AnsweredAssumed Answered

Rotating Shaft FEA Study

Question asked by Baktash Bariz on Aug 3, 2011
Latest reply on Aug 4, 2011 by Baktash Bariz

Like • Show 1 Like1
Comment • 12

Hello all,

I seem to have run into an issue with a FEA study and can't seem to get the correct setup for fixture and load application. I am hoping someone will provide the much needed feedback!!!

PROBLEM:

1. Horizontally rotating shaft with both ends supported vertically (not fixed).

2. Vertically downward force acting at the middle of shaft.

ISSUE: Study fails because of fixture/constraint issues.

I have tried numerous fixture/load combinations but still get a study failure!!! Should I be performing a static study first and then modify to a frequency study? Or is there a another approach?

Thank you and hope to hear back!

Baktash

Baktash Bariz Aug 4, 2011 9:23 AM

Correct Answer

I figured it out on my own. The trick was to fix the ends of the shaft and apply a reversing cyclic load (max to max) on one plane. This would simulate the load applied every 180 degrees. Thank you all for your feedback. Very helpful.

Baktash

See the reply in context

No one else had this question

Outcomes

Michael Feeney Aug 3, 2011 1:10 PM

Are those supports only suppressing the vertical degrees of freedom?

Actions

Baktash Bariz @ Michael Feeney on Aug 3, 2011 1:29 PM

Michael,

The goal is to fix all degrees of freedom except one rotaional at those supports. I require the shaft to rotate along its axis only. The picture does not show all DOFs.

Actions

Michael Feeney @ Baktash Bariz on Aug 3, 2011 1:38 PM

I don't fully understand what you're trying to do. Determine when the shaft will fail due to a static load with a non-rotating shaft? Natural frequency analysis? What exactly is it?

Actions

Baktash Bariz @ Michael Feeney on Aug 3, 2011 2:06 PM

I have already determined the stresses due to static loading as shown on the picture. Now I am trying to study what happens to the stresses when the shafts starts to rotate at a constant rpm with the same load pointing vertically down. I can predict that the load will reverse every 180 degrees. Does that help Michael?

Actions

Tim Goddard @ Baktash Bariz on Aug 3, 2011 2:06 PM

Maybe my SW simulation is license limited or it's my knowledge of the product, but how exactly are you rotating the shaft? (i.e. loads etc). Is this paired with a motion study?

Actions

Baktash Bariz @ Tim Goddard on Aug 3, 2011 2:12 PM

The shaft doesn't have to rotate. It could be simulated using cyclic loading. My problem arises when I get fixture errors which means my part isn't fully fix or support properly.

Actions

Michael Feeney @ Baktash Bariz on Aug 3, 2011 2:27 PM

So you want to follow a single element or point on the shaft and determine the stress fluctuation at that point as the shaft rotates? I wonder why this is important to know from a failure standpoint seeing as the maximum stress will be clear with a static analysis. Tim is right about the being a motion study. You have your model set up there, correct?

Actions

Baktash Bariz @ Michael Feeney on Aug 3, 2011 2:35 PM

Correct, that would be another way of saying it. You can think of it as bearing problem where we need to find out the fatigue stress.

Actions

Tim Goddard @ Baktash Bariz on Aug 3, 2011 3:11 PM

I'm interested in how SW handles this. I'm limited to the standard Simulation with SW Premium so nothing but static-linear for me.

From a fatigue-solution point of view, the static solution is all you need. The cyclic stress is represented by the stress around the circumference. From a stationary point of view, the stress on a rotating shaft with stationary load will never change position (max tensile is opposite load application, compression on load side). Only from the viewpoint of a point on the shaft will it appear cyclical.

If this problem isn't academic then you should consider the dynamic stability of a rotating shaft and see whether you are operating near any critical speed.

Actions

Deepak Gupta @ Baktash Bariz on Aug 3, 2011 5:04 PM

Did you tried using motion studies.

Actions

Baktash Bariz @ Deepak Gupta on Aug 3, 2011 5:21 PM

No but I think I got it.

Actions

Baktash Bariz Aug 4, 2011 9:23 AM

Baktash

Actions

12 Replies

Michael Feeney Aug 3, 2011 1:10 PM
Mark Correct
Correct Answer
Are those supports only suppressing the vertical degrees of freedom?
Like • Show 0 Likes0
Actions
- Baktash Bariz @ Michael Feeney on Aug 3, 2011 1:29 PM
  Mark Correct
  Correct Answer
  Michael,
  
  The goal is to fix all degrees of freedom except one rotaional at those supports. I require the shaft to rotate along its axis only. The picture does not show all DOFs.
  Like • Show 0 Likes0
  Actions
  - Michael Feeney @ Baktash Bariz on Aug 3, 2011 1:38 PM
    Mark Correct
    Correct Answer
    I don't fully understand what you're trying to do. Determine when the shaft will fail due to a static load with a non-rotating shaft? Natural frequency analysis? What exactly is it?
    Like • Show 0 Likes0
    Actions
    - Baktash Bariz @ Michael Feeney on Aug 3, 2011 2:06 PM
      Mark Correct
      Correct Answer
      I have already determined the stresses due to static loading as shown on the picture. Now I am trying to study what happens to the stresses when the shafts starts to rotate at a constant rpm with the same load pointing vertically down. I can predict that the load will reverse every 180 degrees. Does that help Michael?
      Like • Show 0 Likes0
      Actions
  - Tim Goddard @ Baktash Bariz on Aug 3, 2011 2:06 PM
    Mark Correct
    Correct Answer
    Maybe my SW simulation is license limited or it's my knowledge of the product, but how exactly are you rotating the shaft? (i.e. loads etc). Is this paired with a motion study?
    Like • Show 0 Likes0
    Actions
    - Baktash Bariz @ Tim Goddard on Aug 3, 2011 2:12 PM
      Mark Correct
      Correct Answer
      The shaft doesn't have to rotate. It could be simulated using cyclic loading. My problem arises when I get fixture errors which means my part isn't fully fix or support properly.
      Like • Show 0 Likes0
      Actions
      - Michael Feeney @ Baktash Bariz on Aug 3, 2011 2:27 PM
        Mark Correct
        Correct Answer
        So you want to follow a single element or point on the shaft and determine the stress fluctuation at that point as the shaft rotates? I wonder why this is important to know from a failure standpoint seeing as the maximum stress will be clear with a static analysis. Tim is right about the being a motion study. You have your model set up there, correct?
        Like • Show 0 Likes0
        Actions
        Baktash Bariz @ Michael Feeney on Aug 3, 2011 2:35 PM
        Mark Correct
        Correct Answer
        Correct, that would be another way of saying it. You can think of it as bearing problem where we need to find out the fatigue stress.
        Like • Show 0 Likes0
        Actions
        Tim Goddard @ Baktash Bariz on Aug 3, 2011 3:11 PM
        Mark Correct
        Correct Answer
        I'm interested in how SW handles this. I'm limited to the standard Simulation with SW Premium so nothing but static-linear for me.
        
        From a fatigue-solution point of view, the static solution is all you need. The cyclic stress is represented by the stress around the circumference. From a stationary point of view, the stress on a rotating shaft with stationary load will never change position (max tensile is opposite load application, compression on load side). Only from the viewpoint of a point on the shaft will it appear cyclical.
        
        If this problem isn't academic then you should consider the dynamic stability of a rotating shaft and see whether you are operating near any critical speed.
        Like • Show 0 Likes0
        Actions
        Deepak Gupta @ Baktash Bariz on Aug 3, 2011 5:04 PM
        Mark Correct
        Correct Answer
        Did you tried using motion studies.
        Like • Show 0 Likes0
        Actions
        Baktash Bariz @ Deepak Gupta on Aug 3, 2011 5:21 PM
        Mark Correct
        Correct Answer
        No but I think I got it.
        Like • Show 0 Likes0
        Actions
Baktash Bariz Aug 4, 2011 9:23 AM
Unmark Correct
Correct Answer
I figured it out on my own. The trick was to fix the ends of the shaft and apply a reversing cyclic load (max to max) on one plane. This would simulate the load applied every 180 degrees. Thank you all for your feedback. Very helpful.

Baktash
Like • Show 0 Likes0
Actions

Rotating Shaft FEA Study

Outcomes

Related Content

Recommended Content