Eccentricity in Axial Loaded beam to column connection

[learning2banengineer]

Hi All,
I have a question regarding eccentricity on the bolts in the axial loaded connection as shown in the attached PDF.
Do we assume that the pass through load acts along the centriod of the beam and create an eccentricity to the bolt group or can we ignore that and assume that the load passes through the centroid of the group? If eccentricity is considered, it will result in a very bulky connection.
Please shine me some light on this.
Thanks.

 

[paddingtongreen]

Assuming that is a knife plate on each column flange.
You have an eccentricity, it's there, it won't go away because the designer would like it to.
You might try to apply the moment to the beam instead of the connection, the shear being in the weld where the web extender is welded to the beam.
The thing is though, you may already have a moment connection if you have a tee or double angles instead of the knife plate, thick enough to hold the pass through load.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[connectegr]

We do not consider axial load in the extended portion of the connection. We would consider axial load in the beam web bolts and use the extended portion for additional shear capacity.

For additional axial capacity, I would suggest adding additional vertical columns instead of adding horizontal rows. This adds eccentricity to the shear connection, but eliminates the eccentricity due to axial load.

Using a vertical extension for axial load, significantly increases the design considerations and connection cost.

http://www.FerrellEngineering.com

 

[connectegr]

Assuming the beam web is adequate for the axial load, clip angles or endplate would be the standard connection. If necessary the web capacity can be increased with doubler plate(s).

http://www.FerrellEngineering.com

 

[learning2banengineer]

Thank you for the quick replies.
Maybe I should have clarify my sketch a bit more, the column is actually a HSS column and there is a knife plate going through it to support the beams on two sides.
The reason why I added the knife plates to the bottom flanges of the beams was because the axial load exceeds the resistance provided by
w * d * 0.9 * Fy of the beam.
I understand that we can make use of the flange materials even though there is shear lag by only connecting to the beams' webs, but I am not that comfortable just relaying on that to transfer the loads in the flanges and thought that the flange can help take the excess axial load out more directly if a knife plate is added to it. Also, regarding eccentricity in the bolts. I was thinking that the majoritiy of the axial load is coming from the deck welded to the top of the beam so the actual axial load is not coming through the centre of the beam anyways, then isn't there still eccentricity in the bolt group even only the webs are connected, without the knife plate at the bottom?

 

[connectegr]

First...
How do you fabricate "knife" thru-plates through an HSS column? How are the plates welded to the slots, etc.? Typically thru-plates are single plates. This is an option, and there were several comments on this in another thread,

http://www.eng-tips.com/threadminder.cfm?pid=172

If additional tensile strength is required, I would recommend web doubler plate instead of web extension.

http://www.FerrellEngineering.com

 

[connectegr]

Unless the beam is design for the moment, the axial load is transmitted along the centerline. Are the studs on the beam? If the deck is just puddle welded to the beam, the axial transfer would be small. If these are "drag" forces, which collect at a braced bay, then the axial force is in the beams and then the bracing.

http://www.FerrellEngineering.com

 

[connectegr]

Just to clarify, my vocabulary

"Knife Plates" are a two plate connection with a plate near-side and far-side, with the beam web "knifed" between the plates. The bolts are in double shear.

"Thru-plate", single plate, or shear tab is a one-sided plate connection, and the bolts are in single shear.

Maybe we are not using the same terminology.

http://www.FerrellEngineering.com

 

[learning2banengineer]

Thanks connectegr
I should have stated my que more clearly, and I didn't notice that the hidden lines didn't show up in my sketch and a line was missing. Attached is a better sketch

The deck is only puddle welded to the beams, but there is a 275kips thru force at certain locations pass through W61x31 and W16x40 Beams.

I am in canada, so I am only designing the connections, not the beams, but I believe the consultants did not account for any moment that may result from from eccentricity in the connection.

 

[DaveAtkins]

What I generally do is design the connection for the axial load, centered in the connection, and check the beam for the additional moment.

DaveAtkins

 

[BAretired]

learning2b,

I assume you mean W16x31 and W16x40.

Maybe it would be easier to use a W16x40 continuous and splice the column above and below the beam, using half HSS welded stiffeners between beam flanges each side of the web.

BA

 

[slickdeals]

May I suggest transferring the axial forces through the 2 flanges and using the web plate to only transfer shear. That way you can keep the connection concentric.

275 kips can be easily transferred through the 2 flanges. You probably need 1/2" x 9" A36 plates. Alternatively, you can install the flange plate as a seat to the HSS and field weld the top plate to the column.

 

[connectegr]

If you connect the flanges you may have to consider the additional fixed end moment. Although the web only connection can become rather large, it is still technically a flexible connection, and a simple beam.

http://www.FerrellEngineering.com