[time-nuts] Thunderbolt - temperature

[Hal Murray]

> A strange anomaly shows up occasionally in the temperature plot.  You
> occasionally see a 100 millidegree instantaneous positive spike in the
> data.  The temperature then decays over 30 seconds back to the
> original curve.  This occurs on all of my units.  They are all
> different revisions with different power supplies.  The spikes seem
> uncorrelated with any external events. 

Here is a guess.  Maybe the CPU is doing a bunch of work, like shifting to a 
different set of satellites.

Can you monitor the power it's using?  Or check the satellites before/after a 
step?


-- 
These are my opinions, not necessarily my employer's.  I hate spam.