dynamic constraints

[giorgioc73]

I am experiencing dynamic constraints on inventor.in the axieme above there are a rod, a tree to which a disk and two wheels are united, hinged in the disk and bound together and with the rod.
If I fix the disc and rotate the rod, the two wheels move as you expect.
if instead I fix the rod and rotate the disc... "it goes no."
I attach the files so who wants can try it can.
I'd like to know why of this weird behavior.

 

[giorgioc73]

I was wondering if among inventor users, someone had downloaded files and tried the movements. it would be useful even if they had the same behavior.

I forgot to specify:
inventor professional 2020
windows 10

 

[mir]

I'm going. fixed barrel and rotary shaft:
the two wheels rotate solidified to the shaft and rotate in opposite ways than their axis

 

[giorgioc73]

therefore, to have confirmation, turning the disk of some degree in anticlockwise way you are not in this situation. Can you confirm?

 

[mir]

That's how you find me. I understand you say the wheels should be at 12 o'clock, right?

 

[mir]

 

[mir]

What would you expect?

 

[giorgioc73]

Then you don't work too!
look at the figure in #1 and #4. consider the contact point between the joint wheel to the barrel and that of smaller diameter. now compares the path section on the larger diameter wheel and that path on that of smaller diameter.
They're different.

 

[giorgioc73]

what I wrote on it is to let anyone who did not download the file. you who have it in your car can try the movement, see it better in front projection and orthogonal view, and see clearly that the wheels do not gear, but they slide among themselves.
the absurd thing is that if you block the axis and roar the rod, they gear perfectly.

 

[mir]

I think I have an explanation. the management of the corners in inventor makes "schifo" ... and worsens in the dynamics.

If you turn a wheel you realize it becomes a random number generator

 

[giorgioc73]

Good problem. more or less like saying that inventor is unusable for motion simulations.
Since it is one of the three-four most popular and used packages, it seems impossible. There must be some catch in what I did.
We need help.

 

[mir]

even when you think it works doesn't work... However almost certainly is a rounding error on the other hand the ratio 1/3 how many 0.333333333 must have?

above I got it by turning the mouse to "cazzum" for a few seconds in one verse and another

 

[giorgioc73]

I don't think it's a rounding error: I thought the same thing and tried the opposite bond (fixed 1:3 instead of 1:0.3333...). Same thing.
then there is the fact that the wheels go randomly, and otherwise, if you press the movement to the disk, to one of the wheels or to the other.
It just doesn't work.

 

[mir]

However in general I confirm why I have been trying to make a watch for a while and the simulations are painful.

 

[Peppe]

I have no 2020 version so I can't open it.
What constraints did you give to the wheels?

 

[giorgioc73]

the dynamic constraints imposed are two, as visible from the tree belowthe first is between the rod and the wheel 1the second is between wheel 1 and wheel 2the other constraints, three in all, are of insertion, to let free only rotation.
there is then the fixing of one of the parts, as explained in #1.

I tried if you could save the project with a previous version (as you can do in autocad), but on inventor it seems impossible.

 

[Peppe]

If I fix the disc and rotate the rod, the two wheels move as you expect.
if instead I fix the rod and rotate the disc... "it goes no."

But you've got a record connection with the pipe? if you roar the disk you "must drag" even the canton
I did tests and I can do what you ask regardless of what you try to rotate.

 

[giorgioc73]

Maybe I couldn't explain myself effectively: in the real application, the barrel is fixed with the supporting structure, while the disc rotates, tied axially to the barrel. the disk should not drag the barrel, but must rotate free, without any angular bond between them.

 

[Peppe]

All clear.
by experience the constraints of movement if "moves" by hand tend to jump.
have you tried to create an angular bond between the disk and a plan of origin and to animate it?

 

[giorgioc73]

nothing to do: it behaves exactly as with the movement imposed with the mouse.
In the meantime, they told me that even in the latest version, 2023, it behaves the same way.

It really seems impossible for me to be a problem of the program: It's too big a mistake.