Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Tue Jun 23, 2015 1:38 pm Post subject:
I never tested that CV control so can't say if it will work but it will probably do something.
You could also try modulating with squarewaves by adding a diode to the inputs of the NAND gates.
(use a NANDulator to control another NANDulator maybe even put them in a feedback loop) _________________ "My perf, it's full of holes!" http://phobos.000space.com/ SoundCloudBandCampMixCloudStickney SynthyardsCaptain ColliderTwitchYouTube
Posted: Wed Sep 23, 2015 10:04 pm Post subject:
NANDulator
Replace the 100K pots going to the diode pairs with a CD4097.Replace the the 100K pots with a CD4051 resistor network. Gives controlled steps to the Nandulation. Build a dual pot out of the CD4097's and a single pot out of the 4051's. Sorry I don't have a schematic to post, I'll will get to it. Any one recommend a schematic program for Win7?.
Apologizes for the grammar and the English, God knows, parents and teachers tried.
You can certainly do that and it will veer towards the excellent work that Synaesthesia and others continue in other threads.
The complication comes in playing those decisions.
One pertinent point is there is a certain delay in changes in the NANDulator, which can be noticeable and surprising.
I should qualify that these were live improvised manipulations of the two NANDulators and that the complex interactions of the last part of Study 2 were fast touching of banana plugs to jack shoulders. It's a physically demanding apparatus!
The multiple NANDulator is very much an analogue computer.
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Sat Sep 26, 2015 2:49 pm Post subject:
piedwagtail wrote:
NANDulators just laugh at oscilloscopes and spit in the face of data acquisition systems, in fact beyond your ears they're very elusive.
I really appreciate all the work you're putting into it. How did you modulate it btw,. connected directly through some diodes
or did you use vactrols ?
AuDioMiRage wrote:
Replace the 100K pots going to the diode pairs with a CD4097.Replace the the 100K pots with a CD4051 resistor network. Gives controlled steps to the Nandulation. Build a dual pot out of the CD4097's and a single pot out of the 4051's
well you could definitely do so something like that but, and I think Robert can attest to it, because of its somewhat chaotic
na(ndula)ture manual control will get you the best results. Although I am wondering, if you would replace it with
a mux and instead of a resistor network use a couple of pots (like a sequencer) it might actually work pretty well.
You'll end up with a ton of pots though. (just imagine replacing every pot with a 4067 and 16 pots ) _________________ "My perf, it's full of holes!" http://phobos.000space.com/ SoundCloudBandCampMixCloudStickney SynthyardsCaptain ColliderTwitchYouTube
A trio of NANDulators; the left one is a timing capacitance model with led indication
The schematic with numbered capacitors
The six uninsulated banana sockets; the top two are the charging and shorting diode outputs ; the bottom four are the jam inputs to the four numbered capacitors as seen above. There is no input to the final NAND but there is a dedicated screw terminal 'out' underneath after a transistor emitter follower.
The component side with the relevant capacitor options for each oscillator; space for film or electrolytic capacitance in accordance with function. The components were soldered and the board cleaned; the pots were then soldered with rosin-free but this could still be cleaned up. One of the drawbacks of white pcbs are the rosin stains are very visible but after cleaning the results are worth it. 5.08mm screw terminals for supply/ground/output.
As a dedicated player NANDulator the bottom four inputs might be replaced with M6 bolts instead of the sockets so that the input modulation can be manually touched to the contact point. Pushing the input into socket was found to be rarely needed in audio frequency, however in slow movement and very low frequency manipulations the socket method has its merits.
This is the setup used for the previous recording; so to answer PHOBoS's question only diodes were used. I found ldrs to be unnecessary. There are differences between using the charging and shorting directions of the diode outputs, though being NANDulators the precise description of the effect is somewhat difficult to quantify.
Please find two further NANDulator studies No. 3 and 4.
3. Is two audio rate NANDulators with output from one only. The second is purely modulating, there is a change from charging to shorting modulation at 1.00 minute.
4. Is three, two audio, one timing NANDulators. Output from one only again. The useful rhythmic variations are coming from the timing device modulating the second which is inturn modulating the output third.
Phobos! Are you still around? I listened to about 1:40 of your 1st posting of the Nandulator and that thing was AMAZING!
The first new Lunetta circuit that I've come up with in a couple of years 'The Drain Oscillator', yesterday, is based around either the 4011 or 4093. I still have to breadboard it to test. (It's a similar idea to my CD4023 based 'Pulse Width Oscillator' from way back in 2009 or so.
I'll be posting more about it to my forum and Faceboob, once the testing begins.
For now though - I'm blown away by the sonic capabilities from the Nandulator!
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Thu Jul 14, 2016 10:30 am Post subject:
hey Ryk! good to see you back
Looking forward to hearing and seeing (in the form of those entertaining videos) your progress on the next deathlehem machine.
yeah the NANDulator has some nice sonic capabilities for such a simple single chip circuit, I was quite surprised by it myself
I just put it in a new box over the weekend so the easter edition is no more. But I did get some nice use out of it when I played
my set during the June Solstice event. However one pot got pushed through the cardboard so I had to rehouse it. One thing I had
been curious about was if it be useful in combination with my mini 5x5 LED scope. So I did a test and they work quite well together
so I put them both in a (rather flimsy) plastic box. I don't have any DC connectors left so at the moment it just has 2 wires sticking
out from the bottom. Also the boards are just ' floating' in there so there is a potential for shorts.
But the new box with both the NANDulator and the mini scope in combination with a pm I got from Robert got me started on the
development of some PCBs. It will have both the NANDulator and a mini 8x8 LED scope and I am planning on making those available
as full kits (PCB + parts). But I will post more about that later.
NANDulator + mini 5x5 LED scope.jpg
Description:
Filesize:
160.41 KB
Viewed:
523 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Thu Jul 14, 2016 11:03 am Post subject:
ok, so I had noticed something odd. The NANDulator was actually sounding a bit plain like a more standard drone box, which was not
how I remembered it. It also showed the patterns on the display as one would expect with 2 squarewave oscillators. So just for fun I
replaced the chip. oh yeah!, THAT'S how its supposed to sound and the patterns are much more interesting now. So I don't know
if the NANDulator itself wreaks havoc on a CMOS chip, it wouldn't surprise me (The NANDulator, devourer of CMOS chips! ), but
I just wanted to mention it in case someone runs into something similar happening. _________________ "My perf, it's full of holes!" http://phobos.000space.com/ SoundCloudBandCampMixCloudStickney SynthyardsCaptain ColliderTwitchYouTube
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Fri Jul 15, 2016 2:55 pm Post subject:
So here's what I have at the moment.
The NANDulator is the original configuration but with an improved buffered line out (inspired by Robert). It has a stereo 1/8" output
connector although the NANDulator itself is of course mono. This makes it easier to connect to stereo gear without having to use stacked
connectors or use a mono to stereo cable. A mono jack can still be used which will just short the right output to GND, this is the
reason there are 2x 470 ohm resistors in series connected to the outputs.
I also added a switch to mute the output and when muted a momentary button can be used to unmute. so you can trigger the sound
for short audio bursts.
For experimentation and extending purposes I added a header connected to pins 4,6,8,9,10,11,12,13 of the CD4093 and the power supply.
This way you could easily connect it to a breadboard and experiment with additional logic, or use it to patch multiple NANDulator together.
There are also some extra headers on the board with room for additional components. Most obvious components would be diodes but you
could also add extra timing capacitors or LDR's or maybe even vactrols to control the frequency of the oscillators. There are also 2 (shorted)
4pin connectors to use as a passive multiple.
The mini 8x8 LED scope is almost the same as the 5x5 version but with an ULN2804 instead of the CD40106 and of course it
has 64 LEDs instead of 25. It uses a standard 8x8 LED matrix.
Here's the schematic and a 3D render of the current PCB design.
edit: changed some of the resistor names in the schematic
NANDulator + mini 8x8 LED scope - 3D PCB render.jpg
Description:
PCB
Filesize:
510.85 KB
Viewed:
468 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
NANDulator v1_2 + mini 8x8 LED Scope.png
Description:
circuit diagram
Filesize:
56.37 KB
Viewed:
856 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Sat Jul 30, 2016 11:09 am Post subject:
I'll see if I can make a video, I don't really have a good camera though and I am not sure how well it shows up on video because the LEDs
are basically multiplexed.
ok, so I just put one together and it works
however,.. when I powered it up and twisted the pots I immediately noticed the same behaviour on the LED scope as I had noticed previously
with the 'faulty' chip: (diagonal) lines but no chaotic patterns.
I replaced the chip with the one I had recently placed in my other NANDulator and that indeed made a difference. So I just did some more
experiments with some other 4093 chips I have and here are my findings.
tested ok are: HEF4093BP (PHILIPS), TC4093BP (TOSHIBA)
tested not ok are: CD4093BE (TEXAS INSTRUMENTS), HCF4093BE (SGS THOMSON).
now it could be a coincidence and I would need more different types to test with but from those 4 chips both the BP ones work and
both the BE ones don't. Robert mentionend that the MC14093BCP (ON SEMICONDUCTOR) also works, but I have not tested that one
myself. However it does have a BCP suffix.
I recorded a short audio example to show the difference between an ok (HEF4093BP) and not ok (CD4017BE) chip. Both start with
pots P1, P3 full clockwise and P2, P4 fully counter clockwise and I twist P2 clockwise. I also made two (blurry) photos but because of the
multiplexing it doesn't show the lines very well. You can see a big difference though.
Basically the difference in the range of sounds produced by a NANDulator that doesn't properly NANDulate and one that does is this:
So now I need to figure out where I can order some different 4093 chips to replace the ones I had ordered for the NANDulator kits.
edit: Just found some and with a bit of luck I might already have them next week
oh and I also discovered a mistake in the PCB design, nothing major but on the silkscreen for the pin header oscillator C has the in- and output
swapped.
NANDulator chiptest - HEF4093BP.jpg
Description:
NANDulator with a HEF4093BP - OK
Filesize:
169.38 KB
Viewed:
443 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
NANDulator chiptest - CD4093BE.jpg
Description:
NANDulator with a CD4093BE - not OK
Filesize:
166.54 KB
Viewed:
399 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
NANDulator chiptest.mp3
Description:
NANDulator chiptest. first section with a HEF4093BP, second section with a CD4093BE
I replaced the chip with the one I had recently placed in my other NANDulator and that indeed made a difference.
I had a similar experience with 4093 ICs in the Moduletta. What I found is that the hysteresis of the Schmitt trigger differs significantly between chips from different manufacturers. Similar result for me: newer chips from TI did not work so well, PHILIPS and SGS were better. Interestingly, however, some very old TI chips worked best.
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Mon Aug 01, 2016 1:30 pm Post subject:
One of the things I did to create the track that I linked to in the previous post is use an LDR. It was connected across the frequency pot. of oscillator
C and positioned above the LED matrix. I did a couple of test with an LDR in combination with the matrix and in some cases it created a feedback loop
where if you moved the LDR the LED pattern would actually follow it .
This might be interesting for future circuits where a LED matrix is controlled by other logic. A binary counter instead of a 4017 would be worth testing,
but also just different outputs from a circuit directly connected to the rows/collumns. I also wonder what is going to happen when I'll test with multiple
LDRs and use them to cross-link NANDulators. Anyway it is nice that the LED scope already adds more to it than just some eye-candy.
Here is the patch I used for the NANDulator that does the rhythmic part.
NANDulator - Technulator patch.gif
Description:
Filesize:
16.34 KB
Viewed:
447 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
So that got me thinking, instead of combining an oscillator with variable frequency and an oscillator
with variable pulsewidth, combine 2 oscillators that have both variable frequency and pulsewidth.
I haven't tested it so I don't know if it works (it should do something) or what useful capacitor values
are.
i have been starting to make a stripboard version of the NANDulator V2 and now i was wondering if (anyone of) you has come up with suitable capacitors for this circuit? if not i guess i'll just socket them in the first build and try with my own ears. i would be happy if (anyone of) you would have a look over the stripboard layout i got so far, in case i made mistakes...
so, this is unverified so far...
thank you.
rev.1.1 - pot I is fixed, as per PHOBoS hints.thanks.
Last edited by grm on Fri Sep 09, 2016 6:43 am; edited 2 times in total
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
Posted: Fri Sep 09, 2016 6:17 am Post subject:
There is something wrong with how potentiometer I is connected on the last oscillator, for the rest it seems ok. I personally wouldn't
put any wire jumpers underneath an IC socket, although sometimes it works depending on the used socket. As for the capacitors,
yeah you just have to experiment with those. I found it a bit too hard to control with so many pots but it does give you a wider range.
Joined: May 18, 2016 Posts: 91 Location: Texas, USA
Audio files: 2
Posted: Fri Oct 21, 2016 4:56 pm Post subject:
Here I am jumping in way late and not having read the whole thread, I just took a glance at the NAND portion of the NANDulator and it's kind of like what I was planning to do with a 3D printing circuit. My concept is very much simplified however. That's because of the trickiness of 3D printing a circuit design.
In fact, I have only printed some test structures to date. The approach does, however, look promising. In a way it's like the old joke: "If all you have is a hammer, everything looks like a nail", heh! Well there are other reasons to 3D print circuits.
This is not really the thread for detailing my plans, just thought I'd stop by and say "Nice circuit!"...
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