Why does phase margin increases as temperature increases?

[roki]

Hi all,

As seen from the title above, why does phase margin increases as temperature increases?What is the relation of both?

Thanks!!

 

[LvW]

Hi all,
As seen from the title above, why does phase margin increases as temperature increases?What is the relation of both?
Thanks!!

No, there is no direct relation between both.
Of course, when the feedback loop contains temperature-dependent parts/amplifiers (and in principle, no part is really independent on temperature) all the properties of the circuit (including the phase margin PM) may change slightly. And the PM could also change its value in the other direction (decrease). However, normally this is no problem and the associated variations can be accepted.

Added later: In addition, temperature changes and many other unknown/unwanted environmental influences (gain changes, phase distortions, transit time uncertainties,...) are one of several reasons for implementing something like a MARGIN into the feedback concept. The other reasons are, of course, gain peaking and step response overshoot.

 

[FvM]

I would expect amplifier gain changes as the most likely mechanism. Although MOS temperature dependency is rather complex, you'll ususally find a negative gm temperature coefficient, BJT have it anyway (by VT temperature dependency). If you are interested in details, you'll want to analyze which parameters are actually determining phase margin in the circuit of interest.