location of pin support on beam

huangyhg

location of pin support on beam
hello everyone,
i'm looking at the design of beam that is pinned-pinned beam with a uniform load distribution.
the beam is built-up from a thin web with two extrusions fastened to the upper and lower edges of the web to form the beam caps.
the beam will be assembled to the rest of the structure by two pins.
my question is regarding the location of the two pins relative to the neutral axis.  if the pins are located on the neutral axis, then they would just see the vertical reaction load.  but if the pins were located in the upper chord of the beam, would they not see additional horizontal shear components equal to m/h (h = distance betwene beam chords)?  what if the location of the pins were such that they were above the upper beam chord?
hope the problem description is clear enough.  i look forward to hearing the opinions on this design detail.

check out our whitepaper library.
no, the pins are loaded with vertical load and axial load from the offset from na.  is this what you describe as horizontal shear?
yes, axial load due to the offset from the na... that is a better description of the "horizontal shear".
thanks civilperson, that is what i was looking to confirm.

does it really matter ?  i'd be more concerned about putting a pin into an un-reinforced web (you've described a thin web beam with caps but nothing about end fittings).
pinning the ends of the beam makes it as close to ideally pinned as you're going to get, abit surprised you're doing this instead of clipping the web of your beam onto the frame web where i assume you are.  pinning on the neutral axis is "ideal" but not imho truly necessary.  
the pins will not see "horizontal shear" they are thru a hole. do not mix your externals with your internals. if there is no axial load the pins carry wl/2
ok, i've attached a poor sketch to illustrate the problem.
the construction of the beam is irrelavent at this point, because i'm working out the loads on the pin so that i can properly size it.
the poor sketch shows three beams.  the first beam has the pinned supports at the na.  the second has a the pinned supports along the upper edge of the beam, and third has the pinned supports far above the beam.
now assume there is a distrubuted load, causing the beam to bend (up or down doesn't matter).  in case 2 and 3, i would recon that there is a significant axial force (horizontal force) incurred because motion of the outer most point is restricted (usually they would have some relative motion if the pins were on the na).

ash060-
i agree with civilperson. from a theoretical standpoint, the pins located above the neutral axis will see some tension due to the rotation of the ends of the beams about the neutral axis.
whether or not it will see it in practice is a different matter. probably not, as the oversize of the holes to allow for fit-up would likely allow the minor movement to occur at the top of the beam without affecting the pins.
sorry noor, that last pic only made me think about these pattern matching questions ... no clue what you were trying to show.
as to the axial force in the beam, i think if the beam is pinned on the neutal axis, then it's "properly" pinned.  if the supports are off the neutral axis then a horizontal load will develop (opposite at both ends) due to the rotation of the end face of the beam about the neutral axis.  however, i think this'll amount to roughly 0.5 of a gnat.  
a question back to noor, how is the udl being developed ?  we're questioning minor (imho) details about the end restraint but i wonder if the beam is attached to the fuselage ? (which would obviously change everything)
in common beam analysis, on of the supports is though of as a roller, so no tension develops.  if your supports are both truly pinned, with no slop in the connection, you would develop tension in the support even if the pin were at the neutral axis.  to get the tension for your cases 2 and 3, you could model it as a three member frame, with verticle   
rb1957, i know the drawing have much to be desired, but i didn't think they're that bad.  there are three beam configurations represented there (the long rectangles).  the triangles represent the pinned connections (the pin is at the vertex of the triangle).  also you're making assumptions about the application.  please stay focussed on the idealized beam problem.  
miecz, you're correct about the rollers, but in this case the pinned supports are rigidly constrained (not shown in the diagram), and the pins are close tollerance.  you're right that in this configuration the pins would see additional horizontal loads even if they are on the na.  like you say, the fbd can be drawn such that there are vertical   
ok pass on the pic ... how is the udl getting into the beam ?