Innovation, in many cases, does not lie in the invention of completely new ideas, but the right combination of existing ones. These combinations of techniques can lead to the improvement of existing technologies, taking them up to new boundaries as well as the creation of new products.
In this post we’d like to share a bit of that recipe; or at least, some of the tools we use to make some of the highest performing machines in the world happen. So, stick with us and get a few insights into how the cake is made (or rather our pancake and sausage motors).
We will talk about simulation for the real world, the tools we use and a bit on how we use them.
Stress & Thermal
Stress and Thermal analyses are two of the most common types of analyses in industry. Complementing our hand calculations and code sheets, we have a dedicated analysis team that use state of the art Finite Element Analysis tools.
When applied, the design, analysis, and R&D teams work together throughout the whole production cycle in order to refine our product.
At first, stress and thermal analyses are performed to simulate new concepts at pre-design stage. During design, the analysis team validate the concept designs to make sure they are fit for purpose. At this stage, several simulations will be run to check the stress levels across the motor housing and its components.
In applications where the motor assemblies may be subject to heavy vibrations, these are analysed in dynamic simulations, looking for any resonance and increased stresses due to cyclic loading. The most likely part of a motor to fail under such circumstances is usually the housing close to attachment points such as bolted joints, so particular attention is paid to these areas.
A successful motor design relies heavily on its thermal performance. Thermal analyses are usually run with nearly whole motor assemblies being imported into the FEA package. Heat transfer is reviewed by taking into account heat sinks such as the cooling channels, and sources of heat such as stator teeth and some parts of the rotor.
In such cases, all thermal, static and dynamic stress analyses are run in parallel. As large temperatures cause the materials in the motor to expand at different rates, stresses may arise. Ultimately, changes in stiffness and stresses due to thermal expansion need to be explored together with cyclic loading cases as well.
During the R&D stages of the motor production cycle, problems may appear that were not originally foreseen by design or analysis. The purpose of the analysis team in this case is to reproduce failure conditions that were encountered during testing and establish the causes of the failure and correlate this to the test findings.
All our FE analyses are run using Dassault Systèmes Abaqus CAE.
Two of the main focus areas at Integral Powertrain in order to make our motors as power dense as possible are state of the art manufacturing processes and cooling.
Focusing on the latter, we have a patented cooling technology that we use in practically all of our motors.
Since our machines are water cooled, fluid dynamics analysis becomes an important part of our refining process. For this purpose, a combination of calculation sheets, CFD, and testing is applied.
The complexity of the analyses will vary depending on the cooling channel geometry, with the most complex geometries usually appearing in our motorsport and hyper-car applications. The higher the performance of the machine, the more cooling it will require and therefore, fluid dynamics becomes even more important.
The main parameter looked at during this process is pressure drop. In many cases, our customers will have already chosen a pump to drive the coolant through the motor, which often shares a circuit with the battery cooling system. Because the pump has a limit on the amount of pressure at which it runs, the pressure drop in our system has to be less than that of the pump.
Additional challenges arise from having a consecutive set of cooling circuits. This is the case when, besides battery and motor, an inverter is added in series. Our bespoke Silicon Carbide switch inverters, which weigh 5.5Kg, run at relatively high switching frequencies and can handle currents above 620 Arms at 800 Volts nominal. For them to be as efficient as possible, we need to run a cooling circuit, which is often in series with that of the motor.
Getting back to the specifics of the analysis, our first stage involves an estimate of the pressure drop by using in-house tools. As the geometry complexity increases, calculation sheets are no longer enough and a CFD analysis is performed which provides additional insight into the behaviour of the flow inside the channels.
A “simple” channel geometry allows us to use relatively coarse meshes and still obtain meaningful results which we later validate during testing. Complex geometries will have a higher demand in terms of mesh and turbulence models applied during the simulation.
Ultimately, a combination of fluid and thermal analyses can be run to explore the efficiency of the cooling channel at removing heat from the whole system.
For our CFD analyses we use a combination of Abaqus CFD and OpenFOAM, depending on complexity.
Finally, La Piece de Resistance!
As many or some of you would know, a motor’s heart lies in EM (Electromagnetics). It is this area of study that encompasses the physics that make an electric machine possible. Hence, this is one of our core departments, and one that usually comes at the earliest stages of the concept and design processes.
From the moment a customer contacts us and a set of requirements is specified, we move on to select the most appropriate machine for the application. As Integral e-Drive use a scalable core technology concept to create their machines, we can either choose an existing “ready to print” machine, or design/modify a motor using our in-house tools without the need to completely re-think a concept.
Through that initial process, we pick from a variable list of options to specify the ideal rotor and stator diameters and materials, number of winding turns around our stator teeth, type of cooling technology, and so on so that the end result is precisely matched to the client’s system needs. For this it is usual to collect information from our customer such as max desired motor speed, max and min battery voltages, size and weight limits, max and continuous power and torque figures and max available current from the inverter.
Once the concept is defined through our in-house tools, which takes into account EM, thermal, and some mechanical parameters, the EM analysis department will proceed with further actions. These may involve FE simulations where a cross-section of the motor is analysed to look at the motor’s performance under specified conditions. The conditions to simulate will often come from a duty cycle in the form of voltage and current use throughout, for example, a lap around a circuit.
Through simulation we can look at the motor’s efficiency for a particular cycle, and examine the thermal losses that it may experience in both rotor and stator. Likewise, the inverter can be simulated as well.
For the EM simulations we use a combination of in-house tools and Motor-CAD.
We hope this post has been insightful regarding the use of simulation at Integral Powertrain’s e-Drive Division. After all, the modern world is all about optimization, which can’t happen without the use of computers.