when programming new patches the last days I discovered a strange behaviour of the oscillator waveshape pots. I’m starting with an “Init Preset” (disabling all sources in the Mixer exept for oscillator 1 for simplicity) and play a high pitched note. Moving the waveshape pot of oscillator 1 slowly results in a more or less (some gentle digital crackling when going faster) continuous transision of the tone. When I stop at any position, wait for a moment (~2 seconds or more) and then continue rotating slowly, I hear a popping sound. I investigated the output signal with an oscilloscope and noticed a small jump in the waveshape correlated to the popping sounds. Passing the 11 o clock position is almost always followed by a popping noise. Both oscillators are affected and I wonder, if this is just an issue with my device or if anyone else can reproduce this behaviour.
Furthermore I noticed, that the status LED of MOD2 channel flickers at low intensity. The effect ist clearly visible (in comparison with the status led of MOD1) when using the following settings on both modulating channels: Tringangle waveshape, slow rate (~0.7). While the brightness transition of led 1 is smooth, led 2 begins to chop when beeing dimmed.
Greetings
Christoph
Moog Sub37 Tribute Edition
Firmware 1.1.28
Editor Version 0.9.54
Windows 10 (64bit)
Cubase 8 Professional
Thanks for the report! Very good close observation on your part. Both of these things are normal and to be expected.
In the case of the popping sound, there is a discontinuity in the wave shape at the moment when the square/pulse waveform is first crossfaded into the sawtooth wave, which happens at around 11 o’clock on the dial. There are no square “corners” in a sawtooth wave, so at the moment when this square-cornered shape is first introduced into the waveform it makes a little pop.
In the case of the MOD2 LED, what you are seeing is the fact that the MOD2 LED is dimmed using pulse-width modulation (turning an LED on and off very fast to create the appearance of dimming), where the MOD1 LED is using an analog control voltage to set the brightness proportional to the modulation value, which looks much smoother. We would have used a control voltage for the MOD 2 LED as well, but we ran out of them – there are only so many control voltage channels available and we use all the rest of them for the sound engine.
Nearly off topic, but just being curious: ever considered to make the wave generator going from saw to triangle, to trapezoid ,to square and finally to pulse? Transition would be more smooth
Just after installing v1.1.20 (no power cycle) I noticed some strange behaviour of the OSC1 waveshape-pot.
After assigning mod1 to OS1WaveShape, fiddling around a bit and then, turn the mod amount to zero: result the SAW was already at full volume with the WAVESHAPE pot fully counter clockwise. So there was only a fade between SAW and Pulse, TRIANGLE was gone, or masked by the SAW.
I did an INIT (preset).
After some knob turning, I ended up with no SAW at al. There was only a fade from TRIANGLE to Pulse.
I power cycled the sub37 after that, and until now I have not been able to recreate this situation.
Amos, thanks for sharing these technical details! Last weekend I did some further investigations and have a few more questions.
Regarding the popping sounds after interrupting (for ~2sec) a slow rotation of a waveshape potentiometer:
This phenomena seems to be independent of the potentiometer posision. To me it looks like the following: In Idle-state (no control beeing touched) the processor scans all hardware controls for a change in a certain time interval. When a change is beeing monitored the processor sets its ‘focus’ on this particular parameter and follows the hardware control in ‘real time’. Maybe the initial change is beeing missed (due to a finite monitor time interval) which leads to a small but audible parameter jump. But I could be completely wrong!!
Regarding the PWM:
Updating the PWM dutycycle more often (-> smaller steps) would certainly result in smoother brightness transisions - but maybe processor power is here a limiting factor.
Playing around with the led brightness in the editor i noticed a hum noice at about 80Hz (plus harmonics at 160Hz and 240Hz) - most dominant at brightness level ~50%. The hum is subtle but clearly audible with softer mid to high pitched sounds (for example the theremin preset). I guess the leds are also pulse width modulated with 80Hz and there is some leakage of the supply leads into the audio path. Might be the same issue, other users complained about in the past that already has been improved with an earlier firmware update (can’t comment on that as my Sub37 is relatively new). I fear that this is an hardware issue and can’t be further improved via firmware updates
Maybe I’m a little bit picky and should stop looking after the ‘fly in the ointment’. I just want to make sure that my unit is working properly since I had a problem with a bad ribbon cable junction on the controller PCB, causing headphone dropouts…
