Opposite deformed part

[jackorsen88]

Hello everybody,

I have a problem with deformation of plastic parts. the simulation shows to much deformation and we decided to make an "opposite" part, that means that we will deform the part in CAD in opposite direction of the deformation shown by the simulation. I have a .stl from the simulation of deformed part wich I will use to make corrections to original CAD model. I use Siemens NX 11.0. Now I'm asking you wich is the best feature to use? Is global deformation the one? I tried global deformation and I got some results, but I don't know if I have used it the right way.

I made global deformation, with facet bodies. I used .stl made from original CAD and original sheet CAD and deformed stl. is that right?
I don't know what is the difference between compensated and sprung? and how do I get, or what do I enter in deformation factor? it's difficult to get the deformation factor, because the part is deformed in all three directions. And what is transition region?

P.S.:you can see the results I got in the picture attached

I hope I have described my situation clear.

Thanks for your help

 

[racookpe1978]

I don't know the .stl codes, so I cannot comment on their properties.

But, if this is a plastic part, put the original (as-intended final dimensions) into a 3D additive modeling form machine. Deform it according to the "real loads" you expect to verify your "model world" is really going to deflect the way the real world works.
Make a "deformed model" offset opposite the movement you expect based on the test piece, and put that through the 3D additive manufacturing process, then apply the "expect loads" to that model. See it the deformed model moves as you expect, or at least moves close enough to make the final product acceptable for use.

 

[MikeHalloran]

You can only precompensate the deflection for the one load case for which you modeled same.
Most parts are not subject to just one load case.

It might be simpler to just thicken the part, at least in the sections that are most stressed.



Mike Halloran
Pembroke Pines, FL, USA

 

[3DDave]

I've wondered about using this method in actual applications. For example, when creating a sealed cover, to design the cover so that the load between bolts is uniformly distributed. I made a straight edge for use as a paper cutting guide that I formed carefully so that pushing at only the mid-point gives a uniform load along its length which prevents the paper from spinning or shifting; basically forming it according to the curve of a uniform beam with a single center support and uniform load.

However, CAD/CAE systems don't support this, so parts end up heavier than required.

If you can get an STL of deformed part, maybe apply the loads in the opposite direction. For small deflections the deflection with load is linear, so changing the sign of the loads will result in the opposite deflection.

 

[MikeHalloran]

I formed carefully so that pushing at only the mid-point gives a uniform load along its length which prevents the paper from spinning or shifting; basically forming it according to the curve of a uniform beam with a single center support and uniform load.

Recent injection molded windshield wiper arms are shaped to perform exactly in that way. They appear to be a tapered beam with considerable camber in the unloaded state, but since they are prismatical beams, the necessary equations can be found without a CAD system.

A CAD/CAM system becomes useful for transferring the design intent to the tool.




Mike Halloran
Pembroke Pines, FL, USA