You have to use special design transistors which can handle this high voltage which are expensive. Also, they must be able to switch the voltage from 0 to almost 2000 Volts in less than 12.5 us .I've read about high power audio amplifiers running with a supply rail of a couple hundred volts, but can an amplifier be designed to run with a supply of over 2000V and achieve an output swing as near as possible to that rail?
Oops... mistyped. I meant 2*Pi V/uS.Being band-limited to 1kHz sinusoidal, a 2000 Vpp signal requires a slew rate of about Pi V/uS.
Tubes are not an option here. Must be solid state. I'm wondering if there are BJTs that can handle this voltage. I know there are a few MOSFETs that come close. Is there a way to 'stack' transistors to spread the voltage drops evenly across multiple devices?I'm sure these water cooled tubes would fit the bill and give you lots of kW output.
Just to let you know, the filaments were 12V at 275 Amperes. Being you would need a class B or AB1 or AB2 operation, that's two tubes.
I can provide the specs for these ITT tubes ...
p.s. Let me know if you intend to use these tubes. I'd want to buy stock in your electric company.
This project is 'audio' in frequency only. I'm not driving a speaker. Distortion and linearity are less critical here than in normal audio.Possibly, Ebay might still have one of those old Electro-Voice speakers with the response rated from DC to blue light. Too bad it takes 3 train cars to haul one.
by Aaron Carman
by Jake Hertz
by Jeff Child
by Jake Hertz