[time-nuts] WNRO fixup

[Hal Murray]

Bob kb8tq said:
> Simple answer: Run an external piece of software to correct the issue. It���s
> by far the simplest solution. ( no, it���s not trivial ���.. ) 

The code to do the fixup can be simple.  The trick is that you need the time 
in seconds-since-tzero rather than year-month-day.  POSIX uses seconds since 
1970.

If you are on a POSIX system, you can use mktime and gmtime to convert back 
and forth.

Here is the fixup code:
  while (gps_time < magic_constant)
    gps_time += 1024*seconds_per_week

The magic_constant is the time stamp for when the software was built.
You can get that from date +%s

The fixup code will work for 1024 weeks after the date corresponding to 
magic_constant.
That's close to 20 years.

If you are working inside ntpd, NTP time starts at 1900 so the magic_constant 
has to be offset by 1970-1900.

--------


Tim Lister said:
> Given that one of the time servers provides the IRIG-B signal for the 400 Hz
> control loop for the drives for a ~20 ton 2-meter diameter telescope, I'm not
> keen on a homebrew Raspberry Pi solution however much I like them for my own
> time-nut home use...

Does that care about the date?  Or can you modify that code?

If you can run ntpsec, I can teach it to fix the WNRO problem from selected 
servers.  That would let you use the old/broken Symmetricomm gear to provide 
time to Linux or FreeBSD systems.



-- 
These are my opinions.  I hate spam.