Moment on connection

[dccd]

Say that I have a continouos beam resting on top of column spanning across 3 column. The applied load is UDL on the beam, here's how the BMD look like. My question is whether the connection between the beam and column nned to sustain the moment ? Which point need to sustain the moment ? All 3 locations ? Or the edge of the beam only ?


The connection between the column and the beam is the beam will be welded to the plate , and the plate will be welded to the column.

 

[canwesteng]

Well the columns definitely need to resist moment for stability. Which ones are up to, although all three are set up to in your screenshot.

 

[palk7 EIT]

In the case of a fixed connection which it is, the beam & the column both will experience the BM as its seen in the pic. The moment will be there in all the three location. The moment from edge of the beam will be transferred to the top of the column via the fixed connection as you mentioned through the welding of plates.

 

[human909]

My question is whether the connection between the beam and column nned to sustain the moment ?

Short answer; absolutely. Unless you design a connection that is sufficiently flexible for it to allow enough flexibility for it not to attract moment.

Best to design to the required behaviour you want. If you don't want moment on the centre column then best to design it as pinned, detail it appropriately and then you don't need to have a connection to sustain moment. Don't make the mistake of making a moment connection when you don't want one. In the wrong circumstances it can be quite problematic.

 

[dccd]

Short answer; absolutely. Unless you design a connection that is sufficiently flexible for it to allow enough flexibility for it not to attract moment.

Best to design to the required behaviour you want. If you don't want moment on the centre column then best to design it as pinned, detail it appropriately and then you don't need to have a connection to sustain moment. Don't make the mistake of making a moment connection when you don't want one. In the wrong circumstances it can be quite problematic

Just wondering, does the connection of the central column and beam nned to take the moment ? Or the moment at the center will be taken by the beam itself ? The connection between the beam and the column at center doesnt need to take momemt ?


I knew at the corner column, the connection between the beam and column nned to take moment

 

[dccd]

Is it reasonable to design the beam to be pinned pinned at corner location while having pinned base ? (The beam is continuous spanning across 2 spans) .
I dont think so.

So, i procced with the design of beam to take moment at the end and pinned base .


Correct me if I am wrong.

 

[palk7 EIT]

The connection at central column will be needed to designed as a moment connection if your assuming it as a fixed connection cuz the moment from left beam and right has to be connected to the column (centre), however the center column will not experience any moment on the top because the moment from left beam and right beam will cancel with each other (nullify the moment's going to the column), but keep in mind that the connection will sustain the moment its only the center column that will not experience due to the cancelling effect as mentioned above.

Yes you can design the ends (corner) as pinned and the center column as fixed (moment connection) no harm in it, just that your BM diagram will be different and the corner columns will not experience a moment cuz its pinned and the bending moment will be more in the mid-span of the beam.

 

[rb1957]

not my area but I think the moment is in the beam ... if the model assumes pinned supports (so that the column is intended to take load (compressive and lateral) only.

If the column was modelled as a fixed support ... then, of course, the moment is in all three elements (both beams and the column).

another day in paradise, or is paradise one day closer ?

 

[human909]

Is it reasonable to design the beam to be pinned pinned at corner location while having pinned base ? (The beam is continuous spanning across 2 spans) .
I dont think so.

That statement isn't true. I is perfectly possible and could be reasonable to design the system such that the beam is pinned at the corners. It would be a bit abnormal but you should be designing to your structural requirements and constraints rather than what is 'normal'. You need to understand the impact of your design choices.

Just wondering, does the connection of the central column and beam nned to take the moment ? Or the moment at the center will be taken by the beam itself ? The connection between the beam and the column at center doesnt need to take momemt ?

I believe I answered that question. If you are not designing a flexible connection then yes, your connection needs to be strong enough to take moment.

 

[dik]

Unless a portal for lateral loading, I'd treat the end connections as pinned... and use plastic design for the 2 span continuous condition with a Class 1 section.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[Blackstar123]

Just wondering, does the connection of the central column and beam nned to take the moment ? Or the moment at the center will be taken by the beam itself ? The connection between the beam and the column at center doesnt need to take momemt ?

If you haven't intentionally assign a moment release on the interior column, then the bending moment in the interior column is probably coming out to be zero due to symmetrical geometry and loading condition.

In this particular case, since the beam is continous over the column, the connection of beam to interior column need to be designed for bearing only.
Beam designed for these end conditions will be analogous to a two span beam with end supports fixed and a hinge at interior support.

But If you want a frame action type of behavior at the interior beam column joint, for other load cases which results in non zero bending moment in the interior column, then you need to design the connection to transfer that moment to the column, no matter if the beam is continous or not. Otherwise, interior column will act like a leaning column and will not resist any moment or provide lateral stability.

 

[dccd]

since the beam is continous over the column, the connection of beam to interior column need to be designed for bearing only.
Beam designed for these end conditions will be analogous to a two span beam with end supports fixed and a hinge at interior support.

Yes, this is my situtaion. So, the connection of beam to the column in the center need to take moment, right ? The connection od the beam to the center column only take compression (which i dont have to design ) ? Since compression will be directly transferred directly to column? While the bending moment (hogging) will be taken by the beam itself ?

 

[dccd]

That statement isn't true. I is perfectly possible and could be reasonable to design the system such that the beam is pinned at the corners. It would be a bit abnormal but you should be designing to your structural requirements and constraints rather than what is 'normal'. You need to understand the impact of your design choices

I think it's not possible to design the beam to be pinned pinned and having pinned base at the column end since no bracing to restraint the lateral loading. Please correct me if I am wrong.

I think it's only possible to design the beam to be pinned pinned and having pinned base at the column unless there's bracing to restrain the lateral load/ second order effecr. Correct me if I am wrong.

 

[Blackstar123]

You're the designer. You need to decide how you want your frame to behave in certain loading condition. Once decided, analyze the frame by properly modelling the assumed behavior and then detail the connection accordingly.

In this particular case, no need to design for moment because the net moment at beam column interface is zero. Just make sure your frame is laterally stable by other means.

(which i dont have to design )

You should definitely perform all the relevant local checks at the connection such as local buckling, crippling etc.

 

[Lion06]

To answer the question very directly as it's asked, yes, the connection needs to designed to take that moment. I would have some questions, though - What is the lateral system? Is it modeled correctly? If this is not part of the lateral system, then I would not detail the end connections as moment connections and would not track the moment through that connection. Also, the way you have it drawn would not track any moment into the column (because the balance each other out), but you will need to consider unbalanced live load, which will track an unbalanced moment into the connection and the column.

 

[dccd]

Also, the way you have it drawn would not track any moment into the column (because the balance each other out)

Ya, I know the column wouldnt have any moment, my question is whether the connection between the beam and the plate joining the column need to take moment or not ?

P/s : The beam is welded to the plate , and the plate carrying the beam above will be welded to the column below...

 

[human909]

Ya, I know the column wouldn't have any moment,

Oh but it could. As others have pointed out non uniform loads would induce moment. So would lateral movements.

my question is whether the connection between the beam and the plate joining the column need to take moment or not ?

P/s : The beam is welded to the plate , and the plate carrying the beam above will be welded to the column below...[/quote]

So it sounds like you are designing a inflexible connection. AKA a moment connection. Thus you will need to design the connection for the moment and depending on your relevant code a minimum design moment which could be rather large.

Like has been said by myself and numerous other. YES YES YES. If you are designing a connect moment connection then you need to design it for moment. If you don't want a moment connection then than neeeds to be explicitly designed to enable the flexibility required such that moment is not imparted on the column.

 

[JLNJ]

You are asking the wrong question. First, you need to determine how you WANT it to act. You need to have a design INTENT.

After you figure out how you WANT it to act, then you detail it accordingly.

If you have no idea how you want it to act, then you need to show us the whole structure and we will explain your options.