Structural test and shape optimisation

[MrRBall]

Hi,

I may be asking too much here?.... I am looking at doing some simple (?) load testing and shape optimization but I'm a little lost. I've attached a screen shot of a basic plastic hook I plan to 3D print. I wanted to do a test to see the weak points and then to use shape optimization to show how the design can be improved. (Is this what it is for???) Would anyone be willing to show me an idiots guide on how to do this? I've watched a couple of tutorials but still a little lost. Thanks in advance,

Ryan

[todd_alford]

Hi @MrRBall,

Let me try to clarify what you are trying to do, and try to set up some accurate expectations.  Your described workflow sounds like a good start and I think you are in the right frame of mind.

I have had a background in simulation for quite some time, so I'll start by explaining what I would do.  I would start with a Static Stress study (that is where you can find the weakest part of the design, and don't forget the bigger bonus if it would fail (not really worth printing if you are going to have a couple broken parts when you are done).

This tutorial runs through setting up a couple of Static Stress studies.  The examples are a bit more complex than what you are trying to do (you have a single part so contacts aren't necessary, stuff like that).  The basic titles for each of the pages should be the workflow that you would follow though.  It looks like you have some constraints and a force already on the hook.  The step that I'm not sure you have accurately done is applying the material.  That might take a little bit of research to find the properties of your 3D printers material.

I did want to comment on your constraints, it looks like you may have "fixed" the back surface of the hook.  Just to clarify what your constraints are doing, that means the top surface of the hook has no ability to move and neither does the back surface.  Lets say that this was a standard hook that goes over a door.  In this case it would be saying that the back surface has some sort of perfect glue to keep it attached to the door and the top.  In reality most of those hooks would only have the top constraint, as we know the hook will not penetrate the door and the door will not be a point of failure.

That is the best insight I can give you into the static stress simulation that you are trying to do without further details.  So I'll let you tell me how much that helps and what other questions you may have.

Now that we have verified that our part can handle the expected weight from the first analysis let's say that you find out you have a factor of safety of 15 everywhere (that means that the hook can handle a lot more weight than you had prescribed in the static stress analysis).  In that case you would go to shape optimization and use the same constraint and loads and set up preservation regions on the areas of the hook that you want to be kept (the location that the hook attaches and where the object hangs presumably.  The shape optimization will give you an idea of where material can be removed when you go to print it.  Shape optimization only removes material it will not add material

At this point you would want to make modifications to the geometry and make sure that not too much material was removed using the shape optimization by running another static stress analysis on the new geometry. 

Does that workflow make sense?  Am I missing the areas that you are confused about?

Thanks,
Todd

[MrRBall]

Thanks Todd. What you are saying seems logical - especially about the door - glue etc.

I get THIS (attached) as a result - it looks pretty but what does it mean?!!

Also, I constrain the attached parts for the hook and run a test but it fails... Where do I input data? Sorry - I'm very much a newbie with this!!

[MrRBall]

So is this what it should look like?...

[todd_alford]

@MrRBall, don't worry about it.  Simulations isn't the simplest thing in the world, it's pretty far from it.  I dare you to start talking about simulation while in line at the grocery store :). 

Let's discuss your pictures 1 at a time, it looks like you are starting to run the shape optimization.  The first picture was showing the preserve regions, so these  are the areas that the solver will not consider changing.  So these areas must keep their mass, but the solver is attempting to remove enough material to reduce the amount of mass to the specification (what is in the Shape Optimization Criteria). 

The second picture is the actual result stating that if you wanted to remove X% of the mass this is the most efficient path to distribute the load.  I would wager a guess that the part would fail with that much mass missing (it also removed the top of the hook which is pretty important, if you are trying to hang something).

Did you run the Static stress analysis first?  You could use that as a guideline to chose how much mass you are trying to remove.  Let's say the minimum factor of safety in the stress analysis is 6.  That means your design is pretty reasonable and you wouldn't want to remove too much material, or when you go back you'll find that you have removed too much and the part will fail.  It is probably worth reading this page in the help discussing factor of safety there are some additional references listed in the Factor of safety section.  If you want to share your file with me I could show you a couple different set ups.  Do you want me to take a closer look?

Thanks,

Todd

[MrRBall]

Hi Todd, 

I would very much appreciate sharing my file with you! How do I go about this?

[todd_alford]

Hi @MrRBall,

You can export the .f3d file (File Export, to save it local and send it to me) and post it here on the forum.  If you want to keep the file from being publicly available you can click on my name in the top left corner of any of my posts and send it to me in a private message or email.  I'll take a look at it as soon as I can once you get it to me.

Thanks,

Todd