Nose Cone Help

[design-engine]

VSS is one of the most powerfull tools inside Pro/ENGINEER becasue you can tie graphs to it. Its one of the reasons solidworks is weak.Each year SW knocks one more off the Pro/E Shelf and I suspect that in 2009 release there will be graph functionality somehow with a simular VSS tool.Most soldiworks users barly tap into their software as it is. I say keep them in the dark... did notmean to hyjack and talk about sw.

 

[jeff4136]

can you please upload the WF2 part?

It (the simple version I did) is really quite simple once a
few fundamentals are understood (or maybe I should say; sorta
understood, because I don't fully comprehend the reasons for
the relationships). I'll give the (disdain? dorks?) used car
salesman a few more days to try to figure it out, post a model
and redeem himself. If he doesn't I'll post the model and a
'walk me thru it' for Student Ed users if there's a request.
Maybe I'll call it the Nose Job Method.

 

[Arbiter]

Please post a walk me thru it Jeff , I've been working on it and still haven't figured it out despite all of the wonderful suggestions! I know, I am a total noob , but it would help me out a lot! Happy New Year! :-D

-Chris

 

[jeff4136]

We're all noobs here in a sense and in some sense all looking for items with
resale value. Unfortunately there are also a few that spend a disproportionate
amount of time trying to sell it back to us, sell us some software or other,
sell us on how bad some software or other is, ... with a resultant dilution of
valuable content and general devaluation of the forum.


Some of the suggestions were, I'm sure, well meant. Some just leave me
wondering. Which were "wonderful" in your mind and why?


(I normally think it rude to ignore a direct question but sometimes it's
best to just get on with objective discussion, eh?)


I just found these links* looking for clarification of a definition,
trying to figure out if I've been misusing the term 'paraboloid'.
I'm marginally amazed that I haven't seen similar references in CAD forums
for years. Surely it's common knowledge in some CAD user circles and is the
foundation of the shape.
http://www.math.umn.edu/~rogness/quadrics/ellparab.shtml
and (referenced on above page)
http://www.math.umn.edu/~rogness/quadrics/paraboloid.html


The shape can be created with a Constant Normal Direction VSS sweeping a
parabolic curve (either a full symmetric about axis curve or half of it)
section along a parabola. Note that the section is always parallel to the
trajectory conic axis and that the trim plane does not have to be normal
to that axis but can be rotated as shown and still result in an elliptical
trim boundary.


I take it Student Edition can import / export IGES, STEP and Pro/E neutrals?


foundation surf
2009-01-02_165655_foundation_srf-igs.zip
trimmed surf and ellipse quadrants
2009-01-02_165733_trmd_srf-ellip_layout-igs.zip


If you're not familiar with conic arc definitions ...








The rest of it ...
2009-01-02_165909_prt0001-neu.zip
(You should be able to use layers to help pick thru the pieces.)
2009-01-02_170030_wf2-com_ver-native.prt.zip


Holler back when you want to chew on the puzzling pieces.


*(Might also be of or lead to articles of interest...
[url]http://www.math.hmc.edu/~gu/curves_and_surfaces/surfaces/par aboloid.html[/url]
http://mathworld.wolfram.com/EllipticParaboloid.html
http://en.wikipedia.org/wiki/Quadric_surface )

 

[dougr]

This is so simple, why 5 pages of this ????


Split nose into two halves and use a 2 curve Boundary blend:
<UL>
<LI>1st curve use existing surface edge - be sure to select surface edge and NOT datum curve.</LI>
<LI>2nd curve select the one at 90 deg and choose this boundary to be normal to the vertical plane.</LI>
<LI>Select 1st boundary to be tangent and second to be normal to vertical plane. Complete by mirroring and merging as per usual.</LI>[/list]


Not sure about toupees, toucans, two-sures,two wots. This just looks like a plain old cap to me..

 

[mcgowanp]

2009-01-03_063405_nose_cone_wf4.prt.zip


Hey all, I agree with doug.Chris asked for help on a specific problem, he wanted to patch a nose cone, whats a method for this with his existing geometry.


I have attached a model that I hope helps, You have already created your "fuselage" so the only features that are showing the method are features 28-31, they are right at the end anyway. I dont have student edition, do you have access at all to a comm version of WF. Jeff is there anyway way to convert this to a student edition?? If no on all counts I can do a quick video for you, I'm doing these at the moment anywayfor some training classes.


Jeff you ever thought about doing some training classes??


Paddy
Edited by: mcgowanp

 

[kdem]

You'll need to create an iges, step, or neu file as Jeff did for his parts. You won't get the history tree so some directions might be required.

 

[zpaolo]

2009-01-03_063405_nose_cone_wf4.prt.zip


Hey all, I agree with doug.Chris asked for help on a specific problem, he wanted to patch a nose cone, whats a method for this with his existing geometry.

I'm a beginner too as far as surface modeling is concerned, and so maybe I'm saying something very stupid, but I'm learning a lot from this posts and I have a doubt: what's the difference between your patching method if compared with a simple boundary blend using only the two curves (side profile and edge of the fuselage)? they both are curvature continuous but the shape is a little different... is one better than the other for some reason?

 

[zpaolo]

is one better than the other for some reason?

Ok, playing a little more with the model, I can see that in your nose cone patch I can add another boundary blend line, from the "intersected" profile, to better shape the profile of the nose. this sometimes works also with simple 2 side patches, but sometimes fails...

 

[jeff4136]

A 'plain old' (with singularities) cap can work well ...


2009-01-03_162330_bndry_blnd_cap-neu_1.zip



or fair ...


2009-01-03_162526_bndry_blnd_cap-neu_2.zip



or produce garbage, depending on acceptance criteria and input. I
suspect Chris may have problems with input (the bounding conditions
defined by existing surfaces or procedure) that will limit the quality
of a blend cap. Just a guess without being able to examine the geometry
and speaking from personal experience. Creating a complex blend from
arbitrary boundaries presents problems that are hard to comprehend let
alone explain. I think of it as a choral activity where everyone (the
bounding conditions; spans, tangent directions, maybe curvatures and
rates) has to be singing in perfect harmony or the results will be
disappointing unless the sole criteria is closing a b-rep hole, and
sometimes you can't even get them on the same page.
_ _ _


Paddy,


I believe the only way to go between native commercial and student
versions requires a translator that is only available with educational
site licenses (so I've read, anyway).


"doing" a class as in taking (buying) or giving (selling)? Either way
it's too much like buying or selling software or too much like work to
be enjoyable. I have a short attention span and horrible work ethic.

 

[design-engine]

that is the first time i see someone define RHO ....

 

[mcgowanp]

Hey Jeff, for sure there are better methods for creating that geometry and it all does depend on the acceptance criteria.


I just know that when I was learning surfacing i didn't have the moeny to do a course so I was trying pick up what I could from anywhere. I felt it took me a while to understand what I now think of as simple methods. If someone had pointed me in the right direction with regards to getting surfaces built and the various methods for achieving results I know that my progression to understanding surfaces and thus generating more acurate surfaces would have been quicker. I try to keep that in mind when replying to questions like this.


I do like post like this though, I know I lack a lot of the finer understanding of geometry that is displayed by the likes of yourself and doug. it pays to know what you dont know!!!


How are you getting on with it now Chris?? Have we confused you yet??

 

[kdem]

Jeff, Patrick


There is an EDU_Convert option that will convert student version files to the commercial version. However, its only availlable with the University Plus Edition. I've seen discussions on the PTC site about it and it sounds like it is only offer to schools and not to individuals. The purpose was to allow universities that interact with industry to share files with each other.

 

[design-engine]

mcgowanp: the same with me. I never got around to taking a class either.I think i was too cheap or did not believe it would help. In hind site I now realize i would have become the expert much faster if I were to have taken classes.Now I want to take a class and willing to pay and no one will have me. "you work for design engine".Well at least they know us.

I found the stress of learning something new is exhilarating.I designed several new classes (non PTC Certified) and now I have to teach those classes. http://www.proetools.com/courses/pro_surface/level8surfaceed it.htmand http://www.proetools.com/courses/pro_surface/level9.htm These classes are not live into the navigation just yet but the search engines are starting to sort them. Since this is the nose cone post I want to know what is a better name for this air forms workshop.Surfacing Aircraft Forms w/ Pro/Eor> Surfacing Aero Forms using Pro/ENGINEER



Edited by: design-engine

 

[zpaolo]

A 'plain old' (with singularities) cap can work well ...or fair ...


or produce garbage

Ok, this is a good point, playing with different methods I saw the different levels of strength and quality of output surfaces.

I'd like to add one thing: WF5.0 with dynamic edit is really helpful here, for example for the "plain old" cap I keep a trimming plane for the fuselage profile, and add a cap for the missing part. dynamic edit allow me to move the plane seeing when the cap surface degenerates and eventually limiting the capping to the smallest portion of nose cone. This looks really nice to me

Paolo

 

[jeff4136]

Paddy,


I have a sneaking suspicion that if taking a class would help here
this discussion would have been a short page in length.


Chris is probably overwhelmed. ;^)
I know I would be. But I believe the problem is; though he may
'have his fuselage' he probably does not have a suitable foundation
for a blend function nose cone.


(and Paolo,)
The intended purpose of the last two files I posted was to ...
1) illustrate how critical boundaries (differentiating two similar,
even seemingly identical, surfaces*) are when used in blend
definitions and
2) to illustrate the folly of blending / 'lofting' thru a series of
planar sections* (a method commonly advocated by experts**) to get
that fuselage, and similar, shapes. It may be the best a user can
come up with or be expedient and good enough for some applications
but it is rarely preferable. ("Loft" functions are an area where
some programs excel but the ones that do have setup options not
seen in 'MCAD' programs.)

* The first file is a blend based on the (vss) swept 'fuselage' surface
previously posted. The foundation surface in the second (...cap-neu_2)
was created by Boundary Blend thru section cuts created from the swept
surface. (It might make for an interesting experiment to recreate
the (swept section) elliptical conic arcs, Bndry Blnd thru them and
see what comes of it.) If you register the two files in an assembly
analyses show slight variations in curvature at the fwd boundary.
Other than that the only remarkable difference is the surface knot
distributions.

Knot span isoparametric curves:



** There was a time when a briefcase was required to be an expert.
The internet changed all that.
_ _ _ _


While I expect Chris may be overwhelmed (this stuff isn't easy and it
won't be trivialised) I do hope he's not put off. If he's interested
in learning I hope to save him a lot of time hunting snipe. If he
just wants to get a project done and isn't especially interested in
the 'finer points' it would be helpful to expose the data set to the
larger audience via a neutral file (I'd like to see it just to satisfy
my curiosity), let it be known and collect the best results offered.

 

[Arbiter]

Hi Jeff and Company,
I am really sorry for a terse post, I have seen a lot of really excellent posts to my question, and I will be digesting them soon! Unfortunately I am starting class at U. Cincinnati again tomorrow and have been uber busy getting ready over the past week or so! Thank you all so much, and I am not discouraged. I just ordered a book on inside pro/surface from amazon. Hopefully this post and the book will shed some light into surface design! I want to be able to design a smooth fuselage for a home-built composite aircraft I am designing. I will eventually post a neutral file if anyone is still interested. Thanks again!

-Chris

 

[design-engine]

That inside book on pro surface is 10 years old. Surprising they still sell (or people still purchase) that book.

good luck in classes U. CincinnatiEngineering I presume ... or aerospace engineering?

 

[Jacoolo]

hi jeff

I`ve got one question considering the approach introduced in WF2-COM-VER-NATIVE

(I do know this was introduced couple times before - considering others of You posts, and I know it is me just fake off while reading it earlier)

so returning to the case - I understand the idea behind rho value for conic, parabola or eliptical quadrant. I noticed You use parabola arc(Equ len) for nose curve and for VSS section. This is clear to me.



I just do not quite catch a clue considering the cross section rho value of it. You used (sqrt(2) - 1) - eliptical quadrant for last VSS.

The rest of equation I assume is considered to change the section dynamicaly - (.5 - (sqrt(2) - 1)) + (.5 - (sqrt(2) - 1)) * cos(180 + 360 * trajpar) and it could(and does) work well without it, am I right?



so by the use of (sqrt(2) - 1) I understand the section of first VSS is elliptical quadrant. But why? Is this in any moment of this surface(the cross section You created is for vertical line, would it be the same for the other position of the line?)



Second - how about the situation where the section of first VSS is gonna be different than parabola with EQU LEN? How to pretend then the cross section of it?

 

[jeff4136]

Chris,
Sounds like a full plate. Wish you well with it.
We can hope there will always be someone interested in looking at a geometry set and a problem.
_ _ _


(Jacek, let me study the questions and I'll get back with you.)