development and hooks

huangyhg

development and hooks
rather than hijacking the other thread on this i have another hook/ development question. at the top of an external column of a concrete moment frame you get a top bar that needs anchorage into the column. i am taking the standard hook and anchoring that into the outside of the column. then i take the standard hook from the outside column bar and anchor that up into the joint. but really this is a splice of the outside column steel and the top beam reinf. so presumably the bars need to be spliced. how does a bend influence the splice length? i have also been comparing codes on this issue and it would seem that aci gives you very small lengths when you use standard hooks but quite large lengths for development whereas the british standard doesn't give you a great deal of advantage for a hook in terms of anchorage but the general case for anchorage gives you a smaller length than aci
seems like the vertical column bars would simply extend up into the beam depth and hook near the top of the beam with a hook developed past the bottom of the beam.
then the beam top bar would extend over the support and hook into the "joint area" over the column, past the inside face of the column.
this is what i've typically seen.   
normally a standard hook/cog only gives you 50% end developemnt. it does not provide full tension end development. it gives no compression development.
if you need to gain more than 50% end development in tension, you have to change the detail to a larger radius bend (so you lose the extra development from the hook) and use normal development length equations. you cannot get more than 50% with the standard hook radius as it creates too much compression in the concrete in the bend, so a larger radius is needed.
so, for your detail, you should not use a standard hook, but a larger radius bend for the bars from the column into the beam and then lap the beam bars to the column bars.
rapt,
       is this 50% from the aci code?
what rapt is describing sounds a bit like what is in the british standards. in fact the aci standard bends are what the bs would call large radius bends e.g 6d to 8d whereas for the bs the standard for deformed bars is 3d for all bar sizes. so for the bs the compression inside the bend needs to be checked but not for the aci. not sure where the 50% comes from. that is not from the bs. not being in the us i am intrigued to know if that means that all your bends there are those large radius bends. doesn't it cause problems where you have bends coming in from different directions and with large covers at end corners?
i disagree with rapt.  you cannot develop a vertical column bar by bending it and extending it horizontally into the beam.  i agree with jae--you develop the vertical column bar into the joint zone with a standard hook, and the beam top bar into the joint zone with a standard hook.  i've never heard of this 50% stuff (but then, i use the aci code--maybe this issue is treated differently in other countries).
you don't need to splice horizontal beam bars with vertical column bars.  the bars are developed into the joint zone as described above.
daveatkins
pca technical note on the topic. australian code calls up:
where a bar ends in a standard hook, the tensile development length of that end of the bar, measured from the outside of the hook, shall be taken as 0.5lsyt.

dave and jae,
you say you just need to anchor the bars into the joint zone, but what does that joint zone do with the force. surely a top corner of a moment frame is just a bent beam with a large moment at that corner. therefore surely the reinforcement has to be continuous over that corner.
phuduhudu,
i think you have a point - but i've never seen anyone do anything else with it in the hundreds of project designs i've either done or reviewed.
what you are describing is sort of a block of concrete with moment applied to two different faces adjacent to one another causing tension in the upper/outer corner.  i'll have to think on that one.
jae,
exactly. try to draw the strut and tie model and you will see that you get a compression strut going from the outer corner to the inner corner. this means that effectively the tension reinforcement has to be developed beyond that outer corner node which is more than just what you have left of half the bend and the straight end of a hook.
on the other hand if it has always been done as you say and there have been no major problems - why?
in intermediate floors there is column continuing above the floor with continuous steel and compression from the column so this is really only an issue at the top.
one aspect may be that there is less stiffness in a top column and thus less moment developed there.