VCXO calibration

[Mychaela Falconia]

Hello FC community,

I got to perform a manual VCXO calibration procedure (using the CMU200)
on one of my FCDEV3B boards based on my reconstructed understanding,
and it seems to be working: the numbers I got out of my manual cal are
in the same range as those read out of Openmoko-made GTA02 units
(which is expected, as we are using exactly the same VCXO circuit
including the crystal part), and after doing this VCXO calibration I
was able to connect to the local GSM network and make some calls.
(The latter success should not be considered a definite proof of
perfect calibration, though - it once succeeded for me last weekend
before this cal when the temperature in the room was colder.)

I just wrote a document describing what I did, so when Serg gets his
board tomorrow, he will have at least a chance of being able to
reproduce it.  The write-up is in doc/VCXO_calibration in the
freecalypso-tools Hg repository - enjoy!

My next step is going to be trying to operate the CMU200 programmatically
through its SCPI-based command interface, so I can start automating
these calibration procedures - we need automation for repeatability,
and we are also going to need Rx and Tx calibration (beyond the VCXO)
which would be too impractical to perform manually - too many
repetitive steps.

M~