The hydro-hammer sampler is a new type of sampler compared with traditional ones. An important part of this new offshore sampler is that the structure of the core cutter has a significant effect on penetration and core recovery. In our experiments,a commercial finite element code with a capability of simulating large-strain frictional contact be-tween two or more solid bodies is used to simulate the core cutter-soil interaction. The effects of the cutting edge shape,the diameter and the edge angle on penetration are analyzed by non-liner transient dynamic analysis using a finite ele-ment method (FEM). Simulation results show that the cutter shape clearly has an effect on the penetration and core re-covery. In addition,the penetration of the sampler increases with an increase in the inside diameter of the cutter,but de-creases with an increase in the cutting angle. Based on these analyses,an optimum structure of the core cutter is de-signed and tested in the north margin of the Dalian gulf. Experiment results show that the penetration rate is about 16.5 m/h in silty clay and 15.4 m/h in cohesive clay,while the recovery is 68% and 83.3% respectively. The hydro-hammer sampler is a new type of sampler compared with traditional ones. An important part of this new offshore sampler is that the structure of the core cutter has a significant effect on penetration and core recovery. In our experiments, a commercial finite element code with a capability of simulating large-strain frictional contact be-tween two or more solid bodies is used to simulate the core cutter-soil interaction. The effects of the cutting edge shape, the diameter and the edge angle on penetration are analyzed by non -liner transient dynamic analysis using a finite ele-ment method (FEM). Simulation results show that the cutter shape clearly has an effect on the penetration and core re-covery. In addition, the penetration of the sampler increases with an increase in the inside diameter of the cutter, but de-creases with an increase in the cutting angle. Based on these analyzes, an optimum structure of the core cutter is de-signed and tested in the north margin of the g ulf. Experiment results show that the penetration rate is about 16.5 m / h in silty clay and 15.4 m / h in cohesive clay, while the recovery is 68% and 83.3% respectively.