Power op-amp?

[b-rye-an]

Hi all,

I'd like to amplify a 1 volt peak to peak, 30 Hz sine wave I'm generating using a National Instruments USB adapter into a 24 volt signal for use with a fairly powerful electromagnet (it draws about 0.8 amps when hooked up to 24 VDC).

Is there an amplifier that will allow me to pass enough current to make this work? I tried talking to the folks at TI, but they weren't very helpful. Also, is there protection circuitry that I'll need to have to keep the back EMF from the electromagnet from damaging the rest of the system?

Any help would be very appreciated. Thanks.

 

[ronsimpson]

Need more information!
The output of the amplifier is it 0 to +24 volts. OR Is it -24 to 0 to +25 volts.
Try audio amplifiers. LM1875 for a start.

 

[b-rye-an]

The output of the amp is 0 to +24 volts. The input is 0 to +1 volts. Thanks for the tip about the audio amps. I'll check those out!

 

[MikeMl]

Also check out National's Power OpAmps, like the **broken link removed**

Try **broken link removed**, too. (Not sure it will work)

 

[ronsimpson]

If you are conserned about "back EMF" connect 1n4002 diodes from the output of the amplifier to the supply pins so the output can not get above the + supply or below the - supply.

 

[b-rye-an]

For the LM675 I'm a bit confused about how to set up the gain values. It looks like the resistor is on the output represented by R_L. Is that true? If so, wouldn't the gain depend on the resistance of the load? Thanks for the replies from everyone. This has given me a lot of stuff to try.

 

[ronsimpson]

See attached, gain is 21.
I would make +power=30V, -power=-5V

 

[b-rye-an]

How do you calculate the gain?

 

[MikeMl]

For a Power Opamp, you use resistors to set the gain, and then the load is connected between the output pin and the tap between the split power supplies. Since you want a non-inverting gain of 20, then you would use R2/R1 = 19, as shown below:

Note that I answered the question about the "back EMF", too. There is none, as long as the magnet is not suddenly open circuited...

If the input is driven by a 1V square wave, the 1 Henry magnet is driven by a 20V square wave, but the current increases/decreases with a classic exponential waveform, with no voltage peaking.

 

[MikeMl]

How do you calculate the gain?

For non-inverting configuration: Gain = 1 + R2/R1 = 1 +190K/10K = 1+19 = 20

 

[b-rye-an]

Thanks for the response. I guess my question about the diode was because an earlier version of this circuit used a FET. I had to use a flyback diode to keep the the FET's from burning up.

 

[MikeMl]

Thanks for the response. I guess my question about the diode was because an earlier version of this circuit used a FET. I had to use a flyback diode to keep the the FET's from burning up.

Yes, but the FET can only Sink current, then switch off. It takes a diode to provide a path for the magnet current after the FET switches off. The power-amplifier is a push-pull circuit, so it can both source and sink current, so no diode is needed. However, the negative supply pin of the amplifier must have -5V (or more negative) on it for the amplifier to do its work...