Ring Modulator is not ring modulating

[honki-bobo]

I was using the Cherry Audio Ring Modulator module today and I found that it actually isn't ring modulating.

From what I understand, a ring modulator takes two input signals x(t) and y(t) and outputs z(t) = x(t) * y(t), but the Ring Modulator module definitely has a different output.

signals.jpg (91.4 KiB) Viewed 7467 times

Frequency-wise (if the input is a sine wave) a ring modulator should output the sum and the difference of the frequencies of the input. This isn't happening, either.

frequencies.jpg (61.46 KiB) Viewed 7467 times

Correct me if I'm wrong or if I'm missing something here, but the Ring Modulator module does not actually ring modulate the input signals. And I haven't found anything special being mentioned in the documentation at https://docs.cherryaudio.com/cherry-aud ... -modulator.

Best regards,
Martin

[huggermugger]

Agreed. Whatever their RingMod is doing, it's not ring modulation. Two sinewaves should produce a similarly pure, dull tone when ring modded, not that buzzy, overdriven sound that RingMod produces.

I tested a few ring modulators in VCV Rack, and generally got the expected results, although three of them (the three on the bottom) didn't retain the sine wave purity, perhaps I just don't know how to use them properly.

[UrbanCyborg]

I wondered about this, too. Maybe they've just tapped into the confusion there seems to be in general about what a ring modulator is, and coded up something that does what one of the "musical" interpretations has it to be. I do notice that elsewhere in the docs there's some confusion about just what AM and FM do (in the description of FM, as I recall), although I don't think in that instance the code actually does the wrong thing.

[ColinP]

Well spotted Martin.

It's a bit weird as the code for true ring modulation is utterly simple. Yet CA seem to have made it complicated by adding a non-linear transfer curve.

To get true ring modulation just use the Formula module and enter the formula x*y.

I knocked up the following patch to compare the waveforms and sounds.

The green trace shows ring modulation.

The cyan trace shows CA's Ring Modulator output.

ringmod trial.png (416.63 KiB) Viewed 7430 times

[Cherryaudio Danny]

Hello,

The Cherry Audio Ring Modulator essentially multiplies the signals together as expected. For example, feeding +2v and -2v into the ring modulator outputs -4 volts. However, as you'd find in hardware, there is a maximum voltage range of -5 to +5 volts. The ring modulator will distort as you approach the maximum voltage range, which is a part of its character and sound.

If a 'true' ring modulator is desired, attenuating the voltage range to be around or below -4 to +4 volts to avoid engaging the distortion will do the trick!


Danny

[anonyrat]

Looking at the CA blue line I think it should be called a drum modulator. Anyone else see drum music there - bum.titty.Bum.titty ?

[honki-bobo]

For example, feeding +2v and -2v into the ring modulator outputs -4 volts.

Sorry, but no, it doesn't. If I feed it -2V on the x input and 2V on the y input, it outputs about -3.1241267V when both "levels" are set to 50% (which is the default) and -4.06798014V when they are set to 100%.

I also get the same "distorted" output when I scale the oscillator signals down to -1..1V with an Attenuverter set to 20% and it also doesn't make a difference if I use the "Level" knobs directly.

[Cherryaudio Danny]

Hello,

Interesting. I'll reproduce and log a bug report if that's the case. By design, I believe it's supposed to be 'pure' until the threshold mentioned.
Thanks for taking the time to look into this!

Danny

[honki-bobo]

Thanks, Danny, for passing this to the developers.

FWIW, I found that if I set the levels to 0.01 % and scale the output back up by about 120 db I get the expected ring modulation results.

ringmod_with_scaling.jpg (224.58 KiB) Viewed 7391 times

[Steve W]

Interesting effect as two sine waves go in and out of phase.