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.

 

[Tomfh]

It almost seems like they haven’t done it and are extrapolating from the simply supported case, where the F and L results converge....

 

[human909]

I have always been concerned about how little I/we know about the character of the Nastran modelling. In my past experiences messing with full blown FEM software (ANSYS, ABAQUS, etc) there have always been gobs of user selected pre-processing and post-processing variable involved that drastically affected the results.

I'd suggest your past experience is badly skewed. If you are changing settings and variables of the "pre-processing and post-processing" drastically affect the results then I would suggest that you have issues and your results are probably questionable. You could be having convergence issues or other problems. If changing setting here or there DON'T affect your results or have negligible effect then you are more likely to have an accurate output.

Have you watched the hour or so of videos that I linked? You could start there. There really is not great mystery here.

I suspect the Mastan is doing more or less the right thing, but I’m not entirely convinced, as it’s a bit black boxy.

I have faith that Mastan and NASTRAN are both doing the 'right thing' but are obviously constrained by different approaches. From the discussion earlier it is my understanding that it has separate models on buckling on different planes. A generalised FEA buckling analysis isn't constrained in this way and so can readily model buckling along multiple planes.

It almost seems like they haven’t done it and are extrapolating from the simply supported case, where the F and L results converge....

Agreed. And I'm now pretty unconvinced by AS4100's approach. It is interesting though that it is so easy to find literature comparing codes but much harder to find anything that examines how effective length is determined across the codes, specifically regarding AS4100.

 

[IDS]

I've been reading the header, seeing "rafter" and thinking "not my area", but I noticed today it was quite a popular thread, and had a look.

I haven't read all 350 odd posts, but I see people are talking about checking Mastan buckling results. I happen to have been updating my frame spreadsheet to include warping, and cross-checking against Mastan. I would be interested in doing a check on the current problem. Could someone provide a summary of the data, or point me at a post with the required information?

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[KootK]

If you are changing settings and variables of the "pre-processing and post-processing" drastically affect the results then I would suggest that you have issues and your results are probably questionable. You could be having convergence issues or other problems. If changing setting here or there DON'T affect your results or have negligible effect then you are more likely to have an accurate output.

The settings are largely the default. Messing around with the setting doesn't and shouldn't changes things beyond going for further refinement.

I couldn't disagree more. And I'm sorry to have to say this but, frankly, your cavalier attitude towards knowing the details about how your insanely complex software is working its magic strikes me as irresponsible usage.

Have you watched the hour or so of videos that I linked? You could start there. There really is not great mystery here.

You mean the videos that you didn't even bother to watch yourself before you recommended them to me?

In the meantime here are a few videos to cover the basics. (I've only watched the first short one).

Just because I disagree with you about a few things doesn't mean that I'm some thick headed noob that needs to be reeducated from scratch about FEM. I did read the web page that you linked and one the videos that you suggested. It all struck me as high level, rookie stuff not worthy of further attention. None of it delved into any depth on the mechanics of how FEM stability algorithms work. But, then, why would you care when you've got those killer default settings to rely upon?

You can't seem to get your head around the reality..

I can't get my head around your reality, that's all.

 

[KootK]

I've been reading the header, seeing "rafter" and thinking "not my area", but I noticed today it was quite a popular thread, and had a look.

Yeah, after 100 posts or so, I wish that the community had the ability to rename a thread to something more meaningful and more searchable. This should really be called:

"World's Greatest Lateral Torsional Buckling Thread With Particular Reference to AS4100 & Automated Buckling Analyses".

I would be interested in doing a check on the current problem. Could someone provide a summary of the data, or point me at a post with the required information?

Yes please. It's a 36' W10x12. Search your way to these posts for the low down. The sketch below is NOT the example you want. Rather, it's one of it's predecessors that we were using before attempting to move on to something more reflection of normal practice. Still, even with the wrong beam size and an incorrect span, it may be useful for you to see. Just replace W27x84 with W10x12, 32' with 36', and 250 kip with 12 kip.

If there's anything else that I can do to get you started, just let me know.

4 Dec 19 14:52 - Best description of test problem as I prepared it for Human909 for the same purpose.
4 Dec 19 23:59 - Nastran results.
29 Nov 19 18:17 - Mastan file, Mastan results, partial desciption of problem.
7 Dec 19 22:48 - Best summary of the final results from multiple sources.

I happen to have been updating my frame spreadsheet to include warping

I would love to know more about this from a theoretical perspective if it's convenient for you to share any of the details.

 

[human909]

Kootk. My attitude is far from cavalier. Just because acquiesce to your various demands doesn't mean I have a cavalier attitude to it. Go have a try of it yourself rather than spending time throwing around insults. If you had specific questions then sure I'd normally be more that happy to answer them by when you come across like a child demanding to see all the buttons, switches and knobs then it doesn't inspire me to help you.

Each model takes time and effort to set up. If you question my ability or the program then go try it for yourself or read/watch the huge wealth of information available online.

 

[human909]

Btw I respect your contribution to this thread and this forum Kootk. I just aren't thrilled about having my integrity questioned.

 

[RFreund]

Here is the test beam. Please let me know if corrections are needed.

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1576354220/tips/Test_Beam_1_yadtl4.pdf[/url]

EIT

http://www.HowToEngineer.com

 

[KootK]

Thanks for doing that. The only issues I see is with the 4' brace spacing. 36'/8=4.5'. Not that it's likely to make much of a difference to the final result.

 

[KootK]

...throwing around insults.

I object to your characterization of my comments as "insults thrown around". These threads are, in part, debates. And all debates inevitably involve challenge and conflict, some of which can be uncomfortable. The best that can be done to ameliorate hurt feelings is to state challenges as impersonally and dispassionately as possible. And that's precisely what I did:

...your cavalier attitude towards knowing the details about how your insanely complex software is working its magic strikes me as irresponsible usage.

I stand by that challenge can think of no gentler way that I might have issued it while remaining true to the sentiment.

I just aren't thrilled about having my integrity questioned.

I don't believe that I did challenge your integrity. Rather, I challenged your understanding about what is and is not important in complex FEM analysis. And, as with everything else here, that is simply an opinion. One that I've expressed in a gentlemanly manner conducive to a respectful debate. You have chosen to interpret this as a personal attack.

These statements on the other hand, very much fall into the category of impassioned, personal insults:

Do you own research I've held you hand on this one enough.

You can't seem to get your head around the reality...

by when you come across like a child demanding to see all the buttons

It is you who is behaving disrespectfully Human909. You've basically labelled me a petulant, lazy, thickheaded, child because I'd like to know more about your 3D buckling analysis than just "it's FEM, refer to the internet for more info". I challenge you to find anything in my body of work here that would constitute such a personal insult as these statements.

..it doesn't inspire me to help you.

I'm not really asking for your help Human909. What I'm asking you to do is simply supply the bare minimum of detail about of your very complex analysis such that those of us reviewing the results can have confidence in them. If I'd brought one of my former managers the result of a complex, 3D, finite element buckling analysis and proceeded to tell them that I didn't think any of the input parameters were important and that I did not wish to share them... they would have been well within their rights to sack me at my next review.

 

[KootK]

Btw I respect your contribution to this thread and this forum Kootk.

Ditto. I treat these threads like the old sheep dog and coyote cartoons. Within the context of in individual thread, I give no quarter until it's earned. But I treat each new thread as a new day with all past transgressions largely forgotten. Someday this thread will end for good and we can all start anew.

 

[Tomfh]

I rarely see you give a quarter no matter how much someone has earned it. That’s why people get so exasperated.

I suffer from know-it-all-ism too, so I sympathise.


Petty squabbles aside, I think this is a very valuable thread, that many have contributed to in important ways. Certainly both you and Human909 have.

 

[KootK]

I rarely see you give a quarter no matter how much someone has earned it.

I agree that it is rare for me to give quarter. I disagree that is unrelated to how much someone has earned it.

I admit wrongheadedness and logical defeat on a regular basis on this forum. Often it winds up even being me that proves me wrong. But I'll not apologize for being correct a whole lot more often than I wind up being incorrect. I wouldn't do this if I didn't have at least a 0.500 batting average.

"Earned it" is, of course very much in the eye of the beholder. I defend my opinions until I feel that they have been logically supplanted. I don't roll over to just to make friends or to avoid conflict. I also make a conscious but imperfect effort not to hang on too long just for the sake of my own ego. And I see no other rational way to proceed without sacrificing the truth seeking aspect of these long threads that is the primary impetus for my participation.

That’s why people get so exasperated.

Whaat??? Is that where I've been going wrong all this time? Come on, nobody know that better than I. Different people come to Eng-tips for different purposes. My impression is that, among the folks who are predominately "question answerers" you've got three main flavors:

1) Folks who are happy to add their opinion to a conversation and to "agree to disagree" if challenged.

2) Folks who want to lay down their contribution, drop the mike, and not have their work challenged at all.

3) Folks who really want to deep dive to get at the truth and see "disagree to disagree" as wasted opportunity to resolve resolvable technical defenses of opinion in a space where there often really is a "right" answer.

Since we all swim in the same pool and have an equal right to be here, some degree of conflict and frustration are inevitable. As a clear cut #3, I tend to frustrate the crap out of #2's. They get fatigued at having to fend off vigorous challenges but, at the same time, aren't willing to just walk away without having the last word.

So I just do me and leave it to everybody else to do the same. That, again, for lack of rational alternatives.

 

[KootK]

I notice no-one else lasted the distance. Are we really that obnoxious?

Petty squabbles aside, I think this is a very valuable thread, that many have contributed to in important ways.

I view the petty squabbles as part of the natural cost of doing business in a thread like this. This thread exists because you've got some really smart people here who really care. And folks who really care are bound to squabble. When all is said and done here, I expect that the impression that will dominate my memory of this journey will bee something like this:

Thank goodness there is such a thing as the internet and that it connected me to the very, very small group of humans across the globe who are simultaneously:

1) Sharp enough to debate this topic in detail and, more importantly;

2) Crazy passionate enough about this topic to really "get in ring" so too speak.

I live in a city of 2M people and lots of them are structural engineers (oil & gas). I wager that, if I could round up all of them into a gymnasium for an afternoon, I still couldn't reinvent this conversation. There just still wouldn't be enough of the crazy passionate people.

 

[RFreund]

Thanks for doing that. The only issues I see is with the 4' brace spacing. 36'/8=4.5'. Not that it's likely to make much of a difference to the final result.

I'll fix it, but do we want 36' long and 4.5' or 32' long and 4'?

Which matches the models thus far?

EIT

http://www.HowToEngineer.com

 

[KootK]

I'll fix it, but do we want 36' long and 4.5' or 32' long and 4'?

36' & 1/8th point bracing has been my intention.

 

[IDS]

I would love to know more about this from a theoretical perspective if it's convenient for you to share any of the details.

I'll probably blog it real soon now (in the Microsoft meaning of those words).

In the meantime have a look at:

https://www.researchgate.net/public...m_Finite_Element_Including_Nonuniform_Torsion

which is a free download and the basis of what I am doing.

I am not sure how relevant it is to the discussion in this thread though. Warping would certainly have an effect on buckling but at the moment I can't see how you would apply restraints to the top flange only of a 1D beam model, and if you go to a more detailed model that would include warping effects anyway.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[KootK]

Thanks for the paper. That'll do me.

at the moment I can't see how you would apply restraints to the top flange only of a 1D beam model

Can you not do it just as we've been doing it in Mastan to date, by constructing faux frames in the shape of cross sections? See [28 Nov 19 05:13] for an expanded discussion of that.

 

[Tomfh]

Agree with adding the stick elements to grab the top flange lateral restraints. I’d model them as flat elements of equivalent stiffness to the real beam web.

 

[Agent666]

I’d model them as flat elements of equivalent stiffness to the real beam web.

That's what I was attempting to do. Using the same section and applying warping fixity to the faux sections screws with the analysis. Warping fixed for the beam, warping free for the faux elements, and faux elements are another defined section representing the web stiffness. I tend to make this the beam depth by the web thickness. There is of course no true way to represent the restraint to the flanges on a 1D element as its not as simple as just having the web there.

I explained what I was doing in 18 Nov 19 17:42 post with respect to this if you search the date/time.