The Benefits of Validating eAxles with the PXB Power Supply
While much EV testing revolves around battery cycle testing with clearly defined charge/discharge cycles, in essence simulating the motor to test the battery, many test applications are cropping up that instead simulate the battery to test the motor. These tests are often used for electric axle (eAxle) validation. Instead of a central drive configuration where a large motor is connected to the mechanical transmission to propel the car forward, this system is scaled down into a housing that sits on the axle. This leads to a much lighter, more power efficient system that will, in turn, drive down fuel consumption for hybrids and maximize range for BEVs. The tests for these systems will often require compliance with strict industry standards such as ISO 6469-3, IEC60664, and LV123.
The Kikusui PXB power supply are well-suited for these tests with an internal resistance emulator and an IV function that can accurately simulate any battery of your choosing, a stable and clean output under high capacitance loads (often seen with noisy motors), and bidirectionality to help test the regenerative functions of the motors themselves. This article describes the process for testing eAxles and how the PXB power supply can be custom-tailored to any eAxle test setup.
The traditional drivetrain
Typically a vehicle rests on two or more axles, and these can be configured to specific tasks. Axle configurations will vary per vehicle and often follow the following naming pattern:
Nw x Nd = (# of wheels/twin wheels the vehicle has) x (# of wheels driven)
In other words, a 4 x 2 configuration will have 2 wheels driven with a total of 4 sets of wheels (single or double), while a 6 x 4 will have 4 wheels driven with a total of 6 sets of wheels (Figure 1). As shown in the image, a 4 x 2 will have one driven axle while a 6 x 2 will have two driven axles — these two are the most common configurations for long-haul vehicles.
Figure 1: The various axle configurations that are used in commuter cars and fleet vehicles.
In a traditional internal combustion engine vehicle (ICE), this function is part of the powertrain that contains the transmission, differential, driveshaft axles, and the wheels, all of which propel the car forward.
The importance of eAxles
Conventional EVs largely used this basic powertrain topology where the ICE was replaced with the battery pack, the high voltage power distribution (DC-AC inverter), and an electric machine (i.e., electric motor) such as an induction machine (IM), permanent magnet synchronous machines (PMSM), and switched reluctance machines (SRM), connected to the mechanical transmission. This central drive EV configuration led to a dramatic reduction in the number of components, limiting the avenues for a failure to occur. E-machines also will generate much greater speeds and torque than traditional ICE engines.
However, electric axles, or eAxles, have more recently grown in popularity. These systems scale down the larger conventional powertrain of the EV by incorporating a smaller electro-mechanical propulsion system on the axle with an integrated electric machine, inverter, and corresponding reduction gearbox (Figure 2). This leads to much more efficient vehicle operation with a reduction in driving resistance. And, since range maximization is the primary goal for EVs, eAxles are growing in importance in battery EVs (BEVs), fuel-cell EVs (FCEVs) as well as hybrids with an ICE, such as hybrid EVs (HEVs) and plug-in hybrid EVs (PHEVs) where the size of the market itself is anticipated to grow nearly six fold in the next nine years.
Figure 2: A block diagram illustrating the various electrical and mechanical systems within an eAxle.
As with all things automotive, these systems require stringent testing to meet automotive regulations. Even in the process of designing and building prototypes, a thorough characterization is required. Characterizing eAxles involves simulating various types of batteries (e.g., high voltage, low voltage, etc.) that might be used within the vehicle while performing torque and speed measurements, as well as power analysis, vibration monitoring, and temperature monitoring.
Typical test setups involve the use of two load machines, or motors, that mount to the output shafts of the eAxle to simulate the vehicle at various speeds and road gradients in different driving conditions (e.g., urban, long-haul, etc.). Battery simulators are necessary to adequately simulate the power supplied to the eAxle under test. Many systems will include some type of environmental simulation as well, with either an environmental chamber or a thermal conditioning system (Figure 3).
Figure 3: Simple block diagram for an eAxle testing with two oppositely mounted motors, an environmental chamber and a battery simulator.
EV powertrain optimization first requires a driving cycle to indicate how the required torque and speed may vary in EVs or commercial vehicle applications. This will change the amount of energy and power consumed by the E-Axle as well as its efficiency over time or distance. For EV models, the urban dynamometer driving schedule (UDDS) driving cycle or the Highway Fuel Economy Test (HWFET) torque-speed profiles might be used to better represent city driving conditions (Figure 4). For heavy-duty vehicles, the Truckerrunde cycle, Vehicle Energy Consumption calculation Tool (VECTO) long-haul, VECTO-regional, or T2030 might be used .
Figure 4: Various driving cycles are used to simulate the environment that a vehicle is in and properly test automotive components such as batteries, emotors, and eAxles. Image Source: 
Using this equipment will yield a fine-grained view of how the eAxle functions with power loss and efficiency maps to optimize parameters such as torque vectoring (i.e., torque split between motors), transmission speed, and gear ratio for maximum efficiency and range per charge. Other parameters such as noise, vibration, and harshness (NVH) behavior and EMI performance may also be necessary to measure. While NVH behavior is more pronounced in an ICE-based vehicle, vehicles utilizing eAxles might also require NVH testing to characterize tonal noise from electrical machines (e.g., motors), gear train components, and high frequency components from fast switching power supplies. Automotive EMI testing is necessary to ensure automotive systems meet electromagnetic compatibility (EMC) standards in real operating conditions.
The benefits of using the PXB for eAxle testing
Internal resistance (IR) emulator and I-V curve simulator
Ideally, a battery will have a low IR in order to deliver a high current on demand. However, a high IR can cause the battery to heat up, waste energy and cause a voltage drop. An increase in a battery’s IR is a function of several factors, including temperature, aging, and state of charge (SoC). The IR of a battery will typically remain constant during the duration of its operational life — however, IR increases toward end of life. Temperature will also impact the resistive and electrochemical elements within a cell, where resistive elements will increase in resistance at high temperatures while the resistivity of the electrolyte will often decrease with an increasing temperature. As for SoC, when a cell is fully charged the IR stays low. But as the cell discharges, the IR increases gradually. The increase in IR often becomes much more rapid around 25% SoC. All of this may need to be taken into account when simulating a battery (Figure 5).
The PXB can also operate according to any I-V characteristics to simulate any solar cell module that might include a photovoltaic (PV) cell and battery (B) (Figure 5). These curves may be critical for the testing of systems that use solar cells, fuel cells, and more. The ability to simulate an I-V curve diverges from the limited functionality of a constant voltage power supply where external circuits would generally be required to simulate this type of behavior.
Figure 5: The PXB is able to simulate a battery’s IR to better mimic any type of battery at any age, operating at any temperature, and at any SoC (left) and operating according to the I-V characteristic of any solar module with a battery and PV cell.
Stable clean output under high capacitance loads
While it is necessary to have a programmable power supply to simulate the behavior of the car battery and provide power to the eAxle, the power supply will also have to handle the highly capacitive loads or inductive loads that the noisy motor might output. The PXB is designed for highly stable operation, without oscillation or overshoot, even when a load with a large capacitive component is connected. The slew rate and response can be varied to match the characteristics of the connected load, suppressing any potential oscillation and overshoot (Figure 6).
Figure 6: The PXB power supplies’ output waveform with a 500 uF capacitance connected without oscillation or overshoot.
Bidirectional power supply
In order to test the regenerative functions of the motor, it is important to have a bidirectional power supply that can operate as both a source and a sink (Figure 7). The PXB operation is fully bidirectional and offers on-site regenerative capabilities where any power sunk into the power supply is converted to reusable power and regenerated to the AC LINE to supply neighboring racks and nearby test equipment. This is an alternative to the conventional use of resistive load banks, where excess power is converted to heat energy and dissipated in the air.
Figure 7: The PXB’s bidirectionality and regenerative feature assists with testing an eAxle’s regenerative functions.
Summary on the benefits of using the PXB to test eAxles
The PXB bidirectional power supply is well-suited toward the testing of eAxles due to its inherent bidirectionality, allowing users to test the regenerative capabilities of the motors. The power supply can easily mimic the changing IR of a battery, allowing test facilities to successfully simulate any type of battery to drive the eAxle. Finally, when testing eAxles, there is often difficulty in finding a power supply that can handle the high capacitance and high inductance loads that a noisy motor might output. The PXB is able to handle these loads without producing any overshoot and oscillation.