SOL modelling is a key requirement to design the next generation of burningplasma experiments [1]. This is particularly true for Edge Localized Modes,because the energy released by an ELM in ITER could raise the target temper-ature above the tolerable limits. The simplest ELM models introduce ad-hoctransport multipliers. They should be carefully benchmarked against existingexperimental data, before attempting any extrapolation to ITER. The ideal testcase should be sufficiently well diagnosed to provide the input information re-quired to define uniquely all the free parameters in the model, plus a set ofadditional measures to test the output. These are demanding conditions, oftenmet only partially in practice. We illustrate the benchmark of the B2-solps5.0code using an ELM from JET. In section 2 we describe the test case we chose,in section 3 we present our results and in section 4 we draw our conclusions.