beam seat angle

huangyhg

beam seat angle
this is a general question(s), not specifically related to any project.
1.  in a typical beam seat connection, with a steel beam framing into a column using a single angle (although frowned upon) at the bottom of the beam bolted to the column.  the bolts from the angle to the column are designed for shear and bearing, no requirement for tension from the eccentricity of the beam load on the angle to the column, correct?
2.  in the same type of connection but loaded in a tension type scenario, i.e.: drag strut, e.t.c., the tension on the bolts is considered along with the prying action if applicable?
3.  combo the two above situations.  if #1 and #2 are true i.e: there is both a beam reaction and tension force on this connection, they would be calculated separately.  #1 would have no effect on #2 and vice-versa with the exception of the tension/shear interaction of the bolts.  the tension in the bolts would only come from the direct tension of the drag load and prying, not from the eccentricity of the beam reaction, right?
this seemed relatively clear when i was typing it, but if not let me know!
also, any references or design examples/problems are welcome.
check out our whitepaper library.
aisc does have seated beam connection charts set up for your situation for types a thru f. now lets say you have a 4x4 seat angle. i would place the centroid of the beam load or its eccentricty 2 inches off the face of the angle. this does in fact induce a bending moment and thus tension and shear forces would be combined in the bolts.the upper bolts will have prying forces plus the beam shear load. this case is exactly the same as if you were welding the angle. the load would be outside the plane of the welds for which aisc has a specific chart for this special case.
thanks cap.  are the charts in the asd book?  i have seen charts in there for seats, but nothing that really addressed tension on the bolts.  is there a supplement with calcualtion examples or am i just over looking something?
i have the basic steel design textbook by galambos and johnston which very nicely shows the shear force distribution on the bolts. max. tension is on the top bolts to compression or bearing on the lower bolts. the formulas are very straight forward and are very easy to follow.
be very careful applying load conditions not initially assumed in the determination of the failure mechanism model in combination to what was assumed.  
in answer to no. 1, yes it is a typical assumption, and one that the aisc manual tables are built from, that the bolts are designed for pure shear loads (no eccentricity).  this is based on the thought that the angle is flexible enough to provide beam end rotation and the outstanding leg itself is designed for the required bending moment.  
i would be hesitant to use this type of connection with a large tension load however due to numerous factors including the large shear lag effect on the beam.  if you were determined to do so note that tension stresses would combine with flexural stresses in the angle legs and the design should be performed accordingly.  
if your using this connection for both a shear load and a moment say from wind forces use double clip angles on the beam web for the shear with top and bottom angles for the moment. play it real safe when it comes to interaction effects of steel connections. most catastrophic structural building failures come from connection failures and not the beam failing from bending stresses or deflecting to much. be very careful as noted by willisv.
thx willis.  i think i am almost there.  it appears that i was right with #1, the tension is to be ignored.  with #2 also being true, i am still left with how to handle #3.  fyi, this is all steming from a problem in the book "six-minute solutions" for the structural i pe exam.  there is a question, #52 (for those of you who still have your book!) in which a beam seat, like the one i am describing is welded to the column.  they suggest to take into consideration both the shear and the moment induced by the beam reaction.  from what i have seen in the asd book, i am not really agreeing with the solution and would have only considered shear.
willisv,
i think you are correct that for angles bolted to the columns, the eccentricity of the load on the angle-to-column bolts is ignored.
but for welded seats, the eccentricity is included.  
this is stated in the 1986 aisc journal article by garrett & brockenbrough, 2nd qtr. which forms the basis of the seat connection tables in the aisc manual.
.....i agree with jae and one reason for this is that the line of action of force between bolts and a weld on a beam seat are somewhat different.  if you consider the bottom of the angle as the rotation point, a couple through the bolt line yields a relatively small tension in the bolts.  now consider that you have a weld at the top of the angle .... the resulting tension in the weld can be almost twice the bolt tension for the same eccentric load.
for your #1 condition, particularly if the bolts are pre-tensioned, you can consider that all bending occurs in the angle through the bolt line, thus the resulting tension in the bolts is nil and shear prevails.
while it is actually a bit more complicated than the simplified approach model (one significant factor is the relative stiffness of the angle), the error is not large when considering the simplified approach, thus making it much more useful.
jae - agreed...but the question was about bolted seats hence me answer =p.  and actually if you look in salmon and johnson they do consider bolt eccentricity...though since the aisc tables were not formulated that way i do not think it's necessary.  
okay, for welds and looking at page 4-35 in the asd, about half way down "computed using traditional vector analysis".  so tension is considered.  plus ron's comment makes complete sense (thanks ron!).  so does everyone agree that aisc does not address a direct tension and reaction on a bolted seat angle?