Prototype Wednesday :: 25 Oct 23 :: a 16-step analogue sequencer using 2 Baby8s?

 Since I was given some Rakit kits one year for my birthday, I've been wanting to join two baby8s to make a 16-step sequencer. I've returned to this project a number of times but never really got out of the starting gate. (Yes there are one or two published projects but part of the enjoyment of this for me is to work out how to do it myself from scratch.)

Initially I attempted to use logic to direct 8 clock signals into one and then 8 into the other. But then hit on the idea of running them in parallel - offsetting them by inverting the original clock signal. 

(This has the added advantage that it would be possible to 'swing' the rhythm by adjusting the duty cycle of the original clock signal).

This is the furthest I've got with it so far. It makes an eye-catching and ear-splitting demo but there are still too many problems to solve. Some that I had predicted and some that I discovered along the way. 

None of the challenges are unsurmountable but it requires more analogue electronics than I'm comfortable with and it started to feel as if I'd be building so much stuff around the baby8s that I could have started from scratch in the first place.

This led to me sketching out something of my own.

Me being me, this idea is microcontroller-driven, with as much functionality as possible done in software rather than hardware.

Doing this with a microcontroller opens up a world of possibilities:

  • we don't have to be limited to 16 steps - for example we could have A and B patterns as alternative switchable 16-step programs, or run them sequentially to make 32 steps
  • the ability to store and recall patterns (some of these ideas have helped to make my decision on lit tactile switches versus physical switches for 'skipping' steps )
  • the ability to 'lock' the output to a certain scale or mode
  • a 'swing' rhythm, with adjustable swing as an alternative to straight 8's
  • like e-ther, the ability to generate MIDI as well as analogue output
  • adding a small screen and encoder (like MIDISID) to help with selecting advanced features
The first challenge with this project is the large number of analogue and digital inputs / outputs and I think I've got that sorted. I've got two analogue pots operating the tempo and duty cycle of the clock / gate signal. I'm also using an analogue mux to read, currently, four analogue pots but there's capacity here to expand that to 16. (I just haven't got the pots today, or the inclination to wire them all up). 23017 port expanders will help with the digital i/o.

You're not hearing anything in the demo because I haven't yet settled on how to generate an accurate volt/octave control voltage from a 3.3v microcontroller (and I do want it to be accurate). That's the next piece in this puzzle and I have two ideas which I will try out in the next few days to see which works best.

Comments

Post a Comment

Popular posts from this blog

e-ther 2 - Announcement and full rundown of features

Image
I'd like to announce the new and improved e-ther 2. This information is available in video form here . If you're not familiar with e-ther, then it's a small theremin-like instrument with MIDI and audio out . At first glance the differences aren't obvious and indeed the big changes to the hardware are internal.  However, I hope you'll find the list of new features as exciting as I do.  Hardware-wise, I've put the microcontroller and DAC onto the main board.  Although this increases costs (for me, not for you), there are some big advantages. Besides looking more professional, the audio module is no longer a 'buy separately' optional extra. It's incorporated and the audio jack is just there if you want it.  The audio module used to fit below the board but with the audio capability now built-in, the profile is more compact. Most of the new features are down to firmware which means that a large proportion of what I'm calling 'e-ther 2' will be
Read more

Enclosures for MIDISID

Image
 The default configuration for MIDISID  looks fine and is functional. I have fitted wooden cheeks to my own, which gives it a look and tilts the top forward, making it a little easier to use when you're sitting back from it. MIDISID was designed from the start with 'bring your own box' in mind. It's a modular design, the top and bottom parts are easily separated and connect with a 4-way jst-sh cable. In my head I saw it being used in a sloping-top case. Paul of Bad Dog Designs has come up with a different solution. It achieves the same goal - tipping the controls and screen forward. It's a 3D print in two parts. It takes zero effort to fit - MIDISID just drops in. I don't know whether he has made the design files public but  please contact him  if you're interested. Update Nov '23 The current version of MIDISID is now larger, has a tilted front and custom side supports: Owners are still being creative with enclosures. Here's a beautiful one from  @k
Read more

MIDISID - second batch and giveaway

Image
The first MIDISID batch ('earlybird' pricing) have now all gone, I have the materials for the second batch and the first of those are now made.  There are now black and purple editions. At risk of going all Tim Cook, the black really does look gorgeous. Hardware-wise they're hardly altered but the firmware ('MIDISID OS') has come a long way, new features include: you can set glide speed modwheel support; choose vibrato, filter cutoff or pulse width Increases the range of vibrato, to enable spooky effects Now supports RPN 0 messages - for setting the pitch bend range. (This was important for getting my tiny MIDI theremin working - see https://twitter.com/midi_in_out/status/1574870874158764035 ) ( The manual is here in case you'd like to read about all of the features.) I'll shortly be mailing those who have bought MIDISID already with details for the new firmware. Giveaway Because I'm an idiot, I (genuinely) accidentally bought a second copy of @C64Audio
Read more