univibing a small stone

[leonhendrix]

using the layout from tonepad and using these values 1nf 10nf 470pf 4.7nf from
http://www.lynx.bc.ca/~jc/pedalsSmallStone.html

where would i put each value.

leon

[lightningfingers]

Replace the 0.0068uF cap in each stage with the univibe cap. *BUT* make sure you know which phase stage is which so you can put the 'vibe caps in the right order.

Have fun :mrgreen:

[leonhendrix]

I've already built the pedal but i'm not sure where the caps go

leon

[stm]

I don't think the order of the caps in the stages would make a difference. This is true as long as the stages operate in a linear fashion, which is supposed to be.

It is like saying A+B+C+D = D+A+C+B , or whatever other ordering you can imagine.

Just replace the caps and have fun!

[leonhendrix]

thanx

[Mark Hammer]

Its possible that the order of the caps WILL make a small audible difference, though not the sort of difference that might prevent it from sounding uni-vibey rather than phase-ey.

Why?

Each stage imposes an amount of phase shift that is set by the value of the cap and resistance to ground, but ALSO by the frequency of the input signal.  Since the resistor/cap combo is essentially a highpass filter (cap in series, resistor to ground), MORE phase shift is applied in each stage as you go higher up in signal frequency, up to a maximum of 90 degrees per stage.

The notches are produced when you add up the TOTAL amount of phase shift over stages and observe multiples of 180-degree-shift at any given frequency (and mix that phase-shifted signal with a dry signal).

Identical-value caps in each stage cause a "snowballing" effect that produces maximum possible phase shift from that number of stages (i.e., 360 rather than, say 120) at fairly focal frequencies.  As the resistance to ground in each stage (in a standard FET/LDR-based unit) is changed at the same time, all stages change where the maximum phase shift starts to occur in the frequency spectrum.  Of course, this results in the location of notches (where this phase shift has summed to a maximum) moving around as the LFO sweeps the FETs/LDRs up and down in value.

So, back to the univibe scenario.....

What the different cap values do is to *prevent* the phase shift from being disproportionately greater at one frequency than at others.  It DOES produce notches, but they are more spread out, less resonant (i.e., affect neighbouring frequencies) and shallower.  It does this by having the impact of each stage on signal phase start at different points in the frequency spectrum.  Have stage 1 start applying phase shift at a lower signal frequency (because of the very large cap value) and more phase shift is applied across the full spectrum (phase shift in each stage rises from zero up to a plateau of 90 degrees with the "shelf" starting at a point set by the cap/res).  If we applied the same amount of phase shift in subsequent stages that would be one thing, but with smaller cap values, the additional phase shift applied only really starts much higher up.

This is the key to the Univibe - the distribution of phase shift.

If at some given moment, you end up with 280 degrees of overall phase shift at frequency X by the final phase-shift stage, it doesn't really matter whether you got them as 70-70-70-70, 10-120-90-70, or 160-20-70-30.  It's still a signal that is 280 degrees out of phase with the dry signal at frequency X.

[lightningfingers]

While we on this subject of Univibe caps, does anyone have an idea as to why the uneven cap values were chosen?

Is this the only phaser that does that?

[puretube]

it`s based on the findings of Mr. Dome and the filters named after him*.
The univibe was not intended to be a "phaser".
Almost every "vibe" got diff. caps.
Almost every "phaser" got ident. caps.

*most of the answer can be read in and between Mark`s lines...

[Mike Nichting]

Try www.moosapotamus.com. Charlie explains the mods he has done to his small stone. Be sure and check out all the mods he did. He is a very cool guy and the man when it comes to modding.

Hope this helps~!!

Mike N.
NUKLHED