It’s interesting what you see when you graph the data coming off a sensor. Take, for example, the battery voltage in the Tempest weather station. As you’d expect, the battery voltage drops when the sun goes down and the solar panels are no longer charging the battery, then starts increasing again once the sun comes up. What I can’t explain is the steep drop around 11AM every morning, followed by a gradual climb to max.
The charger does not start immediately when the sun comes up. In regards to voltage when you are near full it will discharge to a certain point before starting again, this is to minimise cycling which is one of the reasons a battery doesn’t go long in time. It’s a balance you need to find between keep it as full as possible versus not to do like 10 cycles a day …
I don’t seem to be getting anywhere near those dips:
My tempest seems to be powered by the force rather than electricity lol.
Is your tempest pretty far from your hub? I’m assuming broadcast power is constant, but maybe both of yours needs to increase transmission power to reach the hub? Mine is less than 50ft away.
Mine’s not terribly far away, but it’s gotta go through a couple wood frame walls to reach it. Probably around 60’ away.
If you were to watch the angle of the sun hitting the solar panels at this time of year it makes perfect sense. At solar noon the sun is hitting the solar panels at such a steep angle that the effective solar radiation they receive is reduced. The further south you are the more pronounced the effect will be. In the winter time the effect should be much less since the sun is not nearly as high in the noon sky.
Here is a purposely over exposed picture of one of my field test Tempests mounted by my CoCoRaHS gauge.
Note that there is even shadowing from the case on the panels. If you look at my production Tempest 22968, which appears to have the same charging profile that yours does, you can see that even though it hasn’t reached solar noon here yet, the voltage has dropped slightly. When the sun gets past noon it shines more directly on the side panels increasing the charge rate.
Also, note that the voltage change you are looking at is near 100% SOC and that less than a tenth of a volt bobble represents very little stored energy as I referred to in this post.
I’ve seen the same scenario at my location in Hampton Roads - near Virginia Beach, VA.
That explanation would seem to make the most sense!