static simulation of straight teeth toothed wheels

[andreadenu]

Good morning to all,
I'm learning to use ansys at this time so I apologize in advance if the problem I'm experiencing is simple resolution. I can't find a way out on the internet.

I have to make simple static simulations on dense wheels.
the simulation consists of:
- application of a linear distribution along the side of the tooth at a distance from the known center.
-the wheel is bound by a recess along one side of the tab seat.

the problem is this: changing the resolution of the mesh changes the equivalent voltage state of von mises, which presupposes that a greater mesh finesse is necessary until it reaches convergence. However, I cannot assign to the solutor the task of assigning the mesh dimension as, automatically, a number of elements is created higher than that allowed by the academic version of the software.
At least I can use the automatic convergence function by redefining the mesh cyclically because I get the same error.

then, I thought to redefine the mesh manually in the areas of interest identified by evaluating the "mistake" of the voltage state. (laughing) )

However, I cannot define mesh in this sense, making it more enduring in areas of interest. I tried with the "body sizing" and "sphere of influence" function but I do not get the desired result or get the error that I bring back:

"the selective body meshing is not being recorded, so the meshing may not be persistent on an update. if you want to record the order of the body meshing, please use the mesh buried to track the meshing steps. please see selective meshing documentation for more details."

How can I proceed? I attach some images for greater clarity.
thanks to anyone who can help me!

 

[meccanicamg]

first problem application correct constraints. . .
I would say that the tongue must oppose as a twist, but it is all the circumference that keeps radial in every direction.

then you have the problem of mesh....

 

[Onda]

reduce the thickness, and proportionally the load, so use fewer elements, as you are limited in the number from your version.
then make a small connection radius in the tongue edge. this has two reasons:
- a live edge is seen from the fem (and also in reality!) as a point of discontinuity. tension is infinite to decrease the size of the mesh. in reality there are two effects: the first that there is no living edge, but in any processing you have a minimum radius. the second effect is the plasticization of the material that occurs around the edge.
- the second reason is to force the fem to make smaller elements around the connection radius.
if you look at the uni 6694, you find that in the definition of the quarries for tabs, there is a minimum connection radius and a maximum depending on the size of the tab itself.

 

[andreadenu]

first problem application correct constraints. . .
I would say that the tongue must oppose as a twist, but it is all the circumference that keeps radial in every direction.

with what bond does this concept make?

 

[meccanicamg]

depending on the progry, a radial reaction 3600 and an axial reaction must be provided. rotation is blocked by reaction on the tongue side.
There's definitely nothing to frame. . otherwise...

 

[Onda]

If your goal is to see the stress in the tab you can't frame it, but you have to put a contact. an infinitely rigid bond is not suitable for evaluating stress in its surroundings.

 

[andreadenu]

thanks to the answers,
my purpose is to simulate this test that will be subsequently carried out in the lab.what I care about is mainly the tension at the base of the tooth. in the test the wheel is calettated by a tab on a tree that is fixed. the dot instead applies a linear distribution of pressure to a known radius.

 

[Onda]

I would apply a cylindrical bond to the inner cylindrical surface, a force at the location for the key and a contact between the two components. then you have to stop the axial translation.