Abstract The lack of comprehensive solar radiation monitoring during the longest total solar eclipse of the 21st century at Tianhuangping (Zhejiang), China, on 22 July 2009, has led to this investigation in order to evaluate the cloudiness contribution in estimating the impact on global solar radiation throughout this phenomenon. In doing so, we applied a cloud cover empirical model to obtain the global solar radiation and, at the same time, we deduced a theoretical model to get the direct solar radiation in which both the occultation and obscuration functions of this eclipse are included. We took limb darkening and atmospheric transmission into account. Though the weather during our eclipse observations agreed with the forecasts for that day, clouds and some rain, we were nonetheless able to observe all phases of the eclipse from our observation site at Tianhuangping. This experience suggests that for coming eclipses a record of the in situ observation protocol of cloudiness is mandatory. Our results for comparing global solar radiation models indicate that our total solar eclipse radiation model is quite acceptable and representative of that which could have happened at that time.