The rapid expansion of rooftop photovoltaic (PV) systems in urban areas provides substantial renewable energy capacity while also altering the local urban microclimate through changes in roof radiative properties, surface temperature, and roof-atmosphere heat exchange. Despite growing observational and mesoscale modelling studies, building-resolving large-eddy simulation investigations with direct comparison to rooftop measurements remain rare. In this study, we evaluate the rooftop PV parameterization in PALM for urban microclimate applications using observations from a rooftop PV system near Dresden, Germany. The evaluation uses three clear-sky, weak-wind cases from late summer to winter, with measurements of radiation, PV surface temperature, eddy-covariance sensible heat flux, Doppler lidar wind profiles, and PV power output. The simulations reproduce incoming radiation well and capture the main diurnal timing of PV surface temperature, sensible heat flux, and power output. Simulated PV surface temperature captures the observed diurnal cycle, but with a reduced amplitude (R² = 0.98, slope = 0.79). The model underestimates daytime surface temperature and overestimates nighttime surface temperature, with MB and RMSE values of 1.4 and 4.1 °C. Sensible heat flux shows substantial agreement (R² = 0.86), but the amplitude is reduced (slope = 0.79) and the mean bias is negative (MB = −22.0 W m⁻²), with an RMSE of 36.5 W m⁻². PV power output follows the observed diurnal shape and peak timing (R² = 0.98), but the amplitude is overestimated (slope = 1.15) and the mean bias is positive (MB = 14.0 kW), with an RMSE of 23.7 kW. Compared with the No-PV case, the PV case shows stronger daytime surface heating and roof-atmosphere heat exchange. Sensitivity experiments show limited effects of grid refinement from 5 m to 2 m but stronger sensitivity to PV layout representation. A uniform fractional PV layout provides a useful approximation for roof-mean surface temperature, whereas a layout preserving the PV arrangement is preferable for sensible heat flux and power output. These results provide a basis for further testing the PALM rooftop PV parameterization in more complex urban and meteorological settings.