Hello,
I was looking for a solution to make an articulated union between the pink beams and yellow beams in the intersection nodes. Note that I want the pink beams to be continuous and the same with the yellow beams, so finally I have 4 pink beams and 2 yellow beams. I´ve read about releases but I think it doesn´t work for this case as i want the beams to be continuous. Should I do it with compatible nodes?
@Anonymous Artur Kosakowski edited the topic for better findability
Solved! Go to Solution.
Solved by Artur.Kosakowski. Go to Solution.
Solved by Artur.Kosakowski. Go to Solution.
Solved by Artur.Kosakowski. Go to Solution.
Solved by Artur.Kosakowski. Go to Solution.
Hi @Anonymous
Opps, I did not realized that the model you had attached required modifications of the defined rigid links
As all the beams have free ends to avid instability you wold need to either define artificial supports at their ends preventing rotations about their own axes or replace rigid links with short bar elements with their local axes aligned with the directions of the beams and provide releases at these short bar top ends that allow only for bending about axes perpendicular to the direction of the beams.
If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.
Hello Artur,
Ok, I´ve put a elastic supports with a low coefficient for rotations in the free end of all the bars as you can se in the image. With this elastic supports and with Skyline solver I have no inestability.
Is it ok like this?One question about Skyline Solver, Are the results changing because I am using Skyline solver instead of the Automatic? Or is it only to treat the rigid links?
I mean, can I obtain different results in stress or structure verification if I use Skyline or Automatic solver?
Thank you
Hi @Anonymous
I'd define the elastic supports like this so that they only block rotations about own axes of rafters and purlins.
One question about Skyline Solver, Are the results changing because I am using Skyline solver instead of the Automatic? Or is it only to treat the rigid links?
All solvers should give the same results. The Skyline allows you to not to have artificial (significantly stiffer than bars used in the model) rigid elements to model rigid links.
If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.
Ok, so I understand that in order to block rotation only in their own axis, in case of RX support I should define the local system (alpha, beta and gamma angles) respect the global system. Is this ok?
Thank you!
Hi @Anonymous
IMHO as beans are parallel to the global coordinate system (when you look for the above) for these rotations this may not be required (should be any significant influence on other rotations at these supports) but that would be a must for displacements.
If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.
Hi again Artur, I have one more question about the structure, would it be sway or non-sway? I don´t know how to know that. I want to verify buckling. I know there are other post about this topic but I don´t understand them really well.
You may find the approach described by @RRufino in his post 1 from this topic helpful.
If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.
Hi Artur,
ok, thank you.
I was studying the model with rigid links and I have no instabilities, but i think the results don´t have sense. The nodes of a rigid link don´t have the same translations (note that in the rigid link i´ve blocked the translations). why? I don´t understand. I want them to have the same displacements as I am trying to model an articulated union.
Thank you again.
Hi @Anonymous
The nodes of a rigid link don´t have the same translations
Do you mean master and slave (bottom and top) nodes of the rigid link? If so it is correct as you don't impose the same displacements but you connect these two nodes with very rigid element so this can happen especially when a model rotates (column under the rigid links bends).
If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.
Hi Artur and thank you. When I have no inestabilities the results have sense, I used the option of rigid links blocking translations and in the end of the bar I put an elastic support with a low coefficient in rotations but I had inestabilites too, and I had displacements of 440000 cm, which has no sense.
Is it necessary to cut a bar in the nodes a I want to put a rigid link?
In my model I have a long bar and I made nodes without cutting the bars, and in that nodes I put a rigid link in order to join two beams. The deflectios
but i´m not sure if that´s ok. Please see picture below, do the rigid links placed in the Green nodes work properly? I´ve also attached the deflection which seem to be right Thank you again!
Hi @Anonymous
Is it necessary to cut a bar in the nodes a I want to put a rigid link?
No, it isn't . This is only necessary for compatible nodes due to the way they are defined.
If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.
Hi!
I don't know if this has been answered before, I cannot find the discussion.
Is there a way to connect two beams in a way that one of them lays on top of the other?
From what I have read here I see I should model both bars in their real position and them join them with rigid links that block all displacements (UX, UY, UZ) but allow rotation, is that right?
And in that case, my actual question is; is there a way to block -Uz but allow +Uz? meaning that the beam on top would only transfer vertical loads applied downwards but would lift in case of the load being applied upwards.
Thank you in advance for your response and please excuse me if the topic has been answered somewhere else.
Hi @Anonymous
You can use the compatible nodes (as discussed in messages 2-8 of this topic) with nonlinear function describing the connection between them or define bars at the different levels and connect them with an additional short bar with unidirectional release instead.
If I managed to answer your question(s) press the Accept as Solution button please. This will help other users to find solution(s) much faster. Thank you.
Can't find what you're looking for? Ask the community or share your knowledge.