Section Modulus of Welded Members

[Trillers]

I was asked by a client to provide a design to join two 40' containers for temporary buildings. Our design wind speed is 170 MPH and we are in Category D.

The proposal is to take out one long wall of each container, then join the two together. Sounds simple until you throw in the requirement that the structure needs to be moved every couple of years. The owner wants to move it as a single piece, intact and trailerable.

Our proposal is to use a couple of 3" square tubes connected by a long narrow section (3" wide) of container wall side as the main structural beam running along the joint at the tops of the containers (longitudinal).

We would like to analyze if the setup will meet the required minimum section modulus of a single I or W Beam for the loading and length. I computed the Sx as if the square tubes along each edge and the narrow section were one single member.

My question is would it make more sense to compute Sx individually or, as we have done because the proposal is to weld the components together , to compute it the way we have done as if it was one structural section.

Would like to get your opinions.

 

[Once20036]

Calculating the values of the two tubes individually or as one built up section are both valid answers, and which option is correct depends on the details.

Two tubes individually is the easier of the two methods to detail. The beams behave as two individual flexural members. The difference in strength and deflection of the arrangement that you show would be no different than the strength and deflection of a pair of beams located side by side. The vertical web will act as a compression member to share the load between the two beams. This is the weaker of the two methods because you do not get the benefit of the depth. Check the web and welds for the strength in compression.

Treating the two tubes as one member requires a lot more checks, but will be a lot stronger.
1) You need to think about the horizontal shear and ensure that the welds between the tube and corrugated web are sufficient.
2) You need to think about the horizontal shear capacity of the web. I think that you`re going to have problems here in finding any research to account for the corrugations. I've only seen this setup once in my life, and it was a collapsed structure - designed by a PEMB manufacturer that had long since gone out of business.
3) You need to think about the slenderness of your web member - again, I think that you`ll have problems here.
4) You need to think about localized effects - web crippling, web buckling, and the like. Again - I think you`ll have problems here.
Replacing the corrugated web with a steel plate or 3rd tube will simplify the analysis, and bring you back into the realm of things that have successfully been done in the past.

 

[calvinandhobbes10]

This is a creative approach but agree with the concerns above.

--Two separate 3" tubes are unlikely to work for 40' span.

--A built-up member as you show has a reasonable depth, but as described by Once the corrugated metal web likely won't take the shear flow.

--Shear flow is minimal near midpsan and highest at the supported ends. Could be feasible to add need some additional diagonal and vertical "web" members made out of 3" tube, say along quarter-spans near the ends, and corrugated the rest of the way. However, this starts to add labor and cost, and a W-shape may prove economical after all.

--Is the 3" tube solution so they can build it inside the trailer? Will the 3" tubes need splices along the 40' length? If so, all this can impact your design.

--Also, the lifting requirement and the high winds add considerable complexity. For instance, the unbraced length of this beam in wind uplift is 40'. So you may need to consider transverse purlins to reduce that unbraced length. I'd also suggest considering reducing the wind speed based on expected service life of this structure (5 years?).

 

[dhengr]

Trillers:
I would like to see sections through the containers, with dimensions and material sizes and thicknesses, loads on them, etc. The corners of these containers are really the strong vertical axial elements, and when connected together, I would be tempted to lift at three points at each end, the two outside corners and the middle connected corners area. You may want some structural member, at the roof, across the combined containers, tying them together and for lifting. Otherwise, the side walls and end panels are basically treated at large shear panels, or tension fields. I would look to get two matching containers, and literally put a cover plate over the end posts and the top and bottom chords, on the outside, attaching each pair together. I would want containers which are in good condition, beat up or deteriorated containers will be a design problem at every turn. I would cut the two interior walls out, leaving some vertical wall returns at each middle corner post, at least as wide as a cabinet or shelf system is deep, with a cover plate for a side jamb; flush at the floor, with a cover plate over that joint; and leaving as much of the double wall material (webs) hanging down at the roof, with a plate, a channel or light HSS as a bottom flange, with two webs (a box beam). It would be wise if you could get a copy of the specs. and stds. which the containers were designed and built around. They will give you considerable insight into how strong they are, and in what areas. You likely need some system of providing spanning continuity in the floor structure across the combined structure, or you need a 40' long beam there too.

 

[nonplussed]

What the client wants to do is get the cheapest possible outcome for himself while transferring all the risk onto the engineer. Every few months I get a client who wants to do some ridiculous thing with a sea container, the last one wanted a 2m deep above ground Olympic size swimming pool built out of sea containers and was saying that the sea containers 'look pretty strong'.

There are already transportable dongas made for specifically for what the client is requesting. I hope you're charging many thousands for the design. Don't try and be efficient, just go with a conservative approach for the whole design since the client is already saving huge amounts of money.

 

[Trillers]

Thanks everyone for the input.
Will let you know how it goes but appreciate the input especially regarding your thoughts on the configuration.
Nonplussed - hope you didn't display your reaction too vividly!