Skip to main content
May 15, 2014 1:49 AM UTC
Question

Re: Fastener with preload for 3 components

Hi Scott,

What are you attempting to analyze with a fastener through three components? Are you interested in just the fastener loads? Do you want stress results in the components?

    2 replies

    May 15, 2014 1:36 PM UTC

    Hi Steve,

    What I have is a bracket sandwiched between two bearing caps. The assembly holds two bearings, one on each side of the bracket. When the fasteners are preloaded, the two caps compress against the bracket.

    My main concern is the stress and deflection in the components. As far as the fasteners go, I'm only concerned that I have the correct preload.

    I'm currently attempting to use beams and rigid links to simulate the fasteners. As Shaun suggested above, I used curves for the beams. I then attached the ends of the beam to the bearing caps using rigid links. I can't use weighted links as I don't have the advanced simulate license. Finally, I added a -1° temp load to the beams for the preload. I also made sure to set my coefficient of thermal expansion.

    The issue I'm having is with constraining the bearing caps properly. I've created contact interfaces between both caps and the bracket, but I know that doesn't constrian the assembly sufficiently. I was able to run a simulation by changing one of the interfaces to "bonded" but this overconstrains one of the caps and is not what I want.

    So my main questions are: How do I constrain properly without overconstraining (use springs?) And how do I ensure I have the correct preload? Am I simply looking under the measures section of the summary at "max_beam_tensile"?

    Any help is greatly appreciated!

    May 15, 2014 5:47 PM UTC

    Could you post an image of your model? I'm not quite sure how your model looks, but if you expect very little rigid body motion (due to out of balance forces), then you can used a ground spring to stabilize the model. If there is any symmetry in the model, the this can be exploited as well.

    For the pre-load, you are correct; you can look at the max_beam_tensile measurement and determine the axial pre-load by dividing by the cross-sectional area of the beam.