Author Options:

Half bridge in "A reliable plasma speaker doesn't work"? Answered

Hi! I am a  student and I just started to make my first half-bridge plasma speaker, and I used the schemetic in https://www.instructables.com/id/A-reliable-plasma-speaker/
However, after I finished, it didn't work. I used an oscilloscope and I found the VGS of high side MOSFET was between -3 and -6, and the low side was between 12 and 0. Then I accidentally got it work, but the MOSFETs were very hot, and then they exploded. After I change the MOSFETS, it had the same problem as before again. But if I short the capacitors, the wave form for both the high and the low were correct. can anybody explain why? Thanks.


Its too late now, but you should always be very cautious firing up this kind of device for the first time. Its critical that the gate firing is OUT of phase - or you will do what you have done, which is crowbar the supply and destroy your devices.

Also, check the oscilaltor stage - what frequency is it running at ? How does the modulating frequency affect the waveform ?

If you read my most recent Instructable, there are clues there to why your transistors can melt. https://www.instructables.com/id/Flyback-transformers-how-and-why/

no, the phase is right. The frequency is about 30kHz, and the wave form stays the same as I modify the frequency. And I found only the high side didn't work, and when it was working I could make arcs, so I don't know what's happening.

You're firing BOTH gates at some point in the cycle. Take a look at your trigger transformer - or use a proper bridge driver chip !

Emm..but my oscilloscope cannot probe two MOSFETs because the ground for two probers are connected... but now even if I take the MOSFETs away I still get negative VGS for the high side MOSFET. Now when I close the switch and the MOSTETs are very cold, and the ampmeter shows a current about 0, so it's not working. Unfortunately my oscilloscope is at home, so I can't show you :(