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

Zeroscillator calibration
MUFF WIGGLER Forum Index -> Modular Synth General Discussion  
Author Zeroscillator calibration
worker8
Hi there,

I got a Zeroscillator (eurorack, purchased second hand, not one of the recent shalow run) and ran through the calibration procedure.

I was suprise to see there's a single trimmer for V/oct calibration. Because I've noticed a break in the calibration curve I got using Volta. Usually, those Volta calibration profile are linear. But with the Zeroscillator, it's more a two segment profile. And the position of the break depends on the center frequency I set with the big tuning knob. Drawback, if I use Volta : whenever I change the center frequency of the Zeroscillator, I'll have to calibrate it again ! And if I try a regular midi2CV converter, even after calibrating the ZO, there will be a range where notes are out of tune (lower notes, actually).

I guess I've read the entire calibration procedure, so I don't know what to do next.

Any idea ?
Navs
worker8 wrote:
whenever I change the center frequency of the Zeroscillator, I'll have to calibrate it again !


Isn't that the case with all VCOs? I only have experience with Silent Way, not Volta, so maybe it does it differently. Ideally, it should map the required CVs, rather than calculating an average.

In general, though, I spent some time trying to calibrate the ZO using a tuner and then SW. I finally got my best results using my ears.

If you're trying to get accuracy for the TZ LinFM, here's another observation: I think it sounds better without Silent Way control. At certain modulation depths, the accuracy of the modulator seems more important.
Just me
I found my ZO's to be very limited in span. I could tune them, but they would only be accurate for about 18 semitones and then go wild. (I mean out by full steps or even octaves.) The entire suite of modules need to be dead on for it to work. (The modulators and the ZO.)
Savage
Just me wrote:
I found my ZO's to be very limited in span. I could tune them, but they would only be accurate for about 18 semitones and then go wild. (I mean out by full steps or even octaves.) The entire suite of modules need to be dead on for it to work. (The modulators and the ZO.)


Exactly my experience, as well, with a ZO. I could get an octave to an octave and a half, and beyond that everything goes wacky. So I typically use a ZO for additional coloring to a sound rather than being the primary pitch source.
parasitk
I use mine with Volta all the time and get several octaves out of it. And yeah the curves are a little wackier than with my other VCOs, but considering w/o the software calibration people are saying they don't get much beyond an octave or two, I'll take what I can get!
worker8
Thank you all for your comments,
Navs wrote:
If you're trying to get accuracy for the TZ LinFM, here's another observation: I think it sounds better without Silent Way control. At certain modulation depths, the accuracy of the modulator seems more important.

Thanks to Volta, I can play 3-4 octaves with an FM patch (rarely 4 to be honest). I run two or three instances of Volta, one for the carrier (ZO) and one or two for the modulator(s) (A-105 or A-143-9).
Unfortunately, the calibration process is unpredictable in a way. So the quality of the timbre may vary from day to day. Even if I don't touch the center frequency of the ZO in the meantime. Temperature, humidity, position of the planets ... Sometimes I just can't recall a previous Volta calibration profile. I need to make a new one. I name those calibration profiles with the pitch of the VCO without CV at the V/oct input.
Navs wrote:
Isn't that the case with all VCOs?

Perhaps. But I've been able the change the coarse frequency of my modulators without any major impact on the V/oct tracking quality (playing with Volta). And the timbre was still pleasant over a large range (although I didn't measure it). I think the reason is their calibration profile is more "linear" (when you look at them on Volta).

Actually, I can detune my VCOs within Logic. But I'd like to make it with knobs rather than the mouse.

I see two ways to live with it :
1/ using Volta, setting pitch intervals on Logic.
2/ setting the center frequency as low as possible, to push the non linear part of the calibration profile into sub audio range.

I'll try to post some pictures or a video.
solitaryzen
The poor tracking of the ZO is a real shame, as I love everything else about it. I wonder if the more recent (parallel mounted PCB) versions track any better?
frijitz
worker8 wrote:
Any idea ?

What bias setting are you using? That affects the tracking big time.

Ian
pugix
Have you trimmed the triangle wave shape to be symmetrical? That might impact tracking, too.
Lyonel
solitaryzen wrote:
I wonder if the more recent (parallel mounted PCB) versions track any better?


Just tested a last iteration ZOe. one octave tracking is the very big maximum you may expect from it.
worker8
Bias switch up most of the time.

I've tuned the triangle shape to be symmetrical, before the V/oct calibration. But I can't say if it tamed the two segments I'm talking about.
frijitz
worker8 wrote:
Bias switch up most of the time.

What does that mean? (I don't have one to look at.) It must be labeled, no?

Ian
parasitk
frijitz wrote:
worker8 wrote:
Bias switch up most of the time.

What does that mean? (I don't have one to look at.) It must be labeled, no?

Ian


If it's the same across all formats, up is "high".
worker8
Sorry Ian,
The high position is for a 1V bias.
Low is 0,2 V, and middle position sets no bias.
worker8
The calibration procedure has to be done with the bias switch up.
worker8
Here, you can see several calibration profiles from Volta.

A-110 VCO :

A-105 SSM2044 filter :

A-143-9 quadrature LFO :

Anti-Oscillator (this one was done manually) :

ZerOscillator :

ZO (manually, before I calibrated it) :


On this last exemple, center frequency is set to middle A (440 Hz). Somewhere around 100 Hz, there's a kind of break (that's what I call a second segment). At first, I though the calibration procedure of the ZO could solve this.
frijitz
worker8 wrote:
The high position is for a 1V bias.
Low is 0,2 V, and middle position sets no bias.

Thanks. Definitely you have to calibrate at the highest bias. It's not clear how to interpretate the calibration curves quantitatively, but the ZO certainly looks off by your comparisons.

The first thing I would say is that errors at the low end of the curve are usually not considered to be critical. The reason is that if you are off by, say, 1% at 100Hz, that is only a 1Hz error, ie, you would have one beat per second playing against a perfect oscillator. OTOH, a 1% error at 1kHz would give 10 beats per second, etc.

I could say a lot about how to design the bias control on a TZVCO, since I stewed over it for several weeks for my own circuit. But maybe you could just try something. I believe full bias on the ZO is actually 5V and that you would get better tracking at higher bias. You would do this by adding a DC voltage into the Lin FM input.

I can't give you a specific voltage to try since I don't have all the relevant numbers, but I would be interested to know whether this allows better tracking.

Ian
worker8
Thanks Ian,

I'll try this higher bias trick and let you know.
worker8
I'm a little bit late ... Anyway.

I've just spent a couple hour calibrating my ZO again. I've added a 2v bias, with the bias switch alteady in the up position. Total 3v bias.

I was able to play 5 good octaves, from c2 to c7, with a regular a-190-1. The range switch on the ZO was in the middle position.

Without this additional bias, tracking is not as good. But still 4 pretty good octaves. Something strange : the 1v/oct errors aren't the same for the up and middle settings of the range switch. With perfect a pitch on the lowest note, higher notes will be slightly too high or too low depending on the range setting.

I didn't try to calibrate with Volta yet.

I wonder if it would be possible to modify the internal bias of the module. Actually, I think it would highly improve tracking. But I can't tell if there would be some undesirable side effects.

Thanks Ian for the trick !
robotmakers
I have a pair of the MOTM paneled Zeroscillators from the first run, and they track reasonably well over a few octaves. For FM, I always sync them together so minor differences in scaling matter little.

Cheers,
Roger
Ebotronix
"All audio functions for the ZO should start with the front panel Bias and Range switches in the UP position.
The Quadrature Morphing Switch should be in the Center position.
Please turn all knobs (except the big Tuning Knob) down full Counter Clockwise.

SINE
Sine Wave Output from the SIne Wave Jack has two Trimmers on the ZO Engine Circuit Board.
These Trimmers are for adjusting the sine wave purity. They are marked VR8 and VR9, they adjust SIne Distortion and Sine Symmetry.
We calibrate by listening at about 400 to 900 Hz and then adjust by listening with our ears.
Of course we also view them on a scope while doing this, but by closing your eyes and listening instead, you usually get a the purest sounding result.
The ZO easily makes perfect looking sine waves too. You might try this at a lower frequency than we usually calibrate it at?

QUADRATURE
The ZO has a unique circuit in each of the two Morph Sections of the quadrature outputs,
as you know the perceived loudness is greater with waveforms full of harmonics than a pure sine.
Unique to the ZO is a compensation circuit that diminishes the quadrature outputs somewhat during
the morphing function as it passes from triangle to square to sine. It keeps them sounding the same volume.
For this reason the exact point in the knobs rotation is critical to making comparative measurements.

There are also so many ways in which to measure the quadrature signals.
Here an oscilloscope is very handy, as you can view the output of a "Left side of the front panel" jack (either 0 or 180)...
superimposed on one channel of the scope, Vs. a "Right side of the front panel" jack (either 90 or 270) on the other channel of the scope.
(with the same settings on both channels, you should be able to match-up the hight, or amplitudes of the waveforms by turning the Morph Knobs).

The knob position on the Left side Morph, may not precisely correspond with the exact knob position on the Right side,
so a little hunting for that spot may be required.

I ask her for the calibration
I'm working on a new calibration procedures document for the new surface mount Mark-II ZO like yours,

"Thank you for observing this difference between the two sides of the Quadrature signal levels,
we are doing another run of Zeroscillator engine Boards later this week, and will go over this with Mark Barton
to see if there are any resistor value changes that we can do improve this.

It's not surprising, as the right side has to go through many clever circuitry just to exist at all,
(perhaps a little too clever, lol!)

Thank you for bringing this to our attention, we'll write back to you whatever we find."
Ebotronix
I ask cynthia for the calibration....
"All audio functions for the ZO should start with the front panel Bias and Range switches in the UP position.
The Quadrature Morphing Switch should be in the Center position.
Please turn all knobs (except the big Tuning Knob) down full Counter Clockwise.

SINE
Sine Wave Output from the SIne Wave Jack has two Trimmers on the ZO Engine Circuit Board.
These Trimmers are for adjusting the sine wave purity. They are marked VR8 and VR9, they adjust SIne Distortion and Sine Symmetry.
We calibrate by listening at about 400 to 900 Hz and then adjust by listening with our ears.
Of course we also view them on a scope while doing this, but by closing your eyes and listening instead, you usually get a the purest sounding result.
The ZO easily makes perfect looking sine waves too. You might try this at a lower frequency than we usually calibrate it at?

QUADRATURE
The ZO has a unique circuit in each of the two Morph Sections of the quadrature outputs,
as you know the perceived loudness is greater with waveforms full of harmonics than a pure sine.
Unique to the ZO is a compensation circuit that diminishes the quadrature outputs somewhat during
the morphing function as it passes from triangle to square to sine. It keeps them sounding the same volume.
For this reason the exact point in the knobs rotation is critical to making comparative measurements.

There are also so many ways in which to measure the quadrature signals.
Here an oscilloscope is very handy, as you can view the output of a "Left side of the front panel" jack (either 0 or 180)...
superimposed on one channel of the scope, Vs. a "Right side of the front panel" jack (either 90 or 270) on the other channel of the scope.
(with the same settings on both channels, you should be able to match-up the hight, or amplitudes of the waveforms by turning the Morph Knobs).

The knob position on the Left side Morph, may not precisely correspond with the exact knob position on the Right side,
so a little hunting for that spot may be required.

I'm working on a new calibration procedures document for the new surface mount Mark-II ZO like yours,

"Thank you for observing this difference between the two sides of the Quadrature signal levels,
we are doing another run of Zeroscillator engine Boards later this week, and will go over this with Mark Barton
to see if there are any resistor value changes that we can do improve this.

It's not surprising, as the right side has to go through many clever circuitry just to exist at all,
(perhaps a little too clever, lol!)

Thank you for bringing this to our attention, we'll write back to you whatever we find."
sylvainmoreau
hi,

really never used my ZOdotcom since i bought it... 7 years ago ! ... yeah very strange.
i never got something out of it and when i did i could not recall it next day... had lot of work, started eurorack ... almost sold it last year.

anyway i finally have time to get into it and fak... i love it. i get amazing sounds i was looking for with all my other osc.
so i want to go deeper but can't find any manual online.
like what the hell is time reversal ? vari synch ? how why what...
any idea where i can find that ? strange there's no exhaustive video about it to.
MUFF WIGGLER Forum Index -> Modular Synth General Discussion  
Page 1 of 1
Powered by phpBB © phpBB Group