KVR Audio

janostman · Post by **janostman** » Sat Sep 13, 2014 10:54 pm

Well, when you figure out how the alpha-juno DCO's work make sure you let me know

I have the Juno-6/60/106 covered but this might be something more.

aciddose · Post by **aciddose** » Sat Sep 13, 2014 11:01 pm

janostman · Post by **janostman** » Sun Sep 14, 2014 10:27 am

itoa wrote:I'm trying to reconstruct the Alpha Juno 2 saw pwm oscillator.

My observations:

- Segments have variable steepness, depending on pulse width. Lower part looks like it's low pass filtered - nonlinear filtering?
- Initial saw is rounded - this could be effect of low shelving or ring modulated sine - but it's not, it ends up as a straight line.
- Maybe 'integrated signal' in DCO is stopped during negative pulses?

If you look at the "holes" in your own waveform you see that they are the sawwave but sign flipped.
aka a Wrap or offset.

You show this yourself but still keep mocking people that comes up with theories in this thread.
What answers do you really want?

antto · Post by **antto** » Sun Sep 14, 2014 10:58 am

janostman wrote: If you look at the "holes" in your own waveform you see that they are the sawwave but sign flipped.
aka a Wrap or offset.

i think the OP already got his answer
but it seems you have something completely different in mind, and, by your explanation, i can't see how your solution would reproduce the "pwm sawtooth"
"flipping the sign of a sawtooth" - i don't see this happening

janostman · Post by **janostman** » Sun Sep 14, 2014 12:03 pm

antto wrote:
janostman wrote: If you look at the "holes" in your own waveform you see that they are the sawwave but sign flipped.
aka a Wrap or offset.
i think the OP already got his answer
but it seems you have something completely different in mind, and, by your explanation, i can't see how your solution would reproduce the "pwm sawtooth"
"flipping the sign of a sawtooth" - i don't see this happening

Look at the waveforms in the first post of this thread.
Fold the "holes" upwards and you have the original sawtooth wave again.

But as you said, he probably already got his answer.

aciddose · Post by **aciddose** » Sun Sep 14, 2014 4:45 pm

Sorry for the poor quality, it is not easy to take HQ photos of a scope and these are the best I can produce at the moment.

Waveform: Table of the waves printed on the front of the aJU

All of these waves start from 0.65v when enabled. When all waves are disabled, there is no offset, the output is 0v.

The imgur album: http://imgur.com/a/k6RcS

Square: Nothing exciting here...

Saw: Likewise...

Saw 2: Chopped by square at +1 octave

Saw 3: Like Saw 2, but with adjustable +1 octave pulse width
...

Saw 4: Chopped by square at +3 octave

Saw 5: Chopped by BOTH square at +1 and +3 octaves

It is of note the waveforms follow a binary pattern, as if the bits were used directly to enable a chopper circuit... hmmmmm

Sub 2: Chopped by square (note the sub is -1 oct)

Sub 3: Chopped by square +1 oct (again, -1 oct for the sub pulse)

Noise: The noise is a single bit switched randomly at high frequency. I've moved the camera to try to capture the animated effect, otherwise the image just looks like two solid horizontal lines.

itoa · Post by **itoa** » Sun Sep 21, 2014 9:09 pm

Thanks guys, especially Aciddose - you are my guru!

Yes in fact it's just a saw chopped (or AMed) by pulse wave - so easy. I've been fooled by additional low-shelving filter (hpf at 0) that made the waveform so difficult to decompile

.

Janostman: no it's not wrapping . Checked and proved

LeVzi · Post by **LeVzi** » Mon May 24, 2021 6:13 am

Sorry for the necrobump, but I was wondering if anyone has a wavetable of this for use in the likes of Serum ? In fact wavetables for any of the alpha juno waveforms it used ?

Scrubbing Monkeys · Post by **Scrubbing Monkeys** » Mon May 24, 2021 10:52 am

LeVzi wrote: Mon May 24, 2021 6:13 am Sorry for the necrobump, but I was wondering if anyone has a wavetable of this for use in the likes of Serum ? In fact wavetables for any of the alpha juno waveforms it used ?

A Ajuno saw pwm and square pwm WTs have been recently been added to the Sonic Academy Ana2 factory tables. I am not sure if ithey can be pulled out.

Urs · Post by **Urs** » Mon May 24, 2021 10:57 am

If someone wanted to do it in Hive using the .uhm scripts, it would be a piece of cake for anyone familiar with it, and the exported .wav could be used with any synth.

I'd be happy to give it a go if I find a minute.

Chapelle · Post by **Chapelle** » Mon May 24, 2021 1:04 pm

LeVzi · Post by **LeVzi** » Mon May 24, 2021 2:40 pm

Chapelle wrote: Mon May 24, 2021 1:04 pm
LeVzi wrote: Mon May 24, 2021 6:13 am Sorry for the necrobump, but I was wondering if anyone has a wavetable of this for use in the likes of Serum ? In fact wavetables for any of the alpha juno waveforms it used ?
I don't know exactly how it's supposed to work, but judging from the pictures in this thread, I just made the attached one in Serum. Let me know if that's the right one.

That's pretty legit, thank you, gonna have a mess about with it, now.

Scrubbing Monkeys · Post by **Scrubbing Monkeys** » Mon May 24, 2021 3:01 pm

Hey Chapelle, do you know the number of frames and sample per frame of that file?

Urs · Post by **Urs** » Mon May 24, 2021 3:06 pm

Scrubbing Monkeys wrote: Mon May 24, 2021 3:01 pm Hey Chapelle, do you know the number of frames and sample per frame of that file?

256 frames, 2048 samples long

Urs · Post by **Urs** » Mon May 24, 2021 3:33 pm

Here are version made with Hive, including the scripts to make them and a script that exports them to .wav.

I've made 4 variations. On top of the typical Saw PWM with two pulse cycles I also added versions with one, three and four pulse cycles.

Also, 256 frames, 2048 samples per frame.

The zip is surprisingly small, 54kB for 8MB of .wav files... well, the waveforms are rather simple...

Enjoy,

- U

Roland Alpha Juno-2 pwm sawtooth