Five-parameter single diode models are the most used for photovoltaic modeling, and although they usually apply the same basic equation, there are several variations in terms of parameter dependences on irradiance and temperature. Different methods with variations on the resistances? values with the temperature or irradiance have been proposed, and there is also discussion on the treatment of the dark reverse saturation current with the temperature level. On experimental works, there is even more controversy. Several authors reach distinct con-clusions about the behavior of almost every parameter of the single diode model when extracting them at each irradiance or temperature levels. For instance, in accordance to different publications, series resistance may either increase, decrease or remain constant under irradiance variations. In this paper, we review previous studies and their proposals, discussing also the possible influence of the extraction method on these results. Later, two experiments are performed in a solar simulator to extract all the parameters on a broad range of irradiance and temperature levels with an analytical extraction method for crystalline silicon modules. The results show that the thermal coefficient used to correct the open-circuit voltage for different temperatures may depend on the irradiance level, a finding that is not present in the single diode models found in the literature. The most common approaches for the diode ideality factor and the diode reverse current, constant ideality factor and a temperature variation for reverse saturation current are seen to fit with the experimental data, while the parasitic resistances both presented a significant increase for low irradiance levels and no significant dependence on the temperature. These experimental findings, reached using a set of 27 commercial PV modules, may help improve the devel-opment of methods for solving the single-diode model and improving its accuracy at different operating conditions.