Hey guys
I’m imagining a nontransparent touch-surface that works in 2D and is able to detect whether there is a finger on the pad and where that (one) finger is. No multitouch.
We could use it to send one CC for x-axis and one CC for y-axis similar to what people know from ableton vst host.
This could look like what this guy is talking about at 1:15 to 1:30
Who would be interested in getting his hands on such a module?
I don’t seem to be allowed to start any sort of poll.
Yesterday I’ve done some research on parts and technology. I think I’m gonna design a capacitive touch sensor on a 200mm x 160mm PCB consisting of a touch area of 120mm x 120mm plus some parts.
The main idea of capTouch is that I need a µAmp current source and an AD-converter. I’ve made some good experiences using Microchips CTMU for both.
However, the touch surface is made up of several columns and rows, each of which being 5mm-8mm wide. So we’re talking about 24+24=48 analog channels.
Analog multiplexing seems to be incredibly expensive (8-15 EUR per 16 channels) so I’m thinking of using PIC24FJ128 to perform all the sensor interpretation as it comes with 50 analog input pins for 9-10 EUR.
Unfortunately that thing would be a really small SMD device (TQFP100). THT is not an option anyway since I want the surface to be as close to the front panel as possible.
Analog multiplexing would, however, maybe allow for parts that one can actually solder :shifty:
communication could be done in any way imaginable: the standard approach would be to use SPI or parallel data output for DIN connection. But maybe the PIC is strong enough to adapt parts of TKs code in order to have it be a USB MIDI device of its own (it does support usb). We’ll see about that…
cheers
rava