closed lofts not symetrical

Hi there,

Has anyone else noticed how if you make a surface-loft between to closed
sections which are symetrical about X and Y, for example a rounded
'ob-round', and use four guide curves, which are symetrical in pairs (one
pair in the X plane and one pair in the Y plane) between the two set
maintain tangency and advanced smoothing, this surface is not necessarily
symetrical and tangent about X and Y. I have the base of a blow moulded
bottle which I have constructed in such a way, then I've had to half the
bottle and mirror the halves to avoid doubling my feature tree in some
areas, and then the base is no longer tangent.

I guess a picture would help here, as I'm not being very articulate, but
maybe someone else has seen this and has a comment.

Lee

 

Lee,

I cant quite figure out what you mean, but I bet that if you study these
tutorials it would help understand whats going on...

http://www.dimontegroup.com/tutorials.html

Good Luck

Todd

 

Thanks Todd,

I have studied those tutorials already - and bloody good they are too! If I
get time I'll sort out an example of what I mean.

Lee

 

Sorry I couldnt help you more!!

 

I've seen that. I don't remember the fix, it was probably to do
half of the loft with tangency and mirror it. If you can, you
should send this example in.

Just some questions, were you using any end tangency controls?
Any guide curves or centerlines? What types of sections? (2D/3D,
faces, edges, curves, etc)

matt

 

No, that's probably not the solution. Ed points out in one of the tutorials
that lofts don't maintain tangency on the supposed centerline and explains
why. I know from my own work that mirroring the part can make a break at the
centerline. (On my parts the lack of tangency is small enough that no one
notices.)

Jerry Steiger
Tripod Data Systems

 

I can't explain this, but I can think of a couple of things you might try.
My first suggestion would be to give up on the mirroring and live with the
longer feature tree. You could also try NOT using maintain tangency and/or
advanced smoothing. Their behavior is not always very intuitive.

Jerry Steiger
Tripod Data Systems

 

I, perhaps with less venom, agreee that aesthetics are important. When
challenged on that statement by the technically minded, I ask 'why is your
wife pretty, then? If function is more important than aesthetics, any old
woman would do'. This of course puts the hapless engineer in the bind of
admitting that aesthetics are important in their lives or that their wife is
a 'dog', both of which cause red faces and terrific levels of discomfort.

 

Comment:
Lofts are terribly demanding things. We provide lofts a couple of profiles,
a couple of guide curves, and SW fills in a massive volume based on these
minimal inputs. If we are a little shy on even one of the components of the
loft, it all goes to pot.
With every release, and sometimes with service packs, the rules on lofting
change. I can not tell you the absolute sourceof your problem because I
have too little information. But, yes, I have seen your issue with closed
lofts.

Send me a copy at ed'ignorethebitbetweenthequotes' and
I might be able to root it out. If you want general tips, here they are:
1)guide curves are 'almost' always a mistake. We do a fair amount of
last-minute-fix-it work on other peoples' files, and the majority of the
problems are solved by eliminating the guide curves and creating the loft
'naturally' (which might even take less set up than a guide curve loft). I
have it on good authority that the devlopers are aware of the terrible
issues relating to guides - and they may change things - but its going to be
a while.
2)Guide curve order matters. The first guide curve drives the
parameterization of the loft faces. If that guide curve is a little hinky,
the loft is f***ed no matte what you do.
3)No matter what you do, guide curves take precedence over your loft
sections. That should keep you up at night. If the loft feels the need, it
will follow your guide curves and completely miss your sectrons.
4)I dont' even want to get into start tangency conditions, epecially when
relating to closed lofts (strangely enough, there is no option to match the
start tangency to the end tangency on a closed loft, in my expereince)

There are other issues. Again, I can't get into it all here.... please send
me a sample. If requried, I will sign and NDA



,

 

oooohh yeeeaaahhh

baby.