AnsweredAssumed Answered

Setting up a closed-loop static study

Question asked by Samuel Leith on Feb 25, 2014
Latest reply on Mar 22, 2014 by Samuel Leith

Like • Show 0 Likes0
Comment • 14

Hello,

this is the first time I need to set up a 'closed-loop' static study - see image attached.

This is iteration one of a concept.

The part in white is a torque wrench. It applies a torque to the red part (via a square drive which we can't see). The red part is bolted to the base plate.

You can see there is a yellow block on the torque wrench. This is the reaction arm, which pushes against the back plate when the torque is applied.

The back plate, base plate, and gussets are all welded.

What's the proper way of setting up a study for this?

I was thinking of using the 8 threaded holes in the base plate as fixtures, and apply a force where the reaction arm meets the back plate.

I'd run the one study with all parts as solids, with contact 'No Penetration' for the edges that are welded.

Then, as a second study, I'd simulate the gussets as shells and check for weld sizes.

Is this how you would set this up?

Thank you.

Attachments

Capture.PNG96.7 KB

No one else has this question

Outcomes

14 Replies

Jared Conway Feb 25, 2014 12:52 PM
Mark Correct
Correct Answer
what do you mean by closed loop?

what do you want to learn in this analysis?

is the picture the right picture? i don't see a white part?

is this a commercial or educational project? it sounds like it is something relatively challenging that getting some training might help limit your frustrations and improve accuracy and efficiency.
Like • Show 0 Likes0
Actions
Jerry Steiger Feb 25, 2014 3:29 PM
Mark Correct
Correct Answer
Samuel,

Like Jared, I'm confused. By white part, do you mean the light grey part between the yellow face and the dark part in the center of the red part?

I don't understand how you can apply torque if the yellow part is stopping the torque wrench from moving around the axis of the red part.

Is the torque wrench an assembly of multiple parts? How do they move relative to one another?

Jerry S.
Like • Show 1 Like1
Actions
Samuel Leith Feb 25, 2014 3:38 PM
Mark Correct
Correct Answer
It's true that at the angle shown in the screenshot, the 'white' part looks more like it's light grey.
The torque wrench is the one that's light grey.

It's a hydraulic torque wrench. Once the torque wrench is in place, you activate the hydraulic pressure. The entire wrench spins until the reaction arm hits the backplate. From there on out, torque start to build.
Like • Show 0 Likes0
Actions
- Jerry Steiger @ Samuel Leith on Feb 25, 2014 3:57 PM
  Mark Correct
  Correct Answer
  Samuel,
  
  Ah, got it! In that case your approach of applying a force at the end of the arm and an opposite one on a split area on the back plate is the one I would take. Either that or apply torques on the square drive on the other end. I'm not so sure about checking the weld sizes, as I have no experience with welding.
  
  Jerry S.
  Like • Show 1 Like1
  Actions
  - Samuel Leith @ Jerry Steiger on Feb 25, 2014 4:05 PM
    Mark Correct
    Correct Answer
    Jerry,
    for this concept, both the torque wrench and the red part are purchased components. What I'm really looking to validate is the base plate, the back plate, and the gussets behind the back plate.
    
    With this in mind, how about applying a force on a split area on the back plate, and setting the 8 holes on the base plate (ie. the holes the red parts bolts into) to fixtures?
    Like • Show 0 Likes0
    Actions
    - Samuel Leith @ Samuel Leith on Feb 25, 2014 4:06 PM
      Mark Correct
      Correct Answer
      What I mean to say is the torque wrench and the red part were shown just for representation. I do not want to include them in the simulation.
      Like • Show 0 Likes0
      Actions
      - Jared Conway @ Samuel Leith on Feb 25, 2014 11:51 PM
        Mark Correct
        Correct Answer
        seems reasonable enough, just try it
        Like • Show 0 Likes0
        Actions
Samuel Leith Mar 10, 2014 9:30 AM
Mark Correct
Correct Answer
I realized I should have used the term 'internal forces' as opposed to 'closed loop'.

The majority of the time, when doing FEA, a part/assembly is constrainted and an external load is applied.

In this case, the loads are internal, which is what is throwing me off a little bit.

Please have a look at the picture attached, which is an image I found on google of a case similar to mine.
The blue tool is the hydraulic wrench, which applies a torque on the fixture and also a reaction force (as indicated by the red arrows).
How would you go about setting up an FEA study to validate the fixture (shown by the green arrow)?

I've been thinking about this for a week+, and I'm still not sure. There are no external loads on this assembly. If I was strong enough, I could simply put this assembly in the palm of my hand and crank up the hydraulic pressure and it wouldn't go anywhere, or it wouldn't get any harder to hold.

From what I've tried so far, I know I can't set fixtures on the bottom faces that are in contact with the table because it over-contraints the system. These faces are in fact not constrained and could therefore deform (ie. lose contact with the table).

What contraints would you use?
Thanks for your help.
Like • Show 0 Likes0
Actions
- Jerry Steiger @ Samuel Leith on Mar 10, 2014 4:21 PM
  Mark Correct
  Correct Answer
  Samuel,
  
  I'm no expert in using SolidWorks Simulation, but I would just pick a vertex at one of the corners and fix it in X, pick one at another corner and fix it in Y, and pick one at a third corner and fix it in Z. Since the corners are quite a ways away from the areas where the fixture will be stressed, the fixed points probably won't affect the stress much. Since the back corners are closer to the areas of interest, you might consider fixing one front corner in X and Y and the other in Z. Or use the back corner that is hidden in your picture, since the stresses are probably less there than in the corner closest to the reaction arm.
  
  Jerry S.
  Like • Show 0 Likes0
  Actions
- Jared Conway @ Samuel Leith on Mar 10, 2014 6:10 PM
  Mark Correct
  Correct Answer
  "The blue tool is the hydraulic wrench, which applies a torque on the fixture and also a reaction force (as indicated by the red arrows)."
  
  how are these not external forces to the fixture?
  
  if there aren't any loads on the fixture, why analyze it?
  Like • Show 0 Likes0
  Actions
Samuel Leith Mar 13, 2014 3:24 PM
Mark Correct
Correct Answer
Jerry,
The vertex initially sounded like a possible option, however, I noticed the displacement results were being affected. The corner points are essentially points which have the largest displacement - so fixing them in space affects the result quite a bit.

Jared,
By this I meant that the loads are internal to the system. Yes there are loads on the fixture, but all of these loads are applied by the torque wrench: the torque + the reaction force.
For example, standing inside a box and pushing on one of its walls versus standing in front of it and pushing on one of its walls are two different situations.

Since these loads are internal to the system, there is no constraints to hold the system in place (like bolts for example).
Knowing the torque of the torque wrench, the reaction force can easily be calculated.
Now in simulation, when setting these two loads (the torque & force), the simulation will not work because there is no constraints.

Essentially, I was wondering how to apply constraints without over-contraining.
At the end of the day, this L ange fixture can deform in every possible plane because nothing is holding it down.

I am not an FEA expert either. I just like to solve problems.
Like • Show 0 Likes0
Actions
- Jared Conway @ Samuel Leith on Mar 13, 2014 3:32 PM
  Mark Correct
  Correct Answer
  lets use your box example. how would you propose to analyze that?
  Like • Show 0 Likes0
  Actions
- Jerry Steiger @ Samuel Leith on Mar 18, 2014 6:07 PM
  Mark Correct
  Correct Answer
  Samuel,
  
  If fixing some points doesn't work, you might try simulating the actual test setup. Model the rubber feet with something softer than the steel. Bond them to the plate. Place a floor beneath your plate. Add no-penetration contacts and friction to the feet and the floor. Use a fairly high coefficient of friction, say between 0.5 and 0.8. Add a gravity load to the plate and the torque wrench. Run the simulation. The friction should prevent the plate from moving significantly.
  
  Jerry S.
  Like • Show 1 Like1
  Actions
Samuel Leith Mar 22, 2014 9:19 PM
Mark Correct
Correct Answer
Sorry for the delay in replying to my own thread :s
With your feedback, there are a few test studies I would like to run. I plan on getting to the bottom of this, but I'm doing this during my spare time, so it might take a little longer than usual. I will update the post with my findings.
Like • Show 1 Like1
Actions

Setting up a closed-loop static study

Attachments

Outcomes

Related Content

Recommended Content