G2 surface problem (ISDX)

[Jacoolo]

is this ordinary situation on job market in USA - pro\e in demand, or is it just nowadays?

btw. Bart, question according to post subject - is it possible to solve beone problem with curvature transitions in Pro\e?

 

[design-engine]

want me to take a stab at it? ISDX curves can get to the g2 or curvature continuity with that example. will do that after my class today. maybe upload something by midnight... I have a project tonight that takes priorty.

 

[Beone]

design-engine that would be great! I will be waiting like child for Christmas

 

[design-engine]

I did it g1 today

 

[Beone]

So you are not able to make it G2?

 

[jeff4136]

I wonder, for the sake of a more general discussion, if
this is a good example of a misunderstanding of the purpose
of G2 blending and misapplication of G2 boundary
constraints and Blend functions?


Consider:

 

[carbonmike]

I looked on those models, they look good, but you have visible edges on some places, so surface is not smooth. But its not so bad.


Edited by: carbonmike

 

[jeff4136]

Yes, without a doubt those could both be sweetened

a bit.


How would you improve on them? Maybe they're a good
starting point for G2 blends.

 

[Jacoolo]

Jeff, great job, I took a look on your part, this is intresting attempt to make things different, without boundary blend.

However, everythings looks good until they are mergen, not smooth transitions show up

In the end, really nice tutorials about - how to use sweep tool

Since I am not an expert, I consider this hard to be done with g2 in Pro\E

 

[jeff4136]

> Since I am not an expert, ...


Nor I. I make this stuff up as I go. I wake up every day in a
brand new world and every model is a learning experience.


> ... I consider this hard to
> be done with g2 in Pro\E


I don't think it (G2 + whatever other practical and esthetic
criteria we want to burden it with) will be very easy to get
with any system. It's a simple model that presents some subtle
but fairly complex conceptual problems.


> attempt to make things different,
> without boundary blend


I can't say it's the way it 'should' be done but it's how I'd go
about it until I see something to make me think otherwise. While
the sweeps are where I chose to devote some time experimenting
I'm still looking forward to seeing some Blended examples.


> ... good until they are merged,
> not smooth transitions show up


Despite having chewed most of the flavor out of it there's more
to learn from it, so ...






2007-12-13_000314_last_one-wf2.prt.zip



As with any complex task I'll cast about for useful functions
and technique, then spend a lot of time fussing with surface
contours and trim boundaries to get an acceptable set of
boundaries for subsequent surface definitions, which I contend
is an order of magnitude more difficult when G2 boundary
constraints are to be defined.


I think this discussion is too fuzzy to be useful unless we
qualify and quantify "smooth". I understand the theory and
desire to not see anythingindicating any deviation from
perfection. However; tolerances apply to everything. They, and
compromises, apply to virtual model creation as well as tangible
article manufacturing. Where push button solutions to perfection,
such as those available for prismatic article modeling, are
absent; instead of simply looking at a shaded model, seeing some
indication of a discontinuity and saying it isn't 'smooth'
analysis and evaluation are required to determine suitability
for the given application.


Getting back to G2 continuity; the attached model has some
evaluation included to indicate deviations from more ideal blend
curves. The one or two coats of sanded primer deviations are
typical for 2:1 curvature mismatch.


> In the end, really nice tutorials
> about - how to use sweep tool


Well, thank you. Happy to share the observations and thoughts.
(If you want to try something interesting with VSS; map the
curvature on a relatively simple surface edge then drive a
curvature dimensioned spline with an Evalgraph feature. It can
be parametrically linked using an Evaluate feature.)

 

[Jacoolo]

Sometimes I regret there is still something to learn about Pro\e. Every
time I reach this next level of Pro\e skills, there is next one just
ahead. When it`s gonna stop?



Sometimes I regret I am so curious about all those things. I have no time for other activity))))



ok, jeff if You can - upload

 

[Leonello]

Hi to everybody!


here below I've attached my model.


I used ISDX module, and the complexity of the model is really low.


Bye.


Leonello


2007-12-13_113741_3tubes_connected.prt.rar






Edited by: Leonello

 

[design-engine]

you made you model quite a bit more simple that the original example. Go back and look at the original.

BTW... I speak at the Friday Dec 14th (tomorrow) Pro/ENGINEER user group or BUG meeting
at 11:00 at the County Springs Hotel and waterpark in Waukesha WI if anyone cares. http://countryspringshotel-px.rtrk.com/

My
talk will suggest alternate modeling techniques for Engineers working
with Industrial designers. With WF4.0 there are even more incredible
ways to model forms.


Edited by: design-engine

 

[Leonello]

Hi!


I did it again, using ISDX, starting from the original example. The model seems to be more simple than the original one, one more time.


Bye.


2007-12-14_082058_3tubes_connected.prt.rar



Edited by: Leonello

 

[jeff4136]

A couple of things you might look at and decide if you think
they're relevant ...

 

[Beone]

This is my final version with ISDX... there are some unwanted ,,reflections,, but thats enough

 

[design-engine]

that looks good to me. I am working on a project all night.... maybe I tak a brake and upload my version.

 

[Fegenbush]

I just came here to say that we seem to be spending a lot of time tweaking the top tube / down tube / head tube connection. I would rather see the area surrounding the bottom bracket, as that is where I have all my troubles when trying to model new bike frames.


If the frame is to be made of carbon fiber, the minor surface issues of even the first model will not be noticable in the end product. I only know from actually making the pieces myself. Also, these things can be smoothed and polished in the mold. If the mold is to reflect the finished part with no additional sanding, it will have to be polished to mirror finish anyway.


Just my $0.02.


Fegenbush

 

[jeff4136]

> would rather see the area surrounding the bottom bracket


Post some representative geometry? (WF2 readable, native or
neutral, if you wanna include me in the discussion.)


> If the frame is to be made of carbon fiber, ...
> ... have to be polished to mirror finish anyway.
> Just my $0.02.


Though I suspect Beone's primary interest was trying to get a
handle on ISDX functionality, two cents worth from someone
that's gone from model to finished part is always worth more
than face value. The problem is a good example of the Ninety -
Ten Rule of Perfection (achieving the last ten percent will take
ninety percent of the total effort).


Being, myself, interested in the quest for Smooth I'm sorta
disappointed to see this discussion die with so few proposed
solutions and no ISDX representation. I think the example
problem is ideal for inclusion in discussion and demonstration
of advanced surfacing techniques and Class A surfacing
fundamentals. Having theories regarding (practical?) functions
and techniques I tried chipping away some of the last ten
percent ...





2008-01-04_050746_doin_it_the_long_way__wf2__.prt.zip


... though at considerable investment of time and effort. Far
too much for something that could live with conic rounds, but
it's the concepts that I'm grappling with here. ;^)


Things that I think are noteworthy:

 

[Jacoolo]

jeff you do amazing stuff around here,

your deep interest of bite a thing to the end still impress me. Ohnest. Usually I am lack of strenght to polish things to obtian 100% of what I i`ve planned. Instead I took the result which lay close(90 - 95%) to desired.

I think Beone obtained good result with ISDX regarding G1 transitions.

I would like to see your post much often on my topic regardning sufraces - "Some surface behaviour make me confuse" )))

best regards

Jacek