Full Version: Root Fillet Example Model

Show messages:  1-20  21-27

Hi Anthony, sorry I just don't have any ideas right now on why netgen is having problems with some of those MoI exported files.

So one thing you have mentioned though is a problem with the projecting in Rhino. It might be good if I could see what is going on there. I guess probably the pieces to repeat the projection are in your file, but there are 284 objects in there and it's difficult for me to figure out what's going on. If you could make a simplified version that just has objects in it pertinent to one single projection that is getting messed up that may help yield a little more information on that part of things.

- Michael

---

no problem michael, that's understandable. i don't know what the netgen errors/warnings mean either. it doesn't seem to stop it from meshing all of the time. but it does seem to make it hang on the surface optimization. that's about all i know there.

i tried to strip out what you don't need from the file. there are a lot of steps in making the model, so it's hard to take out everything. but what seems to be the root of the issue, in this particular case, is the projection step. after the projection you have to rotate the airfoils to set the sweep. but don't worry about all that. i saved the model for the projection at the tip. you have to set the cplane each time you do a projection onto a cylinder. the dark blue airfoils are before the projection. the cyan airfoils are after the projection. the red and green airfoils have the sweep set. so all those cyan airfoils that rhino creates seem to be the issue. i have to go and edit them all after the fact, then the fillet issues go away. if you look at the makeup of the red or blue airfoils, you will see they are two ellipses and two curves joined together. everything is fine if you use those. things go south if you use the cyan or green airfoils. the green airfoils are the final step with everything in the right position to do a loft. unfortunately rhino screws those up. so i finally figured out if i edit them, things work as expected. there are different layers to the file. i don't know if those are importing for you. that makes the file a lot clearer. if you aren't getting the layers, let me know, and i'll strip the file out more.

but the main step that seems to be the root of all the problems is going from the blue to cyan airfoils. the cylindrical surfaces are what you need to project onto. i believe the screencast shows the process. i should mention that i rotate those cylindrical surfaces to get the seams out of the way. if the seams enter into to any step then all hell breaks loose and i don't know why. i just figured it out by the school of hard knocks (i.e. rhino kicking me in the nards every change it gets).

---

hi michael,

i stripped the file down all the way to the one operation i'm curious how moi handles. when rhino 5 does the projection onto the cylinder, the resulting closed curve is all messed up. i have to go and edit it after the fact. i'm wondering if moi keeps the original curve structure after the projection. i set the cplane to the appropriate one. all you have to do is project the dark blue closed curve onto the cylinder. the cyan curve is what rhino 5 generates, that's no good.

thanks,

anthony

ps; i deleted all the previous files i posted, to save server space

---

Hi Anthony, I've attached the result of curve projection in MoI, the segmented structure of the curve is preserved in MoI.

- Michael

Attachments:
Anthony_Projected.3dm

---

wow that's fantastic. i clearly made the wrong choice by sticking with rhino. i should have switched to moi when i evaluated it so many years ago now. i was under the impression that both programs were using the same kernel. but you have better filleting and projections. probably a lot more things too. thanks for checking into this. i will definitely be purchasing moi v4 instead of rhino v6.

anthony

---

Hi anthony no they use different kernels, the part that is the same is they both use the .3dm file format as the main "native" format. But that's just for data storage not for any modeling calculations.

It is pretty useful to have both available, they each have different strengths and weaknesses. But yes many Rhino operations can tend to fuse segments together though which can be problematic. I've tried to avoid that with MoI.

With them both using the same storage format you can move objects back and forth between them easily.

- Michael

---

thanks. i think when rhino first started out it may have been using the kernel you are. but then they started with their own kernel. i don't really recall. but whatever the case, moi seems much better at the basics. rhino does have a few features that i thought would be useful. but i'm willing to give those up to stop having modelling problems every single time i use the program. with moi v4 you are adding most of the missing things i was looking for. i think the only other thing is the zebra plots. it would be nice to have the cg location and inertia matrix. but i can live without it. the main thing is being able to make a model without having to do 20 attempts to get something usable. even then, it seems moi on the first attempt is still better than the rhino result.

---

Show messages:  1-20  21-27