Octave transvering puzzzzzzle!

[mike page]

Heres a little something which might sound cool:



The puzzle is to make it so you only get one duty cycle per octave pulse like so:



this example transverses 4 octaves but that is not a critical factor...

I think I have a possible solution but would like to see what you fine folk come up with

[synaesthesia]

Not sure what you are looking for. Is it some effect similar to an arpeggio going up and down through the octaves? Then your solution is good. I would just feed the three selects to the highest three outputs of the 4016 so you can hear more than a few cycles of each octave frequency.

[mike page]

Hey, im after the output in the second drawing. So rather than 16 small pulses follow by 8 twice as big pulses followed by 4 twice as big pulses ... etc ... I was after a configuration that give 1 small pulse, followed by 1 twice as big, followed by 1 twice as big then back down again....

[mike page]

And I think Im kind of after a sort of timbral thingy with it rather than arping.
But of course these things are cool and useful to run at audio or event rates!

[blue hell]

Maybe have a look at rate multipliers like http://www.ti.com/general/docs/lit/getliterature.tsp?genericPartNumber=cd4089b&fileType=pdf or [ http://www.ti.com/product/CD4089B ] for the top level view.

[phobos]

a nice, another puzzle.
Here's something that probably won't completely work, but it's a start.

note: U4 is a priority encoder

---

Octave transvering puzzle v2_1.gif
Description:
Filesize:	59.54 KB
Viewed:	671 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[mike page]

This is what I got, but I cant tell if it works or not!

[synaesthesia]

Haven't looked at the output with a scope, but this is how it sounds.

---

Octave transvering puzzzzzzle!.mp3
Description:
Filesize:	482.05 KB
Viewed:	1530 Time(s)
▶ Play    ⇩ Download

---

[mike page]

Ha! Cool! Is that the circuit I just posted?

I guess its easy for my circuit to go out of sync so it outputs something else. Therefore, proding the reset on the 2nd part of the 4520 should put out different pulse patterns.

And also might get some freaky peaks adding the ol' steppy triangle XOR trick to it...

[phobos]

I did a quick simulation with digital works and it works about 50%. I'll have to draw a timing diagram to see what exactly is happening.
Or maybe someone can use some software that does that. I was thinking about the reset too but sofar it doesn't actually seem to
matter. I think it just changes the phase compared to something else running from the same clock, but the pattern stays the same.

[mike page]

Any recommendations for software to generating timing diagrams?
I tried doing it on paper but it got confusing pretty quickly!

[phobos]

I don't know of any software but I did write down somewhat of a truth table. By the looks of it starts to repeat after a couple of cycles.

[mike page]

once again wow, thanks alot! I see what you mean about 50% working, the principal is sort of there but something is wrong. I'll keep staring at it and try and figure out if it is indeed the right approach...

[phobos]

When I initially tried to solve it I first drew up a timing diagram to see which divisions need to be selected and when.
As you can see there is a bit of a problem which is probably releated to why your version doesn't fully work.
This is also the reason why I opted for a 4029 to do the up/down counting.

I did make a mistake in the circuit diagram: there needs to be an inverter between the carry out of the 4029 and the CLK in of the 4013.

if you do that it counts:
0000
up
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
1100
1101
1110
1111

down
1110
1101
1100
1011
1010
1001
1000
0111
0110
0101
0100
0011
0010
0001
0000

---

Octave transvering puzzle - timing A.gif
Description:
Filesize:	34.72 KB
Viewed:	393 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

---

Octave transvering puzzle v2_2.gif
Description:
Filesize:	60.06 KB
Viewed:	421 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[mike page]

Ah nice. Thanks for explaining the way you approached it. Good idea drawing up which divisions are needed etc v smart!

[phobos]

That timing diagram is also how I came up with using a priority encoder.
While I was looked at it I noticed something:
/2 needs to be selected when /4 is high
/4 needs to be selected when /8 is high
/8 needs to be selected when /16 is high
etc.
I assumed CLK (inverted) can be selected when /2 is high but I didn't really check that, hence why I am not sure if the circuit
fully works or not, I also didn't check yet if it still works correct when counting down.

anyway,
/2 is connected to input 1 of the mux so is selected with CBA = 001
/4 is connected to input 2 of the mux so is selected with CBA = 010
/8 is connected to input 3 of the mux so is selected with CBA = 011

looking at the counter that means that when its output is 001x (Q4 = 0, Q3 = 0, Q2 = 1, Q1 = don't care)
the CBA control inputs of the mux need to be 001.
when its output is 01xx CBA inputs need to be 010
when its output is 1xxx CBA inputs need to be 011
and that's exactly what the priority encoder does.

I think it is also possible to chain 2 4029 counters which would give another 4 bits that can be connected to the priority encoder.
Q2 of the encoder can then be connected to input C of the mux.

[mike page]

ace! You are the king of logic!