Bracing columns with other columns 2

[PEFLWI]

In Appendix 6 of the 2005 code there is a distinction made between relative bracing and nodal bracing. In the commentary the nodal brace is shown braced against a “rigid abutment”. In an existing structure I am attempting to brace a column with a large KL/r against other columns by applying the brace force to the adjacent columns. See attached sketch. It appears that this can be considered a relative brace and the brace to be transferred to the other columns would be 0.004Pr.

Also in Appendix 6 is a required brace stiffness. For an axially loaded brace is the stiffness A*E/L?

A= area
E = 29,000 ksi
L = length of the brace

Thank you for your assistance.

 

[SteelPE]

If columns were large enough I don't see why it is not possible. You would need to check columns 3 4 and 5 for the weak axis moment created by the point load.... and depending on the loading condition you would need to check the columns against the interaction equation.

In terms of brace stiffness. In your instance AE/L would only be part of the equation.... and probably a very small part. You would also need to consider the weak axis deflection/stiffness of columns 3 4 and 5 in order to satisfy this requirement.

 

[JAE]

Don't know if any of these things will help. Dr. Yura from UT Austin, TX has developed similar analogies to this.

http://www.scribd.com/doc/112524281/50/COLUMN-BRACING-PROVISIONS

(see page 304)

http://www.crcnetbase.com/doi/abs/10.1201/9781420039931.ch32

(have to pay $20 for 72 hour access to this book)

http://www.aisc.org/uploadedFiles/C..._Webinars/AISC5 useful stability concepts.pdf

(a short paper by Yura)

 

[briancpotter]

You are correct, this is a relative brace system per appendix 6. See this thread:

http://www.eng-tips.com/viewthread.cfm?qid=194560

You're also correct re: the required brace strength. For a relative system on a column, Pbr = 0.004Pr per A-6-1

However, the required brace stiffness is 2*Pr/(theta*Lb), where Lb is the distance between braces, not the brace length. (eq. A-6-2)

Also, this is for the bracing of a single column. For an overall stability check of the whole system, consult Chapter C.

Brian C Potter, PE

http://simplesupports.wordpress.com

 

[briancpotter]

Realized you were asking about actual brace stiffness, not required brace stiffness. You're correct again, Stiffness (P/delta) = AE/L. See here:

http://www.modernsteel.com/Uploads/Issues/April_2009/042009_si_web.pdf

Brian C Potter, PE

http://simplesupports.wordpress.com

 

[briancpotter]

Hrm SteelPE, wouldn't the stiffness of requirements of the adjacent columns already be taken into account by using a relative braced system?

Brian C Potter, PE

http://simplesupports.wordpress.com

 

[briancpotter]

I keep monopolizing this topic, but so far my answers have left a lot to be desired.

After further research, it doesn't look like this is a relative bracing system at all. A relative bracing system is one braced to at least two points along the unbraced length. So a diagonal going from one floor to the floor below would be a relative brace, but a system spanning straight across like this would not be. Zieman gives a good way to tell if it's a relative brace or not here:

http://books.google.com/books?id=DD...A#v=onepage&q=relative bracing system&f=false

Quote: "If a cut everywhere along the braced member passes through the brace itself, then the brace system is relative".

The system you have here, where the buckling of one column requires all the columns to buckle, is a "lean-on system". It doesn't seem to be found in AISC, but an explanation (and solution) to it can be found in the linked book. Essentially, what must be done is to ensure that the buckling load of the multi-column system is higher than the buckling load of the braced length of a single column. Scroll down to pg. 543 for an example calculation, it's actually rather elegant.

Sorry for the conflicting and confusing previous answers.

Brian C Potter, PE

http://simplesupports.wordpress.com

 

[BAretired]

PEFLWI,
I don't know Appendix 6 or the 2005 code, but would comment as follows:

1. The frame you show on the sketch is unstable with each end of every column pinned.

2. If the tops of Columns 3, 4 and 5 are braced against horizontal movement, then the junction of Columns 1 and 2 could be considered a braced point if Columns 3, 4 and 5 are stiff enough.

3. Columns 3, 4 and 5 must carry the brace force collectively while each carries its own axial load magnified by P-delta effects combined with bending.

BA

 

[PEFLWI]

Thanks everyone. I will sort through the information. Thanks especially to JAE and Briancpotter for the articles.
BAretired: I am only showing a portion of the framing for simplicity. There is diagonal bracing that braces the columns at the 145'-0" elevation. My main concern is bracing columns 1 & 2 with columns 3, 4, and 5 and what brace force is required.

 

[BAretired]

PEFLWI,
I'm not sure what you mean by the 145'-0" elevation, but I will assume all columns are laterally braced at the roof level.

There are two requirements for the brace in the Canadian code. One is strength, the other is stiffness. I believe the U.S. code is similar, but I don't have that code at hand.

The strength requirement in CSA S16 is 0.02 times the compression in Column 1. The stiffness requirement is that the displacement due to that brace force must not exceed the assumed initial misalignment.

BA

 

[haynewp]

Galambos has written some on lean-on bracing that may be helpful as well.