Mad Scientist Lab: From Electronic Projects For Musicians Ring Modulator to the Theramin
ฝัง
- เผยแพร่เมื่อ 17 ก.ย. 2024
- This video is all over the map. I started to build Project 9 Ring Modulator from Andertons Electronic Projects for Musicians, but then I noticed that Ring Modulators are part of AM radio broadcast, and that looked a lot like the beat frequency interactions we hear in musical instruments. It turns out the Heterodyne effect is the same as beat frequency (combination or Tartini tones). Whats even more cool is that Russian Physicist Lav Theramin made a musical instrument by creating an audible sound from the Heterodyne interaction of two high-frequency radio waves. Waaaay too much to cover in a short video. I will come back to this... for now enjoy my ADHD moments caught on camera...
I'm just considering this build, I have the book open in front of me right now.
I never could find the phased locked loop he used in that project. I found another circuit that uses simple diodes and op amps instead. I’m going to make mine that way and show the video here... eventually.
The thing guitar players neglect, or perhaps simply don't know, about ring modulators is that they were developed to modify *pure steady-state* waveforms, generated by electronic oscillators. You feed in a 400hz sine wave, modulate it with 100hz sine wave and you get out 300 and 500hz.
But guitar signals have NEVER been anything like that. First off, they are harmonically rich, Second, the harmonic content changes with time after initial pick attack. Third, the harmonic content is a function of the particular fret (i.e., 2nd, 5th, 9th, etc.). Finally, we tend not to notice it, but we generally introduce a small amount of pitch deviation when we pick. That can be from bending the neck a bit, from lightly, then firmly pressing the string down, from pushing the string to the side a bit when we hit it hard with a pick, and so on. The point is, a guitar string under normal use is miles away from what a ring modulator expects.
Can we help this scenario? Yes. First off, heavier gauge strings that will reduce pitch deviation, and have less harmonic content to begin with. Second, use the neck pickup, with the tone rolled back. Third, use some compression or limiting before the ring modulator. All of this is intended to make the guitar signal come closer to the conditions anticipated by ring modulation.
In the case of the EPFM circuit, there are several things one can do, and which are already done, to further that. The diode pair to ground form a very crude peak limiter, and provide a more constant signal level. But there is NO filtering whatsoever of the input. That first op-amp stage needs a feedback capacitor. A 1nf cap ought to be a good start. Second, there ought to be some additional filtering between the output of that stage and the input to pin 2 of the 565 chip. Here, I suggest divvying up whatever series resistor one used (33k or greater, depending on application) into several smaller series values, adding up to whatever the original resistor was. For example, 33k could be turned into 18k+15k, or perhaps 5.1k+5.6k+22k. Run a small value cap from the junction of the series resistors to ground. For example, assuming the series resistors are 5k1->5k6->22k, use 18nf for the 5k1/5k6 junction, and 15nf for the 5k6/22k junction. That provides 2 poles of lowpass filtering just a bit below 2khz.
Now, following the math, the higher the modulating frequency, the further away from the note will the sideband products be. So, if one wants the end result to sound somewhat "pitched" (i.e., you can follow a sort of melody) and simply a bit "boing-ey", you tune the modulating frequency as low as you can. Again, because of the math, the higher up the guitar notes are, the more the result will have a pitched sound, with a bit of "rubber band" to it, because the sideband products will be closer to the original pitch. And because we have taken steps to make the guitar signal approaching sinusoidal quality, we won't be getting sums and differences of the various harmonics in the guitar signal.
Finally, about the modulator "whine". Sometimes you can fully null out the whine with the 2K5 - 5k trimpot (which should be a 10-turn type). But if that proves to be unfeasible, I use a quick and dirty noise gate. Replace the 4k7 resistor that feeds the ring modulator to the op-amp output mixing stage with a 2K2 resistor, in series with a back-to-back pair of Schottky diodes (1N5817, BAT41, etc.). This will block any low-level signals (
You have way more in depth knowledge about ring modulation than me. I’m impressed. I did find a way around many of the problems I had with the phase locked loop circuit and came up with a fairly useful circuit that could be coaxed into making a ring modulator or a green ringer. I really need to document that build and put it on here. Thanks!
@@MadScientistGuitarLab I look forward to it.
Hello we are new here, Have you've made the Phase Shifter project no.21?
I’ve always wanted to make that project. I got distracted and never got around to building more of these projects. The ring modulator is the one I want to try.