designing a transfer beam

huangyhg

designing a transfer beam
please see attached image.
i have a transfer beam supporting 5 storeys of precast wall panels. i am curious if this transfer beam should be designed for the load of the first storey wall only given that the other storeys will behave like transfer beam themselves expect for the first one because of that large door opening or if i should design the transfer beam to carry the self weight of all of the above storeys? my engineering judgment tells me that the transfer beam will only support the load from the first storey, but what does yours tell you?
also, looking at the plan view, will torsion have to be accounted for? will this beam want to twist in torsion?
clansman
if a builder has built a house for a man and has not made his work sound, and the house which he has built has fallen down and so caused the death of the householder, that builder shall be put to death." code of hammurabi, c.2040 b.c.fff">
i think the openings in the stories above are too large to count on the walls as 'transfer' beams at each level.  depending how much meat is in the wall above or below the opening, the precaster can potentially design and reinforce that segment as a beam, but just counting on the wall to do isn't realistic, imo - assuming there is any substantial load on them.
openings account for about 30% of wall area and the walls only support their own weight.
clansman
if a builder has built a house for a man and has not made his work sound, and the house which he has built has fallen down and so caused the death of the householder, that builder shall be put to death." code of hammurabi, c.2040 b.c.fff">
the upper wall panels can be designed as vierendeel trusses to carry their self weight.
if you removed the transfer beam entirely, then added two horizontal ties between columns, one aligned with wall 1 and the other aligned with wall 2, the upper walls would support themselves with nominal reinforcement.  failure, should it occur, would likely be by excessive bearing stress or poor load transfer from wall to column.
accordingly, i think i would design the transfer beams for the full shear from all of the upper walls and, because of the substantial wall opening, would probably consider all of the upper wall load uniformly distributed on the portion of wall each side of the opening.
best regards,
ba
looking at your elevation, i think the precast panels are going to arch and strut most of their load straight down to the foundation.
regarding torsion, if the load is applied eccentricly then the transfer beam will go into torsion and checks must be done.
i agree with asixth on his first point.  
on his second point, torsion will occur over a very short length of beam, namely the portion crossing the interior columns.  looking at the geometry of the structure (assuming it is drawn to scale), i don't believe that will be much of a problem.
best regards,
ba
as the walls are not loaded by floors, i would design the beam for the full load of the walls, but wouldn't worry about deflections, secondary effects, or torsion (except to use closed stirrups).  the beam reinforcement should still be nominal.  this is quite conservative, but saves a lot of checking of the walls.
hokie66,
i agree we should not worry about deflections, secondary effects and torsion but i think that the transfer beam should be capable of carrying the full shear from all of the walls above.   
best regards,
ba
yes, i agree.
clansman,
the load has to get to the columns somehow. you cannot just assume that the walls arch, you have to provide the load path and internal forces to make it happen.
this has to be done either at each level or at the bottom, it is up to you. each panel (assuming they are single elements) can be designed and reinforced to span the full length to the columns or a transfer beam can do it. but you have to provide the reinforcement and connections to make it work.
but something has to carry the forces to the supports, and "arching" forces are not it!