Re: Armadillo May 9, 2004

At 10:18 PM 5/9/2004 -0400, Henry Spencer wrote:

>On Sun, 9 May 2004, John Carmack wrote:
> > http://www.armadilloaerospace.com/n.x/Armadillo/Home/News?news_id=257
>
>Hmm...  A possible alternative to the "cracked trace" theory is that it's
>got Surface Acoustic Wave filters in it somewhere; they are sonic in
>nature and have at least the potential to be messed up by vibration.
>(I would not have thought it likely, but I don't know the technology
>quite well enough to rule it out.)

Both SAW filters and crystal oscillators are piezoelectric devices
and thus affected by mechanical stress.  My guess is that the main
oscillator crystal is the culprit.  It could be pulled off frequency
by gross changes in G-loading, which would force the signal tracking
loops to re-acquire the satellites, causing data dropouts.  More
likely, mechanical vibration on the crystal is adding electrical
"phase noise" to its frequency output.  Simplifying a bit, this noise
effectively degrades the signal-to-noise ratio of the received
satellite signals.  It might degrade SNR enough in your system to
lose track of the signals completely.

Historically, the g-sensitivity and vibration sensitivity of crystal
oscillators in GPS systems was of great interest to the military.
There were (and may still be) export controls on certain kinds of
crystals and oscillators.

To fix your problem, you might try vibration-isolating just the
master crystal/oscillator from the GPS board, or replacing it with
a better off-board one.

Cheers!
--Stu