Pwm to Voltage circuit

Hi all, I need a circuit to vary a voltage with a 3.3V PWM signal. If the varied voltage was fixed I could simply use an RC circuit followed by a voltage follower. What complicates matters is that the voltage can vary from 10-35V but still needs to be varied from 0-35V by the 3.3V PWM signal. It only needs to source 10mA. It needs to be relatively simple (cheap). I have tried variations on the RC scheme such as a high side switch which outputs the PWM at the higher voltage, to an RC, to a voltage follower. Unfortunatly I cannot find a logic level high side switch, that can handle 35V, in a single channel package.

I am sure there is an elegant solution out there...

Thanks

Rich

Reply to
rich
Loading thread data ...

I am afraid I don't quite understand what you want to do here. So you have a constant 35V somewhere? Or not? Anyhow, if you just want to switch the high side you could use two transistors or a cheap opto coupler like this if there is enough juice on the 3.3V:

formatting link

If the 3.3V is wimpy you need to look for one with a better ratio.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

Do you mean that the supply voltage can be in the range 10-35V, but the output must always be 0->35V? In that case you need a boost converter in there as well as a PWM filter.

--
You win some and you lose some.
As long as I win the ones you lose that\'s fine by me.
Reply to
Hot Jock

Not 100% sure on what you're asking but I have a picture in mind here.

Take the 10..35v which I assume is DC and you want a constant controlled output from that via the 3.3 PWM control signal?

Assuming that. This is what I would do.

Use an 8 volt regulator or LDO 10 volt. This will drive a PWM circuit that is composed of a dual op-amp design or comparator depending on how you design it. One amp for the saw or triangle wave to drive the comparator which in turns creates a PWM signal that drives a logic fet driving a inductive boost inverter coil. The output is returned as a reference back to the comparator against the 3.3 PWM signal you have to regulate the output.

This is a generalized lay out but if you're savvy with circuits, I think you get the picture. A quad package comparator or op-amp may make it easier on the design since you could use an a unit to filter and buffer the 3.3 PWM you now have. Use a second unit as the comparator from the output of that with the return reference to generate a signal to drive the comparator for the PWM of the main circuit. and of course, the last one would be used for the triangle oscillator. I did all of this years ago via a LM324 chip.. I was here the other day looking for a switch mode driver chip when some one reminded me of a much simpler circuit that I had done years back.

--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5
Reply to
Jamie

Not sure this is what you want, but I think you want: NPN common emitter driven by the LV signal. PNP common emitter driven by the NPN collector. PNP emitter goes to the high voltage. Dont forget the resistor from PNP base to emitter, and current limiting resistor from NPN collector to PNP base.

Tam

Reply to
Tam/WB2TT

I made it confusing. There is a voltage, it can be between 10 and 35V (fixed). I have to vary this voltage from zero to whatever it is with a 3.3V PWM signal.

I have a new scheme. Two op-amps, both supplied by the voltage. Let's say in this case it is 35V. The first op-amp acts as a comparator with the positive input connected to the 3.3V PWM signal and the negative to .7V (created by a diode drop). Now we have 35V (ish) PWM signal which we feed to an RC circuit which feeds the created analog signal to the other op-amp which acts as a voltage follower.

Hard to describe in words... Works so far in spice...

Anyone know of a general purpose op-amp that will work with up to 35V on the supply? (Single ended)

Thanks

rich

Reply to
rich

"to whatever it is". Aha! That's different, thought it had to be always to 35V even if there is only 10V.

Tough. First, for best accuracy you'd need a rail-to-rail output. Next,

36V is often the abs max value, way too close to the limits for my taste.

I would take a look at audio amps, some of those can take 40V or more but will look like a sledge hammer to smack a fly. Then again, a sledge hammer can successfully smack a fly. To get full rail-to-rail behavior for your output voltage you'll probably have to design it in discrete, using FETs and a bootstrap supply (to be able to steer the upper one all the way to the 10-35V rail). Or use them drains tied together but that's more dicey in terms of stability.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

I am still going around with this. Right now I need an op-amp that will handle a 35V supply and be able to source 10mA. Rail-to-rail output would be ideal.

Am I dreaming?

Rich

Reply to
rich

--
http://focus.ti.com/docs/prod/folders/print/opa547.html

http://www.prema.com/pdf/op601vhv.pdf

http://eportal.apexmicrotech.com/mainsite/products/pages/op_amps/pa241.asp


Google "high voltage op amp" for lots more hits.
Reply to
John Fields

Yes, pretty much. There are some parts but you'll find that many only seem to exist on paper. Yeah, you may find them at Digikey but it doesn't do you any good if stock is zero. No RR outputs either, usually. Especially the former Burr-Brown appears to have substantial production problems these days.

Here is a specialty power opamp if this is a "one-off" project. Digikey has some but sit down and hold onto something when checking the price:

formatting link

And don't expect the shelves to be replenished the next day after you buy their whole stock.

There is an unorthodox trick to use the supply rails of a lower voltage opamp to steer the bases of a couple of output transistors. That creates an amplifier that can generate higher voltages than the opamp can handle. Take a look at page 256 of "The Art of Electronics", bottom left.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.