Guest OzMax Posted December 3, 2003 Report Share Posted December 3, 2003 First up, sorry to those people thinking... Oh my god, yet another cool thing for MIDIBOX. Alas no, but perhaps in the near future..read on...Driving home in the car tonight, I struck on the idea that a vertical or horizontal touchpad strip would work well as potentially a replacement of an encoder.(Now I should stop right here and say.... Using such things will still be far more expensive than using motorised faders, etc,etc, BUT, may allow for miniature 16 channel mixers to be made, perhaps entire panels of trackpad with every nanometer dedicated to some MIDI parameter, etc,etc.. and also for the "What the hell" buzz.)Ok, let's paint the scene, you use something like a Midibox LC except the motorised faders and the pan encoders are now little one way trackpads. You can increase and decrease the slider positions by simply dragging your finger over it. This might be useful for when say volumes (or whatever is associated with a motorised fader) are moving all over the place so quickly that the poor little motorised fader is having a hernia (does this happen in real life? Haven't personally gotten around to using MFs). The position of the slider is shown on screen in your DAW as well as a strip of LEDS along the side much like we do for the encoders at the moment.WHY.. because it may give you all the advantages of an encoder but with the human interface simplicity of a vertical fader strip. (and of course for the "What the Hell" trip).(One could go on about possible uses, possible advantages, disadvantages, etc, but I will leave that to others)..HOWDOWEDOITI am sure most know how trackpads work, but for the two in the corner cowering under the table, basically (very basically, I have probably even got it wrong) it uses an XY grid of tracks on a PCB and then measures the capacitance between all of these wires. Your finger has a different electrical permittivity than air, so changes the measured capacitance between the wires/PCB tracks (all non contact, your finger never touches the wires, just a protective coating, Â the flip side of the PCB, etc). From these measurements it's possible to workout where your finger currently is, and also if you track these changes to tell where is it moving to. You can also measure the presence of a finger . (Imagine an entire control surface , flat, bare, transparent, but you move your finger over it to turn on tracks, move faders, etc, etc, and LEDS through the panel show you the status of what you pressed. WOA getting a little too Star Trekish for me.. back to reality)..Energise.Ok Because we are only interested in one direction, the PCB would be a heap of parallel tracks (at 90 degrees to the direction of travel. An up down slider would have tracks going left to right) because a finger is a finite size, you could also probably bridge interlacing lots of these tracks to cut down on lines you need to measure the capacitance (like, the bottom line is connected to a track 2 cm up, and then to the next one 2cm up, etc) as we are not really interested in absolute position, just state change.. You could track absolute movement and make it not only a differential control, but an absolute one.Ok to measure capacitance... ok, probably lots of ways to do it, but any of the efficient ways will probably require their own little micro controller, or one that handles ALL of the track pads in the system. You could introduce a clock signal to alternate lines and measure the cross talk, on the ones left over. Perhaps rectify that crosstalk (after the multiplexers - see next sentence) into DC value and into an A/D. You could then have the input lines multiplexed using similar techniques we should be familiar with (i.e. A_IN module). Heaps of ways (will leave for other posts). A little processing and you could workout where you are now, and if you have moved your finger. Send this to the MIDIBOX application as an increment/decrement command (with some adjustable gain/acceleration) and you have a DIY trackpad (well it will take a little more than that, but that's the basic principle).So all in all,Possible - yesDifficult - probablyCool- well I think so, but then I still use wax cylinders for recordingBUT.. is this a waste of time for Midibox, or could it have some actual applications?Ideas anyone... Quote Link to comment Share on other sites More sharing options...
Guest OzMax Posted December 3, 2003 Report Share Posted December 3, 2003 Perhaps it could be configured like this....http://www.usq.edu.au/users/maxwell/webpub/trackpad.jpghttp://www.usq.edu.au/users/maxwell/webpub/trackpad.pdf Quote Link to comment Share on other sites More sharing options...
Nat Posted December 3, 2003 Report Share Posted December 3, 2003 I'm currently testing a commercial pad for midi and I must say having this on a midibox would be loads of fun. I personnaly think that touch surfaces are much more useful than faders, it take ups less space and let's you control more.Looking forward to seeing this.... Quote Link to comment Share on other sites More sharing options...
Guest OzMax Posted December 5, 2003 Report Share Posted December 5, 2003 OK, update... Started construction, testing this weekend... ::)The design is based on + PicAXE chip (Pic 16F84 running Basic, or programmed using basic)+ Multiplexer chip+ Four resistors+ One Capacitor+ a diode(and a partridge..bla.bla.bla)It's a starting point. Then I will change the system over to use the standard 16F84-non basic chip (or if that's not widely available one of the 8 pin chips and a serial shift register to provide the multiple outputs..) Code and schematics to be released after final testing..Let the games begin...Bang (sorry, that was the pic blowing up, not the starters gun) Quote Link to comment Share on other sites More sharing options...
Guest thorstenoe Posted December 10, 2003 Report Share Posted December 10, 2003 After reading this I was uncertain exactly how this would work, but I found a nice description and thought I would share it:http://computer.howstuffworks.com/fingerprint-scanner3.htmSeems to me the software for making this work is the hardest part...The idea is way cool :) Quote Link to comment Share on other sites More sharing options...
rogic Posted December 10, 2003 Report Share Posted December 10, 2003 is there any chance to use old touchpad from some laptop? Quote Link to comment Share on other sites More sharing options...
Guest OzMax Posted December 10, 2003 Report Share Posted December 10, 2003 is there any chance to use old touchpad from some laptop?I guess, so long as the output is either some useful serial data stream, or some other useful signal. However because they are 2D (some are 3D) they would contain LOTS of information in the serial stream, much of which is not really interesting to a MIDIBOX system. I personaly am only interested in recognising an UP or DOWN motion for the first part.... Progress has been slow as I have.. well.. a day job, oh and a night job, oh and 14 bands, a life (no really), scuba diving, and other MIDIBOX constructions going on... 29 hours a day there is.. and all 29 are filled up.. Quote Link to comment Share on other sites More sharing options...
Guest OzMax Posted December 10, 2003 Report Share Posted December 10, 2003 After reading this I was uncertain exactly how this would work...I am sort of an Instrumentation Engineer, and one quickly realises that you can actually measure almost anything, using almost any technique.Capacitance is an underrated measurement technique and that article gives a quick impression as to how it works. I however am bypassing the direct measurement of capacitance as I can simply use indirect measurement. (i.e. instead of measiring capacitance, measure something which the capacitance modifies, which leads the way to things called transduction spaces, and transduction cubes, etc,etc.. standard instrumentation engineering principles) I could however make one of these sensors using, Radiative (i.e. optical), Nuclear, and mechanical methods...I could for instance easily use a strain gague which measures the strain your finger places on the PCB panel and translate that into an UP/Down signal., etc,etc.. I could use a half stripped optic fibre and could measure the refraction when you move your finger along it (which ain't a half bad idea), etc,etc.There are literally a million different ways of making this work. The trick however is coming up with an elegant and simple solution which is easy to make.Some people might be thinking, this is all fancifull thinking, but many of these types of sensors are really easy to use and make.Coming next, the neural implant radio link... Just think it and it happens.. (and you all think I am kiding.. well maybe for now, but just wait after this weekend  :-)  ) Quote Link to comment Share on other sites More sharing options...
d2k Posted December 10, 2003 Report Share Posted December 10, 2003 this all sounds pretty out there!! Hope it works!! :D Quote Link to comment Share on other sites More sharing options...
Guest OzMax Posted December 10, 2003 Report Share Posted December 10, 2003 this all sounds pretty out there!! Hope it works!! :DSo do I...I have to be carefull to not be all hype... the actual design in my head is already different from what I have presented here. Ideas are great, but they need a physical realisation.Perhaps I am not doing the right thing putting my ideas here rather than presenting a final system. I am not sure, but perhaps there are others out there working on similar things?So please bear with me, and all will be revealed (at the 29th hour) Quote Link to comment Share on other sites More sharing options...
DrBunsen Posted December 13, 2003 Report Share Posted December 13, 2003 Here's one way:http://www.wiseguysynth.com/larry/ribbon/ribbon.htm Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.