Propeller Tip (beginner)problems

[Propeller Tip (beginner)problems]

Hi Jonas,

 

you can simply reduce the tolerance of the blade Brep:

 

best regards

 

Carsten

[Parametric Volute - Loss of tangency at the outlet duct/scroll interface]

Hi Simone,

 

yes you are right and we recently fixed this issue. Find attached the updated feature for the diffuser. We also update the tongue feature.

 

best regards

Carsten

 

features.zip

[Propeller modelling with 2d profiles at each radius]

Hi Shuguang,

 

thanks for the feedback. We will think about how the drawing could be implemented in the web app.

 

best regards

 

Carsten

[Propeller modelling with 2d profiles at each radius]

Hi Shuguang,

 

in that case you would have to use a lofted surface, which connects the profiles. Therefor you have to include a definition for skew, rake and pitch inside of the profile definition.

 

Have you already seen our web app for B-Series propeller?

 

https://www.wageningen-b-series-propeller.com/

 

Are you interested in such a model?

 

I guess you can also have the trailing edge at the origin. This shouldn't do any difference.

 

best regards

 

Carsten

[Propeller modelling with 2d profiles at each radius]

Hi Shuguang,

 

the question is what do you try to do? Do you want to have discrete profiles, which you want to modify or do you want to one profile definition, which you want to control with functions along the radius?

 

For the second case you can refer to our propeller demo model.

 

best regards

 

Carsten

[Starting Tsearch from best Sobol result]

Hi Mansur,

 

first you analyze usually the charts, to see the dependency of each design variable on the objective function.

 

Optimization_charts_variable_bounds.mp4

 

If there is nearly no dependency (which is indicated by a light yellow color), then you don't need that variable inside a TSearch. Also make sure that no variable reaches it's bounds for the best designs. if this is the case, then you need to extend the range of the design variable (if this is allowed). Usually you don't need to decrease the bounds of a design variable, because the TSearch will create local search from the current design.

 

To start a TSearch from a best design of a DOE, you can create a new design:

 

global_optimization_best_design_local_search.mp4

 

hope this helps

 

best regards

 

Carsten

[connecting CAESES and FLOW-3D]

Hi Farzam,

 

find attached a sample project with a software connector. Let me know if you have any questions.

 

best regards

Carsten

tumbler_clean.fdb

[Will CAESES read decomposed OpenFOAM cases]

Hi Per,

 

in general it should be possible. We will put it on our list for developments, but we will have to evaluate the effort. We will come back to you.

 

cheers

 

Carsten

[Will CAESES read decomposed OpenFOAM cases]

Her Per,

 

yes you have to reconstruct the mesh to read it in CAESES.

 

cheers

 

Carsten

[Reverse normals on Trimesh with V4.4 doesn't work.]

As a general remark, since this issue was solved via mail:

 

Our export still automatically points the surface normals inwards and therefor ignores the post processing operation "reverse normals". This does also work, if you substract parts from each other.

 

cheers

Carsten

[New member need help]

Hi Alan,

 

no this is not possible at the moment. You would need to create maybe a intersection or projection curve from the camber curve. Or you could also use Brep to trim the hub or shroud surface.

 

best regards

Carsten

[Reverse normals on Trimesh with V4.4 doesn't work.]

Hi Joseph,

 

why don't you combine them into one flow domain in CAESES?

 

I will ask the developers how we can handle this.

 

cheers

Carsten

[Reverse normals on Trimesh with V4.4 doesn't work.]

Hi Joseph,

 

I guess we did not thought of this. Do you export impeller and flow domain extra and merge it in Converge?

 

best regards

Carsten

[Reverse normals on Trimesh with V4.4 doesn't work.]

Hi Joseph,

 

I guess you are using the Converge Export. This one now automatically changes the normals so that they always point inwards, which is needed by Converge. So the post processing functionality might not work for this case. Why do you need this? Is the normal direction wrong?

 

best regards

 

Carsten

[NACA profile with NACA a=0.8 mean line and NACA 66 thickness distibution]

Hi Bastian,

 

Can you try to extend the hub curve a bit, so that the stacking line does not start at the begin of the hub section?

 

best regards

 

Carsten

[NACA profile with NACA a=0.8 mean line and NACA 66 thickness distibution]

Hi Bastian,

can you show the thickness distribution?

best regards

Carsten

[how to modeling to CONVERGE]

Hi Alex,

 

I send you a private message with the current version of the tutorial.

 

best regards

Carsten

[how to modeling to CONVERGE]

Hi Alex,

 

sorry for the late reply. Are you still interested? I can work on a tutorial, which I can provide to you, if you like.

 

bets regards

Carsten

["Maximum number of 2 instances is reached" with no instances open?]

Hi Rusty,

 

please try to remove all files in:

 

C:/ProgramData/boost_interprocess/FS_LICENSE_SHM2
C:/ProgramData/Application Data/boost_interprocess/FS_LICENSE_SHM2
C:/Users/All Users/boost_interprocess/FS_LICENSE_SHM2

 

best regards

Carsten

[Coordinate System]

Hi Suraj,

 

this is not possible at the moment. Maybe you can use the attached custom diagram.

best regards

Carsten

Diagram_custom.fdf

[Adding new parameters to an existing results table]

Hi there,

 

sometimes you have the following problem: You ran a variation and get a results table with your evaluation e.g. parameter "a". But then you realize, oh I forgot to devide my parameter a by the density. So you create a new parameter called "b" in the baseline and then you drag this to the results table. But the value will be "nan". This is due to the fact, that in those older designs the parameter "b" is missing. (You can check that by changing into one of the older designs by double click on this design.)

 

At the moment we don't have a proper way of solving this issue, which is why I created a workaround.

 

You can find the full article here:

https://helpdesk.caeses.com/solution/articles/12000048196-adding-new-parameters-to-an-existing-results-table

 

best regards

 

Carsten

[3D Window lost]

When your are missing your 3D Windows, there are different ways to recreate it:

 

• Reset CAESES to factory settings. Therefor go to file/settings:

• If this did not worked you can try to remove the stored configuration data in following folders:

            - Windows: %APPDATA/friendship/etc%

            - Linux: ~.friendship/etc

• you can first look if you just closed it and if you can not find it you can easily create a new one:

[Tutorial Files Missing]

Hi,

 

the IGES file is located in the installation directory as discripted:

 

 

best regards

 

Carsten

[Export into GridPro]

Hi Keegan,

 

you can also take the export type STL (Extract Colors). With that a stl file for each colored boundary will be created. I will check with the developers why the GridPro Export is not working as expected.

 

best regards

 

Carsten

[SSH Resource Manager Deployment]

Hi Moritz,

 

this problem is probably due to a version clash of your mariaDB database and the corresponding JDBC driver. You probably have mariaDB 10.1. Try to install the lastest mariaDB version (10.3) which is provided on the download page of mariaDB.

 

By the way your current student license will not allow you to use the resource manager. I will send you an email.

 

best regards

 

Carsten