Existing Column Removal 2

[SteelPE]

I have a project where the client would like to remove an existing column from a building in two locations (see attached). The original idea (which had the new support beam framed 90 degrees to what is currently shown) was to frame a new joist girder to the underside of the existing joist and use X bridging to stabilize the top of the new girder. With the new girder direction the options are less desirable.

1) Remove the existing girder and install a new joist girder that can span desired length (sort of the existing columns if necessary).
2) Frame a new girder to the underside of the existing girder utilizing kickers to stabilize the top chord of the new girder. This girder is to run from existing column to existing column
3) Same as option #2 but run the new girder the necessary length (short of the column to column location (as shown in the attached sketch)

I like #1 but I think the client is going to want #3. This is going to require me to figure out a way to stabilize the top of the new column in the middle of the new girder.

Has anyone done something similar before?

Please keep in mind I am aware of the foundation concerns hence the different options.

 

[KootK]

Wild. Must be a lot of headroom. Can you not use the kickers to stabilize the column top as well?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SteelPE]

Kootk,

Not really tons of headroom, but rather a low clear height required. The initial plan called for an 8' deep joist girder manufactured by a joist mfr. I can stabilize the top of the column with kickers, just something I am not enthusiastic about. I have thought about coming off the bottom chord of the existing girder with a vertical slotted connection in such a way that we can if we add kickers to the bottom chord of the girder it will support the column directly below.

I have a meeting with the owner and fabricator scheduled and I am going to float a few of these ideas out there to see how what they think.

 

[KootK]

I have thought about coming off the bottom chord of the existing girder with a vertical slotted connection in such a way that we can if we add kickers to the bottom chord of the girder it will support the column directly below.

The new girder will be pushed up snug against the bottom of the old one? If that's the case, I wouldn't bother with the slotting. Of course, getting snug contact between two cambered things 'aint easy so maybe you've got something else in mind.

I can stabilize the top of the column with kickers, just something I am not enthusiastic about.

While I absolutely would stabilize the top of the column for good measure, in a sense, it's not even technically necessary. You've already stabilized the load that the column supports, by way of the kickers to the new girder top chord. That, plus a little weak axis stiffness in the girder top chord, and you should be good to go.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[KootK]

Never mind the stuff about the clambering and the slotted connection. I bet that you're just grabbing that existing column a few inches below the existing girders.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SteelPE]

Kootk,

The original idea was to keep the joist girder acting as normal and just support the column with a 96G2N100k girder. In this revised scheme the intent was to keep that girder selection, however if it's possible to support the underside of the existing girder continuously would that even make sense (could use a 96G14N13k or something similar). The reason for the slotted connection at the column support was for the column that was 10' away from the existing. The slot would allow the girder to move so we don't get load reversal. However, again, if we continuously support the existing girder with a new girder is this even a concern..... thinking about it some more, if I am concerned with it...... I could add a vertical member into the existing girder above the column and have the fabricator cut a small section out of the bottom chord of the existing girder at the column.

I am now more enthusiastic about option 3. I can continuously support the existing girder with a new girder. Shim the space between the two girders if necessary and then add kickers on each side of the bottom chord of the existing girder to the top chord of the existing joists.

 

[JAE]

SteelPE - take a look at the attached. We did something similar a while back and these are the plans and details of it.
30 ft. bays with a column removed. The girders were WF beams so a bit different than your situation though.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[Once20036]

Great drawings JAE - thanks for sharing them.
Your detail 3 shows w12x19 cross beams between the two 27x84 girders. Out of curiousity, when/how were these installed? It seems these are included in construction sequence item 6, meaning they were slipped between the two girders, rotated into place, bolted up, and then shimmed to the existing girder. There was enough room to perform all this work?

 

[JAE]

Been a while and the project went through OK with no major problems...this has been about 4 to 5 years ago.

I don't recall any issues with getting things installed.
We did install new columns and I think we added to the existing column footings - or at least checked them due to the 50% greater load.

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[KootK]

Great drawings JAE - thanks for sharing them.

Yes, thank you JAE. It's rare that somebody is willing to share this much detail about one of their solutions. It's going in the KootK archive fo' sure.

I am now more enthusiastic about option 3. I can continuously support the existing girder with a new girder. Shim the space between the two girders if necessary and then add kickers on each side of the bottom chord of the existing girder to the top chord of the existing joists.

I'm leaning the other way and would prefer to grab just the column and leave a gap between the new and existing girders. I think that arrangement would be easier to fab/install. I'd also take comfort in not having messed with the behavior of the existing girder much at all. There's a fair bit of complexity and uncertainty associated with a partial, continuous bearing solution in my opinion.

Some other things to consider:

1) You could use a pair of new girders sandwiching the existing column. That would make for convenient column removal and would keep the existing girders behaving as originally intended. One could also install the new girders beside the existing girders rather than beneath them. You might get back 24" of head room if there's any value associated with that.

2) See the sketch below for another solution. It would be pretty efficient and you could dial up the pretension to grab the existing load without incurring a bunch of additional deflection. Also, it would be easy to bring into the facility in pieces I would think. Is it correct to assume that the big new girder would have to shipped as field spliced segments?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SteelPE]

Kootk,

Part of the problem I have is that the owner took my idea with the new joist girder running parallel with the joist and rotated it 90 degrees. Now everyone is sold about going in this direction (new girder parallel to the existing girder). Using two beams/girders was something I thought about, but the budget has been set and unless two girders will save money they are going to want to stick with one.

I thought about your option #2 but while moving the materials into the building may be easy, I believe there is some uncertainty with what is essentially a field built truss. I think there was a discussion on these forums a while back, but what braces the column out of the plane shown?

You are also correct that the joist girder would need to be shipped in two pieces and bolted together in the field. I have talked to a supplier (I think the supplier) and they don't seem to think it is going to be a problem.

 

[KootK]

but the budget has been set and unless two girders will save money they are going to want to stick with one.

Got it. The two girders may well save money:

1) Your design should be governed by deflection. As such, from a material perspective, the quantity may not change much.

2) All the effort of shimming etc may add up to something.

3) If the two girder system will eliminate a bunch of shoring etc associated with the column removal, that may be big money.

Sometimes clients just get ideas in there head and fixate. Clearly, if you've promised it, you're not going to take it away without a good reason.

I believe there is some uncertainty with what is essentially a field built truss.

What kind of uncertainty? You'd be hard pressed to find a more determinate truss. And it offers excellent options for displacement control.

I think there was a discussion on these forums a while back, but what braces the column out of the plane shown?

1) Run a brace from the roof deck to the column at the elevation of the bottom of the existing girders. Let the column cantilever down from there stability-wise OR;

2) Splay the four rods slightly as they leave the column.

3) Install a light, perpendicular cross bracing truss to connect the bottoms and tops of the two severed columns.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SteelPE]

As originally conceived, the intention was to span the girder from column to column and shore just the column being removed....and shore the two columns at each end to install a new footing (I don't like "adding" to an existing footing). One column was moved into the center of the bay more so we could install a new footing eliminate the need for shoring on one end.

What kind of uncertainty? The kind when you go out onsite expecting one thing and end up with anther after specifically telling the people involved what you are looking for. The pieces may not be perfectly straight in the end, the geometry might not be as perfect as is drawn in CAD which will reduce the amount of weld you can count on. I understand what you are getting at, but I feel better with items being build and welded in the shop vs the field.

 

[SteelPE]

So, this job is still going forward. I am meeting the owner, fabricator and tenant onsite tomorrow. After a round of cost savings now the owner, who is my client, is saying that he wants the girder to span from existing column to existing column. He doesn't want to add a column 15' away from the existing to avoid having two column too close to each other. The owner is also worried about what happens when the tenant leaves them with a building with random column placements. So bracing any new columns should be easy (just weld them to the existing column). Now I will just need to figure out how and where to support the existing roof (at the column or continuously under the girder).

 

[jayrod12]

I'm sure you've checked this, but do the existing foundations have the capacity for the new loads? What about the existing columns? You may just be able to fabricate a new connection to the existing column if it all works out as planned.

 

[SteelPE]

jayrod,

No, from what I can tell the foundations were only sized for gravity loads and not uplift like they should have been. This means they were designed to support 1,600 square feet of roof and not much more. The client was made aware of this at the beginning and he was instructed to include adding new footings under the existing columns in his budget. When they were trying to win the tenant and reduce costs, we moved one column 15' away from the existing column (to avoid this in one instance thus saving money). Now the client wants to go back to the original plan.

 

[KootK]

Other than the foundation pain, there might be some advantages. With the concept changing a bit, you might have the opportunity to tweak your solution a bit if you wish without looking like a promise breaking party pooper. Certainly, I understand the owner's perspective. Were I in their shoes, I would definitely be hesitant about any solution that might decrease utility for future tenants.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SteelPE]

OK, so after meeting with the owner and the fabricator we have a slightly different plan of attack. The fabricator would like to sandwich the existing column between two girders and then use some channels to distribute the load in the column over to the two girders. Attached you will find a plan of the proposed stabilization of the top chord of the girders. Do you think this is adequate bracing for these? From my prospective I don't think the top chord of the girders could possibly move horizontally.

 

[KootK]

You've got the top chord of the girder braced but I don't love the moment connections that you've created at the ends of the existing joists. They'll hang up on the braces for future loads. That's one of the reasons that I favor raising the new girders up beside the existing girder and trying to tie the top chords of the new and existing systems together laterally. Similar to the details that JAE posted.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SteelPE]

Kootk,

There isn't enough space to slide the girder up that high plus we have to deal the rerouting some roof drains. The idea the fabricator came up with had me placing the TC of the girder at the bottom of the existing joist. The issue I have with that is the existing joist girders slope causing conflicts with the bracing for the new girders.

And I don't see adding the kickers which are only attached to the top chord of the existing joists causing a "moment connection" in the joist. The idea is to apply the kickers in-between the existing joist and then use a L5x5 to take the stabilization forces back to the joist. This way 4 joists shard this stabilization force not 2, and we have something good to weld to.