3D Solids - Loading and Meshing

[Andyf88]

Hello Everyone,

I am modelling a structure in Robot using solids. At first I have modelled the whole structure as 1 volumetric object by unioning the different objects because this way I can have an coherent mesh.

When it comes to applying the loads on this structure I have to add a triangular pressure (earth and water pressure) which does not act on the whole surface (side) of one wall of the structure. This is when the problem starts because when I use the "Planar load 3p" I cannot apply a pressure on only one part of the surface (side). In fact the pressure goes negative on the rest of the surface that I don't want to load.

Is there any way how to overcome this?

One work around is to split the solid object where the change of loads happen but then another problem arises - if I have different solid objects their mesh is not coherent. How can I achieve a coherent mesh between 2 solid objects (remember the solids are "Object of geometric modeller" type - which means they are meshed in the 3 directions (X, Y, Z).

Another question - how can I build a solid type "Object of geometric modeller" straightaway? The only way I am doing it now is by extruding a face (which gives an solid type "Contour") and then using "Object Combination" to Union or Substract it which changes the "Contour" type to "Object of geometric modeller".

The reason I need "Object of geometric modeller" is because the mesh can be altered in the 3 directions. Whilst for the "Contour" types the mesh in the direction of extrusion is fixed to the number of divisions that you determine when you extrude.

Many Thanks,

Skerdi

[Rafacascudo]

Not familiar with solid modelling , but couldn´t you use planar load 3P on contour??

Can you show some pictures ,sketches of your model??

[Andyf88]

Hello Rafacascudo,

In fact you are right, I can use "Planar 3P Load (Contour)" my only problem with that is that it doesn't recognise the local coordinate system. When I apply the load according the local coordinates it still considers it as global. Is this a problem with my version of software or is it for everyone?

However, thanks for this bit, really appreciate your quick reply!

[Rafacascudo]

It will load in the correct local direction AFTER you run your model or at least run " model generation"

Don´t worry...

Just remember to activate " forces generated automatically" on Display options

[Andyf88]

Thanks a lot Rafacascudo,

I am still having problems with meshing though. With shell elements, if you put a node on the surface of the member the meshing will automatically be generated around that node, whilst I cannot do that with the surface of the solids. I put a node on the surface where the support is located but the meshing does not recognise that point and it turns out to be an isolated node, as shown in the pic below.

Can anyone help pls?

[Andyf88]

Can anyone from the support team please help with the queries above? I really need an answer now. Autodesk Robot must have an expert in 3D Solids.

Thank You!

[Rafal.Gaweda · Accepted solution]

Non ACIS solid example:

http://forums.autodesk.com/t5/robot-structural-analysis/how-to-remove-isolated-nodes-on-solid-s-elem...

ACIS solid example:

http://forums.autodesk.com/t5/robot-structural-analysis/analysis-of-solids/m-p/5555245/highlight/tru...

[Andyf88]

Hello Rafal,

I have faced another problem with applying loads on the sides of my 3D model. Basically it is a slab with a depth variation so it has a number of sides on the top and I am applying a uniform vertical pressure to it.

When I calculate the total force (pressure x area) by hand and compare with the vertical reactions it is not giving the same result.

I checked all the sides separately and for some of them the results matched (force calculated by hand VS reactions from Robot) and for some not. I don't understand, it should be a simple task for Robot - force = pressure x area!!!

Have you ever experienced such a thing?

Thanks!

[Rafal.Gaweda]

It may happen that loads can be applied twice \ doubled on some sides.
Apply nodal forces instead.

[Andyf88]

The forces are coming smaller actually!!

And it is a single solid model which is meshed, so calculating the nodal forces would be a bit tricky because the mesh is random!

[Rafal.Gaweda]

File please.

[Andyf88]

Hi Rafal,

I have attached the model and the loads that I have calculated by hand.

Thanks

[Rafal.Gaweda]

This is what I see - only on arc plane - case 6 - results are not correct

[Andyf88]

Sorry, there was a small mistake on my calcs for Case 1 and 2. So yes there is a small change on Case 4 and a big one on Case 6. For case 6 the difference is by more than 10000kN which is huge.

So why do you think it happens and how can I avoid it?

Thanks again!!

[Rafal.Gaweda]

It must be corrected in software.
Apply different load value in case 6 to get expected sum of reactions.