Cumulus Support

Hi, on two occassions now in the year or so I have run Cumulus I have had sudden extreme entries when opening the program. Today it recorded incorrect low temperature, high wind gust, high wind speed and low wind chill, all at the same recorded time. I luckily had a recent screen shot and was able to reinstate the correct records but after closing and reopening Cumulus I got a 350mm rainfall reading for today (in the past hour) and there has actually been none. This has messed up the totals which hopefully I can edit tomorrow in the dayfile. i also had a similar event some time last year. The odd thing is when I look at the Easyweather data the earlier wind related error shows (suggesting it is a weather station issue) but the rainfall one doesn't(!?) Any ideas?

You'll see lots of threads here about bad data readings with Fine Offset stations - many due to some sort of RF interference - and people's attempts to avoid them. Cumulus does attempt to filter them out where they are clearly wrong. You can't compare the readings in Cumulus with EW because (hopefully) you're not running them both at the same time. A short-lived problem can affect one and not the other if it doesn't last long enough to remain logged in the station memory. If you're getting them at start up, though, it does sound as though they are being logged.

Which version of Cumulus are you using? The latest version keeps a log of all time record changes to make corrections easier.

Thanks Steve,
The errors occured with the previous version of Cumulus but I have since updated it. Unfortunately today I got another extreme rainfall and reading. The rain rate showing is 4606mm /h and there is an approximately 350mm error in the amount. The error light was flashing but I don't see how to find what exactly the error was. Easy Weather has not recorded any errors so I should be able to recover the info but I don't want to have to do this too often so any tips would be appreciated.

equinox wrote:The rain rate showing is 4606mm /h and there is an approximately 350mm error in the amount.

Yes, sudden jumps in the total rain counter are a common problem. If you upload your June log file, I'll have a look at the counter to see what it did.

The error light was flashing but I don't see how to find what exactly the error was.

When the error light is flashing, click on it to see the errors. They may not necessarily be related to the rainfall problem. It could, however, be a 'sensor lost' indication, which might be a clue to the bad data you're getting.

Easy Weather has not recorded any errors so I should be able to recover the info

There's a guide to correcting the rainfall in the FAQ: http://wiki.sandaysoft.com/a/FAQ#My_sta ... it_to_zero

but I don't want to have to do this too often so any tips would be appreciated.

I'm not really an expert at sorting out the problems with the hardware, but if you have a look in the Fine Offset section, you'll see a number of threads about it. The first thing to look at is how your rain gauge is mounted, it may be a simple case of vibration tipping the gauge - although in that simple case you would expect Easyweather to record it too. I have no idea how EW works, and don't have access to full details of how the station works, so I don't know how EW avoids recording large amounts of rain when the counter jumps up, as I know it definitely does. Which, if either, of Cumulus or EW was actually running at the time of the erroneous rainfall?

hi Steve,
I really appreciate your prompt and concise replies. I did wonder yesterday if the error only occured when Cumulus was actually running but just to put paid to that theory I see that last night (when nothing was running) another rogue rainfall entry appeared. I have corrected the records again this morning.

This is where the problem occurred:

11/06/10,14:00,15.8,94,14.8,3.6,22.0,261,0.0,9.3,985.0,409.5,19.3,62,9.7,15.8,15.8
11/06/10,14:10,14.2,94,13.2,2.5,89.3,181,0.0,9.3,985.3,25.5,19.4,62,89.3,14.2,14.2
11/06/10,14:20,15.9,93,14.8,2.5,89.3,179,0.0,393.3,985.3,409.5,19.5,62,7.2,15.9,15.9

Between 14:00 and 14:10, your rain counter went down from 409.5 to 25.5 and then back up again to 409.5. Cumulus assumes that the counter has been reset when it goes down, so then counted the new value as new rain. At the same time, you had a (presumably erroneous) wind gust of 89.3 (which shows on your screenshot). So I guess some sort of interference corrupted the data around that time.

Did you click on the red light - and if so, did it show "sensor contact lost"? If you zip up your diags folder and attach it, I'll have a look.