Building a DRSSTC Pt. 5 - H-Bridge Design

Hello everybody, and welcome back to my blog!

Today I'm going to discuss the bridge. This is the part of the circuit that actually drives the series tank circuit that is made up of the primary coil and the capacitor. In order for the coil to operate, the primary tank circuit needs to be supplied with an oscillating current in order to resonate. This is mainly done via 2 or 4 transistors switched in such a way as to continuously reverse the current in the primary tank circuit. The most common type of transistor to use is the IGBT. While MOSFETs may work, their losses will be much higher due to their on resistance--as current increases, the power increases exponentially, as seen in Ohm's law: P=I^2*R. However, IGBTs only have losses based on their internal diode drop, and thus the power only increases linearly with current (P = I*V). I plan to use IGBTs for this project, so when I reference the bridge transistors, those are what I am referring to.

There are two main methods to drive the primary tank circuit: 1) The half bridge, or 2) The full bridge.

The half bridge utilizes two transistors switched alternately and two bus capacitors to create an oscillating voltage across the primary. The schematic is as follows:

87547

While the part count is low and the half bridge design is generally cheaper to build, it probably isn't suitable for high current builds unless you purchase IGBTs designed for very high currents. It is also very important that you include bleeder resistors across the bus capacitors, as they can store a lethal charge. Now, please note that the above is purely representative. You will need decoupling capacitors and return paths for the transistors, as they are switched based on the Gate-Emitter voltage. We'll get to this later when we look at the final schematic.

The full bridge, often referred to as an H-bridge due to its shape--is commonly used in DC motor drivers to change their direction. In this case we'll be using it to change the direction of the current through the primary. The full-bridge uses 4 transistors instead of only two, and eliminates the need for bus capacitors. This is the design I plan to use for my DRSSTC. I put together this simple animation of how current flows through the primary based on how the transistors are switched:

87549

As you can see, transistors at the opposite corners of the bridge are switched together, and this reverses the current through the center part, which in our case contains the primary coil and capacitor. This, in turn, causes the primary circuit to oscillate at its resonant frequency, which is exactly what we're looking for.

As mentioned before, I will be driving my bridge with 170VDC, which is rectified and filtered 120VAC mains. Your IGBTs should be able to handle voltages must higher than that, at high current. Their switching times should also be as low as possible, preferably turn-on times of only a few tens of nanoseconds, and turn-off times of no more than, say, 120nS. In general, you want to minimize these values as much as possible. I'll discuss the IGBTs I chose more in a future post.

In my next entry I will describe the driver circuitry that controls the H-bridge, and the process for designing a functional gate-drive transformer.

'Til next time!