12:24 <lekernel> larsc, does your company sell a chip that contains both a capacitive sensor controller (CapTouch™) and a slow ADC for digitizing pots?

12:31 <larsc> lekernel: not that I'm aware off

12:36 <lekernel> I'm thinking of using one of those capacitive sensor controllers with conductive paint on glass with a pogo pin to connect to the PCB ...

12:36 <lekernel> this way I can put a LED underneath

12:39 <lekernel> or metal glued to the glass

12:50 <larsc> I never really worked with any of the captouch devices

17:21 <wpwrak> isn't the "quick and dirty" ADC good enough for the pots ?

17:22 <lekernel> it'd need some tweaking, values aren't very stable

17:22 <lekernel> I suspect the tolerance of the threshold of the schmitt trigger, and of the capacitor

17:23 <lekernel> it's quite a problem when you want to reach a clear 0% or 100% close enough to the limits of the pot

17:26 <lekernel> the schmitt trigger threshold can actually vary a lot... I should have checked that before

17:28 <davidc___> For captouch, you can do that with a digital IO or two and a resistor

17:28 <davidc___> (same trick, just measuring RC constant; except you're measuring the C, not the R)

17:29 <lekernel> even through 4mm of glass and FR4?

17:30 <lekernel> and without false triggering from nearby electrodes?

17:32 <lekernel> the ADI chips have all sorts of bells and whistles they claim improve reliability. could they just be marketing? :)

17:34 <davidc___> hah; honestly couldn't tell you. I've never played with em in the depth required. How many touch channels do you need? There are lots of discrete capsense-to-{i2c,spi,one-wire-per-point,keymatrix} controllers out there; and they're cheap too (sub$)

17:34 <lekernel> 6

17:35 <lekernel> for http://milkymist.org/3/mixxeo_preview.png

17:35 <lekernel> just above the screen

17:35 <lekernel> they need to have some light passing through them, too

17:35 <davidc___> lekernel: ITO on glass? ($$$)

17:36 <davidc___> Also, what kind of time-resolution do you want? Capsense isn't exactly the lowest latency

17:36 <lekernel> ha, actually I have a good part of the equipment for ITO at home now =]

17:36 <davidc___> same :). Vacuum/plasma stuff seems to be some kind of universal nerd-snipe

17:36 <lekernel> though ITO is a pesky substance and it'll take a lot of tweaking to deposit it reliably

17:37 <davidc___> Maybe you could do a copper ring around the lit part?

17:38 <lekernel> I'm thinking of gluing a thin PCB (hence the capacitive sensing having to go through FR4) with holes for the lights against the glass

17:38 <lekernel> with pogo-pins for contacting the UI board

17:39 <lekernel> this way I can keep that small board 1-layer

17:39 <lekernel> LEDs would go on the UI board near the pogo-pins, too

17:40 <lekernel> 0.4mm FR4 seems to be easy enough to source, digikey has it

18:39 <davidc___> lekernel: you could use through-pcb LEDs firing right into the FR4 + using it as a diffuser

18:39 <davidc___> (maybe with strategic slot-routes that get filled with black paint for isolation)