Virtual wall

Simulation Resistance of materials
1

Hello

I need to calculate the forces in the ankles in a robot base.

fig1

Stresses and torques are applied at the top and the holes at the bottom are " fixed " and I had added a virtual wall, to prevent the base from sinking into the ground in the middle....

Then I make a measurement of the effort in each of the fixing points.

Strangely enough, the closer my piercings got to the center, the more the plate on the ground decreased, and the less the forces in the ankles were important.... !! This is totally contrary to logic.

 

So to come back to a basic problem, I drew a square, which I consider fixed to the ground by the drilling and the short side supports a 500N force parallel to the ground, just on the ridge at the top

My square is 100mm x 50mm long and the holes are 20mm from the edge.

The theory gives a vertical force in the drilling, of about 312N (500x50=80xF).

Fig2

Simul 1 : I fix the hole. the deformation of the square passes under the plane. Logical but not realistic.

Simul 2 : same thing but I add a virtual wall, to simulate the ground and there I find myself with forces of 700N ... Very far from theory.

Simul 3 : I remove the virtual wall, but add a " fixed pivot" in the angle of the square, and there the result is around 360N. This can be understood with the approximations of the EF and theoretical models.

Simul 4 : I only add the virtual wall and again the efforts fly away....

 

So my question is : does the "virtual wall" connector work ? Do I use it well and if not, how should I do it ?

Because I need to have a reliable answer for my foundation.

 

Kind regards

 

 

 

 

 

2

Good evening

The Paroi Virtuelle works perfectly provided you use it well.

The PV in fact provides a sliding support comparable to the imposed displacement of the "Plane support" type. However, the difference is that if you use it, you can set a coefficient of friction and the elasticity of the wall (like a flat silent-block or an oak plank).

I don't think I understood when you say effort in the ankles.

If it's really the forces in the ankles, I don't think it's the right method because the simulation can only be done on the bodies present.

This is understandable if you use bolt connectors (use "anchor bolt" with virtual wall).

For anchor bolts, the  virtual  wall will replace concrete in the calculations, Note : with PV the concrete does not take into account the concrete strength.
As a reminder, to define the virtual wall, you must create a plane that coincides with the support face in the model.
For  the simulation, contact between set, virtual wall, you must select the face and the target plane.
After: for anchor bolts =>> connection =>> bolt==> anchor bolt, and select the hole edge (where the bolt passes)

Next...: Select the target plane (the one used for the virtual wall)
Then fill in the bolt information ==> material (it is not necessary), ==> nut diameter (see screw head), ==>  preload (tightening torque)  look at the default values which are pre-filled if you have used the drilling wizard to make your holes.

Be careful if you use the anchor bolt connector and you have several of them, you must do the same for each bolt. You can't select multiple bolts at once, or you won't have the other bolts. There are several functions like this in simulation that you have to define one by one.

To read you

Kind regards

2 Likes
3

Hello @ hervé.marchal

Excuse me for going a little beyond your request but simulation can be tricky.

I just saw that you put your holes two by two, which makes the peers quite close to each other.

However, you must also take into account for the anchor the "axial tensile service load of an insulated anchor without influence of the edge".

Take a look at the guide "Metal anchors for Concrete" ETAG N°001 edition 1997 which defines the 12 options that determine the conditions of use of the anchors. In particular, those designed for fixing guardrails in concrete (fixing 50 mm from the edge).

A deep fastening is indicated in relation to the concrete tear-off cone.
Look here at the service limit loads , this also gives you the minimum centre distances and the minimum edge distances for balconies or platform edges, etc...

Look here at the ATE standard , it gives you indications on the center distance (to be confirmed by the anchor anchor supplier because small variations from one supplier to another)

Look at the deep-anchored ones that have an option 7 ATE. In other words, they can be used to fix posts, wall brackets, stairs, etc.

All this to tell you that for your simulation only the tensile strength of the anchor stud or bolt is to be taken into account. That's probably what you put behind the term """efforts in the ankles in a "base"".

Kind regards

 

4

Hello

Thank you for the answers.

For point  2,  yes the position of the ankles is not good but due to a test to verify a hypothesis.

For point 1, at the beginning I fixed the cylinders of the holes and a virtual wall and read the measurement  "resultant force"  and that's where there was a weird behavior. When I redo the manipulation on the square (simple problem) I still have aberrant answers.

What annoys me is that I had the impression of doing well, not perfect of course, but correct, by fixing the holes and putting a virtual wall. I would like to understand where my mistake is?

Thank you.

Kind regards

 

5

Hello

If there are simu bolt fasteners in SW and the variant "anchor bolts, on the  virtual wall",  it is not by chance.

The fixing of the cylinders (the holes) of the holes to a virtual wall does not convince me but hey ;-)

6

Hello

I used the "virtual wall" and the "anchor bolt" fixing, I set to 0 prestress.

Then I measured the force in the connector and I'm at 647N .... which is double the theoretical one!

What am I doing wrong? Can someone test at home and tell me if they find the same value?

Thank you.

Kind regards