Experimental simulation of boundary conditions while carrying out a panel buckling test has been known to be difficult. In this paper, compression buckling tests on 6.0-mm-thick carbon fiber composite (CFC) panels have been described for which a reasonably good simulation of the boundary condition was achieved, especially under the simple supported condition. Carefully designed test fixtures that provided excellent alignment of the panel with the loading axis and adequate lateral stiffness for edge supports during buckling of the CFC panels were used to achieve this goal. Both simply supported and clamped boundary conditions have been studied. Tests in room-temperature-as-received (RT/AR) and hot-wet (H/W) conditions are described. A buckling test on a simply supported panel with prior moisture conditioning was performed under the H/W condition with an environmental test chamber mounted on the test rig to prevent any loss of moisture during the test at 100°C and ⩾85% relative humidity. Based on these results, a test procedure was developed to identify buckling loads through measurement of out-of-plane displacements rather than relying solely upon longitudinal strain measurement. This procedure facilitated estimation of the buckling load of a hygrothermally aged and simply supported CFC panel under the H/W condition, where accurate strain measurements were found to be rather difficult and at times unreliable. Experimental results were compared with those obtained from finite element analysis with regard to the buckling loads and corresponding mode shapes for the boundary conditions considered. The effect of rounding off the knife edges used for the simply supported condition was specifically studied during FEM analysis and was found to be significant.