Unstable sine PWM output

[mister_e]

If you don't have any Logic Analyser handy, I would use a PC soundcard and record the PWM waveform to see if it's a Scope Trigger problem first.

Where your scope get it's trigger from?!? Internal or External?

[pxidr84]

The scope is trigerred to external (CH1), rising edge with noise rejection.

[pxidr84]

News :

I've made some experiments, it's not an oscilloscope trigger problem.

I think it comes from my code. But where?

[Acetronics2]

Hi, Pxidr84

Something caught my attention ... in the reloading, you use DIV32; are you sure there's no possible interrupt while DIV32 works ???

as timer0 is a low priority interrupt ... I'd check that two or three times.

... just an idea after a very quick browse ...

Alain

[pxidr84]

Hi, Pxidr84

Something caught my attention ... in the reloading, you use DIV32; are you sure there's no possible interrupt while DIV32 works ???

as timer0 is a low priority interrupt ... I'd check that two or three times.

... just an idea after a very quick browse ...

Alain

Hi,

Timer0 interrupt is used for the calculation of the inverter paramaters (like the V/F ratio, the timer reload variable, etc.) It's a low priority interrupt because the Timer1 interrupt (used for the PWM generation) is very critical and must be
executed all of the time.

The DIV32 is used for calculate the timer reload value (the timer routine is from Henrik). However this routine is needed for output the desired sine frequency. This routine directly controls the Timer1 interrupt frequency. It is possible to avoid this DIV32 command (because it seems to take a LOT of ressources)?

Maybe it's better to use Timer0 as a high priority interrupt?

Maybe a PWM register is misconfigured?

I don't know

[pxidr84]

No ideas for this problem?

[cncmachineguy]

Not sure of a solution yet, but from watching the posted video, I am guessing the problem is some kind of timing issue. I don't think you have things configured wrong as the output looks great most of the time. Rather I think there are some times when a counter gets reset of an interrupt gets interrupted. It glitches then goes right back to perfect wave.

If you watch it and don't alter the frequency, does the glitch happen at regular intervals, or is it random in nature?