Uniform Force Method Question 1

[shepherd]

I am reviewing calcualtions for a braced frame connection. The contractor took the liberty of fabricating and erecting the steel, before we even recieved design calcs or shops drawings. The connection is what is typically shown in AISC heavy diagonal brace connections where a beam, column, and diagonal brace meet at a single work point. The connection is very similar to what is shown in AISC 2nd ed. LRFD Vol 2, Fig. 11-7. The gusset is bolted to the column and welded to the beam. The beam is bolted to the column.

My issue with the calculations are that the gusset to beam connection is detailed in such a way that alpha and alpha-bar are not equal (ideal centroid for no moment on connection and actual centriod with respect to column flange). This in turn induces a moment on the gusset to beam interface (which is indicated on the calcs i am reviewing). My thinking in looking at the free body of the beam is that this also induces an end moment in the beam as well that we did not design for. The connection designer is stating that the method takes into account this eccentricty and is not affecting the beam. The only thing that I can think of is, that perhaps he is assuming that this moment at the beam/gusset interface is resisted by a couple created at the beam/column connection and gusset/column connection, but it seems to me that the load path still needs to go through the beam first. And the horizontal forces on the beam/column and gusset/column connection do not seem to indicate this.

I would appreciate any comments that anyone may have regarding that if there is a moment induced at the beam/gusset interace, then this needs to be resisted by a moment in the beam. Thanks in advance.

 

[VirtualEngineer]

Shepard,

Sometimes I've determined the optimum alpha location and then consider an effective lengh of weld which maintains a coincident alpha-bar.

If that effective weld is sufficient, then the loading condition is satisfied.

If that doesn't work, you may be able to design it similar to Special Case 2 and design extra moment at the gusset beam interface.

Good Luck


JPJ

[Thumbsup2]

 

[shepherd]

JPJ,
Thanks for the response. The connection has been fabricated and the calculations do indicate there is a moment at the gusset beam interface (alpha not equal to alpha bar). The weld and connection has been designed for this moment;however, the beam (designed as part of a vertical truss) was never designed for any additional end moment. Am I missing something, does the moment not get transfered to the beam as well? As you point to special case 2, would you not have to design the beam for the moment (Mub)shown in the beam free body?

Thanks
Shepherd

 

[aggman]

The gusset must be designed for the moments, as does the welds and the beam. You have to verify that the beam can carry the additional moment induced and also handle the concentrated loads. On page 11-45 under Solution B (Special Case 1) at the bottom of the page states the following where a moment (Mub) of 145 kip-ft is applied to the beam. "The moment, as well as the beam axial load Hu = 149 kips and the moment and shear in the beam associated with the end reaction Rub must be considered in the design of the beam." Seems pretty cut and dry to me. I almost always have some sort of gusset interface moment force, and I always check that the member can handle the additional load. Now I would say that you, or the connection designer, needs to decide if the connection/ beam will fail or not. You may also decide that some of the moment will be transferred to the column and will allow the connection to be satisfactory. Also what I do to calculate the true moments and shears is draw a free-body diagram about the beam to column vertical connection. You can then sum forces to calculate actual applied moments, etc. Say for instance you have a compressive brace force applied to beam. The horizontal shear Vub can either increase or decrease the moment induced in the beam by the vertical component and the moment.

Good-Luck.

 

[shepherd]

Thanks aggman for pointing me to that example. The calculated moments induced by the designers assumed geoemtry are quite large, and we intended the design to not induce additional loads on the member. We'll either need to verify the induced moments can be handled, if not, reinforce accordingly (it has been erected without approval).

 

[shepherd]

After a little more study, i've found the shortcoming in my thinking. This moment at the beam/gusset interface is in fact counteracted by the eccentricity of the horizontal beam force at the interface and the eccentricty due to alpha not equal to alpha bar. The beam experiences no net moment; however, it must be checked for local affects from the moment at the beam/gusset interface on the beam web. Thanks to aggman and Virtualengineer for their responses.