Depends on the load on the column (axial, moment, shear)as well as the location.
I would first suggest that you analyze the column with the portion of bent flange missing. Include the eccentricity that is developed from the CG of unbent column to the CG of the column you are analyzing including the additional moment developed by the axial load times the eccentricity.
If it checks out OK, then no reinforcing is necessary unless you just want to provide additional protection against damage.
If the damage is at free end, it won't make any difference to leave it as it is (may be corrosion problems at worst). In other cases, it increases stresses locally and may affect on stability.
- Has the center line of the column been moved by the impact? Maybe it is not only the flange bent but the whole column distorted. In that case there is now an excentric load, P-Delta efects could be significant.
- How badly affected is the flange. How thick it is? Half an inch is not the same on a heavy section (in which rolling ofr fabrication tolerances are above that anyway) than on a small thin section. If the sectional properties have been compromised (reduced inertia) then the capacity of the column to buckling has been reduced.
- Is the bent local or more gradual?
- Is the column subject to any bending moments? If so, then, with reduced I, reduced capacity.
Half an inch does not seem significant, but it depends of your section size, column length, loading, ... I can not imagine anybody saying: '1/2"? That's all? Not an issue' without consideration to the specific conditions.
If you're in a high-seismic area or you need a lot of ductility in the connection, the buckle can act like a notch and make low-cycle fatigue more of a problem, or a source of fracture.
Besides, if you're hired to look at a damaged column and do nothing, if anything ever happens some lawyer might say it happened because you didn't tell them to repair the column. Even if it works I would do something to it to cya.
kelowna brings up the aditional issue of general displacement which can add to the eccentricity. Another situation that may reduce the capacity is if it is twisted.
but aren't you just as skrew'd if something negative happens in the future and you did repair it in some way ...
in fact, i'd've thought it made matters worse ... "you calculated that my client didn't need to repair the beam, you repaired it, charged him for unnecessary work, and it still fell down ..."
once the lawyers start, the only upside is not going down as far as you might have.
I would recommend a repair, and design a repair that unequivocally restores the original strength, and show that by calculation.
That way, if there's an issue in the future, you are covered, and it is the original design that was at fault.
I bet that if you can show by calculation, that a 1/2 inch of local bending / twist / warp, whatever you want to call it still checks out as OK, you've probably omitted a crucial calculation, unless the loading conditions have reduced from the original design.
