topic Re: Continuous truss chord bending moments in Robot Structural Analysis Forum

Continuous truss chord bending moments

Anonymous — Wed, 09 May 2018 17:37:45 GMT

Hello everyone,

I am a new user of RSA and new to the forum as well, where I have spent the last couple of days reading tons and tons of great useful info. However, I am afraid I have not been able to find a proper solution for the problem I am encountering.

I am currently working on a 3D truss frame model to design a pedestrian footbridge.

The problem is that it looks as if nodes in the truss chords would be supported considering the bending moments that appear.

I have set horizontal struts (top & bottom) releases as pinned-pinned and diagonal bars as truss bars but it does not solve the problem.

I have also a couple of Type 2 instabilities (that I do not figure out yet how to solve) but I believe they are not related.

I am completely sure that it is a problem of bar releases but cannot see the solution clearly. I do not know which is the proper way to model the truss chords so if anyone could give me some assistance or refer me to another topic would be great.

Any help would be appreciated,

Thank you very much!

Re: Continuous truss chord bending moments

Simau — Wed, 09 May 2018 20:46:57 GMT

hI @Anonymous

There is nothing abnormal about bending moment on upper and lower chord.

This is due to self weight for the upper one and to to SW and sobrecarga for the lower one.

IMHO, you don't need any releases for this kind of structure. Without releases there is no instability.bendin

If you don't want big bending moments on the lower chord, introduce loads directly in the nodes.

Re: Continuous truss chord bending moments

Artur.Kosakowski — Thu, 10 May 2018 08:07:22 GMT

Hi @Anonymous

Try to block both RX in the definition of releases you applied.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.

Re: Continuous truss chord bending moments

Anonymous — Thu, 10 May 2018 09:45:57 GMT

Hi Simau, thanks for your answer.

I know there should be bending moment on both chords - that is perfectly fine.

However, I do not believe that the bending moments appearing are realistic. I mean, there is a peak of negative bending moment in each node which leads me to think that the chord is being supported by the horizontal bracing bars, each node acting as a support.

I do not want to model it that way. In any case, the horizontal bracing bars should be supported by the truss chord, not the other way. What I need is a way to tell RSA that I want the truss chord to act as a single beam, but I can not find the how (and I am sure it is possible).

There might have been a slight misunderstanding with my first post - sorry for that.

Cheers

Re: Continuous truss chord bending moments

Anonymous — Thu, 10 May 2018 09:56:13 GMT

Hi Artur, appreciate your answer.

I have managed to avoid the Type 2 instability (Uy) by removing the bracing bar releases and the truss bar consideration in diagonals. The solution you suggest does not seem to change the bending moments on truss chords and, what is more, it brings the instability back.

I must admit I am not very confident with adding bar releases, at least in this model, because nodes represent rigid unions and I might be wrong but I believe reactions on each of them do exist.

Re: Continuous truss chord bending moments

Artur.Kosakowski — Thu, 10 May 2018 10:32:02 GMT

Hi @Anonymous

My proposal was not intended to 'remove' bending from chords (which has to be as they are continuous) but to eliminate them from posts and diagonals. Provided there are no excessive releases there should be no instability reported. Could you attach your model ?

Re: Continuous truss chord bending moments

Anonymous — Thu, 10 May 2018 10:41:11 GMT

Hi @Artur.Kosakowski

The instabilities remain after the changes.

Find the model attached below.

Re: Continuous truss chord bending moments

Artur.Kosakowski — Thu, 10 May 2018 11:11:05 GMT

Hi @Anonymous

You can set all posts and diagonals as truss bars providing you have the chords (including 'end' diagonals) continuous when you prevent horizontal movement of the top of the truss by additional bracing (truss bars 1 and 2). Otherwise you have relay on bending in the vertical elements of the truss to prevent it from collapsing sideways.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.

Re: Continuous truss chord bending moments

Anonymous — Thu, 10 May 2018 11:50:30 GMT

Hi @Artur.Kosakowski

I see the modifications you did.

Setting all bars as truss bars (except for the chords and 'end diagonals') seems to work perfectly nice and it avoids instabilities - and without considering bar releases which I personally prefer not to use.

At the same time, I can not count on the additional bracing you provide on both truss ends because it would affect the corridor and block the clearway of the footbridge. However, I have removed it and the structure does not seem to be compromised.

In reference to the chord's bending moments, don't you think that there should not be any negative bending on the nodes? I mean, the chord should not be supported by the transversal elements so I do not see this behaviour.

Cheers

Re: Continuous truss chord bending moments

Artur.Kosakowski — Thu, 10 May 2018 12:11:25 GMT

Hi @Anonymous

Indeed my mistake. I introduced them at the stage when I had the last diagonals defined as truss bars. They are no longer needed when these diagonals are not truss bars.

Could you attach a screen capture showing what you mean?

Re: Continuous truss chord bending moments

Anonymous — Thu, 10 May 2018 12:13:01 GMT

Oh and another issue,

Would you recommend to apply the live load as set of nodal forces instead of as a uniform distributed load?

Re: Continuous truss chord bending moments

Artur.Kosakowski — Thu, 10 May 2018 12:15:13 GMT

Hi @Anonymous

As is closer to how this load acts in reality.

Re: Continuous truss chord bending moments

Anonymous — Thu, 10 May 2018 12:29:33 GMT

Hi @Artur.Kosakowski

Sure, the capture below illustrates what I am thinking.

Shouldn't I expect the truss chords to behave as a simple model of a continuous beam?

Also, apologies but I did not quite understand your suggestion about applied load. You propose nodal forces, don't you?

Really appreciate the assistance and the time you are spend answering.

Cheers

Re: Continuous truss chord bending moments

Artur.Kosakowski — Thu, 10 May 2018 12:52:29 GMT

Hi @Anonymous

IMHO it is as expected as diagonals work as elastic supports (horizontal posts have got no bending stiffness).

You can define the load as either nodal or uniform load which ever is as this load acts on a real model.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.

Re: Continuous truss chord bending moments

Anonymous — Thu, 10 May 2018 14:32:46 GMT

Hi @Artur.Kosakowski

Thanks for the clarification on the load distribution.

In reference to the bending moments, I guess we are getting kind of into a loop right here 😄

Apologies for my insistence but it is just that I do not understand why the deformed shape of the structure looks exactly as expected (see capture below) but the bending moments on the upper and lower chords do not seem correct to me, and I believed that it was just a matter of treating the truss bars and nodes on RSA in the proper way, which IMHO am not.

Thanks again for your time Artur.

If anyone is interested, further comment would still be helpful.

Cheers

Re: Continuous truss chord bending moments

Artur.Kosakowski — Thu, 10 May 2018 14:49:41 GMT

Hi @Anonymous

Because the deformations of the top and bottom chords despite the similar shapes are not the same. If so the diagonals start to work in tension/compression which causes the similar effect as if you had elastic supports in their end nodes instead.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.