I am doing 3D elastic wave (with 1.5E6 Hz pulse) simulation in Abaqus . The dimension is 40x30x30. The material property is steel with young's modulus 205GPa and poison's ratio 0.28. I have assigned a velocity boundary condition with a tabular amplitude at the middle point of yz-plane at x=0 (shown in the figure below) for the wave pulse and except that I haven't assigned any other boundary conditions. 

I am doing a dynamic/explicit analysis. The total simulation time is 3.33E-5. I have assigned the field and history output at every 2E-7 sec time interval and assigned the output data for stress, velocity and displacement. The mesh size is 1.25E-4 which is in the recommended range for wave simulation. I did a double analysis precision with parallel computing. Now after running the simulation, I was trying to plot pressure vs distance along the center line of the domain. And what I found is there is some pressure amplitudes at x=0 even if the wave has propagated and travelled some distances shown in the picture. 

Then I applied the BC on an area of single element and still I have some amplitudes at x=0 even if the wave propagates far away from the point shown in the figure below. But shouldn't I expect a flat line at x=0?

My question is why I am seeing this and how can I remove it from x=0. Or if it is normal to have this then how can I validate this with theoretical knowledge? I have been struggling to solve this issue since last two months. Please help me or suggest me any guide way to solve it. Thanks in advance.

One thing I forgot to mention that I set the linear and quadratic bulk viscosity in the step module as 0.