[sdiy] Pressing buttons

[Scott Gravenhorst]

Interesting...  I have a mechanical analog of this, something out
of what I believe to be old telephone equipment.  I also have a
table fan that has a mechanical 4 button speed switch which works
like this.  Interesting that this is fairly easy to do with
mechanics and trickier with logic electronics.  At first thought,
it appears that it should be easy, but not!  I wonder if it might
be possible to do with 4 layer devices like SCRs as the "memory"
elements and a more "analogue" approach.  (I realize that CMOS is
the desired output, the SCRs could be buffered easily...)  A
common reset should be fairly easy to implement, cut the current
to all the SCRs and whichever one is on will turn off.

Or maybe that was just some neuron noise in my head...


"jhaible" <jhaible at debitel.net> wrote:
>This is not trivial.
>Seriously.
>Ok, for 4 buttons you can do it like this:
>4 RS-Flipflops, with a 3-input OR gate
>(diodes ...) on each Reset input.
>Button 1 is connected to SET of FF1
>and to RESET (via the OR) to all other FF's.
>
>This solves your problem - Korg did this in
>the CX3 if memory serves.
>
>BUT if you want a large number of buttons,
>it becomes more difficult. Surely you don't want to
>route 15 reset signals for each button to 15 OR-gates
>with 15 inputs each (for a 16-button-circuit).
>
>So the idea woud be to have *one* common RESET
>line to all FFs, and one dedicated SET line from
>a button to its own FF. Of course the SET must have
>priority over the RESET input in each FF. This is
>unusual, but possible.
>
>Now comes the tricky part: Your Buttons produce
>glitches, and your FFs have setup and hold times.
>You *must* make sure that the "button's own FF"
>(which gets a SET *and* RESET, while all the others
>only get a RESET signal) will not fall back to reset
>just the moment you release the button. You have
>to make sure that SET goes *after* RESET, under
>*all* conditions. Button pressed short, button pressed
>long, button bouncing, buttons triggered fast after
>one another. This is not trivial, and I'd like to see a solution
>for this myself !
>
>JH.
>
>
>
>-----Ursprüngliche Nachricht-----
>Von: >>>marjan<<< <urekar.m at EUnet.yu>
>An: Synth-DIY <synth-diy at dropmix.xs4all.nl>
>Gesendet: Sonntag, 12. Mai 2002 21:14
>Betreff: [sdiy] Pressing buttons
>
>
>>
>> I'm in need of nice little cct, there are 4 buttons SPST,
>> I need 4 outputs, when you press button#1 output#1 goes to
>> logic 1 (high) all other outputs to 0, press button#2
>> output #2 goes high all others to 0 and so on.
>> Also output must stay high until active, not just a pulse.
>> And it should be CMOS.
>> I'd make it but my mind spits just sorta priority
>> codec with f-fs solution, with lotsa gates. There should
>> be something more elegant and simple, even inside a cmos
>> chip?
>>
>> I know you know it!
>> And no PIC solutions please :)
>>
>> ThanX!
>>
>> --
>>
>>
>>
>> marjan
>>
>>
>> me    : Marjan Urekar
>> e-mail: urekar.m at eunet.yu
>> s-diy : http://surf.to/marjansystems
>> music : http://go.to/forcemajeure
>>
>>
>>
>
>
>

=========================================================  
- Government: The other religion.
- The media's credibility should always be questioned.
- Lambs who lie down with lions are lunch.

-- Scott Gravenhorst | LegoManiac / Lego Trains / RIS 1.5
-- Linux Rex         | RedWebMail by RedStarWare
-- FatMan: home1.GTE.NET/res0658s/FatMan/
-- NonFatMan: home1.GTE.NET/res0658s/electronics/