Oscillator Sync. How Complex Do You Like It?

Anything modular synth related that is not format specific.

Moderators: Kent, Joe., luketeaford, lisa

Post Reply
User avatar
KSS
Super Deluxe Wiggler
Posts: 4174
Joined: Mon Jan 25, 2016 7:28 am

Oscillator Sync. How Complex Do You Like It?

Post by KSS » Sun Nov 15, 2020 10:50 pm

Is the ability to sync VCOs important to you?
Is hard sync enough?
What about soft sync?

Have you explored Osc sync beyond the basics above?
For example, the ability to sync off of any input waveform.
Or setting a sync level, where two different VCOs can sync diffferently from the same signal.
Likewise, having a waveform edge select switch to choose not only the level but also the direction which applies sync.

Xomrys
Common Wiggler
Posts: 104
Joined: Wed Feb 26, 2020 12:54 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by Xomrys » Sun Nov 15, 2020 11:44 pm

Yes. Combined with FM, and in a 3+ oscillator setup, it is extremely important for getting those complex tones otherwise available through waveshaping Yamaha-style FM, wavetables etc.

Hard sync definitely.. just make sure that the frequency sweep sounds smooth, this is a problem in some implementations.

Soft sync is important. However most implementations are bad. There are two ways to go: a 'hard' soft sync that always locks to many different harmonic and subharmonic relationships, with a timbre change, and a 'soft' soft sync, that locks only when you're very close, with no perceptible timbre change.

In terms of the 'hard' soft sync, the prototype is the Roland system 100 weak sync. I haven't heard any other that sounds as good as this.
The 'soft' soft sync is just an idea, I haven't really heard it. But it should be possible in theory. If the sync window is small, the issue may be getting it to track 8 octaves or whatever.

I wouldn't do a user-controllable, variable sync, I would focus on making the hard sync sound really smooth, and just nailing the other two soft sync settings.

Quadrature sync may be a new option for getting softer sync. This works by providing two waveforms from the reference oscillator 90 degrees out of phase, for instance saw and triangle. I did a PLL patch once using this method. I think there is still unexplored territory here.

All the other sync strategies -- variable sync point, I don't see that they would add much, and it would give maybe too many 'unmusical' choices to the user. A possible PWM effect maybe, but it's better to focus on actual PWM, Saw/triangle morphing, and so on. Maybe 'behind the scenes', it's necessary to do tweaks in this area to tune the different sync options which are exposed to the user.

User avatar
dubonaire
Super Deluxe Wiggler
Posts: 7315
Joined: Sun Jan 17, 2010 10:45 pm

Re: Oscillator Sync. How Complex Do You Like It?

Post by dubonaire » Mon Nov 16, 2020 12:20 am

I use hard sync with FM and when adding oscillators together to avoid phasing issues.

The two earlier versions of Intellijel have a flip sync which flips the waveform, I didn't find that particularly useful.

User avatar
sir stony
Wiggling with Experience
Posts: 455
Joined: Mon Feb 25, 2019 6:09 pm
Location: Bavaria

Re: Oscillator Sync. How Complex Do You Like It?

Post by sir stony » Mon Nov 16, 2020 2:51 am

KSS wrote:
Sun Nov 15, 2020 10:50 pm
For example, the ability to sync off of any input waveform.
Or setting a sync level, where two different VCOs can sync diffferently from the same signal.
Likewise, having a waveform edge select switch to choose not only the level but also the direction which applies sync.
These features would be a dream in any dual LFO (which is likely the alter ego of many a VCO). I'm not sure about the impact of such fine differences in audio range, though.
What I'd be curious about is a cyclic sync, reset for only every nth edge of the input wave, and best have that under cv control.

User avatar
starthief
Super Deluxe Wiggler
Posts: 4820
Joined: Tue Oct 04, 2016 10:39 am
Location: St Louis, MO
Contact:

Re: Oscillator Sync. How Complex Do You Like It?

Post by starthief » Mon Nov 16, 2020 7:22 am

Intellijel Shapeshifter has several different sync modes. Mostly aside from hard sync they sound unusual and kind of off, but when using it for modulation sources it can be interesting (particularly when modulating the ratio). I haven't really played with it at LFO rates, which might be where those modes really shine.

I don't usually like to use sync with FM because the timbre is rougher (unless that's what I'm going for of course). But I found with the E352/E370, it can be helpful to temporarily use sync to get the relative tuning perfect before unpatching sync -- those modules are capable of extreme TZFM depths and also have DC coupled inputs so they are just generally trickier than many.

Xomrys
Common Wiggler
Posts: 104
Joined: Wed Feb 26, 2020 12:54 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by Xomrys » Sat Nov 21, 2020 6:53 pm

Yeah, I don't go for the rough timbres that much. With two oscillators, if one is hard-synced to the other at a very high frequency it sounds like a resonant filter peak. A little sawtooth FM from the first oscillator can help spread out this peak, which makes either a guitar-type sound, or if the synced oscillator is in formant range, a vocal-like formant peak (which then usually needs to be hit with a resonant LPF). A little vibrato you can sometimes get a haunting vocal sound. Then it's often needed to mix in a sine from the first (reference) oscillator in case the sound is too thin.

However, FM from one oscillator only goes so far. This is where a third oscillator, and soft sync helps. With soft sync you can FM the second oscillator by the third, with a pitch that's harmonically or subharmonically related to the first. This opens up possibilities for Yamaha FM-style tones, especially if a triangle wave is used as the modulator and a square for the carrier. Some of these can be metallic or industrial sounding, but also very beautiful and 'acoustic' if the frequencies are higher.

Try for instance: the second oscillator seven octaves above the first, the third (FM oscillator) five octaves above. These frequency relationships seem ridiculous, but the result can be a very convincing, soft bell-like tone. [Just remember to mix in a sine or triangle from the fundamental osc.]

Still more can be done (in theory, I don't have enough oscillators) , you can have two groups of three oscillators, the first 'fundamental oscillator' soft-synced to the second, so they're all in octave relationships.

Or just introduce.a fourth, an octave below the fundamental, to FM the third osc. that's FM-ing the second...

Anyway I have in mind crystalline-clear, FM-style or wavetable tones, and with these three-oscillator patches a lot is possible. You get some of the timbral variety of digital, with the lack of aliasing and other 'sound' of analog. And it's all very traditional 'East-coast...'

I prefer this approach to through-zero FM because of the difficulties I've encountered with sync, which is another issue. My hunch is that PLL sync may give better results here.

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 7279
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Re: Oscillator Sync. How Complex Do You Like It?

Post by Rex Coil 7 » Sat Nov 21, 2020 7:24 pm

KSS wrote:
Sun Nov 15, 2020 10:50 pm
Is the ability to sync VCOs important to you?
Is hard sync enough?
What about soft sync?

Have you explored Osc sync beyond the basics above?
For example, the ability to sync off of any input waveform.
Or setting a sync level, where two different VCOs can sync diffferently from the same signal.
Likewise, having a waveform edge select switch to choose not only the level but also the direction which applies sync.
TOTALLY dig hard sync sweeps. Also use of hard sync ~tones~ (unswept). Soft sync ... hmm ... seems like it might be VCO specific on whether or not it's useful. The DotCom Q106 VCO soft syncs well with another Q106, there's all of these sweet spots (sorry for the overused cliche there). I've not tried a whole wheelbarrow full of other VCOs with the DotComs (hard or soft sync) but the ones I have tried sound differently when soft-sync'd with a Q106 ... not badly ... just differently.

Some VCOs have many types of ~sync~ ... the Intellijel Shapeshifter has some list of sync types ... no clue if any of it is worth the trouble though.

Hard sync is also important to me for synchronizing LFOs to ~whatever~ or VCO to yet other ~whatevers~.

So ... YES .... hard sync ... soft sync (with sweep-adjustable "depth" .. not a depth switch .. like the DotCom)... add them both to the list of features in whichever module you're finalizing in your soon to see system.

Gotta jet ... supper's up!

:)
5U MODULAR NORMALIZING PROJECT (for your entertainment) viewtopic.php?t=78836&highlight=
.. given the choice between conformity and self respect, I choose the latter.
.. dominion - noun: control or the exercise of control . power . possessed and controlled domain . sovereignty . having dominion over the world . supreme authority . absolute ownership . power . authority . jurisdiction . control . command ... power ....... power .......... power.

DSC
Super Deluxe Wiggler
Posts: 2295
Joined: Thu Apr 11, 2013 11:54 pm

Re: Oscillator Sync. How Complex Do You Like It?

Post by DSC » Sat Nov 21, 2020 11:09 pm

...
Last edited by DSC on Thu Nov 26, 2020 8:28 pm, edited 1 time in total.

User avatar
KSS
Super Deluxe Wiggler
Posts: 4174
Joined: Mon Jan 25, 2016 7:28 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by KSS » Sun Nov 22, 2020 12:37 am

@DSC
Your panel drawing shows pos and neg hard sync, along with undefined soft sync inputs. Assume basic 3340 'reference ckt' soft sync? One switch is missing from the first OSC? Or do they all master their left osc? EDIT: What does the Toggle next to the sync PB do? Between it and Coarse knob./edit

That's interesting to have the three jacks input rather than the SPDT pos-neg-both I used below. Not sure how I'd fit it in. 1004A's Panel and IO already filled up. Uses a similar DPDT pushbutton to yours for engage-disengage sync. Left side is current bronze. Gave up sync level and the required dual concentric pot and knobs to support its inclusion. Likely to lose all the pull pots too. A result of using less expensive panel PCB construction. May have to bring them back with a fly-wired version. Toggles seen as filled black rings can mount horizontal or vertical, according to preference. Labels work both ways.
Attachments
APT1004A-5a.JPG

Xomrys
Common Wiggler
Posts: 104
Joined: Wed Feb 26, 2020 12:54 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by Xomrys » Sun Nov 22, 2020 6:45 pm

I'm thinking more, I'd really like to see PLL sync as a third option (to weak and soft), as it's so rare. When tuned correctly, and with negative pitch feedback, it can be a very smooth sync that activates only when you're close to a harmonic. However, when misused (flipped to positive feedback) you get chaotic textures. I like both the smoothness and the unique noisy effect of flipping the phase control.

The only difficulty is requiring both triangle and saw waveforms from the reference oscillator. You need this quadrature relationship to determine whether phase is lagging or leading, at least I don't know of some other way.

In theory, triangle can be derived from saw, through waveshaping. This makes patching easier, but requires more components that are probably duplicated anyway on the first oscillator.

To patch a PLL from scratch is a pain, it requires a number of modules -- which is why it's better to have all this functionality inside the oscillator.

First is XOR (and implicit comparators) for the triangle waves of both oscillators. XOR result gives a thin pulse near the zero-crossings of the triangle -- where one triangle is negative and the other positive, the pulse will be high. Otherwise it will be low when both triangles have the same 'sign'. So the width of the triangle-XOR-pulse determines how close the two oscillators are in sync, but not the direction of the sync.

For the direction, the pulse may be multiplied (Linear VCA) by the difference in the saw waves of both oscillators. This difference will be strictly positive or negative depending on which oscillator is leading in phase. There is no risk of measuring the difference around the saw waves' discontinuity, because the triangle-XOR-pulse gates out this region.

So we have a scaled pulse, the width determined by the magnitude phase difference, the sign (direction) determined by lead or lag.
This difference can be smoothed by a lowpass filter that is DC coupled, and accepts very low frequencies. The output of the LPF is the phase comparator signal. It can then be inverted and set to control the frequency of the second oscillator -- this completes the 'control loop' of the PLL.

So if you put this small 'patch' (XOR, Linear VCA, subaudio-range LPF) inside the oscillator, you get a good PLL. It's up to the designer whether to ask for two sync inputs (saw and triangle), or to derive the triangle from the saw. I think this would open up a lot of possibilities.

DSC
Super Deluxe Wiggler
Posts: 2295
Joined: Thu Apr 11, 2013 11:54 pm

Re: Oscillator Sync. How Complex Do You Like It?

Post by DSC » Sun Nov 22, 2020 7:04 pm

...
Last edited by DSC on Thu Nov 26, 2020 8:28 pm, edited 1 time in total.

User avatar
starthief
Super Deluxe Wiggler
Posts: 4820
Joined: Tue Oct 04, 2016 10:39 am
Location: St Louis, MO
Contact:

Re: Oscillator Sync. How Complex Do You Like It?

Post by starthief » Sun Nov 22, 2020 10:23 pm

Xomrys wrote:
Sun Nov 22, 2020 6:45 pm
To patch a PLL from scratch is a pain, it requires a number of modules -- which is why it's better to have all this functionality inside the oscillator.
If you're just tracking an oscillator with a simple shape, you honestly can patch a PLL with a unipolar VCA -- albeit not a very good one. A bipolar VCA works a lot better, or a bipolar comparator (like the "Box" output of Mystic Circuits Ana). You don't need to get fancy, you just need to:

-- make sure the voltage range of the PLL output signal is appropriate (if the max voltage is too low, or the min voltage too low, it won't be able to push the frequency enough to track successfully).

-- make sure the VCO that receives the PLL output signal is analog, or if digital, samples the pitch/FM inputs at a high sample rate.

-- if you FM either of the oscillators involved in the PLL circuit you might set up a feedback loop that the PLL is unlikely to handle well. You could use FM Aid, or filter FM, or phase modulation with something like Generate 3 where you have a non-phase-shifted output to feed the PLL.

Xomrys
Common Wiggler
Posts: 104
Joined: Wed Feb 26, 2020 12:54 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by Xomrys » Sun Nov 22, 2020 11:28 pm

starthief wrote:
Sun Nov 22, 2020 10:23 pm
Xomrys wrote:
Sun Nov 22, 2020 6:45 pm
To patch a PLL from scratch is a pain, it requires a number of modules -- which is why it's better to have all this functionality inside the oscillator.
If you're just tracking an oscillator with a simple shape, you honestly can patch a PLL with a unipolar VCA -- albeit not a very good one. A bipolar VCA works a lot better, or a bipolar comparator (like the "Box" output of Mystic Circuits Ana). You don't need to get fancy, you just need to:

-- make sure the voltage range of the PLL output signal is appropriate (if the max voltage is too low, or the min voltage too low, it won't be able to push the frequency enough to track successfully).

-- make sure the VCO that receives the PLL output signal is analog, or if digital, samples the pitch/FM inputs at a high sample rate.

-- if you FM either of the oscillators involved in the PLL circuit you might set up a feedback loop that the PLL is unlikely to handle well. You could use FM Aid, or filter FM, or phase modulation with something like Generate 3 where you have a non-phase-shifted output to feed the PLL.
I'm having trouble picturing this. Are you assuming already the existence of a PLL module? Otherwise I'm unsure how the VCA alone can get you the phase comparator signal. My 'patch' creates the PLL module from scratch using a 'normal' VCO and a few extra low-level modules (XOR, CV mixer, VCA, LPF). The second 'normal' VCO is transformed by this means into the PLL, which works also at harmonics and under FM (with some limitation). My proposal is that the extra modules be built inside the VCO...

User avatar
KSS
Super Deluxe Wiggler
Posts: 4174
Joined: Mon Jan 25, 2016 7:28 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by KSS » Mon Nov 23, 2020 1:59 am

DSC wrote:
Sun Nov 22, 2020 7:04 pm
Toggle allows you to remove the attenuated voltage coming from the course tuning knob going to pin 15 on the CEM3340 (Frequency Control). Seems to let it 'roam' freely when removing it(turning off).
Nice! That's a neat trick. The way you describe it you have 4 states? Two switches w/two states each. Would you lose too much to have the 2nd pole of the SYNC pushbutton automatically remove that coarse input as sync is applied?

Does the toggle also remove the fine knob's contribution to the 3340's FREQ input?

More than a little curious how that toggle so close to the Coarse knob works for you in actual performance? I added a toggle like that -close to a CLK rate knob in the 1036 module. Have played with having it real close like yours, where it sort of falls into the 'shadow' of the knob vs having it farther away. But not far enough to let a full finger pass between them as there's not enough panel room. I swapped in a PB, but that only has two states, and I want the three I can get with a toggle. Looks like yours is a normal bat length? Short bats are a mixed bag. Less in the way, but harder to switch. To me, having it in the 'shadow' area lets my fingers sort of lift away from the knob and flow around it fairly seamlessly. Better than when its farther away.<-- but still close enough to need to be avoided as you turn the knob it's near.
You seem to have a Star Trek 'Next Generation' feel to your design. Are you a Trek fan? :tu:
Really? I'm not seeing it as such, but I have been watching Star Trek since first airings of TOS as a child. And did watch quite a few TNG. Lost interest as Janeway appeared<--There's an orange alien on that one who reminds me of a JarJarBinks type character AKA annoying AF. Am a Bakula fan cuz QL, but have only seen a few Enterprise episodes. Chris Pine? Not so much. I own the Star Trek Compendium book but have never gone to a convention or dressed up. So fan? Not sure. I don't know why the Klingon's changed their look -besides the practical obvious reason, that they could!
But I do appreciate Gene Roddenberry's vision and how it played on TNG and this reply shows I am familiar with some parts of the whole genre. Don't care much for what that new director guy who keeps ruining various Sci-Fi genre's has done with his Star Trek vision.

50SQ Looks like TNG? I wonder if it will still seem that way to you once these garish graphics are replaced with their actual panels and parts. Hopefully those arrive soon.

User avatar
starthief
Super Deluxe Wiggler
Posts: 4820
Joined: Tue Oct 04, 2016 10:39 am
Location: St Louis, MO
Contact:

Re: Oscillator Sync. How Complex Do You Like It?

Post by starthief » Mon Nov 23, 2020 10:02 am

Xomrys wrote:
Sun Nov 22, 2020 11:28 pm
I'm having trouble picturing this. Are you assuming already the existence of a PLL module?
Nope!

Leader oscillator output -> VCA input
Follower oscillator output -> VCA CV input
(or reverse those two)
VCA output -> follower oscillator V/OCT or expo FM input

It's literally that simple. Although the VCA might have to amplify, not just attenuate, to give enough voltage range to track, and like I said, a bipolar VCA is much better.

I sold my Doepfer PLL some time back when I discovered I could track just as well only using the logic algo in Disting. I eventually sold the Disting and am using Mystic Circuits Ana or Blinds for my PLL patching now... or Xaoc Tallin just to prove that it works.

Xomrys
Common Wiggler
Posts: 104
Joined: Wed Feb 26, 2020 12:54 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by Xomrys » Mon Nov 23, 2020 4:51 pm

starthief wrote:
Mon Nov 23, 2020 10:02 am
Xomrys wrote:
Sun Nov 22, 2020 11:28 pm
I'm having trouble picturing this. Are you assuming already the existence of a PLL module?
Nope!

Leader oscillator output -> VCA input
Follower oscillator output -> VCA CV input
(or reverse those two)
VCA output -> follower oscillator V/OCT or expo FM input
Great patch! Very simple and elegant, I see now how the VCA works as a phase comparator. and the FM implicitly smooths the result, because phase is the integral of frequency.

Thanks much,, not so often I run across tips this good...

DSC
Super Deluxe Wiggler
Posts: 2295
Joined: Thu Apr 11, 2013 11:54 pm

Re: Oscillator Sync. How Complex Do You Like It?

Post by DSC » Mon Nov 23, 2020 4:52 pm

...
Last edited by DSC on Thu Nov 26, 2020 8:29 pm, edited 1 time in total.

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 7279
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Re: Oscillator Sync. How Complex Do You Like It?

Post by Rex Coil 7 » Mon Nov 23, 2020 5:36 pm

... my wife picked this up for me around the early 1990s ... interesting stuff ... :tu:
star_trek_the_next_generation_technical_manual_us_1st.jpg


...now, if we could just crack that anti-matter powered matter conversion tech and faster than light computer transfer and calculation speeds... all of the ills of this wretched life will just go away ...

:yay:

.. or something.
:hmm:
5U MODULAR NORMALIZING PROJECT (for your entertainment) viewtopic.php?t=78836&highlight=
.. given the choice between conformity and self respect, I choose the latter.
.. dominion - noun: control or the exercise of control . power . possessed and controlled domain . sovereignty . having dominion over the world . supreme authority . absolute ownership . power . authority . jurisdiction . control . command ... power ....... power .......... power.

DSC
Super Deluxe Wiggler
Posts: 2295
Joined: Thu Apr 11, 2013 11:54 pm

Re: Oscillator Sync. How Complex Do You Like It?

Post by DSC » Mon Nov 23, 2020 5:50 pm

...
Last edited by DSC on Thu Nov 26, 2020 8:29 pm, edited 1 time in total.

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 7279
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Re: Oscillator Sync. How Complex Do You Like It?

Post by Rex Coil 7 » Mon Nov 23, 2020 10:06 pm

DSC wrote:
Mon Nov 23, 2020 5:50 pm
Rex Coil 7 wrote:
Mon Nov 23, 2020 5:36 pm
.. or something.
:hmm:
Yep, that's the one. Must be the weed I'm smokin' but don't you see some similarity with KSS's panel art?
Now if you could just backlight all your new panels like in the next generation version, you'd be set :tu:
... hire a bunch of art student interns to paint computer touch screens on glass that's been backlit? (nyuk nyuk .. :lol: ) ... bad wisecrack there ... :roll: ... referencing the way those screens in STNG were done on the sets. I see it .. a little .. maybe .. I gave up on the cannabis stuff a few months ago (that's a rich man's privilege, that shit's expensive!), maybe that's a prerequisite to ~see it~ more clearly or perhaps it... OH HOLY SHIT OF SHITS I SEE IT!!!!!!!

I scrolled down for just a second while drafting this, the post with the proto panel slid by and damned if I didn't immediately see that ~it~ ... that ~thing~ ... there's a sortof proto-sketch STNG feel about it, or perhaps a storyboard level of design refinement pre-completion of the graphics .... polish it up just a skosh, add colors and add backlighting ... and...

ENGAGE !!!

Totally see it now.


... ok wait ... this means I'm stoned? ... woa ... :lol:



.
5U MODULAR NORMALIZING PROJECT (for your entertainment) viewtopic.php?t=78836&highlight=
.. given the choice between conformity and self respect, I choose the latter.
.. dominion - noun: control or the exercise of control . power . possessed and controlled domain . sovereignty . having dominion over the world . supreme authority . absolute ownership . power . authority . jurisdiction . control . command ... power ....... power .......... power.

User avatar
mosorensen
Wiggling with Experience
Posts: 364
Joined: Sun Sep 16, 2018 5:19 pm
Location: New Hampshire

Re: Oscillator Sync. How Complex Do You Like It?

Post by mosorensen » Wed Nov 25, 2020 8:27 am

Sorry to go off topic, but @Rex Coil 7, that post is epic. Thanks!

oldgearguy
Veteran Wiggler
Posts: 664
Joined: Thu May 26, 2016 6:42 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by oldgearguy » Wed Nov 25, 2020 8:46 am

DSC wrote:
Mon Nov 23, 2020 5:50 pm
Rex Coil 7 wrote:
Mon Nov 23, 2020 5:36 pm
.. or something.
:hmm:
Yep, that's the one. Must be the weed I'm smokin' but don't you see some similarity with KSS's panel art?
Now if you could just backlight all your new panels like in the next generation version, you'd be set :tu:
My first thought was ARP 2500 panel homage/reference actually.

User avatar
KSS
Super Deluxe Wiggler
Posts: 4174
Joined: Mon Jan 25, 2016 7:28 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by KSS » Wed Nov 25, 2020 5:25 pm

oldgearguy wrote:
Wed Nov 25, 2020 8:46 am
My first thought was ARP 2500 panel homage/reference actually.
More than an homage or reference. It *is* the 1004 VCO and T,P mixer circuits. In the original size. Fully backward compatible*.
With some additions. Like the sync which is the topic of this thread.
Since the original 1004-P,R,T do not have sync available.

*Using a cage and edge connector adapter since the new APT 50SQ modules are not nearly as deep.

grrrwaaa
Learning to Wiggle
Posts: 10
Joined: Wed Sep 16, 2020 8:00 pm

Re: Oscillator Sync. How Complex Do You Like It?

Post by grrrwaaa » Sat Nov 28, 2020 9:23 pm

Leader oscillator output -> VCA input
Follower oscillator output -> VCA CV input
(or reverse those two)
VCA output -> follower oscillator V/OCT or expo FM input
Just tried this. I'm not 100% sure I follow how this is working, but something is kind of working. It's a bit AM, a bit FM/PM, a bit sync really, and only "works" in certain ranges, but it does generate some very interesting sounds... led to a nice 30 minute noodle-to-tape.

I used Dixie II+ as leader and Electrosmith 3340 VCO as "follower". (Also tried it the other way and it worked too, but wasn't quite as rewarding somehow). I think my VCA only does up to 1x or 2x maybe, I have the feeling that more gain on the VCA would make this even more gnarly. Also my VCA is not bipolar on the CV input.

Seems to work better when follower is similar or higher octaves than the leader. I wasn't sure if it dependended on particular waveforms, but all of sine vs sine, pulse vs pulse, saw vs saw worked, in different ways. Tried different waveforms against each other and it sometimes works, sometimes doesn't. Pulse is interesting because pwm seems to affect how well it works, or rather, the tendency to be sloppy or hit other interesting harmonics. Fundamental (tuning) of the "follower" acts like a kind of spectral brightness. I bet having a bipolar VCA (RM) would do interesting things, but I don't have one.

I was trying to work out *why* this works (ish). I come from a digital DSP background more than hardware/EE, so I think of a VCA as a multiplier of amplitudes. Taking the square as the simplest case, if both leader and follower are "high" then the output is high, whereas if either is low then the output is low -- the VCA acts as an AND at some high voltage (maybe 5v?). Sticking this into v/oct means the follower will run super fast, and very quickly the square will go low again, which will zero VCA once more, I guess then running at its normal fundamental frequency, until that makes it go high again. As long as the follower is higher frequency than the leader, this means the leader will dominate its behaviour (mostly), and we end up with a wave that has an asymetric pulse ratio dependent on the leader/follower ratio. It's really more like a kind of oscillator sync, which means this isn't off topic :tu:

When I try to think about the saw case, my brain starts to melt. I don't have a scope, but I imagine it bends the saw?

Anwyay, thank you so much, this has given me a lot to dig into. I love these kinds of posts/discussions on MW -- like the 'modules that should exist' and 'small discovery' etc. threads.
Attachments
image.png
image.png (527 Bytes) Viewed 87 times
image.gif
image.gif (248 Bytes) Viewed 87 times
It is so nice when audio-rate modulation actually works

User avatar
KSS
Super Deluxe Wiggler
Posts: 4174
Joined: Mon Jan 25, 2016 7:28 am

Re: Oscillator Sync. How Complex Do You Like It?

Post by KSS » Sat Nov 28, 2020 9:40 pm

A balanced modulator aka ring mod is a "bipolar vca". In fact, we've seen modules where the balmod was offset on one of its inputs to more closely fill in for a 'plain' vca. PAIA 4710 and ARP1035 are two examples. MC1496 and 4014 are the multipliers used. ARIES used an MC1596 in similar fashion.

TTSH or ARP2600 ring mod would work well in this patch.

The patch shared is a ring-bal mod patch using a VCA as partial fill-in for the full multiplication of a true balanced modulator.

We see the multiplier used in this way when reading about phase detection in the datasheets for PLL chips too. Like the CD4046.

Post Reply

Return to “Modular Synth General Discussion”