Curds and whey distortion added to a better overdrive?

[jwin615]

That is awesome. I kind of want to take that circuit and throw it in a stompbox. I’m assuming I’d need some sort of charge pump or converter to get that +15 -15 v

Here’s the build docs, under Wavefolder:

Synth DIY

I don't see any reason this couldn't be run at +/-9v as long as you have enough signal exiting U1C

 

[neiltheseal]

Here is a video of my semi successful attempt.

I managed to get it going by breaking the connection between the gain knob and c13 and adding a mosfet boost here.

I think for next time I will make the following changes:

• Use the spare op amp for extra gain stage between gain and c13
• Use a 1m gain pot (only had a 500k)
• Use a better mix pot as the Timmy is way quieter than the prunes
• Remove or use a better tone pot as it only sounds good with full treble

 

[neiltheseal]

And another video with an envelope filter in front. I really like this sound.

 

[Guilherme Collateral]

Now that's great info! Man I love this forum

For what it's worth, the wavefolder section of the Prunes & Custard is adapted from a design by Juergen Haible (alt link). Digging into that might come up with people who have done similar stuff based on it as well.

 

[finebyfine]

I played around with the wave folder section in LTSpice tonight. Instead of working with different gain stages I just varied the amplitude of the sine wave I was sending through the folder. Some random notes, using the pedalpcb curds and whey refdes, take everything with a grain of salt:

• Using lower Vf diodes reduces the amount of amplification needed to get successive folds but with more amplification into those same lower Vf diodes, the harmonic content of the output is reduced. This makes me curious about what using a compressor before it would sound like.
• Even harmonics can be exaggerated by mismatching D1 and D5. These wavefolders are more or less frequency triplers and don't generate a lot of even harmonics. Mismatching lower section diodes has less impact.
• Increasing the unclipped summing resistor R9 reduces the effect of the wavefolder and emphasizes the fundamental. Not much wiggle room here though, compared to stock 56K, 65K-80K basically nulls out the effect.
• Mismatching R14 and R15 has a similar effect
• (I believe both of the two above impact the gain of the summing and difference amplifier?)
• Adding low pass filters in the folding stages leads to some REALLY interesting phase shift and cancellation. They also stop it from generating odd harmonics into oblivion and I imagine would temper some metallic clang. Without these in some configurations the first few odd harmonics can be louder than the fundamental and the amount of additional audible odd harmonics is practically too many to count -- just a couple of caps to ground basically limit this to just the first 3-4 harmonics

 

[xefned]

Oh wow. I would love to hear an almost-all-even-order-harmonic version of the Curds! To the breadboard!

 

[Bricksnbeatles]

We need a wavefolder factory in the style of the muffin factory based on Alex’s findings!

 

[Feral Feline]

Neiltheseal,
Loved the 36sec mark in the first vid and the second vid with filter was a howling good time.

 

[finebyfine]

If it hasn’t been brought up in this thread, it’s worth mentioning that the Escobedo Tripple Fuzz is based on another synth wavefolder and frequency tripler design by Lockhart. The ZVex Machine is basically two stages of SHOs into tripplers. https://docs.pedalpcb.com/project/Gadget.pdf

Ken Stone's Modular Synthesizer

Descriptions of the various modules in a home built music synthesizer.

synthpanel.com

There’s some cool analysis showing how similar the Lockhart tripler is to the Serge based one that appears in the curds and whey: https://acris.aalto.fi/ws/portalfiles/portal/16820117/esqueda_et_al_applsci_07_01328_v2_1.pdf