I wanted to ask if any of you have noticed similar behavior. My PolyBrute 12 is actually super stable regarding tuning. I calibrated it a few times after purchase. Since then, the oscillators and the Ladder filter are super stable. But the Steiner-Parker filter is worrying me. Even after warming up (approx. 30 minutes), I still have extreme fluctuations between the individual voice cards (fluctuations of almost 90 cents). One voice card stands out particularly.
After a calibration, the tuning is then within the expected range (approx. 5 cents deviation). You canāt expect more from a Steiner-Parker filter. But that doesnāt last long. Fifteen minutes later, there are already larger deviations again. This means that the Steiner-Parker filter would actually need to be calibrated continuously on my PolyBrute.
Is this expected behavior with the PolyBrute? I would have thought that at least after warming up, the Steiner-Parker filter should also be reasonably stable. Or do I possibly have a defect, since one voice card in particular always drifts very strongly?
While Steiner Parker is the wild child of the filters, it tracks nigh on perfectly here (the āwildā character only being in the sound now), or as perfect as you can expect from a filter.
Just to test this Iāve had the PB12 powered up for a couple of hours now. Iāve set the SP and Ladder to full resonance and turned off the oscillators. Both filters tracks well across the whole keyboard range, playing in tune with each other with only some slight variation in beating. The voices are pretty much bang on similar. Iām not sure how much the variation is in cents, but itās no more than 4-5 or so I reckon. Itās good enough to play melodies with both filters oscillating without any āsea sickā feeling.
I last calibrated it around the time of its launch, and there has been no need for further calibration. Iām impressed with how well the oscillators and filers stay in perfect tune.
So, if your SP filters are 90 cents out between voices, and you canāt get them to track well by calibrating (have you tried the debug mode calibration?), then I think you should contact support.
Something strange is happening here: Yesterday evening, after several hours of operation, I tuned the PolyBrute again. The Steiner filter was in tune.
Today, after about 30 minutes of warming up, there were strong fluctuations, with voice card 12 being 80 cents above the others.
I then actually recalibrated once in debug mode. After that, all 12 voices were a maximum of 5 cents apart. After waiting 30 minutes, the behavior was similar. Some voices a bit higher, others a bit lower - but overall within a range of 10 cents difference. After 60 minutes, the same picture. I thought it was looking better now.
Now, after three hours of operation, Iāve played through all the voices again. Most of them are still quite stable with a maximum variance of 10 cents from each other. But: Voice card 2 (!) is now suddenly at G#3 when playing a C4?!? Over an octave lower. And above all, a voice card that hasnāt really stood out so far. If it had been voice 12 again, it would at least have pointed suspicion towards one voice card. But like thisā¦
Okayā¦ I really donāt understand this anymore: I opened my Excel sheet to write down the deviations of the individual voices for support. It feels like not even five minutes have passed since the last test. I havenāt changed anything else and just played through all 12 voices again: Voice 2 is back in the normal range. All 12 voices are now back within a difference of 10 centsā¦
Temperature-wise, there can only have been minimal differences, possibly due to playing the test sequence. Otherwise, the PolyBrute has been on for hours, and the room temperature is absolutely stable. Can a minimal temperature fluctuation cause a voice to suddenly ātipā by over an octave?
Hmā¦ No, IMO, itās very stable, small temperature changes shouldnāt cause anything noteworthy. Iāll try to refrain from speculating too much, but could it be something with the CV demux perhaps, or even just a bad connection somewhere. In any case, it should not be like that. Iād run this past the Arturians at this point - Iām sure theyāll help you get this sorted.
I have asked ChatGPT on this (why not have a second oppinion ). Thatās the conclusion that ChatGPT has drawn from my input:
" From my point of view, the brief drop of a single Steiner voice by more than an octave is unlikely to be a classic tuning fault or hardware defect.
When the Steiner filter is driven into full self-oscillation, it effectively becomes a non-linear analog oscillator rather than a simple filter. In such systems, especially when operated very close to the stability limit (maximum resonance, exponential key tracking), the oscillation can momentarily lock into a different stable mode. This can result in period halving or subharmonic oscillation, where the circuit oscillates at a fraction of the intended frequency (e.g. 1/2), which appears on a tuner as an octave-down pitch.
Crucially, this is not the same as detuning:
the pitch jump is discrete (ā octave), not gradual
the voice can return to normal without recalibration
calibration remains accurate and repeatable afterwards
A true defect would typically cause a persistent offset, failed calibration, or permanently unstable behavior. In contrast, a temporary mode change that self-resolves is a known phenomenon in analog self-oscillating feedback structures, particularly in designs like the Steiner-Parker topology.
So while the behavior is unusual and understandably concerning, the fact that it is reversible, voice-specific, and only occurs in self-oscillation at extreme settings strongly suggests a non-linear oscillation mode change, not failing hardware."
The Steiner filter seems to be in tune again, without any further calibration. I will contact the Arturia Support anyhow, just for completeness and because I would like to know the expertise of their engineers on this behaviour.
Yes, that can happen, but Iād think the SP in the PB is not near regions where that occurs. I donāt think this should happen under normal operating conditions, keyword being ānormalāā¦ Iāve never heard anything like that on the PB12 or my other Steiner Parker Filters. Iāve tried to provoke a āharmonic flipā by careful cutoff and resonance tweaks, without success. If I have an oscillator turned up and routed to the SP, then the self-resonating SP filter will lock onto the oscillator harmonics, as one would expect though, resulting in discrete steps above or below the natural resonant frequency (more or less - thereās some warble in between).
This is interesting, if I may say so despite your situation not being desirable. Please let us know how this works out.
A short update from my side: I have contacted the Arturia support team regarding the issue. They also found the behavior strange and wanted to replicate it. However, I have the āproblemā that I havenāt been able to reproduce it again yet. After a warm-up phase, the Steiner filter is now stable on my PolyBrute and there is no voice that is out of tune. I will continue to monitor this for a while longer.
_Need help ?
When you can't find the answer online or in the product manual, the Technical Support Team is here!.
_Stay tuned
Follow us for the hottest sounds, fresh content, exclusive offers and Arturia news as it happens.
). Thatās the conclusion that ChatGPT has drawn from my input:
