In the first year, based on the information obtained in the preliminary experimental study on the proto type 20kN MR damper, a new 40kN damper with the maximum stroke of ±100mm was designed, manufactured, tested and confirmed to be the one with satisfactory stable load-deflection relations which can be adjusted by applying minor electric current. By installing this damper into a small simple one-mass model to represent a base-isolated building system, original semi-active control algorithms by project participants were investigated by means of shake table tests.In the second year, a larger scale shake table test was conducted. The model consists of a 2/3-scale steel frame rested on one-way roller bearings at the foot of the corner columns. Four natural rubber bearings were used to adjust the natural period of the system to be about 3.0 seconds. Instead of a normal passive damper, the 40kN MR damper is installed to perform semi-active control. Five different control algorithms proposed by each project participants were tested. The aim of each algorithm, i.e., to get less acceleration without causing greater displacement than the case of normal passive control, was found to be very successful. It was also confirmed that most experimental results and analytical ones showed good agreement

In the first year, based on the information obtained in the preliminary experimental study on the proto type 20kN MR damper, a new 40kN damper with the maximum stroke of ±100mm was designed, manufactured, tested and confirmed to be the one with satisfactory stable load-deflection relations which can be adjusted by applying minor electric current. By installing this damper into a small simple one-mass model to represent a base-isolated building system, original semi-active control algorithms by project participants were investigated by means of shake table tests.In the second year, a larger scale shake table test was conducted. The model consists of a 2/3-scale steel frame rested on one-way roller bearings at the foot of the corner columns. Four natural rubber bearings were used to adjust the natural period of the system to be about 3.0 seconds. Instead of a normal passive damper, the 40kN MR damper is installed to perform semi-active control. Five different control algorithms proposed by each project participants were tested. The aim of each algorithm, i.e., to get less acceleration without causing greater displacement than the case of normal passive control, was found to be very successful. It was also confirmed that most experimental results and analytical ones showed good agreement