fun octave circuit

[doug deeper]

heres something i came up with, it should work with most npn transistors.
low pass filtering on the input can do interesting things to the fuzziness of the circuit.

[blubman]

So simple, that HAS to be tried out! Could even easily be mounted within a guitar for extra awesomeness!

[earthtonesaudio]

That's neat.  Does it have greater than unity output?

[John Lyons]

Looks neat, have to try that.
I would imagine this would be filed under circuit fragment as it's something
you would want to tack on after a fuzz or with some filtering as eluded to.
Driving this with just a guitar into a clean amp would sound pretty sterile I'd think.

Cool stuff!

john

[petemoore]

  Interesting !
  I don't see any DC bias potential on the actives...
 

[DougH]

Red- 1khz/100mv input

Green- Output

2n2222's and 47n input cap.

This will attenuate a 100Hz signal. But with 470n cap the 100Hz signal will be amplified and "octaved" as well.

Looks like a fun circuit worth trying!

[earthtonesaudio]

Unless I'm way off, is that second transistor bootstrapping the first, AND getting its bias through the bottom 10k resistor?

If so, that's awesome.

[edit] No, no bootstrapping.  Don't know why I thought that.   

[doug deeper]

yeah its a little but of an odd ball.
the second transistor could be a diode as well (ditch the collector basically), but the transistor works just a little better.
its pretty grungy sounding, but does the trick.
its hair above unity gain on my breadboarded version.
the octave is more pronounced with humbuckers (neck pickup of course!)

[petemoore]

  Does disconnecting 9v change the bias orientation ? ie..the difference of voltages inside the DC blocked area between E/B/C ?
  It looks like it would just pull the entire circuit up to it as a reference operating voltage, but the voltage relationships between the Q pins would remain the same.

[doug deeper]

i think you're right, i just recall seeing something set up similar to this somewhere.
i wasnt really expecting much, jut though it was a really interesting simple way to achieve an octave up.
wish i could remember where this idea came from!

[Steben]

I have a hunch that clipping the modulated signal will result in less octave/more fuzz since it can become very symmetric.

[slacker]

i think you're right, i just recall seeing something set up similar to this somewhere.
wish i could remember where this idea came from!

I don't know if it's what you were thinking of but if you make the second transistor a diode it looks a bit like Hemmo's classic the Bazz Fuss.

[jayp5150]

i think you're right, i just recall seeing something set up similar to this somewhere.
wish i could remember where this idea came from!

I don't know if it's what you were thinking of but if you make the second transistor a diode it looks a bit like Hemmo's classic the Bazz Fuss.

Similarities are there, but the output is pulled from the emitter here, and the collector in the Bazz Fuss. I'm interested in hearing this thing, regardless.

Edit: scratch that thought. This would be like pulling the output from a bazz fuss after the diode (between it and the base of Q1). Even more intrigued now...

[mac]

Looks like a bazz fuss, but the diode drop is might not be passive because of q2 base current, my best guess.

However, given that it looks like a BF, at least DC, I thought that both resistors could be made trimmers to find best bias.
Also, ie2 = ib1. What about experimenting with lower gain or darlingtons transistors?

mac

[letsgocoyote]

i just breadboarded this... interesting, the octave was very subtle, but it was there.  although if it soudns better with a humbucker that may be the case, i had a dull squier tele neck pickup to test

anyhow, cool!  thanks for sharing!

[mac]

I experimented with this circuit today. I modified some things.

Bias are quite similar to the Bazz Fuss, lower gain transistors need higher collector resistor Rc and viceversa.
Another fact is that the output  Ve2 is about Vbe1, and the voltage drop across Rc is almost Vcc - Vbe1 -Vbe2.
Note that darlingtons will draw more current than lower gain transistors.

I used a pot as Rc, a pot as feedback resistor Rf (value depends on the gain of q1), and added a gain pot from q1 emiter to ground wired as a standard gain control, with a 6.8uf poly cap connected to the center lug. The value of this pot depends on Rf so as to have a useful range (if Rf >> gain pot then it will not clean up at min).

I tried a couple of darlingtons first, and then a pair of MPSA42 both having a gain of 100.

darlingtons:
I set Rf at 50k (although it sounded good from 10k up to 100k or so), used a 5k-b gain pot (a rev log pot might work better), and used a 10k-b pot as Rf.
At Rf = 5k I got the most octave effect. This pot is very useful since it is like a kind of texture control.

MPSA42:
Rf = 470K, 50k-a gain pot, and 1k-b as Rf.
Max octave at Rf a little less than 1k.

Both sound similar.
The advantage of single transistors is that they will suck less current.
I also tried BC550-C hfe=550, Rf=220k.

I got a nice octave effect given the simplicity of the design, and both the gain pot and the feedback pot interact to give cool tones in between.
The best setting IMHO is Rf halfway between zero and the sweet spot, and gain at 8 or so.
A soft nasal distortion.
I also like the almost clean settings to boost the drive channel.

Cool circuit doug_deeper!

mac

[Johnny Halo]

Can anyone please make a perf layout for this? 

[DougH]

Sounds like a good "exercise for the student" to me. Here's your opportunity to learn how to make perf layouts...

[ppatchmods]

I'm also interested in a perf layout...or vero 

[tiges_ tendres]

Sounds like a good "exercise for the student" to me. Here's your opportunity to learn how to make perf layouts...

Seriously!  This is where you start out.  7 components, nice and easy.

It already practically is a layout there are so few parts!