Making continous beam in grate or 3d frame

[s226612]

Hello, I'm struggling with simple construction as shown in attachment. I've got a problem with defining middle beams (the long ones) as a beam with multiple supports. Can someone explain to me how am I supposed to define it, so the beam would have "+" bending moment in the middle of the span and "-" bending moment above the support (as shown in the picture)? Oh, I almost forgot, that the support should transfer only vertical force on transversal beams.

Now, when I add this middle bean it is seen as simply supported element. When I make it longer than one span it is still seen the same, but the bending moment in transversal beam is "upside-down", not how it is supposed to be (upper part is being "extended").

I know how to define nodes when it comes to connecting main beams with transversal beams (I use pinned-pinned conection on shorter beam and it works as I planned)

[s226612]

oh, what's more i want them to co-operate fully when it comes to displacment (i want them to be the same in node when they connect)

[Artur.Kosakowski]

The bending moment diagram depends on stiffness of the elements of the model (cross section vs. length). You may want to increase cross sections of these beams that you consider as supports for other beams. The displacements in common nodes have to be the same for each of the element of the model.

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.

[s226612]

I am aware of what you have written, but still it won't change the general look of the graph of bending moments. Do I have to use compatibility of nodes? I want those beams to be above the other ones, so the stiffnes of both elements won't matter.

I asked about displacment because I think if compability of nodes won't help I will need to add supports under one beam, but I want them not only to transfer force but also to have the same displacment value (if I need to add 2 nodes in one place, one for each beam).

[Artur.Kosakowski]

No, this will not make the difference. Please mind that the shape of model deformation (distribution of forces, shape of diagrams) is 'based' on stiffness matrix generated for the whole model rather than being based on the gravity analysis (load takedown) approach.

Imagine two beams with identical cross section and lengths forming a cross - which one would you indicate as the support for the other?

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.

[s226612]

Ok, but I do NOT want them to cross, but one beam to lay on another one. So let's forget about their stifnnes for a moment, because it's not the point here.  Things You keep on writing are not bringing me any closer to the solution. I am working on a very simple construction, I ahve added cross sections I want, but still it doesn't work as it should. Do I need to use compability of nodes?

[Artur.Kosakowski]

The way you model your structure is in your hands. I may suggest some solution that you can either take or reject and of course I may be alsototally wrong. In either case you can create some test models and check the results.

[s226612]

Ok i cross them and I want the bending moment to go up near the support, but i dont know how to use compability nodes right, can You help me with that? What should i do?

[Rafal.Gaweda]

 i dont know how to use compability nodes right, can You help me with that? What should i do?

---

Check this:

http://docs.autodesk.com/RSA/2013/ENU/index.html?url=filesROBOT/GUID-774B172C-7704-4D88-8830-DBE3D05...

[s226612]

OK i got to a point when it comes to use this function, but I dont know what am I supposed to do there. I ha ve chosen 2 nodes which should "cooperate" but then on the right side list of elements shows up - what should i do there?

And what kind of compability should i use to get effects shown in one of the attachment from support team?

[Rafal.Gaweda]

Example movei how to appply compatibility : http://screencast.com/t/efFhzVwvc

[s226612]

Wow, that helped a lot, as well as screenshots, when stifness were also taken into consideration. Can You give me an answet to this question:
Is there a way to somehow forget about stifness? I mean when i want to do this kind of calculations on paper I find the reaction on upper beam and then take it to the lower one, can't i do the same here? Or it is only the way in static-determinate (?) constructions?

[s226612]

[IMG]http://i39.tinypic.com/2w3uw54.png[/IMG] This is what I got. On the left there is a construction, where stifness is taken into a consideration, on the left this is not taken into consideration (i change it from IPE100 to IPE 500 and i get the same results). Can I make it the same on the left as well? I need just the vertical force to be transfered down and the node should "imitate" the support to the upper beam. Is it even possible?

[tony.ridley]

I don't understand at all why you need to make this so complicated?

It's just a continuous beam you want to model, one that crosses over another beam?  Just model as such, and forget about compatability of nodes?

Seems like you are making hard work for no reason.

Tony