I'm looking at the data sheet for the tps40075 and it states a minimum on time pulse width of 150ns. I'm running a 12v input with 0.9v output- giving a dc of .075. The frequency vs Vout chart in figure 16 show the max freq for this combination to be below 400khz. shouldn't frequencies up to 500khz be ok? 150ns/2us = .075.
is there a min frequency this part can run at? the chart seems to imply between 100-200khz.
Bob,
Thank you for your interest in TI parts. The charts are typical performance and not worst / best case. Given the strict math, yes 500kHz would be achievable. But running the duty at worst case performance limits would yield no headroom in the duty cycle. This means that transient response would be unusual / unpredictable during a load dump, and it means that the soft start ramp would be erratic. By definition, during soft start the duty needs to ramp up from some minimum (150nSec) towards regulation. If regulation is 150nSec, then the controller would have no choice but to provide somewhat erratic bursts of 150nSec pulses to try and soft start the output. The result is a noisy soft start ramp.
The 150nSec is the worst case capability of the device, but you would normally want to stay relatively far from that limit. I would suggest an Fsw that yields an ON time of no less than 2X the 150nSec. Implied is that only the first half of the soft start ramp might still be noisy.
Based on the graphs in the DS, the minimum recommended Fsw is 100kHz.
MC.