[time-nuts] Position accuracy and Thunderbolt performance

[WarrenS]

Mark

What this looks like is:
First the phase Jumps, (Makes sense if you move the antenna or it is not where it thinks it is), This causes the Dac to jump to start correcting for the Phase error, AND that causes the OSC freq to change which is what causes the Freq measurement difference.

My question is WHY does the Dac react SO fast to the Phase error, what should normally happen is that the 100 sec Time Constant, filters out  and slows down any phase jumps. There would appear to be some kind of additional Feedforward effect in the PID.
I did notice on my unit that the Control loop reacts quite different with a freq jump, and does not seem to go thru the same TC before it starts correcting.

If you could run the same plot with the unit in hold mode, that should show which effects are cause and which are reactions.

My hope was that all that Bad stuff would be filtered out by the 100 to 1000 sec TC control loop.
So I wonder why it is not.

I notice that your plot updates are paused, Any chance that the unit is jumping in and out of track mode or some such weird unnoticed thing like that?

warren.
****************
,
----- Original Message ----- 
From: "Mark Sims" <holrum at hotmail.com>
To: <time-nuts at febo.com>
Sent: Sunday, July 26, 2009 2:49 PM
Subject: [time-nuts] Position accuracy and Thunderbolt performance



I am pretty sure.  I enabled the sat count graph and those big jumps in the signals did not appear to be in sync with the satellite constellation changes.   Here is a plot with the sat count graph enabled.

----------------------------------------
Are you 100% sure it is not correlated to sat's entering and leaving the GPS solution ?


_________________________________________________________________
Windows Live™ SkyDrive™: Store, access, and share your photos. See how.
http://windowslive.com/Online/SkyDrive?ocid=TXT_TAGLM_WL_CS_SD_photos_072009