Proximity Sensors as a Switch Matrix?

Yendor

I saw Beau’s article in N&V·and gave me some additional insight to solving a design concept I’ve been working on...(thanks for the article Beau!)
·
I'm converting an·24 note Organ Bass Pedal·Board into a MIDI controller.
·
The pedal board is mechanical and has about an 1" to 1-1/2" range from an ‘on’ to an ‘off’ position, so I need to design a switching matrix (e.g. 3X8 or similar) to interface it to the mechnical pedals.
·
I've thought about approaching it·with the typical reed relay/magnet switch matrix that I've seen on the web, but feel this is a bit clunky and expensive.

I've notice that some keypads – from remote controls, to calculators, to toys,·have·rubberized buttons·containing a small round metal disc. ·It seems that this metal is non-magnetic, thus ruling out hall-effect detection.
·
Most of the other components with these keypads are SMT’s, and the control lines go right to a specialized 'black dot' micro or IC, but·I’m not seeing the analog components that was in Beau’s article…
·
On these PC boards, just over the button, the traces are etched with what I call an “E3” pattern. ·The traces don’t touch.
·
I don’t think this would pass as a PCB equivalent of an inductor, but would this be the same concept as a Proximity Sensor?· Any thoughts on theory or what is this called?
·
I have a synth that has uses a square-shaped coil in the PCB etching and·used for velocity sensing and polyphonic after touch.· ·I can see that it’s probably using this type of approach, but the "E3" pattern throws me…
·
I can post a jpeg of the the board that hs the "E3" pattern, if necessary.
·
My goal is just on/off switching right now.·

Any info much appreciated
·
Thanks!
·
Rodney

agfa

Rodney,



I believe the disc on the buttons is a conductive material and the logic is detecting a change in resistance.

Yendor

Hi agfa,

Thanks for the response.

I tried ohming it, and it's not conducitive. The keypad membrane is rubber and the 'metal' disc is embedded within the rubber.

I'll try posting some pictures shortly.

Rodney

Zoot

They are like push-buttons switches, where the contacts of the switch are the PCB traces -- the interlocking "E"s are to guarantee electrical contact when the metal disk in the rubber part of the switch comes down on the PCB. Basically, if you put a wire across the E's, you get the same as a button push.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
When the going gets weird, the weird turn pro. -- HST

1uffakind.com/robots/povBitMapBuilder.php
1uffakind.com/robots/resistorLadder.php

Beau Schwabe

Rodney,

As Zoot says these are push button switches and not coils.

http://www.xymox.com/rubberkeypads.htm



Thanks for mentioning the Nuts and Volts article though... you could create an analog switch·using a coil similar to the way that·you describe, but most if not all keypad switches do not use·coils.·


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

kwinn

Rodney, these are almost always capacitive sensors. When the key is pressed the metal foil comes closer to the "E" pattern and increases the capacitance slightly which is sensed by the keyboard encoder. Almost all PC keyboards use this type of switch. It works well on a keyboard, but I am not sure how reliable it would be for your application. Have you considered using LED/phototransistor pairs as photo interrupters on each key. Three rows of eight would be simple to scan.

PS - You may want to take a look at " forums.parallax.com/forums/attach.aspx?a=24522 ". The Piezo Film Vibra Tab sounds like it would also do what you want.

Post Edited (kwinn) : 3/3/2009 3:36:23 PM GMT

Yendor

This is great guys!· Thank you!

The patterns that Beau showed·are spot-on.·

I'm pretty sure, at least in the examples that I'm seeing, that the keypads are non-conductive, hence is probably capacitively coupled, or am still I off base here?··Looks like I'll need·break into my remote again to check continunity of the keypad membrane.

So if this is·capacitively coupled, what would this be·called to allow me to google around?·'capacitivly coupled switching'?··

I guess this would use a RC time constant in a matrix similar to what Beau mentions in his article.··Looking on the 'net, I'm seeing that there are patents applied for this, but with all the products using this scheme,·I wouldn't think it would be that big·of·a deal.

As far as application, all I need is on/off right now.· Would appreciate feedback of·what you (or anyone) is thinking otherwise.·

I guess it wouldn't hurt to start some experimenting in the mean time.

While I'm at it, I'll take some pic's of my keyboard that I believe, uses an inductive approach.

Thanks!

Rodney

·[noparse][[/noparse]edit] the·piezo vibra tab may be cool as a percussion sensor, but I think would really be a challenge to code in·an 8X3 matrix.· Let me ponder on that for a bit.

[noparse][[/noparse]edit2] ok, looks like it would work! Interesting, and·the price is right, too.· I guess I can put·in 3 latches and read the bits accordingly, and prob. will be cheaper to implement than with a pcb board.· Thanks!





Post Edited (Yendor) : 3/3/2009 10:49:56 PM GMT

kwinn

Yendor, in reply to your last post on 3/3/09, the capacitive switches used on keyboards produce a very small change in capacitance and a correspondingly small signal that is more of a fast pulse. If the pulse exceeds a certain threshold it triggers a flip-flop. You could probably use aluminized mylar or aluminum foil with plastic food wrap for insulation to do something similar, but the vibra tab is probably easier and more reliable.