problem with surface quality of G2 patch command

[cekuhnen]

So before startig over in Fusion I wanted to test and compare it with Alias.

Through out this process I noticed a slight surface distortion among the Fusion created surface.

The loft tool with Rails bugs and in addition Rails cannot have G1 or G2 options so Patch is the only option here for me.

Left Alias Square tool (G2 setting), middle Fusion Patch G2, right Fusion Patch G1

The comb shows pretty severe problems like already noticable with the chaos inside the zerba lines.

When I brought everything via STEP into Alias to check there I noticed that the patch tool is just a square surface folded over the corner and not a simplified and clean sweep.

adding continuity checks showed that the alias data transfered back into from Fusion remaind as slim and good as in Alias - everything is still G2

But the Fusion Patch G2 and G1 surface are only positional and not even G1.

Knowing Alias the default square surface option would produce dence surfaces the continuity checks would fail without enough density (extra spans). So thats why you can use the coincident option to aling all cvs well - granted all surfaces have the same degree.

so this concerns me a little that in Fusion with the G2 there is so much noise inside the zebra and curvature comb even while this detail might be minuscle to the human eye.

[jakefowler]

Hi Claas,

I'm afraid the Patch tool is not going to offer equivalent functionality to Square in Alias. It's creating an n-sided boundary surface (taking an arbitrary set of connected curves and fitting a trimmed four-sided surface to these). It's designed to be super-flexible and tolerant to imperfect inputs, but in doing so it sacrifices some surface control and doesn't optimize for clean four-sided inputs such as this.

For creating something like a Square, Loft should have been the right tool to use. We'll certainly look into why that's failing in this example. You should be able to set adjacent curves as profiles & rails, and set boundary conditions at all of these (assuming the surrounding faces 'agree' with the boundary condition).

We'll also look into the why Alias flagged these as non-continuous. These should be meeting the specified continuity within a tolerance. It's possible that this affected by the STEP translation process. We'll look into whether it matched the expected tolerance and whether this is affected at all by translation.

Many thanks for reporting this one, and let us know if there were any further questions.

Much appreciated,

Jake

[xuejun.sheng]

Hi Jake,

You're very right in that the patch tool is designed to handle general n-sided case. In ASM loft offers a similar functionality as Square in Alias but may not offer a comparable quality in terms of control polygon structure and curvature flow. However, there is an option inside the patch tool that is specially designed for 3 and 4-sided patch which may come very close to the quality of Sqaure in Alias. Currently this option is not turned on by Fusion. Can you create a journal file for this case or just give me the smt file?

Thanks,

Xuejun

[jakefowler]

Will do - thanks for looking into this Xuejun! Jake

[cekuhnen]

It would be great to get access to those

while I like the simplified interface of Fusion sometimes I feel really crucial professional tool options are missing making it look to me less powerful.

[cekuhnen]

I am also of course aware of that the STEP data from Fusion in Alias would produce instant cardiac arrest to any Alias cad designer. The same would be with SW or such.

So I do not expect the same quality to come out. However it would be nice.
Like with the square tool you can get in Alias a crap result or a super clean result ICEM will love.

Problem for me is as boring and simple this iPhone shell is in Alias it is a breeze to build and Fusion gives me lots of problems because I need to work around the tool limitations or there only intent.