Electric machines are inherently multiphysics devices. They convert electrical energy to mechanical energy, and produce heat in the process. Because modern machines are pushed to their limits, they require engineering design that delves into electrical, mechanical and thermal domains to ensure robust performance. Excess heat can reduce their efficiency and, in some cases, permanently demagnetize magnets and damage insulation on the windings. Optimizing cooling can be extremely complicated, especially for end windings where oil sprays may be used.
Courtesy of Lucid Motors; created in EnSight
To tackle thermal design, engineers often have a detailed CAD model available before they begin simulation. Other times, they rely on the thermal guidance gained through simulation before they create a detailed CAD file, so they can more quickly narrow the design space to focus on the most promising approaches. For each scenario, ANSYS offers simulation tools and a best-practice workflow for engineering electric machines and accelerating design optimization.
Workflow: CAD file available
The images below depict the first type of workflow. Engineers begin with a detailed CAD model of an electric motor, then use simulation to evaluate two cooling methods. The cooling jacket (left) is used to lower the temperature of the bulk of the machine, while the oil spray (right) cools both the stator windings and the rotor bars. For the conjugate heat transfer (CHT) analysis, an all conformal poly mesh is applied that also allows for a transient volume of fluid (VOF) analysis to account for the oil spray cooling. The optimized temperature solution (bottom) combines the CHT analysis with the heat loss data mapped from the integrated electromagnetics simulation.
Workflow: Simulation-aided design
In this second workflow, engineers begin with ANSYS RMxprt, the electric machine design tool specifically developed to accelerate the design of rotating machines. RMxprt not only sizes the machine’s magnetic components, but also automatically generates a high-accuracy magnetic simulation. Housing definitions within RMxprt allow for the generation of a 3D geometry based on some simple parameters, such as length, outer diameter and height. There are also several styles of built-in geometry templates available including fins, flange and a bolted base. All can be easily defined and exported for thermal simulation. The images below show the geometry created with RMxprt.
Once the geometry is created, the second workflow continues like the first. An all conformal poly mesh (below) is created for the CHT analysis, with loss distribution imported from the electromagnetics analysis. Two-way coupling with the electromagnetics is also an option.
Learn more: Register for the webinar
Join us for the webinar, “Optimizing Thermal Design in Electrical Machines with Simulation,” to see an example of both workflow approaches for cooling electric machines. Register here today.
We’ll also be scheduling a two-day training seminar to demonstrate these solutions at our office in Ann Arbor, Michigan. Stay tuned for more details, and I hope to see you there.
This blog was coauthored with Paul Larsen, a lead application engineer in electromechanical simulations.
The post Optimizing the Thermal Design of Electric Machines with Simulation appeared first on ANSYS.