[time-nuts] Re: HP Z3801A project update

[ed breya]

I did a few more experiments on changing the initial date, hoping to get 
it to accept more recent values, and still give correct date reports. It 
seems like there's an absolute limit somewhere in 2007. 01 Jan 2007 
seems to work, and during startup, it momentarily forces 28 Dec 2006, 
but still gets the right current-date answer as soon as the first 
satellite data arrives.

I still don't get the numeric connections, so have to assume there's 
another hard-wired constant built in there somewhere, limiting the 
setting range, and the total GPS week count rollover. So, it looks like 
it will be good to go for a while (I think until 2026), but at some 
point it will be out of sync again with the actual GPS weeks, and fail 
on the date reporting, if a newer initial date can't be accepted. Again, 
this is not a big deal - I just want the 10 MHz and 1PPS to work properly.

For now, I have it all working pretty well, and the spare GPS RX seems 
OK, so unless the Z3801A itself craps out, it should all be good for 
quite some time. I've gotten off on all sorts of tangents, but can now 
focus on wrapping up the internal battery backup system again.

Ed