Take a look at the FAQ on Cantilever Roof framing, as there's a 1988 roof failure that at least potentially involves this issue, though the articles on the failure do not capture it as a possible contributing factor as it doesn't seem to yet be under consideration as a potential issue.
Compendium of Design Office Problems, ASCE J. Struct. Eng, 1992, V. 118, issue 12, mentions inflection point as one of the problems needing to be addressed.
Compendium of Design Office Problems said:
"4.2 Lateral Bracing of Continuous Beam When does a point of inflection constitute a safe lateral bracing point for computation of the allowable bending stress in the bottom flange of a continuous beam?"
Lateral Bracing of Continuous Beams
Does a point of inflection constitute a safe "lateral bracing point" for computation of the allowable bending stress in the bottom flange of a continuous beam? Are there any other guidelines for this issue?
This question also shows up as early as June 1993 in Modern Steel Construction's Steel Interchange. (I haven't found any response through December 1994). (curiously nobody seems aware of the article in J. Struct. Div as to it being "unresolved".
1999 3rd Edition LRFD (University of Iowa link) contains language prohibiting the use of an inflection point as a brace. (See Davis, below).
1999 AISC Specification said:
For beams with double curvature, the inflection point can not be considered a brace point because twist occurs at that point (Galambos, 1998).
The Galambos reference here is the SSRC guide. Galambos, T. V. (ed.) (1998), Guide to Stability Design Criteria for Metal Structures, Structural Stability Research Council, 5th Edition, John Wiley & Sons, New York, NY
Does a point of inflection constitute a safe ‘‘lateral bracing point’’ for computation of the lateral-torsional buckling flexural strength of a continuous beam? Are there any other guidelines for this issue?
Response
A flexural point of inflection occurs in a beam where there is a change in curvature from negative to positive. Even though at this point the value of the moment is zero, it is not equivalent to a brace point.
The American Institute of Steel Construction (AISC) (1994) notes this distinction in the commentary (Section F1, "Design for Flexure," pp. 195–199) where it states that "the length between braces, not the distance to inflection points, and C[sub]b[/sub] is used in the resistance equation." Also, Ricles and Walsh (1993) note the necessity for providing bracing at the point of inflection.
Section B.6 of the AISC specification requires "at points of support, beams, girders and trusses shall be restrained against rotation about their longitudinal axis (AISC 1994). In addition, lateral or rotational restraint must be provided along the beam length if the beam is assumed to be fully braced. For continuous beams, restraint is necessary for that portion of the bottom flange, which is in compression. Restraint may be provided by the joists themselves depending on the relative proportions of the assembly. A procedure for calculating the restraint is described by Essa and Kennedy (1995) and Rongoe (1996).
Buckling of Beams with Inflection Points, Yura and Helwig, 2010, Proceedings of the Annual Stability Conference, Orlando, Fla., North American Steel Construction Conference, SSRC
