need help or advice for constructing a powerfull signal generator.

[MEGATRON]

i need help building a powerful signal generator or amplifier that can produce a clear sine wave signal from 1 to 20,000 hertz at 1000-2000 watts. i have an experiment i plan on doing involving infra-sound, and maybe later for powering a large acoustic transducer. the experiment involves using car sub-woofers, the "bling bling gangsta" ones that operate at 1000 watts then putting them in a hermetically sealed enclosure about 3/4 the size of a closet so that i can observe the effects of infra-sound. you may or may not know the amplifiers they make for car audio don't work well below 20 hertz, in fact i tested this and at 10 hertz the signal was distorted and the amplifier overheated immidiatly, i already have the sub-woofers and the enclosure. i don't want to spend thousands on an amplifier that can do this, is there any way around this?


maybe get a variable transformer then rectifying its output and using a signal generator to drive a IGBT so that it converts the DC from the variable transformer into a sine wave corresponding to the voltage output of the variable transformer and the intensity and frequency of the signal generator. however i don't know if that will work or not (don't get offended if this is not even possible, i don't know very much about transistors so im probably wrong). i need advice on this. help will be greatly appreciated!

 

[BobK]

No expert on this, but here are my thoughts.

1. I would be very skeptical about the 1000W ratings on your sub-woofers. This market is notorious for over-rating equipment. I doubt that they will handle anything like 1000W continuous.

2. To get 1000W into a 4 Ohm speaker, you need 63V RMS or 90V peak. (180V P-P). Paralleling the speakers to get 2 OHm or even 1 Ohm lowers the voltage but increases the current requirements.

3. At these power levels, I would only consider a class D amplifier in a bridge configuration. Essentially you would be running an H bridge using power MOSFETS and switching a large DC voltage of about 100V with a PWM signal that shapes the sine wave at the desired frequency. You would then need and LC circuit to filter out the switching frequency.

4. Alternately you could think of this an a DC to AC inverter with variable output frequency, and use that type of circuilt, which would involve an output transformer. But I think this is unnecessary complexity since you would still need a a power transformer as well. It seems to me that having two transformers is not better than having one.

5. Why do you set an upper frequency limit of 20KHz when you are talking about infrasound, did you mean to say 20Hz? I think you would have problems producing a good sine wave at 20KHz via a switching ampliier at these power levels.

Bob

 

[BobK]

In my first post, I was thinking only about the electronics. Putting sub 20Hz sound into a small closed box has it's own problems. Basically the box is much smaller than the wavelength, so you are not really creating sound waves that propagate, it is more like you are rapidly changing the air pressure inside the box. I am not sure that is the way to study iffrasound.

Bob

 

[MEGATRON]

thanks for the help bob, im looking into class D amplifier schematics, hard to find one that will go down to 1-20hz. however i came across something else, a variable frequency drive for electric motors, they can do 0.5-300hz at 0.75 KW. output is 200-240VAC at 4 amps 3 phase. at 1000 watts 8 ohms it needs 252.98 volts at 3.95 amps so i think this can work. i dont think it has a square wave output because the frequency is selectable and a square wave is many frequency's. not sure about the 3 phase output, maybe use 1 of the phases. what do you think about this?

 

[MEGATRON]

i dont really need 20,000 hertz, there are 4 10 inch sub-woofers with a 250 watt RMS rating, the idea is to concentrate a vast amount of infra-sound into a small chamber.