Hydrogen-fuelled internal combustion engines are an attractive alternative to current drive trains, because a high efficiency is possible throughout the load range and only emissions of oxides of nitrogen (NOx) can be emitted. The latter is an important constraint for power and efficiency optimization. Optimizing the engine with experiments is time consuming, so thermodynamic models of the engine cycle are being developed to speed up this process. Such a model has to accurately predict the heat transfer in the engine, because it affects all optimization targets. The standard heat transfer models (Annand and Woschni) have already been cited to be inaccurate for hydrogen engines. However, little work has been devoted to the evaluation of the flow-field based heat transfer model, which is the topic of this paper. The model is evaluated with measurements that focus on the effect of the fuel, under motored and fired operation. The experiments were designed with DoE techniques to systematically investigate the effects over the entire parameter space. The results demonstrate that the flow-field based heat transfer model is better able to capture the effect of changing in-cylinder gas properties under motored operation compared to the model of Woschni. However, it is not accurate during combustion either, so model improvements are required to increase the modeling accuracy.