Tips for making a very slow envelope

Anything modular synth related that is not format specific.

Moderators: Kent, Joe., analogdigital, infradead, lisa, parasitk, plord, sduck

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Tips for making a very slow envelope

Post by electricanada » Sat Apr 13, 2019 6:47 pm

What if I want a single eight-minute cycle of a triangle wave, that rises for the first four minutes and falls for the last four minutes--any tips or shortcuts for setting this up? I have a Maths, which according to the manual should be able to last as long as 25 minutes.

I'm not sure how to set it up other than pure trial and error and a watch. I'm hoping there is some kind of shortcut. I have a multimeter and a Mordax Data, if those could be useful in this endeavor.
Eléctrica (electric) Nāda (the yoga of sound).

User avatar
Dave Peck
Super Deluxe Wiggler
Posts: 3242
Joined: Thu Mar 24, 2011 1:22 pm
Location: SF bay area

Post by Dave Peck » Sat Apr 13, 2019 7:43 pm

If it needs to be exactly eight minutes (not 7:52, not 8:07) that's going to be tough to dial in using a MATHs. It would be a lot easier just drawing it out as a midi CC move in a DAW project or some other similar method that is tied to a clock.

Or, if you want to try to use some kind of VC osc/LFO, you could set it to 1.066666667 Hz (if you have a really accurate frequency counter) which equals 64 cycles per minute, and then patch -10VDC to the 1V/OCT pitch mod input, which would drop it ten octaves and give you one cycle every eight minutes. But again, that's asking for a LOT of accuracy in your measuring devices and your VC osc's pitch mod circuit and your -10V DC source.

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Sat Apr 13, 2019 7:51 pm

Dave Peck wrote:If it needs to be exactly eight minutes (not 7:52, not 8:07) that's going to be tough to dial in using a MATHs. It would be a lot easier just drawing it out as a midi CC move in a DAW project or some other similar method that is tied to a clock.

Or, if you want to try to use some kind of VC osc/LFO, you could set it to 1.066666667 Hz (if you have a really accurate frequency counter) which equals 64 cycles per minute, and then patch -10VDC to the 1V/OCT pitch mod input, which would drop it ten octaves and give you one cycle every eight minutes. But again, that's asking for a LOT of accuracy in your measuring devices and your VC osc's pitch mod circuit and your -10V DC source.
Ah, good idea. I knew there would be geniuses who could help me in here. No, I don't need it to be real exact. Conceptually, I'm working with the idea of using a single very slow waveform to structure a piece.

Maybe I could use the readout on my Z3000 to get 1.07 hz? I don't know if it goes that low--will have to check.
Eléctrica (electric) Nāda (the yoga of sound).

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Sat Apr 13, 2019 7:55 pm

Is there a module that would allow me to draw a waveform and then set it to a very slow rate?
Eléctrica (electric) Nāda (the yoga of sound).

ranix
Super Deluxe Wiggler
Posts: 1635
Joined: Wed Jul 13, 2016 9:39 pm

Post by ranix » Sat Apr 13, 2019 10:31 pm

A 3.5 microfarad capacitor would charge to 5V in 240 seconds with a 100 megohm resistor and a 10V supply. If you have electrical safety knowledge and a benchtop power supply, the resistor isn't too expensive and that value of capacitor is used in ceiling fans. That's a lot of energy stored though, like 45 joules, so it would need safe handling. It'd be kind of a project. I think I could get it done in a lab environment. Maybe you can do that better with more resistance and less capacitance but it might get expensive (100 meg resistors are pricey).

I can't think of anything else except goofy stuff like a weight and pulley system like a cuckoo clock.

There are some function generators on Amazon that go down to the microhertz range for around $400 but I don't know if they can make a triangle rise that slowly. Now I want one and it's your fault

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Sat Apr 13, 2019 10:48 pm

ranix wrote:A 3.5 microfarad capacitor would charge to 5V in 240 seconds with a 100 megohm resistor and a 10V supply. If you have electrical safety knowledge and a benchtop power supply, the resistor isn't too expensive and that value of capacitor is used in ceiling fans. That's a lot of energy stored though, like 45 joules, so it would need safe handling. It'd be kind of a project. I think I could get it done in a lab environment. Maybe you can do that better with more resistance and less capacitance but it might get expensive (100 meg resistors are pricey).

I can't think of anything else except goofy stuff like a weight and pulley system like a cuckoo clock.

There are some function generators on Amazon that go down to the microhertz range for around $400 but I don't know if they can make a triangle rise that slowly. Now I want one and it's your fault
I know Maths can do it; it's just getting the setting right that's the challenge. I think that very slow waveforms could be an amazing tool for creating musical form within the modular.
Eléctrica (electric) Nāda (the yoga of sound).

User avatar
Dave Peck
Super Deluxe Wiggler
Posts: 3242
Joined: Thu Mar 24, 2011 1:22 pm
Location: SF bay area

Post by Dave Peck » Sat Apr 13, 2019 11:52 pm

If you have a hardware sequencer module with lots of steps, like 24, you could try clocking that really slow, like one step every 20 seconds, and arrange the steps so the first half slowly increase and the second half slowly decrease, and run the sequencer output through a lag/portamento set at max. That gives you an eight minute cycle.

User avatar
Pelsea
Veteran Wiggler
Posts: 581
Joined: Thu Apr 19, 2018 11:46 am
Location: Santa Cruz CA
Contact:

Post by Pelsea » Sun Apr 14, 2019 12:01 am

I think this is a job for trial and error. Set the knob at 25% and see what you get. Repeat with tiny adjustments of the knob. It will take a while, but not as long as building or acquiring a special module.

Don’t forget to mark the panel when you find the sweet spot.
Books and tutorials at http://peterelsea.com
Patch responsibly-
pqe

User avatar
lisa
Super Deluxe Wiggler
Posts: 4137
Joined: Fri Jun 02, 2017 7:00 am
Location: Lidingö

Post by lisa » Sun Apr 14, 2019 12:22 am

electricanada wrote:Is there a module that would allow me to draw a waveform and then set it to a very slow rate?
Xaoc Zadar gives you a great variety of envelopes and allows you to dial in the exact lenght (up to 30 minutes).
Last edited by lisa on Mon Apr 15, 2019 2:12 pm, edited 1 time in total.
Icy and melodic braindance/IDM winter track. ❄️


User avatar
Dcramer
Super Deluxe Wiggler
Posts: 5388
Joined: Fri Mar 29, 2013 12:11 pm
Location: Canada

Post by Dcramer » Sun Apr 14, 2019 1:01 am

I do this all the time, using the slow functions to open up a whole bunch of things in a complex patch.
I simply set it to trigger via a clock divider and then use a VU meter and a stop watch (phone) to tweak it to the right length. :tu:

User avatar
matcsat
Common Wiggler
Posts: 160
Joined: Thu Sep 26, 2013 4:24 pm
Location: Italy

Re: Tips for making a very slow envelope

Post by matcsat » Sun Apr 14, 2019 6:18 am

Hi,
electricanada wrote:What if I want a single eight-minute cycle of a triangle wave, that rises for the first four minutes and falls for the last four minutes--any tips or shortcuts for setting this up? I have a Maths, which according to the manual should be able to last as long as 25 minutes.

I'm not sure how to set it up other than pure trial and error and a watch. I'm hoping there is some kind of shortcut. I have a multimeter and a Mordax Data, if those could be useful in this endeavor.
lets say you have a waveform (triangle wave) and you know its amplitude (thanks to mordax data) and for example it's 5V pp: you want a 8 min cycle so you know that at 4 min the waveform is at 5V, at 2 min is at 2.5V, at 1 min is at 1,25V, 15sec/0,625V and so on.

In this way you can predict the duration of your waveform cycle in a shorter time.

I dont know the accuracy of this method, but it maybe enought if you dont need clock tight timing.
And probably it will became harder with non linear waveform, unless you use a multiwaveform generator (triangle or saw for the mesurement and then use the others for your porpuse).

Hope it helps.

Marco.

User avatar
matcsat
Common Wiggler
Posts: 160
Joined: Thu Sep 26, 2013 4:24 pm
Location: Italy

Post by matcsat » Sun Apr 14, 2019 6:30 am

... and isnt the mordax data also a waveform generator ?!

Marco.

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Re: Tips for making a very slow envelope

Post by electricanada » Mon Apr 15, 2019 10:16 am

matcsat wrote:Hi,
electricanada wrote:What if I want a single eight-minute cycle of a triangle wave, that rises for the first four minutes and falls for the last four minutes--any tips or shortcuts for setting this up? I have a Maths, which according to the manual should be able to last as long as 25 minutes.

I'm not sure how to set it up other than pure trial and error and a watch. I'm hoping there is some kind of shortcut. I have a multimeter and a Mordax Data, if those could be useful in this endeavor.
lets say you have a waveform (triangle wave) and you know its amplitude (thanks to mordax data) and for example it's 5V pp: you want a 8 min cycle so you know that at 4 min the waveform is at 5V, at 2 min is at 2.5V, at 1 min is at 1,25V, 15sec/0,625V and so on.

In this way you can predict the duration of your waveform cycle in a shorter time.

I dont know the accuracy of this method, but it maybe enought if you dont need clock tight timing.
And probably it will became harder with non linear waveform, unless you use a multiwaveform generator (triangle or saw for the mesurement and then use the others for your porpuse).

Hope it helps.

Marco.
Thanks, that's a good shortcut. Do you think it would work to create complex waveforms if the complex waveforms were created from simpler waveforms individually timed and then added together?

For example, say I create a four minute ramp. I want to add three peaks on that ramp, so I create a second ramp with duration of one minute. Then I add the four minute ramp and the one minute ramp in a mixer. Would this get me a four minute rising wave with three peaks?
Eléctrica (electric) Nāda (the yoga of sound).

User avatar
matcsat
Common Wiggler
Posts: 160
Joined: Thu Sep 26, 2013 4:24 pm
Location: Italy

Re: Tips for making a very slow envelope

Post by matcsat » Mon Apr 15, 2019 11:45 am

Hi,
electricanada wrote:,
For example, say I create a four minute ramp. I want to add three peaks on that ramp, so I create a second ramp with duration of one minute. Then I add the four minute ramp and the one minute ramp in a mixer. Would this get me a four minute rising wave with three peaks?
mh ... i think you can, the timing will remain the same (4 min total), but the amplitude will end up be less predictable (where the two waveforms overlap, the amplitude is the summ of the two), maybe using a min/max module the result will be more coherent with your idea.

btw in your example the peaks will be two.* :)

Marco.

*
For no reasons i figured you'll use single cycle unipolar waveform, otherwise you're right.
Last edited by matcsat on Mon Apr 15, 2019 11:50 am, edited 2 times in total.

User avatar
Dave Peck
Super Deluxe Wiggler
Posts: 3242
Joined: Thu Mar 24, 2011 1:22 pm
Location: SF bay area

Post by Dave Peck » Mon Apr 15, 2019 11:49 am

Yes, but the relative phase of the fast & slow triangle waves would be important. Otherwise you could have a peak of the fast wave coinciding with the lowest point and highest point on the slow wave, which would possibly not give you the shape you want. At least at the lowest point - it would get mixed with a peak from the fast wave and raise the low point of the slow wave, and could even turn it into another small peak with two low points on either side, depending on the amplitude of the two waves.

To avoid this, you would probably want the zero crossing point of the fast wave to coincide with the highest and lowest points of the slow wave.

User avatar
Brendanleespengler
Common Wiggler
Posts: 102
Joined: Fri Oct 14, 2016 8:53 am
Location: Memphis, TN
Contact:

Post by Brendanleespengler » Mon Apr 15, 2019 12:33 pm

I dont' have a Maths anymore, but can't you patch channel 2 to attenuate the rise of channel 1 to get longer attack times, and do the same for the fall to get longer release times?
Brendan Lee Spengler / Electronic Sound Research: The Electric Cowboy, Sinner Frenz, MacMillan & Spengler, Viva L'American Death Ray Music, Teledildonics5000, etc.

User avatar
nostalghia
Ultra Wiggler
Posts: 863
Joined: Sun Apr 07, 2013 2:46 pm
Location: central usofa

Post by nostalghia » Mon Apr 15, 2019 1:28 pm

Brendanleespengler wrote:I dont' have a Maths anymore, but can't you patch channel 2 to attenuate the rise of channel 1 to get longer attack times, and do the same for the fall to get longer release times?
Even easier-patch variable output of channel 2 to the channel 1 "Both" input jack (between Rise & Fall jacks), which will affect both rise (attack) and fall (release) rates equally.

Turn the channel 2 attenuverter knob counter-clockwise to get longer times, as the "Both" CV in works inversely to the Fall and Rise inputs (negative voltages for longer times, positive for shorter). Rise, Fall and Both CV ins take +/- 8 volts, and since channel 2 offset voltage with no input has +/- 10V range, you won't need to turn the knob fully CCW. Also, response on the Both input is exponential, rather than linear like the Rise and Fall inputs.
“If you look for a meaning, you'll miss everything that happens.” - Andrei Tarkovsky

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Mon Apr 15, 2019 1:59 pm

Here is a diagram of a curve that could represent the tension level in a musical performance. If we could make a single cycle of a waveform, and use that as cv to control elements of tension -- such as volume, density, tessitura, etc. -- that would be incredibly useful. The entire basic form of a piece could be encoded in a single waveform. If you then hang some comparators off of that waveform, it gets even more useful for encoding form.

This is the idea I was getting at with mixing one cycle of a very slow rising ramp wave, with several cycles of a slightly faster second ramp wave, to get something like this:

Image
Last edited by electricanada on Mon Apr 15, 2019 2:15 pm, edited 2 times in total.
Eléctrica (electric) Nāda (the yoga of sound).

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Mon Apr 15, 2019 2:05 pm

Here's how we could use a single cycle of a triangle wave, combined with a basic analog comparator, to encode an ABA form in CV. Use the comparator to turn the A and B elements of the composition on and off, thus creating musical form with CV:

Image
Last edited by electricanada on Mon Apr 15, 2019 2:26 pm, edited 3 times in total.
Eléctrica (electric) Nāda (the yoga of sound).

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Mon Apr 15, 2019 2:11 pm

Here's how we could combine a single cycle of a triangle wave with a window comparator to create the CV for a simple palindromic rondo form: ABABA. With a dual window comparator, we could make a more extended rondo form.

Image
Eléctrica (electric) Nāda (the yoga of sound).

User avatar
matcsat
Common Wiggler
Posts: 160
Joined: Thu Sep 26, 2013 4:24 pm
Location: Italy

Post by matcsat » Mon Apr 15, 2019 2:26 pm

Hi,
electricanada wrote: ... If you then hang some comparators ...
i think you would like the Doepfer A-152

Marco.

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Mon Apr 15, 2019 2:40 pm

matcsat wrote:Hi,
electricanada wrote: ... If you then hang some comparators ...
i think you would like the Doepfer A-152

Marco.
Thanks, I actually have a 152, but I haven't installed it yet. I'm not quite sure what to do with it. I like to make music with form, at least occasionally. I'm intrigued by the idea of using CV to create form. How do you think the 152 could fit into this idea?
Eléctrica (electric) Nāda (the yoga of sound).

User avatar
Dave Peck
Super Deluxe Wiggler
Posts: 3242
Joined: Thu Mar 24, 2011 1:22 pm
Location: SF bay area

Post by Dave Peck » Mon Apr 15, 2019 2:42 pm

electricanada wrote:Here's how we could combine a single cycle of a triangle wave with a window comparator to create the CV for a simple palindromic rondo form: ABABA. With a dual window comparator, we could make a more extended rondo form.

Image
Yup. I remember several years ago one of the Nord Modular gurus, Jan Punter from Den Haag (who is a genius at creating self-generating patches/compositions) created a piece that utilized the Nord Modular's 622-second-per-cycle LFO to do exactly what you describe, creating a piece of self-generating music with a duration of ten minutes and 22 seconds.

This LFO was set to start at the lowest point of the wave upon loading the patch, and the LFO's signal was sent to various 'compare to level' modules that were used to initiate various events or changes in the patch as the waveform slowly rose and then fell again. Each performance was unique, yet it always followed the same general progression of events and changes.

User avatar
matcsat
Common Wiggler
Posts: 160
Joined: Thu Sep 26, 2013 4:24 pm
Location: Italy

Post by matcsat » Mon Apr 15, 2019 3:15 pm

Hi,
electricanada wrote: ... How do you think the 152 could fit into this idea? ...
The addressed part is the first that came to my mind.
If you put any waveform into the CV input and offset it so it's only positive* then you have 8 comparators equally spanned.
With some carefull preparation you can organize evolving pieces based from anything you put into the CV in.
Then you have 8 windowed switch, S&H and gates.
From there theres only your immagination (sorry if i'm cheesey).

Marco.

* one thing I really like about the A-152 is that it is able to offset (with the addr. knob) the incoming CV as well as attenuate it.

electricanada
Ultra Wiggler
Posts: 891
Joined: Wed Nov 29, 2017 12:26 am
Location: Norfolk, VA

Post by electricanada » Mon Apr 15, 2019 3:36 pm

matcsat wrote:Hi,
electricanada wrote: ... How do you think the 152 could fit into this idea? ...
The addressed part is the first that came to my mind.
If you put any waveform into the CV input and offset it so it's only positive* then you have 8 comparators equally spanned.
With some carefull preparation you can organize evolving pieces based from anything you put into the CV in.
Then you have 8 windowed switch, S&H and gates.
From there theres only your immagination (sorry if i'm cheesey).

Marco.

* one thing I really like about the A-152 is that it is able to offset (with the addr. knob) the incoming CV as well as attenuate it.
Thanks, I understand some of it now. Because the 152 is a voltage-addressed switch, we can switch between different formal elements of a composition with a single slow-moving CV source, such as a sequencer with a very slow clock, or one of the very slow waveforms discussed above.

I'm still not certain I understand the comparator element of the 152.
Last edited by electricanada on Mon Apr 15, 2019 3:37 pm, edited 1 time in total.
Eléctrica (electric) Nāda (the yoga of sound).

Post Reply

Return to “Modular Synth General Discussion”