Need help with my tens project, can't amplify the signal

Thread Starter


Joined Nov 6, 2017
I'm making a TENS unit for a college project.
I want it to generate pulses between 140 and -140V at its maximum value. Before someone says this is dangerous, let me say it is not. The pulses generated are very far from each other as the pulse width is 200us and the distance between pulses is 10ms at its minimum value. The human skin has high impedance(KOhms or MOhms depending of multiple factors) and acts as an RC circuit.
Im using a STM32F0 microprocessor which gives me a signal between 0 and 3.3V, then I amplify it getting a signal between -14 and 14V.
This works perfectly, but the problem occurs when I connect this signal to a 1:10 pulse transformer (DA2032-AL) because i want a signal between +-140V at max.
I have tried everything possible but only i can get the original signal or 0V.
I dont know if this transformer isn't the right one or if im doing something wrong, if someone could give me some help i'd appreciate it.

Sorry if i made some linguistical mistakes, english is not my main language :D
I attach the main circuit down here v



Joined Sep 17, 2013
The poor little opamp will struggle to drive any transformer, since its output current is internally limited to a few tens of mA :(


Joined Sep 9, 2010
@Alec_t is right. You won't get more than a tiny current at the higher voltage. See the arrangement in the lower right of the picture attached to this post.

I think there could be a problem with your transformer as well. I used an old wall wart transformer from a 12V adapter, in reverse, so that it was roughly 1:10.


Joined Sep 9, 2010
Would you be so kind as to post a link to a reference that supports the safety of such pulses?
TENS devices are widely available commercially and their application is well known and documented. The voltages cited by the TS are typical for these devices.


Joined Sep 9, 2010
How about control of the current output?
My understanding is that the only control of intensity is by the voltage amplitude and length of the pulse. There is not specific control of the current. Clinical TENS devices might be fancy enough to use current control though.

The primary barrier to current is the impedance of the human skin, which is at least about 1kΩ and can be a lot more than that for electrodes some distance apart and less-than-perfectly applied with conductive gel. Delivering 50mA into 1kΩ requires 50V, and I believe 50mA is on the high end of what's expected. It might take well over 100V to achieve that level under normal conditions and these devices can deliver that.

I think the majority of the commercial devices are small and battery powered. In an arrangement like the one I used, the 9V battery cannot deliver much current to the primary of the transformer and that also limits the output.

This all said, I won't deny that there is the potential for a DIY device to achieve dangerous outputs. The circuit I used, if it was supplied with a car battery and modified a little to boost the pulse, would essentially be an inverter. Not something I'd want hooked to me with conductive electrodes.

A pretty good overview here:
or here:
Last edited: