The proposed paper focuses on the issues to deal with when approaching experimental analysis of structures equipped with semi-active control systems. It starts from practical experience authors gained in a recent wide campaign on a large scale steel frame structure provided with a semi-active control system based on the adoption of two magnetorheological (MR) dampers. The latter are special devices able to achieve a wide range of physical behaviors using low-power electrical currents. Changing the current in the damper causes a very fast modification of the mechanical properties of the MR fluids, due to the particular magnetic field applied. In the above tests, four different control logics was adopted to drive the two MR dampers, each one being “co-located”, i.e. they don’t need the full state measured on the whole controlled structure but the actual values of only some quantities (displacements, velocities and forces) measured in the close surroundings of the dampers.

This kind of experimental activity requires special attention in solving several specific aspects that characterize each of the three phases - input, processing and command - performed in each test, most of them being unusual for any other kind of structural testing. This work highlights all these aspect, describes how they were dealt with during the above cited activity and shows undesired and unavoidable effects (e.g. delays of the control chain, leading to a loss of efficiency of the control system; cases of malfunction of the control algorithms) related to some of them and the way to reduce their incidence.