what is the correct model to use for a column to beam connection when analyzing a column?

[ericbuildsit]

I posted this last night, but it looks like it was removed. I reviewed the rules and I'm not sure why it was removed so I am trying again. Thank you in advance for any assistance you can provide.

I would like to use four columns to support three LVL beams. Two of the columns will be used to join the beams. I am planning on using HSS columns, and welding a saddle bracket to the top which will bolt to the LVL beams. I have been using an online engineering tool to size the beams and the columns, but I have some questions.

1. For the beams, I used a pinned connection for the deflection because that seemed like the worse case scenario when sizing the beams. Is this an acceptable approach?

2. For the columns, I modeled the base as a fixed connection as the HSS will be bolted to the concrete footing, but for the top where it attaches to the beam, I am uncertain which model to use. I have the option of fixed, pinned, roller, weak axis brace, strong axis brace, lateral brace (bending). Which model should I use to model a column bolted to the beam? the beams will be supporting a the middle of a flat roof with 2x12 joist resting on it from both ends of the building.

Thank you in advance and please let me know if I need to provide additional information.

 

[271828]

Need more information. What's the lateral force resisting system for this structure?

 

[jayrod12]

1. Yes, pinned is often conservative for deflection design. However, depending on the other details, it may not be always conservative. If the connection is going to be more fixed than pinned in reality, there's a potential for bottom chord buckling when negative moment shows up.

2. If your lateral system is cantilevered column, I don't see why there needs to be anything at the top of that column besides an appropriate connection between the beam and column. There wouldn't be an actual boundary condition there. And ensure your concrete footing is designed to provide the required moment resistance.

 

[phamENG]

It was removed because DIY posts are not permitted. This forum is for engineers to discuss engineering, not for people to get free engineering services.

As George mentioned last night, you are asking good questions, but step one should be contacting an engineer. Find a good one and they'll be happy to explain some of the process to you.

 

[ericbuildsit]

Thanks for the assistance everyone. As far as the DIY comment, I don't see anything about that in the posting policy. I am an engineer (mechanical) so this is engineer to engineer. I will be hiring a structural engineer when it comes time. I have tried three times and all three times the engineers charged me between 300-500 for a consultation and an initial report with promises that the payment can be applied to the plans, but then when it comes time to go further, they seem to not want to do the project. I suspect they love getting 500 dollars for doing nothing with no liability is more attractive. Anyways, right now I am trying to design the floor plan, and this will be dictated by the placement of the columns. This is also a retrofit, so adding footings will have several limitations. I am just trying to get the bearings on the columns and footings so I can do the floor plan.

With that out of the way:

Need more information. What's the lateral force resisting system for this structure?


The beam will go wall to wall along the length of it (75f total). It is the center support beam, and it will have 19 long foot 2x12's connected it to the two load bearing side walls ( the side walls are currently old unreinforced masonry which have been "reinforced" at some point with 2x8 timber walls clad in plywood that are bolted to the slab). The joists are embedded into the old brick, as well as tied to the timber walls.

 

[BAretired]

1. For the beams, I used a pinned connection for the deflection because that seemed like the worse case scenario when sizing the beams. Is this an acceptable approach?

2. For the columns, I modeled the base as a fixed connection as the HSS will be bolted to the concrete footing, but for the top where it attaches to the beam, I am uncertain which model to use. I have the option of fixed, pinned, roller, weak axis brace, strong axis brace, lateral brace (bending). Which model should I use to model a column bolted to the beam? the beams will be supporting a the middle of a flat roof with 2x12 joist resting on it from both ends of the building.

1. Yes, it is acceptable, but my preference would be to move the supports as shown in the lower diagram. This reduces the maximum moment in all three beams.

2. The bottom can be bolted to the foundation using a minimum of four bolts. For analysis, I would consider the column pinned top and bottom, assuming lateral loads are resisted by shear walls or cross bracing.

 

[RPGs]

Based on what you've presented, I probably wouldn't analyze/design the columns as fixed at the base, but rather pinned-pinned supporting simple tributary gravity loads. That way your footings and column baseplate/anchor bolts are not designed for moment, reducing cost.

You've had three engineers come out and all three walked away from the project? Why didn't they want to do the project?

 

[ericbuildsit]

Thank you once again for the suggestions.

BAretired, I understand your what you are suggesting, but we are looking for the large central opening which was why those locations were selected. The outer columns were chosen to be in as close to the center as possible to avoid disruption to any existing footings that are at the ends of building. FYI the original building had the same 29foot middle span, and then 22 foot sections extending from the two columns and resting on brick columns at the ends of the building. The spans were supported by 3 2x12's with steel tension cables attached to increase the effective height of the beam. It was built in 1933 and considering it wasn't maintained (leaking roof, etc), I'm amazed it still is standing. The beam is in horrible shape and it is sagging and twisting. I've put temporary steel supports all along it for some peace of mind during the winters (Utah).

That kind of gets to your point RPGs. We have had a huge construction boom out here like most places, and everyone is busy doing what seems to be less bothersome projects. There are a lot of considerations when trying to save an old building and retrofit the structure, and I think the people around here just prefer to do the new construction projects. I am more than willing to pay for the services, but I can barely get a call back and I hound for months.

My goal is to just use the online software to come up with a rough engineering plan so that I can do the floor plan, and then keep looking for an engineer who can take the work I've done and make it "real". The important thing is that I hand the engineer something reasonable so it won't change much, otherwise I'll have to throw my floorplan out. I would be happier to error on the side of strength than cost, which is why I am trying to use the most conservative models. I feel like it reduces the chance that the floorplan will have to change down the road.

Am I allowed to solicit an engineer from this site or is that against the forum rules?

 

[dik]

I'd treat the columns and beams as pinned and pinned with the roof deck taking the lateral loads to the end walls. Due to the larger interior span, the cantilever drop in span is a great idea.

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

-Dik

 

[ericbuildsit]

It looks like pinned/pinned is the leading model for the columns/footings. I will work with that.

I am going to play around with the drop in cantilever beam idea. With the software I am using I have to analyze each section individually, and then I can link loads from one section to the other at specified locations.

That begs the question...what type of connection is created by a beam hanger? Pinned?

 

[jerseyshore]

I wouldn't go too crazy with the design of each element honestly. I don't know how many engineers are going to take what someone else did and put their stamp on it. I know I certainly wouldn't. Too much liability. Most engineers are going to design all of the members themselves so I'd work on optimizing a layout with engineering in mind and leave the analysis and design work for the PE.

 

[ericbuildsit]

I completely agree I expect all of the work will be done. What I’m hoping to avoid is designing the floor plan around a structure that isn’t feasible or is highly inefficient