Hello, I am trying to analyze if this machined part can be made out of aluminum vs steel. It will be proof loaded in the two scenarios shown. What would be the proper way to apply boundary conditions and loads for each scenario?
I have initially tried doing a static analysis with the bolt bearing faces fixed and a distributed load applied to the lug for scenarios A and B separately, but I am not sure if this accurately reflects the proof load setup. I tried to use the Prepomax tutorial for lug analysis to guide me as well, but felt like I needed extra guidance to get a reliable result.
I also looked at trying to analyze this by hand for verification by using the lug analysis methods in Bruhn’s “Analysis and Design of Flight Vehicle Structures", but I was unsure if was doing it correctly. Any suggestions for how to verify this with hand calcs is also appreciated.
While I did learn about Finite Element Analysis in college years ago, I am a bit rusty with engineering analysis and I am just starting to learn how to use this specific program. It seems very powerful and I am pretty thrilled to get better at using it. That being said, please speak in simple terms so I can understand, thank you for your help
In the new version of PrePoMax (v2.4.0), equation-based non-uniform load distributions can be defined and thus a proper bearing load can be applied using a formula: Bearing load distribution
But a uniform distributed load at a small area can be a sufficient initial approximation. You should just account for the 7.3 deg angle in the second case. Whether they should be analyzed separately or not depends on your requirements, operating conditions and engineering codes if any apply.
Fixed BCs can be ok if you aren’t interested in the bolt area and only want to analyze the lug ignoring the plate. You could replace them with compression-only constraint, but you would have to assume some stiffness.
Analytical calculations for lugs are quite common in literature ao it shouldn’t be a problem. At least basic hand calcs for a simplified setup should be used to make sure the numerical results make sense.
Mesh convergence study should be performed as well.
As you have draw the problem it doesn’t have physical sense. If the base is clamped and there is a hinge the upper piece will rotate up to the point where the load it is vertically aligned with the hinge.
Thank you all for the feedback. The non-uniform load distribution looks interesting to try out. A few follow-up questions:
If I wanted to also analyze the bolt area and the stress concentration at the fillet, would a compression only constraint give better results there? I especially want to analyze the stress concentration at the fillet in scenario B.
For scenario A, can I assume it is a symmetric lug for conservative hand calcs, or should I be doing it another way?
Thanks for pointing that out. Is this something I would do manually by gradually reducing mesh size and rerunning analysis to look for changes, or is there a function in prepomax that aids with this?
One more super basic question:
Under tools>settings>Calculix>parallelization>number of processors, what number should I use here? Should I use the number of cores that my computer’s processor has?
You make a good point. I think it is supposed to be restrained at that angle so it cannon rotate, though I am not sure how. This makes me wonder if a fixed constraint there would be adequate or if it would over constrain it.
You could try different levels of approximation. Maybe even model the bolts as simplified solids (just connected cylinders). O you can play with elastic supports (point spring or compression-only constraint).
Case A is just tension and you should consider a proper load carrying cross-section for F/A.
It has to be done manually.
Yes, physical cores.
Anything can be a beam if you are brave enough But for real, you just have to break it down to fundamental loads - tension/compression, bending, shear and torsion. They usually act in combination but their basic formulas are simple. The tricky part (apart from doing simple FBD and applying approximations) is to determine the proper cross-section, moment of inertia, arm of the force and so on. American books for mechanics of materials can help you develop this way of thinking because they feature many practical examples instead of just bars and beams like European literature.
It’s about Saint-Venant’s principle - even with inaccurate (e.g. overstiffening) BCs, results can be ok away from them.
In my opinion your device would only make sense if there is a second lifting LUG. That changes the reactions not only in the device but in the lifted body too. The base of the triangle will be in compression and each base in shear + tension. The action line passes through the pin and the Lug centers.
But for real, you just have to break it down to fundamental loads - tension/compression, bending, shear and torsion. They usually act in combination but their basic formulas are simple. The tricky part (apart from doing simple FBD and applying approximations) is to determine the proper cross-section, moment of inertia, arm of the force and so on. American books for mechanics of materials can help you develop this way of thinking because they feature many practical examples instead of just bars and beams like European literature.