Cross Bracing Deflection 5

[TxGraeme]

My office was having a discussion about the design of rod bracing as a lateral resising system. The modeling software we are using (RAM Advanse) checks for bending + axial load stresses, which indicate the member is overstressed. The deflection in the model is over 10 inches, which I'm assuming is the cause of the high bending stresses. However, I have heard that braces are typically pretensioned to eliminate sag (and tensioned even further when lateral load is applied) and the brace straightens out so in reality there is no bending.

It is a light seismic zone and the rods are assumed to be tension only. The building is single story. Does anyone have any experience with this or have a reference we can check? I posed the question to AISC steel solution center but that didn't get me anywhere.

Thank you,
Graeme Sharpe

 

[csd72]

Dave,

The pretension doesnt make any difference for the load, but it does for deflection.

clefcon,

I would be concerned about using circular hollow sections for cross bracing unless they are designed for compression. The stiffness means that they will behave more like a beam and will have bending stress, it also means thatt hey are more likely to have a column type buckling failure rather than just buckling as a rod does.

It is also much more expensive to use circular tubes as these need welded connections each end rather than a rod that can just be threaded.

I have seen neoprene sleeves specified to minimise any banging of the rods.

Otherwise I would suggest angles as noted above.

csd

 

[JAE]

Yes, by adding nodes along the length you introduce stiffness terms at each node and the second order analysis will respond to the pretensioning (in RISA using temperature per DaveAtkins above) by straightening out.

But for most rod X-braces, this is a lot of work and I'm not sure of the benefits of all this effort when simply turning off its individual self-weight provides essentially the same analysis results.

Turnbuckles typically used with rods are used to provide just enough tension to initially straighten the rods and for most conditions this tension isn't all that significant in terms of the design of surrounding beams and columns.

With light gage framing and rod/turnbuckles it may be necessary to consider the tension but I've not had to deal with that before.

Besides rods, we've used square tubes - but not pipes.

 

[UcfSE]

I took the question about the pretension in the rod to mean accounting for the extra tension in the rod itself when checking it along with the applied tension from the lateral load.

 

[Lion06]

UcfSE, that was the way I took it also.

 

[sennafan]

TxGraeme,

1 or 2 ksi of prestress should be plenty to pretension a rod brace. Over tightening and relaxing will help assure the rods ends are set. You should specify either double nutting the rod ends and turnbuckles or damaging the threads or tackwelding the nuts to prevent loosening over time. The big thing to watch for is over-tensioning the system - it does not take much tightening torque to generate high rod stresses.

Specifying how many nut turns after snug is the most controlled way to specify - 1 full turn = 1/thread pitch shortening and PL/AE will tell you the internal stress. for a 30' long 3/4" rod with coarse thread ends that equates to about 3.5 ksi internal stress.

Prestressing will help prevent compression rod buckling. You should reduce the allowable stress of the rod by at least the specified prestress amount.

sennafan

 

[OUe]

I'll chime in on one more thing: it seems to me that the braced-bay underconsideration would have to be pretty darn long in order for 10 inches of deflection to occur. I realize that the kl/r can be ignored for rods, but this seems to be a very, very long rod for there to be 10" of deflection on a single story!!!

 

[csd72]

OUe,

The computer was analysing the rod as a simply supported beam, the 10" was a beam deflection of the rod with no pretension.

What relevance is kl/r for tension members?

csd

 

[Lion06]

The AISC spec says that kl/r for tension members should preferably not exceed 200. I believe that it has something to do with vibrations.

 

[DaveAtkins]

I believe the kl/r for tension members, which is NOT a requirement, is to prevent the "slapping" that was mentioned above.

DaveAtkins

 

[JLNJ]

Providing a kl/r less than 200 also gives some capacity in compression.

Many, many, many times I've seen PEB braces bent, or cut, or missing.

 

[Lion06]

The 13th edition has this value at 300. It also states that this recommendation does not apply to rods or hangers in tension.

 

[UcfSE]

The limit for tension members has always been 300, at least since the 9th edition.

 

[haynewp]

If the rod is overstressed with bending from self weight plus lateral induced tension, how can it not be still overstressed if the bending deflection is taken out by pretensioning and then adding the lateral load? I don't think that removing the density of the rod is solving the tensile overstressed problem you have.

 

[JAE]

...well...if the stress from bending via self weight is 99.99% of the total stress

...then by taking out the false bending you can then design the rod for what its true purpose is.

 

[haynewp]

JAE,

I am not sure if I follow. If the bending is taken out by pretensioning, then you still have the bending stress due to the self weight without the deflection, plus you now have added pretensioning tensile stress. I am confused about the 'false bending'. Isn't this a tension overstress situation?

 

[hokie66]

tunacan,

No, a rod which is sagging is not stressed in bending. It is a catenary, and any stress in it is tension. As JAE said, the bending given by the program is false. Just goes to show, engineers should be smarter than computer programs, and structures are always smarter than both.

 

[haynewp]

It is not sagging when it is pretensioned or when it is under tension from lateral, so I don't think it is acting like a catenary in those cases.

 

[Lion06]

I agree with JAE.
There is no bending stress, it is all tension. While you may need to account for the pre-tensioning in the design, the load is still all tension.

 

[haynewp]

I would love to see the proof of it. A rod that is under a uniform transverse load combined with an axial tensile force, high enough to cause an immeasurable deflection, is in pure tension. I would just bump the rod up one size. But I usually disagree with others on this forum anyway so maybe you shouldn't listen to me.

 

[Tomfh]

tunacan,

It's still sagging when tensioned up, it just isn't sagging by much.

It still has some bending stress too, but you just ignore it.