The diurnal cycle of precipitation is an important and fundamental cycle in Earth’s climate system, yet many aspects of this cycle remain poorly understood. As a result climate models have struggled to accurately simulate the timing of the peak and the amplitude of the cycle. This has led to a large number of modelling studies on the diurnal cycle of precipitation which have focussed mainly on the influence of grid spacing and/or convective parameterizations. Results from these investigations have shown that, while grid spacing and convective parameterizations are important factors in the diurnal cycle, it cannot be fully explained by these factors and it must also be subject to other factors. In this study, we use the weather research and forecasting (WRF) model to investigate four of these other factors, namely the land surface model (LSM), microphysics, longwave radiation and planetary boundary layer in the case of the diurnal cycle of precipitation over the British Isles. We also compare their impact with the effect of two different convective schemes. We find that all simulations have two main problems: (1) there is a large bias (too much precipitation) in both summer and winter (+19 and +38 % respectively for the ensemble averages), and (2) WRF summer precipitation is dominated by a diurnal (24-h) component (~28 % of the mean precipitation) whereas the observations show a predominantly semidiurnal (12-h) component with a much smaller amplitude (~10 % of mean precipitation). The choice of LSM has a large influence on the simulated diurnal cycle in summer with the remaining physics schemes showing very little effect. The magnitude of the LSM effect in summer is as large as 35 % on average and up to 50 % at the peak of the cycle. While neither of the two LSMs examined here capture the harmonic content of the diurnal cycle of precipitation very well, we find that use of the RUC LSM results in better agreement with the observations compared with Noah.