With continuously advancing electrification and increasing charging power of electric vehicles (EV) and heavy-duty electric vehicles, high-power DC chargers (HPC) and megawatt chargers (MWC) are commonly set up at public charging stations and depots. Small-scale DC chargers (DCC) are employed in residential applications due to the rise of secondary services, such as bidirectional charging.

With increasing application, new charger designs and control approaches are developed, built, and operated. Subsequently, testing setups and methodologies are needed. To improve the variety of possible testing events, repeatability, speed, and accuracy, test benches imitating EV and the grid side are used.

In general, HPC, MWC, and DCC testing can be categorized into two groups: the communication side and the power transfer side. In communication testing, the interaction between EV and charger is analysed, putting the focus on interoperability, safety and emergency functions, and charger control. Consequently, the power and precision of the EV power component representation within the test bench can be minimized. In the power transfer side, the focus lies, among others, on the charging behaviour, the charger's output stability, and the efficiency. Subsequently, the focus of the test bench is set on the EV's power components, while the communication control is set to minimum.

One central part of these test benches is the battery emulation, taking in the charged energy and providing feedback to the charger via the voltage. Often, bidirectional DC loads are used, feeding back the charged energy at either the same or another grid connection point. This can lead to inaccuracies and problems when analysing the power electronics behaviour of the charger, as the power electronics of the DC loads influence the system themselves.

To evaluate these impact factors, this research analyses the influence of different load representations on testing ability and accuracy. Therefore, a previously built and published laboratory test bench is expanded utilizing different load representations (e.g., different control methods of the DC loads, combined with DC-side filters and batteries). Measured behaviour is compared to a real EV charging session serving as a baseline. The effect on charger behaviour is (hopefully :-) ) shown, classified, and discussed.

In preliminary testing, it is shown that DC loads tend to cause higher ripple on the DC-side, supporting the need for further system investigations.