In any given year, mean monthly irradiation readings can deviate considerably from the long-term mean values indicated in this book, particularly during the winter. However, such discrepancies mainly relate to direct beam irradiation HB. When the H values that differ considerably from the mean values are converted to HG values on the solar generator plane using the R(b, g) factor (see Section 2.4), a major error arises because R factors are determined by the ratio of HB to HD. In such cases as well, using the three-component model (see above) in conjunction with the following simple method allows for a far realistic determination of the HG values that actually occur: 80% of the H value exceedances relative to mean monthly values in the tables is allocated to direct beam irradiation HB, while only 20% is allocated to diffuse irradiation HD. The H values that fall short of the tabular values are evenly divided between direct beam irradiation HB and diffuse irradiation HD. In cases where the H values fall far short of the tabular values, HB can of course never be negative, and in extreme cases HD — H would occur.