With the lack of rain locally this season, I haven't been able to play with this much until yesterday.
This trial was done using an RG11 in drop counter mode. In an earlier unpublished experiment, I logged drop counts every second. That was too fast, but I did observe one characteristic of the RG11. In drop counter mode, rather than one relay click immediately on detecting a drop, when multiple drops are detected in a short time, it appears to batch them up and then click as many as 5 times in a row, maybe more. The most number of counts I could get was 5 per second.
So, yesterday's test was recording the total counts 1x per minute. This morning, I downloaded the log into Excel, subtracted the last minute's total drop count from the next minute to get drops per minute, and graphed the resulting data.
One other note on this graph. In order to just get the drop count data, I had the controller stop logging after 5 minutes of no drops counted. So, zero count flat lines are an indeterminate length of time. Actually, in looking at the graph, that's almost 11 hours of data (at least 649 minutes) so there is very little down time in there...
Like my acoustic disdrometer, it's pretty cool to have enough resolution to be able to see bands of rain pass by.
Next is to try to get some correlation between this drops per minute value and something more familiar like inches per hour. That's actually what started this. The standard inches per hour as measured on one of my Davis stations calculates the rain rate based on the time between tips of the rain gauge. For example an average rate of 0.10" per hour is a tip every 6 minutes. The actual rain fall intensity can vary a lot in that 6 minutes.
The end goal is to use the drops/minute to drive some sort of indicator through an Arduino or other controller, maybe a dial gauge, bar graph, or even a color changing LED.
