What would you consider the unbraced length of this double angle to be?

[TNich]

If I have a typical braced frame where the braces are double angles, if I use a spacer plate with a through bolt at mid-span to connect the double angles, can I consider them braced at mid-span for compressive purposes? I would assume that I could because one is in tension and the other compression, so I could make an argument saying they will not both buckle. Before I assume this, I would like to know if there is any definitive information that leans one way or another on this situation.

Here is a quick picture of what I would like the connection to look like, if it helps. It's circled in red. Any input is appreciated. I know the picture is big, I just could not figure out how to reduce it.

 

[slickdeals]

There's an AISC article on this subject by Subhash Goel on the subject
Read this thread

http://www.eng-tips.com/viewthread.cfm?qid=258454

 

[jrbaus]

What happens if you analyze only the member that is in tension and neglect the compression member? I typically neglect the compression member in bracing sets comprised of slender members such as angles. In saying that, you need to ensure you treat the members as "tension only" in your analysis package or alternatively delete the compression member since it will still attract load under a static analysis.

I doubt you would see enough stiffness at the point where the cross bracing meets to assume any considerable lateral restraint. You could model this in a frame analysis program and do a buckling run if you want but I suspect buckling load factors will be quiet low for any significant racking force applied to the frame. Best practice would be to assume that only the angle member in tension is effective in my opinion (without knowing any details of the actual project).

Jake

 

[KootK]

Consider the braces braced at the crossing. It has been a contentious issue in the past. I'm currently taking AISC's night school class on stability. A couple of weeks ago they touched on this. Apparently, using the modern stability methods, it will virtually always be the case that we can consider these joints as laterally braced.

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.

 

[jrbaus]

KootK,

Interesting point you make there. I would appreciate any references you might be able to point me towards. All the angle cross bracing sets in large industrial structures I have reviewed used tension-only members in their models.

 

[jrbaus]

Sorry,

Just seen slickdeals post with list of references. Most appreciated.

Jake

 

[KootK]

No sweat jrbaus. Most of what I've seen/done had also been tension only. It's a brave new world stability wise!

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.

 

[Hokie93]

The ability of a tension diagonal to provide out-of-plane bracing for a compression diagonal was discussed in a recent (Third Quarter, 2013) AISC Engineering Journal paper by Eric Lui and Xiaoran Zhang ("Stability Design of Cross-Bracing Systems for Frames"). The paper is available on the AISC website (

http://www.aisc.org).

In the two examples provided at the end of the paper, the compression diagonal is only partially braced by the tension diagonal. This paper would seem to cast doubt on a blanket assumption that the tension diagonal can always be relied upon to provide mid-span, out-of-plane bracing for the compression diagonal.

 

[TNich]

Thanks for all the responses. Looks like I have some good thoughts and some good material to look through. More thoughts or references are always welcome, the more material the better... sometimes.

 

[canwesteng]

I always neglect any stiffening effect tension in one brace might have on restricting buckling of the compression member out of plane. With angles, you aren't actually bracing either of principal axis, you're bracing somewhere in between. So it's something like a Lx=Ly=0.7L. With double angles, you are bracing one of the principal axes, but not the other, so Lx=L and Ly=0.5L (depending how you define your X and Y axis).

 

[KootK]

@Canwesteng: your last comment has been nagging at me for a couple of days. Firstly, because I've never checked principal axis buckling issues as you've described. Secondly, I'm not sure that I agree. To elaborate:

With angles, you aren't actually bracing either of principal axis, you're bracing somewhere in between. So it's something like a Lx=Ly=0.7L.

I would argue that, if you have bracing about the two geometric axes, then you have effective bracing about all possible axes, including the principals. See the sketch below.

With double angles, you are bracing one of the principal axes, but not the other, so Lx=L and Ly=0.5L

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.

 

[KootK]

Messed that up a bit:

Quote (Canwest)
With double angles, you are bracing one of the principal axes, but not the other, so Lx=L and Ly=0.5L

As long as the angles are stitched together to buckle compositely, as they normally are, I think that the principal axes are as shown in the section in my previous post. And they're directly braced.

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.