Jump to content
Sign in to follow this  
Roy Duan

Large rigid wall force problem

Recommended Posts

Hi all,

I'm doing a dropping simulation and I find the rigid wall force is strange. So I design a simple dropping model for checking. The rigid wall reaction force is also so strange similar to my problem.

In the early stage of dropping contact, the rigid wall reaction force is very large, I suspect this large force result. Is it real or due to the rigid wall formulation? Or how to solve this problem? At beginning, the distance between the Box model and the rigid wall is about 4mm, which means there is no initial penetration I think. 

I have attached my model. The model is very small, it can be solved in minutes. Please help me find out the reason. Thank you.

2017-10-10_204543.png.ce92a3f44c25fc0ab99a8a6db75e64ba.png

 

Roy

FaceRigidWallPenetration_Example_0000.rad

FaceRigidWallPenetration_Example_0001.rad

FaceRigidWallPenetration_ExampleT01

Share this post


Link to post
Share on other sites
Guest

Hi Roy,

I think this is fine. As the box impact with the rigid wall, with the mass there could be high reaction forces in the beginning.

Share this post


Link to post
Share on other sites
On 2017/10/11 at 1:04 PM, George P Johnson said:

Hi Roy,

I think this is fine. As the box impact with the rigid wall, with the mass there could be high reaction forces in the beginning.

Hi George,

What's your opinion? Do you think this initial large force is a numerical problem or a real contact force in physical? 

In my opinion, I think the contact force should increase from zero to maximum and then decrease to zero corresponding to the real dropping process. So I think the initial large force may be coming from rigidwall numerical problem. Is my understanding right or how to explain this strange phenomenon?

 

Thank you

Roy

Share this post


Link to post
Share on other sites
2 hours ago, gopal_rathore said:

Hi,

Is this force coming at t=0 itself or after some time? Also can you check with impact force analytical calculation using mass, velocity and time of impact.

Hi gopal_rathore,

1. I think the initial large force occurred when the box and the rigid wall began contact. My model is very small, it only takes minutes to solve it, and I have uploaded my model in my first post. Could you run the model and help me debug the problem?

2. What's your meaning impact force analytical calculation? Is there any simple model with theoretical solution?

 

Thank you

Roy

Share this post


Link to post
Share on other sites
Guest

Hi Roy,

I think this is due to the initial contact happening.

Can you please remove the added mass in the model, run the model and review the contact force.

Share this post


Link to post
Share on other sites

Hi,

The issue is due to no proper distribution of mass. The added mass although distributed is taking cg at its visualisation node and thus making overall cg almost at the visualisation node and thus on impact with almost mass near the wall , the forces are coming higher.

I have not run after modification, but this seems to be the problem as the expected cg of the system is not the same as in model. 

Share this post


Link to post
Share on other sites
55 minutes ago, George P Johnson said:

Hi Roy,

I think this is due to the initial contact happening.

Can you please remove the added mass in the model, run the model and review the contact force.

Hi George,

Do you also think this large force due to numerical problem of rigidwall?

I delete the added mass and change the material to steel, and the force is similar (there is also a large contact force in the beginning of contact). I have uploaded the new model. Please try run the model and debug the problem. The total cpu time to solve the model is less than one minute.

2017-10-13_155748.png.79e4f3474f7f18d3a885cf31ac043eef.png

 

Thank you

Roy

 

FaceRigidWallPenetration_Example2_0000.rad

FaceRigidWallPenetration_Example2_0001.rad

Share this post


Link to post
Share on other sites
54 minutes ago, gopal_rathore said:

Hi,

The issue is due to no proper distribution of mass. The added mass although distributed is taking cg at its visualisation node and thus making overall cg almost at the visualisation node and thus on impact with almost mass near the wall , the forces are coming higher.

I have not run after modification, but this seems to be the problem as the expected cg of the system is not the same as in model. 

Hi gopal_rathore,

What's your opinion about this large force? Numerical problem or Real physical phenomenon? I create a new model with no added mass. The contact force result is similar.

Please see attached file.

2017-10-13_155748.png.2651132c736a34f11c8e15ff98c1826d.png

 

Roy

FaceRigidWallPenetration_Example2_0000.rad

FaceRigidWallPenetration_Example2_0001.rad

Share this post


Link to post
Share on other sites
Guest

Hi Roy,

I don't there is any issue with the numerical formulation of rigid wall. Instead of defining a RWALL you can try creating a surface and define an interface between the box and surface, run the simulatiton and review the resultant normal force.

I also tried a simple ball drop  on a rigid wall and the plot looks similar to this.

Share this post


Link to post
Share on other sites
2 hours ago, George P Johnson said:

Hi Roy,

I don't there is any issue with the numerical formulation of rigid wall. Instead of defining a RWALL you can try creating a surface and define an interface between the box and surface, run the simulatiton and review the resultant normal force.

I also tried a simple ball drop  on a rigid wall and the plot looks similar to this.

Hi George,

I will try with a rigid surface you suggested. My main concern is that this large force is real or not. That is to say, if we can get the contact force from a drop test, does it also have this phenomenon? I'm very doubted about this.

 

Roy

Share this post


Link to post
Share on other sites
Guest

Hi Roy,

This is an initial spike happened which you can ignore. You can consider the curve from the next value.

If this was a case like rebounding the curve would have been different.

 

 

rwall.PNG

Share this post


Link to post
Share on other sites
Guest

Hi Roy,

The initial spike was trimmed out from the curve. Normally in these type of cases the initial spike is ignored.

Share this post


Link to post
Share on other sites
1 hour ago, George P Johnson said:

Hi Roy,

The initial spike was trimmed out from the curve. Normally in these type of cases the initial spike is ignored.

Hi George,

How to trim out the initial spike in HyperGraph? Could you show me some details? Thank you

 

Roy

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
You are posting as a guest. If you have an account, please sign in.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...