Svetlana Aleksandrova

Linear stability of Hunt's flow

We analyse numerically the linear stability of the fully developed flow of a liquid metal in a square duct subject to a transverse magnetic field. The walls of the duct perpendicular to the magnetic field are perfectly conducting whereas the parallel ones are insulating. In a sufficiently strong magnetic field, the flow consists of two jets at the insulating walls and a near-stagnant core. We use a vector stream function formulation and Chebyshev collocation method to solve the eigenvalue problem for small-amplitude perturbations. Due to the two-fold reflection symmetry of the base flow the disturbances with four different parity combinations over the duct cross-section decouple from each other. Magnetic field renders the flow in a square duct linearly unstable at the Hartmann number Ha ≈ 5.7 with respect to a disturbance whose vorticity component along the magnetic field is even across the field and odd along it. For this mode, the minimum of the critical Reynolds number Re c ≈ 2018, based on the maximal velocity, is attained at Ha ≈ 10. Further increase of the magnetic field stabilizes this mode with Re c growing approximately as Ha. For Ha > 40, the spanwise parity of the most dangerous disturbance reverses across the magnetic field. At Ha ≈ 46 a new pair of most dangerous disturbances appears with the parity along the magnetic field being opposite to that of the previous two modes. The critical Reynolds number, which is very close for both of these modes, attains a minimum, Re c ≈ 1130, at Ha ≈ 70 and increases as Re c ≈ 91 Ha 1/2 for Ha » 1. The asymptotics of the critical wavenumber is k c ≈ 0.525Ha 1/2 while the critical phase velocity approaches 0.475 of the maximum jet velocity.

Linear stability of magnetohydrodynamic flow in a perfectly conducting rectangular duct

following the jets becomes confined in the layers of characteristic thickness Ha 1=2 located at the walls parallel to the magnetic field. In this case the instability is determined by ; which results in both the critical Reynolds number and wavenumber scaling as 1 : Instability modes can have one of the four different symmetry combinations along and across the magnetic field. The most unstable is a pair of modes with an even distribution of vorticity along the magnetic field. These two modes represent strongly non-uniform vortices aligned with the magnetic field, which rotate either in the same or opposite senses across the magnetic field. The former enhance while the latter weaken one another provided that the magnetic field is not too strong or the walls parallel to the field are not too far apart. In a strong magnetic field, when the vortices at the opposite walls are well separated by the core flow, the critical Reynolds number and wavenumber for both of these instability modes are the same: Rec 642Ha 1=2 C 8:9 10 3 Ha 1=2 and kc 0:477Ha 1=2 : The other pair of modes, which differs from the previous one by an odd distribution of vorticity along the magnetic field, is more stable with an approximately four times higher critical Reynolds number.

An Assessment of CFD Applied to Steady Flow in a Planar Diffuser Upstream of an Automotive Catalyst Monolith

Flow maldistribution across automotive exhaust catalysts significantly affects their conversion efficiency. Flow behaviour can be predicted using computational fluid dynamics (CFD). This study investigates the application of CFD to modelling flow in a 2D system consisting of a catalyst monolith downstream of a wide-angled planar diffuser presented with steady flow. Two distinct approaches, porous medium and individual channels, are used to model monoliths of length 27 mm and 100 mm. Flow predictions are compared to particle image velocimetry (PIV) measurements made in the diffuser and hot wire anemometry (HWA) data taken downstream of the monolith. Both simulations compare favourably with PIV measurements, although the models underestimate the degree of mixing in the shear layer at the periphery of the emerging jet. Tangential velocities are predicted well in the central jet region but are overestimated elsewhere, especially at the closest measured distance, 2.5 mm from the monolith. The individual channels model is found to provide a more consistently accurate velocity profile downstream of the monolith. Maximum velocities, on the centre line and at the secondary peak near to the wall, are reasonably well matched for the cases where the flow is more maldistributed. Under these conditions, a porous medium model remains attractive because of low computational demand.

Linear stability of buoyant convection in a horizontal layer of an electrically conducting fluid in moderate and high vertical magnetic field

Linear stability of buoyant convective flow in a horizontal layer of an electrically conducting fluid is considered with reference to horizontal Bridgman semiconductor crystal growth. The fluid flows owing to the horizontal temperature gradient in the presence of a vertical magnetic field. The main interest here is in the stability of the flow for a sufficiently strong magnetic field, for the Hartmann number Ha > 10, and increasing to high values, of the order of 103–104. The Prandtl number, Pr, has been fixed at Pr = 0.015. It is shown that besides the Hartmann number the instability strongly depends on the type of the thermal boundary conditions at the horizontal walls. For thermally conducting walls the basic temperature profile exhibits zones of unstable thermal stratification, which leads to instabilities owing to the Rayleigh-Benard mechanism. However, the transitions between various, most unstable modes as Ha increases are not trivial. For sufficiently high values of Ha, the most unstable mode cons...

Auxiliary power units for range extended electric vehicles

This paper describes research carried out into auxiliary power units (APU) for range extended electric vehicles (RE-EV), as part of the Low Carbon Vehicle Technology Project (LCVTP) (see acknowledgement). APU requirements are specified and compared to the attributes of a variety of prime power sources. It is concluded that for many applications adaptation of a volume production 4-stroke gasoline engine will be the most suitable choice for a RE-EV, until production volumes make a bespoke engine viable.

The Effect of Swirl on the Flow Uniformity in Automotive Exhaust Catalysts

In aftertreatment system design, flow uniformity is of paramount importance as it affects aftertreatment device conversion efficiency and durability. The major trend of downsizing engines using turbochargers means the effect of the turbine residual swirl on the flow needs to be considered. In this paper, this effect has been investigated experimentally and numerically. A swirling flow rig with a moving-block swirl generator was used to generate swirling flow in a sudden expansion diffuser with a wash-coated diesel oxidation catalyst (DOC) downstream. Hot-wire anemometry (HWA) was used to measure the axial and tangential velocities of the swirling flow upstream of the diffuser expansion and the axial velocity downstream the monolith. With no swirl, the flow in the catalyst monolith is highly non-uniform with maximum velocities near the diffuser axis. At high swirl levels, the flow is also highly nonuniform with the highest velocities near the diffuser wall. An intermediate swirl level exists where the flow is most uniform. To gain further insight into the mechanisms controlling flow redistribution, numerical simulations have been performed using the commercial CFD code STARCCM+. With no swirl, the central jet transverses the diffuser, and a drastic flow redistribution takes place near the monolith face due to its high resistance. Immediately downstream of the sudden expansion, the flow separates from the diffuser wall forming a separation zone around the central jet. Increasing swirl reduces the size of this separation zone, and eventually leads to the formation of the central recirculation zone characteristic of high swirl flows. At intermediate swirl levels, the size of the wall separation zone is reduced considerably, while the axial adverse pressure gradient is insufficient to cause a central recirculation. Such a flow regime occurs at relatively low swirl levels (S ~ 0.23). This may have positive implications for aftertreatment system design with low residual swirl levels from the turbine, which might be tuned by adjusting the distance between the turbine and the catalyst or employing guide vanes. The findings can be directly transferred to other aftertreatment systems with a catalyst or particulate filter. Moreover, swirling flows with an obstruction or a high resistance device downstream (e.g. a heat exchanger or filter) are present in many other applications such as cooling flows, combustion and turbomachinery. Therefore the results are relevant to a much wider research and industrial community.

Effect of grid-filter width definition on implicitly filtered large eddy simulations using OpenFOAM

Turbulent flow over a 3D backward-facing step at a step-height Reynolds number, Reh, of 5100 is simulated in OpenFOAM using Large Eddy Simulation (LES) to investigate how the definition of the grid-filter width, Δ, influences the computed solutions. As part of this study, two different Δ definitions are tested: Δ=ΔxΔyΔz3 and Δ = max(Δx, Δy, Δz). The LES results show good agreement when compared with DNS and experimental data. The simulation results are used to further evaluate the effect of grid non-uniformity on the computation of the subgrid-scale eddy viscosity, νSGS.

Evaluation of transition-sensitive eddy-viscosity turbulence models for separated flow in OpenFOAM

A recently published transition-sensitive turbulence model, k−kL−ω−υ2 [1], is implemented in the open-source CFD package OpenFOAM, and its performance is evaluated in comparison with k−kL−ω [2] and υ2− f [3] models. On T3A and T3B flat plate cases, the k−kL−ω−υ2 model gives accurate transitional predictions. On a flapped NACA 23012 aerofoil, it is found to give only a small improvement over the k−kL−ω model (under 5% reduction in error for lift coefficient) compared with experimental results obtained at the Coventry University wind tunnel, showing limited effects of the extra transport equation which was added to sensitise the model to rotation and curvature effects. Assessment of fluctuating kinetic energy and the new wall-normal turbulent velocity scale shows overprediction near the wall compared to the υ2− f model which indicates a delayed prediction of separation.