"New" Tycobrahe Octavia schematic

[aron]

Confirmed correct:

http://www.diystompboxes.com/smfforum/index.php?topic=39465.from1133295454;topicseen#msg281492

Biasing of transistors are different from our old schematics and capacitor values are different.

[LOTUS]

Hi Aron , can u please repost this link. It  seems to go back to the forum homepage.    tks     Eric

[Eirik]

You'll find it here: http://www.diystompboxes.com/pedals/schems/Octavia.pdf

Note that Q1 is wrong. It should have the emitter "pointing upwards". I know because I spent several hours debugging last night.... 
But now it works great!!!

[A.S.P.]

http://www.elixant.com/~stompbox/smfforum/index.php?topic=39465.0?

[LOTUS]

   Thanks very-much!   I am   breadboarding one today , to see how it sounds.    Eric

[vanessa]

It would be interesting to look under the hood of the F*lltone O*tafuzz. On the 'click here' info page he states:

"IDENTICAL in every facet including layout, size, brand of transistors, Transformer look and specs, etc etc.....identical to the legendary Tycobrahe Octavia"

I wonder if his is based off this or based off the other schematics floating around the web.

[Mr.Huge]

You'll find it here: http://www.diystompboxes.com/pedals/schems/Octavia.pdf

Note that Q1 is wrong. It should have the emitter "pointing upwards". I know because I spent several hours debugging last night.... 
But now it works great!!!

I've traced the 3 original Tycobrahe Octavia units I have here and every time it comes up exactly like the schematic I drew. Q1 is totally correct! The emitter goes to the base of Q2 and a 22k resistor to ground. Just incase you didn't look, but this unit runs on negative voltage. That is the positive of the battery goes to ground. Now if you don't use the right transistors (MPS6519) or mix up emitter and collector I can't say what will happen. But my schematic is absolutely, emphatically , utterly, definitely 100% correct. I've gone over it so many times it makes me wanna puke!

-Mr. Huge

[goosonique]

puke fuzz 

tell me about getting trannies reversed .... especially when you are so eager to try out diff types !!
Mr Huge i second that....its Cooorrreecct

[R.G.]

I have some experience with this circuit as well.

The difference between this circuit and the one I drew up back in ... lets see... 1996? 1997? is that there appears to be a 100uF cap in the gain determining path instead of a 25uF cap, and Q1 has its emitter/collector reversed from this one. The other values seem to be the same.

At the risk of muddying the waters, I too went over and over the single unit I had, and the schemo I drew accurately reflects what I found. I think. It's been almost a decade.

It is possible that the circuit will work with Q1 in either orientation shown. If it does, then in the current proposal, Q1 is running in inverted mode, and does so with much lower gain that it would if it were reversed. The only problem you get into is that the reverse biased emitter/base junction that is acting as a collector will only withstand about 6-8Vdc before it goes into voltage breakdown, as silicon bipolars are doped asymmetrically - the base-emitter junction is made differently from the base-collector. Usually it's epitaxially grown on top of the previously diffused base collector so it can be quite different.

This is pretty simple. For an NPN to work as an amplifier, the emitter must be the most negative terminal, the collector the most positive. In this circuit, the point connected to Q2's base is acting like a collector, because that's what Mother's laws say must be happening. Quantum physics didn't change for this circuit.

Bipolar transistors do work in inverted mode - that is, "collector" negative and acting as an emitter, and "emitter" positive and acting as a collector. Setting them up like this sacrifices the delicate precision that was put into the nominal base-emitter junction and forces the base-collector junction to act as a base-emitter. The collector-base junction was optimized for very different operation, so it does a poorer job if used as a base-emitter.

It is also possible that I got an earlier specimen than Jorge, and they may both be correct, in the manner of blind men perceiving elephants. But there's a Tycho Octavia out there that matches the schemo we have had for these many years, and clones have been built that work just fine that way.

[Doug_H]

I'm pretty happy with the sound of my tyco as is. I suspect the 100u vs 22u will not make an appreciable difference. I don't trust the "backwards" wiring of Q1 (sorry Jeorge... ).  I don't know about the ggg version, I built Justin Philpot's version. Overall, I don't think we're looking at a major difference in sound here. Someone please prove me wrong.

Doug

[edad]

I just finished the ggg version and it sounds great, I cant vision other versions being that much different,but to me it begs the question,When laying out to a schematic,for example the 100uf cap vs the 22uf cap Is there as great a difference in sound or gain in these 2 caps,I guess i mean if I didnt have one or the other and made a substitue to a 47uf say would it make any difference in the mojo?? Where in a layout do you really have to watch your values close,just  wondering??

[thermionix]

Ancient thread resurrection.  But I *think* Mr. Tripps had it right.  It depends on what transistor was used for Q1 in the original Tycobrahe (and new Chicago Iron) units.  If it was an MPSA18 like all the schematics I see, with an EBC pinout (looking at the flat side, pins down), then Jeorge was right, and many schematics found online have Q1 backwards.  Seems it works either way from what people report.  Here's what I'm going by, found elsewhere:

Original 70's Tycobrahe:





Somebody drew this up at the other forum based on the pics above:



Chicago Iron version, Q1 orientation same as 70's original:



Every other gut shot I've seen of 70's units or Chicago Iron reissues are the same.  BUT, it could be that Q1 is not MPSA18, but rather something with a CBE pinout.  Anybody have any further info to share about that?

For reference, schematic drawn by Gottfried Divos, which agrees with Jeorge Tripps' schematic, but has Q1 reversed from the GGG, Geofex, and Fuzz Central schematics:

[thermionix]

Mystery solved.  Just finished a phone+email session with my buddy who has one of the new Chicago Iron-made units, which sounds killer and is a direct copy of the 70's Tycobrahe.  The transistors are a Fairchild MPSA18 for Q1, and two Central MPS6519s for Q2 and Q3.  So the Jeorge Tripps and Gottfried Divos schematics are correct.  For the record, it has the TM022 transformer, and yes the primary is on the diode side (I've seen some confusion about that, though not much).

For my upcoming build I have 2N6519s instead of MPS6519s.  The 2N6519s are rated for 300V, while the MPS6519s are rated for 25V.  Which one is most "vintage correct" is unknown to me at this point, but most sources online specify 2N6519.  I expect that at 9V operation they are equivalent, but I haven't pored over the datasheets.  I don't know if the circuit is picky about hFEs, but I will breadboard first.

[PRR]

The amplifier topology with Q1 emitter "UP" was very well known and much used in the period.

If these pedals were built with Q1 emitter "down", then it is another of those little "mistakes" that got perpetuated.

But I would expect, if Q1 were worked in inverted mode, the input impedance would be very low. Could that maybe be why the inverted mode persisted?

[Rob Strand]

If these pedals were built with Q1 emitter "down", then it is another of those little "mistakes" that got perpetuated.

The problem with a lot of these pedals is trying to battle against:
- weird circuit configurations  (some even looking like bugs to a professional designer)
- errors on original schematics
- production changes over time
- commercial clones with tweaks
- undocumented DIY modifications to original schematics
- bad tracings
- badly redrawn schematics
- few original samples

You mix all that together and sometimes it's hard to know what is what.  No wonder it takes 10 internet years before some keen/hard-core dudes finally get to the bottom of it.   I've poured over thus stuff a lot and I'm sure it would work out to 100's of hours of work.   

[italianguy63]

I build these, and use Q1 (MPSA18) with emitter up... and they seem to be just awesome.

They are very wooley with a lot of gain.

Next one I build I will socket Q1 and reverse it and see what happens...

MC

[thermionix]

I've traced the 3 original Tycobrahe Octavia units I have here and every time it comes up exactly like the schematic I drew. Q1 is totally correct! The emitter goes to the base of Q2 and a 22k resistor to ground.

I just noticed this, he is talking about the 2nd transistor, which is labeled Q1 on his schematic, but is the first PNP, most of us would call that Q2.  He calls the 2nd PNP Q2 and the first transistor Q3, for whatever reason.  I have his schematic in .pdf form, so can't attach it here.  But he still has the 1st transistor "emitter down" like the Gottfried Divos schematic I posted above.

I have it breadboarded this way, and it works.  It's a little fuzzier than I think it should be, with randomly selected transistors.  I think I need different hFEs.  I didn't measure because I had no target in mind.  Can't find much info online about that, but most people build it "Q1 emitter up" anyway it seems.  I'll try it that way too, see what it sounds like.  For the time being, I sent an email to my buddy with the Chicago Iron version, asked him to measure collector voltages for me.  His sounds "ideal" to me, so that's what I'm shooting for.  100% sure it is wired "Q1 emitter down" though that doesn't mean it can't sound just as good the other way.

[thermionix]

Just a little update.  Yes the circuit works with Q1 oriented either way.  That was the first thing I tried in my ongoing BB experiments.  I can see how some report better sound that way, I think I got a slightly stronger octave effect, but I want to dial it in with the "original" configuration.  I tried lower gain Q1 subs, sounded worse.  What has helped is higher-gain PNPs (Q2 Q3) than the ones I started with.

Not quite ready to commit what I have on the BB, but if I run it through my Fuzz Face, neck pickup, tone down, man it sounds amazing.  I could hardly put the guitar down last night.  Band of Gypsies and then some!  But I still think maybe the Octavia could sound better by itself, so I'll keep tweaking for a while.

[thermionix]

More data, for the record.  Voltages from my buddy's Chicago Iron/Tycobrahe.  Positive ground.  Q1 (NPN) is a Fairchild MPSA18, Q2 and Q3 (PNP) are Central Semi MPS6519s.

Batt:  -9.19
Q1e:  -2.548
Q1b:  -4.46
Q1c:  -5.11
Q2e:  -1.986
Q2b:  -2.548
Q2c:  -9.19
Q3e:  -1.327
Q3b:  -1.986
Q3c:  -5.78

My voltages on the breadboard and very close to those.  The tones are equally similar, I think I get a wee bit more octave out of mine, so I'm soldering it up tonight as I have it.

[italianguy63]

Just a little update.  Yes the circuit works with Q1 oriented either way.  That was the first thing I tried in my ongoing BB experiments.  I can see how some report better sound that way, I think I got a slightly stronger octave effect, but I want to dial it in with the "original" configuration.  I tried lower gain Q1 subs, sounded worse.  What has helped is higher-gain PNPs (Q2 Q3) than the ones I started with.

Confusing!  So, which way do you think gave more octave effect?  North or South?

I don't think there is any need to use the Octavia with a FF.  I find the Octavia to be quite wooley all by itself.  The ones I build I have a switch to turn the octave on and off.

MC