Roger Mayer Octavia voltages?

[LucifersTrip]

finally got around to breadboarding this (exactly as schematic) ...sounds very good  right off, though, probably not as rich as the transformer versions.



Before I start tweaking, just curious if anyone has any ballpark voltages for this one. I have:

ebc
Q1: 4.35, 3.16, 2.97
Q2: 2.00, 2.97, 5.80
Q3: 2.50, 3.50, 6.40
Q4: 3.45, 4.06, 9.00

thanx

[LucifersTrip]

Update:

I read on another forum that the voltage on Q4's base should be 1/2 the supply voltage. Can anyone confirm that?

Regardless, I lowered the 2.2M pull-up R on Q4's base to 1.8M to get 4.5V. It did increase the octave effect.

I tried different transistors in the different spots. Q1 doesn't effect the sound. With the few darlingtons I tried, Q2, Q3 had little effect, but Q4 had more. I swapped the MPSA13 (hfe ~ 32000) with MPSA14 (hfe ~ 43000). I got a bit smoother sound, which is better when the drive is turned down. When playing chords, it is less dirty (I believe it tracks better).

Hard to decide what to go with...

I still want more octave. I'll try different diodes next.

[Gus]

A MPSA13 is a darlington

So Q4 emitter will be about two Vbe drops lower than 1/2 the power supply voltage with the two 2.2meg bias resistors used.  You also need to know the loading of the meter you used to measure the base voltage.  This is  because of the use of 2.2meg bias resistors.

Look at page two  3.45VDC(emitter)/10K =.345ma    under 25C and under 1ma graph 2nd down on right about 1VDC,  I would guess somewhere around 4.5VDC-1.0VDC =3.5VDC
http://www.fairchildsemi.com/ds/MP/MPSA13.pdf

[Mark Hammer]

Personally I don't think there is much that is usable in the RMO after the first few kilohms of Gain pot resistance.  Using a 10k seems to me to be overkill...or perhaps something to compensate for specific transistors.  One can probably dial in the best sounds with less than that.

Why Q4 would have any impact on the octaving is beyond me, given how it comes after the volume pot.  I'm not saying it doesn't, just that I can't see why it would.  Seems to me one could stick in the BMP gain recovery stage and do just as well.

[LucifersTrip]

Why Q4 would have any impact on the octaving is beyond me, given how it comes after the volume pot.  I'm not saying it doesn't, just that I can't see why it would. 

I'll tell you what's pretty funny. In my darlington trials for Q4, I tried a BC517. I got an insano gated fat octave fuzz almost exactly the same as what my Great Destroyer does....unusable as an Octavia. The same  BC517 in Q3 resulted in silence...

[LucifersTrip]

A MPSA13 is a darlington

So Q4 emitter will be about two Vbe drops lower than 1/2 the power supply voltage with the two 2.2meg bias resistors used.  You also need to know the loading of the meter you used to measure the base voltage.  This is  because of the use of 2.2meg bias resistors.

Look at page two  3.45VDC(emitter)/10K =.345ma    under 25C and under 1ma graph 2nd down on right about 1VDC,  I would guess somewhere around 4.5VDC-1.0VDC =3.5VDC
http://www.fairchildsemi.com/ds/MP/MPSA13.pdf

thanx much...I guess I'm on the right track. I'll continue tweaking components.

[Plexi]

Old thread reborn alert..!

A bit gated, to be honest.
As I know, the important here are Q1 to Q3.

My readings, V in 9.3:
Q1
E=4.4V B=3.0V C=3.0v
Q2
E=2.1V B=2.9V C=5.8
Q3
E=2.7V B=3.2V C=6.4V
Q4
E=3.5V B=2.2V C=9.3

Not good octave at all, even decreasing the drive a bit (I've followed your advices, Mark H.)

[antonis]

Q1 & Q2 seem to be OK

Q3
E=2.7V B=3.2V C=6.4V  O.K.
(although, Base should sit on 3.8V for +9.3V supply, but never mind..)

Q4
E=3.5V B=2.2V C=9.3
(typo or lower/upper divider resistor wrong value..??)

IMHO, for a decent octave effect output, Q4 Base shoud be biased at about 6V..

[Plexi]

Q1 & Q2 seem to be OK

Q3
E=2.7V B=3.2V C=6.4V  O.K.
(although, Base should sit on 3.8V for +9.3V supply, but never mind..)

Q4
E=3.5V B=2.2V C=9.3
(typo or lower/upper divider resistor wrong value..??)

IMHO, for a decent octave effect output, Q4 Base shoud be biased at about 6V..

Thanks.. !
Tonight I'll re-bias that transistors and check..
I was wrong: so Q4 is relevant in the circuit? Is not some kind of buffer.. I guess..

[Gus]

The voltages at Q3 emitter and collector set how much the diodes are turned on. The 2.2meg and 820K are for allowing DC current to the diodes and to get them to start to conduct a little so the AC coupled signal voltage has less of the the diode turn on voltage to overcome.

So you can adjust Q3 bias(this changes the voltage and therefore the current in the 2.2meg and 820K) for more equal current in the diodes and/or adjust the the 2.2meg and 820k resistor values for more equal currents.

Note the collector output resistance (10K in series with the 2.2meg) is different than the emitter output resistance (emitter resistance with the 10K to the 820K) so this does not have a balanced drive to both diodes. But this is a guitar effect and maybe this is OK or what you want.

[antonis]

I was wrong: so Q4 is relevant in the circuit? Is not some kind of buffer.. I guess..

Of course it's a buffer..
(but a misbiased one here..)

An ideal buffer should have its Emitter biased at exactly half of power supply voltage for equal swinging output signal..
A real word one should have its Emitter biased higher than half of power supply due to Emitter and Load resistors voltage dividing effect when sinking current..
(actually, neither its sourcing current capability is ideal but this isn't our point here..)

A full wave rectified signal has to be undistorted (or symmetrically distorted, at least..) to be perceived as octave effect..
(it already suffers enough from its inherent crossover distortion "pathogeny" due to diode(s) forward voltage drop..)

[antonis]

So you can adjust Q3 bias(this changes the voltage and therefore the current in the 2.2meg and 820K) for more equal current in the diodes and/or adjust the the 2.2meg and 820k resistor values for more equal currents.

+1 

(although, 2M2/820k = 2.7 is close enough to V_Collector / V_Emitter for a practical phase splitter..)

[Plexi]

Thanks Gus and Antonis.
I'm sitting right now to check some bias.

- 2M2 resistors to Q4 are fine, I'll check the transistor.
- Both diodes get the same voltage to anode: 0,75v (0,47v to cathode)

I'll check to re-bias Q3 10k resistor to C and see how it goes..

[antonis]

- 2M2 resistors to Q4 are fine, I'll check the transistor.

I should suggest you to make upper 2M2 resistor a 1M5 one..
(MPSA 13 is a Darlington BJT and exhibits about 1.5V V_BE..)

[Plexi]

Tried everything: bias, different diodes, different transistors, swapping caps.. nothing.
This thing is great as a nasty fuzz, but almost useless as octave-fuzz.

Moved to a better and happy place: Foxx Tone Machine, and (to my surprise) Way Huge Purple Platypus.
There's a lot more of octave up, glitch's and neck pickup's fun.