min currrent to make pin go high???

What is the minimum current required to make an analog (or digital for that matter) pin go high? I’m trying to come up with a design that conserves as much battery life as I can. I’ll be sampling the center connection of a potentiometer and would like to put another resistor between the 5v source and the left connection of the pot. I know I won’t be getting the full range of 5V from my pot this way, but I think I can live with that? I will probably be constrained by the size of pot I can find, as well, since I’ll be using a stacked dual concentric pot for an electric guitar. Any input or clever snide remarks are most welcome!


You need to be careful with the large impedances required to minimize the currents.
You may end up with an erratic system that is sensitive to noise.
For a digital input I would not use a resistor larger than 100K.

Using a larger resistor on the ADC will affect your measurement time. The low current
reduces the charge rate of the input capacitance. Check the Atmega datasheet.
For the ADC I would use a low current op-amp buffer.

(* jcl *)

Thanks for the reply... good point about the noise. I'm only using the pot to pass a value to my circuit that tells the LEDs how bright to glow. I expect it would read a value every time I go through my loop to update the LED states. I tried it this scenario on a breadboard with my 500k pot and the values jump around 2-3 points when i'm not touching it... not sure if this is useful info or not. I've found some dual concentric 100K pots (I need duals) that I guess I could use.... and I'll have to live with the constant current pull of .05 mA... or I could try to half that amount by throwing a 100k resistor in front of the pot since a 512 step resolution is plenty (vs. the 1024).

Anybody know of a small hi-capacity battery I could use to power the Arduino Nano?

Thanks again...

You don't need to apply power to the pot all the time. Only power the pot when
are going to read it. Ground one side of the pot. Tie the other side to a digital output.
When you want to read the value set the digital output to HI otherwise it is LO (and drawing no current). The pot only draws current a small fraction of the time.

For batteries I would checkout www.all-battery.com. For battery capacities I would
checkout the Energizer site.

(* jcl *)

So, if I wanted this pot to be available to a (seemingly) all of the time to read an input to control my "fader" circuit, I can just set the digital pin going to the + side of the pot to HIGH right before calling analogRead() and then back to LOW right after my analogRead() call??? Ingenious... simple... yet ingenious!!! Thanks for the idea. I'll try it out.

I'm planning on controlling 16 RGB LEDs individually with (hopefully) only 14 pins to the array (using TLC5940 LED drivers). I've never attempted to write any code for a multiplexing setup using persistence of vision, so it ought to be a nice little challenge.... I'm also trying to keep the code less bloated as possible... also a chore since I'm no pro, here.

Thanks again... I'll try this out real quick....

Worked like a charm… I did notice (using the millis() function that the analogRead() function added ~50ms to the run time of my program (once through my loop() function). I don’t have a 100k pot on hand right now, just the 500k… and I don’t have enough large resistors to play around with this, but, i was wondering if there is a way to calculate how long the analogRead() function takes depending on the current applied.

Haven't looked at the Arduino analogRead function.

In the ADC section of the Atmel datasheet there is a discussion about the
input resistance and capacitance of the ADC channels. You may be able to get
an idea of measurement time/accuracy vs impedance from that.

(* jcl *)

If your going to be powering the pot with a digital high output pin for only as long as it takes to capture a reading and therefore there won't be a continuous current flow in the pot, can't you then use a more reasonable size pot say 10K ?


I suppose I can, if I can find a dual-concentric pot that is the right physical size to mount on an electric guitar... I was getting pretty good results (from what I could tell) using the 500K pot I have just testing out the "technique" you guys mentioned. The other pot on the dual-pot will be used in the actual guitar signal chain, so that is a consideration, as well.