Aren't virtually all columns eccentrically loaded?

[CDayton]

Pretty basic and general inquiry here. But don't most columns support horizontal beams that have some sort of loading, and unless the load that the beam supports is directly above the column, then the column is eccentrically loaded?

I'm visualizing the basic framework for a standard metal building. Footing-->Column-->Beam-->Purlin-->Sheet Metal Panel-->Load.

If this is correct, then if the beam on top of the columns had a symmetrical distributed load, would you simplify it to a point load in the middle of the beam and split the difference between the 2 columns?

 

[strucguy]

The answer to your last question is "NO". Just as the bending moment of a simply supported beam with uniformly distributed load is not same as that for a beam that will have all the load lumped as point load in the middle span....the same applies to the example you indicated earlier. You need to refer to basic structural analysis book. Most of the frames used in standard metal building sytems use rigid beam-column connections. That will certainly introduce moment into the beam. But, the moment shall be carefully evaluated using basic principles of statics. Not sure if I answered your question. Good luck!

 

[strucguy]

TYPO "That will certainly introduce moment into the column"

 

[briancpotter]

As strucguy says, the bending moment for a concentrated point load is different than that of a distributed load.

As to your larger question, eccentricity comes into play when moment can be induced into the column. Consider a simply supported beam, with a pinned connection at each end. Because it's simply supported, the moment at each end of the beam will be zero. Therefore, at the point where the beam connects to the column, the only moment comes from the distance between the bearing point of the beam and the c.g. of the column - the eccentricity of the load.

On the other hand, if you have, say, a column supporting a cantilever beam, the moment at the end of the beam isn't zero, and by the laws of statics this moment will be induced in the column.

Brian C Potter, PE

http://simplesupports.wordpress.com

 

[CDayton]

Thank you for the responses.

It makes sense that a simply supported beam with pinned connections and a distributed load doesn't introduce any moment into the columns. If you look at the shear and moment diagrams, then there is no moment at the endpoints, and maximum moment in the middle.

Is this still true if the beam supports are welded instead of hinged? Does the moment diagram have the same parabolic shape as the pinned connection? It seems like it would introduce a moment into the columns. Not sure though.

 

[paddingtongreen]

This is not perfect, it is practical. If ther connecting material is flexible, it can only deliver small amounts, so small that they can be ignored.

There is a truism that any connection can be either as long as they are designed for it. Clip angles are, or were, considered flexible up to 5/8" thick but had to be designed for the transfer of moment if thicker.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[frv]

CDayton.

Yes. If the beam is welded to the column, you'd still get a parabolic shape for the moment, with the ends of the diagram shifted down below the "x" axis".

But,as mentioned above, the beam does, to a small extent, introduce a moment into the column even if it is simply supported. The value of the moment at the top of the column is obtained by multiplying the reaction in the beam by the distance from the centerline of the column to the to the centerline of the reaction (for example, the centerline of the row of bolts if the connection is a shear tab welded to the face of the column). In most applications the moment is trivial, but it is not something you can simply disregard in all cases.

 

[msquared48]

If two beams simply frame and bear on the top of a column, unless the loads and placement are absolutely identical, which is realistically not going to happen, the resultant will not be at the centerline of the column

Mike McCann
MMC Engineering

 

[strucguy]

CDayton -

There is nothing standard about a standard metal building frame. They come in various configurations. So, it's pretty difficult to give a general answer or draw a conclusion that that fits all configurations. But to answer your specific question...if you have a welded connection between a column and a beam, it will introduce moment into the column. We may guide you better if you can provide a sketch for the frame you are trying to analyze.

 

[markusgc]

CDayton - All the best in your undergraduate studies

 

[johnbridge231]

Of course they are, but that eccentricity is not significant for simple, linear static analysis. If you performa sample test, you will confirm that. On the other hand, in non linear analysis types, such eccentricities are very important as they introduce imperfections to the geometry of the model.

Analysis and Design of arbitrary cross sections
Reinforcement design to all major codes
Moment Curvature analysis

http://www.engissol.com/cross-section-analysis-design.html

 

[CDayton]

After doing some research into fixed beam analysis, I realized why I've been confused about this. I'm sure it's different with civil and strucural, but for mechanical we never learned about solving indeterminant problems using slope-deflection or moment-distribution method. We only ever solved problems that were determinant by equilibrium. So we never really got into fixed-end moments, which is what introduces a bending stress into a column.