Understanding the susceptibility to spalling of concrete members in case of fire is important to evaluate the residual load-bearing capacity. The investigations of the spalling phenomenon of a concrete mixture using real scale members are necessary but expensive to carry out. Reducing the specimen size leads to an increase of boundary effects that can result in a reduced spalling or absence of spalling. In this study, fire tests were carried out on unrestrained, single-sided exposed, cuboid shaped specimens (0.6 m ×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} 0.6 m ×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} 0.29 m) as well as unrestrained and steel ring restrained cylindrical specimens (Ø = 0.47 m, h = 0.29 m), which induce different boundary conditions. These fire tests were carried out on two ordinary concrete mixtures. The two mixtures differ only in the type of aggregates (quartz gravel and basalt grit) and were used to investigate the influence of the thermal expansion of the aggregate on the spalling behaviour of the concrete. The results show a significant increase of the spalling depth due to the restrained thermal expansion achieved by the applied steel rings. Additionally, the type of aggregate has a direct influence on the spalling behaviour of a concrete mixture. The reduction of the boundary effects by the steel rings recreate the test conditions in the centre of a large concrete member. Thus, this type of specimen is suitable to determine the susceptibility to spalling of a material (screening-tests) as preliminary investigations to full scale fire tests.