[Agent420] brought up this touchpad and VFD hack in the comments on our capicitive sensor guide post. He had broken dell laptop from which he harvested the touchpad and an HP laserjet that contributed the VFD. Though the touchpad communicates using standard PS2 protocol, he wanted to use it with his Atmel 8535 AVR which required him to write some custom code. In the picture above, you can see the VFD displaying the coordinates of his finger. You can download his code as well as the spec sheets for the different pieces on the project thread.
That’s neat. I would love to see a touchpad hacked into something ^^
Im not sure what a touchpad can be used for besides computers, ipods, phones, etc.
How about motor or light control.
Would be a cool control scheme for an R/C chopper. Maybe not practical, but the best hacks aren’t. :)
Umm, does he have a manicure w/ clearcoat? Is this guy in the mafia?
what about making this a sound effect controller?
@baggins: That’s what nails look like when kept trimmed and in a flash.
@dennis: thats what I was thinking, like a homebrew kaoss pad. I know everyone hates on circuit bending, but this could really take it to a new level.
@st.jimmy Oh, thanks. I wouldn’t know because I keep my nail trimmings in a jar and leave the flash on my camera.
Doesn’t the PS2 protocol only describe motion [and clicking]? how could you get absolute position from and old ball mouse, or optical mouse? or is the position information displayed above relative to something your tracking in the avr? ex: if you slide your finger repeatedly in one direction, the numbers keep moving in that same direction. my confusion is if that’s actually the coordinates of his finger, or just a cursor position in the avr.
I know some touchpads expose the absolute position [synaptics/moodpad app], but i am not sure if they also use PS2 interface/protocol, or if the do, if they modify it.
—-PidGin128
Might be a dumb question, but is there an alternative to the touch pad? O.o possibly something you can etch out on a pcb?
@aw
capacitive sensors can work as touch pad but code is harder
@baggins: you know what I meant.
@pidgin128: As far as I can tell, and I might be wrong on this, PS/2 works off of relative position, with the origin (aka (0,0)) being the position of the mouse at start-up. This could be reported as absolute position by a touchpad. Again, I could be wrong.
@aw: I remember seeing something like that on ladyada’s site once, but that could’ve been capacitance sensors, like therian said. You could build a conductive grid, like some remote buttons, or those cloth touchpads.
Wow, made front page from an off-handed post – cool ;-)
@baggins – yes, I’m in the mafia. Nails are trimmed by natural wear on my guitar’s fretboard. And I photochopped the clearcoat effect to compensate for my resin stained fingertips.
From what I can determine, most mice can be programmed to return either relative or absolute positioning. If you refer to the Synanptics tech pdf or my code you will see how this command is sent.
Note that it appears Synaptics moved their link:
http://www.synaptics.com/sites/default/files/511-000275-01rA.pdf
the project turned out to be relatively easy to implement, and given the inexpensive availability of surplus touchpads seemed a good alternative to dedicated capacitance type devices. really this type of hardware could be used for any kind of digital or analog input, from keypads to dimmer switches. though I have not tried it yet, I think the touchpad could be mounted behind a thin enclosure wall or case, so no external access to components would be required.
“Might be a dumb question, but is there an alternative to the touch pad? O.o possibly something you can etch out on a pcb?”
Im not sure… I know that is how other capactive sensors work and that may also be how the touchpad is built.
Imo, the advantage to using the touchpad module as a unit is that it is pre-assembled and provides a known working platform to extend from, requiring only a couple of wires to your controlling circuitry… half the work is already done for you!
plus, it’s a hack ;-)
Also, to clarify, I am explicitly using the absolute positioning function with this project, so all coordinates are exactly those that are returned by the touchpad. this allows me to always know exactly where the pad is touched so that different areas can be programmed to perform different tasks if desired. and you can always create an internal virtual relative position coordinate from the absolute values in software if desired, but you can’t go the other way around.
Very interesting has anyone tried to interface a iPhone multitouch capacitive panel?