Flattened Tube - Resulting Strength - Finite Element Analysis (FEA)

[Hammer335i]

I am looking at flattening the end of a tube for an easier mechanical connection, but I want to make sure the resulting shape is strong enough to handle the applied design loads. I can make the part in Autodesk Inventor and run it through the program's FEA application, but with my limited knowledge of FEA, I am not 100% sure how to analyze the results.

From what I know, I should be looking at the Von Mises Stress, Safety Factor, and Displacement.
Von Mises: I am looking a these stresses and comparing them with the yield stresses of the material (in my case 50 KSI). So long as these stresses are less than the yield stresses, I am okay, correct?
Safety Factor: I am looking at these values to see how many times the applied force can be applied before it fails. I assume I want to keep this at 2 and above?
Displacement: The part should have minimal displacement, as determined by my system's requirements.

Are my assumptions correct? Is there anything else I need to be looking at?

Thanks,

Paul

 

[GregLocock]

Don't you need to consider fatigue? If this structure fails are people at risk? Is it welded or bolted- have you considered the stress raisers ? have you considered the change in material properties caused by welding?

I'm not having a go, I am pleased to see you ask the right sort of questions, I wish more engineers would. But they are a bit scary.

As ever more detail and drawings would be useful.

A typical design detail seen on children's swings is a crossbar for the basic triangle that is made from a tube (hence forming an A), flattened at each end, and then with a bolt through each end into the main tubes. The failure mechanisms are many, surprisingly few in my limited experience associated directly with the flattened bit of tube.





Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Hammer335i]

Thanks for the response. I am just wondering about the flattened end, which will also be punched with a slot/hole for a bolted connection, designed to not be scary per AISC.

This is a part for a solar application that will have minimal risk to life. Attached is a drawing of the part (

http://files.engineering.com/getfil...3-4cf5-b0d8-aaf95973be56&file=PAUL_SCAN.pdf).

Please do not worry about the bracket at the top, or the foundation below.

Thanks in advance for any help!

 

[structSU10]

I imagine this to have some complications. You are flattening the member, essentially forming plastic hinges in the wall of the tube to get it to that shape - I would imagine the residual stress distribution through the tube due to the flattening would have a significant effect on your results. You may have issues with tearing of the walls or cracks and such. The easier detail may be to just weld on a WT, flat plate with a gusset, or something to the end of the tube, attaching your bolts to the stem. No FEA required to analyze that connection. I would also think it would be cheaper and less prone to needing large tolerances than having someone crush the end

 

[jhardy1]

@structSU10:
With respect - you could be over-thinking this - flattened-end tubes are a perfectly common detail (and cheap - certainly cheaper than a welded gusset or similar!) for non-critical structural connections. (I would probably think twice about using this detail in critical aerospace or nuclear applications, but for a simple structural connection - no problem.)

@Hammer335i:
For a simple static application as yours seems to be, do you even need a detailed shell-element FEA? You could just analyse the members using beam elements with a pin at the flattened end. You can also model the flattened end using plate / shell elements if you need detailed stresses - treat the residual stresses from the flattening operation as part of the "member fabrication" process, similar to welding residual stresses, etc. The residual stresses won't really affect the ultimate static load capacity of the member.

http://julianh72.blogspot.com

 

[trainguy]

Some other things to look for:

Buckling strength, which your linear static FEA will not capture.
Applicable standards, which could place factors in front of loads or allowable stresses

tg