Rasp Pi at +/- 1 us from GPS

[Hal Murray]

Nice.  Thanks.

You don't need the RTC.  You can get the initial time over the net.


> - (if I understand properly) pps-client takes a pair of GPIO pins and
> calibrates latency, so its PPS adjustment is more accurate. 

Any numbers?

I've thought of doing something like
  sleep until a bit after a second ticks
  grab time
  turn on GPIO
  grab time
And catching things on a scope that is also watching a PPS.


> I hope to be able to install NTPsec:
> - as a time server
> - leaving system time alone 

I think it should do what you want if you set it up with only a localclock.  
(no normal servers)


> I don't know how to measure if pps-client is doing what it says it is...

On Linux, can generate graphs and such by watching /sys/class/pps/pps0/assert 
and/or /sys/class/pps/pps0/clear
(on non Linux, you will have to work a bit harder, but you can get the same 
data using the PPS API)

Here is a python hack to monitor 60 Hz from the line.  You can tweak it to 
grab every sample.
  http://users.megapathdsl.net/~hmurray/time-nuts/60Hz/60Hz.py

You can also monitor things using ntpd.  Of course, that will include the 
noise from the other system and the USB Ethernet.  Turn on rawstats.  Add 
"server foo noselect".  It will collect data, but not use it for timing.

That will log info from each client-server pair.  (That's when the logging 
system is the client.  You can do the same on the other end.  They should 
match.)



-- 
These are my opinions.  I hate spam.