fun MLCC fact: the crystalline structure of the dielectric ceramic (usually barium titanate) deforms over time, which modifies the material's dielectric constant, causing a capacitance derating effect that follows a roughly log10 curve vs. time in hours.
this effect is essentially innate and continues regardless of most operating parameters, except heat. if you heat the capacitor to the curie point (~150C will do it) the curve resets to t=0.
you can do this as many times as you like.
during the first hour after reflowing an MLCC you'll typically see a 1-2 percentage point drop in nominal capacitance. about the same again over the next day or two. over a period of a decade the capacitance can drop as much as 10%.
but apply some hot air and BAM! back to ~2% above nominal. wait a day, it's back to nominal.
the effect is significant enough that performing a calibration on a board with passive filters immediately after rework can cause it to go out of cal after just a few days!
the nominal capacitance rating that the manufacturer publishes is typically specified for the 100-10000 hour post-heat range, because your soldering during the PCB assembly process will cause a reset of this "aging" process back to t=0.
since a cap might sit in a reel on a warehouse shelf for years before being used, measuring a cap right out of the reel before soldering can lead to misleading results.
good point here: class 1 dielectric MLCCs like C0G don't exhibit this behaviour as strongly due to calcium zirconate's higher stability. they'll hold their capacitance steady for a much longer time.
it's class 2 dielectric MLCCs (X5R, X7R, Y5V) that'll get you. this is usually where your largest capacitances are, too, so the percentage shift represents a larger absolute capacitance change.
@gsuberland
very cool.
Also yet another blow to my fragile grip on reality, it's all so messy gaahhh *runs away with arms flailing*
Graham Sutherland / Polynomial
s-ol
Nate True
xlengineer1
Damien Miller
Attie Grande
Timon ðŸ›