Monitoring water levels in VVER pressure vessels28 February 2001
Before the Three Mile Island accident, the water level inside the pressuriser was used to monitor the water inventory inside the reactor pressure vessel. The TMI accident demonstrated that this was not a reliable method for determining the coolant levels of PWRs and direct measurement inside the vessel was required. This is now also peformed at VVERs.
Together with INKOR of Moscow, Siemens has developed a new level- measurement probe for VVER reactors that operates in a redundant mode independently of the water level measurement inside the pressuriser. The system qualifies to the international standard IEC 323 and RG 1.97 as part of the post accident monitoring system (PAMS). The assemblies are designed and manufactured according to ASME code guidelines to assure structural integrity of the primary pressure boundary.
First backfittings of the new level measurement system in the VVER-440s in Bohunice V1 (Slovakia), unit 1 (1998) and unit 2 (2000), Novovoronezh (Russia), unit 4 (1999) and Kola (Russia), unit 1 and unit 2 (1999) show very good operational results.
The level measurement system was designed according to the following criteria:
•Direct measurement of water inventory inside the RPV under all operational conditions, from start-up to full power.
•Optimal spatial distribution of measuring points to indicate the formation of bubbles under the RPV head, water-filling of the loops, and whether the core is covered with fluid or “swell level”.
•Optimal signalisation, featuring: response time under 30 seconds; a high signal level for reliable processing; response before the core is uncovered; axial resolution of the sensor of +/- 50mm.
•High structural stability and flexibility by using components with proven operating experience.
•Use of LOCA and seismically proven components.
•No additional penetrations in the RPV.
•Integration of level monitoring sensors in neutron flux monitoring probes.
At least two redundant monitoring systems are recommended, each consisting of:
•One KNITU or KITU level probe.
•One electronic cabinet equipped with suitable power supplies for the primary transducers, electronic modules, measuring amplifiers, signal conditioning modules, limit value modules, logic modules, alarm signal modules and one six-line recorder.
•One heating module.
•A communication line (loop cable) between the electrical connector at the probe head and the penetration point of the reactor’s concrete pit.
•Fasteners at the vessel head.
•Communication lines from level probes to the electronic cabinet, and from there to the display panels in the emergency control room (ECR) and the main control room (MCR).
•Display panels in the MCR and ECR.
Level monitoring probe
The KNITU and KITU level probes (see diagram) use the same housing tubes as the KNI probes which are already used for the in-core neutron flux monitoring system SVRK. KNITU or KITU probes can therefore be installed in the same guide tubes without any additional mechanical adaption, and can be installed and removed in the same way as the KNI probes (see illustrations).
The housing tube contains the level probe in the upper part. For the KNITU probes, the lower part of the tube contains the self powered neutron flux detectors.
The probe head is equipped with a Pt-100 resistor to measure the temperature of the cold junction of the thermocouples.
In KNITU assemblies the level probe contains five sensors: three heated thermocouples with heating filaments and two unheated thermocouples. The KITU assemblies can have up to ten thermocouples, eight heated and two unheated. The signals from the unheated thermocouples are used as references which are independent of the water or steam temperature. The measuring principle relies on the fact that heat transfer in water is considerably higher than in steam. A heated thermocouple immersed in water has a lower temperature and therefore a lower thermovoltage than when it is surrounded by gas or steam. The thermovoltage of the thermocouple is therefore a measure of whether the thermocouple is immersed in water or steam. All thermocouples in the system are designed for a temperature of up to 1260°C according to IEC 323 and RG 1.97.
The coolant level is monitored by thresholds. If the voltage difference between the thermocouples falls below a certain threshold, this would indicate that the water level is above the heated thermocouple. On the other hand, should the voltage difference be above a certain threshold, this would mean that the water level has fallen below that thermocouple.
NiCr-Ni thermocouples are used in the assemblies. The heating filaments are made from a combination of NiCr and Ni. Optimal heat transfer is achieved by directly attaching the heated and unheated thermocouples to the housing tube of the assembly. Because the existing guide tubes for the level probes inside the RPV are open, the probes are moistened directly by the primary coolant. This ensures direct heat transfer from the coolant to the thermocouples, and the coolant level is therefore detected without distortion.
Recommended measuring positions
The KNITU/KITU level probes should be installed into homologous measuring positions in the core. The criteria of redundancy have to be taken into account for these positions.
As pressure decreases the steam will accumulate in the space below the RPV closure head, displacing the water there. Should this happen, the water level will drop relatively quickly, although the total water inventory of the primary cooling system will decrease only slightly. If the level decreases so far that the steam forces its way into the reactor coolant pipes, the coolant pipes and steam generators begin to empty. Due to the large water inventory there, a small decline in the water level in the upper plenum indicates a much stronger reduction in the water inventory of the primary cooling system. Once the steam generators and the reactor coolant pipes have drained, the water level in the upper plenum will decrease at a faster rate. Should this happen the remaining water inventory will be very low.
The heated thermocouples should therefore be positioned such that there will be enough time to initiate countermeasures before the decline reaches the top of the core, and the corresponding voltage thresholds will be crossed before the water level is below the outlet nozzle of the coolant pipe.
The recommended positioning of the measuring points and the assignment of the reference points is as follows (see diagram).
Measuring point 1 is used to detect the development of steam bubbles below the RPV closure head. Reference point 1 is located closely below measuring point 1 in order to receive a clear measuring signal, even in case of a reactor scram. During a scram the temperature at the fuel element outlet decreases. If the fuel element outlet temperature were used as the reference temperature an incorrect measurement of the level could result.
Measuring point 2 is used to indicate whether the outlet pipe of the loop is sufficiently filled with water and thus whether natural convection for reactor cooling is possible. The lower unheated thermocouple, reference point 2, is located at the same height as that of the fuel element outlet. Reference point 2 can also be used to measure the fuel outlet temperature as part of the core exit temperature measurement system.
Measuring point 3 is used to supervise the monitoring of the water inventory of the core in case of a leakage in the loop inlet pipe. The measuring point is 60mm below the lower edge of the loop inlet nozzle taking into account the tolerances for fabrication, measuring, and positioning. The fuel element outlet temperature, reference point 2, is also used as the reference point.
The signals from the heated and unheated thermocouples (see diagram) are conditioned by temperature transducers and then corrected with the temperature from the cold junction measured by a Pt 100 resistor in the assembly head. The corrected signals are transmitted to difference amplifiers where the difference temperatures as a measure of water level are formed by subtracting the values of the unheated from the heated thermocouples. The water level inside the RPV is monitored by limit values of the difference temperatures.
The heating current applied to the heated thermocouples is produced by the power supply unit for the heating filament. The heating power is controlled in such a way that it is independent from the surrounding temperature inside the RPV.
One electronic cabinet has to be provided for each redundant level probe. The cabinet contains the necessary electronic modules for signal conditioning, adjustment, limit values and heater power control. The cabinet also has to be equipped with logic modules to perform the algorithms for the generation of the level signals and the equipment fault alarms, a six-line recorder, terminals for input signals, output signals and power supply.