RC Beam. Are they always considered to be fixed?

[rcm87]

Hi, I have a doubt regarding RC beams.

If you are building a rectangular room, which has 4 columns (one in each corner) and 4 beams, how should you proceed with the analysis of the beams?

Should I consider them to be fixed-fixed? My guess is that this is the correct answer, but I'm not 100% sure.

Thanks. =)

 

[msquared48]

If you design and put in enough reinforcing in the correct pattern to develop the full negative moment at the column connection, yes, consider it fixed. Otherwise, consider it not fixed. It's just that simple.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[hokie66]

Assuming a rigid connection of beam to column is not the same as assuming the beam has a fixed end. Maybe you know that.

It is very difficult to completely develop the moment capacity of concrete members around a corner. Most beams at edge columns are assumed to be pinned for design, but with nominal top bars which are usually code required.

 

[rscassar]

No, the negative moment at the column will be somewhere between 0 and wl^2/12 depending on how stiff the column is. It is important to design some reinforcement at the support because it will reduce the deflections mid-span.

 

[JAE]

In the past we've done what hokie66 suggests and assumed a pinned connection at the ends but then came in an just manually added top bars (perhaps 75% of the interior span top bars if there was an interior span). We then got conservative on the column design by assuming a similar moment. This was done due to the limited computer analysis capabilities many years ago.

Today I'd include the columns in the analysis.

ACI 318 allows the designer to assume that the far ends of columns are fixed and implies that the column stiffness should be included in the analysis along with the beam per kikflip's post above.

 

[rcm87]

Thanks for the answers.

Would this change if the construction has two levels? Meaning that the beam is between the first level column and the second level column.

And lastly, in the case JAE commented about. When you have a continuous beam (for example a 4-span section) you consider the connections in the two inner columns to be fixed, and the two connections with the border columns to be pinned?

 

[rcm87]

Sorry. It would be a 3-span beam.

 

[csd72]

In in-situ concrete, all connections between members should be assumed to be fixed. You need to carry out a full frame analysis on this type of structure, not just an individual member analysis.

Foundations are usually treated as pinned unless there is a specific measure to ensure fixity.

 

[rcm87]

Yes, in the case I'm seeing the RC members would be build in-situ. Meaning you place the steel reinforcement and then cast the concrete in-situ.

My doubt started when I saw that the final design of this rectangular room, had more steel reinforcement in the bottom of the beam. When, if the connections should be considered to be fixed, the reinforcement should be more in the top of the beam were it connects with the column.

 

[hokie66]

More reinforcement in the bottom is correct for a single span condition. It is not fixed at the end because the column rotates. Ask yourself if the beam to column connection becomes a pin, what happens to the beam? It becomes a simple span, and that determines the reinforcement required in the bottom, perhaps reduced a bit because of the end column stiffness. I don't like calling any concrete connections fixed...they are continuous.

 

[JAE]

If you don't put enough top steel in at the end you can get some serious flexural cracking down the face of the column and this certainly could compromise your shear capacity there.

 

[rapt]

rcm87

I think we need to define "fixed" to answer this.

In some FEM models they may use the terminology "fixed" as in there is moment connection between them. But they are not restrained against rotation!

They are not "fixed" in terms of restrained against rotation.

For a single span fixed at the ends, the end moment would be wl2/12 and span moment wl2/24.

You will never generate this with columns as there will be rotation as several above have said. You have to dertemine the amount of "fixity" depending on the relative stiffness of the columns to tg=he beam!

For a single span with columns, the span moment will still normally be much more than the end support moments due to the totation at the supports.

And you should not be asking a question like this on this site, ask your mentor/supervisor!

 

[csd72]

Those theoretical fixed end charts are almost never applicable in reality.

When I referred to the connection as being fixed I meant that they are locked to each other so that the column and beam rotation are the same.

These are absolutely fundamental issues that you need an experienced mentor to help you with.