Improvement of PACTEL Facility Simulation Environment (PACSIM)39

The main objectives of the PACSIM project is to enhance the utilization of the TRACE simulation environment of the PACTEL facility with TRACE thermal hydraulic code. The Finnish Radiation and Nuclear Safety Authority, STUK, has required an independent tool to support safety and licensing analysis and decided to use the TRACE code. The use of TRACE enhances the preparedness to give analysis support and improves education in computational thermal hydraulics. During the year 2008 in the PACSIM project TRACE has been used for preparation of complete three-loop PACTEL facility model with horizontal steam generators.

The project has given important validation knowledge for achieving the final goal of the full-

scale VVER-440 model preparation, which will be carried out outside the SAFIR2010

programme. Another objective in the project has been preparation of a new TRACE-model

with vertical steam generators, simulating the new PWR-PACTEL facility set-up. This model

has been completed during first year of the project. The planned TRACE-code calculations

will give valuable analysis and comparison support for the APROS calculations and the PWR-PACTEL experiments.

Specific goals in 2008

Specific goals in 2008 contained preparation of complete TRACE input decks of the PACTEL facility both for VVER and PWR PACTEL facilities. Also, increased practical experience in the using the TRACE code was achieved. Fully functional TRACE code model of the PACTEL VVER facility with three loops, auxiliary and control systems was prepared.

The functionality of the both model was tested and first calculations were carried out in order to validate the model of the PACTEL VVER facility. The model of VVER PACTEL was validated against pressure loss test FLT01 (Runs 4, 5, 6, 8, 11, 14 and 16) and heat loss tests HL22-23. The calculated results agreed well with the experiment data in most cases.

Another main goal was to prepare a new TRACE-model with vertical steam generators for the modified PWR-PACTEL facility. At first, the vertical steam generator models were built separately and tested with inlet and outlet boundary conditions. The existing TRACE model of PACTEL was then modified and connected to the vertical steam generators performing the entire PWR-PACTEL model. Due to lack of experiment data the calculations of the characterizing tests were postponed to future and blind pre-calculations of LOF-tests were loss-of-feedwater made. The calculating results were promising.

Deliverables in 2008

• TRACE model of VVER PACTEL facility.

• Calculation of PACTEL pressure losses

• Calculation of PACTEL heat losses with TRACE code

• TRACE model of PWR PACTEL facility.

Figure 1. Main view of the VVER PACTEL facility model by SNAP editor.

Figure 2. Main view of the PWR-PACTEL facility model by SNAP editor.

2.4.5 Condensation experiments with PPOOLEX facility (CONDEX)

The objective of the project is to improve understanding and increase fidelity in quantification of different phenomena in boiling water reactor (BWR) containment during steam discharge.

These phenomena could be connected, for example, to bubble dynamics, direct-contact condensation (DCC), pressure oscillations, thermal stratification and global circulation and mixing in the pool. To achieve the project objectives, a combined experimental/analytical/computational study programme is being carried out. Experimental part (LUT: CONDEX) of the project is responsible for the development of a database on condensation pool dynamics and heat transfer at well controlled conditions.

Analytical/computational part (VTT: NUMPOOL, KTH: NORTHNET RM3, LUT:

CONDEX) use the developed experimental database for the improvement and validation of

models and numerical methods including CFD and system codes. Also analytical support is

provided for the experimental part by pre- and post-calculations of the experiments. The

PPOOLEX test facility, including both the dry well and wet well compartments of the

containment and withstanding prototypical system pressures, is utilized in the experimental

part of the project.

Specific goals in 2008

Specific goals in 2008 included three different experiment series with the PPOOLEX test facility. In the thermal stratification and mixing experiments the objective was to study thermohydraulic loading of the wet well structures due to stratification processes as well as to get comparison data for evaluating the capability of GOTHIC code to predict stratification and mixing phenomena. An array of properly positioned thermocouples was added to the pool volume in order to measure accurately the characteristics of thermal stratification and mixing.

With small steam flow rates the stratification process started almost immediately after the initiation of the experiment. With higher flow rates the mixing effect of steam discharge delayed the start of the stratification process until the pool bulk temperature exceeded 50 °C (Figure 1)

20 30 40 50 60 70 80 90

0 500 1000 1500 2000 2500 3000 3500

Temperature [°C]

Time [s]

STR-05: Vertical temperature distribution in wetwell water T501

T502 T503 T504 T505 T506 T507 T508 T509 T510 T511 T512 T513 T514 T515 T516

Figure 1. Due to mixing effects the thermal stratification process started only after the pool water had warmed up when higher flow rates were used.

The wall condensation experiments aimed at estimating the amount of condensate generated

with different flow rates and pre-heating levels of the dry well structures and at producing

verification data for CFD calculations. A system for collecting and measuring the amount of

condensate from four different wall segments of the dry well compartment was installed. The

accumulation of condensate was strongly controlled by the temperature level of the dry well

structures. As the dry well structural temperatures increased the condensation process slowed

down. However, the condensation process never completely stopped because a small

temperature difference remained between the dry well atmosphere and inner wall even in the

case of an extended steam discharge period (Figure 2).

90 95 100 105 110 115 120 125 130 135 140

0 100 200 300 400 500 600 700 800

2 4 6 8 10 12 14 16 18 20

Temperature [°C] Mass [kg]

Time [s]

WLL-02-6

T2107 Drywell middle T2108 Drywell bottom T2117 Inside wall T2119 Inside wall Condensate

Figure 2. Temperatures in the atmosphere and on the inside wall of the dry well compartment and accumulation of condensate in a wall condensation experiment.

Specific experiments on the effect of a scaled down collar installed at the lower end of the blowdown pipe on the pressure oscillations occurring inside the pipe and in the pool during the chugging phase of steam discharge were carried out (Figure 3). Different flow rates and pool water temperatures were used. Pressure oscillations were captured with kHz range measurements and filmed with high-speed cameras.

Figure 3. A scaled down collar was attached to the outlet of the blowdown pipe for the third experiment series in 2008.

Work with NEPTUNE_CFD code, developed in the EU/NURESIM project, continued with

the simulation of selected POOLEX experiments. An experiment with thermally insulated

blowdown pipe and a quasi-steady steam-water interface at the pipe mouth were simulated

with 2D and 3D grids. Furthermore, simulation of a chugging experiment was started. The

models of Hughes-Duffey and Lakehal et al. 2008 were compared.

Deliverables in 2008

• A series of thermal stratification and mixing experiments was carried out with the PPOOLEX test facility to study thermohydraulic loading of the wet well structures and to get comparison data for GOTHIC code. Heat-up periods of several thousand seconds by steam injection into the dry well compartment and from there into the wet well water pool were recorded. The initial water bulk temperature was 20 °C. Cooling periods of several days were included in three experiments. With higher flow rates the mixing effect of steam discharge delayed the start of stratification until the pool bulk temperature exceeded 50 °C.

• Five wall condensation experiments, each consisting of several steam discharge tests, were carried out. Steam was blown into the dry well compartment and from there through a DN200 (Ø219.1x2.5) blowdown pipe down to the condensation pool filled with 20 °C water to the level of 2.14 m i.e. the blowdown pipe outlet was submerged by 1.05 m. The steam flow rate ranged from 90 to 690 g/s and the temperature of incoming steam from 115 to 160 °C. Pre-heating of the dry well structures was used in all but two tests. A thermo graphic camera was used in a couple of experiments for evaluating the temperature distribution on the outside surface of the dry well wall. Most of the condensate accumulated during the first 200 seconds of the discharge. However, the condensation process never terminated, because a temperature difference of a few degrees remained between the atmosphere and the inner wall of the dry well as long as steam was blown into the pool.

• A series of chugging experiments to investigate the effect of a modified blowdown pipe outlet on the level of pressure oscillations was carried out. A scaled down collar design was attached to the outlet of the blowdown pipe. Reference experiments were done with a straight pipe. Pressure loads inside the blowdown pipe and in the condensation pool were registered with the help of extra pressure transducers. Strains on the pool bottom and vertical movement and acceleration of the vessel were also measured.

• Test STB-28-4 from the preceding POOLEX experiments was selected to be modelled and calculated with NEPTUNE_CFD code Work and model improvements done in the EU/NURESIM project on steam condensation inside a vertical blowdown pipe formed the basis for the simulations. With the help of system codes (APROS and TRACE) the missing boundary conditions were calculated. The condensation models of Hughes- Duffey and Lakehal et al. 2008 were compared using pressure and velocity inlet boundary conditions at the blowdown pipe inlet. The initial results indicate that the Hughes-Duffey condensation model predicts clearly higher condensation rates than the model of Lakehal et al. 2008.

• NORTHNET RM3 meetings were participated and a POOL-NKS meeting arranged in Lappeenranta. A combined POOL-NKS application for 2009 by LUT, VTT and KTH was written and delivered.

• XCFD4NRS workshop in Grenoble and a training session on TransAT in Zurich were

participated.