I'm basically trying to create a capacitive touch slider, and have already created an algorithm which reads in the sensor data and translates it into an output depending on which direction I'm moving my finger
The only issue is the design of the slider itself. My current design is rather basic, using 20mm wide copper strips spaced 3mm apart with usb power used as the power source which feeds into several MPR121 controller to interpret the information. I've sandwiched the copper tape between two pieces of 2mm thick acrylic (however due to the solder there's probably a bit of a gap between the acrylic sheets)
This works fine on direct contact, however once I put the 2mm acrylic on top the sensitivity drops massively as you would expect, and I'm forced to use the highest sensitivity setting for the MPR121 chips which results in a less-than-desirable performance.
I'm therefore wondering how I could improve the sensitivity? The ways I had found on the Internet which would improve the performance were:
Wider electrodes
Smaller gaps between the electrodes
Hence the current design, however there's an upper limit on how wide I can made the electrodes since wider electrodes require my finger to travel further
The only improvement I could think of at this point was to reduce the acrylic to 1mm thick, but I was wondering whether using thick copper plates instead of this thin copper tape might improve the sensitivity, or are there any other things I could do?
Capacitance is sensed between sensing plate and ground.
But you have no ground near, just other sensing fields.
You could try ground strips between the sensing strips.
Glass should have a higher relative permittivity than perspex.
Made cap sensors that worked fine through 10mm glass.
Leo..
I did happen to read a document today which described surrounding the electrode with a ground actually, so I might make a prototype to see its effect. Do you think there's some sort of minimum width the ground strip should be between the electrodes, or could I get away with making it probably <5mm wide
Using glass is an interesting option, my only concern would be that given my next iteration of the slider would be roughly 60cm x 6cm, acrylic would be significantly easier to work with than glass (and less likely to break in the post)
Seems logical to make three strips and two gaps the size of a fingerprint (~15mm).
Maybe 4mm wide strips and 1.5mm gaps.
Then the finger is always on a sensor and two grounds.
Make the strips short-ish, because long strips will increase capacitance and lower sensitivity.
Please post back with the results.
Leo..
Wawa:
Seems logical to make three strips and two gaps the size of a fingerprint (~15mm).
Maybe 4mm wide strips and 1.5mm gaps.
Then the finger is always on a sensor and two grounds.
Make the strips short-ish, because long strips will increase capacitance and lower sensitivity.
Please post back with the results.
Leo..
I tried putting together a quick prototype to test this new approach
Copper strips are 5mm wide (thinnest ones I could find without importing from China or cutting myself), the gaps as 2-3mm for this prototype to achieve the final 600mm wide slider width. Height of the board is 60mm so the live electrodes are around 50-55mm long
I thought I could cheat here by connecting all grounds together via the long strip at the bottom and then only soldering a single ground wire to it, otherwise I'd have to do a lot of work splicing ground cables together.
The final result does not perform as well as my first design however. I'm not sure if by cheating with the ground I've adversely affect the overall performance, whether the gaps really are too wide or if there's something else
