surface contacts

[dog2k5]

How do you model the surfaces for bolts. do I create surfaces for the bolt shank and the inside of the plates where the shank makes contact? Do I model the surfaces between the nut and the surface it touches as well as the bolt head and the surface it touches? If so how do I do it? I have tried many things and I keep getting a warning in the message file that says:

"contact surfaces "assemeby_bolt head" and "assembly plate1" node
number 1157 instance part-3-1 is overclosed by 116.58 which is too
severe--cutback will result. You may want to change the value of
hcrit (5.19823E-02_ on the *contact pair option."

what does this mean and what is a reasonable value for hcrit? Thanks for the help.

 

[Yoman228]

DO you need to model the bolts in details? where the stress in the bolts and adjance area is the main intrested?
If not, I will not model the bolt and use coupling instead.

 

[dog2k5]

My main concern is modeling bolt slip. I have the experimental data and need to match the results of my model to it. So I would think the bolts would need to be modeled in detail unless couple can take into account when the bolts would slip. if there is a way around modeling the bolts in detail I am all ears. It would be much easier to not have to model the bolts in detail.

 

[gurmeet2003]

dog2k5,

Are you working with a 3-d model? If you are you can model the bolt in 3-d. Partition the bolt in the shank to create a cross-sectional surface for applying the bolt load. It appears that you model bolt, nut and some other parts between the bolt head and the nut. You will need a tie contact between the nut id and bolt threaded area. You will also need a contact between nut surface and one part of your assembly. You will need a contact between bolt head
and one part of your assembly. Since you are looking for slip, at least some of the contacts need to be sliding. If you make all the contacts sliding you may run into problems of rigid body motion. Therefore you may have to make a judicious choice between the sliding and tie contacts, so that the results you obtain are realistic. Since you have expermental data you should be able to validate your model.

Normally you can provide an artificial displacement in the first step to stablize the parts (eliminate rigid body motion). This displacement is deactivated in subsequent steps. However when bolt loads are applied they have to be applied in the first step. Therefore above can not be done in the first step.

The error message you indicated appears to be a result of some interference in your geometry. Check the dimensions of all the parts to make sure that therer is no interference between any of the parts.

Gurmeet