Aiming on the evaluation of SI-engines with extended expansion cycle realized over the crank drive, engine process simulation is an important tool to predict the engine efficiency. One challenge is to consider concept specific effects as best as possible by using appropriate submodels. Particularly the choice of a suitable heat transfer model is crucial due to the significant change in cranktrain kinematics. The usage of the mean piston speed to calculate a heat-transfer-relevant velocity is not sufficient. The heat transfer model according to Bargende combines for its calculation the current piston speed with a simplified k-ε-model. In this paper the eligibility of this model for engines with extended expansion is assessed. Therefore a single-cylinder engine is equipped with fast-response surface-thermocouples in the cylinder head. The surface heat flux is calculated by solving the unsteady heat conduction equation. Through surface-ratio related weighting of local heat fluxes the global wall heat loss can be concluded. This natural-gas engine has a multi-link cranktrain to achieve, based on a compression ratio of 12.2, an expansion ratio of 17.6. The crank train is later modified through a manual mechanical adjustment in order to set the strokes to equal lengths, establishing a “conventional” engine process. This enables the comparison of experimentally determined heat transfer characteristics from one engine with different engine processes. The comparison between experimentally determined and modeled heat flow at the conventional engine process shows a very good conformity. As well, operating at the extended expansion process, a good conformity of measured and modeled data is shown, so the heat transfer model is validated for this engine process. Subsequently the difference in wall heat loss through extended expansion is analyzed by using engine process simulation. In comparison with a base engine with an equal intake stroke, a higher wall heat loss occurs.