MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Using a single CV source for quad panning
MUFF WIGGLER Forum Index -> Modular Synth General Discussion Goto page 1, 2  Next [all]
Author Using a single CV source for quad panning
Hovercraft
I'm in the process of experimenting with quad panning a single mono source. I understand how to pan around four outputs using quadrature LFO's to sequentially open and close four VCA's, but am wondering how to set up panning using a single external CV sawtooth wave. For instance, I'd like to have the CV work like the following:

CV=0V - VCA 1 100% open, VCA 2/3/4 0% open
CV=2.5V - VCA 2 100% open, VCA 1/3/4 0% open
CV=5V - VCA 3 100% open, VCA 1/2/4 0% open
CV=7.5V - VCA 4 100% open, VCA 1/2/3 0% open

All with smooth transitions in between each state. I know the Koma Poltergeist can work like this, and am fairly sure the Buchla 227e works the same way with CV input. Any ideas appreciated!
Navs
Have a look at the Doepfer Morphing Controller or Interpolating Scanners (Haible, Tillman, Toppobrillo) for ideas. The point is that your one CV actually becomes four CVs. There are other ways to do the maths (Wiard JAG) but iirc the problem is not sending your signal to the edges, but equally to the middle of the four speakers.
Hovercraft
Navs wrote:
Have a look at the Doepfer Morphing Controller or Interpolating Scanners (Haible, Tillman, Toppobrillo) for ideas. The point is that your one CV actually becomes four CVs. There are other ways to do the maths (Wiard JAG) but iirc the problem is not sending your signal to the edges, but equally to the middle of the four speakers.


Thanks--will look into those modules, but more interested in constructing something out of basic modular functions. As for spread or radius, I figured I could send an offset to all four VCA's to move a source towards the center.

edit: That Doepfer A144 looks like it performs the function I described. The question is--how can it be replicated with simple modules?
ranix
I would try using the saw CV to control the phase of 2 sine waves 90 degrees out of phase with each other. Patch Sine A to VCA1 and inverted Sine A to VCA3, and patch Sine B to VCA2 and inverted Sine B to VCA4.

I'm not sure exactly how I would control the phase. I'd probably try two FM Aid since I have multiple.

You wouldn't be able to get full range (0% and 100% would be inaccessible) out of phase control via inverted pulse sync, so that method of phase control is out I think.
EPTC
Hovercraft wrote:
I know the Koma Poltergeist can work like this


Yeah, long considered the Poltergeist for a number of reasons, quad panning included.

I opted instead to get the Dotcom Q148 and really love it. https://synthesizers.com/q148.html

I mostly work in stereo panning, and haven't tried quadrophonic with it, but its support of unipolar/bipolar panning CV suggests it wouldn't be difficult to achieve with two of them - or even just one of them with a little math and another send module.
Dave Peck
The SSL Segwencer module is another option that will do this.
Dcramer
Frames SlayerBadger!
Paranormal Patroler
Hovercraft wrote:
Navs wrote:
Have a look at the Doepfer Morphing Controller or Interpolating Scanners (Haible, Tillman, Toppobrillo) for ideas. The point is that your one CV actually becomes four CVs. There are other ways to do the maths (Wiard JAG) but iirc the problem is not sending your signal to the edges, but equally to the middle of the four speakers.


Thanks--will look into those modules, but more interested in constructing something out of basic modular functions. As for spread or radius, I figured I could send an offset to all four VCA's to move a source towards the center.

edit: That Doepfer A144 looks like it performs the function I described. The question is--how can it be replicated with simple modules?


One CV source, four attenuators, four offsets, four VCAs. Mult the CV source to pass through the Att and Offset, set the controls to replicate the percentages you want, plug to the VCAs. It's fairly easy!
ranix
I played with this patch tonight and ended with:

two VCOs up to be 90 degrees out of phase with each other by syncing VCO2 from the inverted pulse output of VCO1 and setting the pulse width at roughly 25%

sine output of each VCO to a VCA with inverted outputs.

+ output of VCA1 to CV in of VCA3
+ output of VCA2 to CV in of VCA4
- output of VCA1 to CV in of VCA5
- output of VCA2 to CV in of VCA6

Then I generated an audio signal with ̶V̶C̶A̶3̶ VCO3 (a square wave) and sent it to the audio in of VCAs 3, 4, 5, and 6.

Then I sent the output of VCA3,4,5,6 to a compressor with blinkenlights that can show the relative channel volume of each of the 4 channels.

Two VCOs 90 degrees out of phase are kind of like having a joystick controller you're moving in circles, it looks like this on a scope:
https://my.mixtape.moe/lwkcfv.mp4

Since VCAs 3, 4, 5, and 6 are silent at 0V and only produce audio when CV is >0, all 4 channels should be silent if the joystick is dead center. Top center is CH1, right center is CH2, bottom center is CH3, left center is CH4 (or similar, depending on how you patch).

Here's a video of the blinkenlights:
https://my.mixtape.moe/eoytwc.mp4

Here's a photo of the patch:



If VCO1 and VCO2 were stopped and you could control their phase with a saw you'd have your requested patch. Maybe there's a way to do something similar to that.
ranix
Paranormal Patroler wrote:

One CV source, four attenuators, four offsets, four VCAs. Mult the CV source to pass through the Att and Offset, set the controls to replicate the percentages you want, plug to the VCAs. It's fairly easy!


This might be simpler, I think my patch idea is a dead end
Hovercraft
Thanks for all the responses--it's interesting seeing what these modules do as background.

@ranix
very cool--the phase shifting approach is something I need to explore. I'm going to try your patch and see where it leads--even if it doesn't solve my problem, it'll probably lead somewhere interesting. I have two quad LFO banks that allow phase to be set on each channel.

@paranormal patroller
Probably missing something, but I can't figure out how your approach works. Each input voltage would require a different offset voltage for each VCA. Could you explain a little more?
moremagic
for sine/cosine LFO duty, you use a filter with multiple outputs set to LF, the different outputs are usually 90 degrees out of phase. the serge vcfa and intellijel dr octature go sub audio, and i would imagine there are others. (maybe not MU since everybody wants to climb a ladder?
ranix
You're right moremagic, the low pass and band pass outputs of my state variable filters are 90 degrees out of phase and make a beautiful circle.

They won't self-oscillate at lfo rates so I might need to try passing an lfo through. The filters are AC coupled though so I don't know how low I can go. I've been meaning to check to see if you can defeat a coupling cap to DC couple a Q107

but that might not even be a thing that makes sense
Navs
Hovercraft wrote:
Navs wrote:
Have a look at the Doepfer Morphing Controller or Interpolating Scanners (Haible, Tillman, Toppobrillo) for ideas. The point is that your one CV actually becomes four CVs. There are other ways to do the maths (Wiard JAG) but iirc the problem is not sending your signal to the edges, but equally to the middle of the four speakers.


Thanks--will look into those modules, but more interested in constructing something out of basic modular functions. As for spread or radius, I figured I could send an offset to all four VCA's to move a source towards the center.

edit: That Doepfer A144 looks like it performs the function I described. The question is--how can it be replicated with simple modules?


Yes, I was suggesting that you study them to find a solution, not buy them wink
Hovercraft
Navs wrote:


Yes, I was suggesting that you study them to find a solution, not buy them wink


Doepfer's Law states that any modular problem can be solved with the purchase of a sufficient number of the correct modules. w00t
Graham Hinton
Navs wrote:
There are other ways to do the maths (Wiard JAG) but iirc the problem is not sending your signal to the edges, but equally to the middle of the four speakers.


That will be the problem with all these suggestions.

Is there any reason why two CVs, e.g. from a joystick X & Y, can't be used? A single CV does not carry enough information.
What is needed is a LR pan law and a FB pan law added at each VCA. This can be extended for 5.1 with a LCR pan law in the front.
Hovercraft
Graham Hinton wrote:
Navs wrote:
There are other ways to do the maths (Wiard JAG) but iirc the problem is not sending your signal to the edges, but equally to the middle of the four speakers.


That will be the problem with all these suggestions.

Is there any reason why two CVs, e.g. from a joystick X & Y, can't be used? A single CV does not carry enough information.
What is needed is a LR pan law and a FB pan law added at each VCA. This can be extended for 5.1 with a LCR pan law in the front.


My question is based on a feature already present in the Koma Poltergeist (and possibly the Buchla 227e). The Poltergeist's pan input modulates the signal's pan position 360 degrees, with a voltage from 0 to 8V. What I have yet to discover--is how that circuit works.

https://koma-elektronik.com/new/wp-content/uploads/2016/03/POLTERGEIST -MANUAL-final.pdf
oneunkind
When they say '360 degrees', do they mean the module is just fading across the 4 speakers? If so, because of speaker placement this would cause the sound to be pannable in a circle around the listener, and, definitely, you could easily place a sound anywhere along the perimeter of the 'circle'(rectangle) with a linear control but never anywhere inside of it. Whether this is really quad panning or just circular panning or whatever ig is some semantics. But, whatever you call it.. it's kind of important to know because if that all the module is doing you already have some answers?
Hovercraft
oneunkind wrote:
When they say '360 degrees', do they mean the module is just fading across the 4 speakers? If so, because of speaker placement this would cause the sound to be pannable in a circle around the listener, and, definitely, you could easily place a sound anywhere along the perimeter of the 'circle'(rectangle) with a linear control but never anywhere inside of it. Whether this is really quad panning or just circular panning or whatever ig is some semantics. But, whatever you call it.. it's kind of important to know?


I think so--just a sequential pan between 4 outputs using a ramp or an envelope. If I can solve how it's done, I think I already have a solution to the issue of amplitude/radius.
ersatzplanet
I have found you need at least 2 CVs to move anywhere in the space. One for front/back, and one for left/right. You can do it with 5 panners. Feed the signal into a panner choosing front/back. The front and back outputs of this panner are each sent to a panner for left/right. The outputs of those two panners are your "corners" in a quad setup.

Another way is to get a Malekko JAG and four VCA. Two CVs into the jag give you the four corner CVs (actually you can get 8 outs from the JAG) that you drive the VCAs with.

You can do the "around the edges" panning with a A-144 and four VCAs. You can also use any sequencer that has a voltage to step function and lag the gate outs it gives. Something like a A-154/A-155 combo or an A-152 voltage addressed track & hold / switch (multiplexer) / digital outputs module that also has that function.
ranix
You need two CV to control the actual panning, but the motion of the two CV is linear and directional (as described in the original first post).

While you do indeed need two CV to control space in a 2d environment (X + Y), when the circle is predefined you only require one CV which represents the distance moved along the predefined circle.

This bit here defines the circle:
Hovercraft wrote:

For instance, I'd like to have the CV work like the following:
CV=0V - VCA 1 100% open, VCA 2/3/4 0% open
CV=2.5V - VCA 2 100% open, VCA 1/3/4 0% open
CV=5V - VCA 3 100% open, VCA 1/2/4 0% open
CV=7.5V - VCA 4 100% open, VCA 1/2/3 0% open


he's basically asking for a function that accepts input radians and returns X position and Y position

This is basically "simultaneous phase control of sin and cos of a stopped oscillator", as best as I can explain it.
kindredlost
EPTC wrote:

I opted instead to get the Dotcom Q148 and really love it. https://synthesizers.com/q148.html

I mostly work in stereo panning, and haven't tried quadrophonic with it, but its support of unipolar/bipolar panning CV suggests it wouldn't be difficult to achieve with two of them - or even just one of them with a little math and another send module.


The Q148 Pan/Fade module is not very useful as it is for panning into two channels of a stereo mix desk. If you send the outputs as separate signals and hard pan the mixer channels right and left then you get cancellation in the center due to the inverted channel output from the Q148. Of course there are workarounds such as sending one of the channels through a signal processor to (re) invert it prior to the final mixing. I built a rider board for my Q148 to invert one of the channels so it works as it should for a final pan module.
Hovercraft
ranix wrote:


...This is basically "simultaneous phase control of sin and cos of a stopped oscillator", as best as I can explain it.


That's one approach--the Serge Quad Locator panel has two quadrature oscillators with hold inputs. I think that's a cool idea, but unfortunately not an easily found osc control.

I appreciate all the helpful input. For a moment, it might be better to forget about any kind of X-Y motion. Think of it as using a single voltage to smoothly pan between four VCA's. The speakers could be in a straight line--or any shape. The voltage controls movement along one axis, and the physical placement of the speakers controls the other axis.
oneunkind
Hovercraft wrote:

CV=0V - VCA 1 100% open, VCA 2/3/4 0% open
CV=2.5V - VCA 2 100% open, VCA 1/3/4 0% open
CV=5V - VCA 3 100% open, VCA 1/2/4 0% open
CV=7.5V - VCA 4 100% open, VCA 1/2/3 0% open


You think you might want all the points at 50% between two VCAs... 0% is center panned front, for example? I haven't switched on my modular or fired up reaktor or anything on this though, (generally need constant empirical testing to keep my head straight so i'm probably wrong.) It seems like it would be easier than trying to get around the discontinuity another way, though.

I do remember adjusting this out on blinds when making sort of wave/filter scanning-type patches and i think i did it something like that. Not exactly the same thing, mixing sources into a single output, i know. But, i was able to fade across three inputs pretty smoothly using a single cv.

(eh, nvm doesnt seem to make a difference )

(pretty sure i remember what i did... multing the ramp into every input of blinds, ring modulating it with itself. if you do the knobs right, it will put out the right sequence of rising and falling voltages. then you stick those outputs into the vcas ... veils, i was using . tbh, though... i just got out of surgery weds night so im bored and loopy on meds so take this with a grain of whatever)

(frames seems like the easiest module solution by far? software is of course simple, with some table/sample lookups.)
Hovercraft
oneunkind wrote:


You think you might want all the points at 50% between two VCAs... 0% is center panned front, for example? I haven't switched on my modular or fired up reaktor or anything on this though, (generally need constant empirical testing to keep my head straight so i'm probably wrong.) It seems like it would be easier than trying to get around the discontinuity another way, though.
...


Aside from this one specific problem, I'm diagramming a full system for quad spatialization, and what you suggest is one of the approaches I've drawn out. No idea if it's a better, worse, or equivalent way to conceptualize and implement quad motion. A friend and I will be testing out different configurations. The small number of artists working in quad are using some different methods, just as electronic artists use different methods for two-channel spatialization.
MUFF WIGGLER Forum Index -> Modular Synth General Discussion Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group