Rotating Shaft with Rigid Element Supports

[kseminsky]

Hoping there is a way to "free" the rotation at the common node for the rigid element? See images attached. Simulating a driven shaft (torque applied to the top end) and a pinion gear meshing into a rack. I am using a constraint at the gear mesh, and constraints the bearing housing bolt holes. Point of the analysis is to check the strength of the housing, and get mounting bolt reactions. 

The problem I am having is with the rigid and line element interface. I want the shaft (modeled by line elements) to rotate freely at the bearing center points (modeled by nodes). If I use end release at the bearing center nodes on the line elements, the line elements spin and my analysis fails. How do I keep the line elements connected to transmit torque, but free to rotate within the rigid element? 

Open to thoughts. Trying to keep the model simple and focus on the housing rather than modeling the shaft and bearings, using surface contacts, etc... Much appreciate any thoughts! Thank you. 

[kseminsky]

Bump. I have some additional analysis coming up, and if this is a proper way to constrain a rotating line element, I'd like to use this as a tool. 

Thank you in advance for any thoughts. 

[John_Holtz]

Hi @kseminsky (Sorry. I thought I had replied, but I found this answer in my draft messages.)

Freeing the rotation at the center node of the rigid connector is the correct idea, but the rigid connector does not have that ability. 

The only thought that I have is to create a short beam sticking perpendicular to the shaft and extending a short distance. (Like 0.1 times the diameter of the hole or even shorter.) The rigid connector would go from the surface of the hole to the free end of this short beam (instead of to a point on the shaft). A beam end release would be added to the short beam to allow the shaft to rotate. Make the short beam "strong" but massless.

[kseminsky]

John, Thank you for your thoughts! This seemed to work perfectly!