What kind of component should I insert here to convert 0-5V to 0-100V

[bill.connelly]

Hi,

I'm trying to build what a biologist would call a "stimulus isolator". That is, I have a voltage follower connected to a system that I want to zap with 0-100V, so that the voltage follower doesn't see the voltage (much).

I have an A/D board that can put out 0 to 5 Volts. I want this converted to 0-100 V. The thing that vaguely complicates matter, is I that the resistance of the output (the input impedance of the system this is driving) will be variable.

I keep thinking op-amps, but then I get confused about how the Earth would be connected so that the output was isolated.

Previously I've used a photovoltaic relay (PVA1354) to do similar things, but now instead of gating the voltage, I want to control it.

Does anyone have any thoughts.

I apologize if this is monstrously retarded, I'm just a humble biologist.

 

[bill.connelly]

Hi, I've had a few thoughts, how does this look?

 

[vk6kro]

You have the resistors the wrong way around. The bigger one should be across the opamp.

Normal opamps probably wouldn't handle 100 volts, but there are some audio amplifier chips that would go close. You would have to give them a split +50 volts--ground-- neg 50 volts supply.

And, sadly, optocouplers don't work like that. They need a small voltage to turn on and this produces light which makes a phototransistor turn on. This can then be used to switch an external voltage.
So, the input voltage is not passed on to the output.

 

[bill.connelly]

Yeah, I though this would blow your average OP-amp...
Double dang about the optocoupler... I thought it seemed to good to be true... thanks for your thoughts anyway.

Okay, say I was fixed the resistor lay out, and reduced the power voltage to 50V (making the resistors 1kOhm and 9kOhm). Is there anything that can simply isolate an input?

 

[vk6kro]

Okay, say I was fixed the resistor lay out, and reduced the power voltage to 50V (making the resistors 1kOhm and 9kOhm). Is there anything that can simply isolate an input?

Can't think of anything. It might be better to have an optocoupler being turned on by your 5 volt supply and then the output of this could turn on a transistor which controlled the 100 volt supply which you already have.

Doing this in a linear manner would pose a problem.

I hope you do not intend to harm small furry critters with this setup. I like furry critters.

 

[bill.connelly]

No, no... no furry creatures... cancer cells removed from a human decades ago...

This guy has posted something similar
http://www.nbb.cornell.edu/neurobio/land/PROJECTS/SIU/index.html
Except it seems very complicated. I can't figure it out enough to simplyfy it (I can't even figure out what the first transistor is for).

Also, he sais that Vcc = 6 volts, but he actually means 100V right?

 

[skeptic2]

The attached is a suggestion for a starter circuit.

 

[MATLABdude]

No, no... no furry creatures... cancer cells removed from a human decades ago...

This guy has posted something similar
http://www.nbb.cornell.edu/neurobio/land/PROJECTS/SIU/index.html
Except it seems very complicated. I can't figure it out enough to simplyfy it (I can't even figure out what the first transistor is for).

Also, he sais that Vcc = 6 volts, but he actually means 100V right?

No, I imagine that he means Vcc=5V (okay, he's probably powering this with 4 AAs, so maybe it is 6V). The output is 90V, however.

The first transistor is just to ensure that the input to the 4066 is around 6V or so (I believe):
http://www.fairchildsemi.com/ds/CD%2FCD4066BC.pdf

The beauty of the DCP0105DS is that it has a fully isolated output--the isolation is built in and the outputs are said to be floating. For that reason, the output from 3 of these (which are outputting 30V between the positive and negative rails) can be put in series to output 90V. TI seems to make this part now:
http://focus.ti.com/lit/ds/symlink/dcp0105.pdf

However, you'll note that the output amplitude adjustment happens at the output via the potentiometer.

EDIT: Reading through the description above the circuit diagram, I believe they state everything I just did above!

 

[bill.connelly]

Thanks Skeptic
I've added in the isolators where I think you wanted them, is this correct?

 

[skeptic2]

Yes, thanks.