Top of column moment connection

[JStructsteel]

Got a moment frame, and its for a fabricator building an addition. They were asking why I was doing a end plate moment connection over having the beam on top of the column.

I have not designed the connection yet, but in my mind, the end plate welded on the beam (W27) is better for strength than having it on top of the column (W18).

I would imagine I will be stiffening the web of the W27 if on top of the column.

The W27 is sloped 1/4"/ft, so not sure if that would complicate it sitting on top

Thoughts?

 

[human909]

Lay people (aka non engineers) always like to see bearing connections. So having the beam bear on the column is seen as "better". The bearing load path is crystal clear. The shear load path through bolts seems to be weaker to a non engineer.

However most of us engineers know better. Bolts are fantastic in shear and so the shear isn't a problem. In the beam to column connection the often narrow profile of the column and the deep profile of the beam makes an end plate normally superior. It also gives good twist restraint for the beam which is often needed.

I'm in the process of designing a beam bearing on a column due. Ensuring suitable twist restrain for the deep beam is going to require specific members. I'd normally frame the beam into the column but in this specific circumstance I can't.

 

[EngDM]

I'm in the process of designing a beam bearing on a column due. Ensuring suitable twist restrain for the deep beam is going to require specific members. I'd normally frame the beam into the column but in this specific circumstance I can't.

Yea you almost need to have similar depth framing members framed flush to prevent the twisting, cause a bearing column isn't doing jack shit without quite the detail.

 

[human909]

Yea you almost need to have similar depth framing members framed flush to prevent the twisting, cause a bearing column isn't doing jack shit without quite the detail.

Yep. I suspect I'll end up with something like they do with bridge girders. Eg something like this.

(Though in an aside, to another thread according to the code (AS4100) web stiffeners and a fixed base are enough to stop twist and call it good. But I'm not trusting that!)

 

[JoshPlumSE]

For what it's worth, I think fabricators (and detailers) usually have a pretty good sense over what details are frequently used successfully and which are cost efficient. Not saying they're always right. But, they are more knowledgeable than an average client.

I would imagine I will be stiffening the web of the W27 if on top of the column.

It think this may be the most significant reason why a End-Plate connection framing into the column could be better / cheaper / easier than a cap plate on the column and the beam running over. If so, then all you have to do is check the bearing / web crippling and such to see if the W27 web would need stiffening.

The W27 is sloped 1/4"/ft, so not sure if that would complicate it sitting on top

I don't think that's a significant difference in cost, or ease of construction.

 

[canwesteng]

You'd want stiffeners matching the column flange full depth in the W27, and a fly brace on one side of the W27 where it bears on the column to a perpendicular member, at a minimum. Then there is the question of if the flange is thick enough for the bending. Certainly can be done safely though, and may be easier to erect.

 

[Eng16080]

Lay people (aka non engineers) always like to see bearing connections. So having the beam bear on the column is seen as "better". The bearing load path is crystal clear.

I'm a big fan of direct bearing connections in general, not that they make sense in every situation (this perhaps being one of them).

The shear load path through bolts seems to be weaker to a non engineer.

I'd say it is weaker, but in most cases it's still more than adequate.

 

[human909]

I'm a big fan of direct bearing connections in general, not that they make sense in every situation (this perhaps being one of them).

In steel design I don't find it doesn't make as much sense as it would initially appear. It is generally inferior form a buckling perspective and it can often push shear load into flanges rather than webs. By using bolted connections you can much better control the load path. Even if it means having a nominal gap (10mm) between the two elements which would otherwise bear on each other.

I'd say it is weaker, but in most cases it's still more than adequate.

I agree it is weaker from a shear perspective. But from a buckling perspective which generally matters far more it is usually superior. Who cares about shear in steel design!

Or in the words of Structural Engineer Mr Gabriel Gardin:
"Shear? I feel like no one really cares about shear. Shear is like the quiet kid in the classroom that nobody talks to."

(I personally found it funny.)

 

[Eng16080]

In steel design I don't find it doesn't make as much sense as it would initially appear. It is generally inferior form a buckling perspective and it can often push shear load into flanges rather than webs. By using bolted connections you can much better control the load path. Even if it means having a nominal gap (10mm) between the two elements which would otherwise bear on each other.

On most projects, I rely on both connection types. I wouldn't necessarily consider one type better, at least based on the work that I do, which is mostly residential and where beam sizes are generally not excessively deep.

I'll often have a steel beam supported by intermediate columns. I use a direct bearing connection in this case, as opposed to splitting the beam up into individual pieces between columns. I typically use web stiffeners above the column, and I'd consider this an overall simpler connection versus two shear tabs.

Who cares about shear in steel design!

In terms of member strength for steel and wood, I almost never have a case where a beam fails in shear and not bending. For bolted shear tab connections, though, it's not uncommon to have issues getting a standard connection to work, especially since I deal a lot with W6, W8, and W10 beams where you can only fit 2 bolts.

 

[Eng16080]

Nice video, btw