Dirty Digital LFO--Done? Time to Move on? Um...no.

Progress! I have the dirty digital LFO, Atmel 328/Embedded C based, in my rack, modded, and it works.  

In the past week or so I've added new features to the circuit posted here.

Question is, am I done?  Nope.

 At the very least, the Dirty Digital LFO needs a new front panel. After further testing I'll do that and update this post with a better looking photo and yes--it's coming--a new "pots board" PCB, replacing the bevy of daughterboards, kludes, and mods with something that is like likely to catch fire.

For the history of the Dirty DLFO design and build see previous posts are here, here, and here.  The "minimal" Atmel 328 PCB is discussed here (github for it is here--the Atmel 328 processor is on its own PCB for this build, so you will need the minimal Atmel 328 PCB to build the entire module).

Here's what I added:

• A waveform "Riley" reset (named after the synth guy who told me I had to add this feature), using the 3904 inverting buffer and an external interrupt, post for this subcircuit and 328 external interrupt basics is here. When the interrupt #1 pin goes low, the waveform is impacted--for instance, saw is reset to full CV.  the idea: on the downbeat start the waveform over. Get on the dance floor n' boogy, Holmes. Cool.  
• 4 new waveforms: saw, S/H (not really a waveform....) and 2x random CVs.

The problem (?) is that the module as racked today is an amalgam of daughter boards, standoffs, 4/40 screws, hookup wire, and snot.  

I have had issues with heavily modded projects catching fire (read more here--one of my most popular posts--why, scotty, why?) so this poor fabrication makes me--uneasy?  But for now it's in my rack.  

I was too lazy to take off the front panel for the Riley reset mod....

 

The inverter board (post here) for "Riley Reset" is installed. thanks to PCBWAY for fabricating this board and sponsoring the AudioDIWHY blog.

A portamento board was added to the output, to cut the grit and provide a "lube" out. Post for portamento is here; I used a .68 cap, not 22uF as in initial post, as well as a 100K lin pot; might sub out 50K down the road.

The thing that's driving me nuts right now is that I need to redesign this whole thing again. In addition to getting rid of the daughterboards as a safety precaution, it would add a lot of functionality (and not too difficult) to add buffered/clamped CV control to the output offset via a jack and additional op amp, as well as BOWAL control (previous post here) to the CV input. That would make the module more useful.

Problem is with my day job, music, family, everything on my bench and all the rest I may not have time. 

Shutup and just do it....

Modified with Riley reset only.  I put a new processor chip on to the dev board before adding the waveform reset and new waveforms to firmware. If I completely screwed up the code I could go back to a chip that I know worked. Turns out I didn't and the new waveforms worked.

In action...

 

As usual I have posted everything for the project on Github....The updated C code, PDFs of schems, gerbers, wiring diagrams etc. etc. used in the project is here. Comment if you have questions about any of this.

Forward? I am not sure what to do next. Part of me wants to rebuild this with a waveform lookup table, but also I am getting ready to start working on a project using STM32's--first time I've worked with ARM.  too many things on the bench--a good problem to have. Update 8-29-21: I've created a new pots board gerber and set it off to fab. Also a slighly improved minimal AVR dev board. If I can get that to work I'll update the github and maybe create a final post for this module. 

Update 10-31-21: Rev 2 of this design is done and works, I've added additional features for even more bizarre waveforms at output. It's also a lot cleaner without the daughterboards etc. Post is here.

See ya soon.