Thermal Performance of Electric Vehicle Inverters
New computational research tool enables faster design iteration testing to save time and money on prototyping.
Given the need to cut carbon emissions, it’s no secret that electric vehicles (EV), including hybrid electric vehicles, have invaded the automotive landscape. In fact, sales topped over two million in 2019.1 So, as the trend toward electrified vehicles continues onward and upward, it’s also no secret that OEMs are making major investments in improving battery life.
Limitations of EV Batteries
Since the introduction of e-mobility, OEMs have relied on traditional lithium-ion batteries to power their vehicles. The problem, however, is that these batters have limitations; namely they have the potential – under certain voltages and temperatures – to catch fire. They’re also big and heavy, which means they take up space and drive down fuel efficiency.
Additionally, most car owners charge their vehicles at home from an electric supply of alternating current (AC) while most EVs maintain a direct current (DC). That’s why every EV battery includes an inverter; a device that transforms AC to DC so the vehicle can be charged. Yet, like today’s EV batteries, these inverters have yet to reach their full potential and performance capabilities.
We talked about it in this blog post. We also talked about how effective thermal design holds the potential to improve safety, reliability, and fuel efficiency. Most importantly, we announced the launch of MES’ thermal characteristics research program, which is being conducted in collaboration with The Ohio State University.
OSU researchers have been making great strides, so we wanted to share an inside look at the latest discovery; this one around EV battery inverters.
Improving Inverters Through Computational Research
Because OSU engineers know that one of the keys to improving EV battery performance is to understand the thermal behavior of EV inverters, they developed physics-based computational models to simulate the 3D thermal transport that occurs within an EV battery inverter.
Thanks to their predictive capabilities, these models are giving researchers a virtual platform for quickly testing design iterations long before they undergo more expensive and time-consuming physical prototyping and testing.
The Model Development Process
Here’s a step-by-step look at how OSU engineers in the MES research program developed these physics-based computation models:
Step 1: Researchers created the inverter geometry with all its relevant components and features.
Step 2: They performed a 3D scan to generate a housing for the CAD model, then cleaned and simplified it for the thermal model.
Step 3: Using available datasheets, researchers approximated the heat source: insulated gate bipolar transistor (IGBT) modules.
Step 4: Each module consisted of IGBT-diode pairs (chips) soldered onto multilayer direct bonded copper (DBC) substrates soldered onto a copper plate.
Step 5: Since the actual geometry of the heat sink was unknown, researchers approximated an ideal heat sink by creating a domain fixed at a constant coolant temperature of approximately 55ºC.
Note: While there are several other components, such as the main controller circuit, gate driver, busbars, mounts, etc., that reside within the housing, researchers only modeled the IGBT modules and capacitor for their analysis.