very cool. cool method. I wonder if you used less fill or less stiff material if you could visually see the stress as the individual triangles deform. maybe even just a thinner plate? when you set up the FEA at the end you are using a fixed bearing support at the rocker pivot? the "seatstay" link is just a two force fixed member? i imagine the shock link is a two force member with a load along the link axis? is this a section view of half of the assembly? or are you not worried about asymmetrical loading by the links only being on one side? i' curious about the proper FEA setup. thanks
Hey John, this screenshot will help make more sense of my setup. I may do a more in depth video about the choices I made in setting up this simulation: i.imgur.com/X4eTNWX.png All the swingarm, seatstay, and "shock" components are defined to be rigid bodies (with frictionless pin connections and fixed hinge fixtures, they end up as 2 force members), leaving only the rockers and front spacers as deformable. Everything is positioned as it would be in bottom-out position, with a 500 lbf vertical load applied at the axle.
I think the only thing I'd do differently is print with the lattice closed in so there is a continuous web. In the test it looked like more of a lattice failure which the fea wouldn't capture and I assume wouldn't be representative of the final part either.
I'm not sure what you mean by "closed in". Do you mean solid top and bottom layers? There are definitely a number of ways I could make the print more closely mirror the FEA. This was a spur of the moment decision to break a mock up print. How the failure propagated through the lattice is nothing like a metal part would fail. But, the areas of highest tensile and compressive stress matched the FEA very well and the failures began at these points.
@@APEX.86 yeah, solid top and bottom. I'm just coming from the angle of trying to get both the FEA and test piece to represent the performance of the final part (not necessarily each other, which I assume would be hard for a printed plastic part anyway). It's just maybe the next guy might not be aware of how to properly interpret/ignore the similarities and differences in results. I am curious if the print failed more in tension on the vertical side or shear on the horizontal side at that high stress bore because that's the main difference I would expect to see. As you say, for just finding high stress areas that's propably enough agreement. It would be more important if trying to use the result numbers to design closer to the mark.
@@thehackerman1234 I actually printed new links fully solid and tried to break them. I ran out of courage at like 260 psi in the shock. It's not worth ruining my frame! I think the solid links could actually hold up to some light trail riding, which is wild.
Are you using normal solid elements to mesh that lattice print? If so, that simulation is not representative at all, and using it in a "FEA vs real world comparison" is just misleading.
What do you mean by "normal solid elements"? That question isn't very clear. These printed parts were meant for fitment checking, not as a direct comparison to the FEA of a metal part. There are a number of differences: the printed part is not a linear-elastic material, it is not a homogeneous piece of plastic because of the lower density infill, etc. But, if you know where to stop the comparison, it doesn't really matter. The location of areas of highest stress don't care what the material is. The sparse infill has the capacity to alter this to some degree, but in this case it didn't. Where will a plastic part yield first? The area(s) of highest stress... just like a metal. The way the failure propagated through the printed part is not comparable to a metal at all. But, we saw compressive buckling right at the two highest areas of compressive stress in the FEA result; and we saw necking in the areas of highest tension. This is not a scientifically rigorous comparison between FEA simulation and real world destructive testing. This was a spur of the moment opportunity to break something in a fun way and see how it compared.
@Antonio Alessandro Deleo Don't take it "the wrong way"? Your initial comment is incredibly insulting. You have no idea about any of the details of my FEA setup, but decide to claim I have done everything wrong, and that I have no understanding of the underlying concepts, and that my results are "garbage". This speaks volumes about you, that you would make these bold claims with absolutely no information to back them up. I challenged you to point out ONE thing wrong with my simulation, and you failed to do that. You made a list of questions instead. If setting up a simulation is so easy, why did Solidworks fail miserably at getting a realistic result in their own software? If you want to have a technical discussion about this video, show some respect. Telling a mechanical engineer that they have no clue about the underlying principles of FEA, and discounting the simulation validity without knowing ANY of the details, is absurd. It suggests that you would rather make false assumptions to judge yourself superior, than to ask questions and have a productive discussion like a real engineer.
@Antonio Alessandro Deleo You won't be "showing off" your credentials? Have you really forgotten what you wrote 45 minutes ago? "However, looking it under the point of view of a structural engineer, many things are wrong" (and then you listed zero things that were wrong) I am perfectly capable of having a technical discussion, but why spend my time trying to correct all the false assumptions you made, after you show up hurling insults? I hope you have better manners in real life. Now, you move on to insulting the software I used, and continuing to claim my results are "close to garbage" with absolutely no knowledge of the simulation setup. Classy.
@Antonio Alessandro Deleo Go ahead and read the updated video description, so you have a basic understanding of what this video was and was not intended to be. Then, if you can muster an ounce of humility and respect, climb down off your high-horse, and have a civil discussion, I can speak to any and all the questions you raised. I hope you do understand that you not knowing which way I did things, is not even close to the same, as me doing them wrong.
Nice catch of the poor boundary conditions! It's not easy but it doesn't look like they put much thought into their example.
very cool. cool method. I wonder if you used less fill or less stiff material if you could visually see the stress as the individual triangles deform. maybe even just a thinner plate?
when you set up the FEA at the end you are using a fixed bearing support at the rocker pivot? the "seatstay" link is just a two force fixed member? i imagine the shock link is a two force member with a load along the link axis? is this a section view of half of the assembly? or are you not worried about asymmetrical loading by the links only being on one side? i' curious about the proper FEA setup. thanks
Hey John, this screenshot will help make more sense of my setup. I may do a more in depth video about the choices I made in setting up this simulation:
i.imgur.com/X4eTNWX.png
All the swingarm, seatstay, and "shock" components are defined to be rigid bodies (with frictionless pin connections and fixed hinge fixtures, they end up as 2 force members), leaving only the rockers and front spacers as deformable. Everything is positioned as it would be in bottom-out position, with a 500 lbf vertical load applied at the axle.
I updated the description to clear up a few misunderstandings. Check it out!
love how you spend so much time bashing solidworks lol
I think the only thing I'd do differently is print with the lattice closed in so there is a continuous web. In the test it looked like more of a lattice failure which the fea wouldn't capture and I assume wouldn't be representative of the final part either.
I'm not sure what you mean by "closed in". Do you mean solid top and bottom layers?
There are definitely a number of ways I could make the print more closely mirror the FEA. This was a spur of the moment decision to break a mock up print.
How the failure propagated through the lattice is nothing like a metal part would fail. But, the areas of highest tensile and compressive stress matched the FEA very well and the failures began at these points.
@@APEX.86 yeah, solid top and bottom. I'm just coming from the angle of trying to get both the FEA and test piece to represent the performance of the final part (not necessarily each other, which I assume would be hard for a printed plastic part anyway). It's just maybe the next guy might not be aware of how to properly interpret/ignore the similarities and differences in results.
I am curious if the print failed more in tension on the vertical side or shear on the horizontal side at that high stress bore because that's the main difference I would expect to see. As you say, for just finding high stress areas that's propably enough agreement. It would be more important if trying to use the result numbers to design closer to the mark.
@@thehackerman1234 I actually printed new links fully solid and tried to break them. I ran out of courage at like 260 psi in the shock. It's not worth ruining my frame! I think the solid links could actually hold up to some light trail riding, which is wild.
Are you using normal solid elements to mesh that lattice print? If so, that simulation is not representative at all, and using it in a "FEA vs real world comparison" is just misleading.
What do you mean by "normal solid elements"? That question isn't very clear.
These printed parts were meant for fitment checking, not as a direct comparison to the FEA of a metal part. There are a number of differences: the printed part is not a linear-elastic material, it is not a homogeneous piece of plastic because of the lower density infill, etc.
But, if you know where to stop the comparison, it doesn't really matter. The location of areas of highest stress don't care what the material is. The sparse infill has the capacity to alter this to some degree, but in this case it didn't. Where will a plastic part yield first? The area(s) of highest stress... just like a metal. The way the failure propagated through the printed part is not comparable to a metal at all. But, we saw compressive buckling right at the two highest areas of compressive stress in the FEA result; and we saw necking in the areas of highest tension.
This is not a scientifically rigorous comparison between FEA simulation and real world destructive testing. This was a spur of the moment opportunity to break something in a fun way and see how it compared.
@Antonio Alessandro Deleo I'll bite... Name ONE thing wrong with my FEA setup.
@Antonio Alessandro Deleo Don't take it "the wrong way"? Your initial comment is incredibly insulting. You have no idea about any of the details of my FEA setup, but decide to claim I have done everything wrong, and that I have no understanding of the underlying concepts, and that my results are "garbage".
This speaks volumes about you, that you would make these bold claims with absolutely no information to back them up.
I challenged you to point out ONE thing wrong with my simulation, and you failed to do that. You made a list of questions instead.
If setting up a simulation is so easy, why did Solidworks fail miserably at getting a realistic result in their own software?
If you want to have a technical discussion about this video, show some respect. Telling a mechanical engineer that they have no clue about the underlying principles of FEA, and discounting the simulation validity without knowing ANY of the details, is absurd. It suggests that you would rather make false assumptions to judge yourself superior, than to ask questions and have a productive discussion like a real engineer.
@Antonio Alessandro Deleo You won't be "showing off" your credentials? Have you really forgotten what you wrote 45 minutes ago?
"However, looking it under the point of view of a structural engineer, many things are wrong" (and then you listed zero things that were wrong)
I am perfectly capable of having a technical discussion, but why spend my time trying to correct all the false assumptions you made, after you show up hurling insults? I hope you have better manners in real life. Now, you move on to insulting the software I used, and continuing to claim my results are "close to garbage" with absolutely no knowledge of the simulation setup. Classy.
@Antonio Alessandro Deleo Go ahead and read the updated video description, so you have a basic understanding of what this video was and was not intended to be.
Then, if you can muster an ounce of humility and respect, climb down off your high-horse, and have a civil discussion, I can speak to any and all the questions you raised. I hope you do understand that you not knowing which way I did things, is not even close to the same, as me doing them wrong.
What material did you print this out of?
PolyMax PLA.
It's a great material that bridges the gap between PLA and ABS.
@@APEX.86 I’ve been starting to print in NylonX (Carbon Fiber filled nylon) interested to see how it performs comparatively
Fantastic display
Stress animation: th-cam.com/video/67qBW9ENQho/w-d-xo.html