Below is a schematic of a German phaser instrument effects unit from the late 60s/early 70s called “Rotor Sound” and manufactured by Schaller. “Rotor Sound” alludes to the sound produced by a Leslie speaker cabinet. A demonstration of the effect produced by the unit can be heard here.
Similar to the vibrato units Hammond made for its organs, the Rotor Sound uses a saturable reactor as part of the effects filter (as seen at the top of the schematic). The effects filter itself is comprised of a 18-rung ladder network of low-pass filters, the series elements of which are the DCR and the inductance of each reactor’s “power” winding and a capacitor to ground as the sole shunt element.
Driving the reactor’s control winding is a vactrol-based oscillator (lower middle of schematic).
The saturable reactors in the device are two-winding P14/8 pot core transformers:
Someone else measured these saturable reactors; the power winding (the “in circuit” winding) is 2.2mH with a DCR of 65 ohms. The schematic itself does not denote the inductance; all it states is that the total resistance of the power windings in series is 800 ohms (800/18 =~ 44 ohms each).
What I need help with is understanding how to look at the effects filter. My understanding so far is that when the current through the control winding is low, each rung of the ladder network is a second-order LCR filter. When the current through the control winding is high, the core of the reactors saturate and the inductance in the LCR drops significantly, changing the filters topology into that of a first-order RC filter. However, I feel like this is the wrong way of looking at it as daisy-chaining 18 low pass filters is not going to produce an easily calculated response at a given frequency. Still, I feel like this shifting of topology twice per cycle of the LFO is key to understanding the function of the effect.
I’m asking for insight because I have managed to salvage a couple saturable reactors from some Hammond vibrato units and would like to repurpose them for something closer to the phaser schematic above.
Thank you in advance.
Similar to the vibrato units Hammond made for its organs, the Rotor Sound uses a saturable reactor as part of the effects filter (as seen at the top of the schematic). The effects filter itself is comprised of a 18-rung ladder network of low-pass filters, the series elements of which are the DCR and the inductance of each reactor’s “power” winding and a capacitor to ground as the sole shunt element.
Driving the reactor’s control winding is a vactrol-based oscillator (lower middle of schematic).
The saturable reactors in the device are two-winding P14/8 pot core transformers:
Someone else measured these saturable reactors; the power winding (the “in circuit” winding) is 2.2mH with a DCR of 65 ohms. The schematic itself does not denote the inductance; all it states is that the total resistance of the power windings in series is 800 ohms (800/18 =~ 44 ohms each).
What I need help with is understanding how to look at the effects filter. My understanding so far is that when the current through the control winding is low, each rung of the ladder network is a second-order LCR filter. When the current through the control winding is high, the core of the reactors saturate and the inductance in the LCR drops significantly, changing the filters topology into that of a first-order RC filter. However, I feel like this is the wrong way of looking at it as daisy-chaining 18 low pass filters is not going to produce an easily calculated response at a given frequency. Still, I feel like this shifting of topology twice per cycle of the LFO is key to understanding the function of the effect.
I’m asking for insight because I have managed to salvage a couple saturable reactors from some Hammond vibrato units and would like to repurpose them for something closer to the phaser schematic above.
Thank you in advance.
Attachments
For contrast, here is the schematic to one of the Hammond units from which I have salvaged a pair of saturable reactors:
This is the above unit, modified for use as a standalone effect for guitar:
This is the above unit, modified for use as a standalone effect for guitar:
My take is the filters are a "T" section all pass filter, (lattice phase equaliser) with up to 180 degrees phase shift per filter, 3240 degrees total.
The saturable reactor changes the coils inductance, so varies the frequency of the phase shift.
When combined with the direct signal, the phase shifted comb filter response sweeps up and down in frequency which varies with the rate of the oscillator.
I’ve been reading about phaser effect pedals and it seems like every pair of filters creates a 180° notch
I’m puzzled how a 100 ohm pot can vary the frequency of the 5V secondary. Surely all it does is affect the luminosity of the bulb by stealing current from it.
Phaser is a filter effect that is modulated.
Done with all pass filters.
180 degree phase shift, but still passes all frequencies.
2 all pass filters in series then summed with original signal create a notch filter.
If the original signal is not summed it is considered a true vibrato effect
By adding more stages you get more sweeping notches.
Basically the same as a single multiband EQ band set to cut, which creates a notch.
The frequency of that cut is modulated.
There is numerous ways to make the notch more narrow like a high Q cut.
Or make it wider by over lapping or changing the all pass filters.
Takes 2 all pass stages to create one notch.
So 2 stage phaser has one notch filter, 4 stage would have 2 notch filters and 6 stage would have 3 notch filters.
being modulated.
Most modern phasers of the 70's 80's started using opamps so no inductor needed.
Much like multi band EQ's which also no longer require inductor filters.
Modulation is more even and symmetrical with opamp LFO and it is easy to retrigger/restart/sync the oscillator.
Can retrigger while running, or most make the on/off switch retrigger the LFO so the beginning of the" Swoosh"
starts when you turn it on. Triangle LFO so you can retrigger the upward ramp or downward ramp.
The phaser can start basically at the high frequency end of the swoosh, or start at the low end frequency of the swoosh.
Of course before that the transistor replaced the tubes needed. So a phaser was able to fit in small box.
I guess the Uni Vibe is well known from hendrix fame. Marketed as a rotary speaker emulator.
Since I guess nobody would buy a " modulated notch filter" what the heck is that?
They are noisy , muddy and modulated with light sensors,
With opamp you get way way way less noise and can just use a fet to modulate the all pass filters with a lfo.
No expensive noisy inductor/ transformers or no light bulbs or light sensors needed.
The overlap of the univibe all pass filters creates a wide low Q notch filters, as with the more known
MXR or roland phasers has a more narrow high Q notch, which is typically better for cut EQ.
Narrow cut, wide boost as are most typical " 7 band EQ" pedals
Done with all pass filters.
180 degree phase shift, but still passes all frequencies.
2 all pass filters in series then summed with original signal create a notch filter.
If the original signal is not summed it is considered a true vibrato effect
By adding more stages you get more sweeping notches.
Basically the same as a single multiband EQ band set to cut, which creates a notch.
The frequency of that cut is modulated.
There is numerous ways to make the notch more narrow like a high Q cut.
Or make it wider by over lapping or changing the all pass filters.
Takes 2 all pass stages to create one notch.
So 2 stage phaser has one notch filter, 4 stage would have 2 notch filters and 6 stage would have 3 notch filters.
being modulated.
Most modern phasers of the 70's 80's started using opamps so no inductor needed.
Much like multi band EQ's which also no longer require inductor filters.
Modulation is more even and symmetrical with opamp LFO and it is easy to retrigger/restart/sync the oscillator.
Can retrigger while running, or most make the on/off switch retrigger the LFO so the beginning of the" Swoosh"
starts when you turn it on. Triangle LFO so you can retrigger the upward ramp or downward ramp.
The phaser can start basically at the high frequency end of the swoosh, or start at the low end frequency of the swoosh.
Of course before that the transistor replaced the tubes needed. So a phaser was able to fit in small box.
I guess the Uni Vibe is well known from hendrix fame. Marketed as a rotary speaker emulator.
Since I guess nobody would buy a " modulated notch filter" what the heck is that?
They are noisy , muddy and modulated with light sensors,
With opamp you get way way way less noise and can just use a fet to modulate the all pass filters with a lfo.
No expensive noisy inductor/ transformers or no light bulbs or light sensors needed.
The overlap of the univibe all pass filters creates a wide low Q notch filters, as with the more known
MXR or roland phasers has a more narrow high Q notch, which is typically better for cut EQ.
Narrow cut, wide boost as are most typical " 7 band EQ" pedals
Last edited:
Technically the univibe is 4 stage with 2 notches. But the second notch was set way out in outer space.
And modulates high frequency not really heard with guitar or the bandwidth of guitar speakers.
So it actually sounds more like a 2 stage.
Hilarious since it would be less noisy and muddy if they removed the worthless upper notch.
Some people like the softer swoosh and not overly dramatic like other phasers.
Another actual 2 stage phaser is MXR 45.
Not the more popular MXR 90 from Van Halen fame
which is 4 stage.
And modulates high frequency not really heard with guitar or the bandwidth of guitar speakers.
So it actually sounds more like a 2 stage.
Hilarious since it would be less noisy and muddy if they removed the worthless upper notch.
Some people like the softer swoosh and not overly dramatic like other phasers.
Another actual 2 stage phaser is MXR 45.
Not the more popular MXR 90 from Van Halen fame
which is 4 stage.