Design Procedure for double angle bolted column with eccentricity

[mark_1155]

I'm looking to do a calculation for a double angle column that is buckling. I know it's because of the eccentricity but all design examples lead me to a concentric approach.

Any software that can add the eccentricity to verify?

Thanks

 

[TLHS]

If you have an eccentricity then you have a bending moment in addition to the compression load. You should be able to use standard design clauses to check a non-double symmetric beam column. It's a pain in the butt to do torsional buckling checks, but not a huge deal if it's a one off.

 

[mark_1155]

This is 140 years old built up angles, 4x3 doubled (3/8), 4x3 doubled (3/8), 3/8 web connecting the two 3" legs with rivets at 7" o.c. (closer at the top). Columns are 16'. They look like they all should have been braced in a wall that was exposed over time. Beam seats which set eccentricity at about 6".

___ ___
| |
| | P out here @ 6"
___ ___

Local buckling at the top is how most are failing. 30 kips, 15 kip ft, that needs to go through the bolts. Surprisingly, the rivets aren't failing. Extreme buckling at the top but also ones with gradual deformation through column. Local buckling calc is a pain from above response, any documents I can look at to get a better idea on the approach? Just a double angle column calculation is not that simple and I didn't even see the bolt spacing factored.

Edit: Vertical lines should be the 3" leg of angle centered. Between the 3" legs is a 3/8 web, nothing is welded, just rivets.

 

[mark_1155]

I think capacity is likely around 10 kips for this configuration. There does not seem to be an easy way to really nail it down, at least the information I have. 25 kip ft of moment I just know the rivets don't work. What d to assume to get a force through the rivets? The rivets aren't failing though, columns are buckling locally up high and other columns deform entire length.

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1567219062/tips/WALL_COLU_ezulhe.pdf[/url]

 

[Settingsun]

Always put the sketch in the first post. I thought it was a channel arrangement from your double-angle description and the ascii sketch.

I'd say the intention of that many rivets is that the section acts as an H-section (aka I-section). Check it on that basis first then follow up with local connection checks. These are also called back-to-back channels and built-up columns which might help you find some literature.

It's not clear to me what you mean by local buckling at the top. Do you have a photo?

 

[mark_1155]

These are some failing ones. I want to be able to put that eccentricity into numbers and analyze the rivet connection and the bending failure in the steel. The floor below that one is similar. I want to know how to analyze the failure so I can analyze the other columns in this building if they were also free standing. Interested in getting that eccentricity accurately into the column, get a deflected shape, then analyze the load on the deflected shape.

An example calculation of bolted angle column where it includes some eccentricity would be a great start.

 

[bob33]

Is there temporary shoring in place?

 

[hokie66]

Agree with bob33. This is not the time for analysis, but rather for steps to stabilize the structure. Reinforcement can follow later.

 

[mark_1155]

I am replacing those columns. The problem is other similar columns are going to be exposed by gc and I want to show, through numbers, it's a very bad idea.

 

[rb1957]

I thought that eccentrically loaded columns required (much) more analysis than simple superposition of the loaded.

A google for "eccentrically loaded columns" should get you there.

Good luck in your endeavours to protect people from themselves.

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