Simulation Mechanical Forums (Read-Only)
Welcome to Autodesk’s Simulation Mechanical Forums. Share your knowledge, ask questions, and explore popular Simulation Mechanical topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Reply
Message 1 of 6
imalleab
1880 Views, 5 Replies

Contact stresses

Hi,

 

I'm currently working on a university project and I'm designing a tube straightener machine for an specific application. Here is a drawing of the machine.

 

imagenforomechancialsim.png

Hoping you can see, the tube (slightly bent) comes into the rollers/wheels and once it makes contact with them, the rollers drags it by friction until the tube passes through. I've been searching info about the stresses occurred on the contact tube-roller and for the specific case I couldn't find it. So I was wondering if Simulation Mechanical has a tool to calculate contact stresses for this particular case (Normal force and tangential force (friction) on Point Contact surface).

 

Thank you in advance

 

Ignacio.

5 REPLIES 5
Message 2 of 6
John_Holtz
in reply to: imalleab

Hi Ignacio,

 

You could perform an MES analysis on the rollers and tubes to calculate the stresses. It would probably take a week to setup and get an answer (depending on what simplifications you want to make. For example, how important is the friction compared to the normal contact.) I would begin with a non-moving tube and bring the non-spinning wheels into contact.

 

The other simulation products like Nastran In-CAD and Fusion 360 can also do a similar type of analysis if some simplifications are permissible.



John Holtz, P.E.

Global Product Support
Autodesk, Inc.


If not provided already, be sure to indicate the version of Inventor Nastran you are using!

"The knowledge you seek is at knowledge.autodesk.com" - Confucius 😉
Message 3 of 6
Keith.Orgeron
in reply to: imalleab

Ignacio,

 

I have had good luck with simulating Hertz contact and friction loading and stresses with deflections with Simulation Mechanical.  I have also had good luck with the use of a free software package called HertzWin by an engineering firm named Vink (Netherlands) for very local linear Hertz contact stress analysis.  Trying the model the Hertz contact behavior with the required, very small element sizes does not leave much processing capacity for modeling the global behavior.  So, I recommend the following approach:

 

GraphsHertzWin.png

 

 

HertzWin_en.png

  

  1. Assuming it is possible to remain within the linear elastic realm (no yielding) this will give an excellent solution
  2. Develop a global model, (and forgive my boldness in giving these recommendations):
    • As a machinery designer, I believe you may need at least one final pair of exactly opposing rollers to develop most of the pulling traction force
    • These opposing rollers would allow pulling without over-bending the wire away from any one single traction roller
    • The remaining, trailing rollers would be responsible for the straightening functions and could add some traction force too
    • Use prescribed rotations (PR's) for all driven rollers which contact the wire with friction
    • Optionally, set the base of each roller on prescribed displacements (PD's)
      • Control the roller bases to move inward from a slightly contacting initial position
      • Given no significant forming occurs, linear elements/materials should suffice and help the MES processing speed
      • Start with the wire and one pair of traction rollers
        • Then, add one or two straightening rollers
  3. Extract the total contact loading at each roller by selection of the contact nodes for each roller
  4. Download the HertzWin software for calculating Hertz contact load, stress & displacement details
  5. Good Luck!  (Send pics and models!)
Message 4 of 6
imalleab
in reply to: imalleab

Thank you both for your guidance!

 

John: As you suggested me, I'm trying to calculate the case with no movement, just the normal force between wheel and tube. To start I'm focusing on the case of a cylinder (wheel) on a flat surface (tube) and comparing the stresses with theory results (or the ones from the program Keith mentioned). I have a couple of questions:

 

Since the tube is expected to suffer a full plastic deformation (due to straightening) should I use a static linear model or a static nonlinear one? Here's a picture of the model.

modeloSM1nonlinear.png

 

 

  • I applied four nodal forces at the base of the lower body trying to simulate a line distributed load (which I didn't find in the program).
  • The wheel is fully fixed in the center.
  • The lower body is fixed on left (fixed Ty and Tz) and right (fixed Ty) edges.
  • Surface contact between wheel and flat surface.

 

Keith: The machine is meant to straightening copper tubes of 1,3 inch outside diameter and they will be mechanically introduced between the wheels. At the end they will fall by gravity into a chute.

 

About HertzWin, I've been trying to use it, basically to compare the results with those from Simulation Mechanicals and see if I'm heading to the right place. At this point I'm not having good results, so I assume  I'm doing something wrong in my FEA model or in HertzWin. Here's what I'm doing.

 

modelo1HW.png

 

I'm not sure about what is the Maximum Stress value and the Length value. I assume the first one is the yield tensile strength and the second is the width of the bodies in line contact (length of line of contact). I also have a doubt about the force, it is a normal point force or it is applied as a distributed load along the line contact area pre deformation?

 

Thanks you again.

 

Ignacio

 

 

 

 

Message 5 of 6
Keith.Orgeron
in reply to: imalleab

Ignacio,

 

Some quick notes that may help:

 

  1. See annotated images below & visit my FEA page at OrgeronEngineering.com/FEA
    • The key here is to refine the mesh to sufficiently sensitive the model to the physics behavior simulated
    • I have a 7-Step FEA™ guide and many FE model images that may help on my FEA web page
      • Your effort represents two separate FEA projects... Tube straightening & Hertz contact behavior
      • Accurate Hertz simulation can be 2D and is not needed simultaneously to simulate tube straightening
        • Tube straightening is an event involving both material & displacement nonlinearities
  2. See the HertzWin Help page - they do a good job with an example and definitions of the inputs & outputs
  3. See the AD SimMech 2018 User's Guide page on loads - AD SimMech - Loads (shortened link)AD SimMech Hertz_B 2017Apr23.pngAD SimMech Hertz_A 2017Apr23.png

 

 

Tags (3)
Message 6 of 6

I have also developed an open-source web app

 

https://foadsf.github.io/OpenHertz/

 

that pretty much does the job. You may read more about it here on Reddit.

 

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report