As per the discussion on wx-forum (Davis flatlining) a suggestion for adding flatline-check for Davis.
The issue:
Every now and then WD goes flatline whats mess up the data. The setup is Davis - Virtual VP - WD. The core problem is not WD itself as also WL stops getting data but WL handles this properly (stops getting data) while WD do not (goes flatline).
The “proper” response to a loss of data was debated a long time ago and the concensus at that time was that it was better to continue recording the last reading during what could be only a short data drop out rather than stop logging. The question is what makes the bigger mess?
Trouble is that some of us may not immediately notice that WD has been flat-line graphing and logging (for up to 5 hours in my case) and if WD carries on graphing and logging then the history data on the logger is not easily recovered. Which is why I asked Brian to make this mod for WMR200: see my (later) posts in this thread Weather Display goes flat line. Works fine (with one new problem that may not have anything to do with the mod!).
I also suggested it might be made an option, so that individual users could choose.
While a pain it’s not hard to recover the “lost” data from the data logger. I know, done it many times since my IP logger also used to flatline. The bigger pain is getting to logger to connect to the network again.
It should stop if the data have been the same say 5 mintues as then its not a short drop anymore. As bitsostring stated it can take hours before a flatline is noted as WD continous also uploading clientraw.txt which again do all kind monitring that file unusable for this case.
See the thread I mentioned above on flat-lining: if I restarted WD after a few hours flat-lining it would read the WMR200 history data saved on the console logger OK but it didn’t seem to do anything with it except write a raw data file that I can’t decode. It certainly didn’t go back and overwrite the flat-line log/graph.
I suggested a minute to Brian, because it’s often 30 secs between live data “packets” from the WMR200, and I think he put in 40 secs. But it shouldn’t detect flat-lining “if the data have been the same” for 40 secs, only if live data has stopped arriving altogether for 40 secs.
The original post in this thread is talking about the Davis, there is no mention of the WMR200. With the Davis it is not that hard to go back and get the missed data.
the issue though is even with the Davis, if WD flat lines, i.e the com port was lost (e.g USB data logger, electrical interference (e.g fluroescent light)
then WD keeps on logging the same data from the last data received
you can set WD to get the missed data from the start of that flat line, but you need to set that time and delete those repeated data in the logfile first, then after the extraction and missed data catch up, convert that logfile to data file
the improvement for this situation is that WD detects no more data arriving (I assume when the flat line occurs, the data received light stops flashing? (I do not have this problem myself, hence why I ask)
and so then WD stops the logging the data and set the time/date for when its restarted to that time that the flat line starts
sort of thin
I can easily test here, by simply pulling the connection to my spare VP console
and so I can work on this over the coming days
Sounds good.
I haven’t checked the data light during the flatlines.
the issue though is even with the Davis, if WD flat lines, i.e the com port was lost (e.g USB data logger, electrical interference (e.g fluroescent light)
then WD keeps on logging the same data from the last data received
As you know, the WMR200 mod was effectively offered to Davis users in WXForum - see weatherc’s original post, above - so I thought I’d mention how well it meets my requirements. Up to you lot to help Brian make it work for Davis!