I need to control an LED driver with analog 0(1)-10V control input.

This control device should be very cheap and compact so until now I made a schematic which uses the fact that those drivers have 100 uA current source at its dimming input. Switching capacitor made this cheap, small and relatively easy.

But now I need to adapt this solution for a driver which has "400...550 uA current at control pin". This is a problem for several reasons:

  • I need to drive about 5...10-times more current in the primary circuit which becomes 4-5 mA (I don't have so much current)
  • Switching capacitance schematic gives a voltage depending on a current. So if this current isn't precise - my voltage will be not precise as well

I'd like to avoid any isolated DC-DC with opamps solutions as this will be big and expensive.

Is there any better solutions?

  • \$\begingroup\$ How about PWM via opto-isolator? \$\endgroup\$
    – JimmyB
    Apr 21, 2016 at 11:37
  • \$\begingroup\$ @JimmyB It would be perfect if I'd know how do I convert this PWM signal to the needed voltage (taking into account the fact that I don't have a voltage source on the secondary side) \$\endgroup\$ Apr 21, 2016 at 11:46
  • \$\begingroup\$ In the other question, you quoted that the driver could be controlled via PWM too. Is this one different? \$\endgroup\$
    – JimmyB
    Apr 21, 2016 at 11:48

2 Answers 2


I'm not sure I'm following what all of your constraints are, but my first thought would be something like this:


simulate this circuit – Schematic created using CircuitLab

This is similar to the previous idea, but it now uses an LMV431 shunt regulator to deal with the variability of the current source. If the switching frequency is zero, the control voltage will be 1.25V, and as you increase the switching frequency, the control voltage will rise.


I initially just wanted to comment, but I don't have enough rep. So I have to go with a full-blown answer.

You could have a look to the devices like ADuM5240. They are digital isolators (two I/Os) that also integrate an isolated DC-DC converter in a single device. So you send both the PWM and the 5V supply to the other side. Then you can do pretty much what you want.

Here is what I came up with. It goes from almost 0V (VCEsat of NPN) to 10V (it could even go above it with a different R2/R3 divider configuration), and its behavior is totally independant of the current source value. The total cost of this may not be more expensive than Dave Tweed's solution, depending on what you use for the "U1 isolation/SW1/SW2" blocks. Note that I think any basic opamp can do the job here. (Maybe you could even use an open-collector comparator instead, and getting rid of the transistor, provided that the comparator can handle 10V at its output with only a 5V supply, and that it has some linear region - which is not usually what a comparator is for, but...)

enter image description here

  • \$\begingroup\$ Thank you for the answer! It looks great but for another case - not mine as ADuM5240 will take up to 100 mA. Which is FAR from I could afford (I'd say 3 orders of ten). \$\endgroup\$ Apr 21, 2016 at 15:45
  • 1
    \$\begingroup\$ Indeed, that's a lot. Note that there seems to be other versions that are more efficient: ADuM6210 et al. The DC-DC part uses ~7mA typ. at no load, and the digital isolation 1.1+2.7mA (for 5V->5V). Still way above your requirement, it seems. \$\endgroup\$
    – dim
    Apr 21, 2016 at 18:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.