about columns

Dear all.

Congratulations first. Karamba is fast on my laptop, it really helps to play with it!

Currently I am trying to resolve a kind of stairs shape structure, thus using short columns between levels, transferring loads with RX, Ry, Rz constrained at nodes.

Would you have some hint about cheating Karamba (with some coeffs applied on material or section properties, maybe), in order to obtain some acceptable compression and bending behavior on columns (and plates, via beams arrays ?...) ?

Also, I am working on OOOcalc, converting some material properties table to use with K :

I like MatWeb,a great place to freak out with funky materials, where the data comes in Newton and Pascal brand for SI... when others provide it with different orders of magnitude ... I always got mad with units.

Would you mind providing in the future a transparent units converter (using prefix/suffix, via your input string parsing methods will sure do it : a little effort compared to your whole GREAT work).

Thanks for your so cool tool.

Replies are closed for this discussion.

Replies to This Discussion

Permalink Reply by Karamba3D on August 1, 2011 at 1:15am

Hello Stanislas,

in which respect do you want to achieve an acceptable compression and bending behavior? Do you want to capture imperfections or instability?

Thanks for the link to MatWeb - the site is interesting.

A units converter component for Grasshopper would be useful - maybe we include one in the next release.

Best,

Clemens

Permalink Reply by Stanislas Petit on August 1, 2011 at 7:50am

Dear Clemens.

> in respect to the result of "columns-specific" calculation methods.

I am not an engineer, but as short as I know about beams and columns calculation, they don't use the same material properties (especially for anisotropic materials like wood), thus neither the same calculation formulas.

My question was :

Would you have some hint about cheating Karamba (maybe with some coeffs applied on material or section properties), in order to obtain some acceptable compression and bending behavior on columns in regard to the result that would be obtained with the use of "columns-specific" calculation methods ?

Best,

Stan

Permalink Reply by Karamba3D on August 1, 2011 at 1:53pm

Dear Stan,

you can use the same material properties for columns and beams for the calculation of deflections or cross section forces. The results will normally be sufficiently accurate.

Bending moments in a linear elastic material lead to a linear variation of stresses over the cross section, normal forces to a constant stres state. Although wood is anisotropic, bending moments and normal forces in a wooden beam act on the individual wood fibres in the same direction.

The special thing about columns is, that they can buckle when under compression. This is the reason why the design equations for beams and columns for the ultimate limit state are different. Compressive forces in columns reduce their resistance to lateral displacements which e.g. somewhat increases horizontal deflections in multi-storey frames.

A way of assessing the influence of second order theory in Karamba would be to do an initial calculation of deflections. Then take the deformed structure (which comes out of the ModelView-component), and recalculate it. Do this until the change of deformation is negligible. In order to get the behavior of individual columns right you have to impose initial imperfections on them:

incline each column by about 1/200 in the most unfavorable direction
divide each column in two and move the point in the middle off the center line (approximately L_column/200)

Tweaking material properties does not result in an appropriate approximation of second order theory effects.

Best,

Clemens

RSS