Is W6x20 a normal section used for Columns? Especially in Hurricane zone 5

[sreeUCD]

Hello All,
I have W6x20 for columns for a 30 x 40 braced moment frame about 18 ft height to resist wind speeds of 165 mph. I have only observed columns bigger than w10 .. was wondering if there is a limitation or column guidelines for low rise buildings? this structure is at a marina,

 

[KootK]

I don't know of a limitation per se. That said, that seems like a mighty small column for the application.

 

[JedClampett]

I'm very old school. I only use column sections in the Tables in AISC. The smallest one is a W8 x 31. You can run all the same compressive stress numbers on any section, including a W6 x 20, but I figure there's some other reason they're not encouraged. For some other W8's it's element slenderness (not for a W6, though). Maybe it's due to difficulty of baseplate design or connections to it.
Alternatively, I'll use HSS sections for smaller loads.

 

[dik]

How much load (axial and moment) and how long? How is the rest of the building braced? Not in a hurricane area, but I've used W6x15 as well as cross shaped built up columns fabricated from W21x55s. It depends on the load and length.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[phamENG]

What is a 'braced moment frame'? They're used in braced frames plenty. Moment frames, not so much. I've never heard of a braced moment frame, though.

 

[sreeUCD]

So the Axial load doesn't exceed 20 kips , the highest moment is around 18 kip-ft. W6x20 was a RISA 3D recommended / suggested section and it passed all the code check. The RISAconnection had completed the baseplate design check as well. when it comes to the moment connections of the frame i started wondering if the column is too small for any kind of endplate or cap plate and bolted connections.
I have L4x4 brace to carry/transfer the wind load in all bays and all connecitons are moment connections. Not the technical moment frame with strong column weak beam one.
So I get that W6x20 is still ok from what 'dik' says

thanks a bunch for your opinions

 

[strucbells]

Make sure you are actually checking your connections beyond the "black box" of RISAconnection unless you fully understand what it does and doesn't check.

Another issue you can run into with small depth W-sections as columns is a lack of clearance for installing connections to the column web. Usually once you get to W10-W12 depth, you won't have any issues with this, but W8 and smaller can be problematic for certain connections. Often for smaller frames with light loading for equipment etc I like to use HSS columns.

 

[phamENG]

If you're using a braced frame, what's the point of the moment connections? They add a lot of cost and, in this case it would seem, are unlikely to contribute much...

 

[JoshPlumSE]

I don't see W6's used for any substantial structure. Not even for stairs. Maybe minor equipment support frames. You know the type where you've got a piece of equipment that needs to be elevated, but doesn't have much load. Maybe equipment skids. Maybe some short augured piles where you drop a W section in the middle of it instead of reinforcing. Something like that.

The reason why I don't use W6s for anything is they're so small that they're tough to weld and bolt and such. Connection costs go up really quick... Bleech! It's probably cheaper to upside to a W8 or W10 for anything substantive.

 

[sreeUCD]

Thank you JoshPlumSE. Where do i find the approximate costs for connection / welding? Ironically i checked for the clearance of W6x20 section moment connection for welding and bolts are they checked out. But following the advice here i will be upgrading the sections to W8x24 just in case,

 

[JoshPlumSE]

You'd have to talk to a contractor.... The ones I've spoken to hate those small sections. That's not a very large sample size. My superiors (earlier in my career) and most engineers I've spoken to have all had similar bias against them. To varying degrees. Companies I've worked for have used them at times. I think we had a 50 lb water pump that needed to be suspended 3 or 4 feet off of a column. Not sure we why we didn't just use some Unistrut sections or such. I would think they'd usually be easier / cheaper.

For something that's 18 ft tall, a W6 just FEELS wrong. What is the KL/r for something like that?

KL = 1.0* 18ft * 12 ft / in = 216

r_weak_axis = 1.52 in max (for a W6x25)

KL/r = 142

That's just a little too high for me. I know we're allowed to have a KL/r up to 200. But, I've always thought that 120 was the approximate upper limit for the KL/r of a reasonably sized column. I believe this is based on the buckling curves. Like 120 may be the upper limit for the capacity to be controlled by inelastic (as opposed to elastic) buckling.

Even if the numbers work out and the cost works out, an 18 ft tall building with W6 columns just doesn't feel right to me. Maybe a six inch pipe or HSS tube might work better. The connection cost could still be an issue, but the KL/r wouldn't be. I've certainly seen HSS and pipes that are this dimension so that they could be hidden (or mostly hidden) inside of walls.

 

[KootK]

But, I've always thought that 120 was the approximate upper limit for the KL/r of a reasonably sized column. I believe this is based on the buckling curves. Like 120 may be the upper limit for the capacity to be controlled by inelastic (as opposed to elastic) buckling.

Sweet. I'm going to write that down and then, later, pretend that I made it up.

 

[KootK]

I would give some serious thought to phamENG's comments here. This sounds very much the like of structure where most engineers will want to model the connections as pinned, such that the columns attract no moment other than, perhaps moments associated with eccentric delivery of beam shear. If you set things up that way, you can make of go of most any shape because:

1) You probably run your beams over top of your columns rather than into the sides.

2) No moment connections.

4) Your braces probably hit gussets welded to the column flanges or simply connect to the undersides of the beams only.

This eliminates any issues with congested connection costs. That said, I still don't love such a slender column.

 

[Flotsam7018]

If there is good reason to use such a small section, such as architectural constraints, then I see no issue with a section in the elastic buckling range - Euler is an old friend after all. Your connections will require more thought for adequate clearances, but it can be done. P-delta, in particular little P-delta, column out-of-plumbness and all those good things are important to consider when approaching higher slenderness ratios. That being said, if you have no good reason to use such a small section other than it works in RISA, I agree with the general consensus to bump it up.