Showers at 14 h and 16 h caused deep minima of illuminance. Very clean air at the and of the night enabled unusually low level of night illuminance. Cold detector led to high voltage before the noon (a need for another daylight calibration at these temperatures is obvious).