Pin Constraint giving incorrect stress pattern.

[Anonymous]

Hi everyone!

I am having a lot of trouble understanding "Constraints vs Contacts" in Inventor stress analysis.

Firstly, when should I use constraints and when should I use contacts? As far as I can understand, I should use constraints for describing the motion/behaviour of a body with respect to the environment(How the body is fixed in space) and contacts for defining the behaviour between different bodies in an assembly. Am i right or wrong or partly right??

http://help.autodesk.com/view/INVNTOR/2014/ENU/?guid=GUID-414FB139-FF74-4724-9EFE-F60F9D7975BD

In the first line itself, it states "You add constraints to mimic environmental conditions."

http://help.autodesk.com/view/NINCAD/2017/ENU/?guid=GUID-53F9F05F-0C67-4487-A702-C703108482AB

It states other than constraints and contacts, Use loads. How am I supposed to use loads to restrict a part from moving??

Coming back to my issue at hand,

Using pin constraint yields no discernable change in stress or deformation behaviour of the assembly. There is also no stress arising from the contact between the two bodies. Irrespective of the pin constraint and the sliding contacts present, the lever is behaving as if it were bonded to the shaft instead of being hinged on the shaft. This is obvious in the stress pattern generated in the shaft. The shaft stress pattern is that of torsional stress induced and not of simple shear as it should be.

Kindly take a look at my screencast and let me know where it is going wrong.

I have made 3 studies,

1- Only pin constraint, with automatic contacts(It considered everything as bonded).

2- Both pin Constraint and manually modified the contact to sliding/No Separation contacts.

3- Only used Sliding/No Sep contacts without the pin constraint.

In all the three cases, the lever is not acting like it is hinged about the shaft. Instead, it is behaving as if it were fixed to the shaft. This is evident from the lack of stress at the point of contact between the lever and the solid bar at the bottom and also by the pattern of stress induced in the shaft. The pattern is that of torsional stress induced in the shaft and not of shear stress induced in the shaft.

Any help/info provided would be of great help.

Screencast will be displayed here after you click Post.

e3013e3c-a78f-4f39-96c9-72dee86fedc3

[JDMather]

I do not see any Analysis set up in the assembly that you attached?

Oops, are you using Nastran In-CAD?

Try the Screencast again (you cannot edit post after inserting Screencast.

If all else fails, post link to Screencast.

[Anonymous]

@JDMather I tried using both, Inventor stress analysis as well as Nastran in Cad. The output from Nastran In cad is even more confusing. I'll create another screencast of Nastran In Cad(or I might take a series of screenshots coz NIC has been crashing a lot after Inventor 2018.3 update) and will upload it within 3 hours.

[Anonymous]

@JDMather,

please take a look at the videos below.

https://autode.sk/2DW82qb

Please ignore the last 7 seconds of the above video and look at the video below instead.

https://autode.sk/2TJaZig

There are 3 subcases of constraints.

1. No constraints on the shaft and lever.(Only the sliding/No separation contact does all the magic).(results look like they are correct)
2. Pin on the lever hole with the coordinate system as the CS of the assembly.
3. Pin on the lever hole with CS as the Local Coordinate system of the Block.

The stresses in the above videos are as one would expect. However, If i deactivate contact analyser and move the lever away slightly(not in contact with the bar at the bottom) and rerun the analysis, the results look just like the results of the Stress Analysis in Inventor.

I have another question.

In https://autode.sk/2TJaZig I have selected the CS of the pin constraint as the Local Coordinate system of the block. Then why is the result wrong? I want to define the constraint such that the lever is free to rotate about the shaft. So that it when placed such that the lever is not in contact with the bar, it initially moves and comes into contact with the bar at the bottom and then the stress will rise at the point of contact.

[JDMather]

I suspect that there are a number of incorrect workflows going on here.

 I suspect that you should be using Dynamic Simulation 2D (or 3D) Contact Joint and then Motion Loads Analysis at various time-steps.

Otherwise drill down and watch the two videos here -

http://knowledge.autodesk.com/support/inventor-products/getting-started/caas/CloudHelp/cloudhelp/201...

[Anonymous]

Thank you 🙂

I'll try using motion loads analysis and couple it with stress analysis and see.

I thought inventor would be able to automatically infer that it is supposed to move. I guess not.

Well, I'll do the motion load analysis and check it.

Thanks a lot sir. 🙂