|
strip method for two-way slabs
the strip method (by hillerborg and wood & armer) seems like the most convenient way to analyze a two way slab with irregular dimensions or boundary conditions (aside from fem modeling).
i have a few questions about the method for those who use it in their practice:
does aci 318 allow strip method for analysis? the two other methods -- ddm and equivalent frame seem too restrictive for irregular slabs.
i find the strip method convenient, but it requires that one makes seemingly arbitrary decisions about apportioning loading and moment between supports (e.g. fixed on one side, strong band on the other). how do you justify dividing the total moment between negative and positive (usually a ratio of 1.5 to 2.5 seems to be used) moments?
thanks
check out our whitepaper library.
i assume that you are referring to pt concrete?
mike mccann
mmc engineering
@msquared48:
no, i mean reinforced concrete.
so noone is using the strip method these days? a coworker referred me to exactly what i was looking for in moments and reactions for rectangular plates by w.t.moody. the tables give coefficients for finding moments and reactions for plates of simple shapes and load distributions. the booklet is found here:
the tables you are using are very similar to what can be found in roark's formula for stress and strain.
aci allows you to do whatever you want, provided you have a rational analysis and you can convince a code official. same applies to all design codes.
if you have sized the depth of slab and steel reinforcing required based on elastic moments of plate then you shouldn't have any problems trying to convince a code official.
every time i am designing a slab which isn't regular, i always perform some form of finite element analysis just to make sure i have made a valid assumption of how the structure is going to behave. sometimes it can be very difficult to make correct assumptions on the stiffness of load paths, particularly when dealing with a bi-dimensional element such as a slab.
@asixth:
do you only use the moments from your fem slab models or do you also accept shear and torsion values? from my experience, people tend to derive the shears for a plate from tabulated coefficients or from some other hand method, rather than using, say, sap2000 values. and yes, roark's is another source for plate stresses.
thanks
yes, i redistribute mxy to mxx and myy as required. for the plate that you have described i would look at the support reactions and use that value in the shear design. but i would only perform this for a non-regular structure that is difficult to idealize into design strips.
i've been interested in hillerborg's method for some time. i have both of his books on the strip method which is quite elegant.
i have similar concerns however:
1) how to safely apportion the span / support moments. this is straight forward for slabs with fixed / simple boundary conditions (moody's charts help here). however, for the much more common case of partial fixity or back-span slabs, i'm not at all clear on how to assign the moments.
2) to my knowledge, hillerborg doesn't do anything to address rebar cut-off locations. i've been developing moment diagrams based on the "assumed" support moments and calculating my bar lengths that way.
it's a shame that the method hasn't been developed further as it seems as though it could be quite powerful. i suspect that the strip method has been largely ignored by codes and designers for two reasons:
1) it doesn't address deflections in any direct way and;
2) the method isn't easily adapted to computer applications. |
|