DC to 100Hz Power Amp - Coil/Loop Driver

You might get a square wave across the 3 Henrys. I don't know. But even if you do, you still need to divide your 48V in half, cutting your power to 1/4 of 1.92W. I assume the signal you are thinking about swings up to +24V back to 0V and then down to -24V. So it is at +24V half the cycle and at

-24V the other half of the cycle. OR at 24V the whole cycle, thus you need to divide your 48V by 2. The average voltage is 24V. If you have an actual sine wave then you need to multiply by 0.707.

24 v 0.707 = 16.968 which is why I said 17 V. Mikek

Give us a clue what you're doing with it.

NT

Or drive it with a bridge.

** No it will not.

Very unlikely to have full output below 70Hz.

** Resonating the coil will involve a *series* capacitor. 3.3uF gives 50Hz.

.... Phil

Oops, sorry, you did say DC to 100Hz! The amplifier's transformer likely isn't good below 25Hz or so. :-(

Anyway, you don't need as much voltage as my suggested solution would provide. The currents are low, and many DC op- amp circuits can do the job.

--
 Thanks, 
    - Win

Yes, I neglected to factor in the waveform. Thanks for explaining.

I am applying a complex waveform with a lot of "on" time, so in terms of duty cycle it would probably be between a square and a sine.

Kevin Foster

And 33uF for 5Hz etc. That would be a non-polarized electrolytic?

Still not sure what the suggestion for a "ballast resistor" was for. The coil is already 1.2K.

Kevin Foster

I am experimenting with the effect of low frequency EMF's on plant germination. No, not those kind of plants.

Keven Foster

** The calculated value is 330uF.

But there is no point, the Q of your circuit at the frequency is way too low at less than 0.1.

.... Phil

An mostly-ordinary audio amplifier, one with bipolar power supplies, such as +/- 35 volts, should be suited for the DC-response trick. These amplifiers don't use output capacitors for the speakers, but they generally have one at the input. So short that.

--
 Thanks, 
    - Win

krw prodded the keyboard with:

NOTE:------------------------------ 0.4mA

Here-------0.04A Still only 0.96W

--
Best Regards: 
                      Baron.

It's been a decade or so since I seriously looked at Class D amps, but back when I did it seemed that the vast majority of them were designed for all-digital signal chains, with a proprietary interface between the amp and the preceding block.

If you wanted to go class D, it might be easier to build your own switching amplifier -- for someone with reasonable embedded SW chops there are some very fine microprocessors designed for this task, and some reasonably good eval boards that could be pressed into service if you're doing a one-off. With a 100Hz "I don't care" frequency, you could keep the switching frequency fairly low if you wanted to.

--
Tim Wescott 
Control systems, embedded software and circuit design 
I'm looking for work!  See my website if you're interested 
http://www.wescottdesign.com

damn, and I was going to offer you some help!

Jamie

Don't think he's looking for that kind of "help".

All of the all-digital ones I know of have an I2C (or some variety of TDM) interface. There are analog-input Class-D amplifiers, as well.

You probably should use some sort of current feedback to have better magnetic field stability (particularly as frequency is varied) than you'd get with voltage feedback. I presume you'll have the plants within a pair of Helmholtz coils? Will you have some kind of magnetic field calibration?

Glad to help, and happy I explained so it could be understood. Mikek