design of column top plate

huangyhg

design of column top plate
hello!
i am looking for design of steel w column cap plate?
how to size the thickness of plate? any help greatly appreciated!!
check out our whitepaper library.
do an elastic analysis and keep the elastic stresses below some allowable (say 0.75fy), or you can do a yield line analysis to use with a factored load.
i would be interested to hear others' opinions as a yield line analysis does require a good bit of deflection and at that point the beam flanges may be helping or redistributing the load closer to the supports instead of uniformly over the plate.
thanks for quick reply! but i am looking for a quick check on bearing as well as punching shear as the beam from both ends rests on column top plate !!
punching shear?  i've never heard of checking punching shear for a steel plate.   
as a side note, before performing any fancy analysis, i would first take a look and see how much of the column below the cap plate aligns with the beam above the cap plate - if that area works in bearing (1.35fy*this area for ultimate strength) then it doesn't matter how thick the plate is - the force can be transmitted directly to the column.  if that doesn't work (and probably even if it does), i'd use a 3/4" cap plate and call it a day.
willis,
that's only true if the beam web lines up with the column web, right?
if the beam web is orthoganol to the column web and the flanges of the beam are above the outstanding flanges of the column, i don't think you could do that (at least not without checking the flanges for bending).
beam web is 90 to column web in this case. but as willis said i would probably go for 3/4" thk plate! any comments??
yes, i am investigating some existing cap plates right now which are grossly overstressed (on the order of 600%!!!) based on an elastic analysis.  it drops down to around 200% using a yield line analysis.
what willis is talking about is great for the bearing look, but you still have to make sure the load can get there.  if the beam is coming in perpendicular to the column web and doesn't run over top of the column then you have little bearing area.  addditionally, the load is going through the beam in bending and shear, getting into the cap plate from a line load over the beam web (or a uniform load over the area of intersection between the beam flange and cap plate), then into the column through bending and shear in the cap plate.   
if there is no extreme loads, usually i call a plate either equal to the sitting beam flange, or slightly thicker. for the worst case scenario, place a concentrate load (beam reaction) at the edge of the plate mid way in between column flanges. this is quite conservative, too expensive for more than a few plates required. for such case, i think fem is justified.  
first trial, match the column flange thickness when stiffeners are used between the beam flanges.
can you throw some light on allowable bearing and punching shear in steel!!