[time-nuts] GPSDO disciplining algorithms

Hal Murray hmurray at megapathdsl.net
Mon Apr 13 21:41:02 EDT 2015

csteinmetz at yandex.com said:
> There are two regimes a GPSDO must deal with -- normal operation, and
> holdover. 

Is there a 3rd regime: Recovery from holdover?

How many GPSDOs even spec what they do in that area?  They may do it 
indirectly by reference to Telco spec xxxx.  What do the telco/cell-phone 
specs say?

I think there are two cases: good frequency and good time.

Suppose while it was in holdover, it drifted 1/4 cycle.  Do you jump back to 
the right phase/time as fast as possible or stay close to the target 
frequency and creep back slowly?  What is the expected frequency offset 
during recovery?

Suppose it drifted 3 1/2 cycles.  Does the PPS skip those 3 cycles so the 
crystal only has to slew 1/2 cycle, or does the system slew the whole 
distance so that a downstream counter stays in sync?

These are my opinions.  I hate spam.

More information about the time-nuts mailing list