I need to develop a small autonomous device that could stimulate a touch screen. For the record I need such a device to be able to run some automatic tests on some specific smartphone apps, where touch related functionalities cannot be tested by injecting touch events from USB.

Here is my first design idea: a first prototype would consist of a grid of copper pad on a PCB, like on the following picture.

enter image description here

Each copper pad could be connected with a transistor to a metallic mass that would simulate the fingertip capacitance. Obviously each transistor would be commanded by the MCU comprised in the device.


simulate this circuit – Schematic created using CircuitLab

However I don't have any idea if this design would work. What I can't figure out is how I should connect my battery to the metal part and/or the copper pad, and how big should my metal body be to yield enough capacitance to the device to be detected by the touch screen.

Thank you for any insight you stackexhangers could provide!

  • \$\begingroup\$ A slightly modified consumer CNC machine or 3D printer might be a reliable and flexible solution. That would give you a high resolution, and some actual touching going on. \$\endgroup\$
    – Dampmaskin
    Sep 7, 2017 at 14:48
  • \$\begingroup\$ Yeah that would be the fallback solution, but I'd prefer much more to have a portable device, that I could manually position on the touch screen. \$\endgroup\$
    – Vincz777
    Sep 7, 2017 at 16:05
  • \$\begingroup\$ Does this answer your question? How to use a capacitive touch screen without a human hand \$\endgroup\$
    – flaviut
    Mar 5, 2022 at 21:58

1 Answer 1


I'd suggest that you will have too much capacitance in the FET to give reliable results.

When the FET (and I'd use P-Chan in your circuit) is off there is typically hundreds of pF of output capacitance.

I'd use a Reed Relay where the open circuit capacitance is extremely low, just a couple of a pF maximum. This gives you the maximum possible change in capacitance to be recognized by your touch matrix. I'd also use round capacitor targets rather than square.


simulate this circuit – Schematic created using CircuitLab

  • \$\begingroup\$ Thanks for your answer! I didn't know the Reed relays. I guess I would need one by capacitor target, right? Do you have any idea if I can avoid the GND plane below the DUT, and just use a big enough piece of metal as alternative ground? \$\endgroup\$
    – Vincz777
    Sep 7, 2017 at 19:28
  • \$\begingroup\$ A metal backing would suffice for the backplane, and yes you would need one reed relay per capacitive target. \$\endgroup\$ Sep 7, 2017 at 21:41

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.