I’m testing the PWM outputs from the R6000 and have a strange issue.
If I use the pwm_set function to alternately set duty cycles of 0 and 1, then every so often (randomly), some of the output pulses are too large. This is observed on a logic analyzer and the effect is seen on the attached hardware too.
If I use duty cycles of 1 and 2, then the pulse widths are as expected.
I’m using the suppress pulse option OUTEIGHTH.
I suspect it is due to the different way the function handles a duty cycle of 0, as the actual register values reflect duty cycles from 1/1024 to 1024/1024; 0 isn’t a direct option.
Anyone seen this before?