TIDUFD2 May 2025
The cycloconverter is a topology that converts one high frequency AC waveform to another AC waveform with lower frequency. This approach eliminates the usage of an additional intermediate stage with the high-voltage DC-Link. This can increase efficiency, reduce size, and save cost. This converter can be implemented in multiple combination as DC-AC, AC-AC, 3ph AC-1ph AC and other combinations. This reference design is focused on the single-phase DC-AC operation. Figure 3-3 shows the block diagram.
The considered cycloconverter consists of:
PWM (Pulse Width Modulation) is applied to all the power devices. On the AC side, two voltage levels are applied across the inductor. On the DC side, three voltage levels are applied. The power transfer is controlled by shifting the phase between the voltage waveforms on the primary and secondary sides.
Figure 3-3 shows that on the DC side, the cycloconverter consists of a traditional full H-bridge to generate a high-frequency rectangular AC waveform from the DC voltage source. This high-frequency AC voltage is applied to the high-frequency transformer. Due to the full-bridge structure, the DC side voltage has three levels: +VDC, 0, and –VDC. The formed AC waveform multiplied by the transformer ratio is applied to the left side of the inductor.
The AC side half-bridge is composed of two AC switches (S1 and S2). Each of the AC switches is composed of two unidirectional GaN power stages connected in common source. The common-source connection allows the AC switch to operate when the voltage of the grid is both positive and negative.
During the operation of the AC switch, one GaN device operates as rectifier and the second operates as a high-frequency switching device. The rectifier device is constantly turned-on for a half cycle of the grid voltage. The switching power stage operates on the same frequency as the DC-side forming high-frequency voltage across the inductor on the AC side. For example, when the AC voltage is positive, S1B and S2B are acting as rectifiers and are constantly turned on when S1A and S2A are switching. On grid zero crossing, the power stages exchange roles: when AC voltage is negative, S1A and S2A are constantly turned on and S1B and S2B are switching. To avoid shoot through, the switching devices have a dead time. During the dead time, both switching devices are turned off. Importantly, during this dead-time period, rectification devices provide a path for inductor current.
On the AC side, the voltage is following the grid and is always changing. The switching frequency for the inverter is significantly higher than the grid frequency. This allows consideration of the AC side voltage as constant voltage within each individual switching cycle.
Due to the half-bridge structure, the AC side voltage generates two level voltage: +VAC/2 and –VAC/2, where VAC is the instantaneous grid voltage level. The half-bridge structure also means 2 × higher RMS current and higher conduction losses than full bridge. Nevertheless, for lower power converters, using the half-bridge topology can reduce system cost. For high-power designs, the full-bridge structure can provide a significant efficiency advantage. Full-bridge topology gives several benefits such as lower RMS current, higher voltage gain, and better EMI. One additional benefit for the full-bridge configuration is that this configuration has an ability to have three level voltage on the AC and increased range of ZVS switching.
The cycloconverter topology can be simplified and considered as a dual active bridge converter (DAB). As an example, for positive grid voltage, S1B and S2B operate as rectification switches and these switches are constantly turned on. Thus, S1B and S2B can be removed from the consideration; therefore, simplifying the DAB schematic with S1A and S2A operating as a half-bridge on the AC side. In comparison to regular DC-DAB, the considered AC-DAB has a few challenges: