Problem too complex for Variable Section Sweep? 2

[treddie]

Hello.

I am having problem with a situation that I thought would be easy to solve with a Variable Section Sweep, but it is slightly different and does not seem to be able to fit "within" a typical VSS. Please see the attached image which is a graphic of the problem. I hope the diagram is not hard to understand as it is always difficult trying to convey a problem like this graphically without making it confusing.

At any rate, the orange squares represent a section I am trying to turn into a solid protrusion that wraps around the outside of the oval cone. The sketch plane that contains this section is placed such that it is always normal to the slope of the two ovals. So as d127 changes by using the relation, d127 = trajpar*180 + cos(trajpar*180)*Offset, the protrusion is generated. "Offset" is a small value that prevents the generated curve from beginning at exactly 0 deg, and stopping at exactly 180deg, since there are singularities at those moments (when trajpar = 0 and 1). I will simply patch those spots if I can get this to work.

The ovals of the cone are NOT ellipses and are NOT of the same degree. I STARTED with an elliptic cone where the capping ellipses had the same degree of ellipticity, but had to offset inward by a small constant thickness since the elliptic cone represented the outer surface of a constant thickness skin. I am working on an oval shell that lies about 2 inches underneath that skin. This means except for that outer skin surface, all other ovals inside those ellipses can not possibly be ellipses, too.
My assumption (until it is obvious the assumption is wrong) is that since the slopes of the ovals where they intersect the sketch plane are equal, then along a line connecting those two points, the corresponding slopes at all intermediate stations will also be that same amount, since the ovals were generated by the same "linear" process.

Any help would be great. This one has me stumped!

 

[treddie]

HAL (from 2001: A Space Odyssey): "Just a moment...Just a moment..."

I THIIIIIINNNK I have a solution. I will post later today whether or not I found the answer.

 

[treddie]

DoH!

No go.

I tried a simple VSS with the small oval as the origin trajectory, and the big oval as the X-trajectory, but there is a subtle issue that makes that the wrong choice. The problem is that the same tangent angle on each oval means a different relative point on each oval, because they have different shapes. So the rate at which the section moves along the origin trajectory (the small oval)may be constant, but the point on the X-trajectory (the big oval) cannot...it must move at a variable rate which means the section can no longer remain normal to the origin trajectory. So the problem is one of having to let the section NOT be normal to the origin trajectory while letting a variable rate point on the big oval steer the section about its y-axis.

It does not seem that a VSS has that capability.

 

[treddie]

Here is an updated version of the image.

 

[treddie]

I think the only way to do this is to nix the idea of using a VSS, and instead start with surface strips made from the surface of the cone, then thickening those strips using the Normal to Surface option.

 

[grimmy65]

Why not create the two end surfaces then boundry blend?

 

[treddie]

The main reason is that those orange squares are just simple representations of the crossection I really needed, which was more complex.

As it turned out the Thicken idea worked really quickly and I was able to move forward. I basically used two Thicken operations to build two solids of different widths and heights and then merged them.

Sure would be nice though, if ProE had an additional Extrude tool that would allow you to set up a crossection's extrusion direction based on an animatable set of parameters locked to the variable (t). Would offer more flexibility than having to rely on the rigid structure of a VSS.

 

[mjcole]

Treddie,

A Star just for the Uniqueness of your situation and for your VSS Loyalty. I have been a fan of VSS for quite a while. I have not delved that far into your question but am quite intrigued by it and understand the problems with achieving a solution.

It's something I plan to attack soon so I can suffer through it like you have. Is there such a thing as a CAD Masochist? CADochist

Maybe you can use CREO Quandry when it gets released.


Michael

 

[treddie]

Yah, it's a strange one alright.

ProE really needs a less rigid method for running trajectory driven protrusions.

What ISSS CREO?

 

[mjcole]

CreO is the new name from PTC's cad solutions CreO/Pro is the Pro Engineer equivalent. There is also CreO/Direct for direct modeling CAD and there is a CreO view which is what Product View has been rebranded as.

Forgot to apply glue to the star now it shows.

PTC's creo -> creo

Michael

 

[treddie]

Oh cool! So Wildfire is no longer Wildfire...Creo now, eh?

 

[aafuni]

The creo installer installs to a different directory by default and acts as a whole new install rather than an update install. I accidentally ended up with both Creo and WF5 installed. The only difference I can see is that the windows say Creo at the top rather than Pro/E.

 

[treddie]

Are there any radically new features in your opinion?

 

[aafuni]

So far I have found nothing new.
I got this email today though:

http://image.exct.net/lib/fef3177473620d/d/1/MM_Active_December_2010_EN.html

According to that there are actual differences. So far I haven't really read through the email completely, but at first glance it just seems like mainly they made a whole other set of software that is now compatible with Creo. So basically there is no difference unless you want to buy a whole bunch of new licences for stripped down "Anyrole Apps". These apps seem to be directed at non engineers being able to look at and annotate models.

 

[treddie]

Thanks for the link. Yah, sounds like mix-and-match.

 

[looslib]

The current Creo/Pro is basically WF5 m65 with new logos.

Real Creo 1.0 won't be released until next summer.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli