8 Replies Latest reply on Nov 25, 2013 12:01 PM by Tony Hepp

Static study of a threaded connection in bending and tension

Tony Hepp Nov 21, 2013 5:27 PM

I'm attempting to conduct a static load study on an assembly comprised of five parts. There are three different local contact types: no penetration, bonded, and shrink fit. The shrink fit contact relates the male/pin threads of one part to the female/box threads of the mating part. The thread is a 3.63" stub acme. Please see the image provided. The left end of the assembly is fixed, while a tensile and bending force is applied to the other end. Having removed 7/8 of the assembly via a cut in the model, I've applied a symmetry restraint to the affected part surfaces.

My question: Is the shrink fit contact type appropriate?

Simplified image of assy.PNG 6.5 KB

Re: Static study of a threaded connection in bending and tension

Mike Pogue Nov 21, 2013 6:09 PM (in response to Tony Hepp)

I'm not sure this directly answers your question, but I am skeptical you can get accurate thread stress from an analysis of this type.

Also, I like you use of symmetry, but you could go even further and use an axisymmetric 2d model.
Like Show 1 Likes(1)

Actions
- Re: Static study of a threaded connection in bending and tension
  
  Tony Hepp Nov 22, 2013 9:50 AM (in response to Mike Pogue)
  
  Hi Mike. My intention is not to determine the thread stress for each part only has cosmetic threads. I'm certain the threads are strong enough. I've never performed FEA using a 2D model. I'll check SW Help and some other sources for info. on this.
  
  Like Show 0 Likes(0)
  
  Actions
  - Re: Static study of a threaded connection in bending and tension
    
    Tony Hepp Nov 22, 2013 10:10 AM (in response to Tony Hepp)
    
    Given the two loads and fixture constraint on my assembly, I don't think the axisymmetric option is applicable. SW says this can be used "when the geometry, material properties, loads, fixtures, and contact conditions have a rotational symmetry about an axis." Agree?
    
    Like Show 0 Likes(0)
    
    Actions
    - Re: Static study of a threaded connection in bending and tension
      
      Mike Pogue Nov 22, 2013 10:58 AM (in response to Tony Hepp)
      
      Good point. But in that case, I'm curious how you sectioned the model. It should suffer the same problem, no?
      
      You might just use bonded contact at the threads. It should be the fastest to solve, it should accurately represent the macroscopic behavior of the screw joint, and the solution far from the threads is not likely to be very sensitive to this decision.
      
      Like Show 1 Likes(1)
      
      Actions
Re: Static study of a threaded connection in bending and tension

Jared Conway Nov 21, 2013 10:31 PM (in response to Tony Hepp)

a quick paint image where you hightlight bonded faces, no pen and the shrink fit would be helpful here

i'm not sure what the shrink fit is for?

we did an analysis like this recently and got decent results but nonlinear is a better fit
Like Show 0 Likes(0)

Actions
- Re: Static study of a threaded connection in bending and tension
  
  Tony Hepp Nov 22, 2013 4:42 PM (in response to Jared Conway)
  
  Hi Jared. Check out the sweet artwork. I chose the shrink fit contact type between the green part and blue part because there is a finite volume of interference between them. The SW Simulation manual that I have seems to recommend this local contact for this scenario. Do you agree with Mike that bonded is better? I anticipate little deflection through the assembly and the material strain should behave linearly.
  
  Like Show 0 Likes(0)
  
  Actions
  - Re: Static study of a threaded connection in bending and tension
    
    Jared Conway Nov 22, 2013 7:31 PM (in response to Tony Hepp)
    
    is bonded better? for starting your analysis, it is probably easier
    is it "right" > its not as close to the physical case
    
    i think you have to take a step back and figure out what you want to learn here. this is going to be a long run time analyiss and it is going to take some time to get setup and to run right.
    
    for example if you just had green and blue and did the shrink fit with no other loads, that is going to be challenging but that is where i would start. with linear analysis if you try and add some other loads to it, it is going to get upset because of all the loads (contacts and shrinkfit are loads in a way) happening at the same time. that is where nonlinear could help.
    
    2d analysis would be ideal, but if you need to bendd this guy, that isn't going to work. what about half symmetry at least? that will add some stability to the problem.
    
    if it was me, i'd step through this analysis component by component and learn along the way. or bond everything and add the contacts one at a time until you complete the simulation.
    
    Like Show 1 Likes(1)
    
    Actions
    - Re: Static study of a threaded connection in bending and tension
      
      Tony Hepp Nov 25, 2013 12:01 PM (in response to Jared Conway)
      
      I got the bonded contact setup to solve and the results are similar to those of a colleague who analyzed the same design via ANSYS.
      
      I will do as you suggested and share the results.
      
      Yes; bending is an input. I did apply a symmetry constraint for modeling only half the parts. ttyl
      
      Like Show 1 Likes(1)
      
      Actions

Go to original post