Hi

The ~ +/- 40 ns correction takes you down to a residual error in the ~+/-10 to 
+/-20 ns range. Using it is an improvement, but it's not day and night. 

Bob


On Feb 18, 2010, at 12:02 AM, Don Mimlitch wrote:

> I have been working with a HP58534A GPS Timing Antenna which outputs extended 
> NMEA sentences including a custom sentence which specifies the Timing Error 
> of the NEXT 1PPS Pulse as a number between +00 and +85 Nanoseconds. I am 
> wondering whether using a PIC to correct each 1PPS pulse using this sentence 
> would be worthwhile to clean up the saw tooth in the 1PPS and if so what 
> would be the best way to correct each 1PPS.
> 
> One solution might be to send the receiver the custom NMEA sentence which 
> lets me set the antenna delay +/-nnnnnn Nanoseconds. I would use the timing 
> error plus the actual cable delay to calculate the nanosecond delay to use 
> for the next pulse and send the cable delay sentence to the receiver before 
> the next 1PPS pulse. I have at least 100ms to send this message at 9600 baud 
> before the specified pulse arrives. This should be sufficient time assuming 
> the receiver acts on the message immediately.
> 
> A second approach would be to use a programmable delay line (about $14 from 
> jameco) to correct the the 1PPS by 85ns - Timing Error from message. I would 
> set the cable delay to 85ns more then the actual cable delay to bring the 
> 85ns in the delay line calculation.
> 
> I'm sure all the timenuts will have better or simpler solutions to this 
> concept and many different views on whether it is worthwhile to do.
> 
> 
> 
> 
> _______________________________________________
> time-nuts mailing list -- [email protected]
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
> 


_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Reply via email to