Steel Lattice Tower Design Guides 1

[CBEngi]

hello,

I'm a Structural Engineer that has been tasked with designing several steel lattice towers. They are approximately 21m in height.

The towers are in Canada and I will be designing to the Canadian Codes. I'm wondering if anyone has any good references for the design such as articles, examples, books etc.

Thanks in advance.

 

[transmissiontowers]

I'm not real familiar with the Canadian Codes, but you should get ASCE 10 and a copy of the PLS-CADD software TOWER which has all the design codes built in, AFAIK. As a minimum, get PLS-CADD Lite and TOWER. You use PLS-CADD-Lite to develop the load cases because it includes Sag and Tension software. If you have several towers to do with multiple leg and body extensions, the TOWER software now has a Family Manager where you can analyze many versions of the tower family. This is their web page:

http://www.powline.com/

If you've never done T-Line tower design, there is a learning curve and our industry is unlike most others. You might try to locate the proceedings of the last few SEI conferences

http://www.etsconference.org/

that had several papers about towers.

How much time can you devote to learning the process? Most of the methods are based on testing results from full scale tower tests and lots of empirical data. For many years, the towers were designed by hand as 2D trusses with many assumptions so the method of joints would work. I started out designing towers with the graphical method in 1973 and have been amazed at the advances in software since then.

After the towers are designed, the fun is just beginning. Are you also tasked with detailing the tower so it can be fabricated? There are a few places in India that have software that does the tower details. There is one guy I know in Pittsburg that does tower details in AutoCAD that does very good work. His web site is here:

http://www.towerdraft.com/

If all this seems like a daunting task, you might consider farming it out to some of the big T-Line fabricators that also design towers.

http://www.jyotiamericas.com/

is one outfit but they only use equal leg angles in their design.

http://www.saetowers.com/

is another fabricator.

http://www.falconsteel.com/

is a fabricator that I think still does towers.

http://www.abeinsa.com/web/en/acerca_de_nosotros/

does tower design.

Post back here and let me know how it goes.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[CBEngi]

Thanks for the reply Transmissiontowers.
I have purchased a copy of ASCE 10-15, which is very similar to the Canadian CSA S37-13.

The line forces have been provided to me by the design team responsible for the line design. Thus I am only responsible for the tower loading/design, and plan on using RISA 3D for the tower analysis.

The project will only require several towers to be constructed, thus achieving 100% efficiency is not a major concern at this point. I will be responsible to provide drawings with connection details for fabricators to construct. We have drafting in house and I feel confident in our abilities to deliver drawings and specifications.

I feel comfortable with the analysis portion of the project; it the part of not exactly knowing what is typical of the industry that makes me uncomfortable at this point.

I was really hoping that someone would make reference to a general book that goes through the various stages of tower design, something similar to the 10+ books I have on my shelf about buildings and bridges.

 

[transmissiontowers]

I run GTSTRUDL for my substation structures that carry moments and I'm not too familiar with RISA. You may be able to do the analysis in RISA but I'm not sure if there is a code check to check the angles per ASCE-10 so you may end up running a spreadsheet to check member and bolt capacities. Pay particular attention to the redundant bracing patterns and account for the redundant area when you do the wind on tower calculations. The corner post leg members are usually done as beam elements with the lacing members and redundant members as truss elements.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[CBEngi]

Quote
"The corner post leg members are usually done as beam elements with the lacing members and redundant members as truss elements."

Transmissiontowers,

When you mention the above statement, does this mean that you generally neglect bending forces caused by wind/ice in your secondary framing members(lacing members), treating them as compression/tension only and ignoring the beam-column action all-together?

thanks.

 

[transmissiontowers]

Well you have now entered the convoluted analysis of T-Line towers. ASCE 10 and it's manual of practice processor (MOP 52 IIRC) were written by the old timers in our industry (even older than me) and most of the formulas were based on full scale testing. Before FEM and computers, we analyzed towers as 2D trusses and calculated loads to be placed on the joints. Wind on the member loads were computed and placed as joint loads and assumptions were made so we could analyze indeterminate trusses with the method of joints using a graphical analysis. Wire loading was put on the arms and in some cases the arms were analyzed separately with their reaction loads placed as joint loads on the main tower body.

Stepping forward to modern software, we still model members as truss elements where possible because the allowable member load equations were derived for truss members from the full scale testing done years ago. Look at the tension capacity for lacing members bolted on only one leg. If the tower has a square body where all 4 faces are the same, the truss assumptions are easier to use. When the tower is rectangular and you have 2, 3, or 4 panels on the narrow faces for one panel on the wide face, the analysis gets complicated. In the old days we just analyzed each face as a 2D truss and added the leg loads together. With 3D computer programs the leg joints are unsupported in one plane so you have to make them beam elements because truss elements would be unstable unless you add dummy or fictitious members to brace the out of plane motion.

To answer your simple question, yes, when doing a tower analysis, we throw away the bending moments in the members when we check the individual member capacities for tension and compression.

Then we have complicating members like Tension-Only members that take a little compression but are allowed to buckle out so the forces are redistributed into the other members when we run the nonlinear analysis. You can end up with an unstable geometry where the model will fail to converge.

Traditionally only the main leg and lacing members were put into the model and the redundant bracing members were left out and designed for some percentage of the leg load based on the full scale tests but mostly sized for their L/r. More recently, the redundant members have been modeled in the tower using the PLS-CADD software, but the more redundant members you add, the more joints the lacing and leg members have to have to attach them to the structure. The benefit of adding them in, is the software can compute the wind on the members so you get a closer approximation of what happens in the real world with wind on the structure. If you analyze just the main members and apply wind on the tower, you miss the wind on the redundants that are not in the structure so you add drag factors to the section to account for the missing members.

Depending on location and tower height, the wind on the structure can account for about 20% of the leg reaction loads in a hurricane region for normal 100' to 200' tall towers. For very tall crossing towers (I designed one recently that was 332' tall to cross a ship channel) the wind on the structure can account for 80% of the leg reaction load.

_____________________________________
I have been called "A storehouse of worthless information" many times.