how to veterans typically approach modelling/analyzing this kind of structure? 4

[teach_me]

im trying to learn to analyze with this truss locally but then i realized that i need to incorporate it in the whole frame because its the only link to columns on the side direction (save from the tension rods). Im so confused. How would you approach this?

 

[milkshakelake]

You could just analyze it in isolation. I don't see any reason it's not just a 5 pin truss. Doesn't look like it takes any moment, so it doesn't need to be modeled together with the columns.

If you're using FEM, this will be a pain and a half to model with the whole structure. There are lots of things that could happen, like instability that's hard to track down and changing member sizes or dimensions for a lot of repetitive members. So I'd just model it as a beam with equivalent stiffness as the truss. And design the truss independently either in FEM or using hand calcs; it's nothing unusual.

Are you saying that these trusses are moment frames? Because I wouldn't design it that way. That's a whole can of worms. The building should have a lateral system. Another option, not great, is to design the columns as cantilevers from the foundation. If you must make the joists moment frames, you would have to design it together with the whole building and pay special attention to the moment transfer between the truss and the column, which I don't really advise.

 

[Celt83]

most modeling packages allowing copying elements.

I tend to land on modeling one of the trusses with pin-roller supports once that modeled has all of the member hinges set properly then expand to include the columns down, then copy n times to create the full building model.

 

[dik]

To start with, I'd change the column heights and keep the truss parallel chord. Why do you need the sloped roof?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[HTURKAK]

Some points which is not clear for me;

- Are the columns SC-1 at Axis 1,3,6 supported on the retaining wall? ( which i will not advise )

- Are the trusses moment frames (trussed beams ) ?

- Are the columns fixed to the base ?

I would suggest , the use of portal frames ( the spans are 8 m and suitable for the standard hot rolled sections ) with moment connection and simple conn. at base . The columns shall not be supported on retaining walls ..Provide vertical and horizontal bracings ..




My opinion..






Not to know is bad;
not to wish to know is worse.

NIGERIAN PROVERB

 

[Standard American]

I would model the truss in staad or some other similar program. Make sure you apply your releases to the web members so you get accurate truss action. The columns could be modeled as pins and rollers. You could also add a support in the lateral direction for your bridging. Typically when modeling something in staad with a lot of nodes I'll use the structure wizard or change my units in my CAD drawing to decimal feet then pull a dimension line in x and y starting at my zero then just move the other grips to each intersection. The model for this truss would only take a couple minutes. A guy I used to work with made a CAD lisp routine that would spit out all the coordinates where he could just copy and paste into staad. That was a little over my CAD capabilities though.

 

[JoshPlumSE]

I'm an advocate of modeling and designing in steps.

Step 1) Model the Truss along (pinned and Roller) as Celt83 suggested. This smaller model is easy to investigate. So, run all your vertical loads and optimize for deflection and stress. Make sure you're comfortable with any program inputs related to design (unbraced lengths or such). You might even run it with P-Delta analysis to make sure you don't have any buckling. Same this model as it's own file.

Step 2) Add in any columns or such if you think that's going to affect the design. Re-run the analysis. Compare to the results you got from the model in Step 1. You might have to modify the design slightly. Maybe not. At this stage, I would generally only up-size elements... not down size them. Definitely include P-Delta effects. Again, save this model as it's own file.

Step 3) If this truss is going to participate in the lateral force resisting system (maybe one of the chords acts as a drag strut), you can build the rest of the structure around the model in step 2. Apply your lateral loads and see how that changes things. My hope is the only thing it changes is the drag members. Again, at this point, I would tend to up-size members only, not down size them.