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]

However, if there are errors in AS4100, then AS4100 may be pretty clear on something that is incorrect or was never intended.

If AS4100 is mistaken I'm sure they didn't intend for it. What I am saying is that if there are mistakes in AS4100 then it's not because of poor wording etc (eg as you previously hypothesized), I am saying the errors are because they are genuine mistakes in the theory, which has thus ended up in AS4100.

My money is currently on them wrongly extrapolating to all cases that L is equivalent to F (when it only applies in some cases), but that this theoretical hole is concealed by real world stiffness bonuses (e.g. purlin cleat rotational restraint), and also the normal strength buffer (phi factors, load factors etc).

I think that you're jumping the gun on your critique of my stuff

Which critque of your stuff? Which stuff in particular?

 

[human909]

PS. I've been not around partly because of thread weariness, partly just busy with work and partly IT troubles; NASTRAN broke and I can't get it running again. :-(

 

[RFreund]

Done. A couple of things to note:

1) While Mastan does have a utility for uniform load, to my knowledge, there is no way to apply a uniform load to a member at any location other than the shear center. Rather than getting fancy with more faux members with difficult to predict behaviors, I just discretized the uniform load as serious of equivalent point loads at the 1/8th points. [500 plf * 36 ft / 7 = 2.571 k] each. Hopefully this is sufficient for your purposes.

2) I don't know about you but I'd expected the switch from a point load to a uniform load to have more impact than it seems to have. Having pondered this a bit, my explanation for this is:

Interesting and thank you. I think I can put this to bed for now. Can you share the Mastan model for this? In general the take away seems to be that there is an increase in buckling strength capacity when the top flange is laterally restrained but not enough that would always get you back to your plastic strength.

I don't think I really understand the current argument between Kootk and Tomfh. I understand you are trying to see when a restraint that restrains lateral movement can be considered equivalent to one that restrains both lateral and rotational. But I'm missing why this is relevant or what the (specific) consequences are.

This might be related to the above question - What in AS4100 says that you can consider the unbraced length to be an inflection point or the brace past the inflection point? If I read through the definitions it seems like you could make an argument that you need to use the entire bottom flange as the segment length. Maybe someone can walk me through how you are able to reach a different conclusion?

EIT

http://www.HowToEngineer.com

 

[Tomfh]

What in AS4100 says that you can consider the unbraced length to be an inflection point or the brace past the inflection point? If I read through the definitions it seems like you could make an argument that you need to use the entire bottom flange as the segment length.

AS4100 defines segment (and subsegment) length as the distance between F, P or L restraints, thus an L past the inflection point counts at an end of sub segment.

No one does it as you are proposing. It’s not the intention.

The issue now is whether that intention is wrong.

 

[KootK]

What in AS4100 says that you can consider the unbraced length to be an inflection point or the brace past the inflection point?

Best be careful... accusing someone's code of allowing inflection point bracing these days is fightin' words. AS4100 most definitely does not support inflection point bracing. As a good way to clarify this, consider the situation of our W27X84 but imagine that the intent is to call the 1/4 points braced and delineating design sub-segments. As I understand it, most AS4100 practitioners would go to one brace past the inflection point but, as afar as I know, going to the brace at the inflection point is also street legal.

It would break down like this:

1) Inflection point bracing = no lateral restraint to either flange but call it a brace point anyhow.

2) AISC = usually lateral restraint to both flanges and, obviously, call it a brace point.

3) AS4100 = lateral restraint to just the top flange and call it functionally equivalent to lateral restraint to both flanges for stability purposes.

I do understand how, at first glance, one might erroneously assume that AS4100 is allowing inflection point bracing however. As is evident from the list, it's 1/2 way there as viewed with AISC eyes. That is, in fact, part of what drew my attention to this thread in the first place.

 

[Tomfh]

Best be careful... accusing someone's code of allowing inflection point bracing these days is fightin' words. AS4100 most definitely does not support inflection point bracing.

Not inflection point Per se, but the lateral braces past the inflection point.

 

[KootK]

Which critque of your stuff? Which stuff in particular?

I mentioned that I was going to take another stab at re-writing / reinterpreting AS4100 and that immediately set off another round of debate about how we feel AS4100 is to be interpreted. We'll have that debate, for sure, but I'd like to forestall that until after I've actually pitched my re-write/reinterpret. I haven't done that yet.

Because I am a ridiculous glutton for punishment, I am eventually going to take another stab at rewriting/reinterpreting AS4100 in a way that I feel would resolve the discrepancies that now seem manifest. At this point, I feel that an error in the writing of AS4100 is at least as likely as:

If AS4100 is mistaken I'm sure they didn't intend for it.

I agree and certainly never meant to suggest otherwise. If there's an error, I see it coming about as shown below/attached: plausibly and devoid of malicious intent.

What I am saying is that if there are mistakes in AS4100 then it's not because of poor wording etc (eg as you previously hypothesized), I am saying the errors are because they are genuine mistakes in the theory, which has thus ended up in AS4100.

I understand, I think, but disagree and will again hypothesize that the issue may be poor phrasing. However, with the next round, note that:

1) Previously, I'd hypothesized that the AS4100 phrasing was mildly ambiguous and therefore fodder for minor misinterpretation by designers. With the next round, I'll be proposing that the problem is something more akin to an error of omission that no designer could realistically have been expected to sort out on their own (unless they'd been part of a 500 post thread on the subject).

2) I am not, and will not be, suggesting that I'm in any way certain that there is a content error in AS4100. Rather, I'll exploring this as one possibility among several that I feel is again worthy of consideration because:

a) at the start of this thread, I might have put to likelihood of an AS4100 content error around 0.5% but;

b) after all that has transpired, I might now put the likelihood of an AS4100 conten error closer to 15%.

 

[Tomfh]

Yeah but guys like Trahair and all the other gurus who wrote the codes and bibles and sample problems do it exactly the same way we do. How could that be the case if things just got lost in translation?

 

[KootK]

Not inflection point Per se, but the lateral braces past the inflection point.

Did I not articulate that sufficiently in the statement below? With the W27x84 exmple, I intentionally set it up so that the IP would also be the first brace point. I feel that makes for the most extreme / salient version for comparing things. If anyone had objected to that, I'd just have moved the quarter point brace over 6" into the compression zone and continued in the same vein.

As I understand it, most AS4100 practitioners would go to one brace past the inflection point but, as afar as I know, going to the brace at the inflection point is also street legal.

True inflection point bracing, in the naughty sense, assumes that a member is F-braced at inflection points when there is no bracing in play. Whether AS4100 L-bracing is at the IP, or one brace past the IP, it's still a massively different thing simply because it's still a real brace instead of no brace at all, even if it is only providing restraint to one flange.

 

[Tomfh]

True inflection point bracing, in the naughty sense, assumes that a member is F-braced at inflection points when there is no bracing in play. Whether AS4100 L-bracing is at the IP, or one brace past the IP, it's still a massively different thing simply because it's still a real brace instead of no brace at all, even if it is only providing restraint to one flange.

Yes, agree completely.

 

[KootK]

Yeah but guys like Trahair and all the other gurus who wrote the codes and bibles and sample problems do it exactly the same way we do. How could that be the case if things just got jumbled in the code writing process?

a) I don't know who exactly wrote the LTB section of AS4100 and therefore do not know if there are textbook examples by those people to indicate how they interpret things. Do you?

b) Since you don't yet know what reinterpretations I'm going to propose, how can you possibly know whether or not any particular textbook examples contradict them?

I'm begging you Tomfh: can you please put a pin in your critique until I've actually put pen to paper on my proposal? That would have to be more time efficient for us both.

 

[KootK]

Can you share the Mastan model for this?

A zip file containing all of the models that I've been using is attached. Ridiculously in the modern era, all 17 only occupy 100 kb of hard disk space. The file naming is not that descriptive but:

a) that's partly Mastan's fault as its naming rules are bizarrely limiting and;

b) upon opening a model, it ought not take anyone long to figure out which it is.

In general the take away seems to be that there is an increase in buckling strength capacity when the top flange is laterally restrained but not enough that would always get you back to your plastic strength.

Yes, as it pertains to AISC procedures. That said, in our explorations, I've not found the bump to be that significant in many cases. It was as low as 17% for the W27x84 I think. I think it understandable that one might leave that on the table in favor of simplicity when not motivated by desperation for capacity in an existing build situation etc.

I understand you are trying to see when a restraint that restrains lateral movement can be considered equivalent to one that restrains both lateral and rotational. But I'm missing why this is relevant or what the (specific) consequences are.

I hope that you'll stay tuned long enough to review one more post of mine. I'm planning to put something together just for us AISC folks and it will elaborate on this aspect of things.

 

[Tomfh]

a) I don't know who exactly wrote the LTB section of AS4100 and therefore do not know if there are textbook examples by those people to indicate how they interpret things. Do you?

Good question as to who actually wrote AS4100. Anyone know the authors?



Steveh49 linked to the Trahair document, where L is deemed equivalent to F. That seems a fairly clear example of the gurus aligning with the practitioners.

 

[RFreund]

AS4100 defines segment (and subsegment) length as the distance between F, P or L restraints, thus an L past the inflection point counts at an end of sub segment.

No one does it as you are proposing. It’s not the intention.

The issue now is whether that intention is wrong.

I agree that no practitioner does it this way (or atleast I take your word for it) and also agree that the issue is weather or not that is correct (I think we have shown that there are cases where this is unconservative). But if we take AS4100 to court and say "look at this design using your code, you way over estimate the capacity of this W10" it seems like they could argue that you should have used the entire length. Meaning if you start with the bottom flange it is the critical flange, what in the code says the sub-segment length stops at the lateral restraint on the opposite flange once you get past the inflection point. I just don't see it that clearly. I will try to walk through this later. Basically making a case (if I can) that the code says you need to use the full bottom flange length. Again not referencing any guides, just the language in the code.

I hope that you'll stay tuned long enough to review one more post of mine. I'm planning to put something together just for us AISC folks and it will elaborate on this aspect of things.

Fair enough.

EIT

http://www.HowToEngineer.com

 

[Tomfh]

ut if we take AS4100 to court and say "look at this design using your code, you way over estimate the capacity of this W10" it seems like they could argue that you should have used the entire length.

Critical flange is assessed at the cross section location, not the start of the flange (e.g. support). That is how the background theory is stated, how the code is written (e.g. "The critical flange at any section shall be the compression flange"), how the textbooks and canonical examples are presented, and how practitioners do it.

How do you mean they could argue the contrary? What specific wording do you believe suggests you need to consider compression in all cross sections as opposed to the cross section under consideration?

 

[Trenno]

I haven't been following this thread religiously, but may be able to contribute something.

Firstly - according to my colleagues back in Straya, the 2019 AS4100 draft code has been released for public comment. After you've posted another 500 times, maybe you can suggest, if any, amendments to be included in the new code.

In regards to the authors of the bending section of the code. The commentary provides some insight. T.J. Hogan also seems to feature heavily throughout the commentary.

Not sure if it helps, but the commentary defines the critical flange as the following:

 

[human909]

Thanks for posting the above comment Trenno from the 4100 commentary. I was meaning to post it myself but I've been distracted and had a bit of topic exhaustion.
-The comment is quite clear that 5.5.2 annd 5.5.3 are more specific clarifications of 5.5.1 rather than simplified choices.

In the meantime:
-I've got Nastran up and running again. Kootk (or anybody else) if you want something specific run through its paces with an FEA approach I'm willing to do it if you point me in the right direction regarding specific scenarios.

-I've also spent a little time digging around academic literature on the topic. Though in general the academics focus on theoretical buckling equations rather than definitions of the effective bending length. AS4100 is apparently more conservative that other codes regarding buckling when when you are comparing apples with apples with beams of the same effective length.

-I've found little academic comments regarding treatment of L,P,F restraints etc...

-A slight aside but in the fine print in Space Gass there a small comment regarding L restraints. SpaceGass a very significant market share in Australia for steel design.
This isn't exactly relevant but it is another hint at how L restraints have confused and been abused by structural engineers. This quote is in the context of MINOR AXIS compression effective lengths so it isn't the same thing.
"there is some recent doubt as to whether lateral restraints on equal flanged I or W shapes can restrain the overall cross section laterally"
(AKA in the past L flange restraints have been used by some engineers as minor axis buckling restraints.)

 

[Tomfh]

The comment is quite clear that 5.5.2 annd 5.5.3 are more specific clarifications of 5.5.1 rather than simplified choices.

I find it a bit odd that they refer to them as “more specific” when in many situations they give different answers.

Though in general the academics focus on theoretical buckling equations rather than definitions of the effective bending length

I’ve been noticing the same thing. The effective lengths seem to be assumed.

 

[human909]

I’ve been noticing the same thing. The effective lengths seem to be assumed.

Call me a cynic but in the realm of publish or perish there is often plenty of papers published that behind the knees of giants rather than on their shoulders.

The effective length is a clear from a theoretical perspective. Once you start messing around with it with qualitive restraints then it all gets a bit murky. (And the definitions of F,L,P restraints are largely qualative!)

 

[KootK]

Kootk (or anybody else) if you want something specific run through its paces with an FEA approach I'm willing to do it if you point me in the right direction regarding specific scenarios.

Hell yes. I vote for the W10x12 example that I created for tomfh recently.

- 36' span W10x12
- Top flange L-brace at 1/8th points.
- Top and bottom flange L-braces at beam ends.
- Ends fixed for strong axis rotation. Longitudinal pins at web to flange intersections?
- Ends free for weak axis rotation.
- No warping restraint to flanges at beam ends.
- 250kip 12 kip downward point load at midspan applied to top flange.
- Weightless beam.
- Fy jacked up to 500 ksi so that elastic bucking is the only failure mode.
- Shear deformation turned off by setting G = infinity.
- E = 29000 ksi

You've mentioned topic fatigue numerous times. Pace yourself with this stuff. Take a week off to refresh. We've got all the time in the world.