Re: Synchronization: PPS and Tx signal in a USRP N200

[Marcus D. Leech]

On 2021-12-13 16:56, isaac mario tupac davila wrote:
> Hello
>
> Thanks for the information !!!
>
> I have a question regarding the parameter unknown pps. I'm focusing 
> wrong, I think so ...
>
> 1. If I read the concept This method is used to sync when the edge of 
> PPS is UNKNOWN. But when I use a N200, a trimble(GPS) and GRC 
> (uhd:usrp source) with this configuration (unknown pps), my N200 gets 
> synchronized using a trimble (known pps)... So , when should I use 
> unknown PPS?
>
> 2. Also says It waits for the last pps and later set time next pps. In 
> another website there is a method 1 to poll the USRP time registers.. 
> Is unknown pps using the same method?...  I think the Method 1 is used 
> for known PPS (trimble and USRP)
>
> Regards
> Isaac T.
"Uknown PPS" has NOTHING to do with whether UHD "knows" what type of 
GPSDO unit you have.

It has to do with the synchronization sequence.

"unknown_pps" waits for the time register to change (get_time_now()), 
and schedules "set_time_next_pps()" for whatever that returned + 1 
second.  The provides good guarantees
  that the timestamp register will be set correctly.

But I should AGAIN emphasize that the 1PPS input DOES NOT provide any 
kind of on-going synchronization of the clocks inside the N200--it's 
simply a "trigger point" to assure
  that the time that is set is aligned with the 1PPS pulse.  It has 
nothing to do with ongoing synchronization of samples etc.

But once your program has high confidence that it knows what the time is 
on the N200, it can then schedule events on the device knowing what time 
the device thinks it is,
  and in the receive case, the timestamps on the incoming samples.

https://files.ettus.com/manual_archive/release_003_001_001/manual/html/sync.html

https://kb.ettus.com/Synchronizing_USRP_Events_Using_Timed_Commands_in_UHD