[sdiy] Sequencer thoughts

[patchell]

    Well, it is probably a bit different from the way you are going, but I did
something sorta like what you are talking about.

    If you dig through this really big PDF
file:http://www.silcom.com/~patchell/seqproj/seqproj.html

    You will find a sequencer controller (I used HC logic) that will do most of
what you want...and may even give you some food for thought.  It might be a bit
more that what you want to fool around with ...

Peter Grenader wrote:

> All,
>
> I am in the midst of my annual 'think about making a counter-based
> sequencer' pilgrimage.  It came about this time because I got two Blacet EGs
> recently (highly recommended to anyone who takes synthesis seriously) and i
> need a means to control all phases of the ADSR or AD with fixed voltage,
> but having a separate, tunable voltage source for each phase (four banks).
>
> High on my list of needs for any sequencer is control of direction, random
> most important among that and I hoping for some input from the group on the
> best (read: easiest) way of accomplishing this.
>
> For clarification, I am mapping out a 4 bank, eight step sequencer.
>
> I was thinking of an engine which consists of 4051 demux which is being
> driven by a 4516.  Easy enough.  For the random, I was going to utilize the
> present inputs of the 4516, which would receive binary data which is latched
> by a 4042 by the same clock which was advancing the 4516.
>
> It's where the 4042 is getting its four bit data that I am scratching my
> head about.
>
> Options I've come up with are:
>
> 1)  A single four bit data stream (another 4516) run at a super-high
> frequency which, when gated by the 4042 at a considerably lower freq, should
> give me quasi-random four bit number streams. A four bit sample and hold of
> sorts. I believe this is how Serge did it, although I have not seen one of
> those schemo's in a few years.
>
> 2) Four independant white noise signals into four different comparitors, the
> output of each  routed into 4042 latch inputs.  Not sure what in the hell
> would happen there, but it seems less random than option 1
>
> 3)  Make four different clock circuits running at four different frequencies
> and route those into the 4042 latch inputs.  This seems like it would give
> better results than option 2, and would require less circuitry (for me, a
> big thing).
>
> 4)  The old LM3914 trick - using one of these bad boys with a single signal
> input which would be attenuated so that only the first four comparators
> would be active.  At this point, it would allow all sorts of direction
> control, not just random.
>
> I am going in the direction of 1 or 4.  Any input, along with any other
> methods you might recommend, would be GREATLY appreciated.
>
> thanks for your time,
>
> PG

--
 -Jim
------------------------------------------------
* Visit:http://www.silcom.com/~patchell/
*-----------------------------------------------
*I'm sure glad Merry Christmas comes just once a year
* -Yogi Yorgensen
------------------------------------------------