"Colour" 500-series Harmonics Generator

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Meathands

Well-known member
Joined
Feb 6, 2009
Messages
319
Location
Philadelphia
Link Simpson of Steller Electronics and I have been cooking up this design over the last couple of months. The big idea is to put various types of harmonic distortion in one line-level box with the ability to mix each in with the dry signal to taste. But here's the kicker: the distortion stages will be modular. So, similar to swapping opamps, people can try out different colour combinations. And much like with the DOA world, we hope for this to become an "ecosystem" in which members of the community will develop their own colours. We'll be offering full kits as well as releasing docs for those who prefer to roll their own.

Here are the input and output stages as well as the motherboard PCB and a tentative front panel design. Pretty simple, just getting the signal to and from the distortion stages as cleanly as possible. There is a single input drive, a dry/wet mix for each stage, and output control. Each pot is dual to control on optional slave board which makes the unit stereo.

http://www.diyrecordingequipment.com/wp-content/uploads/2012/mother-input.png
http://www.diyrecordingequipment.com/wp-content/uploads/2012/mother_output.png
mother-pcb-component.png


Link's working on a few distortion stages to get us started, which I'll post next.

-Peterson
 

Attachments

  • front panel.png
    front panel.png
    18.4 KB · Views: 304
Cool idea! I try to follow the schematics to see where the clean signal goes into the mix, but to me it looks like you have to choose between cleen or color with the bypass relay?? Am I missing something? Look forward to see the distortion part(s). Since there is only one drive pot, will the dist boards have a input trim so they will "break up" at ruffly the same drive setting??
 
Such a cool idea.
I'm actually mostly interested in the modular aspect. Sounds like a great place to take the first step to circuit design.
 
Thanks for the comments and words of encouragement so far. I'll have time to do a proper post about the distortion stages we're cooking up, but in the mean time here is a layout for the prototyping module.

proto-daughter-300x200.png
 
What program are you using to draw up the prototype. Im taking autocad this semester, but I need to buy some software and learn some pcb stuff for next semester!

By the way DIYRECORDING GEAR is GREAT!
 
Not to make this too complex but it might be nice to have sweepable HPF and LPF in series with the input to the distortion path so you can isolate selective bandpasses of the wide band signal to distort.. perhaps a sweepable LPF on the distortion output before mixing it back in. to take some of the harshness off the added distortion.

or not...

JR

 
JohnRoberts said:
Not to make this too complex but it might be nice to have sweepable HPF and LPF in series with the input to the distortion path so you can isolate selective bandpasses of the wide band signal to distort.. perhaps a sweepable LPF on the distortion output before mixing it back in. to take some of the harshness off the added distortion.

A great idea. The main bottleneck for features here is the 500-series front panel space. I'd love to have individual drive pots for each channel, too, but you can only fit so much.

What these limitations do--for better or worse--is put more of these tonal decisions in the hands of the designer. Filtering before and after distortion could of course be done within the "colour" modules. Same with setting the range of input drive, etc. This puts the device closer to the "one-trick pony" and further from the "swiss army knife" end of the spectrum, except you get three ponies per unit and they're interchangeable.
 
This could be used for any sort of plug in module not just distortion, correct?  As long as it fits the footprint and specs you could make a bitcrusher, sub octave, etc.  Maybe an MXR Blue Box module?
 
It would be cool to have a SSL talkback type compressor card...something that just smashes the audio, then you can mix some in.

Also the obvious:
2nd harmonic distortion
3rd harmonic distortion.

preferably a way so when you turn on the knob of each respective harmonic you only get that harmonic and not the associated audio too...otherwise you would continually have to compensate for the added volume (well more so at least). How would one go about getting just 2nd/3rd harmonics for an audio signal? Simple? or hard.

To clarify:

Audio ---> filter --> ONLY 2nd harmonic.

Would you put it through a 2nd harmonic generator of sorts, then flip the polarity of the original signal and then mix both signals? compensating for the delay of the filter if there is any?
 
3rd harmonics = basically fuzz (diodes, LEDs or transistor)
- Elliot Sound Products has a bunch of circuits on distortion (even and odd)
- Aphex built several units just for that
- and then there's the old "tape saturator" circuit 

2nd harmonics = how about feeding audio into the CV In of a THAT VCA (!?)

The rest would be an add/substract matrix to isolate distortion only. Not sure this makes much sense though. Usually you'd mix in rather small amounts of distortion, no? But maybe a 50/50 mix of clean and dry signal or even more is useful as an effects box.

And if faceplate space is confined how about using stereo gang pots, I mean, those that have two individual pots in one (tip and ring)? Makes it more costly though. 
 
Here are the first three daughter board circuits: "Tiode," "Tape," and "Octaver"

Daughter Board #1: Asymmetrical "triode" distortion

daughter1-asymmetrical.png


"This clever circuit has been used in a variety of ways to attempt emulate a triode's typical distortion character. The JFET is used to due to its natural similarities to a tube transfer curves. This circuit uses a constant voltage source on the source of the JFET to round out the tube of the top peak of the time domain waveform. By tweaking the TG pot you can control how drastic the rounding effect is. R24 and R23 set up the range for the TG pot work within. The JFET used here is the Toshiba 2SK170, because it is a great audio JFET. The Output is capacitor coupled and ‘buffered’. If output ‘buffer’ has a gain of 2. You can adjust R6 and R5 to vary this gain.

Gain = 1 + R6/R5"

Daugther Board #2: Soft-clipping "Tape" Distortion
daughter2-soft-clipping.png

This circuit is classic tape saturation attempted. The circuit filters off frequencies bellow 3.2kHz with a butterworth tuned Sallen-Key filter. The 3dB frequency is modified with the following formula:

F = 1 /(2*PI*R*C)

*If and only if R= R12 = R11 and C=C35=C36

The clipper circuit is a simple symmetrical clipping circuit found in many distortion pedals. D5 and D6 clamp the input signal above their turn on threshold (1.7V) below that the gain is 2.

Daughter Board #3: Precision Full-Wave Rectifier
daughter3-rectifier.png


This circuit creates an octave like effect. It is in essence a full wave rectifier, except that the voltage drop across the diodes D7 and D8 is compensated for because they are in feedback path of the op-amp.

More circuits to come...
 
college101 said:
What program are you using to draw up the prototype. Im taking autocad this semester, but I need to buy some software and learn some pcb stuff for next semester!

By the way DIYRECORDING GEAR is GREAT!

Thanks! I'm not sure what Link uses, but I use and like Cadsoft Eagle (free version) http://www.cadsoftusa.com/

mjrippe said:
This could be used for any sort of plug in module not just distortion, correct?  As long as it fits the footprint and specs you could make a bitcrusher, sub octave, etc.  Maybe an MXR Blue Box module?

Absolutely.

abechap024 said:
It would be cool to have a SSL talkback type compressor card...something that just smashes the audio, then you can mix some in.

Great minds think alike. Link just sent me a this last night: "Here is a very cool idea....its a single chip daughter board.  It uses a 4301 to act like a 10:1 compressor with a gain of 10. I purposefully slowed the response time of the of RMS section to not respond to transients and cause more artifacts. I used full wave rectification instead of the normal halfwave to catch the negative pulse.  The ratio is controlled by a simple voltage divider. For the gain I injected a DC offset into the full wave rectifier."

abechap024 said:
preferably a way so when you turn on the knob of each respective harmonic you only get that harmonic and not the associated audio too...otherwise you would continually have to compensate for the added volume (well more so at least). How would one go about getting just 2nd/3rd harmonics for an audio signal? Simple? or hard.

This is essential what the drive control does, right? The harder you drive the circuit, the more THD you get. But adding a mix knob gives you another level of control: you can drive it hard and mix it low, drive it soft and mix it high, etc.

Script said:
And if faceplate space is confined how about using stereo gang pots, I mean, those that have two individual pots in one (tip and ring)? Makes it more costly though.

Dual gang pots for what? We are currently using them to control an optional slave board for stereo operation.
 
Meathands said:
This is essential what the drive control does, right? The harder you drive the circuit, the more THD you get. But adding a mix knob gives you another level of control: you can drive it hard and mix it low, drive it soft and mix it high, etc.

Not to criticize, but the drive control changes the level to all three modules, so it is a bit limiting.  Additionally, it seems like a lot of modules would benefit from having a front panel control.  Is there any room to move the blend pots over and have an additional hole for a trim pot at least?
 
mjrippe said:
Script said:
And if faceplate space is confined how about using stereo gang pots, I mean, those that have two individual pots in one (tip and ring)? Makes it more costly though.

Dual gang pots for what? We are currently using them to control an optional slave board for stereo operation.

I guess I was thinking two mono units (not two mono boards in one unit to make it stereo), both with the same controls but overall more possibilities, using dual pots (independent) for whatever is desired (RING = DRY and TIP = WET, RING = LPF and TIP = HPF, etc.). But agreed, the downside is that one would need two units for stereo processing.
 
vey coool idea!! i also think it would be good to add some form of filtering to the unit, makes it much more versatile.
what about having 1 x motherboard unit then adding the distortion modules/filter next to it in the 500 space rack?
this would make it a truly modular system where different distortion modules can be changed without having to remove the motherboard pcb and and changing it in circuit if you know what i mean???
 
Script said:
I guess I was thinking two mono units (not two mono boards in one unit to make it stereo), both with the same controls but overall more possibilities, using dual pots (independent) for whatever is desired (RING = DRY and TIP = WET, RING = LPF and TIP = HPF, etc.). But agreed, the downside is that one would need two units for stereo processing.

I'm not quite sure what you mean. Are you talking about switching jacks?

leadbreath said:
vey coool idea!! i also think it would be good to add some form of filtering to the unit, makes it much more versatile.
what about having 1 x motherboard unit then adding the distortion modules/filter next to it in the 500 space rack?
this would make it a truly modular system where different distortion modules can be changed without having to remove the motherboard pcb and and changing it in circuit if you know what i mean???

As it stands, you'll be able to swap modules without removing the motherboard pcb. I'm not sure what the other advantages would be to splitting it into two slots?
 
Okay, I think the pots I refer to are called "stacked concentric", i.e. two resistances, two shafts (one inside the other: upper shaft of 3/16" [4.7mm] inside lower shaft of 1/4" [6.3mm]), both independently controllable. Consequently, two knobs are required (mechanical side: upper & lower!? User side: tip & ring [or whatever they are called]).

I suppose, those pots are not so often seen on studio gear (e.g., on SPL SX2 for "Output" and "Dynamic Ratio"), but they can save a lot of faceplate space because it's two entirely independent pots in one, said differently: you can turn the "upper/tip" knob to 4 o'clock and the "lower/ring" knob to 9 o'clock, thus mixing DRY and WET or LPF and HPF or 2ND and 3RD or PRE and POST etc to the heart's and ears' content.

 
Back
Top