topic Re: truss design, instability and displacement in Robot Structural Analysis Forum

truss design, instability and displacement

Anonymous — Fri, 17 Mar 2017 23:58:19 GMT

Hi,

I have a steel truss, and have defined every bar as a truss (geometry/additional atributes/advanced bar properties). I am getting problems with the stability when I run the results (instability in every join).

An other thing that I want to know is when I apply windload my displacement is veeery big lik 6789987mm.. does it have something with realises to do?

Thanks!

Re: truss design, instability and displacement

Rafacascudo — Sat, 18 Mar 2017 11:28:54 GMT

If it´s a plane truss you don´t need to define truss bars , just choose the correct structure type when opening Robot.

Re: truss design, instability and displacement

Anonymous — Sat, 18 Mar 2017 11:49:30 GMT

Thanks for reply. But what do u mean with correct Structural type? If I dont chose truss I will get moment. What about windload, the displacement is huge.

Re: truss design, instability and displacement

Rafacascudo — Sat, 18 Mar 2017 12:06:35 GMT

Can you send your model?

Re: truss design, instability and displacement

Anonymous — Sat, 18 Mar 2017 13:02:56 GMT

Can I send to your mail? or private message?

Re: truss design, instability and displacement

Rafacascudo — Sat, 18 Mar 2017 18:17:53 GMT

Received your file . Your answers

1- It´s "shell" structure type , so all DOFs are available for the bars nodes . So ,for the plane trusses you have on your structure , at least the chords cannot be truss bars!!! . The top chords are linked to the diagonals and the bottom chords have also the diagonals and the beams that transfer the loads to the trusses attached . So ,they cannot be truss bars!!

I made the chords work as normal bars and all the instabilities warnings (and consequent huge displacements) are gone. Instabilities type 1 and 2 cannot be ignored. Watch the webinar about it (and all the others) on Build your Robot Structural Analysis IQ - YouTube

2- No more huge displacements , But your wind load load is giving 212mm displacement on top node. probably due to the lack of horizontal truss linking the trusses top chords . Bucking will be a big factor when designing the top chords

3- Moving loads (alone or in a combination) generate 2 additional load cases(+/-)which are envelopes for all generated moving load components .

Other than that ,I think your moving load step should be 0,5m instead of 1m and your route should be extended in 2 meters so You can "see" your vehicle entering the bridge.

File corrected , as I think it should be , sent directly to your email

Re: truss design, instability and displacement

Anonymous — Sat, 18 Mar 2017 18:27:44 GMT

Thanks!!

Re: truss design, instability and displacement

Anonymous — Sat, 18 Mar 2017 22:38:58 GMT

2 more things;

1. u said that the chords should be "normal bar", can u explain why?

- if the chord is a normal bar then the chord gets a moment.. and thats wrong for a ideal truss?

2. How can I design the top chord for buckling? the chord are not one beam, its devided into 10 beams.. do I need to use superbar, if yes.. how?

Thank you!

Re: truss design, instability and displacement

Rafacascudo — Sun, 19 Mar 2017 10:26:57 GMT

1- There's no ideal truss on the real world. Nobody will build it using a pinned conection on each truss node. And even if they do , you would have loads on midspan due to selfweight and wind at least ruining the " ideal truss"

2- For the out of the truss plane buckling , your top chord is totally free , unrestrained. So you will have a huge buckling length and you will have to use a pretty big section to be obbeying the design codes. And it will be worse because the top chord will always be in compression

Re: truss design, instability and displacement

Anonymous — Sun, 19 Mar 2017 11:45:48 GMT

Thanks, but:

1-I understand the thing about ideal truss, but Why chose normal bar instead of "beam"?

2- when I calculate the tension and compression force on the truss and the chords by hand calculation, can I use same method (method of joints, method of section) as calculation ideal truss?

3-the file u sent, it was (Y) But its a big different between tension/compression on chords and the "truss bars", right?

4-the top chord is devided in many bars, how to check buckling for the total chord?

thanks for reply!

Re: truss design, instability and displacement

Rafacascudo — Sun, 19 Mar 2017 12:22:53 GMT

1- By normal bar , I meant a bar without advanced properties . Beam , columm , they are all normal bars

2- Yes ,You should get aproximately similar results

3- Not sure of what is the question , but yes forces in the chords are a lot greater than in the diagonals

4- check the excelent steel design webinar (3rd webinar) on Build your Robot Structural Analysis IQ - YouTube

You will have to set code paramenters for each of your bars through here.

Re: truss design, instability and displacement

Anonymous — Sun, 19 Mar 2017 15:06:21 GMT

The link u send me was useful, but I am wondering about several things when it come to buckling:

1- What is the difference between continues chord and divided chord when controlling the section?

2- In the file you sent me, are the chords continous since you changed them to normal bars? does this affect buckling?

4- Should I look at all the top flange at the same time when controlling buckling? or separatly for each element?

5- for normal bars, which buckling diagram should I use?

When I control each element separatly for buckling, the section is ok.

but how can I control the whole model for buckling (global buckling)? And how can I know when its okey?

Thanks again!

Re: truss design, instability and displacement

Rafacascudo — Mon, 20 Mar 2017 11:44:38 GMT

1- Don´t know what you mean by " controlled section" , But in Robot all bars are born "fixed" with all DOFs restrained . The results for a 10m bar is the same for 10 x 1m bar .Dividing a bar in several pieces doesn´t make any difference on the results regarding internal forces and displacements.

2- On a 3d structure type enviroment , You wouldn´t be able to perform any kind of analysis(including buckling analysis), setting al bars of the truss as "truss bars" . Structure would be unstable and you would get several instabilities warnings as you did receive.

On an "ideal" 2d truss , yes , all nodes would be , by definition ,restrained on the transversal (out of plane) direction and the chord buckling lengths would be the length of the bar between 2 nodes.

4 ,5 ,...- For the top chord I think that conservatively you should consider the length of each bar for the in plane buckling (Ly) and the whole arc length(41,06m) for the out of plane buckling length (Lz).

for the diagonals , between 0,7 and 1 as K factor .

For a global buckling analysis ,you can do a buckling analysis. I did it considering only the self weight load case . The 1st buckling mode is of course the expected one (Top chord lateral sway)and gives a 2,76 critic coefficient.

Re: truss design, instability and displacement

Anonymous — Mon, 20 Mar 2017 17:10:26 GMT

Hey, thanks for Your reply.

1. Do I have to Control the top chord alone for out of plane buckling or do I have to control the whole structure for out of plane buckling (Global buckling analysis) ... or do I have do both

2. When I Control for global buckling I have to change my load type (moving load) to linear buckling. Why dont I get any Critical coeffisient for moving loads?

Re: truss design, instability and displacement

Rafacascudo — Mon, 20 Mar 2017 19:12:00 GMT

1- Both . You have to check all structure bars against the code of your preference . And you can use the Robot buckling analysis ,for example , find a more precise out of plane buckling length for the chord bars.

I suggested the whole chord lenght (41.06m) ,but seeing the buckling analysis results , it is giving around 16,3m for Lz in the 2nd mode , which is the mode cleary affects only the top chord. And seeing the deformed shape of this mode with its double curvature ,it appears that this value for Lz is not out of reality

Not very familiar with buckling analysis though . Maybe Artur ,Rafal or Pawel , could give us a hint if in this case the Lz value is a good way to determine a trustable out of plane buckling length.

4 - For moving load you probably should create a case from a component and combine with your other vertical loads. There is no point analysing buckling for a moving load case alone.

Re: truss design, instability and displacement

Anonymous — Mon, 20 Mar 2017 19:38:40 GMT

Thanks man, really helpful!

coment on your answer:

1 - Understand your. Hope they see this post.

2 - regarding movable load, I have not chosen components and then combined it with vertical load.. but I have just combined the moving load with vertical static load and have not chosen components.

When I run this combination in steel design (for optimization), Robot tells me which is the most unfavorable position of the car due to that combination, tells me which case it is giving the optimization for... thats right way to do it? or should I run only the moving load first, and then creat a case and then combine it with other vertical load? (this is for steel design)

When it comes to buckling analysis, how can I know which combination is the worst due to structure buckling? or which case from the moving load..

Re: truss design, instability and displacement

Rafacascudo — Mon, 20 Mar 2017 20:48:24 GMT

yes , for steel design you don´t need to "create a case from a component".

For buckling , probably it will be the component that gives the chord its highest compression force

Re: truss design, instability and displacement

structural.engineer79 — Tue, 15 Jun 2021 21:57:04 GMT

Pleasr If changed bar from advanced bar property to truss , how can I changed to normal bar

Re: truss design, instability and displacement

Rafacascudo — Wed, 16 Jun 2021 15:13:13 GMT

unmark everything on "advanced bar properties" , select/list the bars and click apply

Re: truss design, instability and displacement

structural.engineer79 — Wed, 16 Jun 2021 16:20:31 GMT

For the Top chord of truss the buckling length will be equal the distance between purlin, but the bottom chord if you used longitudinal truss the buckling length will be equal to the distance between longitudinal truss