ross/ropez stages switching question

[jdub]

Howdy all-

Been out of the loop for a bit, but finally got some time to get back to building (a Ross/Ropez phaser) and had a coupla questions- used the search function and couldn't find exactly what I needed. First: what is the simplest way to switch between 4 and 8 stages, i.e., how would one wire the extra stages board to be switchable?  Using the Tonepad layout, of course, would one simply wire it so pads X and Y short together (for 4 stages) and connect to the daughterboard for 8, using a DPDT, leaving all other connections between the two boards permanent?  I also read about Mark Hammer's method using the 27k mixing resistor, but couldn't visualize it.  Finally, I assume the phase filter mod would only work in 4-stage mode, correct?  Any tips (or a diagram maybe  ) would be most appreciated. Thanks!

J.

[jdub]

Or would it better to just add the extra stages permanently and not worry about switching?...choices, choices... 

[Mark Hammer]

In some respects, the simplest thing is to do what is suggested over at the Tonepad site, and simply stick the additional 4 stages between stages 2 and and 3 on the main board (which is precisely what the pads on the main board anticipate).  This simply uses a DPDT toggle to engage, and has the added feature of not requiring any futzing around with the feedback path since it goes exactly where it needs to, using that method.

However, in practice, the addition of 4 more stages has the impact of sharing the current from the LFO among more stages, resulting in less apparent sweep (it can sometimes result in greater risk of oscillation if the feedback is set too high, but that's something you just have to play by ear).  All of which suggests that the best wey of making the additional stages something you can switch in and out is to use a 3PDT toggle, that will straddle the 10k resistor on the LFO output with another fixed resistor when you add the 4 more stages.  Reducing the current-limiting would permit for 8 stages to sweep as wide as 4, or at least not less than you'd like it to.

What is the "correct" combined parallel value of that resistance?  Beats the heck out of me.  I know that feeding the OTAs too much current will fry them, and I also know that I've used as little as 9k1 on a 4-stager, and it provides exceptionally wide sweep without ever having fried a chip.  So, that suggests yu might get away with adding a 68k resistor in parallel with the 10k to get a combined equivalent value of 8k7.

I leave it up to the math whizzes, however, to tell us how much we could reduce current limiting when feeding 8 stages vs 4.

[jdub]

Thanks, Mark.  I'm planning on using a 9k1 in place of the 10k with a 10k pot for width- with the pot at minimum, would this provide sufficient current when the extra 4 stages are switched in? 

Any thoughts on the phasefilter mod with 8 stages?

[Mark Hammer]

Any thoughts on the phasefilter mod with 8 stages?

Haven't tried it, myself.

I will say that the phasefilter mod is at its best (IMO) when the dry signal is lifted.  When that consists of 2 allpass and 2 lowpass stages, the amount of pitch variation is reasonable.  The more allpass stages are added, the greater the pitch variation when set to vibrato (dry lifted).  Six stages of phase-shift (allpass) MAY yield a still-pleasing vibrato, or it may yield an annoying amount.  Four stages of lowpass will provide a fairly obvious wah-wah filtered sound.  Whether you view that as an advantage or not is up to you.

As a fan of more restrained effects, that let the melody and harmonies shine through, I'm more inclined to go with only two stages of lowpass for phasefiltering.  YMMV.

[jdub]

Agreed- I'm not looking for obnoxious  ; I was planning on just two stages anyway, with the dry lift.  Thanks for the tips, Mark.  I'll give it a shot & see what happens...