Beam to Column Eccentricity in Column Analysis 2

[marinaman]

If you guys have a multi-story steel framed building, in your column design, do you include connection eccentricity in the analysis of the column?

For example, if I have W12 columns, the connection eccentricity to the column centerline would be roughly 6", each side, in the strong direction and only about an inch in the weak direction. If I include these eccentricities, then the gravity columns become both axially loaded and biaxially bent.

Or

I've heard it argued that straight-up axial only analysis is fine, as beams framing to the column web are nearly to the centerline, and, that members framing to the column flanges offset each other, as there's typically a beam on the other side of the column.....thus only columns with a beam framing to one of the flanges, with no beam on the other column flange, should be checked for eccentricity.

Thoughts?

 

[phamENG]

Should you check it? Yes, yes you should. Do most people? Depends on the project. 2 or maybe 3 stories? Probably won't make a big difference for most gravity columns (corners being the exception, as you noted). 5 or 6 story building? Things are starting to add up. And remember, you don't just assume balanced load all the time. You have to use pattern loading to get worst case conditions.

 

[dold]

+1 for what pham said.

It can be tedious if you're doing it by hand. If you're using software like RAM Structural System then these eccentricities are automatically calculated in the steel column design module. It also considers all pattern loading conditions which is nice.

 

[dik]

It might be that with the higher axial load with taller buildings, the moment has less of an effect. Effect of moment may be more pronounced in shorter buildings.

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

-Dik

 

[KootK]

Danger! Danger! This is a much debated issue here and can be a very nuanced thing.

I would recommend reviewing this thread -- among many -- for a diversity of thoughtful opinion on this exact topic: Link.

I don't bother with connection eccentricity for most simply supported beam-column connections. But, then, in some situations I do. Some countries' codes insist upon it.

Basically, if you're going to ignore the connection eccentricity, you want to be in a position where you are able to confidently rationalize that decision to yourself, and perhaps your boss / AHJ.

 

[KootK]

It might be that with the higher axial load with taller buildings, the moment has less of an effect. Effect of moment may be more pronounced in shorter buildings.

That is certainly the case with high-rise concrete columns which:

1) Don't care incoming floor framing moments at the lower levels and;

2) Are often dominated by roof/floor level framing moments at the top levels.

That said, this particular condition speaks mostly to continuous framing systems rather than simply supported floor and roof framing with slightly eccentric shear connections.

 

[HTURKAK]

Common practice in my zone is , yes.. The eccentricity of the connection should be taken into account in the global analysis. If the conn. is a bolted beam to column connection (end plate, web cleats and so on) to web of an open column in this case may be neglected.
The following doc. gives an idea about the practice at EU.
...

He is like a man building a house, who dug deep and laid the foundation on the rock. And when the flood arose, the stream beat vehemently against that house, and could not shake it, for it was founded on the rock..

Luke 6:48

 

[dik]

I've sealed a couple of straw bale house constructions...

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

-Dik

 

[hardbutmild]

@HTURKAK excellent reference. I see that it is only a part of the whole document, is there a full version?

 

[HTURKAK]

You can reasch the subject doc. at ,

https://people.fsv.cvut.cz/~wald/CESTRUCO/Texts_of_lessons/04-GB_Structural_Modelling.pdf

You may Search the site for other docs.

http://steel.fsv.cvut.cz/

...

He is like a man building a house, who dug deep and laid the foundation on the rock. And when the flood arose, the stream beat vehemently against that house, and could not shake it, for it was founded on the rock..

Luke 6:48

 

[marinaman]

Thank you guys very much. Kootk, thank you for the link.

In my youth, the folks I worked for ignored the column eccentricity alot, but, we mainly designed 1-2 story steel framed buildings.

Flash forward 30 years. I include the eccentricity, but, I've heard it argued otherwise. I'm always going to include the eccentricity, but I wanted to hear opinions on the subject.

For those of you working in RISA Floor like me, we can designate connection types and then "click" to include eccentricity in the column design. Its in the beam data, not in the column data.

Thanks again all.

 

[lexpatrie]

You'd generally need to include consideration of eccentricity (now), back in the 1990s and prior it wasn't probably as commonly done. I remember having various discussions circa 2000 about including them in RAM steel, for example. Somebody did a big project without any eccentricity and after they left, the rest of the group got all irritated about it. Would it have caused a collapse? Don't know. Did they revise the columns later on? Again, don't know.

When you get into more and more convoluted columns, say for multi-story, the pattern loading gets pretty convoluted, and ideally all the load cases have a different eccentricity ("net eccentricity", if you would, from the pattern live load and the layout of the column (corner, edge, center).

Now "consideration" could mean that the eccentricity is "taken" by the connection design, rather than the column. I believe that's still a valid approach and some connection types the eccentricity on the column can be neglected as the testing shows it doesn't show up on the column. Not sure where exactly I've seen this, or which connection types, but the AISC Manual would be a starting point.