Tension Only X-Brace (Slendereness Ratio) 1

[EngSD]

I am wondering in checking a tension only X-Brace with single angles for the slenderness ratio recommended by AISC of L/300, would you use the entire length of the angle or half the length (the X-Brace is bolted at the center)?

I have heard it both ways at the office and would like to get clarification

 

[sandman21]

Full length, I try to not relay on catenary action from the other Lx

 

[frv]

The entire length, in my opinion.

The limits are recommended for handling, erection and vibrational issues (see the commentary to chapter D).

Setting aside the erection and handling issues, if the contention is that IN USE, the compression leg of the X-Brace will brace the mid point of the tension, I would argue that by definition, the compression leg is assumed to buckle elastically in a tension-only brace, thus permitting the tension brace to carry the full load.

BTW.. I believe tension-only bracing is only allowed in seismic design category A.

 

[sandman21]

OCBF can have tension only in any design category.

 

[frv]

sandman21-

I rarely do any seismic design, so I'll take your word for it. I thought I had read somewhere that this was either not permitted or discouraged above SDC A, but I guess the difference is in "Ordinary" and "Special".

 

[JAE]

The problem with tension only is that you concentrate all of the lateral and uplift force at one column .... and that means the ugly monster that shall not be named (App D) rears its ugly head and you end up having a bad day.

I used to use half the length - we just had a thread on this topic a week or so ago - did some research - AISC doesn't have a solid answer on it and the articles they referenced were all over the map.

We ended up using 80% of the full diagonal length as the unbraced length but I wouldn't have a problem with ignoring the tension brace and going full length.

 

[hokie66]

Not familiar with AISC these days, but using rods for diagonal bracing is common practice where I am. Is this prohibited by the AISC provision?

 

[frv]

JAE-

The OP is talking about unbraced length of the tension member, not the compression member.

 

[BAretired]

I was under the impression that AISC recommended 300 maximum but there is no maximum slenderness limit for tension members.

I do not like to use a tension-only X-brace for the reasons given by JAE but when I use compression/tension X-braces I consider the effective length of the member to be one half the length of diagonal.

BA

 

[frv]

BA- you are correct.

There is no maximum for compression either. The language has changed to a recommendation.

 

[JAE]

frv - thanks - didn't catch that. Not sure why you would wonder about the L/ratio for tension member.

AISC has always stated that the compression and tension L/ ratios are "preferred" and recently they've come out even more loosey goosey on it.

I stick with the L/200 for compression but don't usually worry about the tension member.

hokie66, AISC doesn't prohibit rods except if you use what they call a "Special Concentric Brace". This is a higher ductile brace system used in high seismic areas where they state this:

[red]13.2c. Lateral Force Distribution
Along any line of bracing, braces shall be deployed in alternate directions such that, for either direction of force parallel to the bracing, at least 30 percent but no more than 70 percent of the total horizontal force along that line is resisted by braces in tension, unless the available strength of each brace in compression is larger than the required strength resulting from the application of the appropriate load combinations stipulated by the applicable building code including the amplified seismic load. For the purposes of this provision, a line of brac ing is
defined as a single line or parallel lines with a plan offset of 10 percent or less of the building dimension per pendicular to the line of bracing.[/red]

 

[sandman21]

Yes no tension only when designing SCBF

 

[spats]

When you're dealing with angles, slenderness about 3 axes needs to be considered, x, y & z. x & y are not necessarily the same because it can be an unequal leg angle. I would use an angle with a larger outstanding leg, because Lx is the full length of the brace. Ly is one-half the length because the angle is braced at the center by bolting. Lz is also one-half the length because in order for one angle to buckle about it's weakest z-axis, the opposing anngle would have to buckle about it's strongest axis (90 degrees to the z-axis).

 

[ToadJones]

May be off topic, but isn't the 1/2" length argument only present in Compression/Tension X-bracing?

The first office I worked at considered the compression member in the X-brace to be braced at the midpoint connection. The argument was that since while the compression brace was in compression the other member in the "X" was in tension and would tend to keep the compression brace from buckling out-of-plane.

We also would arbitrarily reduce the length of the members to something like 90% of the node-to-node length to account for connection geometry and the fact that the node-to-node length was the worst case.

Tension only bracing is perfectly acceptable and problem free in low rise buildings and/or buildings with very large aspect ratios in terms of footprint. Essentially, if the brace loads are small, shear and uplift at the foundation level are not usually a problem.
In tall buildings with small footprints, tension only is surely not as feasible.

 

[ToadJones]

...I have also sized tension only bracing based on keeping KL/r under 300 when loads were extremely low...as in the case of cutting the unbraced length of large wind columns.
could be wrong here, but sometimes your left with a design that is essentially based on KL/r.

 

[JAE]

ToadJones - we just did a literature search on the concept of using the tension brace to laterally brace the compression element.

It sometimes can do the job but depends on a lot of factors.

What I did learn was that you cannot always say the tension brace adequatly does the job. Do a search on Modern Steel Construction and there are some Q/A responses that lead you to some articles that AISC publishes.

We've scaled back to use 80% - 100% of the total compression length in our compression brace design - with a k = 1.0.

 

[apsix]

From ASCE 10-97 Design of Latticed Steel Transmission Structures

"Tension/Compression system with members connected at the crossover point. If the member in tension has a force of not less than 20% of the force in the compression member, the crossover point provides support resisting out-of-plane buckling.

 

[frv]

apsix-

The OP is talking about the unbraced length of the tension leg, not the compression leg.

 

[apsix]

Yes, but the discussion had morphed into including compression bracing effective lengths.

However, to return to the original topic, and from the same code;

"Tension-only member: member with L/r GREATER than 300, which is assumed to be unable to resist compression."