[sdiy] Sequencer design

[rsdio at audiobanshee.com]

On Mar 10, 2015, at 9:28 AM, Florian Anwander <fanwander at mnet-online.de> wrote:
> On 09.03.15 18:36 , Byron G. Jacquot wrote:
>> I'd nominate anything built around the CD4017 as the "classic" design. It's a sequencer on a chip - just add clock. You can daisy chain them for more steps - there's a schematic in the datasheet.
> Not that this would be wrong, but a big advantage of scanned potentiometer outputs / switch outputs (vs. the 4017-solution) is, that you can "sequence" any material you want. Instead of the voltage you can take audio. You even may use the potentiometers as crossfaders between two signals. Also you may feed different signals in each potentiometer. Have a look at the SND SAM 16 to get an idea of it:
> http://www.s-n-d.com/sam16e.html
> 
> All this can be achieved only if you scan the outputs of the potentiometers.

This is almost as easy as the counter/demux. All you need is an analog multiplexer. The 4051 handles 8 channels, the 4052 handles 4 channels, the 4067 handles 16 channels, and the ADG732 handles 32 channels.

The only limitation I can think of is the voltage. Most analog mux chips will handle 10 V, which is fine for CV. But if you want to handle audio with a high signal-to-noise ratio, then you need to buy the more expensive versions of these analog mix chips that can handle 40 V or more.

As for driving the LEDs to indicate the active step, a logic decoder chip can take the same address as the analog mux and it will activate the appropriate output. However, in this case you're limited to 8 channels with a single chip, and would need cascaded decoding to handle 16 or 32 channels. Again, not difficult.

Brian