JJY Testing Feedback

[koruptor]

Thanks for txtempus! I'm very happy this exists and this was a fun project to practice my crappy soldering skills.

I can confirm JJY works with one of my watches! However, I found something interesting worth sharing. Not sure if this has to do with the code, the JJY signal, or the hardware of the watch.

Background on my watch

The watch I'm using is a Citizen PMD56-2951 (H100 movement). The way this watch works, and I'm not sure if this is the same for all other JJY-capable watches, is it orients to JST.

Instead of setting a UTC offset like most radio controlled watches, a user of this particular watch needs to set a JST offset. What that means is that if I'm on the Pacific timezone, I don't set this watch to -8. I have to set the appropriate offset to JST which is -16 or -17 hours depending on Daylight Savings or Standard. At the moment the watch is set to -16 to JST.

Using txtempus with this watch

So when I run the code without options other than choosing JJY40 or JJY60, it sets my time 16 hours behind my timezone (i.e., 16 hours behind Pacific). In order for it to sync correctly to my watch, I have to set -z 960 (adding 16 hours) for it to give me what I need.

The other option is to set the watch's offset to 0 and not pass any value to -z.

I just wanted to share this in case the Citizen H100 community is experiencing some weirdness.

[AdamLaurie]

I can also confirm it works with Seiko 3B21 movement, both JJY40 and JJY60. The offset can be set from -20 to +4 and although the manual says it's from "Japan Time", it is actually from whatever time signal you transmit, so just works normally.

[hzeller]

I've added a section in the README for JJY to point to this discussion in case someone runs into the same trouble.