Extruded Aluminum Fascia Bracket Design

[JCS613]

Hello all,

I'm tasked to design the section of an extruded 6061-T6 aluminum side mounting bracket which will accept a 2" square aluminum tube or bar post of a guard railing. The post will be thru-bolted to the bracket using up to two 3/4" diameter stainless steel threaded rods. I do New York City Building Code guardrail calculations almost daily and know what load/moment it will need to withstand but want to ensure all the wall thicknesses & dimensions of the extrusion are adequate to withstand the design load. I've attached something I found on CR Laurence's website for a very similar bracket. To me the attached document looks like it focuses mostly on the anchors and is not very thorough in regards to the bracket itself. A few months back I found Autodesk Fusion 360 which does FEM analysis and will show stresses. The issue I ran into with the model is no matter what there will always be some mesh tetrahedrons that are over an allowable stress of about 21,000 psi. I'm not sure if I can use the FEM model in a report since it will always have some finite elements that are over stressed. I'm using ASD design method and the maximum moment the bracket is to be designed for is 37,895 in*lb & 1,763 lb of shear. (Loads perpendicular to plane of wall/surface)

Can anyone advise on the proper calculations to run in order to adequately size the section?

Thanks!

 

[thaidavid40]

I am surprised to that, given some of the leg length-to-thickness ratios shown in your model picture, you are able to use the 21ksi allowable stress limit. If you are getting some FEA meshes over that limit, wouldn't that be a good indication that some elements are maybe too thin? You might have to have some elements thicker than others. Of course, that brings in the possibility of differential cooling after extrusion. These sorts of extrusions are usually a trial-and-error development project. I would be surprised if you could go straight from theoretical design to successful production in one smooth step. Are you the aluminum producer, or are you working for a third party? I would have thought that the aluminum producer would have some experienced production people able to assist you with some experience-based recommendations.
Dave

Thaidavid

 

[Ron]

Consider designing a casting rather than an extrusion. Most of the anchor brackets I deal with are castings, not extrusions.

 

[JCS613]

@thaidavid40
I've played with the thicknesses and even lowered the loading all the way down but it seems no matter what I get small areas around where the constraints are, see this image:

In my past FEM modeling experience I know unrealistic singularities can be ignored for certain cases but this is my first time using Fusion 360.
Also in regards to the allowable stress limit, can you please advise on what your thoughts are for this item?
I work for the railing manufacturer and one of the engineers they use. They have aluminum mill's create new extrusion dies often, but are usually driven by some architect's design. This time it is an internal project for improving their standard mounting which is bolted angles or a welded tee. As it is to be designed to withstand guardrail loads they reached out to us, their structural engineer to design the section.

@Ron
We are set on using an extrusion, casting won't be the way to go for this. But what I am asking is simply what are the calculations to check the shapes for a moment and shear?

Again, I'm asking the forum for what calculations to run to check this bracket if it will withstand the shear, moment & tension loads imparted by the railing post.

Thanks.

 

[KootK]

This kind of thing isn't really my wheelhouse but I do feel that it would be acceptable to ignore some of your localized overstresses. It is a fairly ductile material after all.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[dhengr]

JCS613:
Everyone hears FEA and thinks, Ah, that must be the final answer, the ultimate method and the ultimate answer. Who could possibly argue with or doubt that answer. Then you spend the next seven months explaining the errors of their ways, and the limitations of the FEA methods. Use the FEA to give you some comfort and confidence in your long hand analysis, and don’t even show the FEA results in your report. At most, say “a quick FEA confirms these results” in your conclusions. Your image with the high stress areas shown, is not at all surprising, and 6, 7 - 8ksi doesn’t sound too outlandish though. Those are exactly the locations I would have expected to see the higher stresses, and they are perfectly/easily explainable. This could lead to an explanation of the FEA output is they insist on see those results too.

The two holes in the post & bracket connection will have high bearing stresses with some tension component due to the post cantilever action. They are fairly thin so high bearing stress is anticipated. It consists of a combined force vector due to vert. load and the horiz. reactions which counteract the horiz. load on the handrail (canti. loading). That’s a [(200 - 250lbs., at the handrail)(42" + 6 to 10" to the center of the bolt reaction on the bracket)]/[6 or 8" bolt spacing], 8 or 10 fold increase in the horiz. reaction load, in bearing. There will undoubtedly be some yielding to bring the connection detail into full bearing, and that’s a good part of what you are seeing in the FEA output image. The same analysis logic and FBD (free body diag.) carries back to the four bolts in the legs of the angle shape. You will have shear and tension around the A.B. head, along with bearing of a thin section on the bolt, and then some prying around the bolt head. Finally, I would have them grind a small radius on the inside at the intersection of the two legs of the angle shape, four locations, where you see the high stress running vertically in your image. Any cracking, tearing, stress raisers left from the length cutting operation in this area could be problematic. This area on the bracket has a high tension stress field right at that cut corner/edge in the angle radius.

 

[JCS613]

@dhengr

Thank you for your response, what I'm looking for is what you said by "long hand analysis". I need some design examples of doing long hand analysis on this type of bracket. Can you please provide me with some resources? Thanks.

As for the grinding of the radius, do you mean in this spot?:

I would be able to make that a bigger radius in cross section but since it is an extrusion that will be cut perpendicular to its length, it isn't feasible for the fabricator to do the additional labor to radius that cut edge 4 times per bracket at each post on a job.

 

[SAIL3]

for a rough hand calc I would do the following....I would ignore the front and back plates in the socket as they are not contributing anything meaningful to the resistance of the bracket...so then you are left with two angles to analyze...check out one angle with 1/2 the shear and 1/2 the moment....
if one is going to use allot of these brkts, then I would suggest that they be tested to confirm final design.....the problem with this is one would have to set up the extrusion process for that final design before it was confirmed by testing.....
it might be possible to weld up a carbon steel bracket of equal pl thickness and test it by increasing the loads by the ratio Fy steel/Fy aluminum and also E stl/E al....just thinking out loud here which on further thought may be a non-starter....

 

[dhengr]

JCS613:
You would do well to get out your Engineering Mechanics and Strength of Materials textbooks and do a little studying, or take a couple of these types of engineering courses. You should also be asking these types of questions of your boss, not us here at E-Tips. He should know what you do and don’t know, so he can give you the proper guidance and keep you and the company out of trouble. This is pretty basic Structural Engineering, and you claim to be one, so you shouldn’t need a worked out example which you just plug your numbers into. That’s not the way real engineering is done, and it sure may not be safe for the general public either. I’ve done hundreds of things like this, but not this exact problem. So I would have to do your calcs. for you to show you something, show you how to do them, and that’s not what E-Tips is all about. Talk to your boss, and get some in-office help, so someone knowledgeable can watch over your work, and help you learn. It sounds like you are not qualified to be doing this analysis, so you do need your boss’ guidance. And, he shouldn’t be shoving this off on you, if you and/or he aren’t qualified to do it. Draw some FBD’s and start the calcs., and come back here with some specific questions, not ‘give me a set of calcs. for this problem.’

That is the right location to do the grinding, (4 locations). Grind a small radius on the sharp, cut edge, the sharp edge/corner btwn. the horiz. cut end of the extrusion and the vert. plane of the angle shaped legs. Fair this corner/sharp edge radius out onto the two legs some distance, just as part of the cleanup operation after the length cut. It’s shop work, by the hundreds, to make a good product. If you can’t afford a couple minutes to clean up those edges, then live with the potential high stress or cracking problem. I do not mean make the radius btwn. the two legs of the angle shape larger. You will run into the same problem you did messing around with the wall thickness, you won’t make the potential high stress go away.

Edit: For starters put some dimensions, sizes, thicknesses, loads, moments, etc. on your image so that we know what you are dealing with. Show actual details of the connection and extrusion. Show a section through the handrail, its connection detail and the main structural system of the bldg. so we can see how you are doing this whole system.

 

[thaidavid40]

@JCS613 - I don't have any "thoughts" regarding the stress levels in your extrusion, as I don't have any relative dimensions to compare. However, assuming that the socket is +/- 2" square, or so, I would expect allowable stresses in the 10ksi range (or lower) for the proportions shown. The areas you show as overstressed could, in a casting, be thickened as bosses. But since this is an extrusion, to be drilled later, you don't have that luxury. If you believe your model's results, and you are afraid of the consequences, maybe you could drill and press in some stainless bushings to handle the stress distributions?
Good luck: Dave

Thaidavid

 

[IFRs]

Check the allowable aluminum bearing stress in the Aluminum Design Manual - I think you will be OK. Aluminum is great in bearing, as it yields, grows thicker and stronger and then it all stops.