Shear-Only End Connection for Steel Plate Girder 1

[Arnab Mukherjee]

I am Designing a Plate girder, with hinged ends (shear force only - No moment transfer) This plate girder is connected to the face plate of RCC beams at the ends. The connection is done with fillet welding on the web only. the flange is discontinued 25mm from the support. The basic concept being that the flange takes moment, and no flange at the end means no moment-(only shear).
Are there chances that in spite of this, the web will take some fixed end moment and transfer the same to the supports?
Kindly let me have your comments.
Thanks

 

[HTURKAK]

Fixed end moment is something else .. ( Fixed end moments develop for the cases; the beam supports are fixed support , the continuous beam at mid supports with symmetrical loading , the beam is cantilever . That is , there shall be no rotation ) In your case , the connection may be considered semi- rigid which can transfer limited moment or for simplicity can be assumed simple joint.

it is not clear from your description if the steel girder is a cross beam btw RC beams.

A sketch with some details would be helpful to get better responds .




Use it up, wear it out;
Make it do, or do without.

NEW ENGLAND MAXIM

 

[271828]

The shear connection will transfer some moment to the supporting concrete beams. If you're treating this as a shear (only) connection, then it should be designed to minimize this moment, though.

The following screenshot shows a common choice for this situation. Avoid welding across the tops of the angles. That way, the angle outstanding legs can bend out-of-plane a little and accommodate the simply-supported beam end rotation. The welds can be L-shaped with weld across the bottom of the angle if needed.

The AISC Companion to the Steel Construction Manual, Vol. 1, Example II.A-2A is for this condition.

https://www.aisc.org/publications/s...ion-manual/manual-companion-for-16th-edition/

 

[dik]

I use the same detail as 271828... and extent the weld across the bottom of the angle. I use a similar one if only a single angle connection. It allows a degree of rotation and is slightly less rigid. I review steel connections and have a '*.pdf' stamp to add to the drawings reviewed. Drawings I review usually have full weld on three edges.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[SE2607]

If truly a simple connection (no moment), how does that connection accommodate the rotation at the end of the beam. Are the bolt holes slotted horizontally?

 

[271828]

There will be some moment at this connection. It's hard to come up with a practical zero moment connection. We typically don't need to eliminate the moment, just reduce it to low values. In AISC that's called a Simple Shear Connection. The moment-rotation curve needs to look like:

Because the tops of the angles aren't welded to the support, the outstanding legs bend out-of-plane, allowing the tops of the angles to move a small distance away from the support. See below. The inside legs often have horizontal short slots, but that's now what we're counting on for rotational flexibility.

 

[JoshPlumSE]

Are there chances that in spite of this, the web will take some fixed end moment and transfer the same to the supports?

Well, most shear connections will have to transmit a small amount of moment into the connection. AISC has a whole concept (rotational ductility) that helps you determine whether your connection is flexible enough for you to design it as a pure shear connection. Refer to the following sections of the AISC manual (16th edition) for more information:

Double Angles: Page 10-9
End Plates: Page 10-22
Shear tabs: Page 10-49​

 

[hokie66]

The OP said that the connection is to some kind of 'face plate' on a concrete structure. It would be nice to see a detail, as the effect on the concrete may be the most important part of the decision making.

 

[271828]

Agreed with hokie66.

The double angle shown above is a good choice when it is important to minimize the moment.

In contrast, a single plate shear connection has much less predictable flexibility, even when the flexibility/ductility checks are satisfied. I would not use one of those when it is important to keep the moment low. For example, a connection to an embed plate might be controlled by the stud strength. It would not be good to have high or unpredictable tension in the top studs.

 

[dik]

...by having a weld length equal to twice the weld thickness at the leg of the angle at the top. This allows the angle to 'flex' between the beam web and the leg of the angle.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik