PolySix/KiwiSix PWM-issue

[mikkelb]

I just upgraded my PolySix to KiwiSix.

The synth works almost flawlessly, but I ran into a small issue with the PW/PWM DCO-knob during calibration.

While in PW mode the output dies out, when i turn the PW-PWM-knob past ≈ 7. I checked on a scope and sure enough the duty cycle of the wave hits 0, when the know is turned to around 7. I’ve seen others commenting on the issue online, but never managed to find a solution.
The PW/PWM-knob works as expected in all other modes – and also in PW-mode, if the sub-osc is on.

Can anyone offer some advice?

 

[NoisyIndividual]

Don't turn it past 7.


I believe some people like the idea that you can turn the duty cycle to zero and have the oscillator dissapear.

 

[mikkelb]

Don't turn it past 7.

That fix sounds really complicated.. Not sure if I can manage..

 

[NoisyIndividual]

If the control has full working range doing other thngs, then this surely is a software/firmware issue that Kiwi com need to sort out?

 

[zamproject]

I believe some people like the idea that you can turn the duty cycle to zero and have the oscillator dissapear.

I second that, some synth have a 0% duty cycle before hitting 100% CV, for that exact purpose...make it disappear from an envelope or LFO or whatever CV you use to modulate the PWM. Which IMHO is a good feature...

I have no idea if it's on purpose or not on this P6 mod ?!? maybe ask the designer ?

 

[zamproject]

I just have a quick read at the P6 user manual

Page 5 -PW waveform-

It is clearly explained, at 0 the duty is 50%, at 8 its 100%, so dead from 8 to 10.
So for the polysix this is an -on purpose- design, I don't see why the mod designer will change that.

If you don't like this behaviour you may rescale the PW CV...
...or just don't pass the limit as already suggested by @NoisyIndividual

 

[NoisyIndividual]

If you want to use the fact that the oscillator transitions from silent, to very thin sounding pulse width, to 50% thick pulse width in a real time performance by continuously moving the pulse width control, it would be sensible to some, perhaps, to have some dead space at the top end of the control.