KL/r > 200 - Greenhouse Posts 6

[heleads]

I understand that it KL/r "preferably" should not exceed 200. In greenhouse design, the columns are inherently slender and the factor often exceeds 200. In some cases they exceed 300. Since greenhouse structures typically move much more than a typical building and have a high degree of redundancy, logic seems to suggest that a column with a KL/r of 300 should not be that great of a concern, especially since so many greenhouses are out there and performing under the design load conditions. One reason for a high factor is when K is assumed to be 2 (pinned base and top fixed Rotationally with a truss). Does anyone have any experience or comments regarding this situation?

 

[UcfSE]

When you exceed kl/r = 200, you start getting into situations where the code equations are no longer valid. Things that normally can be negliglected, and are in the derivations of the code, can no longer be neglected. Hence you should not exceed this recommendation or you are on your own. Perhaps they have done more research and have derived equations and done more experiments to justify the use of very slender columns.

 

[271828]

One of the main gotchas would be initial out-of-straightness. If a column is extremely slender, then extra moments coming from OOS might be significant. You might be able to model the column using the new Direct Analysis Method assumptions, do a rigorous second-order analysis (including P-LittleDelta of course) and be ok.

 

[jetmaker]

Couple of things...

1) check to see if the farm building code has different allowable slenderness requirements (if applicable to farm sites/ low occupancy);

2) your boundary conditions may be a little conservative in that often a gutterconnected greenhouse structure has posts which are set in a concrete foundation. This condition is closer to a fixed end condition rather than pinned.

3) as previously mentions, watch out for initial imperfections in the alignment.

jetmaker

 

[DaveAtkins]

I have no problem with kL/r > 200. AISC specifically allows this. The Euler buckling formula obviously controls with columns in this slenderness range.

DaveAtkins

 

[csd72]

Dave is correct.

In the ASD (9th Edn.)section E1 refers to B7 which refers to you to use equation E2-2 if slenderness is >200..

 

[haynewp]

The new Direct Anlaysis Method of AISC does not rely on K factors. You can model your fixed top and pinned base and see what the capacity is if you meet all the provisions of the Direct Analysis Method. I think the Direct Analysis Method will be the recommended method for now on by the AISC, and the effective length method will be thought of as outdated.

 

[WillisV]

haynewp is correct.

 

[271828]

haynewp, you are correct that AISC is heading that way. The next Spec. will have the DAM in Chapter C and the ELM in an appendix. I think that's the slow way of shoving it out the door!

The DAM relies on the engineer's ability to do an ACCURATE second-order analysis to capture instabilities. In this case, he needs to be able to capture both types: P-LittleDelta and perhaps P-BigDelta.

 

[JAE]

That's P[δ] and P[Δ] for you folks in Rio Linda.

 

[haynewp]

There is also the 0.8E that has to be applied to the frame to account for softening due to residual stresses and yielding, and the initial allowed out of plumbness that has to be put into the model (0.002*h) or applied as notional loads. A few other caveats but I believe the Direct Analysis Method is a much more intelligent way to go.

 

[271828]

haynewp, I think the coolest thing about the DAM is how little room it takes up in the Spec. What is it, like 3.5 pages!? Those 3.5 pages more or less say "Do it right." LOL.

I was introduced to all that effective length stuff about 15 years ago and have been using it pretty steadily ever since. I've even taught it once. There are still weird cases that come up and I have no clue what to do with. I say Good Riddance ELM, probably the most misused part of the steel Spec.!