Hybrid electric vehicles (HEV) and electric vehicles (EV) have multiple architectural variations. Figure 1 shows a simplified block diagram of a couple of these architectures. For the strong (or parallel) hybrid and the pure EV (no engine), a high voltage (HV) bus supplied by a large battery, drives the electric powertrain. Power levels of the inverter and motor/generator range from ~ 60 kW up to and over 180 kW. Along with the large Li-Ion battery, a significant investment is required to develop these architectures. Most of the components are bidirectional, allowing for power to go from the battery to the inverter, which turns the motor and moves the vehicle (traction drive). When decelerating, the momentum of the vehicle turns the generator, which drives power back through the inverter and charges the battery (regenerative braking).
Figure 1: HEV/EV Power Train Architectures
In the mild hybrid (MH), the motor/generator, inverter and battery are also bidirectional. They are not large enough to drive the vehicle by themselves (as in the HEV or EV), but instead are used to supplement the engine power during acceleration and recharge the battery during deceleration. The voltage level for MHs is typically 48 V, keeping the bus structure under the 60 V safety rating for HV. The DC:DC converter is also bidirectional for MHs. In addition to charging the 12 V battery from the 48 V bus, the DC:DC converter also needs to convert power from the 12 V bus to the 48 V bus. The main application is to pre-charge the 48 V bus (i.e. input capacitance of the inverter), before contactors connect the 48 V battery to this bus. The pre-charge equalizes the voltage of the battery and the input to the inverter, minimizing arcing across the contactor. So how does one test the bidirectional nature of HEV/EV power converters?
Bidirectional power flow requires test solutions that are capable of sourcing and sinking power to the power converter, for example simulating the Li-Ion battery when testing the DC side of the traction inverter or the DC:DC converter. This is traditionally accomplished by connecting a power supply and an electronic load in parallel. However, external circuitry (i.e. diode to stop current flow into the power supply) and cumbersome ‘two-instrument’ programming typically doesn’t allow for smooth signal transitions between sourcing and sinking power, reflecting an inaccurate simulation of the operating conditions. Reference ‘Conquering the Multi Kilowatt Source/Sink Test Challenge – Application Note’ (https://literature.cdn.keysight.com/litweb/pdf/5991-2873EN.pdf?id=2373832) to learn more about this test challenge.
Electronic loads typically dissipate the power transferred to them from the DC:DC converter. But this dissipated power (heat) can start to add up, especially for test applications where multiple DC:DC converters are tested in parallel. Because of the need to remove the heat from the electronic loads, they are often large with significant forced air (via fans) or potentially even water circulation to cool them. Therefore, it is necessary to increase test system size and HVAC requirements to remove the heat from the facility.
To address these challenges, multiple vendors are introducing integrated source/sink solutions in a single product. The products can seamlessly move from sourcing current (Quadrant I) to sinking current (Quadrant II) without external circuits or synchronized programming of a separate power supply and electronic load (see Figure 2). The integration enables a smooth output waveform that correctly simulates the bidirectional DC:DC converter’s transition between opposite directions of power flow.
Figure 2 – Source/Sink Power System
When the power system sources power to the DC:DC converter, most of the power (depending on the efficiency) is passed through the converter to the automotive load. When the power system sinks power from the DC:DC converter, the power must be absorbed by the power system. At the 5kW power level and above, there are source/sink power systems and electronic loads that regenerate (or return) the power to the AC mains (see Figure 2). This technique is not 100% efficient, but most designs allow ~ 90% of the power to be delivered back to the grid. This leaves only 10% of the power (~500W in the case of a 5 kW product) to be dissipated as heat. The result is a dramatic reduction in size of the products and the HVAC cost to remove heat from the test system environment.
The important thing to note for regenerative solutions is the fidelity of the power being returned to the AC mains. If you work in manufacturing, any distortion in the power returned to the AC mains will be amplified by the number of test systems in the facility. ‘Dirty Power’ can create intermittent problems in your facility, requiring isolation transformers for each test system to help mitigate the problem created by poor regeneration. It is best to check with the product vendor to confirm low distortion power is returned to the AC mains (see Figure 3).
Figure 3: THD & PF measurements on power returned to AC mains from Keysight’s RP7900A Regenerative Power System. Measurements made with PA2203A IntegraVision Power Analyzer
Keysight Technologies provides many solutions for the HEV/EV market, including DC:DC converter design and test. For more information on the RP7900A RPS (Figure 4) go to www.keysight.com/find/ev1003a.
With its recent acquisition of Scienlab, Keysight Technologies has a comprehensive test and measurement portfolio to serve the HEV/EV market. For more solutions, please visit: www.scienlab.com.
Figure 4: RP7900A Regenerative Power System