Hi,
am I correct in thinking the CTR for an opto coupler is basically a scale where the more current on the input side allows more current to flow on the output side up to a certain level. After this level the device is fully one and the current on the output is constant regardless of the input current.
If the above is correct does this mean you can control how much the opt coupler is fully turned on?
"Turned on" doesn't have much meaning in this case. Optocouplers are not digital devices, although they are often used as such. If more current flows through the input LED, more current flows through the phototransistor, up to a point.
Meant to say fully turned on.
Your reply answered my question and confirmed my understanding.
Thanks
Yes that is right.
The main problem is that the transfer is not linear, by a long way.
You can make it linear if you have an opto with one emitter and two transistors. You use the second transistor in a feed back loop with an op amp on the input side to linearise the transfer function.
MartynC:
If the above is correct does this mean you can control how much the opt coupler is fully turned on?
Yes, but it's probably very non-linear and there's probably a lot of variation between individual devices (just like ordinary BJTs). Don't count on having any particular ratio of input->output.
If you want some control, try a H11F1 opto FET.
See As a Variable Resistor or Atomoatic Gain Control section on the data sheet.
https://www.fairchildsemi.com/datasheets/H1/H11F3M.pdf
Also see Resistance vs Input Current graph.
Not really that linear, but not to bad.
The "up to a certain level" in the various answers, is set by two parameters, the safe limit on how much current can be supplied to the LED, and the bias conditions on the transistor. As long as there is a reasonable voltage across the transistor, more current supplied to the LED will produce more current in the transistor - but not necessarily linearly.
In almost all circuits, you have a load resistor on the transistor, so that when it passes a certain current, the voltage drop across the load resistor is such that there is no remaining voltage across the transistor; the transistor "saturates" at that point but a lower value resistor would permit more current to be drawn.
Clearly, the only limit is how much current can safely be carried by either component. The situation is in fact no different to asking at what point supplying more current to the base of a bipolar transistor fails to result in more collector current.