Capacitive touch doesn't work on my board
I shouldn't say it doesn't work but the touch pads give erratic results using the 8-1_Touch_Sense.pbp application.
This problem happened before with a previous board. I am beginning to wonder if my fingers have no conductivity whatever.
My troubles all started the other day I began by wanting to revisit the 10-1_User_Interface.pbp application and discovered that the pads no longer worked.
I reloaded the factory test and ran it and can't get past the Cap touch test which is unfortunately the second test in the program so it hangs there.
All the touch applications ran reasonably well when the board was new but now there is almost no differential (differential of 15 on the one pad that seemed to register a touch) when tested in the 8-1_Touch_Sense.pbp application.
In the 8-1_Touch_Sense.pbp application when I touch a single pad several number values jitter but one seems to change more than the others. Is this normal? I have forgotten which column and row numbers correspond to the 6 pads and that makes things difficult to debug. I would request a refresher on that. I think the far right one is Row 1 Col 1.
Since this problem has occurred some time from the date I received the (replacement) board and the new board worked fine I am trying to determine the cause.
Are the pads susceptible to finger oil buildup? I have cleaned them before with isopropyl alcohol to no avail. Should I moisten my finger tips to make them more conductive? I have been returning the board to the bubble wrap envelope for the last few months for storage. Is there some detrimental effect in doing that? Is there a work around like grounding a selected one of several vias that are labeled on the top right side of the board?
In the 8-1_Touch_Sense.pbp I experimented with the threshold value and when I jumped it up to 200 all the LEDs lit. When I cranked it up to 165 only the A.0 led lit and when I touched the pad it went off. At 150 I could get several leds to lite intermittently by touching a pad but not repeatedly.
To be honest, I am at my wit's end on this puzzle. I have run all the other applications which did not require Cap sense including the exercises and they have all worked just fine. As I said the factory test worked for a while.
It would be handy to have a work around and to know which pads correspond to which Port Pins.
Just an update.
I just discovered that if I moisten my finger and bridged three pads by placing my finger in the adjacent corners of them, I get three leds illuminating bright and steady. I conclude that these are the corresponding leds to the pads. This test was made with the threshold at 150 not 100.
It was more difficult to get the leds to light one by one though. It makes me think that the combined discharge of the three pads contributed to the clarity of the signal picked up by the PIC.
This might be of some use to others.
I kept on playing with the 8-1_Touch_Sense.pbp application and pushed the threshold up to 168 which was a bit higher than I used before but a little lower than all the 6 displayed values on the lcd. I got very distinct responses from the leds this time. Apparently over time the physical sensitivity of the touch pads had changed.
I used the new higher threshold value in the other capacitive touch programs including the Factory Test and the User Interface programs and they worked as designed. In this way I believe I used the 8-1_Touch_Sense.pbp code to "re-calibrate" my trainer board so to speak.
Perhaps my blundering onto this solution will be of help if anyone else has similar issues. I also took a cloth and cleaned the pads again with Isopropyl alcohol which I assume didn't hurt.
If you are intending to engineer a product (or hobby type gizmo) using the Cap Sense, consider using stainless steel touch pad material. It won't corrode. Also, I found that pressing harder on the pads gives different results than when just touching them. You may have to just get consistent with your touch effort. Other than that, you just experienced what I have several times; you got it figured out before others even see the post. Good work on keeping at it.