perpendicular shells 9

[chaboche]

Hi,

I have four shell elements connected perpendicular to one another (in a cross). One face of the cross is subject to a bending load, the other is fixed at one end. A stress concentration is calculated at the interface. Is this a valid stress? It seems very high.

Any comments are welcome.

 

[chaboche]

here is the plot again

 

[chaboche]

I can't see either of the plots I just posted. I shall try resaving it and post it once more below:

 

[cbrn]

Hi,
if in the submodel the fillets are all modeled "as built" and their mesh is appropriately fine (now you don't have non-tangent faces sharing an edge, so you won't get the "mesh paradox"), meaning at least three or four parabolic elements along the fillet radius, then I personally would be extremely cautious as to disregard the peaks as you could do in presence of a sharp edge.
Try clipping the stress results, showing only the portion of geometry whose stresses are higher than 0.9*yield. If these portions are "small" with respect to the net section, and the average stress in the net section is below the correct limits (which are these limits depend upon the rules you are following, be it internal rule, ASME, or whatsoever), then I think you can accept them. Otherwise not.
Regards

 

[rb1957]

i see a couple of potential trouble spots in the pix ...

1) you've cut-off the maximum stress plotted to 300MPa. this gives a misleading view on how much is over 300 MPa ... in the upper left solid pic, everything inside the red perimeter is over 300 MPa (it plots as black, which doesn't look so bad !?) ... that means that pretty much the entire surface has yielded. your section shows that this is localised to the surface, which is good news statically. if fatigue is a concern, i'd shot-peen the fillet.

2) the web supporting the yielded flange has a nasty amount of yielded material, but again more a fatigue concern than a static one (i see lots of material that this overload could be redistributed into wihtout changing the global loadpaths).

3) why do you think the solid model "lights up" on the "innocent" corner (what looks to be a free corner ?

4) i have a little trouble seeing how this detail model fits into your overall structure ... the large "wheel" ?