logo
NOTICE:  This is the new PunchCAD forum. You should have received an email with your new password around August 27, 2014. If you did not, or would like it reset, simply use the Lost Password feature, and enter Answer as the security answer.
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Options
Go to last post Go to first unread
gsjames  
#1 Posted : Saturday, August 04, 2007 11:47:02 AM(UTC)
gsjames

Rank: Junior Member

Joined: 4/7/2007(UTC)
Posts: 27
Location: TX

The 3 attachments are representations of a horizontal tail for a replica of a 1938 world speed record attempt airplane.

The planform is not a trapazoid like the wing planform tool in AeroPack, they represent a realistic planform with rounded tips.

The skinning was done using the two rail sweep tool since there is not an airfoil per-se at the wingtip. The skinning is unusable. What's going on here?

The first attached tail is with the leading and trailing edges as control point splines, the second is with them as interpolate splines and the 3rd is with bezier splines. None of them are satisfactory.

Am I doing something profoundly wrong, or are we looking at a problem. This should be a very straightforward task.
File Attachment(s):
Tail with control point splines.cu (87kb) downloaded 6 time(s).
tail with interpolate splines.cu (86kb) downloaded 6 time(s).
Tail with Bezier Splines.cu (43kb) downloaded 6 time(s).

You cannot view/download attachments. Try to login or register.
gsjames  
#2 Posted : Saturday, August 04, 2007 12:38:58 PM(UTC)
gsjames

Rank: Junior Member

Joined: 4/7/2007(UTC)
Posts: 27
Location: TX

I had the spline directions going tip to root instead of root to tip and after I changed directions on the "rails" it gave me a better result but still not satisfactory. Look at the tip area of the attached file and see how things get "wavy". I've seen this behavior before on other surfaces. Wings don't look like that. Where is the waviness coming from?
File Attachment(s):
tail with interpolate splines and corrected directions.cu (41kb) downloaded 6 time(s).

You cannot view/download attachments. Try to login or register.
Steve.M  
#3 Posted : Wednesday, July 02, 2008 1:48:41 PM(UTC)
Steve.M

Rank: Senior Member

Joined: 6/18/2008(UTC)
Posts: 978

A late reply to thread (well, I am new here), but wanted to show a solution.

The problem appears to be due to the amount and position of the control points on the rails used for the sweep(wing_tip_01). I did (quickly) re-draw the rails, and made the sweep again, and without problems(wing_tip_02)
Steve.M attached the following image(s):
wing_tip_01.jpg (15kb) downloaded 6 time(s).
wing_tip_02.jpg (21kb) downloaded 6 time(s).

You cannot view/download attachments. Try to login or register.
jol  
#4 Posted : Wednesday, July 02, 2008 2:07:22 PM(UTC)
jol

Rank: Senior Member

Joined: 2/26/2007(UTC)
Posts: 2,156

interesting lesson

- thanks for posting !
Steve.M  
#5 Posted : Wednesday, July 02, 2008 2:27:00 PM(UTC)
Steve.M

Rank: Senior Member

Joined: 6/18/2008(UTC)
Posts: 978

Hi jol,
jol wrote:
interesting lesson
I do think the simpler the cross sections/rails can be the better (for any nurbs package), but it is the first time I have really taken time with acis kernel, maybe it is a little particular on control points, but I do stay minimal (with control points) with all nerbs apps when building surfaces. I am not actually sure that it is the kernal that is in control of the sweeps, I am a little lost in that area (I like to build models, I cannot build these types of programs lol)

jol wrote:
- thanks for posting !
I will be around quite a lot now (if OK) as my only real problem is with which version (pro/shark) to purchase, and I do like interaction with other users.


Regards,

- Steve
jol  
#6 Posted : Wednesday, July 02, 2008 2:54:29 PM(UTC)
jol

Rank: Senior Member

Joined: 2/26/2007(UTC)
Posts: 2,156

There are better possibilities with regard surfaces in Shark - though the limitations with the tools that are common are the same

The guys are making really good strides forward lately - and the mass of constructive feedback seems to be a great driver

But there are still areas that need looking at afresh - like issues that have come up again today -

Including surface matching .. you just can't match a matrix of patches with disparate conditions - like those for example that make up the body of a modern power drill.

And surface analysis via opengl is very poor with no real time reflections - let's not even talk about the zebras which are worse than useless
ttrw  
#7 Posted : Wednesday, July 02, 2008 2:55:44 PM(UTC)
ttrw

Rank: Senior Member

Joined: 4/1/2007(UTC)
Posts: 1,583

I see where you are coming from now Steve! Sorry for that confusion earlier. This is a wicked tutorial- thanks a lot for posting- it makes a lot of sense too!

What's that term again...."Knowledge is power"? :rolleyes: :D

Tom :)
jol  
#8 Posted : Wednesday, July 02, 2008 2:58:24 PM(UTC)
jol

Rank: Senior Member

Joined: 2/26/2007(UTC)
Posts: 2,156

>> I will be around quite a lot now (if OK) .. and I do like interaction with other users

There's always been great community around these tools. Having interested users like yourself prodding the limits of the software can only help to improve them
ttrw  
#9 Posted : Wednesday, July 02, 2008 3:07:36 PM(UTC)
ttrw

Rank: Senior Member

Joined: 4/1/2007(UTC)
Posts: 1,583

I gotta say chaps, that I too really love the community here (as you can probably tell? ;) I also tend to love internet forums- its a great way not to do any work!! hehe!).

It's a shame that the bulk of my models to date have been made in Solidworks, otherwise I would have posted by now. But I guess I've just gotta get cracking with VC, and upload some stuff- all in good time. :)
Andusan  
#10 Posted : Wednesday, July 02, 2008 4:03:41 PM(UTC)
Andusan

Rank: Member

Joined: 7/31/2007(UTC)
Posts: 90

Quote:
But I guess I've just gotta get cracking with VC, and upload some stuff-


but as you said : all in good time !
Steve.M  
#11 Posted : Thursday, July 03, 2008 5:11:52 AM(UTC)
Steve.M

Rank: Senior Member

Joined: 6/18/2008(UTC)
Posts: 978

Hi jol;

jol wrote:
There are better possibilities with regard surfaces in Shark -
I do see that they have split the ability of surfacing over the various products.

The surface over 2 rails is limited, as it appears only possible to use 1 cross section, the option for a group is a little confusing, does the "group" mean 1 cross section of a connected group of objects? If I attempt multi cross sections, then I get a "unknown error" popup.
Yes, there is the skinning with rails and also net, but I need to check/compare the surfaces created by those.
jol  
#12 Posted : Thursday, July 03, 2008 5:49:58 AM(UTC)
jol

Rank: Senior Member

Joined: 2/26/2007(UTC)
Posts: 2,156

Net surfs are very good and very flexible

and approximations are made if your 'U' curves and your 'V' curves dont cross exactly -- meaning it's great for roughing in surfaces then refining

dont forget that in most tools you can use points in place of cross-curves .. though watch for for surface degradation

There's a whole bunch of helpful comments in Tim's document on advanced surfacing (I think it's Shark only) .. I wish I could find it for you as it has some dead useful stuff in there
Steve.M  
#13 Posted : Thursday, July 03, 2008 5:57:48 AM(UTC)
Steve.M

Rank: Senior Member

Joined: 6/18/2008(UTC)
Posts: 978

jol wrote:
Net surfs are very good and very flexible
As long as you dont get strange popups of errors.

For a "net" (which I presume is a "network" or "boundary") surface, then I would normally expect a G2 surface, but that can cause bulges. It is sometimes preferred a forced G1.
Users browsing this topic
Guest
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.