Filtering xbee power: 8.2pF+1uF?

[Marc MERLIN]

Yes, yes, I'll probably get the "longuest use of a mobsendat for a totally
non certified use" award :)

That being said, I was looking at how my analog readings on the arduino were
being swayed by the power use of the xbee (yes, if I remove the xbee and
measure non changing resistance, I get a straight line, and if I send data
through the xbee, the readings change by a few percent up and down).
I was actually going to give a talk at LCA 2013 on my work and findings, but
didn't make the speaker cut unfortunately :-/

Anyway, back to my issue, I was reading the xbee documentation, and for a
pro module, page 14 says "use a 1uF and 8.2pF capacitor".
ftp://ftp1.digi.com/support/documentation/90000976_F.pdf

I noticed though that the mosendat only uses 100nF, which doesn't match the
spec sheet at all. Is it a oversight, or is there a reason I'm missing? (I
noticed that 100nF got sprinkled across the board, maybe regardless of the
real need of each chip they were next to).

Thanks,
Marc
--
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
.... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/

[Luke Weston]

The idea that you'd supposedly use an 8.2pF capacitor, which is such a minute amount of capacitance, and is going to be even smaller than the parasitic capacitance between the Vdd tracks and the groundplane in the system anyway, seems utterly bizzare, I've never seen any precedent for any such thing before.