In the last decades many strategies for seismic vulnerability mitigation of structures have been developed through analytical studies and experimental tests. Among these, energy dissipation by external devices assumes a great relevance for the relative design simplicity, even if applied to complex structures, and the effectiveness in reducing seismic demand. In particular, the use of fluid viscous dampers represents a very attractive solution because of their velocity-dependent behaviour and relatively low costs. The application on structures requires specific study under seismic excitation and a particular care of structural details. In this context the use of proper constitutive models for the dampers assumes a fundamental role. Seismic codes, as well as literature models generally provide a velocity-dependent relationship with two characteristic constants defining the shape of the force-displacement curve. In this paper an experimental campaign aimed at the determination of the damper constants in the range of low velocities is presented. A fluid viscous damper is tested with different ramp velocity functions. Results show that the assumption of velocity independence of the damper constants fails for the case of low velocity, giving rise to new issues about the limit of validity of the constitutive models generally adopted for viscous dampers.
