Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

1D versus 3D elements

When analysing beams, how significant is the difference in accuracy between results obtained using 3D elements, and ones obtained using 1D elements (i.e. meshing the beam as a line and assigning a cross section to it)? When would it make sense to make the effort of using 3D elements?

Find more posts tagged with

English

HyperMesh

Accepted answers

All comments

Altair Forum User

When analysing beams, how significant is the difference in accuracy between results obtained using 3D elements, and ones obtained using 1D elements (i.e. meshing the beam as a line and assigning a cross section to it)? When would it make sense to make the effort of using 3D elements?

actually that depends on the beam theory that the program uses (timoshenko, reddy, euler-bernoulli etc.)

you can quickly do some benchmarks with them and compare the results. check von misses, shear etc. and decide which one is ideal for you. also check the beam theories from some technical sources (which one is for what?).

there must be very close results both with beam theory and solid mesh. for the beam usage, integration points must be enough to avoid shear locking etc.

Altair Forum User

Thanks a lot for the input!

So, if I understood correctly, when you assign a cross-section to a 1d component the solver uses one of the beam theories to calculate the required response quantities? Would that mean that the number of elements used for the 1D mesh is completely irrelevant as you would get the same result in any case?

In general, it would make sense to use 1D if the model is close to the ideal case, if not 3D would be more accurate?

Altair Forum User

Thanks a lot for the input!
So, if I understood correctly, when you assign a cross-section to a 1d component the solver uses one of the beam theories to calculate the required response quantities? Would that mean that the number of elements used for the 1D mesh is completely irrelevant as you would get the same result in any case?
In general, it would make sense to use 1D if the model is close to the ideal case, if not 3D would be more accurate?

mesh size of 1d element is important but number of integration points of the 1d element is more important than that because of the carrying of the shear loads. shear locking may occur and you can't get the right values then. check the theory manual for that to be sure of that.

as i said before, do a simple benchmark first, and then you can decide it.

check radioss theroy manual 10 volume 1 page 145

Altair Forum User

Thank you, I'll take a look at the manual.

Altair Forum User

In general, it would make sense to use 1D if the model is close to the ideal case, if not 3D would be more accurate?

Altair Forum User

THANK_FOR_ALL

Altair Forum User

CBARS are based on Eurler-Bernouli beam theory and CBEAMS are based on Timoshenko beam theory and include shear deformations. The number of elements does matter, but the actual math is simply based on the area and inertia properties of the 1-D element wtih stress recovery points being at the extreme dimensions, just as with an undergrad textbook.