anderton ring mod..... insane

[Thomas P.]

Hey Paul,

thank you very much!!! I'm going to read some stuff about it and maybe have some post then :wink:

[Mark Hammer]

Although you can obtain a quieter (i.e., carrier-free) signal using better quality 4-quadrant multipliers like the LM1495/96, the AD633, etc., the advantage that the Anderton 565-based unit has is that it comes with its own internal oscillator and is easily tweaked with a few component changes.  Higher quality ones do not stop once you have the core for modulation; you still need a modulaTOR - i.e., a VCO or other oscillator.  That isn't difficult to provide.  All you really need is a standard 2 op-amp LFO circuit as found in any sort of modulation pedal (tremolo, vibrato, chorus, flanger, phaser, panner), and drop the value of the range-setting cap to something 1/5 to 1/10 the value (e.g., drop a 1uf cap to 0.1uf).  That will deliver an oscillator with a suitable range.  See the Hollis Frobnicator for an example.

One solution which comes up occasionally but has never been implemented in a user-friendly way is a gate add-on for the Anderton circuit.  Flangers and chorusses also have a problem with leakage of unwanted signal into the output.  The way that Boss solved it in the CE-1, and A/DA solved it in their Flanger, was to have a FET-based gate that provided a low-resistance path to ground for the effect signal when the input was below threshold, and a high resistance path to ground once the input was above threshold.  As long as there is something "going on" the high-pitched whine is not as noticeable or objectionable.  Once you stop playing, though, you really notice it....and hate it.  The gate addressed this aspect.  Later pedals, of course, used companding instead of gates since they did not chop the envelope of the signal quite as much.

In the case of the Anderton  Ring Modulator, both modulated and straight signal are fed to the inputs of an op-amp mixer stage via resistors.  The simplest solution to the whine would be to tap the input from the first op-amp stage (I'm using the EPFM-II version here as my reference point), run that to a simple Doctor Q-style envelope-follower, and drive the LED half of an optoisolator.  The LDR half would go in series with the existing fixed resistor where the the modulator signal goes to the mixer.  When you play, the LDR resistance goes low and the modulator signal is mixed in at the level set by the effect-level pot.  When you stop playing, the LDR resistance goes high, and the effect signal is then mixed wayyyyy down, regardless of where the pot is set to.  Naturally, you'd need to play with the envelope-follower parameters, the existing series resistor value, and perhaps a resistor in parallel with the LDR, depending on its value, so that you get the ideal attenuation and restoration.  Given the size of Anderton's PCB layout, or JD Sleep's alternate layout for non-4739 dual op-amps (available at GGG), it would be a simple matter to whip up a small perf daughter-board with the envelope follower and optoisolator, and hot-glue/epoxy it above the existing board on standoffs.  The tapping and insertion points are easily identifiable and accessible.  I'd suggest that the existing 4k7 fixed resistor for the effect path at the mixing stage be on the daughter board instead of on the main board.  That way you can simply run a lead from and to the spot where the resistor used to be without having to modify the main board at all.  Those two leads will now go to the LDR/resistor combo.  Make that an LDR/trimpot (10k) combo for fine adjustment.