Champs,
I am asking this for our customer.
1) Why do you need to use high-resolution period?
Do you have waveforms to show the difference with and without high-resolution period?
For examples, the output ripple is much larger if high-resolution period is not used?
2) Will ISR1 needs to run very often?
On Figure 29. Three-Level Nested ISRs of the tech doc, tidueg2.pdf.
We are aware that ISR1 is used to update high-resolution PWM period, etc. for the first time and then be disabled for the second time. ISR1 is triggered only when ISR2 gets a new period. But we think even under stable states, the calculated period should not be the same. For example, it can be 500.1 KHz, 500.0 KHz, 500.2 KHz, 500.1 KHz in every control loop calulation. In this way, the ISR1 should be triggered almost every time to update the new period. However, On Figure 29, it seems ISR1 is only triggered twice while ISR2 (control loop) are triggered many times.
In our understanding, it's not possible for ISR2 to get the same period command all the time because of output ripples. Why are there only ISR1 twice?
Shouldn't there be many twice-triggered ISR1 after every ISR2?
Would you please explain this in more detail?
Wayne
