WH3080 "Station not Responding"

[Flosex]

I have been using a Nevada WH3080 station since 20 February 2014. It has been working fine, initially with a Windows XP netbook and latterly with a new Windows 8 netbook running 24/7.

Two days ago I got the dreaded "Station not Responding" error message in Cumulus, despite the WH3080 screen continuing to display the correct data received from the sensors outside. Despite rebooting the netbook, I cannot get the station to communicate with the netbook. I also tried Easyweather and got the same error message. I tried the WH3080 on another Windows 8 laptop and got the same error message in Cumulus, confirming the problem is with the WH3080

Cumulus says to remove the batteries in the WH3080 but if I do this, will I not lose all the data in the WH3080? Is there a way to retrieve the data from the WH3080 and re-set it?

Thanks for any help/advice.

Chris
(West Sussex, UK)

[steve]

I'm not aware of any way to do it without losing the logger data. It's often not the data that gets lost, but the 'pointers' get reset. The station may or may not then start overwriting the needed data, depending on which location it was previously using. In theory, if the data didn't get overwritten, Cumulus could remember which memory location was being used previously and ignore the current pointer position until it had downloaded the missing data. I may look at this one day, but it's quite tricky to do reliably.

[Flosex]

Thanks Steve. I disconnected the station, removed and re-inserted the batteries and it is now working again, but I lost my data for the period it was down.

Any idea what caused this to happen and is there anything I can do to prevent it? I do have the Fine-Offset synchronisation set to on in the Cumulus settings.

Regards

Chris

[steve]

No idea, sorry. The 'sync' setting seems to pretty much eliminate the well-known USB lockup problem, so I guess you have a different issue.

[jim-easterbrook]

What margin do you allow for Steve? I've got a station that drifts at about 3½ seconds a day, so a 3 second margin wasn't enough to be sure of not causing USB lockups. I've tweaked pywws to measure the drift rate and continually adjust the time to avoid, but this is probably a step too far. A larger margin would be easier.

[steve]

Good point. The default is 3, and this can be changed by adding a line

FOReadAvoidPeriod=N

to the [Station] section of cumulus.ini (where N is in seconds).

The drift gets logged to the diags file when the re-synchronise takes place, to give an idea of whether it would be useful to adjust the setting.