fillet issue  1-20  21-40  41-56

Next
 From:  Gibs
3105.1 
hi all !

First of all, I'd like to thanks for that great peace of software wich is MOI, really it's intuitive and easy to use. However, I'm actually experiencing some issues with the fillet function. I'm a student and I'm trying to fillet some edges which dont' want to be filleted apparently... It's working when I extrude some curves and stuff but it fails when I'm trying to connect several surfaces I created using different surfacing technics (mostly sweep and planar).

All my surfaces are joined together but edges dont' wanna get filleted :P
any idea ?

Hope someone can help me !
Greetings from Switzerland by the way ! :D

Anthony
www.gibsgraphics.ch

Here are some attachements :











  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.2 In reply to 3105.1 
Hi Anthony, welcome to MoI!

Fillet is one of the most sensitive (and prone to failure) operations that MoI tries to do.

It involves several complex operations in sequence, like calculating surface offsets, and extensions, and then intersections between those pieces. If there are any difficulties in any one of those stages it can prevent the fillet from being formed.

So for filleting it tends to help to have a very clean and regular shaped input into it. Any irregularities in your base objects tend to get magnified and cause problems when trying to fillet them.

It looks like in your case here, the main problem area is this spot:





If you look closely, you can see that the curves are not smooth there, they actually come to a corner point but it is a shallow one, the curve tangents have a difference of 15 degrees between them there.

That kind of situation where things look pretty close to being smooth but actually are not smooth and have a shallow crease between them will tend to cause problems with filleting, it is difficult for the intersection part to work properly in situations like that.

You want those kinds of junctures to either be actually smooth to one another or to have more of a pronounced corner than 15 degrees.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.3 In reply to 3105.1 
Hi Anthony, also another problem is that one of the surfaces is not of good quality, it has some bunching and bad forming in it:






Trying to build a surface where the generator curves change a lot and compress together or go to a sharp point like that is not good - that will make for problems in the surface shape, like possibly small folds near that area and just not a clean shape to it.

Shaping problems like that will also tend to cause difficulty in filleting.

For something like that, instead of surfacing directly to the outline you usually need to build a larger more simple sheet and then trim it with surrounding pieces instead of trying to "patch in" to a non-regular boundary like that.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.4 In reply to 3105.1 
Hi Anthony, here's a kind of quick demo on how you can make a better shaped surface for things like that one curved piece.

When you have outline curves that are not very regular, like one portion is way longer than the other side or stuff like that, it doesn't work so great to try and build a surface directly to those curves. I mean you can do it, but the surface will tend be chaotic in the irregular areas, often times there will be things like small folds back over itself and things like that. You may be able to get away with that for some things, but not if you want to fillet them.

So instead of trying to compress in a single surface directly to the boundary, usually a better way is to build a larger and more simply shaped surface and then trim it.

So in your case that would look something like this:








Techniques like this (build larger than trim) tend to be a way to build a higher quality more evenly shaped model rather than going around and trying to fill in things "patch by patch". You can use the "patch by patch" method where things have a nice even topology but not when pieces are changing rapidly or compressing together or stuff like that.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Gibs
3105.5 In reply to 3105.4 
Hi Michael !
It's a pleasure to have feedback from the creator of MOI itself ! Thanks a lot for taking some of your precious time answering my question, I appreciate !
Your screen shoot are very interesting though, didn't know surface could be a bit messy around those kind of curved surfaces eventhough it looks clean in the viewport.
I'm now trying your technic, hope it's gonna work :D If it does, I'll provide a nice CG render of my swatch once it's done for the MOI user gallery ;)
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.6 In reply to 3105.5 
Hi gibs,

> eventhough it looks clean in the viewport.

Yeah, it can be deceptive.

It's possible for certain kinds of problems to be localized into a really pretty small area at the end of the surface which can make it difficult to see. But Filleting will still be sensitive to such things.

Here's a kind of more gradual zoom in to one juncture you have there, you can kind of see here that you've got a swoop or cusp in there but at a pretty small level. Notice how it looks like it would be smooth from a further out zoom level but it actually is not:




So one of the ways NURBS surfaces work is that a NURBS surface is a kind of quadrilateral "net" of curves, with one set of curves running in the "U" direction and another set running in the "V" direction.

The normal of a surface is determined by the perpendicular taken from both of those directions.

When you have a corner of a surface where the corner is "degenerate" (both U and V directions going in the same direction at that point) it can cause an ambiguity in how the normal is defined there.

So when you want to build a surface with something like Sweep or Network, you need to give a kind of semi-quadrilateral layout with more distinct corners.

So for example this is a clean distinct corner:




On the other hand, this kind of corner is not good for the underlying NURBS surface:



That's because the U and V curves in the corner do not have distinct directions, they have parallel tangents there.


Another kind of problem area is when the U and V directions are parallel but pointing exactly opposite, which is like this:




Now, none of this applies to the arrangement of "trim curves" on the surface, only to the structure of the underlying surface.

In your case, you do have one of these kinds of corners in this area:



So that's kind of another general problem area with that kind of shape.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.7 In reply to 3105.5 
Hi Anthony, here's an even more simple method for the "build extended" type strategy.

One thing that can be good is to try and see if your shape is an overall cutout of a larger more regular shaped piece.

If I understand correctly, you do have something like that here, and this is the larger piece:



So take this edge:



And fillet it, so you get the source of the larged curved portion here:




That's the "underlying form" right there.

Now take 2 curves drawn in the Top view plane like this:



And run Boolean Difference on the main piece and select the curves as the cutting objects. That will slice it up like this:



And now you can delete the pieces you don't want, leaving you with:




So that's even more of the kind of "build larger and then trim out pieces" kind of strategy.

It can just generally be a good way to go if your object has the overall impression that it has some underlying form that has been cut up, try to actually model it in that same way...

This tends to require you to kind of imagine what the missing pieces look like, sometimes it can be hard to get used to doing that but it can be a good technique.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Gibs
3105.8 In reply to 3105.6 
yeah, it's kind of deceptive :( I'm trying everything and nothing can't really get done the way I'd like it to be... I can't imagine I've got something impossible after having seen some object design or car industry modellisation with MOI on the internet. It's crazy !

I think you made it clear about filleting being sensitive to curves angle (thanks for the little animation ;) )
That's why I've tried another way of generating the surface, I mean another structure. (as seen in the screenshoot)
However, in screenshoot no.5 , I zoomed on the little area (the same you zoomed in) and well, there is something weird, I don't know why ? (it's a sweep opération between two edges and two rails)

Is there anyway of filleting the yellow edges ? It's killin'me :)





















EDIT:

sorry, i haven't seen the last post you made :D I'll check this technic ! Thanks a lot for taking some of your time to help me out ;) By the way, How do you know my name ?! :P

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.9 In reply to 3105.8 
> By the way, How do you know my name ?! :P

You signed it at the bottom on your first post in this thread... :)

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.10 In reply to 3105.8 
Hi Anthony, so your new piece has a different problem in it where it folds over itself in this area:



That could be due to a problem in sweep that has been solved in MoI v2 where sweep was sensitive to the control point spacing of the rails and could cause the sweep to kind of "bunch up" if control points had different kinds of spacing between them on the different rails.

There have been a lot of things improved in MoI since v1 actually - the current trial version is actually about 1 and a half years old now, v2 is almost ready to be released with a lot of improvements in it.

One other note - you have a lot of curves in your model on the same spots as edges, that can make it difficult to select edges so you probably want to hide those curves after you are done using them to get them out of the way. If you select a curve and then try to run fillet, it won't work, you need to select the actual solid's edges and not the original generator curves if that makes any sense.

There is also an alternate filleting mechanism that can sometimes help for filleting difficult cases - that's the surface/surface fillet which is activated if you run Fillet when 2 individual separated surfaces are selected. Sometimes doing fillets at that individual surface to surface level can be a way to fillet a model that is not behaving well with the edge-based filleter.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Gibs
3105.11 In reply to 3105.10 
all right, sounds great though !
Well is a MOI v.2 going to be released in a demo version aswell ? To try it out before buying it ?
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.12 In reply to 3105.11 
Yup, there will be a demo version of v2 available just after the release is finalized.

It's getting pretty close... There is one more v2 beta release that I'm working on, then after that I will be updating the documentation and then it will be ready.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.13 In reply to 3105.11 
There's a collection of all the v2 beta release notes available here:
http://kyticka.webzdarma.cz/3d/moi/doc/V2releasenotes.html

That's basically a list of all the new stuff that is coming for v2.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Gibs
3105.14 In reply to 3105.12 
cool !
well hope it'll be ready before christmass cause I'll finish up my internship around the end of December ^^
(doesn't mean I won't use MOI after that deadline :P I do think there are some pretty great opportunity on the market mastering NURBS modelling through MOI, as I'd like to become a Designer. Yeah, designer often got good eyes for colors and stuff or concept, but they rarely know how to modell or represent correctly their though. That's the part I'll enter the game hehe... I think I have to keep it up and don't give it up despite those little... technical issues :D )

Thanks again for thos explanations ;)
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Gibs
3105.15 In reply to 3105.14 
wow, I've checked the link you provided me... awsome improvement !! I haven't read all down yet but there are some pretty cool new things like the Scene Browser helping organising the scene. Good point here ;)

Anyway, I'll try the technic you spoke earlier and I'll tell you if it works.
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.16 In reply to 3105.15 
I think you may actually have a case here where the fillet mechanism does not know how to handle this corner juncture with multiple fillets running into there.



It can do any one of those, like this:








But I don't think it knows how to handle all of them being filleted at once.

In fact, it's not very clear to me how that would actually work, I've combined those pieces together and it's something like this:






I guess it would ideally put a kind of rounded rail on the top or something like that... I'm pretty sure it does not understand this kind of corner condition, so this piece is probably just not going to get automatically filleted by MoI's fillet mechanism.

You probably will need a more "heavyweight" CAD program to do that one, something like SolidWorks or Pro/E maybe...

I think it looks like you would have to do some manual trimming and managing of the individual fillet pieces to make that happen in MoI, that's how I assembled the piece in the screenshot above. But then you've also got to figure out exactly how that corner should actually behave.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Gibs
3105.17 In reply to 3105.16 
yeah i see what you mean, well I tried your technic and it's okey I've got the volume but can't really fillet edges the way I would like. It's a pitty, there's apparently no other turn around. That's kind of sad :(

However, why wouldn't it work to apply a fillet function on the edges I've selected ? (as seen on the screenshoot) Ho yes, I first filleted the problematic edge so that it's more "smooth". By now it doesn't seem to work but is there something I'm doing wrong ?

And yeah, I'm afraid to switch to those kind of heavy CAD package... those software are more likely for engineers and well, I'm more "design oriented" artist though... I don't know if I could learn as easily and as fast as I learnt MOI. Moreover, prizing isn't obviously the same huhu :)

But good friend of mine who's studying architecture in University in Switzerland spoke to me about Rhino or something. He told me it was great for modellisation. In your opinion, is it more likely to work with this software for that kind of fillet task ?
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.18 In reply to 3105.16 
So I guess the corner juncture area is supposed to look like this:



MoI's fillet mechanism does not know how to do that though.

The one here I did by importing the MoI model into ViaCAD which is a different CAD program that has a more robust filleting mechanism than MoI. It can come in handy to use alongside MoI for fillet cases where the geometry engine that MoI uses is unable to figure it out.

- Michael

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
3105.19 In reply to 3105.17 
> In your opinion, is it more likely to work with this software
> for that kind of fillet task ?

No, not really - filleting is not a particularly strong area of Rhino either.

I tried Rhino on this case and it's slightly better in that it doesn't make as much of a mess as MoI (in MoI v2 the filleter will return partial results if it was not able to calculate the full result, unlike MoI v1 where it will just do nothing) but just leaves an open hole in that juncture area.

ViaCAD is much better to use for more difficult filleting than either MoI or Rhino, it's actually very inexpensive ($99).

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Gibs
3105.20 In reply to 3105.18 
ho okey so there is some fillet improvment to do then hehe ;)

it's weird because I managed to get this kind of "triple" edges fillet in the bottom of the piece, but can't manage to get it done in the same way for the top. But that's indeed like you say maybe a limitation of Fillet mechanism in MOI (by the way, filleting looks nice and easy but i can imagine how complex it really is underneath to develop algorithmes that works with such kind of surface)

It might be something really cool to improve for users in v.3 :D

the screenshoot of the bottom part:




Attachments:

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged
 

Reply to All Reply to All

 

 
Show messages:  1-20  21-40  41-56