The objective of this study is to evaluate three different phase change models used to predict the energy behaviour of a PCM cement mortar sample. Reference data are measured on large samples of composite material using a special experimental set-up. The phase change models tested are: the apparent specific capacity method, the enthalpy method assuming a pure body and the enthalpy method assuming a binary mixture. Numerical results are compared to the reference data of heat flux and energy stored/released. The main conclusions of the study are: (1) the thermodynamically inconsistent apparent specific capacity method is not suitable, (2) the enthalpy method gives better results than the specific capacity method and (3) the enthalpy method gives better results with an appropriate guess of the enthalpy curve. (C) 2014 Elsevier Ltd. All rights reserved.