tyco octavia

[numpty]

I built up one of these form a circuit board from general guitar gadgets, although it was only a partial success. the heavily disported sound compresses a lot and then increases in volume when it should tail off, I'm not sure whether its because i used germanium oa90 diodes because the one specified were unavailable.
Also i would like to achieve a more pronounced octave effect which occurs at a higher register similar to a sort of "purple haze" solo tone
and tame the distortion.
Could anybody suggest any mods and ( or) tell me if indeed the diodes i used are unsuitable.

David

[Mark Hammer]

Generally, whenever the core process of combining the same half-cycles of inverted and noninverted versions of a signal are used to mimic doubling the input frequency - whether it involves transformers, op-amps, transformers, silicon or Ge diodes - the level matching between the two signals to be combined is important to achieving a robust octave sound.  If you want to produce two "bumps" on the oscilliscope where there used to be one, the bumps have to be very close to the same height for it to sound like an octave up.

As such, the matching of forward voltage of diodes can often be an issue.  This is certainly not the ONLY aspect to contend with in the Octavio, since that pitiful little Mouser transformer is not exactly wound to Mil or some 1% spec on each of the two sides of the secondary coil, is it?  However, where you can't really easily change the transformer, and probably shouldn't buy up dozens to find one that is tightly matched, you CAN pick through your Ge diode holdings to find a pair that produce equal signal levels for combining.

Again, because the transformer itself is not wound to a very tight tolerance, going nuts about matching diodes may NOT necessarily yield a better octave sound.  What may be better to do might be to measure the signal level coming off the transformer before the diode, and select diodes based on the level matching achieved post-diode using those diode specs.

So, say you have a dozen 1N60s or whatever, and their forward voltages range between 183mv and 228mv.  And say you find that with a constant input signal the transformer output is about 23mv greater on one side than the other.  Since the series diodes would always subtract the amount of signal indicated by their forward voltage, what you would then do, in principle, is to select the diode pair from your dozen that came closest to closing that 23mv gap between the complementary transformer outputs.  Whether that resulted in an overall higher or lower output from the pedal, the two signals to be combined would be mirror images of each other.

Lets say 4 diodes picked from that pile have voltage drops of 185, 203, 206 and 228mv.  The pair that comes closest to a 23mv difference is 206 and 228mv.  The one measuring 223mv goes on the output of the "hotter" transformer side, and the one measuring 206 goes on the cooler side.  In principal, the output levels on the "stripe" side of the diodes should be equal once you factor in transformer and didoe differences.

In theory, that should get you closer to better octaving.  I have my own little Tycho board and 022 transformer perched on it, but lack the time at the moment to follow up.  I'd be curious to hear from people if what sounds good in theory actually works out as well in practice.

[petemoore]

  I thought about that, not in those exact 'Hammerterms' of course...
  Just a socket for diodes, and alot of messing about with such as these.
  But also wondered if there couldn't be a way to add a pot or fixed resistor somewhere in the circuit to fine tune the octave 'sound', say near the splitter output, to 'even out' the signal swings and halves .. by ear or scope.
  ...so..just a thought.

[Primus]

I built the same GGG pedal and although it works it sounds horrible unless I roll the volume on my guitar all most all the way down. I have heard that the octavia always sounds better with the volume rolled down, but this is more than I expected. Anyone have any comments on how this could be fixed? I want to stress that I think this is intrinsic to the design of the circuit, not the quality of the project which like all GGG projects I think is phenomenal.

[LOTUS]

well i've just built-one and it's  starting to sound good. Try different transistor combos.

[petemoore]

  Transistor types do seem to make a difference, I used a couple GE.
  an input pre-gain knob, wired exactly like a guitar volume knob will do about the same/same thing as a Guitvol control.

[numpty]

would adding a preset pot to the higher output diode and adjusting it to equalize both sides be the solution then, or is it not that simple?

[Mark Hammer]

That might work, and it is certainly easy to try out.  You can certainly measure the AC signal level on the volume-control side of each of those two diodes and determine which one needs "rehabilitation" with a trimpot.  It worked for JC Maillet with his adaptation of the Green Ringer.

[DDD]

Since the most of the simple octave-up devices are actually a "set of non-linearities", one can achieve decent octave effect just in a very few level  points and frequences. At the same time it's not necessary to get an "ideal" octave-up functioning. The resulting sound of a "good" octaver is quite "poor", non-harmonic and "empty".
In opposite, if your device varies harmonic contents of a signal with level\frequency, you may find the great tone.

[SeanCostello]

I built up one of these form a circuit board from general guitar gadgets, although it was only a partial success. the heavily disported sound compresses a lot and then increases in volume when it should tail off, I'm not sure whether its because i used germanium oa90 diodes because the one specified were unavailable.

That's what they sound like! The Fulltone Octafuzz I reverse engineered back in 1996 behaves EXACTLY like this. Kinda like an amplifier with hideous amounts of sag. A few possible causes of this effect:

- Since the full-wave rectifier feeds the output directly, there will be a DC offset to the output signal. For certain amplitudes, the Octavia might be driving the input of your amplifier into cutoff.

- The little Mouser transformer in the Octavia might be driven into saturation.

Are you using humbucker pickups? I have noticed this a lot more with humbuckers than with single coil.

Anyway, if you want to "fix" the "flaws" of the Octavia, try using lower gain transistors in the circuit, and slap a DC blocking capacitor on the output of the pedal (0.1 uF to 1.0 uF should work, without changing the tone too much).

Sean Costello

[Toney]

There was a recent post by Aron updating the schematic. The new verified version should be in the schem section. It would pay to run through this and compare it to the one listed at GGG.

[LOTUS]

  I think the Aron & G.G.G schems are the same . U will have less compression if u raise the voltage.    Eric