Ross Phaser More stages... NOW available!

[Fp-www.Tonepad.com]

I'm going to make a board to add more stages to the ross phaser...

And looking at the schematic it's just a matter of splicing the circuit blocks and adding them in between...

The only little issue I have found is the regen control goes to the second phase stage...

Should I 'insert' the new phase stages after the one that has the regeneration or before? I'm thinking after, but I'm not sure.

Any hints? Mark?

Fp

[petemoore]

Both...just thought I'd throw that In...I don't know.
 Preset one and have the other variable...I have no idea how viable these suggestions are.

[Fp-www.Tonepad.com]

hehe...

well, it's very easy to change where the regeneration goes, just a simple wire.

Fp

[Mark Hammer]

Insertion point would depend on how many stages you want to add.  For instance, if you doubled the number of stages by simply duplicating the "row across" arrangement of LM13600s, then there are several viable re-entry points.  If at all feasible, since it doesn't really change that much in the design, why not have a non-dedicated pad at the "regen" end of the 27k resistor coming off the regen pot (on the right of the schem), and similar nondedicated pads tied to pin 13 (or 4, as the case may be) at stages 2, 4, and 6.  This way the user gets to pick max regen or somethng else. The more sophisticated or ambitious user might have a toggle to select regen re-entry point.

If you were just going to redesign for 6 stages (still useful), then the re-entry points become the 2nd and 4th stages.  Keep in mind that, at present, the regen signal "re-passes" through 3 stages (2, 3, and 4).  The scheme suggested here would have it be able to pass through 2 or 4 additional stages  (i.e., 3, 5, or 7 stages).

I forget if someone had contributed a means for adjusting sweep width, but that would be a very useful and appreciated addition if you had it.

[Fp-www.Tonepad.com]

A SP4T switch would be able to switch where the regeneration goes for up to 8 stages...

SPDT if only adding 2 more stages...

You mention stages 2, 4, 6... does this mean regeneration should be at an even number stage?

Fp

[Mark Hammer]

This is one of those screw with your mind things.  Regen needs to come back through an odd number of stages.  Which is why it gets fed to even number stages all the time - because phasers almost always have an even number of stages overall.  So, if you come out through stage 4 and recirculate by opening up door number 2, you pass through stages 2, 3, and 4 again - an odd number of stages.  If you come out through stage 6 and re-enter at stage 2, you still pass through an odd number of stages 2, 3, 4, 5, 6.

The only case I've ever seen where a phaser used an odd number of stages were the original FETs-on-a-chip based Ross Phasers, that had 4 standard swept stages, but fed the regen signal back through a 5th unswept allpass stage back to the 1st stage.  So, even though the phase-shifted output was taken from stage 4, and then fed back to stage 1, it still "played by the rules" by passing through an odd number of stages between the point at which it leaves stage 4 and the point at which it re-emerges from stage 4 after being recirculated.  In effect, they had the same amount of "recirculation intensity", while only needing to sweep 4 stages.  The MXR Phase 100 also used fixed allpass stages and only swept 6, but unlike the Ross unit, it used an even number of stages overall, with no "hidden cards" in the regen path.

Incidentally, as JC Maillet has noted in the past, when one uses a series of allpass (phase-shift) stages to produce vibrato, the degree of pitch distortion/warble that occurs depends on the number of stages employed.  More stages gives a bigger apparent pitch shift.  What this suggests to me is that a supremely flexible and useful layout might also include non-dedicated pads available at pins 8 and 9 so that the person who wanted to could re-direct the 27k resistor from the phase-shift side of the mixing stage at IC1b to whatever phase-shift stage they wanted.

There.  You can take your anti-psychotic medicine now.

[Fp-www.Tonepad.com]

Ok, it's all clear now.

After listening to the 12 stage phasers that some have posted sound samples of, I really want to try using more than 4 stages.

The regen was just a little thing that I wanted to be clear of.

You can expect an add on 4 stage pcb (two LM13600) in the near future (of course expandable to whatever the number of stages the builder wants)

I'll start with 8

Fp

[Fret Wire]

The only case I've ever seen where a phaser used an odd number of stages were the original FETs-on-a-chip based Ross Phasers, that had 4 standard swept stages, but fed the regen signal back through a 5th unswept allpass stage back to the 1st stage.  So, even though the phase-shifted output was taken from stage 4, and then fed back to stage 1, it still "played by the rules" by passing through an odd number of stages between the point at which it leaves stage 4 and the point at which it re-emerges from stage 4 after being recirculated.  In effect, they had the same amount of "recirculation intensity", while only needing to sweep 4 stages.  The MXR Phase 100 also used fixed allpass stages and only swept 6, but unlike the Ross unit, it used an even number of stages overall, with no "hidden cards" in the regen path.

http://www.home-wrecker.com/ross_phaser_orange2.png
I just picked up that version of the Ross phaser pretty cheap to compare soundwise to FP's Ross layout. When did the AM97C11 FET array chip go the way of the Dinosaur? I would've loved to adapt one to the phase 90, or experiment in general with one. It's been gone quite some time I believe.

[Vsat]

Take a look at the Moog ladder filter to see how regen is implemented. At the lowpass cutoff freq, each stage contributes  45 degrees of phase shift. The four stages together will contribute a total of 180 degrees. This will result in positive feedback (regeneration, resonant peaking) when the regen signal is applied to the inverting input of the diff pair at the foot of the ladder. If the signal is applied to the non-inverting input (where the audio signal is fed) negative feedback will result and the response will be flattened, rather than producing peaks.

The typical phaser allpass chain is capable of producing considerably more than 180 degrees of shift at it's center frequency.. 720 degrees, 1440 degrees etc are common.  As a result, peaking will occur for certain frequencies whether the signal is fed to the inverting or non-inverting input, however the peak positions will be different for each case. The fly in the ointment is that the notches will be wiped out unless the regen signal is fed back in the correct polarity. The A/DA flanger and other phasers/flangers that can operate in normal and inverted modes do this in a nice, compact way.

Many "tapped"  phasers combine the regen signal with the direct signal prior to entering the allpass chain. Others feed it directly into the inverting input of the op amp in an allpass stage near the front of the chain. The response for the stage is altered in this case. Which may or may not be a good thing, depending on personal taste.
Cheers, Mike

[RickL]

I really have to finish wiring mine up. I did exactly what you're talking about, added 4 more stages by slightly modifying the part of the board that has the phasing stages on it. I had grand plans to wire it so the extra stages could be added in parallel or series, the regeneration point could be selected and the phasing could be either added or subracted (like the phaser in EPFM).

I got as far as building and testing the original board, building the add-on board, adding selectable filter caps and phasefilter switches to both boards, adding and testing an add/subract switch to the original board, adding a dry lift (vibrato) switch and working out the wiring for the series/parallel switch (4p2t rotary) and the regeneration selection switch.

It's sat cluttering my work bench for several months now waiting for about an hours worth of work to wire the last bits. Maybe this will be the spur I need to finish it.

I love projects that can be expanded by simply adding multiple identical boards. If this one is successful I hope you consider doing the same thing for the Pez 45 and 90.

[Fp-www.Tonepad.com]

Nice input. Thanks everyone.

Rick,

I see what you mean, I'll start with the ross phaser because it's my favorite phaser of them all. I'll try to do it for all the projects including the Rebote2 delay.

Fp

[Joep]

It can be made really interesting by also providing an extrenal LFO for each stage. Like Lovetone does with it's Doppelganger, here you can you select different LFO's for the 1st and 2nd and for the 3rd and 4th stage.

Nice idea FP!

Bye,

Joep

[Fp-www.Tonepad.com]

Yet another board can be made with only the LFO. This way it would be possible to incorporate other(s) LFO(s) to the circuit. Or use them in another circuit

Fp

[puretube]

...The only case I've ever seen where a phaser used an odd number of stages were the original FETs-on-a-chip based Ross Phasers...

there are others, though...  :wink:

[Mark Hammer]

Well, this morning while, um, "reading", I was looking at the service manual for the old Maestro Stage Phaser.  It also uses OTAs (3094), but seems to have only 5 swept stages plus one fixed stage.

[Khas Evets]

I too would love a 6 or 8 stage version of the Pez 90. Can't wait.

[Fp-www.Tonepad.com]

Updated Ross Phaser page:

http://www.tonepad.com/project.asp?id=25

I have made a slightly modified layout that includes the pads for the add-on, but I'll be making this mods to an old version pcb and maybe I'll take pictures, it shouldn't be too hard, you'd need to solder wires to the jumpers, and standing resistors.

Cascading is possible, and anybody who knows a little electronics can figure it out.

Also, switching is possible, the easiest way would be to short pads "x and y" in the main pcb by using a spst switch. More complex switching is going to have to be figured out by the builders.

Fp

[MartyMart]

That's great Fp
I just picked up a very old "purple bud box" Ross "Phase R1"
Its script, one knob, 4x trannies and sounds like somewhere between a
Phase 45 and a Phase 90.
I didn't see it mentioned at the home-wrecker site, so it must
be quite rare !!

Marty.

[Fp-www.Tonepad.com]

Marty,

I'd like to see pictures!

Fp