Author |
Message |
stefanv
Joined: May 09, 2009 Posts: 17 Location: Moorefield, Ontario, Canada
|
Posted: Wed Feb 03, 2010 1:57 pm Post subject:
LFO and ADSR modulation of VCA |
|
|
I'm in the process of building analog-synth-like features into a PAiA Stringz'n'Thingz string synthesizer (keeping all the capabilities of that as well). I plan to post the details on my web site when I'm done (and I'll post some pics and info here too), but I have a question for you all:
I plan to provide both ADSR and LFO modulation of the amplitude, and I'm wondering if people generally use two separate VCAs for that? I realize it's possible to sum the ADSR and LFO voltages, but then one of three things will happen:
1. If the LFO output has its lower peaks at 0V and its upper peaks at say 1V, then the sound through the VCA will continue being turned on and off even after the ADSR has gone back down to 0.
2. If the LFO output has its upper peaks at 0V and its lower peaks at say -1V, then once the ADSR output drops below 1V, the LFO will turn the sound fully off during parts of the cycle. So as the note fades away, a subtle tremolo will turn into a choppy "beeping".
3. If the LFO output is centered around 0V, then both of the above undesirable effects will happen.
It seems to me that one should use the LFO, together with a fixed bias voltage, to modulate one VCA, and then take the output of that VCA and feed it into another VCA controlled by the ADSR. In addition to avoiding the aforementioned undesired behaviors, this results in the LFO modulation remaining as a fixed percentage of the current volume, so that the tremolo won't become more pronounced as the overall volume decreases.
Am I missing something here?
Thanks, Stefan
Description: |
LFO with 0 to 1V range, ADSR with 0 to 5V range, voltages summed. Notice beeps after ADSR has gone to zero. |
|
Filesize: |
28.59 KB |
Viewed: |
563 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
LFO with -1 to 0V range, ADSR with 0 to 5V range, voltages summed. Notice beeps just before ADSR reaches zero. |
|
Filesize: |
22.59 KB |
Viewed: |
469 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
LFO with -0.5 to 0.5V range, ADSR with 0 to 5V range, voltages summed. Notice beeps just before and after ADSR has gone to zero. |
|
Filesize: |
25.26 KB |
Viewed: |
448 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
LFO with 4.5 to 5.5V range controlling one VCA, ADSR with 0 to 5V range controlling the other. No beeps, and the LFO modulation is reduced in intensity as the volume drops off. |
|
Filesize: |
20.12 KB |
Viewed: |
423 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
_________________ Stefan Vorkoetter
http://www.stefanv.com |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Wed Feb 03, 2010 2:06 pm Post subject:
|
|
|
Stefan.
On the Nord Modular (a virtual analog) I'd use two VCAs for this. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
kkissinger
Joined: Mar 28, 2006 Posts: 1354 Location: Kansas City, Mo USA
Audio files: 42
|
Posted: Wed Feb 03, 2010 4:06 pm Post subject:
|
|
|
You might be able to accomplish this effect if you add a little negative bias to the combined ADSR and lfo voltages.
If your LFO is, say, 2 volts peak-to-peak, then with a -2 volt bias, no bleedthrough will occur when the ADSR is at zero.
What WILL happen is that the attack will vary depending on where the LFO is in its cycle.
Incidentally, I used this technique -- to mix an LFO and ADSR, for the lead (the ocarina-like) part in this track:
http://kevinkissinger.com/downloads/music/FluteImprov.mp3
btw -- cool avatar. From one pilot to another, Welcome! _________________ -- Kevin
http://kevinkissinger.com |
|
Back to top
|
|
|
stefanv
Joined: May 09, 2009 Posts: 17 Location: Moorefield, Ontario, Canada
|
Posted: Thu Feb 04, 2010 5:00 am Post subject:
|
|
|
Thanks Kevin. I think that's more or less what I did in the 2nd graph in my original post. The LFO is 1Vp-p, and there's a negative bias so that it remains below zero all the time, thus giving no bleedthrough after the ADSR is done. But, there are still the pulses during the last part of the release cycle, when the ADSR is almost-but-not-quite done. Perhaps in actual use, this doesn't matter, because the release times are relatively short and the LFO cycle times relatively long.
Cool that you're a pilot too! I'm amazed sometimes with the people that write to me that share 2 or 3 or 4 of my interests instead of just one (I'm into planes, model planes, vintage HP calculators, slide rules, Hammond organs, synths, and electronics in general). _________________ Stefan Vorkoetter
http://www.stefanv.com |
|
Back to top
|
|
|
kkissinger
Joined: Mar 28, 2006 Posts: 1354 Location: Kansas City, Mo USA
Audio files: 42
|
Posted: Thu Feb 04, 2010 8:04 am Post subject:
|
|
|
stefanv wrote: | ...there are still the pulses during the last part of the release cycle, when the ADSR is almost-but-not-quite done. Perhaps in actual use, this doesn't matter, because the release times are relatively short and the LFO cycle times relatively long. |
That's true. In my track, the lead part had a good deal of reverb and that masked the note tails.
stefanv wrote: | I'm amazed sometimes with the people that write to me that share 2 or 3 or 4 of my interests instead of just one. |
You should plan to attend an electro-music festival -- I guarantee you that you'd have a great time. _________________ -- Kevin
http://kevinkissinger.com |
|
Back to top
|
|
|
riley smith
Joined: Sep 04, 2008 Posts: 10 Location: santa cruz, ca
|
Posted: Mon Mar 08, 2010 7:38 pm Post subject:
LFO-ADSR modulation |
|
|
Here's another idea. My Emu VCAs have a mode where one control input acts as a multiplier for the other. This allows you to have tremolo that decays properly to zero. This is all done in the control circuit without adding another VCA. |
|
Back to top
|
|
|
stefanv
Joined: May 09, 2009 Posts: 17 Location: Moorefield, Ontario, Canada
|
Posted: Tue Mar 09, 2010 7:10 am Post subject:
Re: LFO-ADSR modulation |
|
|
riley smith wrote: | Here's another idea. My Emu VCAs have a mode where one control input acts as a multiplier for the other. This allows you to have tremolo that decays properly to zero. This is all done in the control circuit without adding another VCA. |
So it's multiplying the control voltages instead of adding them? I think that basically means there are still two VCAs, but one of them is being used to modulate the control voltage of the other (instead of modulating the audio).
I ended up going with two VCAs in my design, since they're pretty cheap and easy to build. Now I just have to find the time to start actual physical work on this project (all the schematics and board layouts are done, and I have all the parts except some of the pots). _________________ Stefan Vorkoetter
http://www.stefanv.com Last edited by stefanv on Tue Jul 06, 2010 8:18 am; edited 1 time in total |
|
Back to top
|
|
|
emdot_ambient
Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD
|
Posted: Fri May 07, 2010 8:19 am Post subject:
|
|
|
Just a little brain storm here...If you use a Voltage Controlled ADSR, you could run the LFO into the Sustain CV of the Envelope instead of into the VCA itself. Then, the LFO will only modulate the VCA when the Envelope is in its Sustain cycle. Not as flexible as both an LFO and Envelope input to the VCA, but it might be all you need.
It's a thought. |
|
Back to top
|
|
|
stefanv
Joined: May 09, 2009 Posts: 17 Location: Moorefield, Ontario, Canada
|
Posted: Sat May 08, 2010 6:59 am Post subject:
|
|
|
Yeah, that would work too, but I've gone with the dual VCA approach instead. The board is all built now, and I'm on the home stretch, wiring everything together. _________________ Stefan Vorkoetter
http://www.stefanv.com |
|
Back to top
|
|
|
stefanv
Joined: May 09, 2009 Posts: 17 Location: Moorefield, Ontario, Canada
|
Posted: Thu May 20, 2010 4:34 am Post subject:
|
|
|
In case anyone's interested, I've posted some pictures with captions of the progress so far on my web site:
http://www.stefanv.com/electronics/paia_stringz_n_thingz.html
I'll eventually turn this into a full blown article like the others on my site, with all the details of the updates and mods I made. _________________ Stefan Vorkoetter
http://www.stefanv.com Last edited by stefanv on Thu May 20, 2010 11:14 am; edited 1 time in total |
|
Back to top
|
|
|
emdot_ambient
Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD
|
Posted: Thu May 20, 2010 9:42 am Post subject:
|
|
|
Dude...man, that's a LOT of work. Looking great, though. Can't wait to see/hear it finished! |
|
Back to top
|
|
|
stefanv
Joined: May 09, 2009 Posts: 17 Location: Moorefield, Ontario, Canada
|
Posted: Thu May 20, 2010 11:13 am Post subject:
|
|
|
Lately, it's just been tedious, but not difficult work. However, just the other day, when the project was just a little past the stage you see in the last picture, I decided to apply power to the three tone generator/keying boards, with the signals connected directly to an amp. Much to my relief, everything worked: the old PAiA boards with new caps, my 8 new note blocks, and my new master clock.
I kind of wish I'd been keeping track of my hours spent on this. Although I spent a fair number of hours wiring and soldering, I think I probably spent even more hours designing and redesigning (including redesigning some of the PAiA circuitry). I've been fortunate enough to get in touch with Marvin Jones, the original designer. _________________ Stefan Vorkoetter
http://www.stefanv.com |
|
Back to top
|
|
|
stefanv
Joined: May 09, 2009 Posts: 17 Location: Moorefield, Ontario, Canada
|
|
Back to top
|
|
|
rage_q3
Joined: Nov 28, 2013 Posts: 4 Location: Moscow, Russian Federation
|
Posted: Thu Nov 28, 2013 3:57 am Post subject:
|
|
|
Hey guys,
I can't get a clue how to mix output of the VCAs to achieve the "correct" effect without bleeps... If I'll mix two via a simple mixer the bleeps won't go away. It would be great if you guys can clarify this moment.
P.S. I am in process of building my first modular synth and totally new to modular synthesizers in overall and now dig deeply in everything related to sound sculpting and synthesis methods thus treat me like a newbie |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue Dec 31, 2013 5:45 am Post subject:
|
|
|
Ah oh .. oops ... sorry rage_q3 .. I just now see you posted here.
I would use one VCA to modulate the output of the other one .. so .. not add them in a mixer but multiply rather. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Wed Jan 01, 2014 9:21 am Post subject:
|
|
|
Blue Hell wrote: | I would use one VCA to modulate the output of the other one .. so .. not add them in a mixer but multiply rather. |
some more thoughts about it ... you seemed to want to add two VCA signals, where I think they should be multiplied ... a VCA essentially being a multiplier ... so .. two cascaded VCAs where one applies an envelope and the other applies the LFO. The LFO would need an offset added to it to make it a positive only signal. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
rage_q3
Joined: Nov 28, 2013 Posts: 4 Location: Moscow, Russian Federation
|
Posted: Wed Jan 15, 2014 11:05 pm Post subject:
|
|
|
Blue Hell wrote: | so .. two cascaded VCAs where one applies an envelope and the other applies the LFO. The LFO would need an offset added to it to make it a positive only signal. |
it would be great if you can post an image of the patch in nord modular g2 showing this idea. i cant get a clue how to multiply the two vca signals? the mixer will just sum them. |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Thu Jan 16, 2014 10:22 am Post subject:
|
|
|
VCA ==multiplier ... so you just put one after the other, no need for anything additional (except the LFO and the envelope) .. hmm not even that is needed as the Envelope modules have a VCA built in already.
Ok, waiting for the next backup failure here at work ... will download the Clavia demo SW. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
|
Back to top
|
|
|
rage_q3
Joined: Nov 28, 2013 Posts: 4 Location: Moscow, Russian Federation
|
Posted: Mon Jan 20, 2014 4:57 am Post subject:
|
|
|
Thank you for your reply, kinda helpful! Now I really understand that VCA is working like a multiplier. Was unaware of that.
The leftmost example is really simple, the AM input allows to modulate the whole level of the envelope using any external source. Now i understand how to make a sequenced pattern with note gaps since the note sequencer doesn't provide a way to skip a note to get a silence instead like in all DAWs the piano roll allows. In g2 I can just send a zero value using a value sequencer in the AM input of the envelope and viola ! and it allows to control the sequencer velocity as well.
But what is the downside of this method with modulating the AM input? I didn't hear the difference between all the methods though I understand that the middle example is the same as the leftmost but with external VCA applied. And the 3rd one is really a closest to the original description.
Btw, you said that the lfo is biased but I didn't notice that...its just a regular unipolar lfo at a 2.72 hz frequency. would be great if you can explain it. |
|
Back to top
|
|
|
rage_q3
Joined: Nov 28, 2013 Posts: 4 Location: Moscow, Russian Federation
|
Posted: Fri Jan 24, 2014 10:48 am Post subject:
|
|
|
i got it, the lfo is a positive only! thx! |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Fri Jan 24, 2014 11:31 am Post subject:
|
|
|
All three are the same ... _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
|