank.
Analysis has been completed which shows that leakage back into the tank on the order of 20 gpm for 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> is acceptable.
Based on this analysis, we are confident that higher interim leak rates could be justified if necessary.
On an interim basis, using the plant specific LOCA analysis the effect of leakage into the SIRW tank is not a significant contributor to control room operator or off-site doses.
As part of our long term solution, we will verify the validity of the possible scenarios and include their impact on our final analysis and design as required.
Iodine Removal Systems are designed, in general, such that the pH of the post accident containment sump is maintained from 7 to 8.
This assures that the maximum amount of iodine will be retained in the containment sump water.
Studies have shown that lower pH (i.e., less than 7) sump water allows iodine to re-evolve to the atmosphere over time.
This contributes to containment atmosphere activity, a percentage of which is required to be calculated as a release from the containment.
This release contributes to both the off-site dose and control room operator dose.
The higher the containment atmosphere activity, the higher the containment dose release component.
Presently Palisades is designed for hydrazine injection, along with the injection of the SIRW tank water, to help in iodine scrubbing from the containment atmosphere.
As the accident progresses the plant chemistry department staff is tasked to monitor the post accident containment sump water and recommend to the operations staff to add sodium hydroxide as necessary to provide long term sump pH control. Since hydrazine is no longer considered effective for iodine retention, we have modified our emergency operating procedures to add a calculated amount of sodium hydroxide (200 gallons) via the engineered safeguards system to the containment sump, beginning immediately after recirculation has been verified. This amount of sodium hydroxide will buffer the containment sump to a pH of 7 to 8.
6 The control room ventilation system is designed to maintain a habitable environment for the operators through a post accident plant condition.
One system function is to minimize the introduction of radioisotopes such that the operator's exposures are kept below GDC-19 limits.
Upon our initial discovery of the additional source term from post accident leakage into the SIRW tank, we determined that the GDC-19 limits for control room operators exposure may be exceeded.
In our February 5, 1992 telecon with the NRC, an agreement was made to use a plant specific LOCA analysis to predict doses to provide a bases for a justification for continued operation.
We are using the plant specific LOCA source terms to predict doses on an interim basis until we can complete a revised MHA analysis and adjust plant configuration, as necessary.
Based on the results of the plant specific LOCA analysis, the control room ventilation system provides the required protection for the control room operators.
- 4.
Conservatism and Margins The MHA analysis is not a realistic prediction of the impact of a major LOCA.
It is, instead, an extremely conservative bounding calculation.
As stated in Palisades FSAR Section 14.22.1, "The Maximum Hypothetical Accident (MHA) is postulated to release substantially more fission products and result in more severe consequences than any incident considered credible.
The evaluation is only meant to determine a reasonable upper bound of the consequences of an incident involving the release of radioactive material from the plant site.
The radiological corisequences of the MHA are determined in a manner independent of any specific plant transient sequence that might be postulated.
To this end, the evaluation is performed in accordance with the guidelines and recommendations put forth by the NRC staff. These guidelines are nonmechanistic in nature and are meant to maximize the consequences of the MHA."
FSAR Section 14.22 discusses the current MHA analysis for Palisades.
The site boundary and low population zone thyroid doses calculated in this analysis are 69.72 rem and 39.63 rem respectively (as compared to the 10 CFR 100 limit of 300 rem).
Use of International Commission on Radiological Protection (ICRP) 30 vice ICRP 2 dose conversion factors, would lower these results even further.
Because of the large margin between the limits and our present calculations, off-site exposures from the MHA are judged not to be of concern.
Two sources of conservatism in the MHA analysis are of particular note because they directly impact calculated control room doses.
First, the MHA analysis assumes instantaneous release of fission products from the fuel at the time of the LOCA.
The total core inventory of iodine and noble gasses is also assumed to be released to containment. Second, the MHA analysis assumes that containment leakage is at a full 0.1 weight percent (w/o) per day for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and 0.05 w/o per day thereafter for 30 days, regardless of containment pressure.
In reality, the leakage would be significantly less, especially in later stages of the accident due to
the reduced containment pressure.
The control room habitability analysis uses the same MHA releases that contribute to off-site dose to calculate a theoretical dose to control room operators.
One extremely conservative assumption in this calculation has the operator stationed in the control room for 100% of the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, 14.4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> per day for the next three days, and 10.4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> per day for the next 26-days.
It also assumes that no protective measures such as use of respirators and potassium iodine are taken.
Actual exposure of control room operators to whatever control room activity might be present can be expected to be considerably less, if only because there are five operating crews which would be on rotating shifts during this period.
7 As an interim analysis, the source term from a plant specific LOCA fuel failure analysis will be used.
A LOCA analysis is already per,formed to show conformance to 10 CFR 50, Appendix K, which justifies that fuel temperatures will not be high enough for fuel melting to occur. After discussing with our fuel vendor the benefit of calculating the percentage of the cladding that would be predicted to fail for this scenario, it was decided to assume 100% core cladding failure.
An analysis was then performed for the off-site and control room doses, assuming 100% core cladding failure with 20% of the core iodine and noble gas inventory in the pellet-clad gap.
The interim analysis included the administratively limited 0.1 gpm total SIRW tank in-leakage through CV-3027 and CV-3056, that was routed to the spent fuel pool tilt pit as directed by the current operating emergency operating procedures.
Most of the parameter assumptions, except the source term and valve leak rate, were discussed with the NRC staff during our meeting on January 15, 1992.
The source term and valve leak rate assumptions were discussed with the NRC in later conference calls.
The interim calculated doses are shown below.
Site Boundary Low Population zone Control room Total Interim Doses Thyroid Whole Body 7.645rem 4.504rem 8.410rem 0.057rem O.Ollrem 0.45lrem Limits Thyroid Whole Body 300rem1 300rem1 30rem2 25rem1 25rem1 5rem2 The interim calculated dose values are well within established limits.
1 10 CFR 100 2 GDC-19/Standard Review Plan 6.4
- 5.
Probability and PRA Reviews Operating the facility as described in our justification for continued operation would have no impact on the current PRA results since the failure, as postulated, would not cause a failure of the injection function and, therefore, it would have no direct affect of the frequency of core melt.
8