Problem with ellipsoidal section ducts - screenshot provided

Good evening Zozo and GT22,

I note your enthusiasm developed around this problem not very well posed by an author who seems to have disappeared from the radar... It must be the butterfly effect.

As for me, I'm still looking for the CAD file that seems to exist.

To return to the 20° twist, I would lean towards a sudden value, perhaps even without the knowledge of the author who would misunderstand the twisting of a 3D curve, rather than a stated desire to twist the geometry. So before putting these 20° in the specifications, we should make sure of their legitimacy.

I will finish by side with GT22 who is quite right, Zozo_mp is very close to winning the cup.

Good night

1 Like

here is the file in SW 2018 @ pierre

If you can convert it to parasolid I take thank you

I'm missing the SP5

@+

PS: but in view of several images

I have the impression that the journey of the ovoid tube is very far from over?????


tronxy-xy2pro_nozzle-fan_without-plate.sldprt

Hello Wednesday

As an attachment you have the file in Parasolid format. If anyone wants the SW 2019 version, just ask.

I think that the form is easily explained and understood if you look at all the sketches. In fact, if you look at the attached image and look at all the sketches for future volumes, you can see that the part we made is extended by a straight part that connects to a hollow (like a large countersunk head) see attached image. The same goes for another part.

Note that once you know how to twist and change the slope, it is possible to make the two main hollow channels in a single extrusion with the Smoothing and Material Removal function instead of making end-to-end extrusions. But hey! It's not what we're asked ;-) (hihhihihihi!!!!!  )


tronxy-xy2pro_nozzle-fan_without-plate.x_t
1 Like

thanks @ Zozo

Is it your file that you posted?

I haven't been able to recover the databases of the requester's file with the sketches

so I created another profile to join the other gaping part via a 3D spline on your @Zozo file

See attached image, it will be noted that the sections are constant and this without any guide curves

other than the parameter of the construction line my 3D sketch which is a spline

it's done quickly in surface

See attached file SW2017

@+ ;-)

 

 


tronxy-xy2pro_nozzle-fan_without-plate_test1.sldprt

@gt22

Errare humanum est

Your interpretation of the rest is in my opinion wrong.

Look at the attachment with the explanation. The previous one was not very clear when it is not attached

For the file I have been working since the beginning on the @ 30187066b5 file without changing anything. The one I sent you just contains my proposal.
But the sketches are not transmitted to the Parasolid file.

Kind regards


v4_elipsoide_courbe_et_vrillee_2020-08-05_11_15_09-window.jpg
1 Like

Yes, I know very well that it's completely 

not the type of part requested

it is simply an example 

I have As explained draw a 3 d curve on the ins and outs 

on which I created a surface smoothing

for explicit that it can work even if it's not on the same level 

That's all 

and it allows you to always have exactly the same profile all along the guide curve

What from the view of your part does not seem quite the case on the geometry of the profile in theory we should have the same profile for each perpendicular plane along the guide curve

@+

Ha!!!!! OK!

A very good example, by the way.

On the other hand, you know that I'm a marble in surface, so could you tell me how you fill the space between Surface-Smoothing 1 and Surface-Smoothing 2.

In other words, I only work in volume while you go through the Surface phase which is simple. But what about the empty space to be filled (what function allows this.)

By the way, I don't understand why I was forced to go through two guide curves when you didn't need to. I did a little surface still ;-) ;-) to make my guide curves.

To try to understand I made an angular offset like the model and then your surface-smoothing to a funny head. (see Attachment) Just like mine before I did two guide curves.

It is the angular offset that would ultimately force us to make two guide curves.

What do you think?                                                                                We don't get paid a lot but we have a lot of fun;-)

Kind regards
 

 


v6_elipsoide_courbe_et_vrillee_2020-08-05_11_15_09-window.jpg

@ Zozo

it's not going to do it ................ ;-( ...........; -(            (dispute mode with ear pulling and period blowing on the fingers)

I don't know what you did

but my internal part is like that

so well as it should be and in line with the initial profile and arrives ;-(

you had to move the internal curve line to the smoothing with the 2 green dots which must absolutely be on the right fiber from end to end

then to make a full body you just have to create the surfaces that are missing, sew them and say I want a volume (check the cell)

In addition, you'll see that I deliberately left a space between the exit of your elbow and my sketch resketch via a sketch offset and create a surface that doesn't touch your elbow

but quickly adjustable 

but I repeat it was only for the example (but it also works in volume) although sometimes he doesn't like splines

simply out of habit and as I know that we go a hair + far in surface area than in volume (deformation of the material) in the handlebars and we can have smaller radii

I often do this ;-)

if you need + explanation tell me 

To confirm my words, reopen the file I'm posting and hide the smoothing surface 2

What I did on 1° screen 

on the 2° a transparency 

and in theory and practice if you create a perpendicular plane(s) via a constrained point on  the construction line (spline in 3D sketch) in yellow

you will always find the same profile

@+ ;-)

 

@ Zozo, please forgive me.............. I understood after X reflection (PHEW)

that you had twisted the internal smoothing by X degree so in the opposite direction of a clock

I think it was for the proper understanding of the feasibility of your elbow

well after thinking about it having managed to turn your elbow

I had fun modeling it, it doesn't go without guide curves

The radius is too small and this creates continuity breaks

 

see attached file SW 2017

@+ ;-)

 


tronxy-xy2pro_nozzle-fan_without-plate_test1.11.sldprt

Yes! but my fingers hurt a lot after your period   ;-)

Not easy to use the mouse now ;-)

We could almost set a ruler if I take your last image with the pink extrusion, it works because the sides of the squares  of the sketches are parallel to each other, but if you tilt one of the two squares then you need guide lines. This is even more visible with the ellipse.

Indeed for the ellipse if the angle of arrival is slightly twisted there too you need guide curves that cannot be done freehand, hence my tip to have an ellipse keeping the 2.5 x 1 all along the curvature. This is so that there is no brake or turbulence for the fuide.

Here is an example to conclude the topic   (pouf, pouf, pouf)

Kind regards

 


v6_elipsoide_courbe_et_vrillee_2020-08-05_11_15_09-window.jpg

Did you see that I redid your elbow?

Wow!  I saw ;-)

Kind regards