Steel moment connection - which one? 3

[n3jc]

Hey guys, i need a little help with this.

Im wondering which moment connection should I choose and why.

All other connections in a model are considered as pinned.

Im also open for other type of connection if you think it would be better.

Steel construction is supported / braced in X directions by existing building.
In Y direction it is braced by steel bars (roof plane).

 

[Lnewqban]

It seems that you are only asking about A.
I would suggest the second option.
Being dimensions of plates and bolts equal, the second option has the friction between the plates and the resistance to shear of the bolts resisting the weight and the wind uplift loads.

"God will not look you over for medals, degrees or diplomas, but for scars." - Elbert Hubbard

 

[kingnero]

Seeing as the shear strength is only ~ 60% of the tensile strength, and I wouldn't take friction into account when you're not explicitly talking about slip-resisting connections, it appears option A would be the better choice when resisting vertical/uplift forces. Also, erection is much easier when choosing option A.

http://www.fusionpoint.be

http://be.linkedin.com/in/fusionpoint

 

[Stenbrook]

I would say the first option is better and more likely to happen. In general for one story steel structures, I have seen that the majority of the time they will detail the structure to have the beam sit on top of the column. I also believe it is easier to build that way. I don't think you can go wrong with either one though as long as you design it correctly.

 

[MotorCity]

I vote for the beam running over the top of the column. There will be less demand on the bolts since they will primarily resist tension. In the case where the beam is connected to the column flange, your bolts need to resist shear and tension. Also, in the case where the beam runs over the top of the column, there is a reduced chance of slippage of the connection.

 

[n3jc]

Tnx for answer guys. Also Id like to know, where/how do you determine the point of rotation (for bolts design)? Is my assumption alright?

 

[canwesteng]

No, your assumption is not correct. Compression should occur in the compression flange of the column, but you can still use that third row of bolts to resist shear.

 

[dik]

Typical pre-eng buildings in this area have a level plate at the top of the column with the 'rafter' bolted to this. Bolts do not normally extend beyond the flange of the column on the inside.

Can you consider running the roof 'purlins' over the top of the rafters and cantilever a tad to reduce the mid span moment? Connection is generally cheaper and easier. Can you use CFS channels in lieu of rolled sections?

Dik

 

[Nor Cal SE]

I agree with those who suggest the beam running over top of column choice.

One more relevant note: if you have any difficulty getting the beam-col detailing to work as a moment connection, or are unsure about that in any manner, you could switch gears to a cantilever column system. Because the 'R' value is smaller, your lateral load will be higher (for seismic), and your base connection and footing would need to resist the effects of rotation about the base. But if your structure can sustain this, then it might be the easiest way to go.

 

[wannabeSE]

In the original post, it says the new structure is braced in the x-direction by the existing building. If the existing building can handle the new loads, why are moment connections needed? Does it save much in the beam size for resisting the gravity loads?

 

[n3jc]

wannabeSE - good point, was thinking about that... because the existing building provide lateral support to steel construction, there is no need for moment connection in X direction. Sure, with moment connection there are smaller bending moments and deflections, but in that case this is not problematic.
But it makes me think what happens to a steel construction when existing building deflects/moves(earthquake)for lets say 2 cm (in X direction). In that case steel columns act like cantilever (if supports are fixed of course, but what if they are not designed that way)?

So I think I can make fixed supports at the bottom of columns and pinned at the top (instead of moment connection). What do you think about that?