accounting connection stiffness in rigid frame design in software 1

[AlexImmanuel2014]

i have read few articles on semi-rigid connection modelling for creating lighter rigid portal frames.

but i havent seend any examples in any design guides where frame models with rotational springs stiffness applied and accordingle analsysis results are published .

which brings a question whether it is economically viable option to use these methods in a consulting engineering practice .

neither do any software documents such as staad ,scia ,risa or sap200 has provided any examples .

i am a young engineer and i have been constantly making attempts to improve my knowledge in this area .So experts please kindly advice me on this .

 

[Guastavino]

Alex,

This will sound cynical, but I don't mean it to be. I hear the terms semi-rigid a lot in practice. But, I've never fully modeled a semi-rigid connection or semi-rigid diaphragm. Frankly, it's more academic than practical. With that said, you could create a model with a reasonable (key word reasonable) set of assumptions and then test it and verify if it makes sense. My point is, you are a young engineer. Many times us older engineers like to throw around big words like semi-rigid, etc. but we don't know any better at how to do that than you do. We just know it's likely BS when the term is used, but man does it sound good and is easy to tell other engineers to do. Much like "modal analysis", etc. I hear a lot about it, but don't know anyone who uses it (other than likely KootK who could derive and explain every structural method since the dawn of time!)

Now, with that said, this is where engineering judgment comes in. Take for example load paths. Say you design a moment frame lateral system in a building that has a brick veneer. Well, you don't count on that brick veneer to do anything, and yet it is likely more stiff than the moment frame you put in there to support it. The point is though, that regardless of where the load goes you have at least one REASONABLE load path to get the load to the ground. My opinion is that would be the same for a semi rigid connection. Make sure the connection can take the loads you model, and then compare it to a fixed/pinned connection to envelope it to make sure you're in the ballpark. As for getting stiffness correct, no way to know for sure, but if you design the connection for the moment in the model, it's likely going to be in the ballpark.

Lastly, someone can wax poetic about how to model that, etc., but then you have to remember job-site conditions and the fact that our material assumptions are quite a guess too. And remember, that we engineers treat steel columns with 14"x14" base plates with (4) 3/4" anchor bolts to be a pinned connection. Really? Pinned?? Yes, and it works because it's generally conservative.

Never forget the stiffness that suspended ceilings, brick veneer, and "non-load bearing" masonry walls bring to a structure. This likely exceeds the impact of a semi-rigid connection. Point is, the KISS method. Keep it simple stupid. Simple allows for hand checks and verification.

Nick

 

[XR250]

Excellent Post!

 

[GregLocock]

" Much like "modal analysis", etc. I hear a lot about it, but don't know anyone who uses it " Oh well, I've been using it since 1983.

In the automotive industry all joints in the body in white could be called semi rigid, I suppose, by which I'd mean that they are not pinned, but are typically still less rigid than the beam type sections which they join.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[canwesteng]

Modal analysis probably shouldn't be lumped in with semi rigid connections. Unless you work at a pre-eng builing fabricator and can test assemblies, use the tried tested and true connections and call them either all pin or all fix as appropriate.

 

[XR250]

In the automotive industry all joints in the body in white could be called semi rigid, I suppose, by which I'd mean that they are not pinned, but are typically still less rigid than the beam type sections which they join.

Cars are a different animal than buildings and much more highly engineered (as you know). The economies of scale make an in-depth analysis cost effective for the automotive industry (and PEMB as canwesteng stated).

 

[JoshPlumSE]

I have seen a good number of people do semi-rigid connections for storage racks and such. But, not for buildings.

My thought on the subject:
1) To do it "properly" you need to have a moment - rotation curve for your connection.
2) You then model a small "link" member between the column and beam. I use half the depth of the column. You adjust the stiffness of this link beam to get the moment - rotation behavior you desire. It will be a linear behavior (at least in RISA). That may be good for storage racks. But, it may not be good for commercial buildings
3) Often times, the real behavior is very much non-linear.
4) The design procedures that I get asked about for "wind" connections involves assuming a pinned connection for gravity loads and a fixed connection for lateral loads. While I understand where this procedure comes from, I very much dislike it. Maybe that shows some west coast / seismic bias.... because I would never dream of using one of those connections in seismic country.

 

[wannabeSE]

Semi Rigid Connections - The Steel Construction Manual has a chapter on semi-rigid moment connections. They call them Partially Restrained Moment Connections. I never use them.

Semi Rigid Diaphragms - I have analyzed semi-rigid diaphragms on a building with special steel moment frames and bare metal decks. Since the building was irregular, I don't know whether or not an envelope of rigid and flexible would be conservative.

Modal Analysis - This is frequently used for seismic design. If nothing else, is used to calculate fundamental period. Also, the mode shapes can tell an engineer a lot about how the structure will perform.

 

[Trenno]

I've read about various software packages that can calculate the correct rotational spring stiffness.

However, I agree with the KISS method. Contractors won't give you a pat on the back for reducing a plate thickness from 20mm to 16mm.

Link to said software, if anyone is interested.

[URL unfurl="true"]http://www.idea-rs.com/documents/tutorials/idea-connection-stiffness/
[/url]