I’m thinking about possibly using an atan() for the clipping. We could use a clamp but it’s pretty abrupt, although the clip on your traces looks pretty brutal.
1 Like
Yeah I think hard clipping is the way to go for this one. I have an idea for how to make the glitch but it’s gonna take a bit of experimenting.
Here’s the waveform from the spline with a “wiggle” at the end. With this approach the wiggle frequency will change with the input frequency.
Not too far off. If the glitch is truly resonance the the frequency of the glitch shouldn’t change with the input frequency. Ableton has a pretty nice spectrum analyzer, you might be able to see the glitch on a sine.
1 Like
I just checked it out on the oscilloscope an the resonance does seem to have a static frequency, not expanding when I sweep the frequency knob.
I tried to use the spectrum analyzer built in to spot a static spike but was below the threshold. I increased the frequency until the wavelength was approximately the same as the resonant wiggle and I and getting 11.7 Khz
That would seem to indicate that it’s a resonance rather than something introduced as result of the JFET non-linearity. There are enough capacitors and active elements in the circuit that it would be hard to say what’s resonating even if we had the actual schematic. You might be able to use a resonant filter, kick it with a pulse when the signal exceeds some threshold and add the output to the output waveform.
What would be the most efficient resonator for the purpose?
It don’t mean a thing if it aint got that PENG! Seriously, I would probably use the built-in filter with resonance or a bi-quad with feedback. The built-in rings nicely and is pretty efficient. Be careful! The built in will happily go into self oscillation. Best to keep the resonance a bit below 1. Remember that the resonant frequency won’t be quite the same as the cutoff frequency.
1 Like
I noticed that the clip is asymmetric. That implies that the clip is occurring after the JFET. Since the JFET is already attenuating one side of the waveform, a symmetrical clip will chop off one side before the other. It’s not clear from the traces whether there is also a DC offset on the signal going into the clip, but I think probably not. You should be able to vary the gain (multiply) of the output of the FET stage and feed it into a clipping function. clamp() will give you a hard clip.
1 Like
I just had a thought. I wonder if the dirt control is a feedback loop. Maybe back to the gate or possibly the drain. That could explain the resonance at high dirt levels. Certainly an area for exploration. I think I still need to build a new model with the signal on the drain. It will necessarily be a VCA rather than just a waveshaper since varying the bias will change the gain. Since the JFET has two distinct areas of operation I may need to use two equations. I downloaded the spec sheet for the JFET used in the Gristleizer so I’d have some parameters to work with.
2 Likes
That would definitely be something to mess with!
2 Likes
Playing video games with my sons in the AM maybe in the afternoon if I feel up to it. BTW David is now the proud owner of a MiniBrute 2. $500 at Guitar Center.
2 Likes
Ahh, the modern family. 
1 Like
Good way to spend some time together with us scattered all over.
1 Like
TG-4 proto.audulus (43.4 KB)
Ok, I whipped something up!
I combined both spline and the uJFET models and added a enveloped sine wave that is threshold triggered. I am reasonably satisfied with the results.
1 Like
Glad to see you’re having some success. I worked on the JFET model today and I’m doing something wrong. When it’s connected, it goes into oscillation for some reason. I think I have the formulas correct, but I’m obviously missing something. Probably a simple error, but I can’t see it at the moment. The voltage at the drain is dependent on the current through the FET, but the current is dependent on the voltage at the drain. It’s not a linear relationship (hence the distortion).
I haven’t given up but I need to rethink my approach.
1 Like
I like the thought process! Does the oscillation sound interesting? I’d be curious to check out the patch regardless for potential glitch sound use.
2 Likes
Unfortunately it oscillates from + to - infinity (well overflow anyway). Probably not a useful signal. I’m sure it’s the result of a logical error on my part.
2 Likes
Thought I was over my bug. Yesterday I felt fine if maybe a bit sore, but today the chills are back. Off to the doc in the AM. David posted his first efforts with the MiniBrute on SoundCloud. I know they’re probably a bit lame by your standards but I know he’d like a listen
2 Likes
Feel well soon, friend.
2 Likes
It occurs to me that the technique I used for making the glitch could be used for making a sub harmonic generator.
3 Likes