Sine Subosc

isak · Posted: Sat Oct 19, 2013 2:20 am Post subject: Sine Subosc

Hi Guys.

after listening to the Minibrute Subosc i was very thrilled and wanted to try doing this my self.
so i was looking around and have this idea, hope its a good idea and if not i'll be happy to hear any suggestions about this.

i have a square subosc which works great, my thoughts where to chain it to a lowpass filter and kill all the high freqs, this action will leave me with only the low freqs and if im not mistaken it should look in the scope like a sine wave (not sure about the sound but if it look like a sine it should sound like a sine?).
my next action will be to split the signal with toggle switch, to choose between square sub or going out after the lowpass (sine)
if its a good idea i will need a simple lowpass filter scheme with no need for for FM or reso pot, only with cutoff pot.

what do you think?

cheers,
Isak E.

analogmonster · Joined: May 30, 2011 Posts: 90 Location: Germany

I wouldn't do it with a (lowpass-) filter, as a filter is frequency dependant. This means you play a high note, you get a sine, you play a low note, and the square is back, or one note disappears, or whatever.

A wave shaper would do a better job. There should be example circuits on the web. Another possibility might be to use a frequency divider cascade à la octavdivider. I use this in my Formant modular.

See also http://www.analog-monster.de/oktavdivider_en.html

If you are interested I send you the schematic (just send me a PM)

isak · Posted: Sat Oct 19, 2013 3:42 am Post subject:

thanks for replying and for the input.

analogmonster · Joined: May 30, 2011 Posts: 90 Location: Germany

isak · Posted: Sat Oct 19, 2013 6:04 am Post subject:

mmm...complicated...
I have the Berfotron saw to tri to sine waveshaper which works great but then again if I want it to be 1 or 2 octave down it should be extra vco.

I will google this sine wave suboctave and see if I find any hint in this matter.

thermionicjunky · Joined: Dec 07, 2006 Posts: 90 Location: san francisco

http://www.bernacomp.com/elec/og2/og3_sine_suboctave.html

Dave Kendall · Posted: Sat Oct 19, 2013 1:21 pm Post subject:

FWIW - on the OG3 schematic posted above, R11 doesn't actually need to be 49.9K - just use two 100K resistors in parallel (= 50K). Similarly, use 300K for R10 instead of 301K- it's a much more common value.

cheers,
Dave

isak · Posted: Sat Oct 19, 2013 11:06 pm Post subject:

Thanks for the input guys.

Out of curiosity I did an expirement yesterday with LP filter on square using vst and vsti, just to give my self an idea of how it should work.
It was success, the only thing is when I play higher octaves/notes then the sound is fading out obviously.
I found a solution for that but didnt try it yet.
Well, connect the CV keyboards to the VCF CV IN and when play higher notes it will open, the cool thing about this is that the max cv keyboards is 5V so it won't open the vcf cutoff all the way and if it will open it to sound a bit like square ill increase the vcf cv in resistor Smile

All i need is a very simple LP filter with CV in, any ideas?

Cheers.

gdavis · Posted: Sat Oct 19, 2013 11:15 pm Post subject:

For this idea to really work you need an LPF with exponential CV which isn't simple to do with much accuracy.

If your source osc is a saw, the circuit posted by thermionicjunky is probably your best bet.

isak · Posted: Sat Oct 19, 2013 11:32 pm Post subject:

not looking for the easy way to do this but in the name of experiment i must try this, only out of curiosity Smile

i'll experiment this today and if i'll succeed i'll upload a Tube clip.

wish me luck Smile

isak · Posted: Mon Oct 21, 2013 2:29 am Post subject:

Ok, just did the test with the ody vco and the mfos svf.
It was almost perfect.
What happens now is I lower the cutoff pot to about 50hrz and the shape in the scope was a perfect sine wave, when going up the scale the sine faded out naturally, so I connected the cv out of my midi2cv to the vcf cv in, it was ok and still got a sine wave shape, but the amplitude was low, when raising the cutoff a bit the amplitude was great and still got the sine wave shape.
5V from the midi2cv is not enough, Ill reduce the 100k of the vcf cv in and see if that will work.
As promised I'll make a vid clip as soon as I fix the key tracking.

Cheers.

gdavis · Posted: Mon Oct 21, 2013 12:25 pm Post subject:

Are you saying that with the keyboard CV the low notes are good but as you go up the keyboard the amplitude still decreases? This would mean you're not tracking 1V/oct accurately enough, there are trimmers to adjust this on the filter. I wouldn't touch the 100k unless you've maxed out the trim.

The V/oct on that filter isn't temp compensated like most VCO's are, so you may never get it perfect. This is why I said this solution wouldn't be simple Wink

Might be able to get pretty close though, will be interesting to see how well you can make this work.

isak · Posted: Mon Oct 21, 2013 2:29 pm Post subject:

hi gdavis and guys Smile

i think you misunderstand me..
i connected the cv out of the midi2cv to a regular cv in of the VCF.
the tracking to the vco works great, the VCF is not 1V/oct, i didnt tune it to be 1V/oct.
its fading out couse the cutoff is closed, this is why, not because the reason you think Wink

now, for the success of my test and i must say im very happy with the results!
it works as i thought it will work!
as BORAT says...."great success" Laughing

now in the vid youll see im playing from C0 to C4, thats more the enough for me as subosc, with more fine tune i can get it to go higher and lower.

enjoy the vid, any comment will be more then welcome.

http://www.youtube.com/watch?v=ORev4PsR7BI&feature=youtu.be

cheers,
Isak E.

gdavis · Posted: Mon Oct 21, 2013 3:37 pm Post subject:

I understood that you were connecting the KCV to the VCF cutoff CV. Both the svf's circuits I found on the MFOS site have V/OCT cutoff CV inputs. I think you should be able to trim that so that once the cutoff frequency knob is set properly, the VCF cutoff tracks the VCO frequency without changing the 100k.

isak · Posted: Mon Oct 21, 2013 3:53 pm Post subject:

Ohh ok, sorry for assuming you didn't.
I know the svf can be 1V/ocv but I didn't trim it for that.
In any way I'm not going to waist the svf to this matter, I'm going to use a simpler vcf I can find, svf was only for the test to see how it will work.
You saw the vid clip? It works well, don't you think? Smile

BTW, eventually I didn't changed the 100k, I replaced it with a 100k pot.

I found here in the forum a simple scheme of vcf, what do you think about it?
http://electro-music.com/forum/phpbb-files/ultrasimplevcf_132.png

And this is the original post..
http://electro-music.com/forum/topic-45231.html&postorder=asc

gdavis · Posted: Tue Oct 22, 2013 2:39 pm Post subject:

isak · Posted: Tue Oct 22, 2013 9:32 pm Post subject:

Do you think that maybe I can reduce the 100k to say 50k and put a trimmer in series with it the impadance will be ok?

gdavis · Posted: Wed Oct 23, 2013 11:05 am Post subject:

Are you still talking about the MFOS filter? Why don't you just leave the 100k and use the trimmers that are already there?

If you're talking about the super simple filter then it's less than ideal but if you're hard wiring it to a subosc output it should be fine. You're substituting simplicity for a degree reliability. It's a judgement call based on the particular circuit and what you expect to get out of it but based on what you're trying to do I don't see any problem.

isak · Posted: Wed Oct 23, 2013 1:32 pm Post subject:

Talking about the simple VCF.
Thanks for the info and your help Smile