Pseudo-Lunetta Noise

dk · Joined: Feb 12, 2019 Posts: 115 Location: Europe

Hi all,

Just sharing something I'm about to put together. It almost stays CMOS... I did cheat and use a noise transistor, as using a shift register (LFSR) would have been equally parts intensive, taken up a lot more space, and less random. I've breadboarded everything except the 4017/4094 part and so far, so good.

The white, low ("sort of pink"), and granular were somewhat inspired by Ray Wilson's Noise Cornucopia. Instead of using comparators for the "grainy" noise, I just made a schmitt trigger with adjustable hysteresis to change the density.

The bottom part was inspired by this. I don't have the panel space to bring all the outputs to the panel and didn't want to divide the clock by 8, but theoretically you could hook up a switch (or a mux? Twisted Evil

) between the 4017 and change how often the 4094 gets strobed. Likewise, you could hardwire an r2r or the like to the outputs of the 4094 and get random stepped CV.

None of this is re-inventing the wheel, but sharing is caring! Smile

Pseudo-Lunetta Noise.png

Description:

Filesize:

260.43 KB

Viewed:

267 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

PHOBoS · Posted: Mon Apr 27, 2020 9:37 am Post subject:

nice use of the 4069 for the density control.

And yes shiftregisters work great to get some randomness. You don't even have to use an actual noise source.
Even just using a high frequency oscillator seems to be pretty random. Would expect a repeating pattern but it's
probably due to small fluctuations in frequency that it seems to become random.

dk · Joined: Feb 12, 2019 Posts: 115 Location: Europe

A quick update and a quick question:

I've finally built this out to put in my machine. The 4017 needed to be connected differently in order to keep the shift register from running two steps each time the strobe was enabled. I've attached the updated schematic below.

The noise generator I had working on breadboard seems to be acting up, though. It sort of sounds like self oscillation, with the White and Low outputs significantly quieter and a distinct whine mixed in with the noise. The digital (Grainy) has a tiny bit of said whine when there is no hysteresis (pot at minimum), otherwise it sounds fine. I can also slightly hear changes on the density knob on both the White and Low.

I don't have a ferrite bead to try, as the local shop only sells them in large quantities and by order. Is there another way to go about taming this? Could proximity to the 4069 be part of the problem? I left a ground trace between the transistor section and the 4069, but they are physically a lot closer than when I breadboarded it.

EDIT: I tried pulling out the 4069 and listening to the output of the noise generator directly from the 10uF cap. No dice... the whine is still there. Also, I'm using the exact same parts in the same positions as when I had everything breadboarded and working.

Pseudo-Lunetta Noise.png

Description:

Updated Schematic

Filesize:

160.09 KB

Viewed:

182 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.