Steel beam end plate connections 6

[phuduhudu]

I am wondering about the design of standard steel end plate and cleat connections for simple construction. I understand that the connections are designed to allow rotation of the beam ends and this principally happens by flexing of the end plates at the top of the connection between bolts. Hence the use of thin plates and large cross centres between bolts on the end plates.

All the design guidance I have seen refers to examples with pure shear for beams and then examples with pure tension for vertical bracing systems. However in most structures many of the beams are also part of the vertical wind bracing system and so they also have axial forces. Would the flexibility in these connections under axial loads render them inappropriate for taking significant axial forces. If so what can you do short of creating a fully fixed connection?

 

[Lion06]

You would typically have a gusset plate to take the horizontal load out of the beam and get it into the brace.

 

[phuduhudu]

Yes, but what about where you have the vertical bracing moving across a bay and you need to transfer the axial forces through two beam end connections across a column. Another case I have is a cantilever column on one side which takes quite a bit of lateral load and the braced bay is two bays across so again the steel edge beams need to take the axial forces across

 

[Lion06]

Can you post a sketch?

 

[phuduhudu]

How about this.

http://files.engineering.com/getfil...4fd9-a942-7ee8a4b51c44&file=Steelworktemp.pdf

You see how the wind shears from the top level have to travel through the beam end connections at the mid level to go across to the next bay.

 

[Lion06]

If wind is coming right to left, it goes into the top beam, through the brace, into a gusset and straight into the mid-floor beam. If wind is coming left to is where I see a potential problem. Is it possible to provide adequate connection to the mid-floor diaphragm to get the roof shear back into the mid-floor diaphragm and drag it back into the frame beam in the bay behind it?

 

[phuduhudu]

No, there is a void behind the left frame so no diaphragm. Also this is an open grid structure so there is a limited diaphragm consisting of horizontal trusses.

Even with those horizontal trusses where they span across two bays you end up with beams needing to take tensions through their end connections. Surely this is not such an uncommon occurrence.

The only guidance I found was in a book on connection design that said that the bracing cleats where they fix to the column and beam should not have unduly large flanges as they tend to restrict end rotation. They then show pictures of extended beam end plates which extend above and below the flanges with bolts all the way up and down. That seems like you would really end up with a fixed end connection for your beam.

 

[DaveAtkins]

When I have faced this situation, I have used double angle connections and checked the angles for prying. Usually I end up with 5/8" thick angles, the maximum thickness allowed by AISC for a pinned end connection. It almost always raises eyebrows, but I don't know another way to do it.

DaveAtkins

 

[youngstructural]

I'm with Dave on this one... I've only seen this once in practice and was happy it was not my job, however a similar solution was used.

I have seen a quite old factory building where a custom bracket was used: Picture a semi-circular block that fit into a C-shaped socket. Axial load yes, but free to rotate, and as such quite litterally a "pin". It was in drawings, but was apparently built in Boston in the late 1890s.

Weird how problems can just stick around; Then again it is a physical reality to be faced, so I guess the laws haven't changed with time. ;0)

Cheers,

YS

Cheers,

YS

B.Eng (Carleton), P.Eng (Ontario), MIPENZ (Structural-New Zealand)
Working in Canada, and missing my adoptive New Zealand family... at least I brought the little Kiwi with me!

 

[PUEngineer]

Am I missing something here? This condition occurs frequently in simple braced steel buildings. A typical braced steel building with roof deck and perimeter beams would have this condition if the deck in the bays bracing doesn't have the capacity to get the force into the beam of that bay. Hence a drag strut condition. This could also be possible in a braced steel building with non-composite floors where there would be a floor beam required to drag load into the bay of bracing. These connections would have axial load through a shear connection.

I think Dave's response about double angles is appropriate. But if the forces are small then a single shear tab is good for some tension force along with shear. Maybe my confusion is in the end plate connection that you describe, but wouldn't you have tension and shear in the bolts, along with some prying action?

 

[nutte]

You do get tension and shear in the bolts, along with prying. This is an extremely common connection, and I see it used a lot. I don't understand the hesitation.

Perhaps there is a disconnect between the theory that the plate flexes, providing slight rotation, but it is not enough to concern me.

 

[phuduhudu]

Thanks for all the responses. I am glad others agree that it is not such an uncommon condition. Which is why I was worried when I looked at all the guidance that I couldn't find a single example of a nominally pinned end connection taking axial forces as well. Hence you never see that prying check done in these examples. I will simply design as usual with combined shear and tension on the bolts and then do the prying check in addition.

 

[csd72]

phuduhudu,

If the transverse load is small compared to the axial then I have typically designed it based on the axial loads since the rotation requirement is small.

I always try and get the bolts toward mid height of the beam to minimise rotational restraint.