Abaqus - Mises Stress values are too high

[Error.n.n]

Hello,

As practice, I have simplified a steel-concrete composite bridge to one 3D beam model with effective width of concrete. The beam length is 21.3m and concrete effective width is 2.7m. I used tie constraints to simulate the interaction between the concrete deck. I am following the base units of ton, mm and N for abaqus and currently modelling using elastic materials only. However I am not comfortable with he final values I am gettign for the von mises stress. In addtion to the selfweight, I have applied a UDL load of 1kn along the area of the flange girder which is 300mm wide. I am getting values of 1.117e6N/mm2 for the von mises stresses. can some one help me? Am I interpreting the units von mises values wrongly? I think the value I am getting is too large because when i change the steel material to trilinear material with yield stress of 355MPa and ultimate strain 0.18,, I get the error of strain increments are too large. I have triple checked the input units multiple times. The other input values are mass density for steel is 7.85e-9, for concrete 2.4e-9, acceleration -9810, E (Youngs modulus) concrete = 30000 and v =0.2, for steel E = 210,0000 and v =0.3.

The girder is UB beam 1016x305x272 with concrete deck thickness of 250mm. I have been trying to compare the values when i add stiffners to the web.

 

[SWComposites]

How are you getting VM stresses from a beam model?
Show a picture of your model. and a picture of the stress results.
Maybe you need larger beams to carry the load.
(note I don't think in SI units so can't tell if your numbers make any sense).

 

[Error.n.n]

Hello SWComposites,

Thanks for the quick reply. I apologise for my mistake in wording. I am using shell elements to model the beam section and deck in 3D space. I have attached the file with different values. In each picture, the only variable is the positioning of stiffeners. I sent all so you can view the consistency in the value of results. I must add I am fairly new to abqus and finite element modelling in general.

Charl

 

[Error.n.n]

[/url]

I have added the assembly and the unit conversions for your reference. thanks for the help, bacause i am completely baffled. Because if my units are correct, the max displacements I am getting are 239mm which seem excessive for 1KN udl

 

[FEA way]

Can you share the .inp or (better) .cae file ?

In the first message, you said that E for steel was E = 210,0000 MPa but in the table, it’s E = 210,000 MPa (correct value).

 

[Error.n.n]

Hello,

Yes, Sure, I can share the cae. file. Thank you. I am sorry for the typing errors. Actually that post has too many. I will do better.

C

 

[Error.n.n]

Any luck in helping me with this problem? I have finalised my final bridge model, and the results from it are similar in exaggeration to the ones in this simple model. Any idea what is causing the stress results to be excessive, or is it my unit conversion that is the issue?

 

[FEA way]

I think that I've found your mistake. You defined the UDL as a pressure load. This means that its magnitude (1000) is not in N but in MPa...

 

[Error.n.n]

Thank you so much. Now I completely understand the phrase "rubbish in, rubbish out" when it comes to software. I had a feeling it was something so simple that I completely missed. Thank you once again, and my apologies for using your valuable time.

C