Learning Forum

[tumulpurwar]

I have solid hollow tube 50 mm thick and 20 m long, standing vertically(like electric poles in streets) and it is connected with beam element rods with some angle say 45 degree(kind of diagonally connected)....in this case share topology will not work I think(means beam element connection with solid element) and why?
And node merge will work but why?

2nd question
For bonded contacts...if I want connect beam element with solid element..how much gap work..as I am not sure there they provide tolerance limit or not in bonded contacts

3rd question if I have two solid concentric hollow tubes one 50 mm thick and another 8 mm thick which is inside 50mm hollow thick tube...and if I convert both in shell elements by applying midsurface..than they have gap between them of 25mm + 4mm = 29mm..which is out of tolerance limit(max 3mm) for share topology in spaceclaim.. so it means we cant use those hollow tubes as shell elements ?

4th question
Is solid elements are fine for showing bending in 50 mm thick , 20 meter long hollow tube?

5th question...if my hollow tubes 50mm is made up of wood(say douglas fir)
It comes in non linear analysis?...as it is wood?

6th question...two hollow concentric tubes 20 m long with 50mm thickness and 8 mm thickness...are there whole cylindrical length of 20m are bonded or glued with each other by share topology or else I should use bonded contacts?

7th question
Normally if hollow solid Element tube is connected by beam elements at some angle say 45 degree( like street electric poles are connected with electric wire at some angles)...to know the streses in beams(say electric wire)it is good to use share topology or else bonded contacts?

8th question...normally solid dont support rotational moment..for example if in our case beam rod connected with solid element hollow tube....and if heavy load acted over solid tube which in gives bending to solid tube and than it gives bending to connected beam rods...so in such case beam will show bending or not( as solid dont support rotation)

Thankyou

[peteroznewman]

1) It's easy to share topology of a beam vertex with a beam vertex - see #4.

2) It's easy to bond a beam to a beam - see #4.

3) Use a Beam model - see #4.

4) Beam model is better for a 20 m long tube that is only 50 mm in diameter (is that what you mean by "thick"?).

5) Make an Orthotropic material using measured wood properties.

6) Do the two tubes have different material properties? If so, compute equivalent beam properties to use beam elements.

7) Build a beam model to represent the overall structure.  Take the forces from that overall structure to apply to a small, local, detailed, 3D solid model (submodel) of a specific joint between the pole and the wire.

Beam model supports rotational moment.

Maybe you could include some images of the overall model.

[tumulpurwar]

Thankyou for reply ...
I am attaching some pics

It's actually a sail mast...made of 2 hollow tubes 50 mm thickness (wood material)and 8mm thickness(carbon epoxy material)
As you said above ,use beam element....but I cant use beam element for 8 mm as its material is carbon epoxy.
So I can use one hollow tube of 50mm as beam element but other 8 mm I can only use as shell(I think because of carbon epoxy).
For the replies
4th it is not diameter..its hollow tube thickness
6th yes hollow tube with thickness 50mm and 8 mm has different material
For 50mm wood
For 8mm carbon epoxy

Can you elaborate 6th point reply a bit.

Right you are for 8th point (50mm thick wood hollow mast tube PLUS 8mm thick hollow carbon epoxy tube)
Along 20m long length, loads are acting at 3 locations in a pair of transverse and longitudinal load at 19m
17 m
And 5 m
These loads will create bending of hollow tube (or mast) and beam connection will take away axial load..actually I need use truss element instead of beam element.

Now what is better in such case
Means how connections will work
Between shell element 8mm thickness carbon epoxy hollow tube inside 50mm hollow thick tube made of wood(which can be beam or solid element)( I don't know which one is better to use)
And contacts between solid element / beam element tube with truss/beam element rods.

My one more big issue is both hollow tube 50mm thickness and 8mm thickness are having cross section which increasing uniformly with height....so I cant extract beam from solid tube easily and if I use beam element than it doesnot represent correct shape of tube geometry.

Also can I have conformal mesh in such case by having share topology for this complete multibody parts?

[tumulpurwar]

[tumulpurwar]

[tumulpurwar]

[tumulpurwar]

[tumulpurwar]

[tumulpurwar]

[peteroznewman]

Is the CF epoxy tube on the inside of the wood?  I would have thought it would be the opposite configuration with the CF epoxy on the outside. How is the wood bonded to the CF tube?  What is the shear strength of the adhesive? Is the wood applied in sectors?  Does the wood have flats as shown?

If the mast is changing cross-section along its length, then you would have to divide the length up into segments and use a beam with a different cross-section on each segment.  The diameter will step down segment by segment as the height increases. You don't have to do that if you use solid or shell elements.

Re #6, the equation for beam bending uses EI as the stiffness of the beam.  You can build a solid model of a uniform hollow cross-section composite beam of length L of two different materials bonded together. For a cantilever beam, make a fixed support at one end and apply an end load P and obtain the displacement d at the end. Use the formula d = PL^3/3EI to solve for EI.  Choose some nominal radius r for a circular solid beam and compute I.  Now divide EI by I to get E. Create that material and assign it the value of E.  Create a beam cross-section with a solid circle of radius r.  That beam model will have the same bending stiffness EI as the solid model of a hollow composite beam.

What version of ANSYS are you using, 2019 R2?  Are you on a full license or the limited Student license?

[tumulpurwar]

1) wood applied in sector
2)adhesive is epoxy resin
3) CF is inside
4)yes wood have flats called transverse frames 50mm thickness
5)if use shell element for both outer and inner tube , the tolerance is 3mm maximum , but in our case it is (50mm/2 + 8mm/2 = 29mm)..its out of limit.
6) in your 6th reply..composite beam of length L of two different material??? Means wood and CF ?

Can you elaborate also my first post question over point (2), (3) , (6) and (7)

Limited student version..I will do this project in university in academic license version

[peteroznewman]

Since the wood has flats, you can't use shell elements since the thickness varies across the width of the plank as the side opposite the flat is curved. The wood has to be solid elements until you condense the composite beam into a simple beam element with an equivalent EI.

Re #6) Attach a SpaceClaim .scdoc file with a uniform cross-section of the solid model of the CF inner tube and the faceted wood outer solid. I will use steel for the CF and aluminum for the wood to demonstrate how to evaluate EI for a composite beam.  Note that EI will be slightly lower when bending the composite beam parallel to the flats on the wood versus bending through the "points" on the wood because the structure has a larger radius to the point on the wood than the flat.

[tumulpurwar]

attached below file

[tumulpurwar]

i dont have flats in my file because of bodies size restriction in student version...normally flat(trasnverse frame) are 0.5 m spacing, they will start from top length and go upto bottom of mast....they are inside 8mm hollow composite tube( as i attached in above photos, you can see those flats(or transverse frames with 50 mm thickneess)

 

i can provide you complete model by monday through university...this model is without frames(flats)...if you can incorporate in your model to show the results it will be greatful...thankyou

[tumulpurwar]

frames photo attached inside carbon tube. frames are in this photo are shell elements(thats why have no thickness, normally they are 50 mm thick, and i thing as you suggested shell elements will not be connected properly with CF tube becasue of Uneven shape nad width, so i wll use solid elements for it from next time.

I want ask one more question, by approximating composite solid tube as beam...how much can be error approximately (+ or -) in estimating correct  bending stresses ? 

[peteroznewman]

Equivalent Beam Method

Can be used only on solid bodies that have a uniform cross-section. Cannot be used on tapered beams. To do tapered beams, chop the beam up into short segments of uniform cross-section which will step up to smaller uniform cross-sections on each segment.

Solid Bodies meshed with Solid Elements

Nodes = 31,800, Elements = 7,056

A composite beam has an inner octagonal steel tube and an outer aluminum octagonal tube. It has been meshed with solid elements and shared topology allows the two materials to share nodes at the interface between them.

 

To go from nearly 32,000 nodes down to 43 nodes, an equivalent beam is calculated from the displacement of a cantilever beam solution. 

A material called Equivalent Beam is created with this value of Young's Modulus.

Line Body meshed with Beam Elements

Nodes = 43, Elements = 21

Line Body displayed showing circular tube cross-section.

Mesh shows 8 facets, but that is just a low resolution rendering of the line body with the circular tube cross-section.

The equivalent beam has the same tip displacement as the composite solid element beam.

[tumulpurwar]

Thankyou very much for your attention over my problem..thankyou for visual photos and equivalent beam concept.