Sum of reaction and sum of force not equal.

Anonymous · ‎06-09-2014

Hi there,

I am modeling a 85m high chimney. The top 65m is cylinder (shell) which is supported on a 20m high support structure (bar elements). At this stage I have only applied self weights. I am checking the reactions from self weight and superimposed dead load.

My question is why the "sum of reaction" and "sum of forces" are not equal? My understanding is these values should be same with some minor errors. Your help is appreciated. I can update the model if required.

Thank you,
Raju

Artur.Kosakowski · ‎06-09-2014

Lack of convergence? Ignored instability warning? Could you attach the model?

Artur Kosakowski

Anonymous · ‎06-09-2014

Hi Artur,

Appreciate your prompt reply. Yes there was messages regarding the instability but I thought it would not matter for the linear analysis. Also there were messages regarding cohesion (definition of kinematic constraint) and finite element. I have noticed that the on panels 20, 65, 315, 77, eventhough I did meshing, I can't see the meshing. I though it might be the graphics issue. Please advise.

Please find attached dropbox link as I could not attached the file in this forum due to size limitation. I would appreciate If you could find any other issues the model have (if any) and advise.

https://www.dropbox.com/sh/ee1o45y3qd4zj52/AAB4hid1fYoeFcL78lMkysl6a

Thank you, Raju

Artur.Kosakowski · ‎06-10-2014

What would be the point to display the instability warning if it doesn't matter?

You need to create a coherent model. I'm attaching the corrected one (mind that generation of nice meshes has not been my objective here).

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.

Artur Kosakowski

Anonymous · ‎06-10-2014

Hi Artur,

Thank for your feedback and support. My understanding was that the instability and convergence issues would only affect the member forces locally but the reactions should be equal to the applied forces. However this is not the case for Robot, which in a way is good.

We have the Robot license for last two years in our office but we never use it as there are many things myself and my colleague do not know how the Robot works. Partly due to limited documentation and partly no local support here in Australia. Therefore, my aim from the model was to calculate the natural frequencies only. I would be preparing a separate model with the support structure only on Robot as a starting point and apply loads from the stack above. I have gone through the Robot help/online threads and I am confident with a model with the bar elements only. However due to your prompt reply I am now interested to utilize the Robot in its full capacity.

Thank you,
Raju

Anonymous · ‎06-13-2014

Hi Artur,

I have a question in relation to modelling a bar elements and shell elements, how they connects? For example, I have a Portal frame of I-Sections (Two columns and a beam). If I want to have a plate cantelivering from the top flange of the beam with continuous weld all around, how do i modell this on robot? Do I model plate first and then offset the beam from plate or vice versa. What happens to the nodes at beam and columns locations, do they lie in the plane of plate for CL of beam.

Thank you

Artur.Kosakowski · ‎06-13-2014

They are connected by common nodes. You have to think about bar elements as lines rather than solid shapes. In other words there is nothing like top and bottom flange for a bar element with I section assigned (unless you create a steel beam as two horizontal panels at the top and bottom of the vertical one) and all loads (either defined or coming from connected objects) are applied at the center line of the beam. Personally I'd rather avoid use of offsets as they influence results in bar elements (you will get axial force plus small bending instead of larger bending). If you want to consider the effect of in-plane forces from the plate being applied to the top flange of the beam define the plate above the line of the beam and use rigid links to connect them together (generate number of nodes along the beam which correspond to the mesh of the plate).

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.

Artur Kosakowski

Anonymous · ‎06-13-2014

Hi Artur,

Thank you for providing the options. Depending upon the situation of loads, I may have to create the plate offset from the beam. In this case rigid link will he helpful as you mentioned. But, how do I define rigid link between two nodes at certain distance? Does the nodes on the beam and plates have to be in line?

Also is there any command which converts a I beams into corresponding plates (like exploding a 3D object). I know you have mentioned bar element is just a stick only, Just asking if there is?

Thank you,

Raju

Artur.Kosakowski · ‎06-13-2014

Thank you for providing the options. Depending upon the situation of loads, I may have to create the plate offset from the beam. In this case rigid link will he helpful as you mentioned. But, how do I define rigid link between two nodes at certain distance? Does the nodes on the beam and plates have to be in line?

See the message 6 from http://forums.autodesk.com/t5/Robot-Structural-Analysis/Beam-on-Beam-load-transfer/m-p/4877198 and check http://help.autodesk.com/cloudhelp/2015/ENU/Robot/files/GUID-00FFC3EC-257C-41DA-97C4-EDB9EF9C3464.ht...

Also is there any command which converts a I beams into corresponding plates (like exploding a 3D object). I know you have mentioned bar element is just a stick only, Just asking if there is?

No, you can't convert a bar element into number of panels.

Artur Kosakowski

vanvan11 · ‎09-25-2015

10x for info i will try to use Rlink more often when i need to include the shell thicknes to reduce bending moment

but why you not use offset ?

I often model a new section to which I add a small pice of the panel (cooperating width of the panel) which helps to reduce the bending moment and not have to offset and do not need Rlink Is this correct?

vanvan11 · ‎09-25-2015

first you need to understand FEM basics
it's easy to learn software which use FEM with basic knowledge for FEM

really I do not want to offend

have a nice day

eng. Ivan Georgiev

meskmo13 · ‎06-11-2018

Hi,

what changes did you made to the model?

I downloaded both but I cant tell the differences.

10x,

@Artur.Kosakowski wrote:
What would be the point to display the instability warning if it doesn't matter?

You need to create a coherent model. I'm attaching the corrected one (mind that generation of nice meshes has not been my objective here).

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.

Community

Sum of reaction and sum of force not equal.

Sum of reaction and sum of force not equal.

Sum of reaction and sum of force not equal.