Ext. Crane Way Support and bracing 2

[levey12]

I am an EIT working on a preliminary design for an exterior crane system. The crane bay is 80’ wide with rail beams that span 30’ and an approximate height of 25’. One side of the crane bay is supported by A-frame columns; however, the other side has obstructions, which forced me to consider different support options.

In the attached document you can see the proposed option. I have been questioning myself about how to distribute the forces to the column system. I initially designed the rail beam support column (col. 1) for all of the vertical wheel loads and then distributed the entire thrust force through the strong axis of the remaining column (col. 2), which resulted in a basic design for a cantilever beam. However, I know that the thrust force will induce moment in both columns. What moment should I design each column for? It is my understanding that less moment will be resisted by the rail beam support column (col. 1) due to its weak axis orientation and thus should be designed for less.

Also, when designing the braces to tie the columns together, they have to be able to resist the compression flange (prevent LTB) and thus decrease the unbraced length (I have been treating Col. 2 as a cantilever beam). Is it OK to design these for 2% of the compression flange force as you would for lateral stability bracing? Currently, I have been using AISC appendix 6 to design them as nodal column braces.
Any suggestions are appreciated.
Best Regards

 

[Stillerz]

There is a lot that can be said here.
First, runway girders with web diaphragm tie-backs are known to cause problems with fatigue. A tieback at the runway girder top flange is a better way to go.

As for how much moment each column takes, it is really a matter or relative stiffness and depends GREATLY on how you tie the two columns together.

I design cranes columns as you have shown as leaner columns with the "bracing" being small simply supported beams. This makes the laod sharing b/t the columns to almost nothing. If the "braces" are moment connections, you'll definitely have moment being shared and will also result in a coupled reaction at the column bases.
Thrust forces are usually not very high compared to wind forces. Of course, you have an outdoor structure which will most likely reduce wind (& make it more complicated).

 

[Stillerz]

It is also very important when designing the crane column (the one supporting the girder) to take into account the eccentric reaction on the cap plate and the moment it induces in the column (strong axis, in this case). With heavy cranes and girders this moment can be very significant.

 

[ishvaaag]

My suggestion is that you make a model of the colum with battes using as elements the tying plates and segments of colums between them. Use P-Delta and apply the forces and restraints. Ensure that if or the (in your drawing) out of plane you are telling the program for the segment members a length between braces in accord with the wanted behaviour according to the restraints. This way you will get appraisal of the benefits of the connection. For more precision, divide even the segments between tying plates in 4 and even give then an in-built initial imperfection. That way you will cover as well P-small delta. For the deformations in the plane of your paper, K=1 for every segment defined at input, whichever its length. Not so for out of plane behaviour, that you control stating restraints and Length between braces directly.

 

[BAretired]

One side of the crane bay is supported by A-frame columns; however, the other side has obstructions, which forced me to consider different support options.

(1) It is not a good idea to use crane columns with vastly different lateral stiffnesses. If possible, design both sides the same way, i.e. use the same detail each side of the craneway.

(2) Turn the inside column so that the web is parallel to the runway beam, i.e. both columns are oriented the same way. Tie the two columns together to act as one combined section. Lateral torsional buckling will not be a problem.

BA

 

[Stillerz]

BA,
The inside column supporting the runway girder does have the web parallel to the girder.

There is nothing at all wrong with having the outside column or "building shaft" oriented to take wind and thrust loads. Design for LTB like you would anything else.

This will be a tough design since there is no roof structure to make a complete moment frame and also no bottom chord bracing which helps share thrust loads between bays in a typical mill style building. In fact, having a free standing runway like this, like BA said, with two different frames is scary.

Typical free standing type outdoor runways are made with A-style frames. This cantilever idea in the PDF will almost certainly result in runway alignment problems due to lateral stiffness.

 

[cntw1953]

Just a few notes here:

First to agree with stillerz that a top flange tie-back is more desirable than the one in the girder web. Ideally, the tie-back shall be stiff enough to transfer crane horizontal thrust from column 1 to column 2, and yet flexible enough to alow the girders movement, due to longitudinal thrust, in the direction of runway. Pre-assembled/engineerred tie-back system is available through Gantrax.com.

I tend to agree with BAretaied on providing identical lateral load support system rather than mixed. However, I think your columns layout is fine as long as you can provide x-bracings in the direction of runway.

Note there are thrusts in two directions, as well as environmental loads for outdoor cranes. Avoid layouts that lead to, or weak in, torsion. Suggest to obtain a copy of AIST Technical Report No. 13 as a guide on your design.

 

[cntw1953]

correction - "Gantrex.com".

 

[levey12]

Thank you all for your comments, they have been very helpful. I agree with all of you that it is ideal to use identical lateral support system rather than a mixed system. I am currently investigating other options that use A frame systems on both sides.

Stillerz,
I also agree with you that free standing outdoor runways are typically made with A-style frames. In fact, All of the outdoor runways that we currently have use A frame columns on both sides for support. If there were no constraints (building perimeter) I would ideally use A frame columns on both sides as well.


cntw1953,

I also have an A frame system at the end of each bay to help with the longitudinal (Braking, accelerating) force parallel to the runway beams. In this case, I believe X-bracing may not be needed if the connections provide the proper resistance. Do you agree?


Does anyone have any useful references for A frame column design? I am currently only referring to "Engineering and Hoist Design" by Shaw-box. Are there any up-to-date references?

Thanks again

 

[cntw1953]

levey:

I am not aware of any runway to have A-frames in direction of runway girders, except mobile/gantry cranes. But, I don't see why it wouldn't work.

I rember a book titled "Derricks & Cranes", or something like/close to that, I would check ASCE website first.

 

[ishvaaag]

Code NBE EA-95 gave K (=beta therein) factors for the compressed columns.

http://tocasa.iespana.es/descargas.htm

look p. 121 in the pdf, the K values are at the second column of the table Tabla 3.2.4.3C, there's a misrepresentation at the headers of the columns in the table.