Clip for Cladding _ Interpreting FEA static stress analysis stress results

[Marko04]

Hi,

I am currently in process of designing a clip with a thermal break for attaching the cladding on to. The clip is suppose to handle a dead load of 90 lbs. The clip's body consists of 6063 T5 aluminum alloy and
the thermal break is Polyamide. The clip is 5 inches long. Since aluminum has a greater tensile strength than the polyamide strut, the critical factor in in this analysis was to see whether the bending stresses caused by the vertical acting load of 90lbs acting at a distance of 5 inches away from the fulcrum would cause the polyamide strut to deform. A safety factor of 1.7 was used in this analysis.

After viewing the results of the analysis ( looking at the Von Mises Stress chart and corresponding stress colors found in the model), I noticed some small "hot spot" areas that only appeared
on surface of the polyamide strut indicating stresses higher than polyamide's tensile strength. ( Please see the attached images). I am not sure how to interpret these results. Could these hot spot localized
higher than yield stresses be neglected and that the overall design is safe? I would assume that if these higher than yield stresses went through the whole thickness of the material that the design would then
not be safe, however, since these localized stresses appear only on surface of the polyamide strut could they be neglected and design deemed safe?

NOTE: The areas in red indicate stresses higher than yield

 

[DaveAtkins]

Why are you using a 1.7 factor on DL?

Does the clip resist WL as well?

13,000 psi does not seem very high for aluminum--am I missing something?

DaveAtkins

 

[Marko04]

Hi DaveAtkins,

Thanks for your response. I was told by a structural engineer to aim for a FOS of 1.7. He told me that the limit state design LRFD would be a better way to asses the structural integrity of the component but since i
use a FEA that doesnt have LRFD I asked him if I could use allowable stress method instead. That is when he suggested for me to aim for a FOS of 1.8 in the conceptual design phase.

The 13000 psi is the yield strength of the thermal break strut which made out of Polyamide so it is not Aluminum. Yes, the clip should also be able to resists the withdrawal forces imposed by the wind pressures uf up to 60lbs/sqft. But my main concern is the DL. In the stress analysis I performed, the red hot spot stress areas that appear on the polyamide thermal break indicate stresses that exceed polyamide's tensile strength of approx 13000 psi. These stresses appear only on the top surface of the polyamide break and don't penetrate deep into the material. I am little but unsure how to interpret this FEA graphical stress data in terms of whether or not the polyamide thermal break is able to safely resist 90 lbs of DL without undergoing deformation. Any feedback would be greatly appreciated

Thanks,

Marko

 

[andriver]

It might help if you show your mesh. I am not FEA expert but mesh size can give some more info. Also how are your joints connected from one element to an other (polyamide connected to clip etc). Are you connect along the entire length of the polyamide and clip or just at the ends?

 

[structSU10]

How is this 1.7 FoS applied? are you amplifying the load on the material? are you able to apply the actual load and have it plot a FoS contour on the assembly?

When you state the hot spots are over 13,000 psi, I assume that is with 1.7*90lb of dead load plus 1.7*WL? A different way to do it is apply they actual loads, and limit the stress plot to 13,000/1.7. It should give they same basic results.

Besides all that, is the polyamide material ductile or brittle? that will affect what type of stresses you want to consider. A von mises stress is better suited for the ductile type material, such as the aluminum. For a more brittle material, you would want to investigate the max/min principle stresses.

 

[TLHS]

The general behaviour of the material will also affect how you treat localized stress concentrations. A localized stress concentration that can redistribute to elsewhere in the structure may not be an issue in a ductile material when you're looking at non-cyclic ultimate limit states. In that case, your material can yield and your loads can redistribute (assuming you can prove or easily see that this redistribution capacity exists). In a brittle material, the same situation might result in cracking and failure.

If you decide that the loads are acceptable due to redistribution, you should also take a look at what the situation is under service (non-factored) loads. You shouldn't really be yielding in service limit states except in some very specific scenarios (maybe around connections and other locations where it's acceptable to locally go to ultimate rather than yield stress)

The other way to deal with this is to use a non-linear finite element model that will model yielding and load redistribution.