Repulsor Sound from 2 chips

Uncle Krunkus · Posted: Thu Jun 18, 2009 3:47 am Post subject: Repulsor Sound from 2 chips

Yesterday, I was researching a circuit to power the blue LED electrostatic repulsors which I plan on incorporating into a scratchbuilt model of a hovercraft ship, inspired by the Nebuchadnezzar from the Matrix movies.

It's just 4 * 40106 free running oscillators, approx 15hz each with the outputs sent to 4 * XOR gates so that the gates are run from oscillators - 1&2, 2&3, 3&4, 4&1
With a LED on each output, held close together, they look like a semi random pulsing blue light which is just right.
Of course I'll need at least 4 LEDs in each of 8+ replusor drives, so the whole idea will be scaled up,.........
But,..........

I decided to have a listen to what it sounded like, and I've got to say I was very impressed.
Instead of simply mixing the outputs, I added them together in a simple 4 bit R/2R network.
Then,..... in an effort to "slide" the transition of each bit inside the R/2R, I put a 100nF cap around each of the R segments. (I'm using 10K/20K)

And the result?
A sound which sounds exactly the way I wanted it to look!!! Shocked

In fact, if I can somehow add audio transducers to the model, I think I will, as it would be the icing on the cake to "hear" the blue glow of these repulsors as well as see them.

Here's an mp3. No effects. Not even an op-amp buffer! DC coupled straight off the MSB!! Very Happy

Repulsor.mp3

Description:

Filesize:

141.53 KB

Duration:

0:09

Viewed:

2796 Time(s)

RF · Posted: Thu Jun 18, 2009 4:01 am Post subject:

Sweet, Krunkus!
I agree - that's just about perfect!

bruce

Uncle Krunkus · Posted: Thu Jun 18, 2009 4:04 am Post subject:

I just tried replacing the 100nF caps with 470nF, and I think it's even better. More rounded bottom end.

The way the caps interact with the fluctuating bits in the R/2R is really nice. I can hear occasional flanging / phasing of the XOR PWM effect, but it's more than that. It's like the caps don't know if they're charging or discharging, cos they're bleeding into each other, and in and out of the XORs.
It almost sounds like it has a nice resonant filter built into it.
It would be debatable as to how "musical" this sound could be made.
I s'pose it just depends on how much you control the oscillators, which I'm not doing at all. The only reason this sounds as complex as it does is the tolerance on those 40106 oscillator caps and resistors.

If anyone wants a schem, I can knock one up pretty quick.

RF · Posted: Thu Jun 18, 2009 5:08 am Post subject:

If you are willing to draw it - do it! I may or may not build it, but it would be interesting to see the circuit...

Thanks

bruce

Uncle Krunkus · Posted: Thu Jun 18, 2009 6:28 am Post subject:

Here ya go,

Repulsor.gif

Description:

Filesize:

16.4 KB

Viewed:

891 Time(s)

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

RF · Posted: Thu Jun 18, 2009 6:47 am Post subject:

TU Uncle Krunkus!

It would be interesting to play with this at higher frequencies - or for your application, set them initially a tad higher and then controlling the frequency slightly by the v+ on the 40106.
You know - that old here come the "squidies" - "power up the repulsors and get out of here!"... thing.

bruce

slacker · Posted: Thu Jun 18, 2009 10:36 am Post subject:

Looks interesting I'll give that a try.

droffset · Posted: Fri Jun 19, 2009 2:32 pm Post subject:

That's cool, I need one.

Uncle Krunkus · Posted: Sun Jun 21, 2009 4:35 am Post subject:

By adding these four 240K resistors, I've managed to make each oscillator "bleed" some current into the next one's capacitor. Thereby creating this kind of cyclic syncing effect which makes the waveform more regular, but with unpredictable shifts in timbre.
Oh, hold on, (I'm listening to it through headphones right now) and it seems like after about 10mins it's finally dropped into a locked state.
Okay, well I didn't see a locked state when the resistors were up around 270K. That might be a better value.
I've also noticed that the stability is affected by what you do with the two unused buffers. I've currently got them figure 8 style with two caps. I'll fix that bit of the pic later.

Repulsor-2.gif

Description:

Filesize:

18.84 KB

Viewed:

832 Time(s)

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

synthmonger · Posted: Fri Jul 10, 2009 10:13 pm Post subject:

Did you try replacing the resistors for diodes in either direction? Or R+D combo? I use those for a sync effect with a whacky vco im working on as one of the 40106 inverters as the core.