I m trying to control 3V, 40mA laser brightess digitaly but not with PWM
When I use analog potentiometer I get the whole rangle with 0-40 ohm resistance
Is there a digital pot that fits requirements or something simple I could make. 16 steps is enough

 

Hello,

You could use some fets that switch some resistors.
The resistor values could be 40, 20, 10 and 5 Ohms.
When all resistors are parallel the value will be 2.66 Ohms.
When only the fet of the 40 Ohms is switched on the resistance will be about 40 Ohms.

Greetings,
Bertus

 

I doubt that you will be able to find a digital pot capable of that kind of current. and low resistance digital pots are quite rare too.

 

Here is an example of a high-current "digital" pot:

The original is from the National datasheet for the LM317. I am suggesting this as a concrete example of the approach suggested above by bertus, which could be adapted to mosfets.

John

 

How about a programmable current source?

 

Here is an example of a high-current "digital" pot:

The original is from the National datasheet for the LM317. I am suggesting this as a concrete example of the approach suggested above by bertus, which could be adapted to mosfets.

John

In your schematic example, do you simply connect the LED's to the different transistor outputs? The resistor values connected to each collector will vary the current to each output, correct? So I could have a rotary switch connected to an LED to vary the current. This doesn't seem digital at all...

 

Check this thead out.

http://forum.allaboutcircuits.com/showthread.php?t=11792

 

You can make use of PWM to vary the intensity continuously. Apply PWM through a suitable RC to gate of a FET. That way you can control the ID and hence the intensity of LED.

 

You can make use of PWM to vary the intensity continuously. Apply PWM through a suitable RC to gate of a FET. That way you can control the ID and hence the intensity of LED.

no reason to convert the pwm back to an analog voltage through a rc network. just apply it direct to the gate so the fet stays cool.

 

no reason to convert the pwm back to an analog voltage through a rc network. just apply it direct to the gate so the fet stays cool.

Well, it depends on application. In photometric application it would not matter since persistance of visions doesn't perceive the ON-OFF operation of LED but the average current through it.

While in radiometric application where the constant intensity matters for few tens of microseconds, you need a very high frequency PWM. Instead, RC does this job of averaging.

 

In your schematic example, do you simply connect the LED's to the different transistor outputs? The resistor values connected to each collector will vary the current to each output, correct? So I could have a rotary switch connected to an LED to vary the current. This doesn't seem digital at all...

While trying to refine the term, "digital," I stumbled on this ap. note from Analog Devices:

It seems to be right on point for the subject of this thread.

John

 

There we go, that looks digital to me!