Photodiode, photoresistor, or phototransistor

Thread Starter

coinmaster

Joined Dec 24, 2015
502
Hello, I'm trying to figure out what would be better for a digital voltage control.
Photoresistors are resistors so that's an obvious choice but optocoupled LDRs are expensive and/or difficult to find.
Phototransistors adjust current which in turn adjusts voltage but they don't work with negative voltage right?
Photodiodes I haven't figured out yet.
 

tracecom

Joined Apr 16, 2010
3,944
Are you trying to switch a voltage on and off based on light levels, or are you trying to adjust a voltage level based on light levels?
 

dl324

Joined Mar 30, 2015
16,839
Hello, I'm trying to figure out what would be better for a digital voltage control.
It would help if you gave some design context to bound solution space.

Phototransistors adjust current which in turn adjusts voltage but they don't work with negative voltage right?
Phototransistors respond to photons; so as long as the transistor is properly biased, negative voltages can be used.
 
Last edited:

Thread Starter

coinmaster

Joined Dec 24, 2015
502
Are you trying to switch a voltage on and off based on light levels, or are you trying to adjust a voltage level based on light levels?
I want a digitally controlled potentiometer so I can adjust power supply voltages for voltages around +/- 300v.

I realized that photoresistors are not good for high voltages so that leaves phototransistors and photodiodes. I'm not quite sure how a photodiode works. Is it like a variable zener?
 

tracecom

Joined Apr 16, 2010
3,944
An optoisolator is just an on/off device that provides a means to switch one circuit with another while providing galvanic isolation. From the limited information provided, I think you need a circuit that responds to changing light levels with a linear output. That linear output would then control yet another circuit that is able to handle the voltage levels at whatever current you require.

There are lots of people that post here with ideas that they want to keep secret. Sometimes they succeed in getting what they need, but often they don't provide enough information for eliciting the help they want. Oftentimes, better help is available if the overall goals are stated instead of asking about specific components.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
An optoisolator is just an on/off device that provides a means to switch one circuit with another while providing galvanic isolation
Are you saying the above examples will not work? They do fine as variable voltage control in the simulations.

That linear output would then control yet another circuit that is able to handle the voltage levels at whatever current you require.
That is what I intend, but the fact is I still need a digitally controlled resistor for a divider to adjust the power supply voltages.

I realized that photoresistors are not good for high voltages so that leaves phototransistors and photodiodes. I'm not quite sure how a photodiode works. Is it like a variable zener?
I'm still wondering about this.
 

dl324

Joined Mar 30, 2015
16,839
I want a digitally controlled potentiometer so I can adjust power supply voltages for voltages around +/- 300v.
Optical devices are probably a poor choice if you intend to use them as analog controls.
I realized that photoresistors are not good for high voltages so that leaves phototransistors and photodiodes. I'm not quite sure how a photodiode works. Is it like a variable zener?
A photodiode is operated reverse biased and converts photons to current.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
What other options are there? Mosfets and BJTs can't be used as high voltage linear resistors across a huge range of resistances.
There's an entire section of ultralinear photodiode optocouplers http://www.mouser.com/Optoelectroni...gh-Linearity-Optocouplers/_/N-6x5jj?P=1z0iwb4 but I still don't understand how photodiodes are used.
Photodiodes are operated reverse biased, but they're not used for voltage regulation.
Why?

LEDs aren't linear and the CTR of phototransistors (et. al) isn't constant.
You said the reason optos aren't good is because the CTR is not constant, I looked it up and this is only the case as they age and with changing temperature, It wouldn't be too difficult to compensate. Also the LED linearity is internally compensated with feedback in many designs so that is not an issue.
 
Last edited:

crutschow

Joined Mar 14, 2008
34,280
Photo diodes and phototransistors act in a similar fashion, converting the photons to a current, with the phototransistor having more gain. A phototransistor is similar to to photodiode whose output is driving the base of a transistor.
So a phototransistor optocoupler acts similar to a transistor by itself. The output current increases in proportion to the input current with the gain given by the opto transfer function listed on the data sheet.
[The device you listed has a gain of 40 (4000%) and a Vce breakdown of 350V so it should work in your application. The reverse Vce of 0.6V is a problem only if you reverse bias the transistor].
It's nothing like a variable zener.

So you use a phototransistor just like a regular transistor with a resistor in the collector or emitter to convert the current to voltage.
The resistor in the emitter will give in increase in output voltage with input current, and the resistor in the collector will give an inverted signal.
Make sure you keep the transistor output resistor at a high enough value to limit the power dissipation in the opto output.
You should probably limit it to about 150mW, which means the output resistor should be no smaller than 150kΩ for a 300V supply.

If you show exactly how you expect the output will control the supply voltage, we can suggest how to best do the circuit since making an optocoupler act like a potentiometer is not simple.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
If you show exactly how you expect the output will control the supply voltage, we can suggest how to best do the circuit since making an optocoupler act like a potentiometer is not simple.
Well one of the designs will be a series regulator in which I was thinking of using something like this

to determine the voltage output of the regulator, I haven't determined the details of the regulator itself yet, probably a power transistor with the "zener" at its base.
The other circuit is

with the variable resistor being R3.

I'll also have a variable resistor controlling a CCS but I haven't decided the exact CCS design yet.

The reverse Vce of 0.6V is a problem only if you reverse bias the transistor].
That means my earlier example of 300v on the emitter will not work then right?
 
Last edited:

crutschow

Joined Mar 14, 2008
34,280
So when you decide which circuit(s) you are actually going to use, post that.
It looks like you are wanting some sort of constant-current source, not constant voltage(?). :confused:

Your example shows -300V on the emitter, which is the same as +300V on the collector, so you are okay.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
It looks like you are wanting some sort of constant-current source, not constant voltage(?). :confused:
Not sure where you are getting that idea, one is a variable zener circuit and the other is effectively a CVS at DC.
Both require a variable resistor for the dividers in order to be adjustable.
 

crutschow

Joined Mar 14, 2008
34,280
Not sure where you are getting that idea, one is a variable zener circuit and the other is effectively a CVS at DC.
Both require a variable resistor for the dividers in order to be adjustable.
I guess because the second circuit describes it as a constant current source.
So how would you adjust the second circuit?
Is this to be some sort of variable load for the plate of a tube?
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
So how would you adjust the second circuit?
R3 sets the voltage.
Is this to be some sort of variable load for the plate of a tube?
No, it's just a gyrator load. There's plenty of gyrators to choose from, I just picked that one because I felt like it. The important thing is that it acts as a CVS, I'm converting it to be adjustable, I have a specialized use for it.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
Hah, I forgot about that thread, I feel like you are showing a picture of me in high school with 80s hair. Yeah I am that coinmaster, I made that thread while I still knew crap about electronics (in the grand scheme of things I still do but I know x10,000 more than I did then). Some of the people from another thread kept wanting me to make it so I did.
I've spent the last 3 1/2 months studying up.
The plans, ideas, and thruths from that thread have shifted almost completely. My current project is vastly different I should probably ask a mod to delete it.
 
Last edited:
Top