1
\$\begingroup\$

I’m trying to create a circuit for iontophoresis that I can control the current of with a raspberry pi or eventually a microcontroller. What would be the best way to control the current?

Iontophoresis is a treatment for sweaty hands or feet where the hands or feet are placed on two separate electrodes and a current of between 4-20mA is driven through the body. I would like to be able to take an input voltage of 15-30 V and produce a 4-20mA current depending on what value I specified ahead of time. I also want to increase the current over a few seconds rather than go straight from 0 to max.

I did a session yesterday and found that the current from 2 9V batteries wired in series started at about 18 mA and then slowly dropped down to 8mA or so. I was using tap water with salt and baking soda in it with aluminum foil sheets as electrodes and paper towels to separate my hands from the electrodes.

For the polarity switching I plan to use an H-bridge circuit. And I have an INA219 sensor to measure current and voltage with, but I’m not sure what exactly I should use to control the current. Digital potentiometers seem to often be only able to handle voltages of around 5V. I found this one that can handle up to 30V but it also claims that max current is only 5mA at 30V and Rwb < 6k ohms and A open. Which I’m not sure if that’s a problem or not?

Another option seems to be variable voltage regulators. I found the LM317-N and a video talking about how to set it up as an adjustable constant current source. My only concern with that is the minimum load current is listed as TYP: 3.5mA and MAX: 10mA. Does that mean I’d have issues running it at less than 3.5mA through the hands during the startup and shutdown? Constant Current Circuit Design

Any advice on the best course of action here? Is one of those a good solution or is there a better way of doing this?

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Pretty much any common op-amp like an LM324 can source or sink 20mA, however I'm not sure what you are doing is safe. \$\endgroup\$
    – Spehro Pefhany
    May 14, 2020 at 1:01
  • \$\begingroup\$ Long-term exposure to DC current definitely is not safe: DC current causes chemical changes in the tissues that it flows through, and any exposure to current (DC or AC) that passes from one limb to another carries a risk of messing up your heart rhythm. \$\endgroup\$
    – Solomon Slow
    May 14, 2020 at 1:31
  • 1
    \$\begingroup\$ P.S., "4-20mA?" For real? Now, where have I heard those numbers before?... Sounds like maybe whoever invented this bogus snake-oil idea maybe found some components in the Grainger catalog that they thought would make it sound and look "scientific." \$\endgroup\$
    – Solomon Slow
    May 14, 2020 at 1:33
  • \$\begingroup\$ @SolomonSlow Not saying there aren't potential issues but it's actually a very widely accepted treatment for hyperhidrosis with multiple devices approved by the FDA for use. Some more info on it: ncbi.nlm.nih.gov/pmc/articles/PMC5705354 sweathelp.org/hyperhidrosis-treatments/iontophoresis.html \$\endgroup\$
    – Elderberry
    May 14, 2020 at 23:21
  • \$\begingroup\$ Found the web site of the company that makes the machine described in that paper. I can't read Korean, so I don't know what it claims, but just based on the pictures, I'd say it's pretty sketchy. \$\endgroup\$
    – Solomon Slow
    May 15, 2020 at 1:04

1 Answer 1

Reset to default
0
\$\begingroup\$

Google for "Howland current pump": https://www.allaboutcircuits.com/technical-articles/the-howland-current-pump/

It's a lovely op-amp circuit that turns a voltage (from your MCU DAC) into a single-ended current. You can use a "power op-amp" to implement it: https://www.st.com/en/amplifiers-and-comparators/power-operational-amplifiers.html

\$\endgroup\$
5
  • \$\begingroup\$ Doesn't that still run into the issue of having a higher voltage across the resistors than most digital potentiometers seem to allow? And that you still need regulate the input voltage? Or would you adjust the current by adjusting the input voltage? And in that case how would you suggest doing that? \$\endgroup\$
    – Elderberry
    May 17, 2020 at 1:12
  • \$\begingroup\$ "Doesn't that still run into the issue of having a higher voltage across the resistors than most digital potentiometers seem to allow?" --> My suggestion has nothing to do with digital potentiometers. " And that you still need regulate the input voltage? Or would you adjust the current by adjusting the input voltage?" --> The latter of the two assumptions is correct. \$\endgroup\$
    – Alex Lopez
    May 18, 2020 at 11:04
  • \$\begingroup\$ "And in that case how would you suggest doing that?" --> As described on my response, you would feed the (analogue) voltage generated by your MCU DAC into the Howland circuit, which will output a current that's directly proportional to said voltage. You would choose the resistor values to fit the application (supply voltage, DAC and output current ranges). The circuit is inherently current limited, which you would want given its intended use. \$\endgroup\$
    – Alex Lopez
    May 18, 2020 at 11:04
  • \$\begingroup\$ Ah I saw the op amp portion of it and got caught up in that and missed the DAC part. This does seem like it would work well for what I want. I assume the benefit of the DAC over an adjustable voltage regulator like the LM317-N is that the voltage regulator would need to dissipate quite a bit of power at the lower voltages? \$\endgroup\$
    – Elderberry
    May 20, 2020 at 23:59
  • \$\begingroup\$ One thing I was wondering is how much headroom you'd want on the power ratings of the resistors. I have a sense of what the max resistance of the hands tends to be but it does vary some. And I wouldn't want it to increase a bit and all of a sudden have the power of the resistor exceed its rating. I guess what I'm really asking is is there some rule of thumb on how much I should add to my best guess on the power rating it'll need when selecting a resistor? \$\endgroup\$
    – Elderberry
    May 21, 2020 at 0:07

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.