Octave down schematic.

[R.G.]

Are there any sound samples of Rocktave Divider? E.g. digitalized original "Roctave Divider Demo Tape" would be nice I tried to find some, but did not succeed so far..

The Rocktave is a DIY pedal article published before the internet. It comes from a time before sound samples were common. You'd have to find someone with a Roctave and get them to make a sample.

I'd do it but my Rocktave is packed away in a box... somewhere...

[El Heisenberg]

I have the book. Its a huge schematic though. Never bothered. Maybe cuz i dunno what itll sound like.

[Thomeeque]

Are there any sound samples of Rocktave Divider? E.g. digitalized original "Roctave Divider Demo Tape" would be nice I tried to find some, but did not succeed so far..

The Rocktave is a DIY pedal article published before the internet. It comes from a time before sound samples were common. You'd have to find someone with a Roctave and get them to make a sample.

For Roctave Divider there was demo tape created by Mr. Anderton himself And it seemed to be quite popular project (e.g. it is one of projects on GGG and even Line6 emulates it in higher models of their PODs!)..

[El Heisenberg]

Hey, you have heard this tape?? Where! Id love to hear anderton play!

[doug deeper]

pick up a mandrake memorial record!

[El Heisenberg]

A whos who record?

[Taylor]

Does that iPhone not have Google? 

I didn't realize Anderton was in a band before now. It's kind of goofy stuff, but sort of illuminating in a way. So far, I'm not hearing much in the way of effects, just some volume swells. Now whenever I read his articles, I'll think about him bringing his newest creation to band practice, then having his bandmates tell him, "yeah, sounds cool, but let's try it clean instead". Poor guy.

[Thomeeque]

Hey, you have heard this tape?? Where! Id love to hear anderton play!

Nope, I only know it was made - it's listed here (Partial Discography/Tapes section)

Edit: There is CA's Music subpage with some samples, unfortunately all links there are down You can try to google for those WMA's however (at least one could be maybe downloaded via eDonkey), good luck

T.

[GFR]

Hey, you have heard this tape?? Where! Id love to hear anderton play!

I saw him playing live He was using a midi guitar BTW.

[andromeda]

Hi, here is an abbreviated 2 knob version of the OC-2 schematic I put together for people looking for a simpler, cleanish (ie not a square wave) octave down. From the nature of the effect, it can't really get any simpler than this, boss's fundamental extractor (the bit on the bottom) is quite elegant and works better than most.



Some explanation of how the boss circuit works and whatnot:
http://www.diystompboxes.com/smfforum/index.php?topic=58609.0

Hi
is this project verified?
and if i want add to second stage of octave down,how can i do?
Thanx

[StephenGiles]

I am completely befuddled by this circuit, but I'm sure it's very clever.

[Processaurus]

Hi
is this project verified?
and if i want add to second stage of octave down,how can i do?
Thanx

I haven't built it personally, but a few other people have here.

If you wanted the 2nd octave down take a look at Boss's original schematic for the  OC-2, and substitute half of a CD4013 dual flip flop IC for the rare roland single flip flop chip they use (IC6).  I never had much use for the 2nd octave down, but maybe it's good for something.

I am completely befuddled by this circuit, but I'm sure it's very clever.

I don't understand where the magic comes from either, the part with the two half wave rectified signals being compared to the low pass filtered signal with the two comparators, driving the set/reset on the first flip flop is quite strange.  Maybe it does some kind of hysteresis so little harmonic burbles don't clock the flip flop doing the octave division.  I'm curious if it is a circuit used in signal processing electronics outside of musical instruments, or if it is an invention of Boss's.

[StephenGiles]

Still, if it works don't break it - or words to that effect!

[merlinb]

I don't understand where the magic comes from either, the part with the two half wave rectified signals being compared to the low pass filtered signal with the two comparators, driving the set/reset on the first flip flop is quite strange.  

If you just have a single comparator driving the flip flop then you always get a tracking problem because the natural second harmonic of the string changes phase compared with the fundamental. You get a little "blip" that rides on the main signal, and when it gets to the right point on the waveform it will be detected by the comparator, causing a false trigger. EDIT: This usually happens about 1 second after you pluck the string, not just during the final decay period.

By using another comparator that triggers on a different part of the waveform you fix this, since the 'blip' can't be in two places at once.
One comparator in the Boss circuit produces a pulse at the positive peak of the audio waveform, and the other at the negative peak. So although you will still get false triggers from each comparator, they almost never happen at the same time, and that's what counts as far as the flip flop is concerned.

Incidentally I discovered all this by trying to design a simplified octave down. Eventually I realised that it was impossible to do without two comparators, so my final circuit eneded up extremely similar to the Boss, much to my chagrin  
(Actually my circuit has slightly better tracking than the Boss since one of my comparators detects the zero crossing rather than the peak, but hey...

[Mark Hammer]

The Rocktave addressed the blip-n-sputter at the end of notes, during that portion where fundamental detection was hit and miss, by using two different time-constants for the compression and expansion, such that the expanded signal decayed before the sputtering began.

There are effectively two paths in the Blue Box that have to be co-ordinated.  The 4013 flip-flop requires that the fundamental signal feeding it be consistently above a given threshold in order to actuate the change in flip-flop state that produces the frequency division.  The signal is brought up to that critical level by the huge gain of stage 1, and the comparator-like action of stage 2. 

At the same time, the Blue Box uses two transistors as gates.  The envelope-follower path from stage 1 feeds the collectors of Q2 and Q3, treating them as voltage-controlled gates, to shut down the output before the signal feeding the 4013 falls below the point of stability and the sputtering begins.  Of course, as anyone who has built one can tell you, the approach adopted by the Blue Box is NOT one size fits all.  Sometimes it is effective at sputter-elimination, and sometimes not.  I keep meanng to tinker with the envelope follower section to see if there are some simple mods that can be readily adapted to a 3-way toggle for a 3-sizes-fits-all arrangement.

In the meantime, the solution that Anderton came up with for the Rocktave, using a compander chip, more reliably tackles the de-sputtering across a wider array of signal inputs, by reducing the differences in those inputs through compression.  The fact that the clean signal continues after the octaves are gated out, makes for a more natural-sounding decay.  The BB, of course, has only distorted and sub-octave, so the abruptness of even well-functioning gating shows through.

The huge gain of the input stage in the BB that feeds the envelope follower sort of achieves some compression by reaching the headroom maximum, but not flawlessly.  The problem is that the dynamic range below the critical threshold for triggering the 4013 is still maintained, even though the dynamics above a certain point are compressed.  The compressor half in the Rocktave does us the service of not only reducing the dynamics of the peaks, but increasing the level of the softer portions as well.  I wonder if some combination of a soft clip in stage 1 (e.g., a diode pair in series with a small-value fixed resistor in the feedback loop) and a compensating change in the collector resistor values of Q2/Q3 might succeed in applying some dynamic range constriction across a broader range of signal amplitudes.

[merlinb]

.

[earthtonesaudio]

I don't understand where the magic comes from either, the part with the two half wave rectified signals being compared to the low pass filtered signal with the two comparators, driving the set/reset on the first flip flop is quite strange. 

If you just have a single comparator driving the flip flop then you always get a tracking problem because the natural second harmonic of the string changes phase compared with the fundamental. You get a little "blip" that rides on the main signal, and when it gets to the right point on the waveform it will be detected by the comparator, causing a false trigger. EDIT: This usually happens about 1 second after you pluck the string, not just during the final decay period.

By using another comparator that triggers on a different part of the waveform you fix this, since the 'blip' can't be in two places at once.
One comparator in the Boss circuit produces a pulse at the positive peak of the audio waveform, and the other at the negative peak.

Incidentally I discovered all this by trying to design a simplified octave down. Eventually I realised that it was impossible to do without two comparators, so my final circuit eneded up extremely similar to the Boss, much to my chagrin 
(Actually my circuit has slightly better tracking since one of my comparators detects the zero crossing rather than the peak, but hey...

The fundamental extractor portion of the OC-2, as well as the MFOS SubCommander, and possibly also Merlinb's circuit (if I'm understanding him correctly), is what I would call a window comparator.  However, where a "textbook" window comp uses two fixed voltage references, here the upper and lower references are the positive and negative signal envelopes, respectively.  This is probably what gives it good tracking, because it has an envelope detector (actually two!) built in.  The signal is compared to its own envelope, and the envelope always lags behind the signal by the time constant set by the parts after the rectifiers, so you nearly always have a definitive difference in the two voltages which helps avoid false triggering.

By using another comparator that triggers on a different part of the waveform you fix this, since the 'blip' can't be in two places at once.

That seems like a really good explanation, thanks!

[Processaurus]

Thank you both, Alex, Merlin.  That sheds some interesting light.  I still might need to see it at work on a gross input signal with harmonics to get a feel for how and when the pair of comparators trigger.  Merlin, did you mean the comparators trigger right at the positive and negative peak of the waveform?

I wonder how fast the sort of DCish envelope voltage charge and decay on pins 2 and 12 of IC2, does it drain out completely during the second half of the input signal frequency's period? 

Also, the precision rectifiers are curiously different than the textbook circuit, here the input signal goes to the non-inverting input, rather than through a resistor to the inverting input.

[merlinb]

Merlin, did you mean the comparators trigger right at the positive and negative peak of the waveform?

Yes. Well sort of. They compare the original audio signal with the smoothed output output of the precision rectifiers, so you get a pulse 'close' to the peak of the audio. Here's a simplified diagram of what I mean.

Also, the precision rectifiers are curiously different than the textbook circuit, here the input signal goes to the non-inverting input, rather than through a resistor to the inverting input.

Not sure what you mean by that. They look like the standard precision rectifier I have seen in the textbooks...

[fuzzo]

Sorry to cut this intersting talking but , no one , made a PCB for the OC2 chopped ? (make it on veroboard would require a lot of bravery )