sanity check regarding generalized dc-dc&LDO combination math-wise..

[sparkydonkey]

assuming the PWM dc-dc regulator had a 2.5% accuracy and the LDO regulator had 100mV typical to 160mV maximum dropout..if going by the worst operating condition for both components, the dc-dc side outputting 3.6V would at any time have as little as 3.51V going toward the LDO, and since 3.51V minus 160mV dropout equals 3.34V the overall power circuit is pretty much reliably stable for a 3.3V-demanding regulated circuit. yes?

I know that load sags is another thing to factor in but I just wanted to see if I was somehow overlooking anything with regarding to calculating just the supply side itself alone for now
(especially with that in this case the LDO is more particularly for PWM ripples reduction than to actually regulate any large voltage difference hence the rather close matching voltage-wise?)

[David Hess]

Nothing prevents using an external error amplifier and reference to improve the accuracy of the switching pre-regulator.  Or in some cases just an external reference is required.

A more complex option is to adjust the pre-regulator's output based on current so that it follows the post-regulator's dropout voltage which varies with current.

[Peabody]

You would need to look at the LDO datasheet's definition of "dropout voltage".  I think it's typically referenced to a drop in output voltage of no more than x% from nominal at a particular current.  If so, then you would also need to consider that drop.

But I would strongly suggest you watch Dave's video on removing ripple:



It turns out that LDOs are not very good at filtering out ripple.  But better ways exist.

[MrAl]

assuming the PWM dc-dc regulator had a 2.5% accuracy and the LDO regulator had 100mV typical to 160mV maximum dropout..if going by the worst operating condition for both components, the dc-dc side outputting 3.6V would at any time have as little as 3.51V going toward the LDO, and since 3.51V minus 160mV dropout equals 3.34V the overall power circuit is pretty much reliably stable for a 3.3V-demanding regulated circuit. yes?

I know that load sags is another thing to factor in but I just wanted to see if I was somehow overlooking anything with regarding to calculating just the supply side itself alone for now
(especially with that in this case the LDO is more particularly for PWM ripples reduction than to actually regulate any large voltage difference hence the rather close matching voltage-wise?)

Hi,

You would be better off posting your part numbers so we can take a look at the actual LDO.  The dropout voltage spec can be tricky sometimes.

[sparkydonkey]

thanks for that video peabody

and mral I'll happily share that soon this week, still looking for new manufacturers so I haven't quite settled on a possible shortlist yet