axial member compression

huangyhg

axial member compression
i'm designing a welded connection for an axially loaded member.  a solid reference that i'm using says "in the design of welds connecting tension or compression members, the welds should be at least as strong as the members they connect"
i think that means that if the tensile capacity of my member is 100 k then the capacity of my connection should be 100 k.  others at my workplace think that the connection should be designed for the calculated load in the member and that the sentence above just means that the weld material (ie. fy & fu) must be as strong as the base metal.
what are other people doing?  are you guys designing connections that develop the full capacity of the connected members??

acoording with my knowing your workplace have the reason your have to design the welds take in consideration the stress.
adam,
by far this is not the standard, but what i do is when slenderness controls i design a weld, especially in a low seismic zone or for small wind forces, what i do is design for 1.5 to 2 times the actual load.
when the stength of a member controls, i design for the capacity of the   
adam,
for all seismic design the philosophy is to design your structure to behave inelastically during the earthquake but not fall down.  to do this, codes are written to force you to design elements that can easily go into an inelastic deformation (such as a brace member, a beam in bending, etc.) for the required seismic response load.  for elements that tend to be more brittle (such as connections) that cannot become inelastic without a quick failure, the code compels you to design to a higher load.
the aisc required connections in concentric braces, for example, to meet the smaller of three forces:
1.  the full tensile capacity of the brace.
2.  the seismic load multiplied by a seismic force amplification factor.
3.  the maximum force that can be dragged into the brace or connection (for example - a brace may be able to take 100 kips of load but the floor diaphragm may only be able to drag 80 kips into the brace before it fails.  therefore you only need to design to the 80 kip limit).
the concept of designing the welds to at least the same level of the   
adamp - are you welding a splice?  if so, i would say that the weld must be equally as strong as the section.  if you are welding a beam to a column (or similar) i would say the connection does not need to be as strong as the column.
say your welding a w8 strut to a w14x455 column, i wouldnt think the capacity of the weld needs to be that of the 14x455.
the two criteria stand somewhere and in my opinion it is the enforceable code which must say how to proceed, what, as jae elaborates, can mean other cases as well.