Meeting AISC relative & nodal bracing requirements - How are engineer's checking this is real li

[jochav5280]

Good day,

When designing columns & beams, we must determine what the un-braced lengths are for flexure and compression. AISC has specific strength and stiffness requirements that must be observed. My question is how are engineers actually checking verifying that their braces are strong and stiff enough to meet these requirements? For large structures composed of thousands of members, you certainly can't create a model for each situation to test this, so I'm wondering what other practicing engineers are doing.

My experience is that nobody seems to check this; generally it is assumed that if there is a beam framing to a girder that the beam braces the girder from translation and is thus relied upon for determining the girder's flexural capacity. Our office uses STAAD.Pro for structural analysis & design, and they put out a few good references on structural analysis & design. They put an example together where they created an additional model to check both strength and stiffness of their brace to ensure that it met the AISC requirements, however, this doesn't seem practical for large structures. Are there any good quick ways this is done in practice?

Many thanks in advance!

jochav5280

 

[Guastavino]

Are we talking average everyday beams or some kind of truss with specific bracing?

Average everyday beams if there is a diaphragm of any sort I usually neglect.

With specific bracing, if the loads are small and I have something in there, judgment tells me it's ok. If the loads are higher I check the brace for the load.

Now it's not "judgment" based on a gut feeling. I've run the numbers and the bracing requirements are small for most of the smaller type buildings I do. Therefore, judgment tells me that it will be OK. Much like I don't do calculations anymore for a W8x10 lintel with a 1/4" bottom plate for a 3'-4" retrofit lintel in an existing masonry wall, been there done that.

So, ultimately, run the numbers for one, see what they are and then get a feel for which of the thousands you really do need to check.

I certainly don't check them all...at least not with numbers.

 

[KootK]

My experience is that nobody seems to check this

You can put me in this category I'm afriad. Thank goodness for the anonymity of the internet.

I guess I'm doing a bit more than utterly nothing:

1) 80% okay by judgment and do nothing (don't ask about the nature of my judgement).
2) 15% of the time, 2% rule and no check on stiffness assuming same material for brace and member.
3) 5% fancy AISC provisions.

The level depends on the degree of KootK paranoia in any given situation. I've got some spreadsheets that cover the bracing requirements for some of the very simple/common cases. But then, those are the very cases that I usually end up going "by inspection".

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[TLHS]

Bracing strength and stiffness rarely control. For overall buckling of a column, slenderness or the lateral system design will heavily control sizing. For lateral torsional buckling of a wide flange, the loads are normally small enough that constructability is more likely to govern.

The spots where you get into trouble are trusses, light framing (including detailed design of joists), and situations where you're mixing materials (e.g. bracing steel with wood). Other than that, it's when you've been forced into a corner and are doing something non-conventional.

I put it in the class of code requirements that is meant to stop people from doing stupid things and then arguing that the code doesn't require them to do anything else. It's so you don't forget that bracing actually has to stop the braced object from moving to be effective. In that sense, it's similar to the half percent notional load for stability in the Canadian steel code. I don't normally use it, because it's a rounding error compared to the lateral forces that affect stability in the types of structures I work on and my dead only case isn't even close to being a problem. But it's there so that you can't pass the code by running a mathematically perfect model for a stability check on a gravity only structure because you forgot that the real world doesn't work like that.

 

[sandman21]

All lateral frames beams and columns are checked, tapered steel girders, trusses, girders greater than 2x the beam depth (if your detail does not properly cover), girders/beams which do not have traditional diaphragms, all other unique situations.

 

[MotorCity]

Unfortunately my experience has also been that nobody checks this. I get the impression that most engineers consider this to be an academic exercise as opposed to a practical, everyday check. In fact, a few engineers have outright told me that it is an exercise in futility and used the age old argument "the buildings designed 50 years ago when this concept was not around aren't falling down are they?". On the bright side, while the concept can be difficult to enbrace, I believe many of the software programs will check it.

 

[jochav5280]

Thank you for your responses.

MotorCity, I get the same responses from my colleagues, yet many of them don't know how to check these requirements. We use STAAD.Pro and to my knowledge, it doesn't perform any check in this regard. Is there a program that you know of that does?

Trying to glean some take aways here; a lot of the time these forums lead to a lot of non-definitive suggestions and I wished something as important as this topic would be black and white.

Best regards,

jochav5280

 

[KootK]

I get the same responses from my colleagues, yet many of them don't know how to check these requirements.

Same goes for conversations here. As soon as it gets remotely complex, like a steel beam braced by a wood diaphragm, you get a whole lot of silence. And those who keep talking struggle to even figure out if what they're dealing with is nodal or relative bracing.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[JoshPlumSE]

One thing I really like about these provisions is that they give us the ability to check things for situations where we have a concern. Is this member sufficient strong or stiff for bracing the main chord of my truss? Well, we can use these provisions to get a good answer.

What I don't like is that I feel like these should be used for "guidance" rather than requirements. The provisions are a bit confusing and it's not always clear how they'll apply in certain situations where we are unsure.

I might like them better if there were a design guide that showed a number of real world type examples where the numbers make sense. I have seen attempts to use this provisions when the "by inspection" answer is obvious to any engineer with experience and these provisions seem to say that the "by inspection" conclusion is incorrect.

 

[DETstru]

1) 80% okay by judgment and do nothing (don't ask about the nature of my judgement).
2) 15% of the time, 2% rule and no check on stiffness assuming same material for brace and member.
3) 5% fancy AISC provisions.

I'm pretty close to this but more like 80/19/1.

AISC has the equations you need in Appendix 6 but using 2% is almost always conservative and whatever member you typically brace with almost always ends up very conservative. The few times I've actually ran through the check was for some large braced frames in high seismic conditions and some really long spanning trusses.

What I've usually done for these special conditions is find my worst case brace force, find a member that works for it, and apply it to all the locations. Usually whatever member I come up with for bracing "feels" too small and I use something more reasonable (and easier to detail connections with).

 

[WARose]

I usually don't have a lot of worries there.......because unless I've run the numbers, nothing counts as bracing. My approach when I am counting on some bracing (always from the lateral force resisting system) is to superimpose a net (lateral) bracing force at each level (that acts along with the other load cases as per the code) and then design for the outcome. The STAAD model also serves as a check for adequate stiffness for bracing.

Most of the time, it's really no big deal (as you are not going to have full dead, live, and controlling lateral all happening simultaneously)......where I have typically run into situations where it must be accounted for is the big, multi-story process buildings where you are supporting equipment that weighs 100 kips or more. There you might see it control for a member/connection or two.

I have to admit however, for some of the smaller stuff, I don't give it much thought.