[ivanbakel]

>A quick googling did not turn up more material on the idea that voltage regulators depend on the temperature like that, and it would be surprising (generally electronics performs better when cooled).

Not a hardware expert either, but Wikipedia points at this TI doc[0] which claims ambient temperature is necessary for the quiescent current. There's no mechanism described there, though.

[0]: https://www.ti.com/lit/an/slva079/slva079.pdf

[formerly_proven]

These have ground pins (as opposed to floating regulators), so they draw whatever bias current they need (which mostly depends on temperature, input voltage, output current and the lottery). The explanation is bogus, and doesn't explain how heating once would help with a problem caused by too low ambient temperature during operation anyway.

[jonsen]

> ambient temperature is necessary for the quiescent current

No it does not say that. It says ambient temperature is a factor contributing to quiescent current:

"The value of quiescent current is mostly determined by the series pass element, topologies, ambient temperature, etc."

In an LDO you normally want as little as possible quiescent current when idle. You certainly wouldn't design stand by operation to be dependent on temperature. If it turns out to be so with time, it's an aging problem.

[thomasthorpe]

Old thread now. But often a minimum load is needed for an LDO, particularly on older parts, to work correctly. Such that these are given in datasheets. This is continuously required so is effectively quiescent current. I think this is what the OP was trying to get at.

But failed capacitor reulting in instability seems much more likely.

[jonsen]

You are right that a small quiescent current is necessary for correct idle operation. But much of the discussion here is about the misconception that a certain ambient temperature is necessary for correct operation by design.

By the way it is in fact imaginable that aging has caused the quiescent current to become too small at low temperature.

There are several possible temperature dependent fault mechanisms caused by aging, so I wouldn't make a guess in this case.