PWM to 1V out supplying at least 200 mA

Dear Forum,
I need to control a device with a 0-1V input that supplies at least 200 mA of current (500 mA max). I have limited electronics knowledge and have scoured the net to find a simple solution but as far as I can see, most voltage regulators tend not to work at this low voltage range.
Can any one help with a circuit?

I would like to do this using the same voltage source that supplies the arduino so I am looking to use a single rail, low voltage power supply.

I know how to convert PWM to analogue out and how to use an OP amp to buffer the output but this is not sufficient current for my needs. I need to know how to wire up a suitable voltage regulator circuit to supply enough current.

If it is easier, I could live with a system that just switches on 1V rather than modulating between 0 and 1 but would prefer the modulation if it is reasonably straight forward.

Thanks,
Nick

Do the pwm and then buffer the output with an audio power amp chip, like LM386.

Thanks for the suggestion. The data sheet for this chip suggests the gain is set to a minimum of 20. Is there a way to ensure that the gain is actually reduced since I only want to go from 0 to 1 Volt. If not, the accuracy controlling it with PWM will be very low since I will be working with only a 50 mV voltage from the PWM output to get to 1V.

Perhaps I have misunderstood how to wire up the audio amp.
Thanks,
Nick

NickH:
I need to control a device with a 0-1V input that supplies at least 200 mA of current (500 mA max).

What is the device?

Hello,
It is an acousto-optic modulator that controls a laser. I found that a Nidaq card can't supply enough current and this control is important because it determines the relative transmission of a laser through the device. I need maximum transmission hence have to make something that supplies plenty of current.

Best,
Nick

As long as you have a PWM or AC signal you might be able to use an audio transformer to trade your voltage for current. A 5-1 winding ratio would divide your voltage by 5 and multiply your current by 5.

Since you are just talking high and low you could also do it with transistors and a separate 1V supply.

Also remember that because something like a buffer amp has very high input impedance you can use a simple resistor voltage divider to attenuate the input as much as you need.

Is there a way to ensure that the gain is actually reduced since I only want to go from 0 to 1 Volt.

In that case, you can use a power opamp that you can configure the gain (tda2030 for example), or use an opamp + a buffer. Both are fairly easy.

I need maximum transmission hence have to make something that supplies plenty of current.

Then none of the suggestions so far will work.

You need to feed the filtered PWM output into the gate of a FET with the load connected between the source and ground. Connect the drain to your 500mA capable power supply. The voltage does not matter but the higher it is then the hotter the FET will get because it will be running in linear mode and be burning off the power not used by your device.

or use an opamp + a buffer.

A simple opamp buffer would consistent of an opamp, 1 base resistor (optional) and an output device (npn or n-channel) rated for this. Connect the opamp's output to the npn's base through the base resistor. npn's collector to your power source. npn's emitter to the load and the opamp's inverting input. opamp's non-inverting input to your pwm output signal.

Another solution that you may consider is a DAC (buffered), or a digital pot (buffered).

A simple serial DAC for example would be HC595 + a R2R network.

I can't see what the op amp will be doing for you, the voltage is sufficient already.

I can't see what the op amp will be doing for you, the voltage is sufficient already.

It is not about the voltage. It is about impedance matching and linearity.

Look I don't know what you think you are doing but in this and several other threads you have been less than helpfully and less than knowledgable about electronics.
The impedance matching is done by the buffer not the op-amp!

To the OP, I can post a schematic for you if that helps.

Also, a NPN-PNP pair can do this as well but with a dead zone at the low-end, which may or may not be acceptable for your application.

Assuming that your load is to be connected between the 0-1V output and ground, then I would use an op-amp feeding an NPN power transistor configured as an emitter follower. See attached schematic. The power transistor will dissipate up to 2.5W so it will need a heatsink.

Nice schematic by dc42 - it graphically reproduced what I described ealirer - sorry that it confused some poster.

The output voltage on the emitter will be the same as the input voltage on the opamp's non-inverting pin.

Howerev, it is not without potential issues:

  1. You want a base resistor to save-guard the opamp. It also helps with stability should you use a mosfet there.

  2. For opamps that cannot swing with 07v of the negative rail (gnd in this case), you want to use a darlington or a diode to pad the Vbe. More diodes to pad the Vbe more. This will linit the max positive swing, however.

  3. As is, the output current is determined by the input voltage + load resistance. To the extent that the load has non-linear v-i characteristics, you can float the load to the collector or introduce a sampling resistor - this type of circuit is often used in a current loop trnasmission.

Hope it helps.

@dhenry

  1. You want a base resistor to save-guard the opamp. It also helps with stability should you use a mosfet there.

Wrong on both counts.

Hope it helps.

I have yet to see thread where you do help.
Your advise is flaky at best, I suggest you finish learning about the subject before offering advice here.