[sdiy] Percussion sequencer idea

[Dave Magnuson]

At 04:48 PM 10/3/01 -0500, you wrote:
>Dave, 
>
>what power supply do you plan to use for the perc sequencer?  


I'll build my own supply, most likely.  Most of the logic will run on 5V,
but I'll bump up the final gate outputs to around 10V, so I'll need a +12V
or +15V rail as well.  I personally like mounting a transformer in the
chassis, but I suppose you could use an AC wall wart if you don't like
messing with line voltages.


>Also, how
>would you set the pattern lengths? 
>Andre Stordeur has a nice page on counters.  A neat idea here is to use a
>combination of switchs and manual trigger jacks.
>http://www.angelfire.com/music2/theanalogcottage/diyelpro.htm


Yes, I've seen this page before.  I assume by the switch and trigger jacks,
you're referring to his binary switch setup on the divide-by-n counter?
Since my sequencers are going to be 8 steps long, I'll probably just use a
rotary switch to select the length.  I'll probably use a 6 position switch
providing selections for 8, 7, 6, 5, 4 and 3 step lengths.  Combined with
the clock speed switch, this can provide almost any common time signature
(3/4, 5/4, 7/8, etc)

The circuit could easily be modified to have jacks instead.  When setting
sequence length on the 4017 all you're doing is sending the gate from the
last step back to the reset pin of the counter.  Because of this you
couldn't use the binary method, unless you wanted to add extra circuitry.
Instead, you could provide a gate output for each step and a reset input.
You could then run a patch cable to select the length you want.  I just
prefer the rotary switch.  I'm trying to squeeze this all into a 2U rack
chassis, so I'll be saving a lot of real estate by omitting the extra jacks.  

Of course, with the extra jacks you can do some neat stuff... using some
creative VC switching you could alternate the pattern lengths.  An example
would be make a voltage controlled SPDT switch (you could use reed relays
or CMOS) connect one input to the step 4 out and one to the step 8 out.
Connect the switch's output to the reset in on the sequencer.  With the
switch's CV input "low", it would repeat the 4 beat loop, and when it
receives a "high" it would continue through the full 8 steps.  Since all 4
sequencers can be reset independently, you can come up with some very
complex sequences.


>I know that banana plugs shouldn't be stacked at a single input, but
>perhaps sequential gates is an exception?  What do you think? 


It depends on how the circuits are designed.  If the gate outs were run
through a diode, they could be summed together with stacked bananas at an
input.  If the gate outs weren't somehow protected from backfeeding, all
sorts of odd behavior or damage could result.

Also, for anyone that's interested in my design, I've been creating logic
symbols in AutoCAD for all of the ICs I need.  As soon as I make all of my
logic symbols, I'll start piecing together the schematic.  So far I've
finished the MIDI to DIN-Sync portion.  This is virtually identical to the
datasheets/notes for Blacet's custom IC.  I've decided to add MIDI thru
jacks since it's so trivial to add to John's design (and you can never have
too many thrus)

Well, back to logic-land in AutoCAD :)

Dave



Resonant Frequency:
resfreq at hoohahrecords.com
http://www.hoohahrecords.com/resfreq/index.html