2d blade profile to follow arbitray curve while remaining in specified plane

[Mr. Bram1 Kerkhofs1 0]

Dear

 

I'm currently trying to model a helical (Gorlov) turbine.

After creating a blade profile (in the Z-plane) using the Fnaca4ds function and using FsweepTransformation to sweep the profile along a helical path.

In this particular case the helical path covers only a quarter of a helix (so it goes for about 90 degrees). The blade profile therefor should also turn 90 degrees and remain in the Z plane.

For now this isn't what happens. I believe that the orientation of the 2D profile to the sweep-line is defined in respect to the line and not to the coordinate system of the blade profile. Therefor the blade profile twists and turns and doesn't stay in the Z-plane.

I've tried already to get around this by using an image to create a pressure and suction side of the profile. therefor you can define the "start" end "end" on the attack and trailing edge. Using these points in order to use metasurface with skinning and two rail functions, in order to keep these points on the rails didn't work, it gave some issue's with overlapping triangles.

 

So the question is: how can I make a blade profile follow a certain path and control the orientation of the blade profile.

 

picture "selection_003" is the view in the z-plane, where the rotation of the profile isn't the desired 90 degrees.

 

Picture "selection_004" shows that the profile twists a bit out of the z-plane

 

[Willy Maschen 1]

Hey Brami,

 

to solve this problem you can use a translation instead of a sweep transformation. The key is to use the .getpos() command onto the curve. If you have a setup like in the picture it should work just fine.

As an input for the getpos() i used a parameter that goes from 0 to 1.

Hope this is what you needed, if not feel free to ask again.

 

Greetings

Willy

[Mr. Klaus Straka 0]

Hi Willy,

 

i have a question on a similar topic like Barmi. I want to sweep a profile along a helical path and at the same time keep a certain orientation of the profile. The goal is to use this sweep transformation to create a meta surface as describes in the "Sweep Transformation" tutorial.

 

The screen shot below shows the result of the profile movement with the sweep transformation. The transformed profile is somehow rotated along the path.

 

 

However, for me it is important that the two parallel edges of the profile always stay in direction of the z-axis as shown in the following screenhot.

 

 

For the transformation in the latter picture i used a translation (as you decribed above) and a rotation.  However, i wonder if this type of transformation gives me the possibility to create a meta surface via the curve engine. I far as i understood i have to use the sweep transformation in this case. Right?

 

Is there a possibilty to control the rotation of the profile along the path when performing a sweep transformation?

I attach a small example project with witch the screenshots have been created. Maybe it helps ;)

 

example.fdbc

 

Many thank for your help and best regards,

 

Klaus

[Jörg 29]

Hi Klaus,

 

Yes, for sure, this is possible. The "sweeping" parameter is then the curve parameter t which runs from 0 to 1. While sweeping according to your needs, you can furthermore define functions in the range [0,1] to control your rotation as well.

 

Does this help?

[Mr. Klaus Straka 0]

Hi Jörg,

 

thanks for your reply. I already played around with the "Sweep Transformation" and tried to control the rotation before my previous post. I attach a small project which is based on the Sweep Transformation Tutorial and where the sweep conture i twisted during the transformation. In the screenshots you can see that for this planar sweep path the conture is not rotated if the angle phi is set to zero.

 

 

If phi is set to 90 degree, then the conture is rotated as expected.

 

 

However, if the sweep path is not planar but a a curve in 3D space, then i get the unwanted rotation mentioned in my previous post. Can you give me some idea what the reason for this rotation is. Hopefully it is than more easy to compensate this behaviour.

 

Thanks again and best regards,

 

Klaus

SweepTrans.fdbc

[Jörg 29]

Hi Klaus,

 

Not sure if I understood correctly. However, here is a try: Please find attached a slightly modified version where I simply created a function for your rotation, to change the rotation along the path.

 

Best regards

Joerg

SweepTrans.fdbc

duct_rotation.mp4

[Mr. Klaus Straka 0]

Hi Joerg,

 

nice video, thank you :) . I try to reformulate my question.

Actually the project i attached in my last post was my first try to check if i can control the rotation of a conture along a sweep path and, although my solution was not as fancy as yours, it worked.

If the parameter phi is set to zero  the conture at the beginning of the path has exactly the same orientation as at the end. If phi is set to 90° the conture is twisted. So far, so good.

 

However, if i apply the sweeping to a helical path (see. post #3) the conture rotates during the sweep transformation (not very sure but i think it rotates around the tangent vector of the path) even if i don't set any rotational transformation. That's what i tried to explain in the first picture in post #3. In other words: With the finding from my example above i was in the opinion that the orientation of the sweep conture is exactly the same at the beginning and the end of the helical path (as long as the helix angle at the end is a multiple of 360°). However, this is clearly not case :unsure:

 

The question is now: what is the reason for this behavior?

 

Hope its more clear now,

 

Thanks for your help and have a nice weekend,

 

Klaus