1
$\begingroup$

I'm designing a frame for supporting some agricultural equipment that will be composed of welded square hollow section members. The top of the frame will be a rectangle of SHS members, with four vertical SHS legs - one in each corner.

When it comes to the corner joints, I'm not sure which is the best way to go: (A) weld the top members together with 45° mitred corners, then weld it on top of the vertical members, or (B) weld the horizontal members in between the vertical members, and cap the ends of the vertical members. See the following image for a comparison.

Image of corner joints

My feeling is that B (on the right) would be stronger, but I'm not sure why. I'd like some information - perhaps what terms I should be searching for - so that I can come to an informed decision based on the requirements of my application.

$\endgroup$
1
$\begingroup$

Typically, your welded joins are stronger than your base material itself. Looking at your two joint configurations, I would approach it from an ease of assembly/build perspective.

Joint A Wins in my book. You could fabricate the upper rectangle as a sub-assembly. Ensure that it's built square and correct and then weld the "legs"(the vertically oriented section) to it. Joint A also inherently caps the end of the section which is probably preferable. Most saws should be able to cut your 45 deg mitre. It's a pretty typical cut.

Joint B would have you eventually fabricating in 3 planes at the same time. You would need more fixturing, manipulating, and checking as your welding to ensure you're positioning and alignment is correct.

$\endgroup$
0
$\begingroup$

All else being equal the isn't a huge amount of difference as long as in case B the open end of the vertical tube is capped (ie has a plate welded on the end rather than just a plastic bung).

In practice the best method will really depend on the fit you can achieve. Often it is easier to achieve a consistent fit with square cut ends as it is both easier to cut the ends in the first place and easier to adjust the fit as necessary. Bearing in mind also that SHS is not always quite as straight as we might ideally like.

There is also the fact that SHS, especially thicker walled stuff often has quite rounded edges which can make cutting mitres to accurate length quite difficult although this does of course depend on exactly what process you have available.

So the short answer is that you'll get the biggest advantage from whatever will give you the best and most consistent fit to start with. Which depends on the processes you intend to use for cutting, finishing and welding the tubes.

There may also be production consideration as to what is the most convenient way to assemble and weld the various parts and in what order. Typically with these sorts of box section frames there is quite a lot of effort involved in getting everything held together and in getting at all of the welds which often involves turning the assembly several times or welding in awkward positions.

Equally weld order matters in terms of minimising distortion which can be hard to control in this sort of space-frame construction without a very substantial jig. It is hard to eliminate this entirely but you can at least mitigate its effects.

Overall I would say that for thin walled small section tube (say 25x25mm x 2mm) I would probably tend to go for mitred joints, favouring capped ends more as the section and wall thickness increases.

Equally if the actual manufacturing is going to be sub-contracted it may be worth discussing with them what they think the best approach would be.

$\endgroup$
  • $\begingroup$ Hi Chris, thanks for the detailed reply. (I thought I replied yesterday, but can't see my comment.) The SHS I'm intending to use is 100x5, so moderately large but with a fairly thin wall. Good point about straight cuts vs mitred cuts. I think the next step now is to get onto the fabricator and see what their opinion is. Thanks again! $\endgroup$ – Chris Apr 12 '17 at 9:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.