I’ve used the AT42QT1010 capacitive touch sensors in another project and they work pretty well, but I’m attempting to use them with a small sheet of ITO and the sensor reads a HIGH signal even when I’m not touching it (I can tell because the LED illuminates constantly even when nobody is touching it). I connected a small wire to the pad input, attached a gator clip to the wire and ITO, and it reads a consistent HIGH input with no touching.
It almost seems that the input is floating. I’ve tried tying various points of the sensor (VDD, OUT) to ground with a resistor (tried a 10k and a 1k resistor, had no effect) and I also wanted to try adding resistance between the electrode and pad input connector, but when I connect a wire from the pad input to an empty breadboard channel, it also stays HIGH with no touching.
What am I missing here? Must any object connected to the pad input be heavily insulated somehow? If I must alter the capacitance, how do I go about doing that?
Another thing I should mention is that I desoldered the bridge for fast mode and changed it to slow mode but that had no effect on this outcome.
While we have not tested these breakouts with ITO sheets/glass, it theoretically should work. We have only tested the external electrode with things like copper tape or, in one project, with a large metal bar for pull-ups. Just to confirm the electrode pad is working, have you tried using a different material for your electrode on this specific capacitive touch breakout? Also, from what I found in a brief search on ITO sheets, ITO sheets have the conductive coating on just one side. How are you connecting to the ITO Sheet and have you measured with a multimeter to verify you’re connecting to the coated side of the sheet?
I used a plastic-coated gator clip connected to a 9" length of stripped copper wire which is soldered into the pad input on the sensor. The gator clip is then clipped to the ITO and yes I did distinguish the conductive side, although the gator clip is touching both sides so I didn’t think it would matter.
I wonder if I instead need to use copper tape and a connector that only touches the one conductive side instead of a gator clip that touches both sides? I didn’t think it would matter because I thought the touch aspect would be what set it off and not the instance of various materials.
I have a static mat on my bench which might be part of the problem, but capacitance is hard to measure so I can’t be certain what’s really happening here.
I should add that when it’s just the length of stripped copper wire with no gator clip or anything else attached, the sensor works as it should. As soon as I connect a gator clip, the signal goes high (sensor LED illuminates) whether I’m touching it or not.
These sensors do a self calibration when turned on. I think what’s happening in your case is they self calibrate with no wire attached and then when you attach the wire, the sensor then sees a change in capacitance and thinks someone is touching it.
Try attaching the wire with the power off, then apply power. Then touch the wire and see if you get a reading then. Try to avoid the wire being close to anything conductive or close to a grounded object as those can trick the sensor and make the self calibration fail.
I’m still curious about adding capacitance, or is it not really necessary? In the data sheet for the sensor, it mentions changing the value of Cs but it’s unclear how that is done.
The way the board is setup, you shouldn’t need to worry about changing anything.
If you were changing the capacitance on the board, you’d need to calculate the new value needed and then desolder that cap off the board and solder a new one on.
TS-Chris:
The way the board is setup, you shouldn’t need to worry about changing anything.
If you were changing the capacitance on the board, you’d need to calculate the new value needed and then desolder that cap off the board and solder a new one on.
I’m having more issues with the sensors in that they seem very unstable. When I first power them on, they work well only responding to touch, and then after a short time, a few minutes or so, the sensors start triggering on their own with no touch.
What might be causing this? If they are self-calibrating, and they work well on power-up only to drift into unstable and unpredictable territory, how might I correct this? Could it possibly be my anti-static mat offsetting the calibration?
In the datasheet, it goes into depth about increasing/decreasing sensitivity without explaining how and that seems to be what is needed here.
The only thing I can think of to try would be to eliminate the ITO sheet you’re using and see if the pad on the sensor itself will work. If you need a sheet, you might try a different material.
Reviving this thread… Had some luck with using the Serial protocol and libraries for Arduino. I was able to get it all working but only with 1 touch sensor and my string of LEDs. As soon as I hook up more than 1 touch sensor, it all goes haywire with sensors triggering randomly and unpredictably and I’m not finding many answers on the datasheet or hookup guide.
I’m not sure how to move forward from here. I’ve definitely burned out some touch sensors and hoping to not burn any more. I may be failing to understand some key concepts relating to capacitance and how the HIGH/LOW signals on the OUT pin are sent.
I am having the same issues with phantom triggers. I’ve tried a couple different boards, tried changing one to low power mode, but they still keep triggering the output with nothing anywhere near them.
The LED is dimly lit when nothing is touching the sensor, and lights up brightly when touched. When it triggers on its own, the LED does not increase in brightness as it does when actually touched.
I don’t know anything about the device you have this connected too, but make sure the power supply for the touch sensor is a steady voltage between 1.8 and 5 volts. If the voltage changes or is interrupted, the sensor can go into a re-calibration and that could cause phantom triggers.
You might also try testing with only a regulated 5 volt power supply connected to see if you get phantom triggers there. If you don’t, the device the board is connected too might be doing something odd that causes issues.
