«Item 7b Severe Accidents Related Issues Preliminary Monitoring Report Report to the Federal Ministry of Agriculture, Forestry, Environment and Water ...»
In the operating NPPs the means available for RCS depressurization are not as diverse and reliable as those required of future NPPs. As long as there is water injection to the secondary side of steam generators, opening relief valves on the secondary side is an effective method of RCS depressurization. However, under severe accident conditions loss of the main feed water system, auxiliary feed water system and even emergency feed water system has to be postulated. Under such conditions forced opening of the secondary side relief valve to lower the secondary side pressure has been shown in PN7 MELCOR calculations to bring only temporary RCS pressure drop. In the long run this strategy can result in faster loss of secondary water inventory and faster core uncovery. It is recognized, that secondary side depressurization without availability of sources of water to inject to SGs is not sufficient per se to depressurize the RCS under severe accident conditions.
In the aftermath of TMI-2 accident, in which the presence of non-condensable gases hampered coolant circulation in the RCS, the requirement to install high-point vents was imposed in US plants. In WWER 1000 NPPs of 320 type the high vent points are installed on the top of the RPV, of the pressuriser and of each of the primary collectors in the steam generators.
Since after gas removal to the pressuriser bubble tank and eventual burst of the bubble tank membrane the gases are released to the containment, the vents, which avoid that, are preferred for use as a means of depressurization of the RCS. The effectiveness of the system depends on the flow capacity.
TACIS calculations for WWER 1000 NPP showed that in case of maximum mass flow rate of 20 kg/s the Emergency Gas Removal System is not enough to assure reliable RCS depressurization under all severe accident conditions, while at 30 kg/s the RCS can be depressurized reliably before RPV failure [TACIS 02]. In Kozloduy NPP the tests at low-pressure difference and their extrapolation to high pressures suggest that the maximum EGR flow rate will be below 30 kg/s. Thus the question of reliability and capacity of EGRS lines is important.
If the EGRS is not sufficient to bring the plant to stable steady state conditions then the opening of PORV is necessary. In some WWER 1000 NPPs (e.g. in Balakovo) PORV cannot be opened below the operational RCS pressure [Morozov 03], but in most Western and Eastern NPPs PORV can be remotely opened at any pressure. The requirements regarding the reliable opening of PORV at low pressures are very high. In some NPPs that also have PORVs capable of being opened at low pressures (e.g. in Loviisa NPP), special dedicated depressurization systems have been designed and built.
In the process of depressurization the lines and valves of the secondary side of SGs, EGRS and PORV would be operated with water-steam mixture and solid water flows, while their design basis involves only steam flow. High mechanical stresses connected with water flows in those lines can result in their failure. In the case of the line leading to PORV this would correspond to SB LOCA and loss of control over the PORV opening and closing. In the case of the valves on the secondary side of SGs the consequences would be worse, because in case of PRISE a bypass of containment would result. Therefore, qualification of these lines for steam-water and water flows is necessary.
Current plant status Temelín NPP realizes importance of depressurization. It is on a leading place among its SAMG strategies. The main technical means to achieve depressurization is to use the power operated relief valve (PORV) on the pressuriser or the emergency gas removal system (EGRS).
PORV is designed so that it can be remotely opened at low pressures and kept open. The minimum overpressure needed for its opening is 0,5 MPa. The design with an electromagETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues netic valve used as the pilot valve ensures reliable opening whenever the operator cuts the current to the electromagnetic valve. The loss of current to the pilot electromagnetic valve results in its opening; it is a passive arrangement, not subject to any failure of active elements.
Once the electric current is lost, the pressure difference will open the valve. The value of 0,5 MPa overpressure is needed to overcome the force of a spring. Since the RCS pressure should be brought to about 1,0÷1,5 MPa in order to avoid HPME and DCH, the PORV can be maintained open throughout the whole period of RCs depressurization. Thus the opening is sufficiently reliable to fulfil the requirement of TSO that “the opening of PORV at low pressures should be as reliable as opening of a safety valve at high pressures” [TSO 01].
The lines leading to PORV have been checked to sustain dynamic loads due to possible steam-water mixture flows. The calculations made by ÚJV Řež for Temelín with RELAP code show that the opening of PORV is enough to assure timely depressurization of the RCS. The analyses performed within PN7 with MELCOR code confirm that. It should be noted that feed and bleed strategy is part of new symptom based EOPs and the calculations of depressurization with PORV were competently checked within the related process of EOP verification and validation.
The second line of defence is using EGRS. This system is designed to remove gases that can be accumulated in the high points of the RCS under accident conditions, in particular in the PRZ, RPV and primary collectors in the SGs. It can be also used for coolant removal from the RCS. According to Temelín staff, the calculations of Energoproekt Prague showed that EGRS is sufficient for RCS depressurization. This statement is based on calculations performed using RELAP code, which is a recognized tool for thermal hydraulic evaluations.
On the other hand, according to PN7 calculations, and also according to Russian calculations within TACIS, the EGRS capacity is just at the limit needed for successful depressurization. If this system were the only means of pressure reduction in Temelín, the matter would need further consideration, but since the primary means of action is opening the PORV, this is sufficient.
In addition, Temelín NPP stated that the calculations of all scenarios would be repeated this year, using MELCOR 1.8.5 (so far only MELCOR 1.8.3 was used). After that Czech experts will review the question of effectiveness of depressurization again.
If the depressurization is not achieved fast enough, there might be a danger of consequential ruptures of the primary system pressure boundary within the steam generators with resulting PRISE and containment by-pass. The prevention of this failure is important for successful SAM. No information was obtained on the question of weakening of steel in high temperatures and possible thermal shock on SG tubing after EFWS recovery. However, the SAMG stress the necessity to keep the SG tubes covered with water, so that the conditional probability of accident induced tube break is maximally reduced. The formation of water seals was considered in the RCS model used in MELCOR. No information was obtained on the model of flow inside the core, the heat transfer coefficients used during fuel candling, corium stratification and the RPV weakening with high temperatures. The Czech side provided information about the special model of RPV bottom rupture that had been used in MELCOR calculations to reflect special WWER geometry.
It is worth to be noted that all these parameters will be determined again using the latest MELCOR code version 1.8.5 in the calculations to be performed by the end of 2003. This version of MELCOR has already been implemented in ÚJV Řež and is ready for use once the order of Temelín NPP is finalized. In version MELCOR 1.8.5 the features characteristic for WWER 1000 geometry can be fully modelled, including mechanistic determination of the bottom head rupture of the RPV.
ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues 113 Evaluation The measures available in the plant are sufficient for timely depressurization of RCS. Temelín NPP has two lines of defence in this respect (PORV and EGRS), which is better than in many other NPPs of similar vintage. The WOG SAM strategies being implemented in the plant recognize the importance of depressurization.
The ETE-PSA estimate is that the probability of the PORV to fail is 2%, which is equivalent to the overall failure probability of the depressurization system.
The Specialist’s Team would recommend the Austrian Governement to consider monitoring he calculations to be made by Temelín NPP with MELCOR 1.8.5 to see confirmation of capability of PORV to depressurize reliably the RCS. If its reliability at low pressure is lower than that announced during the Prague meeting, the credit taken for EGRS operation should be checked.
5.4 Containment Failure Prevention 5.4.1 DCH and other early threats to containment integrity VLI No. VLI title / description 8.5.1 Has DCH been considered?
What was the amount of corium assumed to be dispersed in the containment?
8.5.2 Is reactor cavity well connected with the containment, or do the ventilation channels provide only limited flow cross section area to release steam from the reactor cavity to the containment after RPV failure?
8.5.3 After high pressure RPV failure, has the impact of molten corium against containment liner been considered and possible damages to the liner taken into account in determining containment leak tightness?
8.5.4 Has the impact of short-time high temperatures in the containment been considered while predicting that the containment integrity will be kept? Has a stress-temperature analysis been performed for ventilation system valves providing containment boundary?
8.5.5 Are there any reliable means of opening the door to the reactor cavity to facilitate corium spreading?
8.5.6 Has the danger been analyzed of the door being blown out by overpressure in the reactor cavity and hitting the containment wall?
8.1.10 What assessment has been made concerning the continued integrity of the spent fuel cask transfer hatch at elevation of the area outside the reactor cavity?
8.7.5 What is the physical and experimental basis for the conclusion that parallel hydrogen burning and DCH are not predicted for Temelín in case of HPME/DCH due to vessel failure at high pressure in a severe accident?
State-of-the-art requirements and practices The worldwide-accepted state of the art practice is to demonstrate that the likelihood of early RPV-failure due to HPME as well as DCH as a consequence is very low.
Current plant status Once Temelín has implemented the planned depressurization systems and procedures the plant status will be sufficient. Furthermore, calculations indicate that even in the case of HPME and subsequent DCH the containment failure pressure is not reached.
114 ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues For the cavity door opening procedure it should be mentioned that, if a pressure scenario results in corium ejection and hot gases flow through that failed door opening the containment liner on the wall opposite the door opening can be damaged. It is also possible that the molten corium will be distributed non-uniformly causing non-uniform rates of penetration through the concrete.
Another hazard connected with HPME is due to the presence of a large metal hatch (opened during refuelling to transfer spent fuel casks to a rail car at elevation +0,0m) in the floor outside the reactor cavity, which serves as part of the containment boundary during normal operation. If HPME occurs, core debris could be deposited adjacent to and onto the hatch, degrade the seal between the hatch and the containment floor, or degrade the hatch itself, which might fail due to melting and/or a combination of thermal degradation and the ambient containment pressure.
Failure of the hatch cover would allow release of radioactive gases (as well as combustible gases such as hydrogen and carbon monoxide) and aerosols (and possibly core debris) outside the containment into the reactor building, which is not a hermetic structure. Czech side acknowledged these concerns during the Prague workshop and stated that it is planned to install removable walls that will prevent ejected corium from flowing outside a limited area, and in particular protect the containment liner and the spent fuel transfer hatch from contact with the molten corium.
Evaluation Similarly to other NPPs with large dry containment, the hazards of early containment failure due to DCH in Temelín NPP have been evaluated as negligible and the strategy of depressurization included in SAM in Temelín reduces further such hazards. On the other hand, several technical measures are planned to be implemented in the plant, but have not been installed and no details of technical design have been disclosed.
Further monitoring is needed for the implementation of the mechanism for early opening of the steel door between the reactor cavity and the neighbouring equipment room and the installation of steel walls preventing ejected corium from flowing outside the designed area.
For monitoring the additional measures related to DCH prevention or mitigation, information is needed about the design requirements and the installation for the flow-deflecting wall (e.g.
ability to withstand long term heat, flow impact and diversion under reduced pressure melt ejection etc. are of interest).
ETE Road Map - Preliminary Monitoring Report – Item 7b: Severe Accidents Related Issues 115
5.4.2 Basemat penetration
VLI No. VLI title / description 8.6.1 Is it possible to flood effectively with ECCS water or with other water sources the molten corium pool in the reactor cavity or after spreading into the instrument room?
8.6.2 Can the reaction of molten corium with concrete be stopped before complete basemat penetration? Would power recovery be enough for it?
8.6.3 What is the minimum thickness of the basemat necessary to keep the weight of molten corium and the pressure in the containment?
8.6.4 Are there analyses to show the consequences of complete basemat penetration including possible hydrogen burning in the rooms below the basemat and fission product releases to environment?
8.6.5 How was the result reached quoted by [Sykora 01b], that with late ECCS recovery the depth of concrete ablation after 300 000 seconds is only 1.4 m?
What mechanisms of water-cooling of corium were considered?