big cruddy amp

Discussion in 'The Projects Forum' started by madjack, Feb 14, 2012.

  1. madjack

    Thread Starter New Member

    Feb 14, 2012
    Ok, so I suck at electronics; I am learning but please forgive me if everything I say in this post sounds totally stupid.

    The project I am working on requires that I drive an electromagnet (~ 3 - 5 ohm) from a fairly small logic level input signal, -5v to +5v, 250 MA. I would like the output signal to be on the order of -24v to +24v 20A continuous output. While I think this project is similar to an very big audio amplifier, I am in fact not driving a loudspeaker. The general gist of the project is to drive ferrous objects (not speaker cones) in time to music. A simple example would be making a iron ball on a string swing back and forth in time to the bass signal in a song.

    Because the output of the system is not a sound wave, I don't care very much about distortion at all. What I need is a simple push-pull amp design with gain, gain, gain. From my early electromagnet experiments I have realized that because of the inverse square law moving objects from any distance at all requires a very powerful magnet. To give you some idea, the power supply I made for this project is built around this transformer.

    And yes, I realize that pushing 600 watts continuously though a electromagnet is going to get very hot very fast. The electromagnet is liquid cooled.

    I am slowly making my way through "Designing Audio Power Amplifiers" by Bob Cordell but it dense material (for me a least) and I am having trouble figuring out what exactly I can/can't do without.

    My questions are:

    What is the simplest circuit I can get away with?
    What transistor / FET pairs can handle this sort of power dissipation even with a cooling fan?

    Thank you for any help you wish to offer, even if it is simply to tell me that I am going to electrocute myself.
  2. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    The problem you will encounter is mass and the Laws of Motion.

    The more massive the moving object the closer to impossible it becomes to reverse its direction of travel instantaneously. Moving a metal ball any visually detectable amount at anything close to an audio frequency will be difficult for almost everyone.

    A 'beat' detection circuit could generate pulses and the resulting output could push an oscillating ball to keep time with a song. Anything faster than that and you'll be up to your neck in mechanical/physics type problems. Many of which are not solveable with electronics
  3. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    OP's link is a virus, I think.

    I am getting a freakin DLL file
  4. williamj

    Active Member

    Sep 3, 2009
    I got a Digikey page on OP's link.
  5. madjack

    Thread Starter New Member

    Feb 14, 2012
    I'm sorry if the link is somehow malformed. It should go to a digikey page.

    Yes, I understand that I can only move a large object at a low frequency. I am already doing a fourier transform to pull off just the very low frequency bass signals. The +5 volts is the output of that system. I have a system right now using a single N channel mosfet to amplify the signal. It works reasonably well but is only a "push" amplifier and does not ever "pull". I want both. I have also made a push / pull system using relays that works ok, but I really want to be able to scale the power output more then just on or off. Because I suck at analog electronics it is taking me a long time to translate the delicate audio amplifier circuits I read about into something more beefy but less accurate.