The Idea
Google recently released the Littlebox Challange, a challenge for participants to design a 2kW inverter in the smallest volume they could (around the size of a small laptop). Stanford is currently working on an entry and this project investigates a potentially beneficial topology for this converter. In order to minimize volume in a converter, it has to be both efficient and run at high frequencies (upwards of 1MHz) to minimize the volume of the magnetics. However running at such high frequencies comes with several drawbacks. The first is "switching loss". This is the energy dissipated to turn on and off the buck switch which at over a million cycles per seconds can amount to tens of watts or more. Additionally, the cost of your components increases since they must be fast enough to handle the rapid operating speeds of the circuit, yet beefy enough to not catch fire. Enter soft switching. If the converter only turns off the FET when there is zero voltage or current across it, then there will be no switching loss! This is exactly what the quasi-resonant converter (QRC) does, eliminating the switching loss and lowering the stresses on the FET. As a result, the converter needs less cooling, can operate at higher frequencies (smaller inductors) and can have slower device turn-off times (cheaper parts).
The Goal
This project aims to demonstrate the advantages of soft switching in a 2kW micro-inverter and outline the feasibility of moving to such a solution. Our investigation will include the main buck converter design and component selection, initial layout of a PCB, detailed SPICE simulations, and designing a basic controller in MATLAB to regulate the system.
Google recently released the Littlebox Challange, a challenge for participants to design a 2kW inverter in the smallest volume they could (around the size of a small laptop). Stanford is currently working on an entry and this project investigates a potentially beneficial topology for this converter. In order to minimize volume in a converter, it has to be both efficient and run at high frequencies (upwards of 1MHz) to minimize the volume of the magnetics. However running at such high frequencies comes with several drawbacks. The first is "switching loss". This is the energy dissipated to turn on and off the buck switch which at over a million cycles per seconds can amount to tens of watts or more. Additionally, the cost of your components increases since they must be fast enough to handle the rapid operating speeds of the circuit, yet beefy enough to not catch fire. Enter soft switching. If the converter only turns off the FET when there is zero voltage or current across it, then there will be no switching loss! This is exactly what the quasi-resonant converter (QRC) does, eliminating the switching loss and lowering the stresses on the FET. As a result, the converter needs less cooling, can operate at higher frequencies (smaller inductors) and can have slower device turn-off times (cheaper parts).
The Goal
This project aims to demonstrate the advantages of soft switching in a 2kW micro-inverter and outline the feasibility of moving to such a solution. Our investigation will include the main buck converter design and component selection, initial layout of a PCB, detailed SPICE simulations, and designing a basic controller in MATLAB to regulate the system.