by Valentin Bonnifet
Field Application Engineer chez Nextflow Software
Oil-cooling is a very efficient strategy to maintain systems to their nominal temperature. Like other liquid cooling techniques, oil cooling is often more compact and allows better temperature control than air-cooling. Contrary to water-cooling (or other dedicated cooling fluid) where a specific cooling circuit has to be added, oil-cooling uses a fluid already present in the system most of the time. The oil non-conductivity allows to use such system directly on electric devices which makes oil-cooling a very good candidate for vehicle electrification technologies: e-motor cooling (see previous articles https://lnkd.in/dwhaUpi and https://urlz.fr/dAPL), battery or electronics thermal management (which will be addressed in an upcoming article). In the present article we focus on a generic cooling approach: the oil spray.
When entering a spray nozzle, a continuous flow is split in droplets during the so-called atomization process. Hence, a spray is a discontinuous media that widens the jet aperture angle compared to continuous jet. This feature provides a versatile approach to target the hot areas to cool down in a system. Apart from cooling, sprays also fill a wide range of industrial processes needs from washing to coating.
The spray thermal qualification is a key point of an efficient thermal design. Several spray parameters can be tuned to reach the desired behavior. Amongst them are aperture angle, incidence angle, wall distance, velocity, oil temperature, oil physical properties… These multiple parameters, along with the system geometry own complexity, makes an optimal spray predesign costly when relying on classical experiments and complicated when using analytical considerations. For these reasons, CFD (Computational Fluid Dynamics) is a valuable tool to improve spray design process. Among CFD approaches, SPH (Smoothed Particle Hydrodynamics) is one a the most efficient methods since it is intrinsically suitable to deal with flows involving a lot of free surfaces.
The computations presented in the following images have been obtained with the SPH-flow solver. We consider three 2.5 mm radius oil sprays with different aperture angles: 0 degree, 15 degrees and 30 degrees. The inlet sprays velocity is 0.5 m/s. Each spray is spraying over a wall located 14 mm apart. This configuration is representative of e-motor cooling, as depicted below. On the figure above, the jet is colored by fluid lifetime. One can guess the particle path with a flat falling tail bounded by thicker corners.
The wall heat-flux is depicted on the right. The 30-degrees spray is far more efficient than the others. Note that the flowrate is the same for all three sprays. Surprisingly the 0-degree spray catches more heat than the 15-degrees one once the thermal convergence is achieved.
The two fields presented below depict the wall mapping heat flux at 0.4 s and averaged heat flux, respectively. This wall mapping is a key postprocessing which can be used for thermal coupling with external thermal solver, for which temperature, heat flux or heat thermal coefficient (HTC) must be exchanged. The coupling with external thermal solvers allows to fully compute the cooling of the system.
With heat flux mapping, one could figure out how much the wall surface is cooled. As expected in the present case, the higher the spray aperture angle, the lower the heat flux density.
As shown in this article, simulation results obtained with SPH-flow can help e-motor designers analyze different design choices and their impact on the cooling capability of oil spray.