We discovered yesterday that one of our buck converters is running in an unstable mode. What we are seeing is pulses alternating between wide and short, every other cycle. I've read about this before, and recall that it's a loop stability issue, it just happens to synchronise with the operating frequency. The data sheet refers to sub-harmonic oscillations, which I believe to be the same thing.
I ended up manually crunching the equations in the data sheet. The numbers that I found for the compensation circuit were substantially different from what's in our circuit now. Most of the rest of the circuit is fine and doesn't need to be changed. It's just the stability issue that was bothering me.
I swapped in some new values to the compensation network. The pulse width has become stable, which is good, but the oscillation frequency is now a lot lower. For the life of me, it doesn't make any sense that the two could be connected in any fashion. So, now, I have to try to explain why the frequency is different and does not follow the relationship outlined in the data sheet. Should I change it, or is something else now broken? Is it running skip mode? Roughly, it's loaded to about 1.6 amperes.
Please see the attached images.
You'll see the before persistence image with two obvious pulse widths over top of one another, and the associated wide span showing about 375 kHz. Then, after changing from 487 ohms and 100 nF to 4.99k and 10 nF, I get different results. Operating frequency is now more like 190 kHz, but the pulse width is consistent. Subsequently, it's gone back to full frequency with no change, but there is some jitter going on.
What could be causing this?
