[time-nuts] PPS phase between two GPS units.
attila at kinali.ch
Tue Dec 22 07:19:10 EST 2015
On Tue, 22 Dec 2015 11:57:55 +0100
Mike Cook <michael.cook at sfr.fr> wrote:
> > A sampling difference of 1 cycle with 48Mhz clocks would give you ~21ns
> offset if not using sawtooth correction. Can you do the sawtooth correction
> and see if the offset is reduced?
This is the most likely explanation. The LEA-6 modules use a 24MHz oscillator
which is doubled for the processor, leading to the 48MHz clock.
> Good thinking Brian. But I would expect the difference to wander from 21s
> UNLESS the the two receivers clocks are lock stepping. Injection locking
> might do this if the two module are physically close enough.
Injection locking wont work by simply playing the modules close to eachother.
The oscillators are shielded well enough that quite some injection energy would
be needed. Unless the TIMEPULSE2 output of one module is used to generate a
8MHz signal which is directly injected into the power suplly of the other, i
don't see how injection locking would work. And even then, I wouldn't be sure
if injection locking could be acheived.
What could be done, though, is to lift the shield over the cpu side of the
LEA-6, unsolder the oscillator and use an external 24MHz oscillator to
supply both LEA-6 modules.
If you do this, you could even go as far as to use the sawtooth correction
message to stear the reference oscillator. That way you wouldn't need to
measure the PPS output and could to a "fully" digital control system instead.
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson
More information about the time-nuts