Another Loft Problem on a fan blade

[gettinnerdy]

I've read through the forums and tried various solutions posted for wind turbine blades and other similar geometries, but nothing seems to help loft this.
I'm trying to import and recreate at 3DM file that you can see in the features. My general approach is to take the 3DM file and slice it into 0.1" sections, create sketches on those planes, then loft them all together. Before I go through the whole fan blade I want to try to get the first 9 or so sections to work properly.

Depending on how I try to achieve the loft I get one of these errors:
SKIN_BAD_UV_SKIN_DIR: Surface U and V are in the same or opposite direction. (ext)
or
SKIN_SELF_INTERSECTION: Self intersecting or cusping surface found. (ext)

I have tried "smoothing" the individual sketches more, I have tried creating guide lines from the 3DM file, I can add more "slices" along the loft path, but I'm not sure that will be helpful. Any suggestions?

Also if my approach to getting a working drawing from this 3DM file is wrong please let me know a better way.

 

[simonb65]

If you zoom right in on the sketch nodes, you see that they aren't actually on the b-spline ...



Maybe precision or rounding errors with Alibre are causing your guide curves and aerofoil sketches are not creating valid sweep criteria. If you don't get an answer from the Forum, I would possibly post this to the Alibre team to see if they can internally pinpoint the problem.

It may be related to the recent update of the core ACIS ... or maybe not!

 

[gettinnerdy]

If you zoom right in on the sketch nodes, you see that they aren't actually on the b-spline ...

View attachment 25838

Maybe precision or rounding errors with Alibre are causing your guide curves and aerofoil sketches are not creating valid sweep criteria. If you don't get an answer from the Forum, I would possibly post this to the Alibre team to see if they can internally pinpoint the problem.

It may be related to the recent update of the core ACIS ... or maybe not!

Thanks, but I'm not sure that's it. I just went in and re-drew by hand the sketches and it seems to be acting the same way. Side note, any time I zoom in far enough on a sketch the nodes don't appear to be on the spline...this is true even of other designs that have working lofts...
Sam

 

[simonb65]

Have you tried just using the two end aerofoils and adding more guide curves by placing more nodes along the chord of the upper and lower surfaces?

 

[gettinnerdy]

Have you tried just using the two end aerofoils and adding more guide curves by placing more nodes along the chord of the upper and lower surfaces?

View attachment 25839

View attachment 25840

Thanks Simon I'll give that a try.
Are those guide curves all part of the same 3D Sketch or is each Guide Curve a unique sketch?
Did you do anything special or use any special logic or knowledge to choose where you place the nodes along the chords?

 

[simonb65]

The guide curves are separate 3D sketches. I just split the chord in half, then each half into thirds or you could use some planes to 0.254" spacing in the x direction as a guide to place them, i.e. ...



Repeat for each cross section, then create the 3D sketch guides by connecting the nodes along each plane. If you don't get the guides in a straight(ish) line on such a compound surface, you get some weird distorted artefacts creeping in !

It really needs a function to evenly place a number of nodes along a spline, between the end ponts !!

 

[gettinnerdy]

The guide curves are separate 3D sketches. I just split the chord in half, then each half into thirds or you could use some planes to 0.254" spacing in the x direction as a guide to place them, i.e. ...

View attachment 25841

Repeat for each cross section, then create the 3D sketch guides by connecting the nodes along each plane. If you don't get the guides in a straight(ish) line on such a compound surface, you get some weird distorted artefacts creeping in !

It really needs a function to evenly place a number of nodes along a spline, between the end ponts !!

I love it, thanks I cant wait to try it!
I have no idea how to write a function for this, but I look forward to working that out.

 

[bigseb]

While probably do-able (with a lot of effort) AD might not be the most suitable software to do this type of flowing/organic modelling. Perhaps one of the higher end packages like Catia, Creo or NX and be a better choice?

 

[gettinnerdy]

While probably do-able (with a lot of effort) AD might not be the most suitable software to do this type of flowing/organic modelling. Perhaps one of the higher end packages like Catia, Creo or NX and be a better choice?

You may be right. I am getting no where with the equation editor figuring out how to distribute nodes across a spline. I'm not even sure that adding more Guide Curves (even if I can get them distributed properly) will make this loft work.

 

[bigseb]

Lofts are not AD's strongest area. You can do it but most likely you will need to use the '100s of sketch planes' method you described in your first post.

 

[bigseb]

Alternatively a surface modeller like Moi or Rhino would do a great job of this too.

 

[Lew_Merrick]

You may be right. I am getting no where with the equation editor figuring out how to distribute nodes across a spline. I'm not even sure that adding more Guide Curves (even if I can get them distributed properly) will make this loft work.

GettingNerdy -- Assuming that you have (x,y,z) values for your Loft Contours why not Sketch them using Wizoscript?

 

[Oldbelt]

See this thread
https://www.alibreforum.com/forum/i...ft-from-airfoil-data-points.12255/#post-74144

or search Oldbelt : Propeller.

 

[Oldbelt]

 

[oldfox]

Nerdy, here is a wind turbine blade I made in 2011. I don't remember a whole lot about the thought
processes which were at work but what ever they were they were successfull. This model also ran
through a CAM simulation without error. Looked great.
Slice and dice it and use it as you will. Good luck. Just think of it as a wordless tutorial.

It was still early into learning Alibre so be nice.

You may want to take a look into Wizoscripts. Andy Ayre, the author has done a fantastic job. There
is a script in there for plotting an airfoil section (think wing rib). If you don't know about it, the
University of Illinois at Urbana Champagne. (UIUC) has a great airfoil library with all of the coordinates. Just enter them into a CSV file and Wizoscripts will draw it for you. You can then go on
and build the hub, root or what ever. You just need 2 files: one for the root and the other for the tip.
And of course the ubiquitous lofts.

 

[Oldbelt]

It is not needed to make a script to input datapoints from the airfoil library.
Delete the first line in the air foil .dat file and read in with alibre point import, found in sketch>figures>insert from file.
Thats simpel, and keep it simpel, this thread has shown a lot a not nessasery " inovative methods".

 

[oldfox]

read in with alibre point import, found in sketch>figures>insert

Thanks for the tip. A whole lot simpler. Since I just use Alibre at the hobby level, I just keep using what I know works. My bad. Always
good to learn more ways of doing things.

 

[Oldbelt]

getinnerdy.

Please inform foilname, rotcord, tipcord, span, washout (twist) and I wil show how to do.

 

[GIOV]

Hi all,
I did the blade with loft through each profile.
I hope this help
Here is the file:

 

[GIOV]

And this one have some guide curves:

To get a smoother surface you would need more curves guides.