Separation contact but act like bond contact

[yta10]

Hello

I used separation contact between two mating surfaces. and they are connected though bolted connection.

but they act like a  bonded contact.

Could anyone help me with this?

Thank you

[John_Holtz]

Hi @yta10 

I have a few questions:

1. What type of analysis are you performing? (Some analysis types cannot handle separation contact and convert it to bonded contact. But Linear Static and Nonlinear Static stress can handle separation contact.)
2. How are you expecting the results to be different? Or what result indicates that separation contact is not working? All I see is that the 3 bolts are holding the flange together, and that is what I would expect to see when using either bonded or separation contact. The image does not show to me that separation contact is not working. (There may be a very small gap opening at the tip of the toe, but the displacement would be much smaller than at the top of the channel and would not be visible in the image you provided.)

[yta10]

Thank you for your reply.

1) It's linear static mode

2) Numbers mean Contact types and locations

All separations 

1- 3EA of bolt washer locations to the orange surface.

2-bottom gray surface without 3 washer locations to the orange surface.

3 and 4 - fillet surface to the orange surface. 

As you can see, depending on the contact areas.. I got all the different results. I expected all the results would be the same. Especially, when I applied contacts (1,2,3,4).. it acted like boned contact.

What contact location is the proper contact to simulated separation?

[John_Holtz]

Hi @yta10 

Thanks for the images. I understand the problem now.

I do not know why it would behave like that. If you want me to look at it, please provide the Inventor files (the assembly and part files, .iam and .ipt, compressed and posted to the forum) for model "D".

Since you are performing a linear analysis which is based on infinitely small displacements, no nodes on the fillets (3 and 4) will contact the top of the orange piece. Therefore, separation contacts are not needed on the fillet. You only need to defined the separation contact between the flat face (or faces) of the gray part to the flat face (or faces) of the orange part.

[yta10]

Hello,

Thank you for you reply.

Please find attached for the contact problem.

And, I would like to ask you one more question. 

Contact : separation (without fillet area)

Test A : 1 Bolt

Test B : 3 Bolts

For Test A, I thought the bolt stress would be at least 1/3 of the TEST B.

But the stress on Test A is even less than Test B.

Would you please let me know why this happens?

Test A

Test B

[John_Holtz · Accepted solution]

Hi @yta10 

Contact

Thank you for the model. I am getting similar results when using version 2022. I have notified the developers.

My original suggestion of not including the rounds in the contact are one solution. Another solution is the following:

1. Change the analysis type to Nonlinear Static Stress.
2. On the "Analysis > Edit > Options" tab, set "Large Displacements" to "Off".
3. On the "Nonlinear Setup", set the "Number of increments" to 1.
4. Run the analysis.

Items 2 and 3 make the analysis very similar to the linear static analysis, so the analysis runtime is similar. 

Bolts

In a properly designed and assembled bolted connection, 90% of the stress in the bolt is due to the bolt preload. The load added to the components increases the bolt stress by a small percentage. These are the results that I get using nonlinear static in version 2022:

• 1 bolt: bolt stress = 8177 psi.
• 3 bolts: bolt stress =8336, 7712, and 8380 psi.