Octave fuzz without the octave part unfortunately

Started by cherler, December 14, 2015, 06:16:34 PM

Previous topic - Next topic

cherler

Hey guys! I'm trying to build an octave fuzz, as the title may suggest, but I'm not really getting the octave sound I was looking for. The circuit doesn't sound bad actually, it's kind of a warm fuzz/distortion type of thing. Unfortunately, that wasn't what I was going for. I started from a schematic I found for the Ace Tone Fuzz Master FM2. Anybody have an idea where I went wrong? I've double checked a few times, and what's in the diagram below is what's in my breadboard.


Mark Hammer

Can one simply substitute JFETs for bipolars in such an application?

cherler

Hopefully we'll find out! Yay learning! On that note, I will give this a shot with some regular BJT's tomorrow. I just thought it might be neat to try the JFET's. I was getting an octave sound with one configuration but I didn't like the over all tone, change something and I can't get it back now so I have a feeling it is possible.

Cozybuilder

Quote from: cherler on December 14, 2015, 06:16:34 PM
Hey guys! I'm trying to build an octave fuzz, as the title may suggest, but I'm not really getting the octave sound I was looking for. The circuit doesn't sound bad actually, it's kind of a warm fuzz/distortion type of thing. Unfortunately, that wasn't what I was going for. I started from a schematic I found for the Ace Tone Fuzz Master FM2. Anybody have an idea where I went wrong? I've double checked a few times, and what's in the diagram below is what's in my breadboard.



Have you tried using a couple of diodes? Maybe try 1N4148 between the 1K and 100nF cap feeding each BJT base (Anode to R, cathode to C). I haven't breaded this, just an idea.
Some people drink from the fountain of knowledge, others just gargle.

cherler

Totally worth a shot! After work tomorrow I'm gonna give this a shot with BJT's everywhere and I'll try throwing some diodes into this circuit as well and see what happens!

duck_arse

just before you go silly w/ bipolars, you have NO gate bias on those fets. none at all. add the resistor gate to ground, they will thank you for it.

then you can go haywire with the tunings.
I feel sick.

PRR

> you have NO gate bias on those fets. none at all.

+1

Also I bet the 2nd stage wants to be biased up above zero volt level. AND either wants P-type JFETs (costly) or should be inverted.

I'm not sure why you even want JFETs in the middle. The doubler stage "needs" BJTs to get decent sensitivity and significant doubling. The driver and phase splitter could be either way, but no real difference. And there is a known-good plan for BJTs here.

The only place I would much with a JFET is the first stage.

Why does your drawing lack arrows on emitters and gates? Makes proper/expected polarity totally dubious.
  • SUPPORTER

cherler

So I totally forgot to put arrows on my part diagrams, oops... That will be fixed.

I guess the reason I decided to go with the FET's here was just to see if I could make it work. Probably kinda dumb but it seemed fun to try and figure out. So for clarity's sake here, when you say second stage are you talking about the phase splitter or the gain stages on the two split lines after that?

cherler

#8
Ok so update! I replaced the phase splitter with another 2N5088 and I added all of the bias resistors to the JFETs. This got the octave going and I'm pretty happy with it! I tried with and without the extra gain stage/s after the phase splitter and decided to leave them in. Things are a little more gross and a little more weird with them there which is fun.

Quote from: CozybuilderHave you tried using a couple of diodes? Maybe try 1N4148 between the 1K and 100nF cap feeding each BJT base (Anode to R, cathode to C). I haven't breaded this, just an idea.

I tried adding some 1N914's like you mentioned and I just got that farty sound you get when transistors aren't biased properly or you forgot the input caps (sounds I'm relatively familiar with). I guess the signal isn't hot enough at that point to get over the voltage drop on the diode?

Thanks for the help though everybody!

antonis

Are you sure for the apropriate connection of diodes..??

Maybe a drawing would help..

Alternatively, you may ommit 1k resistors (or you may not) and connect diodes from base to GND or base to +9V (depending on your choice for positive or negative full-wave rectification
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

duck_arse

Quote from: cherler on December 15, 2015, 07:58:49 PM
I guess the reason I decided to go with the FET's here was just to see if I could make it work. Probably kinda dumb but it seemed fun to try and figure out.

excellent! and exactly so. the main reason for half the stuff that gets done around here.
I feel sick.

antonis

Quote from: duck_arse on December 16, 2015, 08:48:28 AM
the main reason for half the stuff that gets done around here.
If you suggest him to get rid of the 3th & 4th transistors (and the respective circuitry) I have to admit that I totally agree with you, my down under friend.. :icon_wink:
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

duck_arse

get rid of the .....

my dumb idea was to lose the phase splitter, run the Q3 Q4 in series, take their outputs as the inversions. but then I thought "all that gain" and decided not to mention any of it. and you could lose Q2 and Q3 and run Q1 and 4 in series, same, less gain.

or renumber Q2 to Q4, then "get rid of" Q3 and Q4. done.
I feel sick.

antonis

And maybe a replacement of 2N5088 with a pair of pnp, like 2N5087..??
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

cherler

Quote from: duck_arse on December 16, 2015, 09:13:12 AM
get rid of the .....

my dumb idea was to lose the phase splitter, run the Q3 Q4 in series, take their outputs as the inversions. but then I thought "all that gain" and decided not to mention any of it. and you could lose Q2 and Q3 and run Q1 and 4 in series, same, less gain.

or renumber Q2 to Q4, then "get rid of" Q3 and Q4. done.

That's actually an intriguing idea, the two signals are definitely out of phase at that point but would there be problems if Q4 started clipping and the waveforms were different? There would be a lot of gain at that point with three stages in series. Sounds fun to try!

Quote from: antonis on December 16, 2015, 09:20:01 AM
And maybe a replacement of 2N5088 with a pair of pnp, like 2N5087..??

Are PNP's usually used here? What advantage would we get? Either way I'll probably try this tonight too.

antonis

#15
@ cherler: I was thinking of some kind of pnp positive full wave rectification, like this:

"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

duck_arse

cherier - I honestly have no idea what one clean one distorted would sound like, but if I had a spare couple of centimeters of BB, I'd give it a go.
I feel sick.

cherler

Quote from: antonis on December 17, 2015, 07:05:22 AM
@ cherler: I was thinking of some kind of pnp positive full wave rectifiacation, like this:



This looks like a cool idea! I'll bread this up as soon as I get the chance!

Quote from: duck_arse on December 17, 2015, 09:15:08 AM
cherier - I honestly have no idea what one clean one distorted would sound like, but if I had a spare couple of centimeters of BB, I'd give it a go.

It definitely seems worth the try. Having different levels to each half of the double may cause problems, though right? I think with some attenuation on one stage to match some clippers on the later stage could get them pretty close, and then maybe something cool will happen!