Increase estimation accuracy with "panel" temperature


Is anybody uploading temperature from sensor at the bottom of or close to a panel?
Still debating is it worth extra trouble or just stay with personal weather station data close to the array…


I think this would be hard to standardize as some systems use the weather temperature while others use inverter temperature. Perhaps you could log panel temperature as a function of output in the extended data fields.


IMO It’s probably not worth the trouble to monitor back of panel temperatures. PV Module temperature coefficients are in the range of 0.3%/°C. By the time you consider the accuracy of the devices you’re using to measure your output and temperature, the temperature affect would be noise.

I’ve been involved with several utility scale PV Projects (~125 MWac) where we use very accurate devices to measure power, energy, temperature, humidity, irradiance, wind speed, etc. and we would typically only monitor the back of module temperatures of less than 0.1% of the modules. We would monitor ambient temperature, irradiance and wind speed at several locations throughout the site to account for variations in the conditions as the area covered would be in excess of 1500 Acres.


You need to enter “Panel Coefficient” then I thought the pvoutput system had two temperatures “air temperature” v5 and “temperature” v10 one of which is used to create a correction in the calculation. I think the v5 and v10 is correct
If you send the weather station data to temperature then the calculation will correct based on the “Panel Coefficient” you entered. I assume there is no simple way to relate ambiant to panel temperature.

From they say the difference in ambient to panel temperature is 25-35 degrees depending on how you mount - That would generate an error of greater than 7%. I of feel this is significant enough to be of interest.
I have thought about doing the same thing


With solaredge systems you can log the temperature of the optimizer which is most of the time very near to the back side of the solar panel.

I sort of assume it does add some temperature of its own.
For example
My system started at 05:30 and the temperature of the optimizers was 24.8 degrees Celsius.
Out side temperature was 13.5 Celsius.

At 06:30 the system produced 644 watts, the optimizers report 40.1 Celcius, the outside tempeature was not much more, perhaps 0.2 degrees.

Peak temperature was 45.5 at 10:45 when it produced 3300 watts
Than heavy rain came and temperature dropped to about 42 degrees.and 100-500 watt production

On very sunny days the temperature goes up to above 50 degrees.

But i doubt you can use this for measurement as i see no direct relation between panel temperature and optimizer temperature.