[time-nuts] When does NIST change to DST?

Tom Van Baak tvb at LeapSecond.com
Tue Mar 15 15:57:59 EDT 2016


Nick, Andy,

There are hundreds of different WWVB RC clocks & watches, made by dozens of companies over the years. It turns out that many do not handle DST, leap seconds, or other boundary cases perfectly. I've never seen the problem caused by WWVB itself, though poor reception can contribute to a poorly designed clock IC making a mistake.

Since you're interested in this topic, I would encourage you to do some fun experiments. You can wait twice a year to check how your particular clock(s) handle DST and then guess what happened and why. Or you can generate your own in-house WWVB AM test signal and find out for sure. You can craft a series of DST tests and get results within hours instead of just twice a year. You can even recreate incomplete, incorrect, or noisy subcode bits.

There's a bunch of pages on my web site that try to do this. Basically all you need is a sig gen for the low-power 60 kHz carrier, a dangling wire for the antenna, a 10 Hz timebase for the subcode, and a PC or Arduino or whatever to send the subcode bits.

The 60 kHz carrier doesn't need to be perfect since RC clocks don't track the carrier. If you don't have a bench instrument try using a 192 kHz sound-card instead.

The 10 Hz needs to be accurate to 100 ppm or the RC receiver won't properly track the subcode across several minutes.

One trick is to use a FTDI serial/USB converter -- set the baud rate to 10, yes 10 baud, and then send 0xFE or 0xF0 or 0x80 bytes for the WWVB 0, 1, or P bits. Some sample code:

http://leapsecond.com/tools/tco1.c
http://leapsecond.com/notes/wwvb2.htm

See also:
https://www.febo.com/pipermail/time-nuts/2011-November/060937.html

/tvb

----- Original Message ----- 
From: "Nick Sayer via time-nuts" <time-nuts at febo.com>
To: "Discussion of precise time and frequency measurement" <time-nuts at febo.com>
Sent: Tuesday, March 15, 2016 10:45 AM
Subject: Re: [time-nuts] When does NIST change to DST?


Well, they flip the “warning” bit at 0h UTC (4PM PST the afternoon before), so I think the explanation is that my clock was able to sync up at that time, and it reacted to the warning bit rather than using that as a cue to make the change at 0200J like it should have. Either that, or the receiver erroneously received a ’11’ rather than a ’10’, or WWVB transmitted the bits wrong (which I’d think unlikely, except that it wasn’t just my clock that reacted early).

> On Mar 14, 2016, at 8:23 PM, Andy <AI.egrps+tn at gmail.com> wrote:
> 
> Yes, what I meant by "switch to DST" was that it flips the DST bit.
> Not that the timecode itself changes.
> 
> Since the time many clocks use to sync up with WWVB is (from what I
> recall) around 2AM, it seems kind of dicey whether they would make the
> change on the right day.
> 
> I think that doesn't explain the change happening at 2000J -- unless
> your receiver was receiving and syncing to WWVB at that time and they
> had already flipped the bit.
> 
> Andy
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

_______________________________________________
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.


More information about the time-nuts mailing list