A little problem with today rain

[Matt.j5b]

Hi

We got some rain this afternoon after a few days without rain, but I got a slight problem with that rain in Cumulus.


It has been a few days since we had rain and after I cleared that rain from a few days ago, the rainfall counter in the Nov11log and Dec11log has been 0 since then. When we got the rain today the rain today is 1.6, rain in last 24hrs is 19.1, rain in last 1hr is 0.5 and this month is 1.6. The 1hr and 24hr which I understand comes from the station is correct and 1.6 for today and this month should be 19.6 based on the station.


In edit, Today rain and also in today.ini, the start of day total is 18, equal to the 19.6- today of 1.6. The question is that I don’t understand why if the log file and the station haven’t started with a rain figure above 0 how can the start total begin at 18 when there was no rain yesterday and none until this afternoon. There was 0 on the station for a few days so the start of day should be 0.


Am I seeing this right? I have been having trouble with rain totals when we get rain. Apart from that all the data is downloaded with the correct figures so that seems to be not a problem. It is seems like it how the numbers are treated. I have solved this by changing the start total to 0 in today.ini and today total becomes the correct total of 19.6.


Just wondering why this is happening and it would be easier if I didn’t have to change the figures. But to tell you the truth it only is minor matter and can easily solved, it just a little annoying.

[steve]

Cumulus reads the "rain counter" from the station and calculates everything (including the last hour and the last 24 hours) from that. That rain counter only goes to zero if you explicitly reset it at the station (I don't know how you do that with a La Crosse, on some stations you have to remove the batteries from the rain transmitter). Cumulus stores the current value of this counter at the start of day (the Start figure in today.ini) and calculates everything based on the difference between the current value of that counter and the start of day value.

If you zip up the diags folder and attach it, I'll have a look.

[Matt.j5b]

Hi
Thanks for your help and the quick response.

The rain counter that Cumulus stores for the start of the day total, is that the total from the base station.

According to the Nov11log and Dec11log, the day started with 0 for the rain counter figures carrying over from yesterday but when I downloaded the data today it shows that start total as 18 and today rain as 1mm at that time.

La Crosse station can be easily reset at the station which I do for my seperate totals in my spreadsheets. The counter in the gauge itself doesn't reset in the history file of the Heavy Weather software that comes with the station.

Please find Diags attached.
Thanks

[steve]

The rain counter that Cumulus stores for the start of the day total, is that the total from the base station.

It's the counter from the gauge. However, La Crosse don't make it easy to determine the value of that counter when the data is coming from the logger, it's not in the logger entry and has to be calculated. In this case the counter came out slightly negative, which caused Cumulus to later think that it had been reset, so it started counting again. I'll see if I can fix this.

As an indication of what I'm up against, here's the description from the Open2300 site of the rain value in the logger entries:

14,13,12: Rain. (RAINCOUNTn)
The value is binary and steps by rainfall/impulse (from 0437) mm/step. Every period the current 12-bit rain count value (from 043A) is stored as history data.

Call the value for a specific history record 'RAINCOUNTn'.
The first time a program is started without a history data file but connected to the station, the program should fetch the total rain value from addresses 04D7-04D2. We can call this RAINref.
Then save the "Last rainfall total counter" value from 0440. Call this RAINCOUNTref. The program always saves just ONE value when it starts and this becomes a reference value for the entire time the program is running. From then on each new history record (past or future for the run of the program) is fetched by the program and the total rain for a single history record is calculated using RAINCOUNTn.
RAINtotal = RAINref + (RAINCOUNTn - RAINCOUNTref)* rainfall/impulse
If the total rainfall has been reset, or the remote sensor is reset (eg battery change) then the historical counter values will be out of sync and the program will need to handle accordingly.

[Matt.j5b]

Hi

Thanks for taking a look.
So it is the total from the gauge that has to be calculated when downloading from the logger which came out negative. I have noticed this a few times and editing the start total it if it is today or dayfile and yesterday if its yesterday.
I understand what you are up against because it not the best for pc connection and few software works well with the La Crosse. I not sure if would affect the gauge counter but would resetting the base station create problems with the start total even if it is not used.

Thanks for your attention and your support for the La Crosse

[Matt.j5b]

Hi
I didn't see new the build uploaded to try to fix this problem ( build 1022). I give that a go and see how it goes. The rain has gone and wont be back until next week. Until then I hope the attempt works. See how it goes.

Thanks for your help.

[Matt.j5b]

Hi

Got some rain today and the today rain figure was calculated correctly. The attempt to fix the rain counter calculation has seemed to work hopefully.
Thanks Steve for your help.

[steve]

That's good, thanks for the feedback.

[Matt.j5b]

Hi
No Problem Steve, I like to be of help, give the credit where its due and follow things through.

[Matt.j5b]

Hi

As a update we had rain since Thursday morning finishing up last night and the rain start was calculated correctly and so there is no problem with that, until I reset it in base station and start total goes from 38.3 with a today rain of 6.something to -6.7. I resetted the station with cumulus stopped and when I restarted Cumulus the start calcuation went wrong. That was last night.
After the rollover last night yesterday rain is 0. Of course this trouble has been caused by resetting at the station, which I have to do because once the rain finishes I want it to start at 0.
If there is no solution I will have to edit the start today to be correct. Thank you

[steve]

The rain calculation for La Crosse stations is difficult enough, even without resets. I'm sorry, but I don't really understand why you want to keep resetting it, and I don't think there's anything I can do.

[Matt.j5b]

Hi
I reset the rain total in the station when the rain finishes so that the total is 0 for when rain comes back and so the total doesn’t go up and up. For example the total of the gauge total in the heavy weather history file is at 170 and for my purposes I only interested in a total of each period of rain so that at the time I get rain I know how much rain we have got.
It is all right if nothing can be done, do only what is possible but I guess I have to either not reset it and cope with increasing totals or reset it and edit the today total.
It seems that there is no way to reset the total without affecting the today rain total. Thanks anyway for your help.

[steve]

Ah - so you're not interested in hourly/daily/monthly etc totals, but what are sometimes called 'storm' totals, i.e. the amount of rain in a particular rain 'event'? This is something I may add to Cumulus at some point, but most people are interested in rain over a particular period, so that's what I've concentrated on, and while it does cope with the occasional reset as best it can, it relies on the station's rain counter behaving sensibly in order to calculate those figures.

[Matt.j5b]

Hi


I have been resetting the total because for me it just hard to remember the total was this and now is this and the total for this lot of rain is that. I have been doing that my other La Crosse non pc station and maybe it’s a habit that didn’t have any consequence then but now I am using this one with pc connection.
Now it seems that it affects the counter I may have to shift. I am interested in the rain event totals because I do make my own calculation in my spread sheets for the daily total and monthly total ending at 9am using a manual gauge because it only comes in 0.5mm steps. But I do use Cumulus for max rain rate, hour rain, and the graphs, etc etc.
But I am certainly interested in 24 rain total because that is why I calculate my 9am total as I compare it with the Met Bureau. But as you say the storm total may a useful addition one day but I can easily work this out manually, I have been doing this for years.

Ah - so you're not interested in hourly/daily/monthly etc totals, but what are sometimes called 'storm' totals, i.e. the amount of rain in a particular rain 'event'?

I don’t have objection the hourly/daily/monthly totals that Cumulus creates, I use them in conjunction with the total amount.
I realise that resetting causes problems working out the figures, which is what the whole problem is occuring. It works perfectly fine thorough the whole rain event and the rain we getting at the moment is testing that, it is when I reset it is problem. If only I could reset without trouble, maybe I have to shift my ways.
Thanks Steve