[time-nuts] time-nuts Digest, Vol 174, Issue 35

[Andre]

Hi,
This is also relevant to my experiments with SDR.

For my purposes I added a reverse biased blue LED to each pin on the crystal.
On my radio reversing it so the crystal was on the back revealed the two pads someone
had forgotten to put capacitors on explaining one of the problems.
This seems a good approach and as blue LED capacitance is a factor of applied voltage it
gives me a further parameter to play around with.

Note, this sent the offsets to hell but I can correct for it in software.
Next step: hacking the memory and connector so it can work on USB 3.1 just for the heck of it, 
as the chip does in fact support this mode but the pins aren't populated.

DISCLAIMER: If you blow up your SDR/laptop/etc please don't come crying to me.

Hi

On a ?chip" based setup like this, it?s a matter of pulling the TCXO and putting in some
wires. Whatever frequency the TCXO used to supply, that?s what your magic external
device needs to provide. The designers do not seem to allow you to provide a variety
of input frequencies. Given how the input gets used for all sorts of things, that?s not
as crazy as it sounds.

The option to pull the TCXO and fiddle has been there for quite a while. At some point
they will move the TCXO function inside the ?chip" package (with MEMS or some other
magic). Once they do, the ability to tear it off will become a bit more difficult?. :)

Once you do replace the TCXO, you then are very dependent on a sawtooth correction
to run your GPSDO. The PPS becomes one big long hanging bridge and thus is not useful.
Carrier phase or (like in the TBolt) code phase data could be used as an alternative to
a sawtooth correction word.