Author |
Message |
Starspawn
Joined: Jun 14, 2013 Posts: 85 Location: Oslo
|
Posted: Sun Jun 23, 2013 5:40 am Post subject:
1V/Oct Keyboard suggestion, help to confirm choice. Subject description: Confirm if possible. |
|
|
I want to use this keyboard as it is small and cute and has a motion sensor.
However it is not single bus, but I figured if I A: removed all diodes and connected resistors between the keys at the holes nearest the keys instead Id get a resistor line as in the schematics.
B: If I then connected the 6 pins on the other side (lower side here on uploaded schematic) Id have a single bus connected to all switches.
Have I missed something?
Also its not that many keys, is a transpose switch as easy as adding resistors for an octave and switching them in or out? |
|
Back to top
|
|
|
analog_backlash
Joined: Sep 04, 2012 Posts: 393 Location: Aldershot, UK
Audio files: 21
|
Posted: Sun Jun 23, 2013 7:47 am Post subject:
|
|
|
Hi Starspawn.
It's a bit hard to say for certain without seeing the PCB, but it seems possible. My biggest concern would be the resistance of the individual contacts. If you're going to use Ray Wilson's keyboard controller design, with 100R per semitone, the resistance of the contact needs to be (near as dammit) zero (i.e. much, much lower than 100 ohms). On a lot of cheap keyboards with rubbery contacts, this is not the case. Assuming the contact resistance is very low, it sounds do-able (see attached).
Also, switching a 1K2 1% into the chain should drop the keyboard range by one octave, so the switching idea seems sound.
You may get a reply about not bothering to remove the diodes and using it as a digitally controlled keyboard, but I don't have any experience at that so I'll keep quiet
Gary
Description: |
|
Filesize: |
124.74 KB |
Viewed: |
333 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
Starspawn
Joined: Jun 14, 2013 Posts: 85 Location: Oslo
|
Posted: Sun Jun 23, 2013 7:59 am Post subject:
|
|
|
Cheers, that diagram is how Ive done it (got bored waiting )
Is there a clever way to measure the contact resistance before I build the MFOS PCB? |
|
Back to top
|
|
|
analog_backlash
Joined: Sep 04, 2012 Posts: 393 Location: Aldershot, UK
Audio files: 21
|
Posted: Sun Jun 23, 2013 8:33 am Post subject:
|
|
|
I'd just connect my multimeter to the KB1-6 common and the end of the resistor which is in the D1 position, press the first key and see what resistance it shows (or go to the other end of that resistor and see if the combined switch/resistor resistance is much different to 100 ohms when the key is pressed). I suppose that you should test more than one key, to see if the switch resistances are consistent, but then the resistance will be n x 100R plus the switch resistance.
Gary |
|
Back to top
|
|
|
Starspawn
Joined: Jun 14, 2013 Posts: 85 Location: Oslo
|
Posted: Sun Jun 23, 2013 10:32 am Post subject:
|
|
|
Bah, looks like 450 ohms resistance on just the basic switches.
I can replace them with something else I suppose, any tips for disc type push switches for such? |
|
Back to top
|
|
|
analog_backlash
Joined: Sep 04, 2012 Posts: 393 Location: Aldershot, UK
Audio files: 21
|
Posted: Sun Jun 23, 2013 11:28 am Post subject:
|
|
|
Yes, I thought that might be the problem. When I built the MFOS keyboard controller, I used 6x6mm tactile switches like these:
http://www.rapidonline.com/Electronic-Components/Miniature-tactile-switches-71703
I made up my own PCBs to mount them on, but you might be able to fit them to your PCB after a little surgery! The length of the actuator will depend on the gap between the PCB and your keys (I think I used 7mm). The only problem with using tactile switches, is that you feel the keys clicking, but I've had no problems (so far) with reliability. To get a click free keyboard, you can go down the magnet plus reed switch route, or if you clever, you might be able to make your own contacts with phosphor-bronze wire (again, I'm no expert on that...).
Gary |
|
Back to top
|
|
|
capicoso
Joined: Nov 19, 2012 Posts: 128 Location: Argentina
|
Posted: Sun Jun 23, 2013 11:40 am Post subject:
|
|
|
i also built mine with tact switches, on a 4octaves tonebank. It feels weird yes, but there are other tact switches, a bit more expensive, that doesn't have the click, you could check those. |
|
Back to top
|
|
|
Starspawn
Joined: Jun 14, 2013 Posts: 85 Location: Oslo
|
Posted: Sun Jun 23, 2013 11:42 am Post subject:
|
|
|
Thanks, the pcb is as of now just a row of holes for the resistors, so I can just fit the 4.3mm high switches instead of the plastic row, should be about the same height, and if not I can adjust with felt on the front edge.
Cheers for all your help, Ill come back when Im done |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Sun Jun 23, 2013 12:03 pm Post subject:
|
|
|
analog_backlash wrote: | If you're going to use Ray Wilson's keyboard controller design, with 100R per semitone, the resistance of the contact needs to be (near as dammit) zero (i.e. much, much lower than 100 ohms). |
Sorry to break in ... but ... actually that is not true.
The resistor chain sets a voltage divider (and you want that to be as low in impedance as possible). Then it is the loading circuit that sets bounds on the usable resistance range for the switches.
I've looked into this in the past and remember the loading resistance to be quite high, and so the actual contact resistance of the switches does not matter much. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Sun Jun 23, 2013 12:19 pm Post subject:
|
|
|
Cynosure was kind enough to find back a previous discussion of the matter (with a bit more detail) http://electro-music.com/forum/post-360207.html#360207
Thanks Jake :-) _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Starspawn
Joined: Jun 14, 2013 Posts: 85 Location: Oslo
|
Posted: Sun Jun 23, 2013 1:37 pm Post subject:
|
|
|
Im still wrapping my head around that, as I dont get the math of all this yet, but it seems like its a go after all
Because resistors increase the whatever on their line, but the contacts resistance for output between resistors is relative to another thing, not the thing the resistors do
Sigh, must read theory finally ...
(Resistors affect voltage, but the resistance on the contacts affect how much of that voltage the circuit can draw, and so must be large in relation to max load to change the values?) |
|
Back to top
|
|
|
capicoso
Joined: Nov 19, 2012 Posts: 128 Location: Argentina
|
Posted: Sun Jun 23, 2013 2:15 pm Post subject:
|
|
|
Starspawn wrote: | Im still wrapping my head around that, as I dont get the math of all this yet, but it seems like its a go after all
Because resistors increase the whatever on their line, but the contacts resistance for output between resistors is relative to another thing, not the thing the resistors do
Sigh, must read theory finally ...
(Resistors affect voltage, but the resistance on the contacts affect how much of that voltage the circuit can draw, and so must be large in relation to max load to change the values?) |
As long as that contact resistance is the same on all buttons it's fine i think |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Sun Jun 23, 2013 2:30 pm Post subject:
|
|
|
capicoso wrote: | As long as that contact resistance is the same on all buttons it's fine i think |
Even that is of minor concern _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
analog_backlash
Joined: Sep 04, 2012 Posts: 393 Location: Aldershot, UK
Audio files: 21
|
Posted: Sun Jun 23, 2013 2:49 pm Post subject:
|
|
|
Blue Hell wrote: | analog_backlash wrote: | If you're going to use Ray Wilson's keyboard controller design, with 100R per semitone, the resistance of the contact needs to be (near as dammit) zero (i.e. much, much lower than 100 ohms). |
Sorry to break in ... but ... actually that is not true.
The resistor chain sets a voltage divider (and you want that to be as low in impedance as possible). Then it is the loading circuit that sets bounds on the usable resistance range for the switches.
I've looked into this in the past and remember the loading resistance to be quite high, and so the actual contact resistance of the switches does not matter much. |
Thanks for that information Jan - that's really useful to know. I too will need to think about that one! It opens up a few possibilities with some old keyboards I have...
Gary |
|
Back to top
|
|
|
analog_backlash
Joined: Sep 04, 2012 Posts: 393 Location: Aldershot, UK
Audio files: 21
|
Posted: Mon Jun 24, 2013 4:45 pm Post subject:
|
|
|
I've read the attached thread and I think I understand it now - the key switch is not actually part of the resistor chain, so (within reason) it's resistance is not affecting the voltage tap at that point and the voltage follower deals with any impedance problems (I think - I'm very slow). I'm sorry for leading you up the garden path Starspawn As I said before, this is a very useful post, as I've been deliberately avoiding these types of keyboards in the past. I can now buy a crappy old car boot sale keyboard and get 1V/octave out of it
Thanks Jan |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Tue Jun 25, 2013 11:39 am Post subject:
|
|
|
Yes, that's the correct way to look at it - glad to have been helpful there :-) _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
|