I have just come across a discussion of the rainfall problem, where the EW software corrupts the rain data. There was a URL (below) to download a later version which is supposed to resolve the problem.

http://home.exetel.com.au/daygle/Fil...v2.0(4.16).zip

I have just installed, so we will see.

I'll give that one a spin too - thanks

===post install===

Program size is identical to the version with the rainfall bug - 1,262,568 bytes so my betting is that this is the same code. It also highlights the ridiculously lax way version control has been implemented (or more accurately hasn't been implemented)

__________________

There is another discussion I have found where they have said V4.16 still doesn't solve the problem completely and they have been in touch with the manufacturer. The manufacturer suggests reducing the update interval to 30 minutes as a means to reduce the occurance of the problem, but they are aware of the issue and are getting their software engineer to provide a fix.

It seems the problem is due to some error in the dat file where wrong dates and times are assigned to the rainfall data.

I have run the V4.16 for 24 hours at 5 minute update intervals with no errors showing as yet.

It doesn't fix the problem - I have just suffered another ficitious monsoon. I don't think it is a different software version.

__________________

My EW V4.16 has been running whenever my PC has been on, since last Wednesday (7 days) with plenty of rainfall and none of the 'false rain' recorded yet. My impression is that the false rain only appears when there has been some actual rain fall.

7 days without any corrupted rain data is in itself a record.

That's my impression too - no rain no corruption but as soon as some gets recorded there is a chance it will go haywire.

I think I'll move out the next duff file rather than just deleting it then I can compare it with the new one created from the unit memory. I think the unit only records instantaneous rainfall and the software calculates the other figures - you can see it doing that when it first starts with no dat file. If that is so then I could maybe modify my Python script to redo the calculation and maybe return to reporting rainfall through UIView.

__________________

Speak of the devil...

There had been no corrupted upto my turning the PC off late last night. I got home today, turned it on, ran the EW software and todays rain data was corrupted during the course of downloading it from the displays memory.

I shut it back down, then used Wordpad to edit out (delete) all of the data downloaded from the display today in EasyWeather.dat - all the date lines showing todays date and time as the first entry of each line. I then ran EW again, which redownloaded the same set of data again, but this time no corrupted rain data.

During the deletion of todays corrupted download of data, I noticed that a good half of the data comprised that which it had downloaded from it today, the other half being the data it had downloaded for the previous 7 days worth of data. So the corruption also stores in the file as a much larger set of data.

You guys might want to take a look at this in the projects section Cumulus weather station software - Midlands Weather Forum - A UK Weather Forum - Projects and let me know how you get on.

__________________

Odd, I received seven emails advising me there had been a follow up message to the thread, but with nothing added until the above a few minutes ago.

I run Cumulus with Easyweather as a backup and to make resetting the unit easier. I've just noticed that Cumulus recorded 20.1mm of rain for December where Easyweather recorded only 15.6mm which is a reverse of the 'Monsoon effect. I ensure that Easyweather is updated regularly and, as an extra precaution, before clearing the unit memory and I've checked that there's no missing days in the record. Don't ask which version of EW I'm using as I agree that the version numbering is a joke! BTW, I've been warned not to have both programmes running simultaneously!