bevel weld on bar pos

huangyhg

bevel weld on bar post
i need to design an eccentric welded connection for a steel rectangular bar rail post. this is for a rail on a stair stringer.
the bar post is being welded directly to a stringer channel flange.  the bar post is 1/2" x 1-1/4" and the width of the channel flange is 1-1/2".  thus, based on these dimensions, i do not have the room to develop the moment required at the base by using fillet welds (i.e. the thickest weld i can get on the short sides of the post is (1.5-1.25)/2 = .125" welds, which is not adequate).
so, i have determined that i am going to need to use a bevel weld in which a portion of the base of the bar post will be beveled in order to make room for the weld.  i am just unsure of how i determine how much needs to be beveled off.  do i treat it like a fillet weld and bevel off the width required to make the design work as a fillet weld??
i hope this question clear, i can draw a sketch if necessary.  
you can't just use welds on the sides of the bar to resist the post moment?  seems like they would be more than adequate.

jae,
i would really have to pile up the weld on the sides , since the weld on the side so much less efficient than the welds on the top ((i forget the exact requirement right now - not at the office).  i would like to know what my options are with the bevel weld.  either way, i may need to use it at some point in the future.
you could use a full penetration weld or just chamfer the ends of the post so the weld at the ends are at least as large as those on the sides.  i doubt, however, that the welding of the posts will be critical.  bending in the channel web will probably control.
the 1 1/4" shall across the 1 1/2" flange, it gives 2 1/2" in total weld length. if it is still not adequate, then you need to design side (web) mounted posts.
draw a sketch of what you are thinking of: something doesn't seem right.  when i build rails like that, i've never found that a 1-1/2 x 1/2 flat bar has any strength against bending.    a "section" of 1/2 x 1/2 pickets, or a "welded section" of ornamental iron castings isn't strong either: the sections have to be anchored into the base structure at the posts very, very firmly.
you're required to oppose a 200 lb horizontal force sideways at the top of the 36" (or 42") high handrail/guardrail.   your fillet welds up the side of the channel will be strong enough - unless you are trying to just use the "top" of the channel?   
how are you opposing the "rotation" force on the channel?  what size channel?   what size stringer, post, picket are you thinking of?  how far between posts?   how are you anchoring the posts?   commercial or residential construction?   
here's a trick that you can do. flush one side of the rail post to the stringer and chamfer this side of the post to deposit the welds. you'll end up with 1/4" of space for your fillet welds on the opposite side of the stringer. also if welds will be at 4 sides, you can use some of the welds at the other 2 sides to increase your capacity.
some of the posts here are right about the stringer flange. that might not calc out for bending and might need to be stiffened with a gusset plate underneath. the 200 lbs. requirement at the top of the hand rail/ guard rail, that doesn't need to be resisted by a single rail post though. assuming all the rail posts are connected at the top with a continuous piece of steel that acts as your hand rail/ guard rail and they are not spaced too wide apart.  
if i'm understanding your configuration correctly, you have a 1/2" x 1-1/4" flat bar serving as a handrail post for a stair.  you are welding the flat bar to the top flange of a stair stringer. the orientation of the flat bar is that the long dimension is parallel to the loading......
if so, your numbers are not good. your bending stress is somewhere around 55ksi.  understanding that many codes still allow a 1.33 overstress for handrails and stair sections, you are still using a lot of the capacity for any level of comfort, assuming you are using 50ksi yield strength material.  if you are using 36 or 33 ksi material, you're sunk.
your other option, assuming you can get the stress levels to work out, would be to provide a gusset on the back (stair) side of the bar, to increase the weld area and to reinforce the probable overstress condition.
a sketch would be helpful (as racookpe1978 noted) to see if we are making correct assumptions.
ron,
your assumptions are exactly correct as you have described it.  you are also correct in that the 1/2" x 1-1/4" flat is inaquate by itself as a bending   
abusementpark...changing the spacing might not help...you have a two-fold requirement, either a 200 lb lateral load at any point (worst case is at the top of the post), or 50 plf in any direction.  the spacing will help the 50 plf condition, but not the 200 lb condition, unless you can share the load quite a bit with a rigid top rail connection.  your end post will probably prevail.
the channel web is probably not a significant concern.  the tread supports (assuming typical configuration) will stiffen the web a fair amount.  a slightly smaller stringer section is common in this area (mc10x8.4).

i would change to side mount support - a tee with pipe sleeve to hold the post. the tee can be welded or bolted to the web.