See blog post for detail: http://audiodiwhy.blogspot.com/2021/10/dirty-digital-lfo-done.html

Holy smokes (not literally--no smoke yet!!) this one was a lot of work.  

But it's finished; much improved from its thru hole build predecessor, and I am extremely happy with the results.

If you want to get the history for this module, there have been many previous posts about the D-DLFO leading up to this finished module:

Leaving Arduino Sketch for Pure C: here (it was probably easier to build a module of this complexity with embedded C vs. Arduino Sketch--I am glad I went down the embedded C road).

Minimal AVR board: here

Previous DLFO with simpler buffering (and less features): here and here.

Adding waveform reset via external interrupt, with 3904 based buffer: here.

Fabricating and troubleshooting the SMD "pots board" for this build: here.

Also I want to thank my faithful sponsor PCBWAY for helping me realize this build. This was a tough fab, using 0803 SMD parts and intricate stencils and is an advanced project. Their support and quick turnaround for the prototype boards was extremely helpful, so please help me continue to improve this blog and check them out.

Getting Dirty? Overall I am really happy I hung in there and rebuilt the original Dirty DLFO module to do more. The waveforms you can create with it go from normal tri/sine/ramp/saw to truly bizarre. Cool! 

Check out a few scope shots. That's the wonder of DIWHY I guess, you end up with things that can be--well--different?

Generous lube makes the signal analog like....

Portamento turns tri into a sloppy sine. Good enough.

My stock portfolio?

Hebrew from an LFO?

Hello?

More Cowbell!

the path not taken, right?

Using the Dirty Digital LFO: The module has two outputs: "Lube" and "Out", the latter is the output through a portamento subcircuit.  

The lube gets rid of the some of the digital grit which may or may not be what you want. 

There are 3 ways to change the frequency: the frequency pot at the top right, a slow/fast switch, and a 0-5V CV in. This is standard LFO stuff. You can set the output bias offset with the OFFSET pot (if I had more room I would have made this voltage controlled, but enough already right?) the WOMBO input mixes CV to the DAC out (see below); finally the CV bias in adds DC offset to the incoming CV in case a boost is needed.

What is heck is a "WOMBO"?  I had to call it something....some of the strangeness you see in the scope shots above is due to blasting CV in parallel into the DAC buffer output. We are not mixing the additional signal into the negative feedback side of the output op amp buffer--we are mixing it into what is normally used to set the bias offset (often tied to ground, but here treated as a mixer).  

I got this idea working on a friends' Ken Stone Dual Wasp Mixer, and thought it would be interesting feature to add to an LFO....it turned out to be the case. 

I had no idea what to call this input, so I chose 5 letters at random: "WOMBO". There you go!

The PCB is wired for +/- 15V but should work on 12 with no problem; the PCB board I laid out accommodates a Euro power connector. 

As it's coded now, the board supports 6 waveforms and has a S/H feature:

• Tri
• Ramp
• Square
• Saw
• Random 1 and 2 (CV controls frequency, or, CV controls frequency and randomness)
• S/H (reset is used to sample whatever is at the CV in and put it at the output).

There is also a "Riley reset". What? See previous posts here and here. This didn't work at first, after about 3 hours at the bench I realized that I had wired the interrupt to D1, but coded it for D0. Gotta love them dumb mistakes. 

Since this interrupt starts the waveform over, you can get further complexity, such as turning a triangle into a saw, doing psuedo duty cycling of the square wave, and so on. 

Thanks to Max, Dave and the guys at BAM for insisting that I put this in. It was a really good suggestion.

On to this post's useless bench build photos:

  Showering off: Overall, this experience was AudioDIwhy at its best (for me anyway...)  

After hours of enjoyable and peaceful C coding, fantastically frustrating hours fixing and debugging, and much time spent soldering, wiring and having dumb bench fun, I end up with a voltage controlled LFO that may be like nothing else out there--if you have an LFO that can display hebrew characters on a scope, lmk--I didn't have one for sure; but now I do.

I feel I've come a long way since I was perf boarding little op amp circuits and seeing if they smoked. Improvement! In spite of my psychiatrist girlfriend wondering about me and if everything is OK upstairs, I carry on--this is a good thing. 

This is a pretty tough build, but if you are up for a challenge, you can get all the schems, files, gerbers, PDFs etc., from my github (here) as well as my sponsor PCBway's project page, here.  

Let me know if you come up with anything fun.

Here, now, I am happy. Rock on!!!

You will also need my Minimalist 328P AVR board--here: https://github.com/cslammy/AVR328-minimal

Project for AVR board: https://www.pcbway.com/project/shareproject/Minimal_Atmel_328P_MCU.html

Note: this is a complex build

not recommended for anyone undertaking SMD fab for the first time.

If you want to build this, comment on the blog post, I can reach out to you there or via some other means.

Have fun!