Continuous Beam Unbraced Length

[MatthewC]

I would like to know if anyone has any information on calculating an unbraced length when a beam is continuous over a column, and only the top flange is braced by framing members. It is my understanding that the inflection point of the beam should not be used for an unbraced length of the compression flange, but then what should be used? It seems to me that the beam-to-column connection could be used as a brace point if web stiffeners are provided, and the cap plate/bolts/column are designed to prevent twisting of the section. Also, any comments on the applicability of Cb in these conditions?

 

[JAE]

MatthewC:
This has been a big bugaboo issue for AISC for some time and they have consistently refused to directly address it in their Specifications. By intuition, or logic, the "correct" procedure is to use the full length (column to column) in your case, and calculate the Cb based on that full length (using the 1/4 point moments across that full length). Cb will generally be greater than 1.0 and you can then calculate the phiMn value for your beam and compare it against the NEGATIVE moments at either end for your capacity check.

For the POSITIVE moments, use the unbraced length between joists (I'm assuming joists here because you said that the top chord was braced but not the bottom chord) and a Cb = 1.0 - or you could go to the trouble of calculating a Cb over the short top chord unbraced length for the section between joists at the max. positive moment...but that is work that is usually not required....just use Cb = 1.0.

In past years, many engineers didn't use the above method due to the fact that Cb was calculated differently and many used the inflection point to column distance as the unbraced length - perhaps adding 20% to be safe...and ALWAYS using Cb = 1.0. Some papers have indicated that this is not always conservative but I have not seen anything published to prove it. Many beams have been designed through the years by this method.