The determination of the rigid-body dynamic properties of structures is an important task to be performed in the field of structural modification or coupling analyses. In the system representat ion, a modal matr ix which incorporates up to six rigid-body modes may be used to describe the models pertaining to unconstrained structures. In analytical terms, these modes can be easily obtained, a fact that does not happen in the experimental route as, for each FRF measurement of a free/y suppotted structure, a// the rigid body information is condensed to one single value, i.e. to the inertia restraint or mass line value. In this work a method is presented to make use of those vakms, each one measured for a pair of response/excitation, in order to extract a// the ten values to fu//y characterize the rigid-body behaviour of any unconstrained structure. A geometrically simple object, a rectangular cross section beam was used to simulate the measurement responses in different points~and for several excitation conditions. These analytical tests were performed to determine the effects of the measurement and excitation locations on the accuracy of the rigid body properties obtained by using the proposed methodology.