Trimmable current sources

[Radical Edward]

I'm currently using this simple circuit to power an organic LED that I want driving at constant current, using a 330 ohm resistor (so I supply 0.2mA)

 

 

putting the OLED in the position of R_L

 

Now if I want to be able to change the current, I suppose I can just put a variable resistor in there, but what I really want to do is to have a single control that allows me to change the current for a whole load of these circuits with OLEDs in them. I guess I could just put the OLEDs in series, but as I am testing devices, if they're in series and one goes, then they all go and my experiment is ruined. Does anyone know how I might construct such a circuit?

[lesolee]

The usual simple way of driving an LED at "constant current" is to just put a resistor in series with it to a power rail. So if your LED has a nominal volt drop of 2V at 1mA you might power it from a 5V rail and drop 3V across the resistor at 1mA (3000 ohm resistor). The volt drop across the LED is going to be pretty stable so the current will also be stable. There is no need to get complicated about this unless there is some unusual circumstance you haven't mentioned. Obviously the higher the power supply voltage you use, the more stable and known the current will be. One resistor per LED will give a pretty cheap solution and then you can use an adjustable power rail to change all the currents at once.

[Radical Edward]

Well it is a bit unusual - as we're using OLEDs, devices from different areas of the glass substrate have slightly different thicknesses and operate at different voltages, however the current which flows at a given brightness is always the same. Hence if we have a fixed voltage supply of say 3V, then all the devices will be a different brightness, but if we have a fixed current supply of 0.2mA, then they will all have the same brightness. That's the reason that I am using the circuit above. The problem is that sometimes I want to run a load of devices at 0.2mA and other times 0.3mA and so on, and trimming every device manually would be an utter pain.

[lesolee]

Well it is a bit unusual - as we're using OLEDs, devices from different areas of the glass substrate have slightly different thicknesses and operate at different voltages, however the current which flows at a given brightness is always the same. Hence if we have a fixed voltage supply of say 3V, then all the devices will be a different brightness, but if we have a fixed current supply of 0.2mA, then they will all have the same brightness. That's the reason that I am using the circuit above. The problem is that sometimes I want to run a load of devices at 0.2mA and other times 0.3mA and so on, and trimming every device manually would be an utter pain.

Well ordinarily that wouldn't be an issue, but as I don't know the details of these devices I have to assume that when you say "different voltages" you mean something significant like 2V across one and 3V across another. Obviously the more data I have the more I can help you.

 

This next idea is at increased cost/complexity but improved current matching.

 

Obviously you can get better matching accuracy by using a higher base drive voltage and you could add temperature compensation either using a diode or a transistor base-emitter junction or wrapping an op-amp around one of the drivers. If any of this needs more explanation then feel free to ask.