Twisted rudder with Costa bulb

[Matthias Maasch 21]

Hi folks,

 

please find attached a fully parametric and CFD-ready twisted spade rudder geometry, optional with costa bulb. This design offers you to use your own profile definition for each part of the rudder - the blade, the skeg and the costa bulb. Therefore the respective auxiliary scope consists of a feature which allows you to load a point-data file from your hard disk. In order to provide you a working setup I detached the imported NACA0018 profile (you can find the profile as a NURBS curve "imported_profile" in each auxiliary scope). If you want to use your own profile, simply load the data file and set it as input for the curve engines, respectively the image-profile for the costa bulb.

 

 

A few requirements have to be considered: The imported profile has to have its leading edge at x=0 and its trailing edge at x=1. Moreover the z-coordinate should be =0 for all points, the profile has to have an open trailing edge and the profile has to be a full profile (e.g. starting at the suction side trailing edge and ending at the pressure side trailing edge).

 

 

A fully implemented delta shift function twists the imported profile in a way, that the forward part of the profile is shifted in y-direction defined by a twist angle. Furthermore the start of the shift in x-direction, "twist start", as well as the intensity, "twist weight", of the shift can be controlled. The twist angle will twist the lower rudder part half the angle in one direction and upper rudder part half the angle in the other direction.

 

 

The rudder blade, and the skeg are modeled using metasurfaces, thus the geometry is described by functions, which can be customized as well.

 

 

The rudder blade is defined by a few typical parameters, such as blade area (= shadow area), blade height, profile chord length at the upper edge and the length of the profile from the rudder shaft to the leading edge at the top and the bottom. In addition you can define the propeller shaft height, the twist angle and the twist transition length, meaning the twist transition between lower and upper blade part in % of the blade length. For a reasonable surface this value should not be smaller than 0.2. All parameters can be found in the parameters scope, sorted by the rudder parts.

 

Finally all surfaces are meshed and joined in a CFD-ready STL. If you want to get rid of the costa bulb simply remove the object from the boolean sum: 03_STL -> twisted_rudderSTL. The idea is, that the rudder STL intersects with a hull STL so that both can be joined in a final STL. Therefore you have to make sure, that the upper part of the rudder skeg lays within the hull overhang.

 

 

If you have questions, do not hesitate to post them!

 

Cheers

 

Matthias

twistedRudder_costaBulb.fdb

[Mr. Daehwan Park 1]

Thanks for this!

 

The trimesh of rudder_blade is combined with both parts.

Is it reasonable for any purpose?

 

Daehwan.

[Matthias Maasch 21]

Hi Daehwan,

 

thanks for your question. Well, I got my "inspiration" from a Google search :)

See for example this rudder sample: Twisted Rudder

Actually my particular setup is not so crucial, as it is easy to change the twist-angle function.

So every user should be able to adapt this setup to get the desired shape.

 

For those of you who are waiting for the next release. I created basically the same rudder setup using the new Free Form Deformation.

This function is easy to handle and allows for remarkable form shifts. I started with a simple profile. I put this profile in a Brep and

extruded it to a certain height. Then I closed the open faces automatically. After that I used the Free Form Deformation to apply the

twist angle and another Free Form Deformation to change the dimensions of the rudder. The setup is really simple and easy to control.

Finally I added the shaft, the skeg and the bulb. All the single objetcs are joined in a Solid in the end. So, this setup is ready for

CFD simulations.

 

 

*above is a gif animation (not so easy to see, as it starts with a white screen :D )

 

Please note again: This function will be avaiable in the upcoming release CAESES 4.0. Furthermore, it will not be available in the

CAESES Free version.

 

Cheers

 

Matthias

[Guest Mr. YEONUK KANG]

hi maasch,

Brep rudder bulb looks good, I'm also want to use that model.

 

Now I check bulb size and shape what is best perfomance. (each ship)

 

we also check bulb shape (just cylinder type with fillet).

 

can I get the brep model as you upload picture , I use pro edition.

[Matthias Maasch 21]

Dear Mr Yeonuk Kang,

 

sorry for the late reply. Please find attached the above mentioned Brep model based on a free form deformation. You can find the parameters in the particular sub-scopes. Again I included a feature which allows you to import your own profile data from your hard disk. If you want to use it, simply go to baseline > 00_blade > 00_profile > import_profile > readFile and load your profile. Then set this profile as the source for the image curve baseline > 00_blade > 00_profile > import_profile > profile.

 

If you have further questions do not hesitate to post them here.

 

 

Best regards

 

Matthias

rudder_freeFromDeformationBased.fdb

[Guest Mr. YEONUK KANG]

hi , I have one question. 

in 02twist/FreeformDeformation -> 2 Translate ->we set tan(angle)*0.5*chord

 

There are not any range about that but it's automatically applied below to shaft center height and above to shaft center height.

Is it related with relative plane postions? It's also confused. 0, shatheight/bladeheight:20, shatheight/bladeheight,1:20 

 

variation applied to shatheight/bladeheight:20 and shatheight/bladeheight,1:20 

[Matthias Maasch 21]

Dear Mr Yeonuk Kang,

 

 

in 02twist/FreeformDeformation -> 2 Translate ->we set tan(angle)*0.5*chord

There are not any range about that but it's automatically applied below to shaft center height and above to shaft center height.

Is it related with relative plane postions?

 

I created two translation in opposite y-direction respectively in order to deform the BSplineVolume in positve y and negative y direction.

In order to apply the twist angle to the initial profile, I used the above mentioned line "(+-)*tan(angle)*0.5*chord". This sets the rotation point for the twist at 0.5 of the chord length.

If you want to start the twist of the profile in chord-wise direction earlier, then use values smaller than 0.5, if later, then use values larger than 0.5. You can also create a parameter for this for variation/optimization.

 

The setup allows so far only for a constant twist angle above and below the propeller shaft. If you want to have a variable twist angle along the whole blade it would require a change in modelling. It is an interesting question though, as the propeller wash out reaching the rudder is of course not constant over the blade height. Therefore it would be quite beneficial to have a fully parametric twist angle along the blade.

 

 

It's also confused. 0, shatheight/bladeheight:20, shatheight/bladeheight,1:20

 

This line basically creates the BSplineVolume for the FreeFromDeformation. It sets planes below and above the propeller shaft. I tested a few different setups at found to have 20 planes below and above to be the best and most stable modelling approach.

 

Best regards

 

Matthias

[Guest Mr. YEONUK KANG]

Thanks for reply. 

 

But I mean in Translate there are only dx,dy,dz . These is no any postion where is changed. but It automatically change below shaft heigth and above shaft height. 

 

I want to know how we select where is changed. I think it's related with bspline volume relative plane... but I don't know exactly

[Matthias Maasch 21]

Dear Mr Yeonuk Kang,

 

please have a look into the tutorial Free Form Deformation. Step 6 in the tutorial explains how to apply an operation to the deformation box. This is done by selecting points in the 3DView to which the particular operation will be applied. Going back into the project you can set the FreeFormDeformation to visible (00_blade>02_twist>FreeFormDeformation). This will show you the points to which the translations are applied. Switching between the operations will highlight the different points. Please see the attached pic.

 

Best regards

 

Matthias

[Guest Mr. YEONUK KANG]

oh...  I see,,, just click point is needed.

thanks ! ^^

[Karsten Wenzke 12]

Dear Mr. Yeonuk Kang,

 

here  (on our website) you can watch the video "Free From Deformation turorial" which shows how to use and setup the ffd.

Here is the youtube link not sure if you can access youtube from your location if not use the first link.

 

Maybe this helps you to get a better understanding.

 

Cheers,

 

Karsten