Article on Significance of Inacti=6 in Type 7 Interface in RADIOSS

As the title suggests, this article is going to deal with the significance of Inacti=6 and what that particularly does to the model. We already know, there are two important contacts: type 7 and type 11. Type 7 is the node to surface contact and type 11 is the surface to surface contact. The recommended properties for both the type 7 and 11 are the same. In the recommended properties, Inacti is something like a silent guardian. Inacti is setting the gap as variable with time. Particularly, Inacti=6 means that the gap is going to be varied and if there are any initial penetrations then it will be de-penetrated (moving the nodes move one another)

Now you may ask me, I am seeing the model and I don’t find any penetrations or intersections here, how Inacti=6 will find that penetration/intersection? Penetration is just the nodes being inside the master element gap of the contacts that we have defined. To take the contact related penetrations, inacti=6 takes things into its hands. To understand much better and see the power of inacti, we are going to do an experiment and understand that. This is not something a new model, but with the same challenge 5 crush box model. The crush box model has 4 components and 2 kinds of thickness assigned to it as shown below,

We are running the model now for two cases. Case 1 is going to be recommended properties with inacti=0 and case 2 is going to be recommended properties with inacti=6 as shown below,

We have already studied that the slave must be out of the master segment gap. If it is within that gap then, penetrations occur. Now let's calculate whether there is penetration or not. We have given the Igap=2 which is nothing but the variable gap calculation as mentioned below,

Where,

gm: master element gap I.e tm/2 where tm is the thickness of the master.

gs: slave node gap I.e ts/2 where ts is the thickness of the slave.

Now let’s calculate this Igap for our components, we have pshell_3 and pshell_2. The GAPmin values are given as 1 for this model.

For pshell_3, it is calculated that,

Max{GAPmin, min[Fscalegap . ((3/2) + (3/2)), GAPmax]}= max{1,3}= 3 for pshell_3

For pshell_2, it is calculated that,

Max{GAPmin, min[Fscalegap . ((2/2) + (2/2)), GAPmax]}= max{1,2}= 2 for pshell_2

So we have got the Igap values as 3 and 2 for the pshell_3 and pshell_2 respectively. This penetration can happen between the overlapping of components. Let’s find its distance, as shown below,

The distance between the overlapping components is 2.1 mm. Now when it is compared to the overlapping distance, pshell_2 is safe as 2 mm < 2.1 mm. For the pshell_3, it is 3 mm > 2 mm. The nodes and elements present in the 3 mm thickness component will face some issues with the penetrations. We can even get a much clearer idea by running the penetration check. The penetration check is carried for the contact that is defined. We get around 300+ penetrations as shown below as in the image,

As mentioned earlier the 3 mm shell thickness components are going to have some penetrations in the model. This is what is called the initial penetrations in the model due to the created contacts. It is not over yet, lets run the model for the mentioned cases.

The model is simulated for both the inacti=0 and inacti=6. Shown below is the energy error obtained from case 1,

The energy error in case 1 is around 99% because of the initial penetrations present in the model. The presence of this penetration provided a high energy. As mentioned earlier, penetration creates a positive energy error in the model. The whole energy of the model is going towards the removal of this penetration in the model. This penetration produced in the model is one of the major reasons for 99.9% energy error which makes the simulation pretty unstable and you get weird animations too.

Now when we look into case 2, we can see that inacti=6 flag creates repulsive forces, which will de-penetrate the nodes from the gap. The energy error in case 2 is around -5.7% which is under accepted levels.

As shown above, even before the original impact happens, the penetration and its relations does create the highest amount of contact forces. The contact forces occur only in the above mentioned 3 mm components. The below image displays the contact forces in case 2. You can see that the inacti removed the penetrations in the model and the contact forces were created only due to the contact of the components in other ways.

Let’s plot the internal energy and contact energy for the two cases. The contact energy skyrocketed even before the impact started to happen. Because of this reason, the internal energy also raises up immediately and then increases slowly due to the impact in the model.

Down below is the case 2, the internal energy and the contact energy in the model is very less and close to reality. The contact energy is also less, due to the initial de-penetration by inacti flag well before the impact, the contact energy is very less and it is even less than 5000. The negative energy error is due to the sliding of the nodes. In case 2, the penetration is removed by inacti, the removal is nothing but sliding of the nodes from the model. This sliding leads to negative energy error in the model.

To conclude, the initial penetration in the model makes the model weak and gives out undesirable and unrealistic results. These kinds of issues are very much simulation setup based and it does not happen in real life. To avoid these kinds of issues, the model setup must be proper. The initial penetration in the model, weakens the overall model. To avoid this kind of errors and unrealistic results, the model must be first checked for penetrations and intersections. Both kinds of component wise and contact wise penetration and intersections must be checked and it must be removed. From this experiment, we can see the significance of Inacti=6.