Saturday, May 18, 2019

A Rolling Stone Gathers no CMOS--the 4069UB Chip

Time to get back into CMOS stuff, this time the 4069UB (unbuffered) IC used in an analog realm....




What got me started down this particular AudioDIY rabbit hole was an Electro-music forum user "Hammer's" 4069 VCF design--see the post here.  He also has a cool but unassuming Youtube vid of his 4069UB LP filter here. Good work! Even more intriguing: he's from Russia and writes that his creation "screams much wilder" with a Russian version of the 4069. Indeed--got to check that out.

How can something simple sound so--gnarly?  Must be those 4069s?

I dug in and found a really good Hackaday write up about 4069UB's used in analog audio work here and here. Hammer contacted me about his design, and suggests another good source about 4069s used in audio, here.

The 4069UB chip, in this get up, is configured somewhat like an inverting op amp. There's a negative feedback loop needed to do useful things--again the Hackaday article lays out how it works; I won't repeat that here.

OK, then, why bother using a 4069 at all (besides getting the equivalent of 6 op amp stages for 50 cents US?)  Why not use a damn op amp? Bing! distortion!  From messing around on the bench: them 4069UB's love to distort.  You crank down the positive rail--distortion. Crank up the incoming audio--what was that then? Um, distortion. I figure you can make a 4069UB **NOT** distort but for squeaky clean applications I'll use an op amp; when I want some grit I might consider a 4069UB. Best of all: to my ears, the 4069's distortion is a bit tube-amp-sounding and OK, sometimes I like that.


4069UB buffer! You need a 4069 with UB in the name for any of this to work, must be the un-buffered IC.

Hours of fun! Experiment with different P/P audio input voltages and DC offsets into a 4069 buffer to hear 50 shades of grunge. You can read more AudioDIwhY posts about DC offsets here and here.


Scope a dope! Purple is original ramp signal from my Siglent 1025 (Siggy plays Guitar) Yellow is the buffered signal at output.  Yes, it inverts....

Wiring up a cap in the feedback loop has some of the same vibe as an op amp active filter:

100K pot is for cutoff frequency....

And running it through more buffers might make the sound gnarlier!


The 4069 punishes that ramp signal!
Try more 4069 inverter stages for a bit more grunge during cutoff sweep. Xtra credit: put 2 of these entire filter fragments in series. Yeh!! 

OK this weekend coming up I will see if I can get this EM 4069UB based LPF to work. I laid out a prototype PCB capturing the design (since I hate breadboarding or stripboarding. If the PCB doesn't work I'll just pitch it). Got it fab'd at my usual spot.

I am wondering how this will work. Hammer feeds positive CV into N channel FETs (5457's); N channel JFETS are opened up by having their gates see voltages below source and drain right?  It's a safe bet the filter works at an elevated DC offset; all the analog 4069 designs I've seen so far do. The key: 4069UB, mind the offset.

UPDATE: I got hammer's Low Pass Filter to work! It sounds good n crazy! Details here.

That's what's fun about analog, you get to ponder these things. Nonetheless, I'll say it again: don't breathe the fumes. See you next time.

No comments:

Post a Comment

EFM/KORG770 VCF--BUILD: yes, SOUND: no

Ahoy!  This time, I wanted to refine the quick prototyping idea discussed midway through this previous post : Minimal breadboarding--I hate ...