Nyquist Aliaser Project

[Eb7+9]

One pretty cool mod I discovered is that you can inject some pitch modulation at the 100K sampling rate pot. I wired that pot with the wiper to the 100K resistor, one leg to the 0.1u cap, and the other leg left unconnected. Then, for the external modulation input, I added a mono jack with the tip going to the unconnected leg of the sampling rate pot and the sleeve to ground.

So far, I've used a CV pedal, a sequencer, and an LFO. All sound great. You can hear the LFO and the sequencer in the clips.

wow, that's brilliant Charlie !! ... I thought maybe somebody was gonna mount it in a wah pedal and show us what this thing can do rocked ... but I didn't think it was CV'able ... holy !! 

(indeed - the Dirty-Love bass-LFO riff put a smile on ma face ...)

thanx for all the feedback y'all - great looking box Clay ! (how you do that ??) ...

---

... back to simming that S/H filter ... 

[moosapotamus]

Thanks, jc! Ultimately, instead of or in addition to the external CV input jack, I would like to have an LFO and an envelope follower both integral to the circuit. So, I'm on the hunt... Actually haven't had any luck at all with the envelope followers I've tried. But, something tells me it's not the fault of the circuits I've tried as much as it might be how I'm trying to tie them in to the clock. Would love to hear other's ideas about how an EF could be implemented here.

Hmm... interesting.

I notice that it's the same signal as my stutter pedal but with the LFO kicked way up. Serial switched signal with a JFET, a 0.001 to smooth out the switch transients.

Hmmm, indeed... that is interesting. I remember spending some t-t-t-time with your s-s-s-stutter p-p-p-pedal a while back.

I wonder if adding a wet/dry control would make it useful for a wider variety of styles.

I actually recorded a clip of bass with some dry blended in and it didn't sound as good (to me) as the straight clips. As I've been tinkering with it, it also seems that there is already the capability to kind of get a wet/dry blend sound out of it depending on the combined interaction between how you set the sampling rate and the 10K trim in the clock section.


~ Charlie

[moosapotamus]

Putfile is out. The clips are here now...
http://users.adelphia.net/~cwbarth/Nyquist/
... should be mac friendly, too.

~ Charlie

[nordine]

jc., i have to tell you: THIS IS THE DIY PROJECT I WAS EXPECTING TO SEE FROM LONG TIME AGO

it's so simple and effective, it's hard to believe... it's like when you're a kid, get your chritsmas present, and go to bed planning to sleep about 2 hours, to go again and play with that great toy... (i know it may sound "a bit too much", but is how i feel it right now)

i like the stripped down version, certainly moosapotamus CV and LFO modulations are inventive, but the raw sound of an aliased bass guitar is unpayable... instant mario tunes, even then, classic ARCADE tunes

man, you made my summer (we're in summer down here)
thanks a lot!!

[soggybag]

I like the sound clips. Especialy the LFO and Sequencer clips. I can hear from these that an envelope could be great addition.

I just drew a layout for this one. Now I need to find the time to build it!

[moosapotamus]

I can hear from these that an envelope could be great addition.

Yes! Any suggestions?

~ Charlie

[soggybag]

I have been working at adding an envelope follower to the Frobnicator for a few months now with little results. The envelope follower is stumping me. I got the best results building the envelope from the Dr Quack. I put this project on the shelf for the time being, maybe I should dust it off again...

[moosapotamus]

Yeah... it's stumping me on this one, too. How & where were you linking the Dr.Q CV with the Frob?

I started off ultra simple, a transistor LED driver. Removed the LED and wired directly to the mod spot on the Aliaser... nothing.

Then measured the output of the LFO I used. That was peaking at around 3Vdc. So, I tried just tapping off of the input signal and going to an EQ pedal that had to be putting out more than 5Vdc. I wired that to the same spot where I hooked the LFO, sequencer, CV pedal... and still got no action.

Next try... Both the CV and gate outputs from my sequencer worked. So, I breadboarded Ray Wilson's Audio Signal to Gate Converter, just the gate part.
http://www.musicfromouterspace.com/analogsynth/signaltogate.html
Still nuthin'.
(But, that's a pretty damn cool circuit, anyway. Gonna box it up and play bass thru my ProOne )

So, maybe I'll check out the Dr.Quack, too. The search continues...

Thanks!
~ Charlie

[Eb7+9]

I can hear from these that an envelope could be great addition.

Yes! Any suggestions?

~ Charlie

Maybe the way to go is by CV'ing the bias ref of the oscillator ...

the way I have it, the circuit providing a constant low-Z voltage reference on the oscillator block is made up of a 10k pot, NPN follower device and 4k7 emitter bias resistor ... it produces a voltage source of at most 50 (AC)ohm drive assuming a (conservative) BetaAC figure of 100 for the device ...

if you can modulate the voltage at the Base of that NPN ref device you'll modulate the oscillator frequency while retaining the low-Z drive of the ref circuit ... if you tried modulating the voltage after the buffer you'd have a rougher time doing passively as I'm about to suggest ...

for a side-chain signal gain circuit use what you feel like ... then follow it by two half-wave rectifier circuits, creating a positive charging and negative charging circuit ... both with their own diode and cap (diodes and caps opposite in polarity) ... this gives a crude first order time averaging of the signal envelope, in two versions - one producing a positive base-line drift, the other negative ...

from either one of the charging caps switch in a resistor (try 10k maybe more/less) to link the hot side of either cap to the sweeper on the 10k bias trimmer ... this should modulate the bias reference enough to create a deviation in frequency ... if you want more deviation just increase the gain of the side-chain amplifier ... switching between up/dn modes will give you the same but opposite amount of DC deviation on the sweeper node but how it translates in frequency shift I'm not sure ... as long as your side-chain gain can be adjusted it should give you the range you desire ... if that doesn't work too well try replacing the 10k bias trimpot by 50k or 100k ...

perhaps crude but it should do it, lemme know ... cheers

[kissack101]

Apologies if I sound like the "new-guy" that I am, but any chance of a parts list or even better, a perf board layout? My band is going into the studio in 10 days time and i'd love to have something like this for that, don't quite trust my abilities enough yet to just dive in on my own. Any help would be much appreciated...

Adam.

[hank reynolds 3rd]

it's at the beginning of the thread
http://www.lynx.net/~jc/NyquistAliaserSchem.gif
http://www.lynx.net/~jc/NyquistAliaserStrip.gif

Sam

[theblackman]

there was an effect similar to this in the old dod tec4 multi effects, apparently designed with some dude from nine inch nails. the thing I liked about it was with high pitches you could get this bassy square wave overtone happening. it also had a random setting which I guess was some kind of pseudo random sequencing.

[moosapotamus]

Maybe the way to go is by CV'ing the bias ref of the oscillator ...

Well, I didn't think to try that. Might be dam cool if it works. Could maybe be able to combine the envelope control with an external CV mod source (pedal, sequencer, LFO...) at the same time.

the way I have it, the circuit providing a constant low-Z voltage reference on the oscillator block is made up of a 10k pot, NPN follower device and 4k7 emitter bias resistor

Right, I think it sounds to me like that 10K pot is (technically speaking) more of a "wackyness" control. In fact, near the mellow side of it's rotation, it starts to sound to me like it's just blending in some of the clean input signal.

I thought it sounded like the 100K sampling rate pot is what varies the frequency/pitch. I guess that's why I went after the sampling rate pot as the point for inserting some external modulation. But, I'm totally game to give it a shot at the 10K bias pot.

for a side-chain signal gain circuit use what you feel like ... then follow it by two half-wave rectifier circuits, creating a positive charging and negative charging circuit ... both with their own diode and cap (diodes and caps opposite in polarity) ... from either one of the charging caps switch in a resistor (try 10k maybe more/less) to link the hot side of either cap to the sweeper on the 10k bias trimmer ...

So, for the half-wave rectifiers, is that basically all they would be... each just a cap and a diode in series, but with their polarities oriented opposite? That sounds really simple. Maybe I need to draw and post it to see if I've got it right.

perhaps crude but it should do it, lemme know ... cheers

Absolutely! Thanks again, jc!

~ Charlie

[Eb7+9]

ooops ... just realized a glitch in my thinking there - I was overlooking the fact the bias trimmer wouldn't allow those caps to stay at ground potential when at rest (silly me) ... looks like you'd need something more like the envelope detector from the mutron-III, one that maintains a non-zero DC at rest and then goes up or dn from there with envelope ...

[Mark Hammer]

Kind of late joining this one.

The circuit is delightfully simple, and effective.  The general outcome is to "misrepresent" the entire frequency spectrum.  Do you think there would be any advantage or disadvantage to using some sort of simple frequency-splitting to run ONLY the upper or lower range through this, and have what's left over blended clean at a mixing stage?

Understand that this is coming from a guy who is obstinate about being able to hear a note underneath everything, if not all the time, then most of the time.

[moosapotamus]

I think there are some settings that already sound like some clean signal is being blended in. But, the idea of freq-splitting is interesting. I have an adjustable crossover I could try it out with.

I also (finally) got some envelope control action happening.
Sound clips... http://users.adelphia.net/~cwbarth/Nyquist/
Scheme... http://users.adelphia.net/~cwbarth/Nyquist/NyquistENV.gif

Kinda freequie soundin'.
One thing about adding the envelope follower, tho, is that the gain control seems a bit less effective at nulling out the clock feedthrough. Anyone have any thoughts about why that might be?

~ Charlie

[Doug_H]

I'm late to the party...

Charlie, that first clip with the guitar (no env control or etc) sounds like a clavinet. Cool!

Doug

[bwanasonic]

Not sure how I missed this thread first time round, as it certainly seems right up my alley, but Charlie's clips seem to be unavailable. Anyone have clips? I have quite a backlog of *things to build*, as I have been more focused performance / recording, and getting the most out of my existing pedal board. This might be the project to get me to clean my work area and settle down for some prolonged soldering sessions!

Thanks JC , as well as Charlie and Clay.

Kerry M

[lovric]

just try:
http://users.adelphia.net/~cwbarth

[Arn C.]

Is there anyway we  could hear a vocal(voice) sound file?  Very curious what someones voice would sound like through this....

Thanks!
Arn C.