tDCS Schematic

Thread Starter


Joined Aug 27, 2012
Hello out there! I'm a new member and I am looking for a easy to follow schematic on how to build a tDCS. I looked over diagrams I've found but they are incomplete and/or withholding key information. Does anyone have reliable information they wish to share? thx


Joined Sep 9, 2010
Do you mean this?

I've built one that works nicely, but I've been reluctant to post it to the "Completed Projects" forum because I'm not sure how the moderators feel about such things.

For the benefit of the mods:
It's a circuit to apply a µA level of DC that reverses polarity at about 0.5Hz. This is driven through the ear lobes to provide a profound mood alteration. A 9V battery is more than enough power to do the job.

The treatment is FDA approved and there are commercial, patented devices available on the market. I can supply more information as needed.


Joined Dec 26, 2009
I want one.....
Actually I want a lot of them...
And can I get a direct wire electrode kit? Is it an option.???
Mind control.... Man I can only think of ever so many diabolical uses...
Deffinatly want a case or two?
How much did you say they cost?
Phoenix, AZ


Joined Sep 9, 2010
The parts are cheap, but the "cost" could be high! :p

But seriously, these things bring a lot of relief to certain people. I prefer gin.


Joined Aug 13, 2012
Hi wayneh,

I am VERY interested in complete tDCs Schematics. Can you email me with them or could we work something out. I really have been doing a lot of research on this and cannot get my mind off making one. That said, I want to make this correctly. I really hope you can help me out. Thank you.




Joined Sep 9, 2010
I'm going to post the circuit I built. If the mods decide to remove it, I won't feel bad. Maybe I'm overly cautious anyway. I don't want to encourage anyone to apply magnetic or electrical currents to their body.

This may be slightly different than tDCS; it's actually referred to as MENS (Microcurrent Electrical Stimulation) or CES (cranial electrotherapy stimulation), or MET (microcurrent electrical therapy). Acronyms seem to be important in this industry.

Anyway, I based my project off this device, described in patents US2010145410, US2010047834A1 and others. I also added TENS capability since most of the circuitry was already in place to easily add this capability.

My circuit uses a dual timer 556. The first timer feeds an RC tank to deliver a slow changing voltage onto the control pin of the second timer, which operates primarily at 0.5Hz but is adjustable up to about 120Hz. The changing frequency feels somewhat random and prevents a "burn in" in the patient. It's more effective than a steady beat, but probably this is a small refinement.

The timer output is fed to a 4017 counter and the "0" and "2" outputs are compared. This provides a way to alternately remove and reverse the direction of current flow. An LED on "3" is flashed at the completion of each cycle.

Output from the counter controls the state of an LM358 dual op-amp controlled-current circuit. Intensity (less than 1mA) is adjusted using a dual gang pot, since there are essentially two channels, plus and minus.

The TENS output is developed by scaling up the voltage of the pulses with a transformer. A switched pot turns off the TENS output, which is delivered from a different output jack. I wanted to make it a bit difficult to accidentally apply TENS to your earlobes. :eek:



Last edited:


Joined Oct 2, 2009
Firstly, do you need working brains for tDCS to work, or will it work on dead brains?
If so, that would be very useful for AAC and most of my students.:p

I can feel my legs twitching already.


Joined Sep 9, 2010
This is an intentional hijack of an old thread. The TS never returned, and this was basically my thread way back then, so today I'm claiming it as my own.

The circuit in #7 and the build in #9 worked and made my friend very happy. A more recent build without the TENS portion worked great for an ex-Marine friend with mild PTSD. But I never had the luxury to verify that the waveform was really doing what I designed it to do. Before I make more of these, I'd like to be sure it's "robust".

Since then I've learned to use LTspice, and I've made a simulation of my circuit. This simulation uncovered some oddities with my op-amp H-bridge. (The timing waveform is fine.) The purpose of using the op-amp instead of the direct output of the 4017 counter, is to allow for a constant-current output of the timing waveform.

The output waveform has a few problems:
1) The "zero" segments are not at zero. I believe this is due to the offset properties of the op-amp and I can make this problem nearly disappear using a different op-amp, a TLV272 instead of LM358.
2) The waveform falls apart when the load R9 is increased to, say, 2kΩ versus the 540Ω shown in the model I've uploaded. The real load is unknown but is more likely in the higher range. Could even be 10K or more.

I've been trying lots of combinations of op-amps and resistors to solve these issues and have become quite confused and frustrated. So I'm here to ask if there is a better way to accomplish a controlled, constant-current output of 0-600µA into a load of 1k-50K that follows the timing waveform. (I understand that it won't be possible to maintain full current into a high impedance with just a 9V source. That's OK as long as I can cap the current at a fixed level.)

Here is the op-amp part of the circuit. The signal coming in from above is the waveform put out by the 4017, see #7.
Screen Shot 2016-08-09 at 11.55.49 AM.png

Here is the output using the LM358 as shown. Note the shoulders at ±250µA that should be at zero.
Screen Shot 2016-08-09 at 11.52.56 AM.png
Last edited:


Joined Sep 9, 2010
WTF!? Now I cannot make the model above misbehave when the load impedance (R9) is increased. It seems to be fine up to at least 10K.

Something about preparing this model file for upload seems to have fixed it. :confused: Definitely confused.

Screen Shot 2016-08-09 at 12.19.34 PM.png


Joined Sep 17, 2013
I agree the shoulder problem is down to opamp offset, and perhaps the ouput not pulling right down to the ground rail? Try adding pull-up resistors of ~100k from the non-inverting inputs to the 9V rail.