Posted: Fri Feb 10, 2006 5:10 am Post subject:
An idea for a new kind of DIY VCO
I don't know whether this is a new idea or not (probably not) but it's been brewing in the back of my mind for a couple of weeks now and I decided it's time to see what you guys think.
Okay, basically it's a low res hand made DAC Oscillator. A set of three 4066 switches short out (or not) sections of a resistor ladder whose values are chosen to make a unique voltage at the output for every combination of switches. ( the example uses 500R, 1K & 2K which means there are 8 possible resistor values which create 8 different voltage dividers (output voltages) depending on which switches are closed)
Now if each switch is turned on and off over time using a sequential counter like a 4017 and switches with diodes then each switch becomes the equivalent of a bit in a 3bit DAC.
If the 4017 had 8 outputs it's clock would need to run at 8 * the fundamental frequency of the wave output. This could possibly be a CV clock taken up by 3 octaves?
It's not elegant, it's not chip resourceful, but I think it could work, and might be a lot of fun. The output would be a modified square wave with (8* 64 possible timbre structures. This could be expanded easily to (16*16) 256 possible timbre structures.
I know it's dependant on how fast you can run the 4066 switches, that may mean only lower registers would work. Filtering/buffering of the output may be an issue.
The really cool idea is what happens when you replace the diodes and switches (they would have to be DILs for cost reasons) with separate binary counters and arrays of AND gates. Depending on the speed of the counters, you would get a constantly morphing waveshape. Maybe they could be clocked by something else again? Maybe they could be speed controlled by 3 more CVs?!?!
Anyway here's the first pic of the basic idea.
I especially would like to hear what Scott and Jan reckon could be possible issues with the basic idea.
Looks interesting, but why not just run a step sequencer at audio rates? Maybe you don't want any pots in the circuit. _________________ --Howard
my music and other stuff
Looking and what you are trying to actrually build here, going stepseq seems like a smart idea. You could integrate both schemes? _________________ A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"
Posted: Fri Feb 10, 2006 10:02 am Post subject:
Re: An idea for a new kind of DIY VCO
Uncle Krunkus wrote:
Depending on the speed of the counters, you would get a constantly morphing waveshape. Maybe they could be clocked by something else again? Maybe they could be speed controlled by 3 more CVs?!?!.
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 97
Posted: Fri Feb 10, 2006 11:19 am Post subject:
I was trying to figure out why you wouldn't just use a step sequencer until I noticed the last post there pointing out:
Quote:
Depending on the speed of the counters, you would get a constantly morphing waveshape. Maybe they could be clocked by something else again? Maybe they could be speed controlled by 3 more CVs?!?!.
Ah! Looks to me like something you should try out! This could be used to modulate VCO's at audio and subaudio rates as well, which would be very interesting, I think.
I patched it out for the G2, and it will run in the G2 demo software - it runs at 25% usage at my PC. The G2 is great for prototyping ideas.
I may have misunderstood the idea, but it is certainly capable of weird sounds as it is. Anyway, its just the basics, and there is some aliasing going on that you may, or may not, want to filter out.
Yeah, that last bit was actually the main point. I was just making sure I didn't skip the foundations. Imagine Mosc having a little gnome hanging off each pot, re-adjusting it automatically, maybe from one value to the next each wave out, and back again, or randomly, or a combination of both.
The thing I really like is the potential to produce any wave shape at all.
Maybe it could be useful as a completely separate type of VCO. If not then it should still be on par with another type of WSG, which would be cool too. In fact hanging something like the WSG filter section off the end could be all it needs to smooth out the aliasing as much as you like.
I need to finish soldering up the Roboride sequencer/driver board so that I can get some real-estate back on my breadboard, then I'm gonna get this thing squeaking ASAP.
Keep throwing back any ideas which come up about the actual implementation of it.
It's really great being able to bounce ideas off you guys before I even get the chips out! _________________ What makes a space ours, is what we put there, and what we do there.
Hey Jan,
I love the look of that screen dump. It's gonna take me a while to work out what's going on (and whether that's what I mean) though.
I'll have another coffee (or 4) and get back to it! _________________ What makes a space ours, is what we put there, and what we do there.
Its just simple sweep of the "overall spd" knob from its lowest value to 24. Meanwhile the righthand side LFO's play up/down sequences on the four VCO's. It's not filtered or anything.
Cool!
Pretty unmusical, but also very f@!?%d up!
Just the kind of thing I love!
I imagine that actually putting it together using discrete components would tend to calm it down a bit, yeah? Maybe that would be a good thing. Maybe not.
What are the chances you could do another run, leaving the fundamental frequency constant (~440Hz?) and just slowly modding the waveform oscillators?
Thanks heaps for your help Jan _________________ What makes a space ours, is what we put there, and what we do there.
Uncle K, why don't you download the free G2 demo from the Clavia site and try it yourself. It's a great too to prototype circuit ideas, like Jan said. _________________ --Howard
my music and other stuff
Main oscillator running at 8 * 440 Hz = 3.52 kHz, the three sub osc's tuned like :
f (weight = 4),
f + 5 (weight = 2),
f + 11 (weight = 1)
and starting with f at sub audio (a few Hz or so) sweeping up into the audio ranges (f ~ 330 Hz) at about 12 s into the track - later falling down again.
I added a filter at the output: LP, 12 dB/oct, f0 = 3.73 kHz, Q = 0.81.
Don't think it will get really smooth when building it in hardware, the circuit is pretty digital in nature, so it will have digital artefacts and when the oscillators have no harmonic relation it will be pretty harsh :-)
I've already got it Mosc. In a situation like this I'd prefer to keep the iron hot and leave the theory to the experts. I will give it a go soon though, just need to make some space on my breadboard! _________________ What makes a space ours, is what we put there, and what we do there.
BTW,
Love the sample Jan.
If I use all four switches in a 4066, I'd be able to get 15 voltage levels out.
I've got four 8 way DIP switches, man this is gonna be one nasty looking breadboard! _________________ What makes a space ours, is what we put there, and what we do there.
I've already got it Mosc. In a situation like this I'd prefer to keep the iron hot and leave the theory to the experts. I will give it a go soon though, just need to make some space on my breadboard! :)
How do you do "click me to explode me mode".
I was thinking the only way of posting anything with any detail in it would be to zip it. _________________ What makes a space ours, is what we put there, and what we do there.
By putting the image on a webserver somewhere and using a link to the image in the message (using [img] tags) instead of an attachment. Such images will be auto-sized by the board's software. The thing is you can't delete the image later on, or move it.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum
Forum index » DIY Hardware and Software
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
64 possible timbre structures. This could be expanded easily to (16*16) 256 possible timbre structures.
