Trouble with IGBT Circuit

[mpicco]

I am having an issue with what I thought was a simple circuit. Here is how the circuit is set up:

I have a 120VAC line with a resistive load (light bulb) that passes through a full wave bridge rectifier. The rectified voltage is then passed through an IGBT to turn on and off the load. The IGBT is controlled by a microchip MCP1405 mosfet driver with a 500 Ohm gate resistor between the output of the driver and the gate of the IGBT. I am using 500 ohms since I do not need to switch the IGBT very fast (120Hz). The input to the driver is a 3.3VDC input. The source voltage for the driver (and the input to the IGBT) is 12VDC. All of the grounds are tied together (3.3VDC, 12VDC, and the rectified 120V). When I change the input to the mosfet driver from ground to 3.3VDC it turns on the driver and IGBT fine. I also turns off the driver and IGBT when I go from 3.3VDC to ground on the input to the driver. When i try to have a PIC uController control the driver it does not work.

I noticed something wierd. When everything is connected the 3.3VDC bus is actually a sinewave centered around 3.3V (2.8V min and 3.9V max) - not 3.3VDC as it should be. When i disconnect the circuit the power supply produces a perfect 3.3VDC voltage. This circuit is currently on a breadboard with no decoupling caps - could this be the issue. I am confused as to how the sine wave is appearing on the 3.3V bus. Any ideas? How I found this was by putting a probe on the 3.3V bus and the ground.

Thanks in advance!!

Mike

 

[Miles Prower]

"This circuit is currently on a breadboard with no decoupling caps - could this be the issue. I am confused as to how the sine wave is appearing on the 3.3V bus. Any ideas?"

That's your whole problem right there. IGBTs like to oscillate, especially with breadboard construction. It probably could use a gate stopper. Try a 1K resistor mounted as close to the gate terminal as possible.

 

[mpicco]

Thanks for the quick response. The IGBT is not oscillating but it seems to be the cause of an sine wave on the 3.3VDC bus. Is that what you understood from my description? If so does the gate resistor go from the gate to ground? I had a 10K resistor from the gate to ground earlier but that didn't help.

thanks again,

Mike

 

[Miles Prower]

A gate stopper of about 1K or so is connected in series with the gate, and installed as close to the gate as possible.