Lateral/torsional buckling is similar to column buckling in that the beam has to start translating before a load is implied on the lateral support - it's a conundrum.  The load on a lateral support member is empirically assumed at some percentage of the actuating load, such as 2 percent (reference AISC Manual of Steel Construction, LRFD Second Edition, Vol II, page 12-22).  For a beam in flexure, the lateral support load is determined based on the maximum compression of the top flange.  Once the lateral support load is determined, you apply it as a lateral load perpendicular to the masonry wall and determine the capacity of the masonry to resist this load.

Hi, lextoo.

"How much support....?"  Probably more than you think.

Lateral torsional buckling requires simultaneous lateral and torsional deflections.  Without either one, that mode of buckling cannot exist.  

If you have a sound and reasonably wide masonry wall supported by a WF beam, then ask youself how easily can it deflect torsionally?  As soon as there is a tendency for the beam to rotate at any section (which it can only do by developing a gap under the wall at one edge of the flange), then the contact between wall and beam will become eccentric and the load from the wall will provide a self-correcting torsional moment.

Restraining a beam torsionally is every bit as good as restraining it laterally (but not normally as simple), and I would ignore all talk of resisting nominal lateral loads.

ie, I believe that a sound wall (about as thick as the WF is wide) should provide a sufficient continuous torsional restraint, and you should have no buckling problem at all.

Don't be too surprised if others tell you that this view is heretical.  Just ask yourself whether it is theoretically sound.  

HI again, lextoo.

Of course, what I should have added was that I was only answering your question as put - "how much support can you get from a grouted and reinforced wall".

During construction of the wall you have an entirely different state of affairs. Until the grout and mortar has achieved substantial strength, your "wall" will be merely a loose assembly of blocks, with weight but little structural integrity.  In that condition, you will get no buckling restraint worth considering.

If the lintel is to be supported until the wall has achieved structural capability, then that would not be a problem.  If not, then the lintel will have to be capable of taking the full load without any assistance from the wall.

yes, i agree with  bhub and austim-- this confirmation is
helpful to me...external/eccentric loads on spandrel
beams are a challenge...

there was a paper mentioned in "modern steel construction"
may 2002:  "Economical Design of Shelf Angles"  by
R. H. Tide  and N.  v. Krogstad --it sounds like it would
be relevant to this subject...

once again thanks to bhub and austim