Peridynamic ves instability

Hello,

I have tried to simulate a tensile test on a cube using Peridynamic VES.

Dimension: 1 x 1 x 1

Lattice distance 0.1, 0.05, 0.02

Young modulus 1e9

S00 = 100 ( i used a very big condition for the damage to remove the damage from the simulation )

Lambda 0.99 to have a visco material

I used different tau from 0.01 to 1000

The load is applied in y axis and the deformation stopped at 1% elongation.
The force is monitored at the top layer of particles.

The idea of the simulation is to observe the effect of the tau on the decreasing of the stress with time.

What I observed is that:

When I use high tau value the energy increase and lead to instability of the simulation.
“ compute peratom all pe/atom
compute pe all reduce sum c_peratom”

I use smaller timestep, different loading but nothing works.

What remove the instability is the viscosity, but It could affect the final result.

Do you have any idea of the reason of it?

Kind regards