10 Replies Latest reply on Jan 3, 2014 11:27 AM by Jared Conway

Bolt Preload and Unusually high Axial Force

Alexander Lam Jan 2, 2014 7:51 PM

I'm currently trying to simulate the deformation of a prosthesis whilst connected to a support block with a force of 2500N. This involves connecting the bolts with a torque of 103Nm for the preload for an M12 bolt. But when i run the simulation to see if the bolts would hold or not, but the bolt force comes up with a ridiculously high number upwards of 40000N. I then compared it with a simple load of 1N and it still had a ridiculously high load. Is there anyway of finding the axial force without having to take into the preload?

Thanks,
Alex

Re: Bolt Preload and Unusually high Axial Force

Jared Conway Jan 2, 2014 8:02 PM (in response to Alexander Lam)

can you post the bolt connector definition?

if you make a simple block with just the bolt connector do you see the same behavior?
Like Show 0 Likes(0)

Actions
- Re: Bolt Preload and Unusually high Axial Force
  
  Alexander Lam Jan 2, 2014 8:24 PM (in response to Jared Conway)
  
  Similar thing happens with the simple case where the axial force is still around 30000, which suggests that all the torque comes from the torque but I'm currently using the recommended values for preload.
  
  Like Show 0 Likes(0)
  
  Actions
  - Re: Bolt Preload and Unusually high Axial Force
    
    Jared Conway Jan 2, 2014 8:55 PM (in response to Alexander Lam)
    
    Can you post the simplified version and a screenshot of your bolt definition? What friction factor are you using?
    
    Like Show 0 Likes(0)
    
    Actions
    - Re: Bolt Preload and Unusually high Axial Force
      
      Alexander Lam Jan 2, 2014 9:24 PM (in response to Jared Conway)
      
      The friction factor is 0.2
      Sorry the computer im using wont let me compress the folder so i cant actually post the zip with the results, but I can dropbox the link with it all inside if its not too much trouble.
      https://www.dropbox.com/sh/6euw88yeywb0gzc/xiyzJ7gJtl
      
      Like Show 0 Likes(0)
      
      Actions
      - Re: Bolt Preload and Unusually high Axial Force
        
        Jared Conway Jan 2, 2014 10:23 PM (in response to Alexander Lam)
        
        With those parameters, the axial force lines up with what is expected.
        
        http://help.solidworks.com/2012/English/SolidWorks/cworks/Connector_-_Bolt.htm?id=b15a148f1e0a48019f2703d74b96af60#Pg0
        
        Check out that article. Near the bottom it talks about preload calculation and how torque is converted to axial load.
        
        Like Show 0 Likes(0)
        
        Actions
        
        Re: Bolt Preload and Unusually high Axial Force
        
        Alexander Lam Jan 2, 2014 10:36 PM (in response to Jared Conway)
        
        Ok so i now understand that the torque generates a significant amount of axial force, but is there a way to determine the impact of the actual external loading on the bolt. Because I'm assuming that this impacts the factor of safety, as the bolt itself with just the preload already shows a pretty low factor of safety even though it should able to handle a much higher load.
        
        Like Show 0 Likes(0)
        
        Actions
        
        Re: Bolt Preload and Unusually high Axial Force
        
        Mike Pogue Jan 3, 2014 2:27 PM (in response to Alexander Lam)
        
        The tensile part of the external load adds to the pre load like this:
        
        Pi = Pe*kb/(km + kb)
        
        Pi = the external load taken by the bolt (N)
        Pe = the total external load (N)
        kb = the effective spring rate of the bolt (N/m)
        km = the effective spring rate of the joint (N/m, can determine with FEA by applying 1 newton to washer area and measuring displacement)
        
        The shear part of the load can be applied directly to the bolt
        
        The von Mises stress is
        
        s_vm = sqrt(St^2+3*Ss^2) < SF
        
        St = the tensile stress
        Ss = the shear stress
        SF = your safety factor
        
        NASA uses this failure criterium (TM 106943 widely available for download, see also NSTS 08307) which may be the same as the von Mises stress if you do some algebra:
        
        Rt^2 + Rs^3 < SF
        
        Rt = tensile load ratio
        Rs = shear load ratio
        
        All of this is doable in Simulation--you can extract the results from bolted connectors--but I understand there are some flaky things about it, which Jared is familiar with, so it would be very good to understand the methodology.
        
        PS: Don't use my equations, which I ripped off the top of my head. Look them up and read the texts. They are short and digestable.
        
        Like Show 1 Likes(1)
        
        Actions
        
        Re: Bolt Preload and Unusually high Axial Force
        
        Mike Pogue Jan 3, 2014 10:41 AM (in response to Mike Pogue)
        
        Also, you likely have a fatigue problem, not just a yield problem.
        
        Like Show 1 Likes(1)
        
        Actions
        
        Re: Bolt Preload and Unusually high Axial Force
        
        Jared Conway Jan 3, 2014 11:27 AM (in response to Alexander Lam)
        
        just to add to what roland and mike said
        you know that the bolt axial force is X
        now you're going to run an additional loading case that will make the axial force X + Y
        once you implement roland and mike's suggestions about knowing better what the actual axial force is, you'll be able to determine the bolt's condition
        
        Like Show 0 Likes(0)
        
        Actions
Re: Bolt Preload and Unusually high Axial Force

Roland Schwarz Jan 3, 2014 10:33 AM (in response to Alexander Lam)

In reality, using torque to determine preload is fuzzy math at best. Torque to tighten bolt is affected by friction from many different quarters, and getting an exact preload value is nearly impossible.

Preload values calculated from torque in a frictionless (e.g. CAD) environment are going to be higher than what one sees in reality.

If you're really interested in controlling preload force, you should match the actual force and not hope that the torque-to-preload resulting from some random default equation coincidentally matches what is happening with your real parts.
Like Show 2 Likes(2)

Actions

Go to original post