through-zero FM

Anything modular synth related that is not format specific.

Moderators: luketeaford, Joe., lisa, Kent

User avatar
authorless
Super Deluxe Wiggler
Posts: 2320
Joined: Tue Feb 17, 2009 10:41 am
Location: Wilmington, NC

through-zero FM

Post by authorless » Wed Sep 16, 2009 8:51 pm

could someone give me a technical description of what through-zero FM is? i certainly like the way it sounds though i am not 100% sure i know what it means. and the modicum of description i have found is archaic at best. thanks.

User avatar
Fnord
Discordian Wiggler
Posts: 1997
Joined: Wed Apr 15, 2009 10:08 pm
Location: Dirty Jersey

Post by Fnord » Thu Sep 17, 2009 6:01 am

Ian Fritz on Richard Brewster's page explains it nicely:

http://pugix.com/synth/teezer-through-zero-vco/
Last edited by Fnord on Thu Sep 17, 2009 6:18 pm, edited 1 time in total.

User avatar
worker8
Ultra Wiggler
Posts: 792
Joined: Mon May 04, 2009 3:53 pm
Location: .fr
Contact:

Post by worker8 » Thu Sep 17, 2009 6:11 am

The oscillator stops (frequency = 0) and go "backward" (frequency below 0).

With a regular VCO, if the modulated frequency reach or go below 0, you won't hear anything. With a thru zero VCO, you'll hear "negative" frequencies.

Example : 600 Hz thru VCO carrier, modulated to +/- 800 Hz : you'll hear (-) 200Hz and 1400Hz. With a standard VCO, that 200 Hz sideband won't appear.

User avatar
Roycie Roller
Death Row Wiggler
Posts: 1434
Joined: Sat Jul 05, 2008 6:57 pm
Location: Sou-westerly

Post by Roycie Roller » Thu Sep 17, 2009 6:27 am

Thru-zero is beautiful in that respect.
With the right kind of modulation, eg. speeding up & quickly slowing down an LFO, the + & - frequencies can sound like delay.

frijitz
Super Deluxe Wiggler
Posts: 1637
Joined: Mon Aug 31, 2009 3:37 pm

Post by frijitz » Thu Sep 17, 2009 8:58 am

Fnord wrote:Richard Brewster explains it nicely:

http://pugix.com/synth/teezer-through-zero-vco/
Not to be too picky, but I believe if you look closely you will see that I wrote that explanation.

:grin:

Ian

User avatar
authorless
Super Deluxe Wiggler
Posts: 2320
Joined: Tue Feb 17, 2009 10:41 am
Location: Wilmington, NC

Post by authorless » Thu Sep 17, 2009 3:04 pm

but what exactly does "time reversed replica of the waveform" mean? is my thinking here: much like aliasing in digital sampling the lowest frequencies will become the highest in the harmonic spectra but they will also be phase-inverted?

frijitz
Super Deluxe Wiggler
Posts: 1637
Joined: Mon Aug 31, 2009 3:37 pm

Post by frijitz » Thu Sep 17, 2009 3:26 pm

authorless wrote:but what exactly does "time reversed replica of the waveform" mean? is my thinking here: much like aliasing in digital sampling the lowest frequencies will become the highest in the harmonic spectra but they will also be phase-inverted?
No, much simpler. Think in real time, not in spectral space. As frequency goes through zero, the wave runs slower and slower, stops, then reverses direction. Here is a photo of the TZ saw core in action:
http://electro-music.com/forum/topic-29149.html
It starts with down ramps that slow to a flat slope, then change direction to accelerating up ramps.

The other way to think of it is that a wave's phase is frequency times time, so reversing the sign of frequency has the same effect as reversing the sign of time, i.e., running backwards.

:grin:

Ian

User avatar
Karl Jeffers
Wiggling with Experience
Posts: 390
Joined: Fri Jul 18, 2008 2:13 pm

Post by Karl Jeffers » Thu Sep 17, 2009 3:46 pm

I was thinking, can you do "thru zero FM" with normal lin FM and soft sync?

User avatar
authorless
Super Deluxe Wiggler
Posts: 2320
Joined: Tue Feb 17, 2009 10:41 am
Location: Wilmington, NC

Post by authorless » Thu Sep 17, 2009 6:16 pm

thanks, fijitz. quite clear now.

User avatar
Fnord
Discordian Wiggler
Posts: 1997
Joined: Wed Apr 15, 2009 10:08 pm
Location: Dirty Jersey

Post by Fnord » Thu Sep 17, 2009 6:17 pm

frijitz wrote:
Fnord wrote:Richard Brewster explains it nicely:

http://pugix.com/synth/teezer-through-zero-vco/
Not to be too picky, but I believe if you look closely you will see that I wrote that explanation.

:grin:

Ian
Sorry about that Ian, I kinda suck when I roll out of bed and haven't had any coffee yet.

frijitz
Super Deluxe Wiggler
Posts: 1637
Joined: Mon Aug 31, 2009 3:37 pm

Post by frijitz » Thu Sep 17, 2009 8:51 pm

Fnord wrote: I kinda suck when I roll out of bed and haven't had any coffee yet.
LOL! Coffee first -- always -- around here.

:grin:

Ian

User avatar
bubblesound
Veteran Wiggler
Posts: 649
Joined: Tue May 12, 2009 10:38 pm

Post by bubblesound » Thu Sep 17, 2009 9:31 pm

it took spending an hour with the zeroscillator at scooter's house during a meet up for me to really get what these things are doing. the downside is now i think i have to have one...
i've tried nothing and i'm all out of ideas...

User avatar
Luka
Super Deluxe Wiggler
Posts: 5375
Joined: Tue Jul 15, 2008 9:15 am
Location: melbourne

Post by Luka » Thu Sep 17, 2009 9:39 pm

why exactly is fm synthesis more like digital fm synthesis when using tzvcos apposed to regular vcos with fm inputs?

User avatar
dkcg
I pity the fool w/o enough VCAs
Posts: 9142
Joined: Mon Jul 14, 2008 12:35 pm
Location: LA

Post by dkcg » Thu Sep 17, 2009 10:03 pm

bubblesound wrote:it took spending an hour with the zeroscillator at scooter's house during a meet up for me to really get what these things are doing. the downside is now i think i have to have one...
They are a LOT of fun. :)

Last time I talked to Cynthia on the phone I asked what through zero means. Some of it went a little over my head but that scope capture on EM explains a lot of what I'm hearing on the ZO. I'm not sure if other designs do this, but she told me that the ZO's triangle core actually flips (not sure if that meant it inverses) when it crosses into negative time. Whatever it is, I love how it sounds.

That reeeaahhhwooooo (if that makes sense) when the frequency is swept is so lovely, sounds like robotic cats in heat to me at times.

User avatar
incekt
Veteran Wiggler
Posts: 587
Joined: Tue Jan 20, 2009 12:30 pm
Location: Milwaukee WI

Post by incekt » Tue Sep 22, 2009 11:55 am

could you get the AFG to go through zero if you run the same signal into the exp and lin inputs? or would you have to invert one of them?hmm...

User avatar
urbanscallywag
Ultra Wiggler
Posts: 937
Joined: Sat Sep 05, 2009 10:20 pm
Location: San Diego, CA

Post by urbanscallywag » Sun Sep 27, 2009 4:24 pm

Luka wrote:why exactly is fm synthesis more like digital fm synthesis when using tzvcos apposed to regular vcos with fm inputs?
because the digital synths also use through-zero oscillators.

in addition the regular vco with fm inputs probably have an exponential rather than linear response.

User avatar
decaying.sine
Broken <--> Fixed
Posts: 5398
Joined: Wed Jun 03, 2009 10:05 pm
Location: New Haven, CT USA

Post by decaying.sine » Sun Sep 27, 2009 9:26 pm

frijitz's TZFM SAW VCO thread at electro-music is simply great. i'm trying to wrap my head around some of these concepts, and it is all very thought provoking. why are triangle and saw cores ideal for this application?

sorry if that is a dumb or irrelevant question.

gzifcak
Common Wiggler
Posts: 112
Joined: Sun Nov 23, 2008 2:03 pm

Post by gzifcak » Mon Sep 28, 2009 10:40 pm

i believe the reason that thru-zero FM sounds like classic digital FM is that digital FM (at least Yamaha FM) is actually phase modulation, which would explain the going "backward" when the modulation drives the oscillator past zero.

User avatar
Karl Jeffers
Wiggling with Experience
Posts: 390
Joined: Fri Jul 18, 2008 2:13 pm

Post by Karl Jeffers » Tue Sep 29, 2009 10:46 am

incekt wrote:could you get the AFG to go through zero if you run the same signal into the exp and lin inputs? or would you have to invert one of them?hmm...
I was thinking that you could lin fm an oscillator with eg a saw and then put a pulse of that same oscillator into the softsync input to inverse the waveshape. You need to set the pw so that when the saw is 0, the pulse softsyncs the fmed oscillator. It needs inversing probably to.
But I'm not that smart, so...

frijitz
Super Deluxe Wiggler
Posts: 1637
Joined: Mon Aug 31, 2009 3:37 pm

Post by frijitz » Tue Sep 29, 2009 10:38 pm

decaying.sine wrote:frijitz's TZFM SAW VCO thread at electro-music is simply great. i'm trying to wrap my head around some of these concepts, and it is all very thought provoking. why are triangle and saw cores ideal for this application?
It's also been done with Sin and Trap cores. The Trap core looks to be very reliable in operation. However it is quite complicated, and it's not clear how to make a Trap -> Sin shaper. It's pretty important to have a Sin output, IMO.

Offhand, I think the Sin core design will be very sensitive to OTA non-idealities. (Based on my work with quadrature oscillators.)

:grin:

Ian

User avatar
decaying.sine
Broken <--> Fixed
Posts: 5398
Joined: Wed Jun 03, 2009 10:05 pm
Location: New Haven, CT USA

Post by decaying.sine » Tue Sep 29, 2009 11:21 pm

thanks Ian. you just fucked up my brain worse :hihi:

Now I got some more to think and read about!

User avatar
tthogs
Common Wiggler
Posts: 203
Joined: Fri Nov 13, 2015 2:22 am
Location: Asheville, NC

Post by tthogs » Sun Mar 05, 2017 4:11 am

As far as I know, none of my vcos have thru zero. When I fm any of my vcos, the pitch goes up more and more as I increase the index. If a vco is thru zero does the pitch stay centered and just "orbit" further from the center pitch or does the pitch go up just like standard linear or exponential? Thanks

User avatar
listentoaheartbeat
Super Deluxe Wiggler
Posts: 2588
Joined: Sun Mar 30, 2014 7:42 am
Location: Berlin

Post by listentoaheartbeat » Sun Mar 05, 2017 4:41 am

I thought this discussion on the subject turned out to be very interesting:

viewtopic.php?t=136716

User avatar
Navs
Super Deluxe Wiggler
Posts: 4185
Joined: Fri Sep 18, 2009 3:49 am
Location: Berlin, Germany

Post by Navs » Sun Mar 05, 2017 4:49 am

tthogs wrote:As far as I know, none of my vcos have thru zero. When I fm any of my vcos, the pitch goes up more and more as I increase the index. If a vco is thru zero does the pitch stay centered and just "orbit" further from the center pitch or does the pitch go up just like standard linear or exponential? Thanks
Firstly, if they did have TZ, you'd know about it because few do and the manufacturers advertise the fact loudly! :hihi:

But the pitching you're experiencing is not an indicator of TZ capability. Even a TZ VCO can be affected by pitching if the modulator is not symmetrical, meaning it is riding on or contains a DC offset. That is what causes the pitching artefacts.

You can test this if your VCO has vanilla linear FM: make the carrier and modulator the same frequency and then apply some FM. If the modulator is clean, or if the input is AC-coupled (meaning it blocks any DC), then all you should hear is a change in colour.

User avatar
ranix
Super Deluxe Wiggler
Posts: 1763
Joined: Wed Jul 13, 2016 9:39 pm

Post by ranix » Sun Mar 05, 2017 6:04 am

one of the big problems that prevents analog linear FM from getting that "dx style" sound is that it's impossible to keep two oscillators in sync if one is modulating the frequency of the other. You can't do fancy stuff where the frequency of the modulator is an integer multiple of the frequency of the carrier, and especially not with multiple modulators. And you can forget using the FM'd output of one carrier to modulate the frequency of another carrier with everything in sync, that's just not going to happen.

the best module I've found so far for doing really super cool FM stuff is the happy nerding fm aid. I have several. I have kind of a big system and plenty of oscillators to mess with them but even one of them can do some really nifty stuff with a mixer and 4 oscillators. It takes a carrier saw and a modulation source as input and outputs your phase modulated sine wave. Since the carrier oscillator isn't affected by the FM, you can keep the carrier and as many modulators as you want in sync and do all the fun fancy FM shenanigans.

it's a little noisy but some of that is probably caused by my mistakes since I haven't had these for very long

also it sounds pretty freakin' crazy when you sync sweep an fm modulator

Post Reply

Return to “Modular Synth General Discussion”