Analysing the problem with my rain sensor

During winter period I encountered for a longer time problems with my rain sensor.

Even when weather was dry I got indication off rain from my cloudwatcher

I tried to analyze the problem, perhaps I found a better setting

What happened with my rain sensor during longer period of rain and cold weather.

I found that the measured rain sensor value was at 0 and indicating status unknown with a violet colour

I tried to change setting values for the rain sensor heating.
The heat impulse values did not bring any change and it seemed that heating pulse was not responding.

Later I understood that for a heating impulse the measured rain sensor value needs to be higher than 400, and mine was well below this value.

Then I tried to increase the Delta low value and as result I noticed the heating to increase and temperature rising from the rain sensor. This looked good to me and after some time I noticed the rain sensor value to indicate a value of 164, so much higher than the 0 from before.

Also I noticed that increasing the heating resulted in drop of the rain sensor value which felt wrong.

I decided to leave the heater to a value that it heated to about 20 degrees Celsius a night over.

The next morning the rain sensor value had risen to about 1600. I then had a day with nice sunshine dry weather and a max temperature of 7 degrees.

I still noticed that increasing the heating level the rain sensor value dropped.

It still needed a day before the rain sensor value stopped rising at a level of 2304 as maximum.

At that level I again tried to increase the heating level and then I found that the highest rain sensor value was almost stable at 2304.

I changed my heating parameters in that when the heating impulse came up it would have a much stringer and much longer effect in heating

These are my settings, I still need to experience if this will work for a longer period but for now this gives me a much better performance.

I expect that during a longer rain period, the inside of the Cloudwatcher box was getting a high humidity and even perhaps moisture within the box.
At starting up the heater this moisture resulted in a higher humidity at the rain sensor and therefore decreasing the measures value.
After a longer dry period with also some increased heating power the inside from the box dried up and moisture or humidity dropped to a level not influencing the rain sensor measurement.

The impulse heater seemed to me as not strong enough during long periods of rain at low temperatures.
So my solution was to increase the heating during cold weather and to increase the value and period of heating impulse.

Will have to wait for some rainy days to see if this setting will work and keep the rain sensor also within working limits.

Unfortunately this did not help very much and I still had problems with the rain sensor

I tried to test the rain sensor for error spots using very small water drops and see the response from the sensor

I found several spots where the sensor went into error, see the picture were these spots in my sensor are

Mostly on the edges off the sensor I found places where a drop of water would lead to erroneous readings

I decided to make a new glue layer on the edges of my rain sensor and then tested again for error spots.

Rain sensor with new glue on the edges and some water drops for testing

With the new glue on the sensor I did not find any spots leading to errors.

After this I mounted the cloudwatcher outside again and with the increased settings for heating the cloudwatcher seems to work fine again.