HSS Beam Moment Frame

[BadgerPE]

I am retrofitting an existing facility and I am attempting to remove two adjacent building endwalls and combine them into one. Geometry dimensional restrictions abound so I am left with using an HSS beam to carry the torsion of unbalanced flange loading to the frame columns. The frame takes building lateral load as well. My question is related to the welding of the HSS beam to an extended end plate. The weld of the upper flange to end plate is required to transmit all tensile forces resulting from the moment, correct? Welds along the side should not contribute to the tensile resistance correct?My plan was to require a fillet weld around the perimeter of the HSS beam, but preliminary sizes indicate that a CJ weld may be required. See attached sketch.

 

[BadgerPE]

Forgot to mention that columns are W-Sections with strong axis oriented in direction of wind loading, so end plate is connected to column flange.

 

[structSU10]

That sounds like a conservative and acceptable way to look at it. You could always look at the total load and get a resultant force from the shear and moment, and analyze the unit force on the entire weld. A good reference for that check is from blodgett. Another reference for the end plate connection is AISC design guide 4. I don't know if it has anything specific on the weld in particular though.

 

[BAretired]

I don't agree that the side welds contribute nothing to tensile capacity, but the column may probably needs stiffeners cut from HSS material.

BA

 

[msquared48]

Conservatism is good here. Assume all the tension goes through the top flange, but reinforce the column to cover the situation that some of the tension does get transmitted through the sides too.

I agree with BA that this is most likely the case in the real world.

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[JoshPlumSE]

AISC design guide 4 (end plate moment connections for wind / seismic) only covers wide flange connnections. However, they have a requirement that the web weld must develop the strength of the web in tension (not shear). I've interpreted this to mean that as the moment goes into the inelastic range and approaches Mp (or Cpr*Ry*Mp) the web welds need to be sufficient to cause tension yielding in the web. Otherwise, you risk initiating a weld fracture at the top of the web.

That rational may not apply as directly to HSS moment connetions. Especially if this is not a seismic application where you expect to get significant non-linear behavior in your moment connection.