Continue to Site

Welcome to MCAD Central

Join our MCAD Central community forums, the largest resource for MCAD (Mechanical Computer-Aided Design) professionals, including files, forums, jobs, articles, calendar, and more.

Graph of Cross Section?

jimshaw

New member
I have imported a long skinny partof varying cross section. It is for a fluids application; internal flow. If I rebuilt it in Proe, it could be a swept blend, where the trajectory is a planar curve. Or a vss of course.I would like to create a graph which has location along curve on the X-axis, and cross-sectional areaof the model on the y-axis, and I would like to automate the creation of this graph as much as possible. Has anybody done this? Any help would be much appreciated.


Thanks In Advance!!!


Jim
 
I tried once to do that for internal flow.I ended up doing it with cross sections and boundary blends.... figured out doing that approximately what i was looking for and recreated the internal structure for a compressed air flow... but the design changed so much so fast that I never got a suitable geometry. It ended up being a bifocated intake duct which I broke up into two separate VSS features and blended both with boundary blends.

I wanted to incorporate the cross sectional area so that data drove the graph but never took the time to complete it. I bet jeff could figure that out.

Drive a VSS with the cross sectional area so the duct gets larger or smaller at a specific rate so to manage pressure loss or gain.We wanted to control the speed of fluid.


Edited by: design-engine
 
Yeah, I am afraid that the graph might have to be the starting point of the design, however in this case, I have existing geometry that I need to analyze.


To add to the challenge, the cross-section shape changes significantly- worse than bifocated. I could extract the cross-section from a velocity profile, as it is incompressible flow, but it's far from laminar... for now anyways!
smiley4.gif
 
I think to get this to work we will have to create a simple example. once that is illustrated then apply what we learn to the specific examples we have.

Can you discuss your specific application? Is your project a gas or fluid application?

I am not sure how to control the cross section.or better state: drive a cross section with respect to area. I can do it the other way around tho. ;)
 
Okay, imagine a bullet floating inside of a curved tube. The cross section (as the flow follows the curve) can be described as: circle, donut, circle.


I would like ProE to tell me what the cross-section is as a function of the position along the curve. Since the trajectory is planar, maybe this can be done in 2-D?
 
that is easy if i understand what your asking...

What I wanted to do is control 'like behavioral modeling' ... the shape of the duct work by controlling a graph or equation. Drive shape with area as opposed to display area every 10mm along the duct.

If you want to do the first i think i can help.
 
Yes,I want to display area every 10mm (hopefully continuously!) along a curvy duct -I need to analyze existing geometry.


I know it sounds very counterintuitive, especiallyfor ProE,but the actual geometry does not allow me to drive the features from a graph. UnfortunatelyI can'tshare the geometry...
 
I did that very thing a few years ago an a drone inlet. You use an annotation feature to display Cross section area in a note. See if you can figure it out from there otherwise i have to open Pro/E
Edited by: design-engine
 
Jeff,

I think he wants to regulate the laminar flow http://en.wikipedia.org/wiki/Laminar_flow pressure... ie. duct gets larger by a a constant rate so the duct does not back flow (which is one reason the boundary layer is needed. (see F4J Phantom
[url]http://upload.wikimedia.org/wikipedia/commons/8/82/F-4J_Phan tom_II_VF-31.jpg[/url] image) In this image the inlet has a large panel forward of the inlet. See the Arrow pointing to the 101 on this oldy but goody.

Common problem in inlet design for jet engines and turbofan especially for high altitude take offs like above 15thousand feet. pratt&whitney, rolls-royce etc.

I guessed all this from the above description BTW becauwe I think about aircraft forms so much lately to design this new class.

http://www.proetools.com/courses/pro_surface/level9.htm guess i should add boundary layer and laminar flow to this class?
Edited by: design-engine
 
Thanks for the helps guys. We ended up doing it the old fashion way... lots of datum planes. If we get a chance to remodel it with vss and paramterize the X-sec, I'll share it fo' sho.


Bart - This isn't your first rodeo I see...


I'll start saving for that class... looks awesome!You might want to cover "how a wing works" at least, to give students the big picture idea, but a couple videos of smoke blown over a profile should do the trick :) Any more might be an overload. If you want to couple it with an intro toCFD class, you know where to find me... (BTW - that's not an F1 car on the bottom of the link above :)
 

Sponsor

Articles From 3DCAD World

Back
Top