The Phaser algorithm isn't going to have the multi-phase LFO control that you're going to need for a 'barberpole' phase shifter. As I understand it, you have to ramp the all pass filters up (or down) – offset in phase every 45 or 90 degrees. Then you have to fade each of the filters in & out (triangle wave) to produce the constant rising (or falling) illusion.
You may be able to come somewhat close with a BiPhaz setting, reversed operation, and a very slow Speed value. Still, I don't think that there's enough there to produce a convincing simulation; even with the secondary S-Mod and D-Mod controls. I don't have the Phaser algo – yet – but maybe I can try out some settings in under the 5-minute demo time. I'll let you know what I can come up with.
That's a fascinating proposition for a H9 Special algo feature request, though. It may be too much of a one-trick pony alone. But a quadrature-type LFO, with a parallel LFO controlling amplitude, could be applied to different types of effects:
- All pass filter = barberpole phaser.
- Lowpass / high pass / notch filter = barberpole filter?
- Short delay line = barberpole flanger.
- Short delays & channel amplitude = pseudo-quad panning.
- (Any effect) rate control = Risset-type increasing / decreasing tempo illusion.
- Pitch = Shepard-Risset "scales".
You can hear some of the latter effects using the MicroPitch or H910/H949 algorithms. The pitch shift is in the feedback loop, so regenerated tones constantly rise or fall. It depends on whether there is a slight detuning above or below unison. Not quite the same as Shepard tones, because there's no offsets or fading in & out of pitches.