Patch 22

synaesthesia

Playing around with filters lately, I had a great sounding Op-Amp filter on my breadboard and wondered if the same could be built with CMOS inverters. Indeed, there are some circuits on the web already. So I built one of these using three 4049 inverters and it works just great. Of course, there needs to be some musical input, so I put together a simple pattern generator from a 4060 counter as well. The outputs of the 4060 counter are mixed via 22K resistors and produce a voltage sequence that drives a VCO built from the other three 4049 inverters. This produces several (2^6=64) different melodies, depending on the counter frequencies selected via the DIP switch. The two 4060 counter outputs with the highest frequencies are reserved to modulate the filter via a DIY vactrol.

I built several vactrols with the LEDs and LDRs I had at hand. The best working ones do not use super bright LEDs, but normal diffused red ones. Rather old LEDs, that are not as bright as modern ones, tend to work better. The LDR is a cheap standard type like the GL5516 or GL5528, that you can get in quantities at ebay. It is very important to seal them together in a light-tight package. I used a 15mm long piece of heat-shrink tube for the package. Because your mileage may vary considerably with these DIY vactrols, I put two pots into the circuit (RV2 and RV3) that can be used to compensate the variations. Nevertheless, you might need to experiment a bit here.

Looking at the almost complete schematic, I noticed that most of the resistors used were 22K. Even the capacitors had the 2.2 signature. So just for the fun of it I aimed at making all of them 22K. I could hardly resist to use two 22K resistors in series instead of R18 with 47K, but that would have violated my minimum-parts doctrine. Smile

The name of the circuit was obvious then: Patch 22.

In the recording I start with a slow tempo and fixed melody. Then I change the resonance of the filter from minimum to maximum and back to a middle setting. After increasing the tempo, I changed the DIP switch settings at A2 to A8 for the melody. Because of the slow tempo most of them are not complete, and so sound rather similar. At the end I changed the filter settings at A0 and A1, and finally turned the filter modulation off completely for the last few seconds.

Lots of fun and room for experimentation with just two CMOS chips.

DIY Vactrol.jpg

Description:

Filesize:

329.71 KB

Viewed:

435 Time(s)

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

Patch 22.PNG

Description:

Filesize:

64.76 KB

Viewed:

601 Time(s)

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

Patch 22.mp3

Description:

Download

Filename:

Patch 22.mp3

Filesize:

2.17 MB

Downloaded:

899 Time(s)

piedwagtail · Posted: Sun Nov 29, 2015 10:51 am Post subject:

Super!

I bought and played with a lot of those LDRs.
I'm not sure what you found to be the best resistance range vs brightness.
fyi
My findings with a modern yellow led were lowest resistance for

and highest resistance being

GL5528 and GL5537 were similiar to 1k then changed linearity with further decrease in current

Ebay codes were

R

richardc64 · Posted: Mon Nov 30, 2015 1:12 pm Post subject:

Here's another CMOS filter that might be fun.

I used a 4069, but it might work with 4049. Vary the caps smaller or larger; the 470k has some control over resonance; for VC Freq, slap any old NPN transistor in parallel with the pot. No finagling around with LDRs or LEDs required.

cmos_bpf.gif

Description:

Sorry, no 22ks.

Filesize:

5.33 KB

Viewed:

9401 Time(s)

synaesthesia · Posted: Tue Dec 01, 2015 11:51 am Post subject:

@piedwagtail
In my experiments I arrived at the conclusion that it is more the LED than the LDR that is critical.
I got the best range from using 3mm red diffused LEDs. Preferably old ones that really need 10mA. Mine must be around 20 years old. Smile

Nevertheless, I just measured the LDRs that I have and the table below is what I got. Even when covered by one layer of black heat-shrink tube, the resistance grew to >20M for all LDRs when I switched the workbench light off.

LDRs.PNG

Description:

Filesize:

11.86 KB

Viewed:

408 Time(s)

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

synaesthesia · Posted: Tue Dec 01, 2015 12:14 pm Post subject:

@richardc64

Ah, saw that schematic before with an op-amp, but didn't try it with a DIY vactrol. This could be an improvement, because it needs only one inverter. Thanks!

The "great sounding filter" that I played around with can be found here: https://www.youtube.com/watch?v=yZSkNASlMjM
The link to the schematics is in the description. I modified the circuit for 5V, changed the input to an LDR controlled oscillator, and changed the PWM modulation section so that it only requires two LM324 chips (8 op-amps) in total. This was meant to become a Theremin-like instrument. It works great and sounds exactly like the one in the video. I didn't post it, because it is purely analog and doesn't qualify as a Lunetta.

There-Wah.PNG

Description:

Filesize:

32.68 KB

Viewed:

496 Time(s)

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

synaesthesia · Posted: Wed Dec 02, 2015 12:24 pm Post subject: