[SOLVED] igbt/mosfet h-bridge shorting, missing something basic?

[cslarson]

The eventual aim of my project is a constant V/f inverter that can use multiple renewable energy power sources to run 1 or 2 (initially single phase) water pumps. The output frequency will vary depending on the power available in the wind as calculated simply by the rectified dc buss voltage. I have worked out how to control the h-bridge using 2x msp430 launchpad controllers as well as 2x ir2184 half-bridge drivers. One controller determines the frequency and the other controller is responsible for the sine wave pwm output to drive the h-bridge. Yes, it is quite possible this project is far beyond my capacity. I worked for a number of years in the developing world and believe such a controller would allow for cheap, easily available water pumps to be used with renewable energy power sources such as wind and solar, without a battery bank. Power output for the controller will be around 3-4kW max

The problem I am having is with my h-bridge. Initially I used the IRFP460 mosfet. When I connected the h-bridge to the power source the mosfets failed even though no load was connected and no gates being driven, just the negative to the ground side sources of the h-bridge and the positive to positive, high side drains. The power source was rectified 220vac, though, so may have been too high for those mosfets. I have since bought the following IGBT module: F430R06W1E3 IGBT,MED POW,QUAD WITH NTC,600V,30A. I am also not longer attempting to connect such high voltages to anything I make until I have more of an idea of what I am doing. Clearly I am missing something, though because the module draws current and warms up even when I connect to a 40vdc power source I have (+ to high side collectors, - to low side emitters). No gates are connected to anything. Is there something fundamental that I'm missing or not seeing from the schematics of h-bridges?

Thanks very much for any assistance with this.

 

[cslarson]

A previous post suggests putting a 1k resistor between the gate and source for all mosfets in an h-bridge to prevent accidental turn-on. I can try this with my igbt h-bridge module.

 

[FvM]

No gates are connected to anything.

You should never operate IGBT or MOSEFTs with floating gate. The transistors will switch on arbitrarily and may be permanently damaged by electrostatic and other voltages.

 

[Orson Cart]

FvM is quite correct, floating the gates of an igbt/mosfet will allow them to turn on by themselves (this is not good). Also if you have significant capacitance across the DC rail of your converter then you will need a soft start resistor and shorting relay when connecting to the mains, otherwise when you switch on there will be large currents through the internal diodes of the devices as the mains charges up the DC rail caps, certainly large enough to destroy the IRFP460 you mention, these are only good for about 10A at 100 degC at say 35kHz switching (even with very good heatsinking) so IGBT's are a better choice for the powers you want to get to - good luck..! Regards, Orson Cart.

 

[cslarson]

Thanks, and thanks again for the additional tips.