Undesirable Flow Behavior in a Proposed Validation Data Set

Abstract

The next generation nuclear plant (NGNP), whose development is supported by the U. S. Department of Energy, will be a very high temperature reactor (VHTR). The VHTR is a single-phase helium-cooled reactor that will provide helium at up to 800 °C. The prospect of a coolant at these temperatures circulating in the reactor vessel demands that careful analysis be performed to ensure that excessively hot spots are not created and that sufficient mixing of the coolant is obtained. Computational fluid dynamics (CFD) coupled with heat transfer will be used to perform the desired analyses. However, primarily because of the imperfect nature of modeling turbulent flow, any CFD calculations used to perform nuclear reactor safety analysis must be validated against experimental data. Experimental data have been taken in a scaled section of the lower plenum of a prismatic VHTR at the matched index of refraction (MIR) facility at the Idaho National Laboratory. These data were taken with the intent that they be examined for use as validation data. A series of investigations have been conducted to assess the MIR data. Issues that have already been examined include the extent of the required computational domain, the outlet boundary condition, the inlet data and the effect of the turbulence model. One of the jets that flow into the model impacts on a wedge, which represents a portion of a hexagonal graphite block that is part of the inner wall of the lower plenum. The nature of the flow below this particular jet is such that a randomly varying recirculation zone is created. This recirculation zone is seen to change in size, causing a relatively long-time scale of motion or disturbance on the flow downstream. It is concluded that such a feature is undesirable in a validation data set, firstly because of its apparent random nature and, secondly, because to obtain an appropriate longtime average would be impractical because of the compute time required. It is found that by eliminating the first of the four inlet jets into the scaled model, the resulting recirculation zone is rendered stable.

CFD Simulation of Proposed Validation Data for a Flow Problem Reconfigured to Eliminate an Undesirable Flow Instability

“…Improved results were obtained for these inlet conditions, as reported in Ref. [4]. However, it was also discovered that an undesirable flow condition was present in the flow of the scaled model [4].…”

“…Figure 7 plots the streamlines for the original four-jet case and for the new three-jet case for the SKW model. As can be seen for the four-jet case, the recirculation zone below Jet 1 flows back and impinges on itself with some energy, causing the separation point to move violently up and down as plots at other times show [4]. For the three-jet case, the recirculation zone occupies the whole height of the model and remains stable as plots at other times show [4].…”

“…[4]. However, it was also discovered that an undesirable flow condition was present in the flow of the scaled model [4]. This was a recirculation zone below Jet 1 that was unstable with a characteristic time an order of magnitude greater than the vortex shedding behind the posts inside the model.…”

“…This was a recirculation zone below Jet 1 that was unstable with a characteristic time an order of magnitude greater than the vortex shedding behind the posts inside the model. This unstable zone has a very significant impact on the flow field inside the model; it was determined computationally that turning off Jet 1 should render the recirculation zone stable [4]. It is expected to be able to obtain new MIR data for the scaled model of the VHTR lower plenum with improved PIVbased inlet data and with the first jet turned off.…”

“…[5]. While the mesh was deemed somewhat coarse, it was found to be able to provide important results in the study of the MIR data [3][4][5][6]. The mesh, denoted mesh a4, is a hexahedral mesh, with 5.78 million cells, created using GAMBIT 2.4.6 [7].…”