Simulation - beam element attached to another

Out of category
,
1

Hello

I am on a preliminary project where I have to size a load-bearing structure for an overhead crane. In this type of trade, it is common to weld angles on HEAs to increase the inertia of the profiles without increasing their size.

In beam calculation, on simulation, when I represent an angle welded on the edge of a HEA wing, I can only modify the connections to make the angle attached to the HEA by the ends.

But how do I tell the software that my angle is soldered all the way?

(captured in PC)

Thank you


capture_corniere.jpg
2

Dsl of the double post.

For the moment, I managed by putting a folded sheet metal with the sheet metal function, taking into account a shell (see image in PJ). But it was impossible to do it with a beam element, but I will have to find a solution: :)

Thank you for your help.


caputre_2.jpg
3

Hello

It should be noted that unlike second-order volume tetrahedral elements or each element with ten nodes with three DDLs each:  the beam mode only reasons on two points at the ends.

Indeed, it is a question of vocabulary in your profession and that of metalworkers, we talk about beams so it seems logical to choose a simulation with beams.

In short, to answer your question , you need to change all of your parts to volume. From there, if all the parts are declared as solidary, you will have solved your problem.

I say solidarity because simulation cares about welds like Colin-Tampon.
In any case, the welds are not transmitted between the model and the simulation so there is no need to put any knowing, knowing that for the simulation you must always have a simplified model (without screws, or useless holes, the absence of holes facilitates the meshing and reduces the time for the simulation etc...).

As is the case for an overhead crane, be careful to place the load correctly. In your case, prefer a downward-facing remote load with a surface area limited to about the rectangular surface of each wheel of the trolley.

Kind regards

1 Like
4

Hello Zozo,

I hadn't seen your answer before posting by giving my way with the folded sheet metal.

The beam mode only reasons on two points at the ends.

  • Indeed I noticed it. As a result, with beams that are a little special, we are quickly annoyed, because making a large structure in volume hello the calculation time. 

 In short, to answer your question, you need to change all of your parts to volume. From there, if all the parts are declared as solidary, you will have solved your problem.

  • I'm going to avoid, I don't know if my results are really reliable, during my training on simu I was told to avoid wedging meshes that are too big in relation to my part thickness. Putting 2-3 elements per thickness is good, but on a set like this impossible to mesh, even if you have quite decent pcs. Not to mention the problems of meshing with the adjusted profiles...

As is the case for an overhead crane, be careful to place the load correctly. In your case, prefer a downward-facing remote load with a surface area limited to about the rectangular surface of each wheel of the trolley.

  • Thank you for the clarifications.

Now, I'm going to try to feint the software by creating a .sldflp of the HEA with the angles directly integrated into it. It will be recognized as a beam with the right inertia in theory. I'll try and I'll give you feedback ;-)

5

So, I tried to make a sldflp with the angles, new problem, see the screenshot:

https://image.noelshack.com/fichiers/2020/40/5/1601642431-1.jpg

I can't mesh correctly and generate a constraint plot. The travel routes work but with a delirious view of the profile... Could someone try to see if it comes from my home? A simple simu with a gravity see if it goes through your home.

Thank you


Part4.SLDPRT
6

Hello @Jwunder

Everything works properly

On the other hand, you have to use the right parameters everywhere :-)   The "delusional" deformations are only so if you don't bother to indicate the visualization of the desired deformation. If we set to 1 we do not see the distortions other than those given by the color, if we indicate 20 ( = 20 times more than normal) we will have a greater distortion. But this is only visual information that does not change the result Von MISES To see real life you have to set to 1.

Kind regards

PS: If you don't mind, I don't agree at all with what you said about the mesh and even less about the computation time.
But hey!
In the attached example, I took a mesh of max 130 and min 6, which means that I have a mesh in less than two seconds and a simulation in less than 2s. What we can remember is the number of elements 2476 and the number of nodes 5084, in other words nothing at all for a simulation, I regularly process 5 million nodes and it doesn't take hours.   I have attached the report to you in PDF so that you can have all the parameters used.

 


piece4_from_zz-statique_1-1.pdf
7

Hello

Dsl for the response time, I had to return my AVP!

I know about the deformation scale, I'm used to modifying it, often customers call me back in panic to ask me if it's the real scale, so I always end up putting captures at scale 1.

I was talking about creating a sldflp (sketch file for the welded construction) with the profile I'm interested in, but it doesn't seem to work, I don't understand why.

After creating the file, I use it to make a beam via the welded construction function, everything works:

Then, I mesh in beam element, we should see sections of mesh, but the model remains smooth I had never seen that before:

I launch the study, I have the following alert:

that I can't solve by rummaging through the menus indicated. Then I have this kind of vision on the trips:

(I have enabled beam rendering), and the constraint drawing does not work. I would like to find the solution to this problem because it is important for what I do.

After the volume I use it a lot when I have plates welded in all directions or for lifting equipment for example (spreader beam etc. that kind of thing) but on structures we spend 5 times the calculation time on volume compared to beam (taking into account the drawing of the model) because on a set of beams the junctions are made in 5 seconds, in volume we have to treat the fittings, put the plates, check for collisions...

If I have this problem again, I will try the support directly.

For my problem with the post, I did with folded sheets that I made integral with the connections menu, it worked. Thank you for your help in any case.

Have a nice day