MXR Phase 90 feedback mod

[bassmeister]

I'm thinking about adding adjustable feedback to my stock Phase 90. I don't just want to replace the feedback resistor with a pot. Instead I want to try out a solution similar to the one in this schematic: http://home-wrecker.com/ross_phaser_orange2.png. However (as far as I can tell), that solution gives only positive feedback. I'm imagining an all-analog version of the phaser module in the Nord Modular synth, which has a knob for adjusting positive/negative feedback. In middle position, there's no feedback, turning it CW from middle position adds positive feedback and turning it CCW adds negative feedback.

Any ideas on how this could be accomplished in the analog domain? You could have both a positive and a negative feedback path and use a switch to select between the two, but it's a bit more slick to have continous adjustment, right? Perhaps those pots with a "click" in the middle used to control balance on an amplifier could be used? I'm not sure on how they work, but isn't it sort of made out of two pots where the resistance of one is zero (or full) and the othor is adjusted during one half of the turning range and vice versa?

What do you think?

[Mark Hammer]

A similar solution is sometimes used for panning between noninverted and inverted envelopes.

Essentially it works like this.  Your signal source is split, with one path going to a unity-gain inverting buffer.  The other path goes to one side of a linear pot and the output of the buffer goes to the opposite side of the pot.  When the wiper is in the middle the two versions of the original signal cancel, with one superceding the other as you pan out to one side or the other.  The sensitivity control on the old E-H Y-triggered Filter works like that.

Of course, when it comes to positive vs negative feedback in phasers, there is the added complication that positive feedback yields a different final volume level than negative, the same way that summing yields a different level than differencing when phase shifted and dry signal are combined.  In general positive feedback gets you more volume and a more resonant sound with more feedback.

I'm not sure where the signal is tapped from and fed to, but in my DOD FX20 (very similar to the LM13600-based Ross unit, but with 6 stages instead of 4), increasing feedback sounds great in the one type, and positively awful in the other.

I guess the moral is, be careful what you wish for...you might get it.

[bassmeister]

Well, I guess the volume differences could easily be accounted for when working digitally, and a bit more difficult in the analog domain. I'll have a go at your solution when time is given... I wonder what it will sound like at extreme resonance settings... Should probably distort, but it might sound cool!

Thanks a lot!