Introducing Inertia with Rigid Groups

Altair Forum User

Hi,

 

I am trying to run a dynamic impact analysis problem with bouncing involved. In my model, I have added inertia specified in global skew frame to replicate the actual case. I want to reduce the computation time when the body is off the ground by introducing a rigid group consisting of all the nodes in the model. For the same, I am using a sensor of TYPE: interface. According to the RBODY definition, is SENS_ID is non-zero, the added inertia is ignored. Please confirm if this is true? If so, how do I ensure that the added inertia is incorporated in the model when I make SENS_ID non-zero.

 

Thanks and Regards,

Sai Phanindra

gopal_rathore

Hi Sai,

I have worked on similar problems where I have introduced a global rigid body when the object is in free motion. 

I have used RBODY ON and OFF option in the engine file. Thus by making two engine files , one for the rigid body motion where global rigid body is on and another for the impact simulation where the global rigid body is OFF.

Also for global rigid body use ICoG=2 option, this allows calculation of inertia and mass from the slave nodes. Also when we use RBODY ON in engine file, it automtically takes ICoG 2 for that particular rigid body.

Altair Forum User

Hi Sai,

As suggested by Gopal can you try with RBODY ON and OFF option. This option can lead to the activation or deactivation of elements entirely connected to a given rigid body

Altair Forum User

Hi Gopal,

 

Thanks for the info. It would be great if you could elaborate the following:

1. Suppose I add some external inertia to one of the rigid bodies which are deactivated during free-fly, What happens to the inertia? Is it also deactivated.

2. If that is the case, can I add external inertia to the global rigid body to compensate for the deactivated Inertia?

 

Thanks,

Sai

Altair Forum User

Hai Sai,

Can you please coordinate this issue with the Indian Support Team.

If you are commercial customer, please contact your local support.

You can contact your local support @ http://www.altairhyperworks.com/ClientCenterHWSupportProduct.aspx?_ga=1.248867588.26683431.1427861035

gopal_rathore

Hi,

In my case , I was making the global rigid body with CG node of th system to the rest of the system. This makes inertia of the system about the CG same, irrespective of we use global rigid body or not. What type of analysis is yours?

Altair Forum User

Hi,

 

I have added mass and Inertia at the CG of the system, which I want to retain with the global rigid body. My analysis involves impact of a body on hard ground and the objective is to predict the subsequent dynamics. For the same, I want to use a global rigid body to reduce the time taken during the bouncing phase.

gopal_rathore

Hi,

Have you checked the inertia values in the .out files of global rigid body. This will be obtained after running the starter 0000.rad file.

Compare this value with system having no global rigid body.

 

 

Altair Forum User

Hi,

 

I have checked and compared the .out files for the model with and without global rigid body. The model with global rigid body does not account for the added inertia.

gopal_rathore

Hi,

I think you can add that difference value of inertia to the global rigid body element itself by using card edit for the element so that it accounts for the complete inertia. Try this I think this might work. 

Altair Forum User

Hi,

 

I tried giving the additional mass and inertia to the global rigid body. However, during the RIGID BODY INITIALIZATION of the global rigid body (in _0000.out), the added mass and inertia are ignored and only the slave node mass is considered. Is there any workaround where I can add mass and inertia externally?

 

Also, I am facing another problem with global rigid body:

 

The model is initially off the ground, so the global rigid body is activated. Once it reaches the ground after free fall phase, the global rigid body is turned off and the computation is carried out as expected (with a lower time step and all elements active) for 30 ms. Beyond that, the error (which is -5% at this point) starts increasing and reaches +99% at which point the dynamics become nonphysical. i.e. the kinetic energy of the system increases to 10,000 J even though the initial total energy of the system is 460 J. Is there any reason for this sort of instability?

Altair Forum User

Hi Sai,

I think using  Rbody initialization  all the parameters related to it will be ignored. As a workaround, you can add mass externally using ADMAS which will assign additional non-structural mass to nodes or a group of nodes

Check the model for any incompatible kinematic conditions and also check for the time step, which can be due to high distortions or penetrations, that can result in high energy error.