[USRP-users] gps_locked sensor throwing exception in latest master

Josh Blum josh at ettus.com
Thu Feb 7 00:06:31 EST 2013



On 02/06/2013 04:37 PM, Perelman Nathan (Nathan) wrote:
> In experimenting with the latest master and the addition of the
> gps_servo sensor, I noticed some anomalies in the gps_locked sensor. In
> my application I am asking for it every second or so to provide a status
> output. Every so often it will print a warning three times and then
> throw an exception similar to:
> 
> UHD Warning:
> 
>     locked: ValueError: Invalid response "$GPGGA,2113-02-06 5104 108872
> 189.29 5"70E-12 15 15 6 0x0
> 
> terminate called after throwing an instance of 'uhd::value_error'
> 
>   what():  ValueError: Timeout after no valid message found
> 
>  
> 
> This will usually happen within 10 minutes of the gps locking. In
> addition, the gps_locked sensor appears to sometimes return false even
> when the gps is locked (from looking at the nmea strings I can see that
> it is actually locked). I can usually observe this at least once within
> a minute of the gps locking the first time.
> 
>  
> 
> All testing was done with a USRP N210 r3 with a Jackson Labs GPSDO
> installed.
> 

Are you also actually querying the gps_servo sensor? I just want to
confirm if enabling that was the cause of the error.

The gps_servo sensor is off until it queried (I believe). When its on it
tends to add a lot of extra serial traffic. Which isnt actually a lot
for the baud rate, but I think the micro inside the GPSDO might end up
overflowing a uart buffer, and causing the parser on the host to slip
up. Something like that...

-josh

> -Nathan 
> 
> 
> 
> 
> _______________________________________________
> USRP-users mailing list
> USRP-users at lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
> 




More information about the USRP-users mailing list