metal part, STATIC ANALYSIS 1

[minny]

I'm sizing a fitting of civil aircraft, primary structure.

I've started with hand calculation, considering the fitting as a rigid body and I've compared the axial stress with Ftu and Fty (if compression) obtaining Reserve Factor Rf. Then comparing the acting shear with Fsu I obtained the Reserve Factor Rs. At the end the Margin of Safety combining the two effect.

Now I've to use an f.e.m. model. With a 2D model I simply replicate the same calculation, the fem help me to read the internal load, with a freebody.

I want to have the internal stress at each critical section from the fem.

If I use a 3D fem, I don't know which kind of stress I have to plot: Max Principal, Von Mises, Averange ...
There are same stress peaks that are no good for static amnalysis.

Please suggest me the right way, I've to finish the report ...

 

[rb1957]

sounds a little excessive ... hand calc, confirmed by 2D FEA, and you need to do 3D FEA as well. any ways, from the 3D model i'd use either principal stresses or von mises and compare with fty or fcy (or whatever you have for a compression allowable).

stress peak in the FEA ... typically these are part of the model ouput ;cause you're applying point loads and analyzing them as ideally elastic material (unless you're doing non-linear FEA). thus you can easily get unreasonably high stresses. thus should be highly localised and thus you should be able to reasonably say that these have no effective (localised insignificant yielding ...)

 

[EngineerMicah]

For fittings, it is hard to beat the analysis presented by Nui, in chapter 7.

FEA should always be taken with a lack of trust untill you can prove otherwise with model specific testing or a conservative analysis. The FAA does not accept FEA/FEM without model verification based on test results.

Micah Hamley, PE
Aeronautical Engineer & FAA DER
- Keep those old planes flying! -

 

[flash3780]

This is going in an aircraft? Sounds like a sound approach... it's hard to pull over on the nearest cloud if you get it wrong.

When you're looking at stresses, the von Mises stress best represents the relation to the yield point in many metals, so if you're dealing with a stainless steel part, compare the von Mises stress to the allowable stress.

What's the allowable stress? Glad you asked.

The allowable stress depends on the function of the part. In parts subjected to repeated loading, the allowable stress should be determined with a S-N diagram. In the absence of an S-N diagram, steels under 30-35 Rc generally have an endurance limit between 1/3 and 1/2 of the UTS. Be sure to consider stresses in fillets and other stress concentrations when doing fatigue analysis. Concentrated stresses need to be considered when parts are loaded cyclically.

If you're dealing with a part that isn't subject to repeated loading, then you can often neglect high stresses in fillets, around holes, etc. and look at only the bulk section stresses. Often these features will yield and strain harden. If the loading isn't cyclic, that's as far as things go. In this case, the bulk section yield stress (with a factor of safety) is often used to determine the adequacy of a part.