Page 1 of 1
GMT/BST question
Posted: Sun 27 Dec 2009 10:04 am
by cats_five
I have setup the weather station to synchronize it's clock with the PC. and am wondering what happens when BST comes into effect and the PC moves the displayed clock forwards - I want the weather station on GMT (Zulu) the whole year round.
Re: GMT/BST question
Posted: Sun 27 Dec 2009 12:48 pm
by steve
cats_five wrote:I have setup the weather station to synchronize it's clock with the PC.
I'm not familiar with how you arrange that - I do have a Fine Offset station, but it's just for testing and experimentation. I assume it's a function of Easyweather?
and am wondering what happens when BST comes into effect and the PC moves the displayed clock forwards - I want the weather station on GMT (Zulu) the whole year round.
As far as Cumulus is concerned, it timestamps everything with the PC clock, so to have everything timestamped with GMT year-round you would have to arrange for the PC to stay on GMT.
Re: GMT/BST question
Posted: Sun 27 Dec 2009 2:54 pm
by cats_five
Yes, you are correct - I was getting confused.
It would be a nice enhancement to Cumulus 2 - to be able to ignore local time and always use GMT.... After all most of the Aviation world uses Zulu for all reporting purposes.

Re: GMT/BST question
Posted: Sun 27 Dec 2009 4:20 pm
by JohnG
I had a similar question about the data from the Lacrosse 2300. Is it is possible to use the timestamp that is presumably in each data record, and is based on DCF77 radio transmissions?
Re: GMT/BST question
Posted: Sun 27 Dec 2009 4:46 pm
by steve
JohnG wrote:I had a similar question about the data from the Lacrosse 2300. Is it is possible to use the timestamp that is presumably in each data record, and is based on DCF77 radio transmissions?
Each history record does have a timestamp, so Cumulus could in theory use that. But then it would have to use the station clock for 'live' data too, and it just doesn't work that way - some other stations don't timestamp their history entries, and I have to try to cope with them all using as much common code as possible.
Edit: Actually, what I wrote is nonsense (I have a bad cold and my head is even more fuzzy than usual). It
does use the timestamp from the WS2300 logger entries - it has no other choice. But it uses the PC clock for 'live' running; the assumption being that your station and PC are set to the same time.
Re: GMT/BST question
Posted: Sun 27 Dec 2009 6:28 pm
by beteljuice
BTW Yesterday my La Crosse decided it was the 22nd Dec - not for the first time, the 'time-stamp' got corrupted. So Cumulus actually (live) rectifies date errors untill I eventually notice the station clock !
Re: GMT/BST question
Posted: Sun 27 Dec 2009 6:34 pm
by steve
beteljuice wrote:BTW Yesterday my La Crosse decided it was the 22nd Dec - not for the first time,
I think time-signal clocks do sometimes throw a wobbler. My bedside alarm (MSF) used to set itself to weird dates and times on occasion, which can be a bit of a nuisance. Since they moved it from Rugby up to the Midlands, it seems to be a lot better

Re: GMT/BST question
Posted: Sun 27 Dec 2009 7:25 pm
by JohnG
So Cumulus uses a mix of radio and computer clocks. Couldn't that lead to confusion (e.g.jumps in the data intervals), after all, PC clocks aren't necessarily very accurate either?
Re: GMT/BST question
Posted: Sun 27 Dec 2009 7:39 pm
by steve
JohnG wrote: PC clocks aren't necessarily very accurate either?
My PC clock is quite accurate; it's synchronised to an internet time server (aren't most?), and is therefore within a second or so. But ideally I'd provide an option to use either PC time or Station Console time, for those types of station where this is feasible.