In developing countries, the increase in the electricity access and therefore in the energy consumption through distributed energy solutions, leads to searching energy saving strategies. Solar Thermal Systems (STS) have been widely used to increase energy efficiency in residential sector in many countries, and several models are being developed to improve their performance and deployment. However many of existing software tools somehow fail to provide more sensitivity analysis to different parameters of STS performance, given their wide applicability.In this paper, two dynamic approaches to model STS are discussed: a developed hourly model based on energy balance of the STS components, and the simulation tool Polysun. A sensitivity analysis is made to different domestic hot water demand profiles and tank sizes, using as case study, a typical 5-person family house in Maputo, Mozambique.The results show that the developed hourly model can be used to evaluate and size solar thermal systems, having similar results to Polysun. However, regarding the auxiliary heating, Polysun approach leads to a higher or lower electrical auxiliary heating, depending on the control layer, resulting in over-sizing or under-sizing of the system, given that it considers the temperatures stratification while the developed model considers the energy balance in the tank as a whole.