Testing the electrified vehicle

David Maliniak, Teledyne LeCroy


EVs encounter different issues when it comes to test and measurement

In the popular usage of the term, “electric vehicles” are those whose propulsion systems are under at least partial electric power. Long touted as environmentally beneficial, hybrid vehicles such as Toyota’s Prius and Chevrolet’s Volt have been rather successful in capturing market share initially. Relative newcomers to the market such as Tesla have reinvigorated the excitement around electric vehicles by pushing their performance to another level. However, the all-electric vehicle has seen a slower rate of acceptance than expected.

In the meantime, though, our vehicles are gradually undergoing a transformation in which subsystems that have long been electromechanical or hydraulic in nature are becoming purely electromotive. Vehicle efficiency standards and mandates to reduce CO2 emissions are forcing automakers to abandon old-school belts and gears for powering compressors and pumps in favor of new all-electric approaches.

Rather than using one rotating system (the engine) to drive other rotating systems, we’re now seeing vehicles with electrically driven subsystems, including power steering. By doing so, automakers can often reduce the weight of their products substantially, which goes a long way in terms of improving energy efficiency, while also shifting more of the total energy burden to batteries and away from combustion engines.

Steering the ship

With the trend toward electrification of the complete vehicle comes a somewhat different set of requirements when it comes to test and measurement. Consider, for example, the power steering system. Traditional power-steering systems are hydraulic in nature, with a belt-driven pump that applies fluid pressure to a piston. Turning the steering wheel applies that pressure to one side and the piston moves, which in turn influences either a steering gearbox or a rack-and-pinion system. A complex system of valves and reservoirs regulates everything. The hydraulics does all the heavy lifting of controlling the whole system. Test of such a system involves little more than pressure gauges.

An electronic power-steering system is a very different animal with very different test and validation requirements. The steering system’s ECU calculates the amount of steering assist required based on the amount of torque applied to the steering wheel by the driver, the steering wheel’s position (angle), and the vehicle’s speed. The steering system’s electric motor rotates a steering gear with an applied torque that reduces the torque required from the driver.

Thus, the power steering system relies on data from at least three sensors: a vehicle speed sensor, a steering angle sensor, and a torque sensor, with the sensor data provided in analog signal form or embedded in a CAN (or other) serial data message. Design and/or debug of such a system would require test equipment that can acquire all of the relevant signals and correlate them with each other in a unified view.

One such instrument that serves this need is Teledyne LeCroy’s MDA800 family of motor drive analyzers, which deploys eight analog input channels for capturing multiple signals such as those from the sensors in an electronic power-steering system. Moreover, the MDA800 can provide analysis of the interactions between the analog or serial data sensor data, the steering system’s ECU, the inverter drive, and the three-phase motor under complex, dynamic operating conditions, ensuring that the system is behaving as designed. Such capabilities bring great benefits for any in-vehicle drive and motor analysis.

Managing the batteries

With so much more of the vehicle’s overall power coming from batteries, automakers are spending a lot of time and resources on design of battery management systems. There has been a big push toward 48V power buses within vehicles to accommodate the current draw imposed by electrification of other vehicular subsystems. So-called “mild hybrids” that will extend today’s 12V stop-start technology into the 48V realm are expected to be broadly available by 2020; according to some estimates, as many as one of 10 vehicles sold globally will be a 48-V mild hybrid by 2025.

Stop-start technology uses energy that is typically lost in braking to aid in launching the vehicle in its subsequent acceleration. An “e-charger,” or electric turbocharger, uses a dedicated battery to capture that wasted energy and applies it in the form of low-RPM torque. The result is a quicker engine start with less vibration.

Automakers are now developing vehicle architectures in which the main propulsion batteries, which are in the 300-500V range, will have their voltages down-converted to drive both 12V and 48V buses. The 48V bus will power the large energy consumers such as the e-charger, electronic power steering system, roll stabilization, heaters, pumps, A/C compressor, and 48V starter/generator. The classic 12V bus will handle the lesser loads such as lighting, infotainment, smaller motors, and the rest (see Figure 1).

Cliock image to enlarge

Figure 1: Automakers are moving toward 12/48V electrical architectures that pose a more complex DC-DC conversion scheme.

All of the above amounts to a complex DC-DC conversion architecture, with one or more large dc-dc conversion systems and numerous smaller DC-DC converters. As with the steering system and its multiple sensor inputs, the battery-management system in a 48V/12V vehicle is best characterized and analyzed by an oscilloscope with eight analog input channels. This is because the full H-bridge topology of the system’s inverter includes four power-semiconductor devices. For each of these devices, the design/debug effort ideally permits examination of both the gate-drive signal and switching signal for all four devices within the H-bridge.

Once again, the channel count for such a test setup is higher than the typical four channels one finds on most oscilloscopes on the market. A device like Teledyne LeCroy’s HDO8000 Series, an eight-channel oscilloscope, is well suited to serve this application space. HDO8000 Series brings 12 bits of vertical resolution to the table (15 bits with enhanced resolution applied) as well as bandwidths of up to 1 GHz.

For comprehensive motor-drive analysis capabilities, including three-phase numerical and waveform power analysis from motor-drive input through motor mechanical output, the MDA800 Series of Motor Drive Analyzers, built on the HDO8000 Series platform, provides additional benefits.

Vehicle propulsion

While simple static power analyzers are still used for end-stage validation of the propulsion system in electric vehicles, research and development engineers require more complex dynamic analysis of propulsion system performance.  A common example is the drive and motor performance during a regenerative braking event in which the power flow through the motor and drive reverses so as to load the propulsion system instead of drive it, recapturing energy that would otherwise be lost to braking. 

A conventional power measurement tool provides little value during dynamic measurements such as this, because the short capture time and averaging of power cycles would result in averaged numbers that have little relevance for so dynamic an event. However, the Motor Drive Analyzer easily captures and displays many thousands of power cycles and plots complex power behaviors as per-cycle Waveforms, making it easy to understand system performance. 

Probing concerns

When probing the power electronics of electric or hybrid vehicles, there are concerns regarding the acquisition of small signals floating on a high-voltage bus. The gate drive circuit in a power-electronics design is a series RLC circuit with parasitic capacitances across the semiconductor devices. An upper-side gate-drive device has an applied gate voltage floating above ground.

Any measurement probe with high tip-capacitance in parallel with the parasitic capacitance from the gate to emitter (CGE) or from gate to source (CGS), and/or high impedance and low loop inductance in series with the gate-drive impedance, will at best undesirably load the gate-drive signal and at worst cause the circuit to malfunction. Measurement interference from the low-side device switching may also interfere with proper measurement of the upper-side gate voltage.

For such probing applications, Teledyne LeCroy has announced its high-voltage, fiber-optically isolated (HVFO) probe, which provides optical isolation between the probe tip and oscilloscope input to reduce adverse loading of the DUT. The probe also reduces noise, distortion, ringing, overshoots, and transients on measured signals. It easily surpasses the measurement capabilities and signal fidelities of conventional HV differential probes and acquisition systems that rely on galvanic channel-to-channel and channel-to-ground HV isolation (Figure 2).

Click image to enlarge

Figure 2: The HVFO (yellow trace) and a high quality HV differential probe (blue trace) measure an upper-side low voltage sensor signal. The HVFO lower loading and higher CMRR more accurately represents the signal, whereas the conventional HV differential probe loading and lesser CMRR performance less accurately represents the signal.

With its low tip capacitance and high input impedance, and because its amplifier is optically isolated, the HVFO’s tip has only to measure the small signal gate-drive voltage. As a result, loading is about 1/100th of that of conventional HV differential probes (Figure 3). Expected bandwidth for the HVFO, which will be available in January, will be in the range of 60 MHz. Input impedance will be 5-10 MΩ for most applications. Common-mode rejection ratio will be ~100 dB.

Click image to enlarge

Figure 3: The HVFO measures only the low voltage sensor voltage across its leads.  Therefore, the total RC load on the circuit is ~100 times less than that of a conventional HV differential probe.