Can you tell me how it works?

[gasmask]

I have one problem with understanding Tim Escobedo's Rambler circuit.
I mean, I know how 386 chip works.

1. But why there are 2 diodes?
2. Does this second one (connected to pin 3) is necessary?
3. How does the sound 'behave' in this two diodes?

gasmask

[thelonious]

Not really familiar with the circuit, but I'll take a stab at it. It looks to me like the series diodes half-wave rectify the signal. One cuts off the top of the wave, and the other cuts off the bottom of the wave. There is some good information about how that affects the sound at http://www.geofex.com/effxfaq/distn101.htm. That creates an octave sound on its own.

One diode goes to the + input of the 386, and the other to the - input (through the pot). The - Input is out of phase, so it flips that half-wave over and you end up with close to a full-wave rectified signal, which has a lot of octave sound. The 500k pot controls how much of that second half-wave makes it through to the 386. Or something like that.

I'd be interested to have some of the more experienced guys chime in and fill out my answer a little bit---you have me curious now, too. 

[gasmask]

Well, I understand now. So the diode going to + doesn't have to be there... ?
What if I disconnect pins 1&8 (in theory)? 

[thelonious]

So the diode going to + doesn't have to be there... ?

You would lose most of the octave up effect if you removed it, but it doesn't have to be there.

What if I disconnect pins 1&8 (in theory)?  

The jumper connecting pins 1 & 8 bypasses an internal resistor that sets gain. The jumper sets the gain at all frequencies to 200. If you disconnect it the 386's gain will go down to 20.

[gasmask]

The jumper connecting pins 1 & 8 bypasses an internal resistor that sets gain. The jumper sets the gain at all frequencies to 200. If you disconnect it the 386's gain will go down to 20.

And will it destroy octave effect?

[GibsonGM]

The jumper connecting pins 1 & 8 bypasses an internal resistor that sets gain. The jumper sets the gain at all frequencies to 200. If you disconnect it the 386's gain will go down to 20.

And will it destroy octave effect?

The jumper connecting pins 1 & 8 bypasses an internal resistor that sets gain. The jumper sets the gain at all frequencies to 200. If you disconnect it the 386's gain will go down to 20.

And will it destroy octave effect?

Remove it, and see!!  )     I think it will still octave, but will be very low volume.

Thelonius' analysis of how the circuit works looks correct, especially the part about the phase difference (which has very much to do with the octave effect).   And as the diodes clip, they will also remove signal level (of course), so you would need to have the extra gain to recover from that.

[pinkjimiphoton]

the diodes are in large part what creates the effect.  if you remove one, you'll get a boost... i don't think it will even matter if there's one remaining as shown in tim's circuit.

you are clipping the top and bottom of the waveform and compressing the signal some, generating harmonics.

when you turn the knob, you're literally phase cancelling out the fundamental as you turn it, resulting in more octave up sound.

i've been meaning to try this circuit out.

hope i didn't muddy the waters further...but most octave ups use a pair of diodes like that, well, not LIKE THIS, but to help form the octave effect.