Kl/r preferably <200 2

[smokiibear]

I provide engineering for light guage steel covers with Solar atop. They are light gauge purlins, light gauged metal roof. When columns reach over 14.75', I've typically followed the recommendation to not exceed Kl/r < 200, and consequently, forced to go to 6" std pipe. I'm curious if you could comment. I don't think it's warranted on these tiny structures, but don't want to make a poor choice. Could you comment.

Here is a sample of the product.

 

[Flotsam7018]

I don't believe 200 is a magic number. If the analysis has been done in accordance with Appendix 7 (360-16), and the utilization of the columns is within a comfortable range, independent of the loading conditions, I don't see an issue.

Additionally, one could use DAM are your KL/r would drop by ≈48%.

 

[saplanti]

I agree with KootK on the issue since today seismicity, wind and flood becoming more and more news on the media. Some other things are related to the foundation and their effectiveness. However the life expectation is another subject to think about.
I guess that these structures are not regulated, everything is for the happiness of clients.

 

[dauwerda]

dik, one thing to note on round HSS. The sizes that are typically produced and readily available (as readily available as anything is right now) coincide with standard pipe sizes. So, unless you are willing to wait for an intermittent mill roll, you are essentially stuck with the same wall thickness as standard or extra heavy pipe. You do get the advantage of the stronger material however.

It looks to me like the columns will see mainly bending and tension loads (due to wind uplift) with relatively low compression loads - especially if snow isn't an issue. I would be ok going above KL/r = 200 as long as everything else still checks out.

 

[KootK]

Food for thought: from the perspective of damage during fabrication, handling, and erection, how does the column know that it's a cantilever in service with a K=2.1? Over the years, I've found structures to be notoriously bad listeners.

 

[dik]

Thanks dauwerda... not the case in these environs... simply because thinner wall thickness is generally available. There are often 'gaps' in the sizes available.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[glass99]

@KootK and other kl/r>200 haters: What kind of damage specifically are you talking about? If you are talking about a bow or tilt in a column, the issue is P-delta (i.e. extra moment) less so than Euler buckling. We have looked at that extensively with our glass structures because there are large variations between fabricators for straightness. If you have a column which is primarily in bending and with low compression loads (such as the example in this posting), P-delta isn't a major concern. kl/r>200 columns are only ever a consideration for relatively light compression loads. Can a flag pole have kl/r>200? We have had a kl/r=270 structure standing in the middle of Times Square carrying up to 500 tourists at a time (and >10 million cumulatively) for the last 13 years without issue.

 

[dauwerda]

I know KootK does not need me to come to his defense, but his stance is on the other side, he is in favor of allowing the columns to go past the 200 mark.

2) Given my expectations for the loads on these posts, I feel that they will be almost more like beams than columns. As such, if designed properly for lateral, I also struggle to support KL/r < 200 on any kind of design / accidental eccentricity basis. The dominant load on these posts may well be applied moment.

I vote for disregarding the compression slenderness in this instance.

Note the less than, not greater than symbol (and of course his last statement).

 

[KootK]

KootK spends rather a lot of time defending himself and will gladly accept any help offered.

 

[phamENG]

In terms of handling and fab damage, I certainly agree that the column doesn't care or know if it's a cantilever or not. In those cases L/r would be an eminently reasonable metric to compare the old limit.

glass99 - I had a horrible experience there. The salesperson was nasty and unhelpful, and had I known my half price ticket to the Phantom would only let me see half the stage, I would have gladly paid full price. Though I suppose that's not the sort of issue you're talking about, is it?

 

[glass99]

@KootK: God sorry, should have read your earlier post. Turns out you are on the right side of this.

@phamENG: Let me know who you dealt with at TKTS and I will have them fired immediately. But at least when you shoved your credit card through the glass the wall didn't buckle under the weight of the people above.

 

[KootK]

@glass99: all good.

 

[phamENG]

Fair point, fair point. Though I must say, many of the handling considerations for the KL/r are sort of moot when it comes to glass, aren't they? Let's face it - even some of the most robust glass is far more likely to break in handling than some of the least robust steel elements. So any glass handling operation will likely fall into the 'special care' clause mentioned in the steel manual.

At the end of the day it's a judgement call. If there's a compelling reason to go higher than 200 that outweighs the risks (such as they may be when considered case by case), then by all means be an engineer. But if there's not a particularly compelling reason...why give the other guy's lawyer something else to point to? Because it doesn't matter what the problem is, if there's a lawyer involved that's going to be low hanging fruit.

 

[glass99]

@phamENG: Yes you definitely need to commit to doing your homework if you are going to exceed 200.

We have been playing with the idea of ultra high modulus carbon fiber for some columns holding up a roof. The young's modulus is 81,000 ksi if we use unidirectional fibers. The roof is 26ft * 17ft with columns 1.2" square, which would put them in relatively high compression stress under snow load, and a kl/r of ~240.

 

[KootK]

In some respects, one could argue that, for your carbon fiber, it would be kL/r x E_steel/E_carbon = 86. Given that it's an entirely different material, however, it's probably just best to stick with that commitment to homework doing.