Modeling thermoelectric generator (TEG) performances plays an important role in guiding the design of TEGs to achieve better efficiency. However, a rigorous 1-D TEG modeling performance has not yet been conducted, which prevents reliable prediction of TEG performance. In this work, a detailed 1-D model has been developed to take into account temperature-dependent thermoelectric material properties, heat loss due to radiation and conduction, and Thomson effect. A Lead Telluride (PbTe) TEG was chosen as a sample module and the modeling results agree very well with the experimental results, which proves how powerful the presented detailed 1-D model can be used to predict and validate TEG experimental results. TEG power and efficiency were found to have a respective decrease of 10% and 31% from the simplified model at a temperature gradient of 570 K. While heat loss attributable to conduction and radiation were found to be small, the Thomson effect, which is often neglected, was found to significantly reduce TEG performances. The deep analysis enabled by the new model provides useful guidelines to improve the performance of TEGs. (C) 2017 Elsevier Ltd. All rights reserved.