Pre-Switch has published the most efficient data on the CleanWave200 200kW inverter that was recently released by Pre-Switch. In an interview with EETimes, Bruce Renouard, CEO of Pre-Switch, explained that the efficiency can be as high as 99.3% (Vacuum Vector – Modified) at a switching frequency of 100 kHz with a flat profile where the load varies, resulting in an increased electric vehicle (EV) up to 12%. “We have a massive amount of data published as of today that shows how we can achieve 99.3% with 0.01% accuracy,” Renouard said. Utilizing AI-based DC/DC and AC/DC soft-switching technology, Pre-Switch demonstrated how this was achieved by using three separate, low-cost 35-cubic-meter SiC FET modules per switch location. “We focus primarily on silicon carbide, with the goal of virtually eliminating nearly 100% of shift losses,” said Renouard. And as a result, [by] To reduce switching losses, we can reduce the amount of silicon carbide needed for each system by about 50%. The amount of SiC saved depends on how much switching losses the alternative system has, but it’s certainly a big part. This is a significant cost saving.” The CleanWave200 inverter (Fig. 1) provides fast switching frequencies that create a near-pure sine wave that makes electric motors efficient. Increased switching frequencies also reduce the size and cost of DC jumper capacitors, out of proportion to the increased switching speed, and have the benefit of An additional feature is to enable the low weight induction motors needed in aviation Figure 1: CleanWaveTM rating system (top view) Power electronics needs AI Pre-Switch AI allows users to move from costly and wasteful implementations that require constant switching to designs Efficient and soft switching with 10 times higher switching frequency results in a quasi-pure sine wave output.Artificial intelligence technology analyzes its parameters in real time, and makes necessary adjustments to small resonant transistors, resulting in soft switching even in difficult and changing environments.The algorithm takes AI before switching into consideration a range of parameters such as temperature, device degradation, input voltage change and sudden current fluctuations. To hard-switch simply the transistor on and off by adding current or voltage to enable the modified states. Hard switching is known to require a lot of hardware on transistors, and also shortens their lifespan. On the other hand, the concept of soft switching uses an external circuit to avoid interference of voltage and current waveforms when switching transistors. Inverters for Electric Vehicles In the automotive sector, research into the efficiency of electric vehicles focuses on the performance of the battery and the efficiency of the inverter and electric motor used. Strict automotive safety and quality standards drive technological innovation to methods that increase the efficiency and autonomy of electric vehicles while reducing battery size and weight and lowering costs. AI provides a key support in the push towards autonomy and efficiency of electric vehicles, including efforts to eliminate switching losses in order to ensure fast transistor switching. Scaling up an electric vehicle requires improving both the efficiency of the motor and the inverter known as driveline losses. Driveline losses dominate most EV losses around 50 mph, at which point the wind resistance takes over. But driveline losses account for the largest share of all losses in electric vehicles, so it is necessary to monitor both the inverter and the motor, with a trade-off between switching losses and higher motor efficiency. Motor iron losses decrease with increasing switching frequency, but inverter losses increase. Renouard noted that SiC assists the inverter at lower power levels but many EVs still use SiC devices at lower switching frequencies – on the order of 10 kHz. However, increasing the switching frequency does not always solve the problem. Faster switching results in higher switching losses, which reduces the efficiency of the inverter. Furthermore, Renouard said that if you want to try to switch FETs faster and keep the inverter efficiency high, you need to add more FETs to reduce conduction losses in an effort to compensate for the higher switching losses. This results in increased cost, and the higher voltage/dt associated with fast switching frequencies often requires thicker motor insulation and ceramic bearings to make the motors more powerful. Pre-Switch addresses this challenge by incorporating artificial intelligence into the FPGA that is used to precisely control the timing of the auxiliary resonant transistors, as shown in S1 and S2 in Figure 2. The result is the virtual elimination of all switching losses in the main SiC transistors (Q1 and Q2). ). Figure 2: Pre-switching integrates artificial intelligence into the FPGA, which precisely controls the timing of the auxiliary resonant transistors S1 and S2. Figure 2: Pre-switching integrates artificial intelligence into the FPGA, which precisely controls the timing of the auxiliary resonant transistors S1 and S2. During each switching cycle, the auxiliary resonant transistors S1 and S2 are timed to ensure that Q1 and Q2 have virtually no switching losses. The algorithm calculates and reduces the dead time based on the full knowledge of how and when each key travels. “Let’s take a look at Figure 3, which shows 20 switching cycles,” Reynward said. “When turned on, the algorithm starts the learning process, then in the fourth switching cycle, the first correction provided by the AI is made. In this case, a decrease in the resonant current of the inductor [shown in green] Observed. Going forward, the algorithm will adjust the inductor’s buzzing current independently to ensure that it oscillates briefly above the load current [shown in blue]. All adjustments are fast enough to ensure accurate and smooth switching with any PWM input and can be used to create a perfect sine wave using an AC/DC converter. The system also works perfectly in reverse.” Figure 3: Switching cycles showing power, algorithm learning process, and continuous corrections for enhanced soft-switching. The AI in the CleanWave inverter evaluation system continuously adjusts timing conditions within the switching system required to force an echo to offset shapes Wave current and voltage This reduces switching losses, enabling step-down functionality at higher switching frequencies while improving inverter efficiency CleanWave200 at 99.3% at 100 kHz Published data plots system efficiency for switching speeds from 50 to 100 kHz, input voltage, output power , and current output, allowing system designers to compare pre-switch results with their requirements.Renouard highlighted that from an EV application perspective, the analysis allows for improved efficiency requirements, with a net improvement of up to 12% over EV range from the same battery size (Figures 4-7). From 50 kHz to 100 kHz, the efficiency is relatively constant, with a variance of about 0.2% (at 60 amps).” What makes this particularly important [is that] “All of these curves were made at 800 volts,” Reynward said. Figure 4: CleanWave200 system efficiency results before switching Figure 5: CleanWave efficiency at 100 kHz vs. amps per stage Figure 6: CleanWave efficiency at 100 kHz vs. power Figure 7: Switching pre-switching frequencies increase from EV range 5% to 12% . “We’re moving toward a new FPGA system on a chip that will enable us to run much faster than we’re doing here, and we’ll soon have a new CleanWave200 update,” Renouard said. High frequency inverter switching (Fsw) reduces motor losses. At 5 kHz, you have the worst case, as Renouard points out in Figure 8, for the Nissan Leaf IGBT silicon inverter. “If you go 20 times faster, you can see that the sine wave is very clean, without any additional output filters,” he said. Figure 8: High inverter Fsw reduces motor losses. Renouard said that this system can also be applied to GaN solutions. Although it hasn’t been tested yet, the system’s flexibility makes it doable. SiC and GaN switches share some similar properties; They both run very fast and stop very quickly.” “We think, in fact, that this silicon carbide algorithm will work pretty much the same as it does with GaN, and it will work with silicon IGBTs. But we will soon introduce versions of the algorithm that are optimized for each switching technology.” When considering the benefits of the motor, switching inverters in the application are always compromised by reducing switching frequencies in order to maintain high efficiency of the inverter. The result is a large amount of output ripple encountered by the motor, which Represents, in effect, the inductive heating that occurs inside the motor.Obviously, this heat must be dissipated, which is another cost.Renouard noted that a near-perfect sinusoidal output is provided by a 10× to 20× increase in the switching frequency, resulting in an improved Engine efficiency and reduced engine cooling required. “When operating at 100 kHz, engine efficiency is greatly improved at low torque and low to medium speed, where most of the driving is done,” Renouard said. % to 12% in the EV range by reducing switching losses.” The increase in motor efficiency also translates to savings in battery space as well as costs.