3
$\begingroup$

Say I have a simple reinforced concrete frame (3as shown below) with all beam ends pinned:

enter image description here ( the keyplan)

This is the detailing output:

enter image description here The beam elevation detail ( left), beam cross section detail ( right)

enter image description here Column cross section

I understand that due to the pins, all beam end moments will be 0. However, at the column joint, will there be any moment? I think no.

But one of my engineering friend think that because this is a reinforced concrete beam, and we design our beam reinforcement bars for the designed moment, this designed moment should somehow be transferred to column (even though the frame analysis shows that there is no moment at this beam joint), and thus the column joint will have moment.

In other words, the presence of reinforcement bars in the beam means that we need to design for moment in the column at the joint, as the reinforcement will create a moment in the column, even though in the analysis this "reinforcement induced" moment is simply not there ( and the analysis shows us that column moment at joint is a 0). Is this a common practice ( as stipulated by convention or demanded by code of practice), or is this something completely unheard of?

Could he possibly be right?

$\endgroup$
9
  • $\begingroup$ This was very unclear originally. Based on your middle paragraph, I think I can understand your question, and so I have edited all the question to be consistent with this. Please review and let me know / edit it if I have misinterpreted. The main thing I was unsure of is whether you have pins at the central column. $\endgroup$
    – AndyT
    Aug 18, 2016 at 8:13
  • $\begingroup$ @AndyT, I think you misunderstand my question; my bad. I have edited the question ( especially the bold part) to emphasize on my actual question $\endgroup$
    – Graviton
    Aug 18, 2016 at 10:46
  • $\begingroup$ By what mechanism do you believe reinforcement induces a moment in the joint? $\endgroup$
    – AndyT
    Aug 18, 2016 at 10:50
  • $\begingroup$ @AndyT, I am not too sure; one of my engineering friends says so ( without being able to provide a mechanism or a code practice reference), so that's why I'm asking here, so that I don't appear too ignorant when I discuss with him, lest he calls me stupid. Question also updated $\endgroup$
    – Graviton
    Aug 18, 2016 at 10:54
  • 1
    $\begingroup$ Could you please edit your question with a drawing/sketch of the connection you're asking about (including reinforcement)? This has become very generic and theoretical, while your question seems to be far more down-to-earth. $\endgroup$
    – Wasabi
    Aug 18, 2016 at 12:06

2 Answers 2

2
$\begingroup$

Just as there is conservation of energy and mass in the universe, so must you also think of "conservation of internal forces". I totally made this term up, it doesn't exist. What I mean by that is that forces and moments never simply appear or disappear.

As you've stated, your beam will present zero bending moment at nodes 4 and 6, where it connects with the left and right columns, respectively. Now, for those columns to present any bending moment at those nodes, the bending moment would have to come from somewhere, in such a way as to explain the discontinuity which would be present at the node (column at node 4 with bending moment, beam at the same node without bending moment). In this case, there is nowhere for such a bending moment to appear.

Therefore, no, the left and right columns will also present zero bending moment at the nodes. The central column depends on how it's linked to the beam. If the column is hinged to the continuous beam, then it'll present zero bending moment. Otherwise, if the column is fixed to the beam, then it will absorb some of the beam's negative bending moment.

Notice that this has nothing to do with the structure's material or with structural design. The solution for internal forces is defined entirely by structural analysis, which is mostly agnostic to such considerations.

As to how to calculate the reinforcement for a column with zero bending moment, one adopts minimal reinforcements defined in codes or, if your country's codes have no such definitions, common practice.


After your edit showing your actual situation, however, let me quote a comment I previously wrote beneath this answer: "A fixed node [...] wouldn't have the narrow throat, but would be a monolithic connection with almost all the column and beam rebar passing through." This connection has no throat and is entirely monolithic (where does the beam end and the column begin? You can't tell because the connection is, well, both) and is therefore clearly, beyond a shadow of a doubt, a fixed joint. Your model shouldn't have any hinges on those connections and, therefore, the beam and the column should both present a (possibly small) bending moment at that point (negative bending moment on the beam, which generates tension on the top fiber. In the column, this bending moment will generate a tension on the external fiber). The bending moment in both will be equal.

$\endgroup$
6
  • $\begingroup$ by "conservation of internal forces", you are saying that the internal bending moment/forces at beam and column ( at the joint) must be the same, if there are no extra bending moment/forces at that joint? $\endgroup$
    – Graviton
    Aug 18, 2016 at 3:49
  • $\begingroup$ Wasabi, according to your knowledge, is there a code of practice in the world( or common design practice) that simply has an "reinforcement induced" moment term at the column joint, for reinforced concrete structure? $\endgroup$
    – Graviton
    Aug 18, 2016 at 10:48
  • $\begingroup$ @Graviton: Exactly. For there to be a discontinuity in the bending moment diagram, a concentrated moment must be applied at the point of the discontinuity. They may be an external force applied at the point or, in the case of a grid structure (as opposed to frame structures as you're currently designing), torsion and bending moments can "transform" from one to the other when perpendicular beams are connected. $\endgroup$
    – Wasabi
    Aug 18, 2016 at 11:43
  • 1
    $\begingroup$ @Graviton: As to your second comment, I don't understand what you mean. As AndyT mentions in his answer, your reinforcement and your analysis must be compatible. If your beam is hinged, then the reinforcement must be adequate for a hinged connection (which implies in no principle reinforcement continuity between beam and column). If the beam is fixed, then there must be significant reinforcement continuity. $\endgroup$
    – Wasabi
    Aug 18, 2016 at 11:53
  • 1
    $\begingroup$ @Graviton: If you look at the examples of different hinges I showed in an answer to a previous question of yours, you'll notice that it is impossible to have any significant reinforcement continuity. Freysinnet hinges have no reinforcement at all, while other hinges have specific reinforcement for the hinge but the majority of the beam and column reinforcement are terminated at the hinge. A fixed node, however, wouldn't have the narrow throat, but would be a monolithic connection with almost all the column and beam rebar passing through. $\endgroup$
    – Wasabi
    Aug 18, 2016 at 11:58
2
$\begingroup$

You seem confused between how a structure behaves in real life and how it behaves in a structural analysis model. They key point you seem to be missing is that one must reflect the other.

If, in real life, there will be reinforcement continuity between the beam and the column, then the analysis model should not use pin-ended beams; it should use fully-connected beams.

If, in the design, the analysis model is based on pin-ended beams, then the detailing should be done to avoid moment-fixity.

$\endgroup$
10
  • $\begingroup$ Andy, so what you are saying is that if Analysis shows that there is no moment at column joint, then the column design/detailing should have no designed moment as well? $\endgroup$
    – Graviton
    Aug 18, 2016 at 8:41
  • $\begingroup$ @Graviton - No, that's not what I'm saying. There is a difference between what you tell the analysis, and what the analysis tells you. If you tell the analysis that there is no moment at the column joint, then you need to make sure in your detailing of the joint that it cannot transmit moment. If you tell the analysis that there can be moment at the joint, and the analysis tells you that there is no moment occurring you a) think your analysis package is broken; but if you check the results and are happy with them then b) you can detail the joint however you like (as long as it meets code). $\endgroup$
    – AndyT
    Aug 18, 2016 at 9:01
  • 1
    $\begingroup$ {cont} Rather than use an empirical method to give the moment in the joint, the analysis should be re-run using fix-ended beams. $\endgroup$
    – AndyT
    Aug 18, 2016 at 11:54
  • 1
    $\begingroup$ @Graviton: To quote AndyT's answer, "The key point you seem to be missing is that one must reflect the other". It is utterly nonsensical to analyze the structure one way and design it another. One must choose. If the structure is to be pinned, it must be analyzed and designed accordingly. Likewise if it should be fixed. The only exception is in the paranoid case suggested by AndyT's other answer, where the structure is continuous and then you analyze and design it as BOTH pinned and fixed. $\endgroup$
    – Wasabi
    Aug 19, 2016 at 1:38
  • 1
    $\begingroup$ @Graviton - "I just want to be sure that there is no code practice basis for him to analyze it as pin but detail/design it as fixed?" I can't answer for every standard/code as I've not read them all. I have never come across such a requirement, and would be very surprised to find one. $\endgroup$
    – AndyT
    Aug 19, 2016 at 8:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.