designing with channels

huangyhg

designing with channels
i am designing a cantilevered channel and i am looking for a methodology to determine the allowable bending stress in the channel.  
asd chapter f (page 5-45) says this chapter deals with "singly or doubly symmetric beams loaded in the plane of symmetry.  it also applies to channels loaded in a plane passing through the shear center.."  my situation does not have the load passing through the shear center, so i will have some torsion.  i have check the torsional stresses in the channel (which are ok).   
i am not sure what to compare the bending stress to because i have an unbraced length equal to 6' and i am not sure how to calculate lc to compare my unbrace length to.  is it as simple as 0.6fy?
sigh... asd 9th; takes me back to the days of big hair and shoulder pads.
from memory; there is the "channels bent about their major axis " equation that you have to figure your unbraced length into, equation f1-9 or something??. then look for an interaction equation combining shear, torsion, and bending.
or better yet:
haynewp is right.
bottom of page 5-46:
for channels bent about their major axis, the allowble compressive stress is determined from equation (f1-8).
also, +1 on the 2005 spec., lol.  deal with it for a while and the green book seems devoid of logic.
thanks, i must of read right past it.  i guess you look for something too long, its best to put it away and pick it up at another time.
cpbiv,
the unbraced length for lateral torsional buckling may not be 6' for cantilever case. it all depends on support condition at root and tip of cantilever.
also, please note that for the cantilever case it is more important to brace the top (tension) flange.
271828: i disagree that the green book is devoid of logic.  look at steel structures by salmon and johnson.  the green book is what you get when conservative and simplifying assumptions are made to the theoreticaly "precise" equations.
bjb, i've gone through all the derivations in s&j, etc., and understand where the simplifications come from, etc.
here's the easiest example showing why i think the newer formulation is more logical.  this is just one engineer's opinion.
say you're trying to check a compact, unbraced, w-shape for strong-axis flexure.
with the 2005 spec., you would:
1. see how your lb compares to lp and lr.
2. if lb<=lp, then phimn=phi*fy*zx
3. if lb is between lp & lr, you have the inelastic buckling straight line equation for phimn.
4. if lb > lr, then you have the elastic buckling equation for phimn.
an engineer immediately knows what type of failure he's investigating, whether it's yielding, inelastic buckling, or elastic buckling.  he can tell from looking at those equations exactly where they came from, where fs are applied, etc.  anybody can immediately understand what fcr, lp, and lr represent.
with the 1989 spec., you would:
1. see how your lb compares to lc.
2. if lb<=lc then fb=0.66fy
3. if lb exceeds lc, then we completely change slenderness parameters to l/rt.  compare it to the two limits at the top of page 5-47.
4. if l/rt is less than sqrt(102...), then 0.6fy applies, but one has to work at it to see exactly why.
5. if l/rt is between the limits, then use f1-6 (inelastic buckling)
6. if l/rt exceeds the higher limit, then use f1-7 (elastic buckling)
7. but wait, f1-8 can apply whether or not we have inelastic or elastic buckling!  calc that and compare to fb obtained so far.  use the larger.
it's impossible to see where the fs are applied in most of the equations.  it is totally impossible to look at those equations see where they came from--pure torsion, warping torsion, etc.  going from .6 to .66 is a covert way to go from s to z.  the asd formulation completely masks the underlying mechanics, decreasing physical understanding.  
just try drawing a flowchart of the two methods.  the 2005 one is a straight line.  the 89 one looks like spaghetti!
the asd formulation looks the least bad when we consider an easy example like this.  one has to really start mis-applying equations to handle more complicated cases.  the 2005 spec. has special sections for many of these.
anyway...that's just how i see it.  i've designed bldgs both ways and i've never understood why anybody would prefer the 89 stuff.
271828
to each his own i guess.  i too have used both in design, but still find the 89 asd easier to use, probably because i'm more familiar with it. the 89 asd is a design manual to be used by qualified engineers, it is not a text book.  therefore, the fact that the safety factors and underlying mechanics are hidden is not an issue, in my opinion.  
it's not my intent to debate the merrits of one spec versus another, nor am i saying that i don't like the new black book.  i just disagree that the green book is devoid of logic.  to compare the two isn't really fair, the green book is basically the 60's spec with a few tweaks, written for engineers whose primary calculating tool was a slide rule.  i've never used a slide rule, but i imagine trying to calculate equation f2-6 of the black book would not be pleasant.
bjb, ok "devoid of logic" is a bit too harsh, lol.  
i'd argue for "less logical," but i suppose that's all i could really do.
have a good evening!