Corona challenge

[Fancy Lime]

Hey everyone,

are we going to utilize this mess right now fr something worthwhile (while also taking our mind off it), and have a little design challenge, or what? I can't build right now but my brain still sort of works and I have a KiCad. So I'm game for collaboratively designing something crazy. Any suggestions?

Cheers,
Andy

[Kipper4]

I'm up for a project. I can't build much right now too but I'm in.

[swamphorn]

I'm down as long as it fits in a 1590B! 

[Ripthorn]

I've got three new boards coming from fab, including my first two smd boards. I've also got one idea about to go on the breadboard. Kind of my own design, kind of not. Just think "legit amp in a box"...

[iainpunk]

id love to see if we can collaboratively come up with a new type of octave fuzz, that provides that octave up without an already existing schematic

edit, are there any wave folder-like octave up schematics out there? something like pic related?

[Fancy Lime]

id love to see if we can collaboratively come up with a new type of octave fuzz, that provides that octave up without an already existing schematic

edit, are there any wave folder-like octave up schematics out there? something like pic related?

Funny you should mention that... I have been mentally toying with a concept of a hex-inverter octave fuzz based on halfwave folding. I only have a few minutes right now and did not draw a schematic yet but here is the description:

Inverter stage 1: Just a bit of boost, maybe 2x to 10x. After that, the signal is split in 2 paths.
Path 1 goes into Inv stage 2, which is a halfwave rectifier with a gain of 2. To get the rectification, place one 1N5817 in the feedback path and another one after the stage in the signal path. Both diodes need to point in the same direction (either left or right). The output of that stage goes via a resistor (10k) into a B100k mixing pot.
Path 2 goes straight into a 10k resistor and to the other end of the mixing pot.
The wiper of the mixing pot goes into the input of a 2-inverter fuzz, like the Tube Sound Fuzz or Red Llama.

Pretty simple all in all. Basically an octave version of the Tube Sound Fuzz. The mixing pot blends between (almost pure) non-octave signal and a half wave rectified signal of opposing phase and twice the amplitude. A 50/50 combination of both gives a full-wave rectified signal.

If someone wants to try that, please go ahead. I probably won't be able to draw a schematic before Tuesday, though.

Andy

[iainpunk]

i like your thinking, but im not a big fan of hex inverters, im going to use an SRPP jfet stage, into some sort of wavefolder like opamp stage, and into another jfet stage into clipping diodes, im going to draw the schematic in falstad circuits in a jiffy, but first coffee

[iainpunk]

i had this thing in mind, maybe with a cool tone1/tone2 switch, just like a superfuzz

[bushidov]

Anyone find it ironic that this corona challenge was started with a twist of Fancy Lime?

[Fancy Lime]

Anyone find it ironic that this corona challenge was started with a twist of Fancy Lime?

Hey, someone did get it!

What I suggested to do with CMOS inverters could of course be done just as well with opamps. It would require more parts, though.

Andy

[Transmogrifox]

Lime and Corona...an irresistable pun.

On octave up -- here's an idea to throw into the mix.  Maybe implement an x^2 function using an OTA to get the octave up harmonics?

[EBK]

What about an effect that slowly mutates over time, like with a very low frequency oscillator?  Just throwing that out there.  I'm in a weird mood.  We all are to some extent, I suppose.

[Ice-9]

Count me in for the task of building something for while we are all having to sit at home.  I have started a little idea to build a Copicat 3 head tape delay in a 1590B. See how this turns out I have some of it breadboarded for now.

[amptramp]

What about a pedal that adds in audience noise and cheers?  Because stadium rock isn't coming back in the foreseeable future and it would be nostalgic.

[Fancy Lime]

What about a pedal that adds in audience noise and cheers?  Because stadium rock isn't coming back in the foreseeable future and it would be nostalgic.

+1 from me.

For similar reasons: how about a Temple of Kukulkan or Temple of the Feathered Serpent delay. Should not be too difficult with a few PT2399's, right?

Cheers,
Andy

[swamphorn]

i like your thinking, but im not a big fan of hex inverters, im going to use an SRPP jfet stage, into some sort of wavefolder like opamp stage, and into another jfet stage into clipping diodes, im going to draw the schematic in falstad circuits in a jiffy, but first coffee

I have an octave fuzz for the CD4007 already breadboarded. I'll edit this post with the schematic when I find it.

... Maybe implement an x^2 function using an OTA to get the octave up harmonics?

That would make a nice, smooth octave effect. One thing to keep in mind is that OTAs (or at least the LM13700) are noticeably noisy; I would recommend including a noise gate in the design. Perhaps we can use one section to square the signal and overdrive the second section as well as use it to gate the signal?

[snk]

Interesting challenge !
I like CMOS inverters, as well as wavefolders (and octave down... maybe using CD4013s ?)

[ElectricDruid]

... Maybe implement an x^2 function using an OTA to get the octave up harmonics?

That would make a nice, smooth octave effect. One thing to keep in mind is that OTAs (or at least the LM13700) are noticeably noisy; I would recommend including a noise gate in the design. Perhaps we can use one section to square the signal and overdrive the second section as well as use it to gate the signal?

One way of doing X^2 is to feed a ring modulator with the same thing to both inputs; X*X = X^2. For a ring mod, the 13700 is a possibility, but there are plenty of other options out there now too. I've been looking at some of the Alfa CEM-clone chips for pedals. The AS3360 is a decent VCA, or you could use a SSM2164 clone like the V2164 or AS2164. These are exponential VCAs, but can be linearised and are a lot quieter than the 13700 will ever be. Just because it's a guitar pedal it doesn't *have* to be lo-tech, right?

The ring-mod approach makes one thing clear: it *won't* be a nice smooth octave up effect unless you feed it a pure sine wave. Those pesky guitars and their pesky harmonics are going to play hell with any X^2 function however it is generated. That's just physics and there's no getting away from it. But we're only interested in a musically useful result, not some abstract technical perfection, so that's no reason not to give it a try.

[swamphorn]

...

One way of doing X^2 is to feed a ring modulator with the same thing to both inputs; X*X = X^2. For a ring mod, the 13700 is a possibility, but there are plenty of other options out there now too. I've been looking at some of the Alfa CEM-clone chips for pedals. The AS3360 is a decent VCA, or you could use a SSM2164 clone like the V2164 or AS2164. These are exponential VCAs, but can be linearised and are a lot quieter than the 13700 will ever be. Just because it's a guitar pedal it doesn't *have* to be lo-tech, right?

The ring-mod approach makes one thing clear: it *won't* be a nice smooth octave up effect unless you feed it a pure sine wave. Those pesky guitars and their pesky harmonics are going to play hell with any X^2 function however it is generated. That's just physics and there's no getting away from it. But we're only interested in a musically useful result, not some abstract technical perfection, so that's no reason not to give it a try.

Well, smooth is relative but it would be smoother than a full-wave rectifier. Squaring a harmonic series results in a harmonic series with the fundamental cancelled out (as well as DC). As for using a VCA, I worry about availability; I would prefer the LM13700 because it's available from Tayda but I have to admit it performs rather poorly. Not to mention it's several times cheaper.

[amptramp]

One of the National Semiconductor application manuals had a couple of log-antilog generators in AN-30 that can create the original signal raised to any exponent you wish.  There was one version that was very accurate but had limited frequency response and another that was less accurate but had audio response.  You could signals squared, cubed or square rooted and any exponent in between.  Want a tube sound?  Use 1.5 as the exponent.  FET sound?  That would be squared.  These manuals are now online at this location on page 92 but it may also be on diystompboxes:

https://archive.org/details/NationalSemiconductorLinearApplicationsHandbook1994/page/n91/mode/2up

Of course as ElectricDruid says, intermodulation of all the input frequencies could give very non-musical output frequencies but if you had a single-frequency source like a theremin, this could be an interesting experiment.