C Channel Beam 3

[medeek]

I have a client who wants to weld two c channels together to form a rectangular tube/beam. Apparently this is cheaper than having to purchase the equivalent sized HSS rectangular tube.

I kind of want to model this up in RISA3D to check a number of things but the built up beam from two C channels has me scratching my head. Has anyone ever run across this or a similar situation? Modeling the C channels as separate members is probably the conservative route, but maybe too conservative. I really don't want to have to generate a custom section.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[Trenno]

Continuously welded together or hit and miss?

 

[KootK]

How will the beam be loaded? From the top with load applied to both channels? From one side only?

1) If load will be applied to both channels, the only demand on the weld is the ability to develop the shear strength / stiffness required to use the combined section for lateral torsional buckling. And that demand would be very small and easily satisfied by any practical weld.

2) If load will be applied from one side, I would design the welds for #1 longitudinally and also ensure that they could transfer at least 1/2 the applied load across the flanges such that you could plausibly claim to have both channels participating in load resistance.

3) The weld quality has always been a concern for me in this situation. I don't see how one would get a reliable weld connecting the channel flanges unless they were prepped for partial penetration welding which would be costly. I'd prefer to weld the two members together with stitch plates if those plates wouldn't interfere with other elements.

Re-reading the original post, I see that you're interested in the modelling rather than the weld design. Oh well, no sense deleting good information. I would model it as a custom tube section with the side walls matching the channel web thickness and the top and bottom walls matching the average channel flange thickness. Easy.



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.

 

[medeek]

@KootK I appreciate all of the good advice and I apologize for not giving more information on the problem.

The "beam" will be loaded from the top onto both channels which in my estimation will be acting as a single member provided the welded connection is sufficient. There will be two steel HHS 3.5x3.5x0.25 posts supporting it. The intended configuration is 4'left overhang, 12' middle span, 4' right overhang -> 20' total beam length.

I had not even got to thinking about the welding of the seam between the two channels. I was thinking probably a skip weld for both top and bottom seams. The composite beam will be fully braced at 4' intervals along its top (compression) face, bottom will be unbraced except at the supports.

Pretty much as I expected a custom tube section, I guess I need to dive a little deeper into RISA3D.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[wannabeSE]

If the webs are vertical, model it as two separate members and decrease the unbraced length (or specify continuously braced). Then look at the results and see if the limit states look appropriate.

 

[JAE]

And don't put a vertical hanger rod through the ends at the seam weld. Seen that before and not liked it....

(Hyatt Regency Collapse)



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[medeek]

@JAE

I'm not even going to place the sill plate bolts through the seam, they will be offset into the solid C channel flange.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[KootK]

At 4'oc with top loading, there will be no advantage to composite behaviour. I'd orient the channels back to back and connect them with bolts. This will make sill plate bolting much easier. Potentially, you could run your channels either side of your columns with thru bolt connections and avoid the need for rollover bracing at the supports.

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.

 

[jwilki]

I don't see an issue with doing this and its common to build up sections as required (client requirements - cost, aesthetics etc never end). Ensure the welds are designed for what the beam is meant to accomplish. Is it loaded in bending and shear about the strong axis and load distributed evenly between the two members (such that the load and weld are parallel) - a "small" amount of stitch welding will do the job, just to connect the two members together to help restrain it from lateral buckling. If it is loaded on the 'weak' axis", such that the load and weld are perpendicular, design the welds for the longitudinal shear flow. If the member is in torsion... well consider using a prefabricated tube or you may be forced to provide a continuously weld

 

[XR250]

Wow, how can this be cheaper? By the time he pays you to analyze this and pay the fabricator to weld it, it has to be more expensive unless the tube steel is just a weird size that is hard to get and has to buy a 50 ft. stick.

 

[medeek]

I strongly suggested just to use a nice clean rectangular tube but some people have their own ideas.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[XR250]

Just make sure it is not going to be used in a Hyatt Regency..

 

[Brad805]

Be cautious about whom the owner is having do the welding. These type of ideas tend to come from the un-informed that think anyone with a welder can do the job.

I think Koot's bolted idea might not be too bad cost wise.

 

[rb1957]

I'd second using back-to-back C's (making a I beam) ... another advantage is you can add a web where the load is applied, and save on flange bending. If Really wanted, tack weld cosmetic face plates so it looks like a box beam.

another day in paradise, or is paradise one day closer ?

 

[racookpe1978]

25 Sep 15 02:28
If the webs are vertical, model it as two separate members and decrease the unbraced length (or specify continuously braced).

Remember NOT to assume you have only one web with twice the thickness of the two separate skinny channel webs. A weld at top and bottom -even a "big" weld - means both of the two webs reacts and bends and twists independently from its neighbor.

 

[medeek]

This may be a stretch but does anyone have a good resources for typical weld details for these type of built up beams. I've got my steel books I haven't opened forever but I'm not seeing much discussion on built up beams, or at least not in the texts that I own.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com