Learning Forum

[Grant Stidham]

I am trying to model a simple flat belt drive system. After much research, I've heard the advice that you have to start with a circular belt, then have both pulleys tangent to the inside face of the belt. After this, fix one of the pulleys with a revolute joint and then a remote displacement to keep it from rotating. Then place a revolute joint on the other pulley, and a displacement joint load on that same pulley to displace it in the x-direction and tighten the belt. I have heard this advice and tried to implement it on my own, but to no avail.

As you can see, I've used a remote displacement instead of a displacement joint load, because a joint load requires a joint! And I'm unsure how to set that up correctly. The situation is pretty tight right now so if someone could please show me how exactly to stretch the belt using this method, it would be much appreciated! Thanks!

[Ashish Khemka]

Hi,

Please see if the following forum thread helps: Belt and Pulley Assembly (ansys.com)

Regards,

Ashish Khemka

[peteroznewman]

Is this one of the discussions you found? /forum/forums/topic/transient-structural-converge-on-a-solution-for-the-nonlinear-problem/

One of my replies in that discussion has a link to an Ansys 2022 R2 archive that you can download and examine if you have that version or later.

[Grant Stidham]

@peteroznewman I ended up figuring out the answer to this question the same day I sent it, several hours later. Just in case anyone will work on this type of project in the future, here is how it is done. The answer is inspired from an old ANSYS motion ACT app download (only available in the 2019 version, but can still open the file and view the model, contacts and joints in newer versions, just can't view solution results).

Answer:

Consider that you have a flat belt drive model with two pulleys and a circular belt (unstretched). The pulleys should be tangent to the belt, one on the left side and one on the right, but yet inside the belt as shown in the image at the top of this thread. IMPORTANT: there should also be a bearing part (a simple concentric cylinder) placed concentrically on the inside of what we will call the mobile pulley. 

After adding the appropriate frictional contact regions between the pulleys and belt (search "ansys flat belt drive" on YouTube), the user should create a revolute joint on the pulley that we will call the fixed pulley. This revolute joint should be placed on the pulley's inside diameter face, and should be designated "Body to Ground". Secondly, apply a "Body to Body" revolute joint between the outside face of the bearing and the mobile pulley's inside diameter face. Finally, create a translational joint (Body to Ground) for the bearing. Select the inside diameter face of the bearing when you do so.

In order to move the pulley, insert a "Joint Load" object in the "Transient" section (the same section where you setup loads and supports). In the details section, select the translational-to-ground joint. Then, apply the displacement type (according to @peteroznewman, using force, velocity, or acceleration is not recommended). Select the displacement you want, setup the time steps in the analysis settings, and the simulation should show the mobile pulley stretching the belt until the indicated displacement.

I hope someone can find this helpful, it's the only reason I'm including the details of my solution.