Rafter without fly brace? 22

[fourpm]

I am designing rafters to AS4100 and wondering what if I don't use fly brace. I understand that with fly brace it will give you full restraint. But if I don't use fly brace, will the purlin above be considered as lateral restraint for rafter under uplift? If so. can I take the purlin spacing as segment and the only factor that changes without fly brace is kt?
I have the same question when it comes the continuous steel floor beam design where Z/C floor joints sit on top of the beam. What segment should I take for the beam near the support? Can I take the floor joists spacing as segment with lateral restraint? Can anyone give me some examples? I have read some manuals but the examples they have are simply supported beams only. Thank you.

 

[RFreund]

FWIW, still working. I've cheated and have bounced some questions off higher powers, but everyone is still trying to get their feet under them after the holiday break.

EIT

http://www.HowToEngineer.com

 

[human909]

The 610UB example under self weight appears to work under LTBeam, but not in NASTRAN?

Does it? Your previous statement suggested LTBEAM: 1.526 (FLF) (Mcr = 266.4kNm) which is not the same as PLP. Additionally your critical buckling moment is not beam capacity. So from what I can see it doesn't clearly work.

From back of the envelope calculations you need to reduce you Mcr by a 44% to calculate capacity according to AS4100. Though that is very back of the envelope.... I thought we had moved on accepting the fact that AS4100 calculates incorrect reference buckling loads in cases of moment reversal due to using incorrect effective lengths.

 

[Tomfh]

Human, you are right that the LTBeam example is FLF not PLP. LTBeam doesn’t have P restraints as far as I can tell.

Reference buckling moment exceeding capacity by 52% is far less clear cut than an example where capacity exceeds reference buckling moment.

If using cl5.6.4 buckling analysis with a buckling moment of 266kNm the the 610UB factored capacity is ~210kNM, which exceeds capacity of the the simplified Le approach.


I’d like to identify a clear example with F and L restraints where that isn’t the case, ie where the AS4100 predictions using the Le approach is unconservative.

 

[Tomfh]

I've cheated and have bounced some questions off higher powers,

Did you get anywhere with this? Who did you ask?

 

[Agent666]

For what it's worth I went through and analysed the original 70' beam case both with F restraints and L restraints to determine the differences in both the hand and buckling analysis routes.

Agree that the use of L restraints gives very large differences in capacity between buckling analysis and hand analysis methods. My thoughts are outlined in the post above. But in general I think if you're considering L restraints in regions where you have moment reversals you might be better making sure there a F or P restraint is used either side of the reversal to prevent the twist that seems to significantly reduce the reference buckling moment.

In a way without the twist restraint it's perhaps a little like a cantilever where both flanges are critical in a way. You restrain one, then the other is still unstable to a degree.

In my mind I don't understand how the code writers might have got it wrong if this is indicative of a larger issue, keeping in mind of course that we're only looking at selected hand picked cases which are a very small population in the scheme of things. These methods tend to predate modern methods of buckling analysis (30+ years), where in the past they were available to researchers, while now, they are more mainstream. Also the benchmarking for developing the provisions was based entirely on simply supported beams. So it's entirely possible that something is wrong in L restraint land.

 

[human909]

Something is very wrong in L restraint land.

Nice blog post Agent666.

 

[Tomfh]

Yes it seems very likely that the code is mistaken about the effectiveness of L restraints. Agent’s article shows that a code compliant L restrained beam can buckle well below the code predicted capacity.

 

[Blackstar123]

I'm at almost 20% mark of this post and plans to continue reading it tomorrow.
This debate is as entertaining as (but definitely more educational) than the current fiction I'm reading. I laughed out loud at several quips made by contributors all in the noble cause of better understanding.
But this debate is not only entertaining, it may answer one of the most troubling question in my 6 year design experience. I'm learning alot. I hope it gets better.

P.S. kootk almost made me a believer that full length between the fixed supports should be considered for negative moment if bottom flange is not laterally restrained. Until now, for negative moment I've taken the unbraced length equals to the length from fixed support to first lateral support (at top flange) after IP. I believe that the tension force in the bottom flange due to positive moment will try to straighten the buckling produced due to compression near the supports. But I could be wrong and I am reading this post with open mind.

Euphoria is when you learn something new.

 

[KootK]

I believe that the tension force in the bottom flange due to positive moment will try to straighten the buckling produced due to compression near the supports.

Search for my post on [1 Dec 19 22:10] which contains the clip shown below.. In that post, I considered the phenomenon that you're describing in some detail. I concluded that the improvement is real but generally insufficient to resolve matters entirely.

 

[Tomfh]

Blackstar, much has happened since the 20% mark, and most of us who’ve been active in this thread view it differently to how we did back then, so don’t come to any conclusions too soon.

As for tension in the bottom flange - it is often insufficient to constrain buckling. I.e. insufficient to maintain an uphill energy gradient.

 

[Blackstar123]

Search for my post on [1 Dec 19 22:10] which contains the clip shown below.

I’m at 11th November, 19 right now and I can see there is some very interesting discussion going on between 11th Nov and 1st Dec. I’m only a tiny bit tempted to skip these middle posts. I notice some very good FEM results when I quickly scrolled down to bottom. I’m hoping to have a better understanding till I reach this point.

Blackstar, much has happened since the 20% mark, and most of us who’ve been active in this thread view it differently to how we did back then, so don’t come to any conclusions too soon.

Can’t wait to read all. It took me more than an hour to reach that 20% mark.

As for tension in the bottom flange - it is often insufficient to constrain buckling. I.e. insufficient to maintain an uphill energy gradient.

This was how I used to rationalize my view of thinking regarding the unbraced length for negative moment. But I’m starting to see that buckling in bottom flange is not as simple as I presumed. I remember posing this very question to my professor but his answer was not very satisfactory.


Euphoria is when you learn something new.

 

[RFreund]

Did you get anywhere with this? Who did you ask?

Dr. Ronald Ziemian
He seemed interested, but was traveling when I originally sent him the question/summary. I will reach back out to see if he had time to get through any of my questions. I tried to summarize the debate as best I could.

EIT

http://www.HowToEngineer.com

 

[RFreund]

@Agent666 - Great blog post. Seems like we are starting to get on the same similar page.

EIT

http://www.HowToEngineer.com