In the GeoApp it is either possible to directly set a uniaxial tensile experiment, a combined compressive/tensile experiment, or to use the current settings of ElastoDict Deformations.

The tensile experiment is chosen by selecting Tension in the dropdown menu. The experiment is defined by the maximal tensile strain (in percentage with respect to the original geometry) and the number of steps to apply this strain.

For the Compression and Tension experiment, a compression is applied to the geometry, followed by a tension. Again, both deformations are defined by the maximal strain (in percentage with respect to the original geometry) and the number of steps for each deformation.

For both experiments the Load Direction can be chosen (X, Y or Z).

To keep the accurate volume fraction information for each material ID and update these in the memory, the Keep gray value image checkbox can be chosen. This option needs more memory, but leads to more accurate geometry deformation if many (small steps) are computed. If this option is not chosen the volume fractions are segmented into a structure in each deformation step and the volume fraction information is lost.

The Resampling options are explained under ElastoDict Deformation - Geometry Handling in more detail.

If CurrentElastoDictSettings is chosen, the settings, which were defined in the ElastoDict Deformations dialog, are taken over and used with the GeoApp. No additional adjustments can be made in this GeoApp dialog.

If the option Allow Contact Detachment is checked, a detachment between different objects during the simulation run is possible.

Detachment is detected as soon as the stress changes from compression to tension in the contact area. If Decompose Stress Tensor is chosen, the stress tensor is decomposed internally into a compression and tension part. With the Contact Penalty Factor the detachment can be controlled. The higher the penalty factor, the later the detachment. As this option is still work in progress and the adjustment of the penalty factor highly dependent on the structure settings, it is recommended to leave this option unchecked.

By choosing Detach Initial Contacts, a detachment is allowed for all objects, especially also for those, which are attached to each other initially. Otherwise, only objects can be detached, which were detached initially but get attached to each other during the simulation.

For this simulation, information about the objects in the structure is required, either in the form of GAD objects (choose Object Mode - UseGad) or via a G32 file, which can be loaded by choosing Object Mode - LoadFile.

Since GeoDict 2026 it is possible to run the simulation with Adaptive Time Steps.  

In the Edit dialog you can define the Initial Time Step, with which the solver starts the simulation. The number of iterations from the first step is kept as reference value. For all subsequent simulation steps the number of iterations is compared to this reference number of itereations: if it is much smaller, the time step size is increased. Is it much larger, the time step is discarded and restarted with a smaller time step size. As soon as the minimal time step size is reached, the solver finishes the step and proceeds with the next step.

The time step size is always kept within the predefined bounds: it will never exceed the predefined Maximal Time Step and never fall below the Minimal Time Step.

Since GeoDict 2026 you can define in the Edit dialog of the Contact Detachment GeoApp to check if a solid path through the structure exists before running the simulation.

If Analyze Geometry is checked, GeoDict checks if a solid load path exists before starting the simulation. This ensures that the geometry is properly connected, which is usually intended.

For special cases, it is possible to disable this check (by unchecking Analyze Geometry).

After closing the Contact Detachment Parameter dialog by clicking on OK, the Contact Detachment simulation is started by clicking Run. The results are stored in the result file previously entered in Result File Name (*.gdr) of the Contact Detachment Parameters dialog.

After a successful run, the Result Viewer opens automatically at the Results - Report subtab.

In this example, a Tension experiment is simulated in which two objects were pulled in the Z-direction resulting in an object detachment. In the Report tab of the Result Viewer, the strain, which the objects undergo, is documented in the first table. As the Uniaxial Experiment – Tensile with a maximum of 55% strain was defined in the Macroscopic Load tab, the values for the Average strain for the column ZZ start with 0% and end in 55% after 55 steps.

In the table below, the Average stress is displayed.

The figure in the Plots subtab of the Results tab shows the tensile stress plotted over the tensile strain (which is the graphical representation of the table content of the Report tab). The tensile stress has its maximum for 11% strain and becomes zero at about 51% strain. At this point the separation of the two objects begins.

In the Strain/Stress Visualization tab, the different result files can be loaded according to the description for ElastoDict Deformations.

For this type of experiment, the Box Structure Rendering might be the better choice (as in the pictures below) since the detachment can be seen more precisely.

Of course, the detachment process can be visualized in a video using options found under the Create Videos tab.