Is there a simple octave up and octave down in one pedal?

[Ben Lyman]

Good point Rob, thanks! I really don't know, I will try. But do you mean reference to ground after the caps? I did leave off the usual 47K to ground after C10 because it was going to the... well... I don't really know why.
Of course, C10 is coming from the GR mixing buffer and already has a 10k to ground before it.

[robthequiet]

The mixing part throws me a bit but usually you would have at the output a cap to block DC then your output resistor or a volume pot, from most of the circuits I've seen. Not trying to suggest adding separate knobs for each section, but the grounding at your R22 output pot has me scratching my head a bit, because you do have the ground point under your volume knob.. .

The phase thing would require inverting one of the outputs, such as taking the output from Q6 emitter instead of collector, something I would get a second opinion on. Then again, maybe having a phase swap might be an effect unto itself.

Somehow I think the outputs should be pots to ground then your final output would be like a master volume, then you wouldn't have one pot doing the blend. Like on a guitar blend pot, you get a better result if you have stacked pots. Something to explore. Would having an output buffer mess with the tone?

[Ben Lyman]

Yes, I see what you mean Rob. I like the idea of two volume controls for mixing but I don't like the idea of having to add more pots if I can avoid it.
Have a look at the original Shocktave schematic and you can see where I added the GR circuit, I basically just inserted it without changing the original mix pot set up. I'm off to tinker with it now.

[robthequiet]

Yes I can see how you made the GR fit into the Shocktave. Can't see any way to do it differently, really.

You mentioned R9 and C12 in the context of tuning the Shocktave -- I think from Joe's description the tweakable parts might be C6 and C9 that would affect the tuning of the oscillator, but I would think that Joe had it pretty well dialed in, although changing them might have some effect.

I was thinking about phasing in this circuit but you have a half-wave and then a double-wave mixed together so I'm not sure if phase is really the issue here. Just a question of how the two waves come together, if one is partially canceling out for some reason, thus the dip at the center of the blend.

[Ben Lyman]

Cool, thanks again Rob, that's a big help, I didn't know C6 and C9 would be the things for tuning the oscillator. I will look into that tomorrow for sure.

I tried taking the GR from the spot where the Shocktave takes the dry signal but then I immediately realized it is the gain booster and I would have it driving the GR as well, which I already have a booster on. Anyway, I currently have it coming off the input buffer and it is working well, so I can have a separate gain for each octave.

I also took off the mix pot and figured I would try using the gain pots as a means of mixing the signals. At first they didn't blend well at all but I put a couple 100K's in place of the 1M mix pot and it is working out great. I can get straight sub octave or straight Green Ringer or mix the two to taste.

All that remains is fine tuning that sub octave for optimum performance but I think it's very close now.

[duck_arse]

an observation on the Q5//Q6 flip/flop - when one collector is high, the other is low. the collector signals are anti-phase, so if the Q5 C is wrong-phase for the ringer, take the output from Q5 C instead.


hmmmm. I wonder if a pair of diodes coming of those collectors would make a doubler, like in the ringer .....

[duck_arse]

geeze, I just listened to the V2 clip, it sounds like you've got Dario Argento in a box. that first version was a killer.

[robthequiet]

^^ DA saw the phase thing better than I did, taking notes here...

It's true, the pedal has a real saxophone grit to it on the Shocktave. Have you tried running a gentle boost into it to beef up the octave down or does it like low levels?

[Ben Lyman]

Duck, you like the first vid? It has the Fuzz Face in it and I felt like it was just too much. I think it will be more versatile if I can just stack it with an external fuzz pedal when wanted.

My second vid (breadboard) has no Fuzz Face in it. I think it's pretty good like that.

I will try taking output from Q5c, that's easy enough.

Diodes coming from the collectors you say? And going where? You lost me...

Rob, I have tried the Fuzz Face before it and I've tried a SHO clone before it. Didn't work out so well but I have not tried a gentle boost. The completed version on my pedal board sounds pretty good with my Dynacomp in front of it though. That's another reason I want to keep it simple, so I can just add other pedals to it at will. It sounds crazy with a vibrato and comp in front, a delay after. It's almost like some kinda mad Farfisa

[duck_arse]

mad farfisa? I thort hammond in places, but mad, yes.

see how ringer Q4 does phase splitting, with opposite signals on collector and emitter? well, same goes for Q5 and Q6, the emitters are opposite phase. so, on the breadbord, as a kwik test of the theory, you could just connect C7//C8 to the Q5//Q6 instead, and see if anything comes out of Q7.

maybe smoke.

[Ben Lyman]

Ah, yes. Okay, I see what you mean.. partly. I see the GR phase splitter but I fail to grasp the concept of Q5 and Q6 as phase splitters because I can only picture two identical contraptions connected in parallel to Q3b. With the exception that one of them has a cap attached to the collector for the output. Do you see what I mean by "identical contraptions"? Going back to your other idea of taking the output from Q5 or Q6, I don't see how that could make a difference.

[robthequiet]

I think I get it, but a bit hazy on the signal appearing at the base of Q3. Does the Ge diode do any rectifying or is it just there as a biasing aid? I think I hear some guitar signal mixed in with the low octave, but could be imagining things.

So you use the flip flop as the phase splitter, but it's a pulse wave as opposed to two out-of-phase guitar waves. Gradually becoming clear. So you can do away with the first 2/3 of the GR ckt. If you feed this into a second octave divider do you get two octaves down or does it just mush out?

[Ben Lyman]

Maybe the magic lies somewhere within Q3, R9, C12 and the diode? it looks to me like the signal goes straight from C5 through the diode to the oscillator, with the Q3 contraption doing something. Diode in Q3 and Diode acting back to back? 

I would try putting the last part of the GR onto the oscillator emitters but then I won't have that cool "Upper Octave Gain Control" anymore. I like the way it mixes right now, just needs a fuzz pedal after.

For tracking, I tried a 150k trimmer in place of R13 but the best sound was still right about 100k. I don't know why I did that, I guess it was just convenient at the time.

[robthequiet]

Lots of possibilities -- from what I read last night, the voltage coming into the flip flop thingie from the base of Q3 is the "control voltage" that can raise or lower the flip flop frequency. Joe hacked this thing into a half-wave rectifier/pitch-tracking oscillator but I'm wondering if the Ge diode is also doing something magical. Remember how he used it in the Vulcan? I think the diode compression helps to trick the oscillations so they only have time to trigger on every other peak. The timing is set by the R13+C6 and R14+C9 values -- it's a two-cylinder engine -- which is why trimming R13 probably threw the symmetry off and upset the timing. If you move both R13 and R14 to the same value up or down you might find some tuning opportunities, theoretically.

Here's some nerdy links that explain the theory way better than I --

http://www.electronics-tutorials.ws/waveforms/astable.html
http://www.electronicshub.org/voltage-controlled-oscillators-vco/

This is a really cool box, Ben. Keep V1, too.

[Ben Lyman]

Thanks for the links, that is some very interesting reading. It's mostly above my head but I can grasp little bits here and there.

I like the first version too but I think the Fuzz Face was overkill, it just needs a little extra fuzz on top off the already fuzzed out octave circuits.

[duck_arse]

I'm certainly not advocating anyone do any of this guff I mention, it's just fluff. you should be able to recognise these building blocks when they pop up, is all.

I hadda bituva go on the bb today, w/ sig gen, not real world, and only the shocktave section (from V1/2). it seems the D1/Q3 rectifies the gained signal to provide a supply for the cross-coupled flip-flop - and a gating method: no input signal, no rectified DC, no mad flipping//flopping. on the 'scope I saw little diference between 1N4148 and a BAT46, the highest gain transistors you have at Q1 and Q3 (V3 numbers) may provide some benefits, and Q3 as a darlington might track better.

some fun/madness with a led as D1 might be apparent, at least on the scope.

and - messing with the F/F parts, I put in 18nF or so instead of the 100nF caps, and it tripled where it should have singled. I think, in theory, it might be possible to run two F/F's, one tuned high, one tuned low, from the one rectifying transistor, and mix those results.

[Ben Lyman]

Cool Duck, thanks for trying it out. I am still happy with everything except the Shocktave stability, it's good but needs a little fine tuning.

I attached the diodes to the floppy emitters and the results were interesting but I got more normal signal and less octave signals.

Two FF's is probably too many for me to try and tackle but I might try two Electras or something similar

[amptramp]

Getting an octave up with fullwave rectification is relatively easy and the Green Ringer section does the job easily.  Getting an octave down is another matter.  You have to determine what the frequency of the signal is and this may not be simple when the harmonics are higher amplitude than the fundamental.  If you look at the Boss OC-2 which gives you one and two octaves down, they have a steep lowpass filter going into the divider logic to avoid "wolf notes" where the output suddenly goes up an octave.  Unlike the Shocktave, it has separate paths for the signal and divider logic.  The Valve Wizard U-Boat also separates the logic from the signal but it uses different logic from the Boss as shown here:



They both reverse amplifier polarity to give you a signal consisting of the original cycle then an inverted cycle.  The difference is the Boss circuit uses the peak of the signal as the beginning of the cycle whereas the Valve Wizard uses the zero-crossing as the beginning of the cycle.  The Shocktave appears to add a divided signal to the original waveform and cannot use lowpass filtering to achieve stability since there is no distinction between the signal path and selection circuitry.  The Boss circuit should sound a little better than the Valve Wizard but the Valve Wizard is a bit simpler.  Both of them should handle signals with higher second or total harmonic than fundamental amplitude, something the Shocktave would not do, so both should sound better than the Shocktave.

[Ben Lyman]

Slowly but surely, I'm getting closer. Just gonna drop this here for future reference:
http://www.diystompboxes.com/smfforum/index.php?topic=38212.0

[Mark Hammer]

Make the two diodes in the Green Ringer half both Schottky types and you'll be happier.  Trust me, it makes a positive difference.