5

I'm trying to replicate part of this experiment. Specifically, I want to touch the screen with a cable and some aluminum foil whenever I send a signal from an Arduino pin. From the blog post linked above:

For communication from the Arduino to the Android tablet I'm using a relay to vary the capacitance of a aluminum foil taped to the screen. [...] A relay allows me to float the foil or pull it to ground, effectively changing the capacitance. This is much more reliable, but requires a bulky relay. I think with some more experimentation I'll be able to find a solid sate solution.

This is how my circuit looks like right now:

Basically I have a relay switch switching from ground to float every second.

This works intermittently. Sometimes the screen stops accepting real touch from fingers until the cable is disconnected from the screen.

How can I make this circuit more reliable? How can I make this cable 'touch' the screen whenever I send a signal from the Arduino board?

EDIT 1: I was able to isolate the problem a little bit. If I put my hand behind the phone, even a few milimiters away from it, the trigger works. Otherwise, it won't. See: https://vine.co/v/bIeWP0pdvuP.

EDIT 2: If I connect the charing cable to the iPhone and touch the other end (the USB shield that gets connected to the computer) it works. I tried to connect that end to ground and works intermittently. It looks like a ground problem. The arduino is connected via USB to my laptop so I guess that's where my ground comes from.

miguelrios
  • 51
  • 1
  • 4
  • Some related questions: http://electronics.stackexchange.com/questions/40156/mimic-human-capacitance http://electronics.stackexchange.com/questions/60070/how-do-i-make-a-micro-controller-act-as-a-finger-on-a-touch-screen http://electronics.stackexchange.com/questions/23036/how-can-a-capacitive-touch-screen-be-triggered-without-human-contact http://electronics.stackexchange.com/questions/21119/conducting-plastic-for-capacitive-touchscreens – miguelrios Mar 31 '13 at 21:00
  • Have you tried adding some capacitance and maybe a resistor parallel to the `com`->`no` pins? – drzymala Mar 31 '13 at 21:06
  • The experiment you linked to was for an android phone to be used as a flashlight - I don't see the connection with your question. – Andy aka Mar 31 '13 at 21:16
  • Please check your relay connections - there is an anomaly in your circuit - the contact symbol you have used doesn't match your NC label. I'm not saying you've done a mistake in the real circuit but who can tell? I think it might be OK actually. – Andy aka Mar 31 '13 at 21:56
  • @Andyaka Part of that experiment is communication between the tablet and the arduino board by using 'touch'. That's exactly what I need, to 'touch' the screen programmatically. Here's an earlier version of the experiment: http://www.youtube.com/watch?v=JDgDMBquBw0 and another similar one that uses an iPhone: http://www.youtube.com/watch?v=z4eTjH4zjwE – miguelrios Mar 31 '13 at 22:51
  • @martini I'll try that and report the results. Nevertheless I'd like to take a more empirical approach and understand why it's not working right now and what would it take to make it work reliably. Experimentation is part of that, but I'd like to avoid guessing as much as I can. Thanks! – miguelrios Mar 31 '13 at 22:53
  • If touch sensing works by extracting a small current (via capacitance) to ground then the relay needs to have low contact-capacitance when it is open circuit. This may not be true of the relay you have used. On the other hand, if the extracted current is meant to "return" to the handset via the palm of the hand then maybe you need to rethink how this should be done? – Andy aka Mar 31 '13 at 23:02
  • In other words you're trying to simulate a human tap? – Kevin Mark Apr 07 '13 at 22:02
  • @KevinMark yes, with no human interaction and ideally not moving parts. – miguelrios Apr 08 '13 at 01:12

1 Answers1

1

The critical point here is to control the ground of the phone.

When you manipulate your phone for its normal usage, one hand is touching it and sets the ground. Trying to do the same manipulation with a phone sitting on a table will be less reliable.

To have the same effect in the lab, you should have your phone laying on a conductive layer that is connected to the ground (or any other fixed potential) of your circuit. A simple way to do so is to have your phone laying on a copper foil that is used as the common ground of your circuit.

Your phone does not need to have direct ohmic link to ground, a good capacitive coupling is ok too.

MAC
  • 361
  • 1
  • 5