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

R*S NTO Pitch Drift on LIN FM?
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author R*S NTO Pitch Drift on LIN FM?
wavecircle
Hi Folks,

I have an R*S NTO in 4u LW format but have noticed that when the modulation depth on the LIN FM exceeds around 2 o'clock there is a very perceptible shift in the fundamental pitch. This really annoys me as it's an awesome oscillator but it sounds bad (to my ears) when using an envelope/LFO to modulate FM index or indeed, just trying to get high modulation indexes manually.

My PCO which sits right next to it doesn't exhibit this behaviour. Can someone with an NTO, euro or 4U, test this out for me to see if it's a fundamental behaviour with this PCB?

Just put a periodic signal (sine) into the LIN FM input and raise the index slowly, listening for pitch drift. I would be hugely appreciative if I can stop this drift or to know that everyone experiences the same thing.
wavecircle
Resolved, I have heard the same thing on a Euro version.
J3RK
I'm not sure how the NTO's FM input is coupled. (I have a schematic somewhere, but haven't looked in quite a long time.) If it's DC coupled, it's possible that an offset is creeping in somehow. If it's AC coupled, if the cap is a large enough value this could still occur a bit, though not as much, and it would probably eventually settle. Just as a test, you could try running the FM signal through an AC coupled module like an audio mixer or AC coupled VCA, and see if it does the same thing.

One other thing, and I'm not sure what the exact effect would be. If the FM input is AC coupled, and you're FMing with a PWM wave, the pitch might shift as the signal shifts at various pulse width settings. That's something that would need to be tested though.

Just guesswork here since I don't remember how it's set up internally.
thetaflux
Hey wavcircle. You may find this passage from the R*S NTO for Eurorack build guide of use to you.

"Linear FM: Unlike the PCO, the NTO offers voltage control over the LInear FM (this is a unique NTO feature).
The Linear FM section comprises the 2 jacks above the LIN FM knob. Input is the top (black) jack underneath
the 1V/Oct in the center, VC input is the jack below. The FM trimmer on the pcb eliminates bleeding of a CV signal
for the LIN FM section. To calibrate, feed a signal at audio range into the white VC input (middle jack) for
LIN FM without any audio (FM input) signal going into the black input above and listen to the SAW output (pitch
of the VCO). Adjust the trimmer so that the effect of the CV signal is as little as possible."

the 4u version should share this trimmer.
wavecircle
Nice fellas,

I'll get the thing open soon and see if the trimmers do anything.
wavecircle
Hey Thetaflux,

I tried your patch and it seems that I do get a fair amount of modulation when doing it, could you tell me what you experience when you make that patch?
J3RK
thetaflux wrote:
Hey wavcircle. You may find this passage from the R*S NTO for Eurorack build guide of use to you.

"Linear FM: Unlike the PCO, the NTO offers voltage control over the LInear FM (this is a unique NTO feature).
The Linear FM section comprises the 2 jacks above the LIN FM knob. Input is the top (black) jack underneath
the 1V/Oct in the center, VC input is the jack below. The FM trimmer on the pcb eliminates bleeding of a CV signal
for the LIN FM section. To calibrate, feed a signal at audio range into the white VC input (middle jack) for
LIN FM without any audio (FM input) signal going into the black input above and listen to the SAW output (pitch
of the VCO). Adjust the trimmer so that the effect of the CV signal is as little as possible."

the 4u version should share this trimmer.


Good information!
kashmir
I had eurorack version, I did the calibration and it behaved exactly like that. I wrote Ralf from R*S and he assured me that it is normal behavior.
wavecircle
Ralf also told me it's normal but I was wondering if there is a calibration which minimizes it.

Ideally I'd like to reduce the range of modulation if I can not prevent this pitch drift at the higher end of the modulation index.
Navs
Is it CV bleed as described or is it the VCA itself that is emitting DC, i.e. is it OK up to a point and then become unusable?

I heard this on a Euro NTO too but had limited time with it and couldn't really come to a conclusion. There's no way of bypassing the lin-FM VCA, is there, like a direct lin-FM input on the VCO?
wavecircle
Hey Navs,

I think it's the VCA within the NTO itself. The pitch drifts up once the modulation pot exceeds around 1-2 o'clock, no CV applied at all, just manual control. The same thing happens under CV control, the top end of the modulation range exhibits the same pitch drift.

There is a LIN FM audio input and I could patch a VCA before but that is defeating the purpose of having the depth control on the module. I have a small system and only a couple of VCA's so getting the full functionality of the module is important.

It's a pain because it sounds great until around 2 o'clock on the manual LIN FM control. Maybe I can just double the value of a resistor somewhere to restrict the modulation range?

I have a video of the behaviour, I'll upload it to my instagram. Link in my sig.
J3RK
That manual control is likely summed in with the CV input. If there's even a tiny amount of CV feeding back through (what that trim is likely for) then as you turn that knob it will become a larger and larger offset. (which will result in the shift you're seeing) If the trim doesn't remove that, then there may be a secondary source of offset (depending on how the VCA is coupled).

Increasing the summing resistor will decrease the modulation range, but it won't actually remove the source of the problem. Did you try the trim? Just curious what effect it had if any.
wavecircle
Thanks for the feedback Dustin, I'm away from the serge for a while. Will try some calibration routines when I get back. Just haven't had time these past few days.
J3RK
wavecircle wrote:
Thanks for the feedback Dustin, I'm away from the serge for a while. Will try some calibration routines when I get back. Just haven't had time these past few days.


No problem! Just wasn't sure if I missed something where you tried it. Hopefully that will do the trick.
Navs
The video is helpful - it doesn't sound as bad as I thought and it raises two questions for me:

- does this pitch still rise if you open the VCA without a modulation signal? If so, then it may indeed be something to do with the VCA

- if it doesn't, then it might just be the amount of modulation pushing the linear FM too far, i.e. to a point where it is no longer linear.

My Plan B Model 15s do this and iirc, the AFG did too. Ian Fritz posted something about this years ago. I'll post the link if I can find it. I can also push my lin-FM-modded A-143-9 VCO to a point where it goes 'sour'.
wavecircle
That's a great help Navs, I think the issue is overmodulation and the through-zero problem... this has actually forced me too look into the maths and realise the value of through zero FM.

Maybe the best solution is a simple attenuation of the cv modulation. The through zero issue could happen at any depth if the frequency difference was high enough. I still wonder if I can optimise the range with calibration or resistor changes.
thetaflux
So I did some tests with the euro version and confirmed the same behavior. unfortunately on my two NTOs it was impossible to remove the behavior entirely. there are two sweetspots on the trimmer, that allow the minimization of the behavior either when the linear FM attenuator is fully open, or when the linear FM attenuator is fully closed. You can't(or at least I couldn't) find a sweetspot that did both. So you have to pick between having your dynamic linear FM signal subtly affecting the pitch either when the attenuator is closed or when it's fully open. none of the settings of the trimmer allowed for elimination of the effect over the full range of the linear FM manual control.


personally, i find this pretty annoying. it makes the internal VCA on the NTO effectively useless unless you want to FM the NTO with 2 signals at once :(

EDIT: i also found that turning the manual FM knob also affects the pitch. does that mean there's an issue with the VCA?
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group