Nice to see you in this discussion forum for Home work Level-2.
Regarding the geometry for this task, the blade angle is the same as Level-1 , Blade design-2. But, there are some minor difference in blade thickness. This geometry for Level-2 has a slightly thicker blade that is exactly 0.5 mm uniformly thick ( i.e from top to bottom surface) with a filleted leading edge and a flat trailing edge. In comparison the level-1, design-2 blade had both sharp leading and trailing edges.
Well, Should you come across any issues or problems with this homework, please feel free to post them here.
I hope that you all stay tuned.
About kinematic viscosity used in the model. It looks like we are using the default value 0.000001 m2/s. The kinematic viscosity of air seems to be about 0.00001 m2/s (see, for example, Properties of Air). Did I miss something obvious? Thank you!
The homework is really interesting, just a small question; why do we need to add layers in the rotor blade faces if we already made a mesh refinement on it? how this layers helps us to improve the simulation? thank you very much for your answers
I guess the layers are useful to define that just the surface has a fine mesh because thats where the air streams (it is therefore legitimate to have a “coarse” mesh inside the blade). So I guess the surface is more important than the inner part of the blade. Reminds me in some ways of the Boundary-Element-Method where you do the same. Just Mesh the surface (for instance crankshafts etc…) and you will get a pretty accurate result (sometimes even better than in FEM but depends on the case).
Just an idea, maybe @Milad_Mafi has a better explanation for us
Simulation for Level 2 is taking a lot of time. Its only around 9% complete in 120 mins. I expected it to be slow due to large no of cells. However, I just want to confirm. Please comment on the same.
In my case it took 76 minutes for mesh around 4.4 mln (downloaded results file about 850 MB).
I guess you missed with the Number of computing cores. If so, then please change it to 32.
As you might have already noticed and as mentioned by Milad in Topic " Step-by-Step Tutorial: Homework of Session 1 - Level 1", the platform was updated yesterday. The updated added a New Feature of “Material Library” to the simulation setup under ‘Model’.
So, all your projects and simulations have automatically been updated to the new setup. Your Drone projects and simulation setups created before 21.10.2015 will now look as shown below.
By default, a ‘Material 1’ with properties of Air will be there. Under ‘Topological Mapping’ the 1st entry of ‘volumeOnGeoVolume’ (i.e the main mesh) would be selected by default.
IMPORTANT : Please make sure that you also select ‘RotatingZone’ as shown above ( so that both are selected) and save it for any further Runs. All previous projects and runs will remain valid.
The above TUTORIAL will shortly be updated and show you how to add materials from the library (…its rather easy though ). Until then, make sure you have made the changes as shown. Ok
Hi @jigarparekh279
By “4.4mln” I meant 4.4 million elements. Generation time was 993 seconds (according to the mesher’s log), but actually it was more than 17 minutes. It takes the mesher some time to start and finish parallel processes.
However, 9% of a simulation in 120 minutes is really strange. Are you sure you did not miss anything? No misclicks?
What is your End time value and Time step length anyway? Maybe there is your problem. If you set the first one at 2000 [s] and missed to change default step value then you lengthened your simulation by 200 times (!).
How do your convergence plots look like? All below 1e-4?
There was a typo in the default value of 0.000001 m2/s. You are correct that the viscosity of Air is about 1e-5 or 1.5e-5 to be more exact.
This has been updated after the platform update yesterday. You will now also have a ‘Material’ sub-tree under ‘Model’ ( please see my latest post below for details).
Nevertheless, you do Not need to do the simulations again and your previous runs / Homework submissions are Accepted ( as the quantitative results are not so relevant ). For further simulation runs, you can now use the default value for Air.
If you have entered the exact values under 'Simulation Control ’ as shown in TUTORIAL at the top, the simulation should take about ( or similar with 16 cores):
Simulation Runtime (min)168 min
Simulation Result size 1,622 megabyte
( given you followed the exact values in meshing aswell to get a meshing time of ‘Finished meshing in = 791.9 s’ and ’ volumes/cells/3D elements:4400346 ')
Thank you for the timing details. I checked my simulation and couldn’t find anything different from the tutorial.
According the your timing details, I think I should wait for some time as I can see the residuals (although relative) falling and approaching 1E-4 for ~700 iterations.
@JhonnyFlash If you have a CFD simulation and the fluid does affect the stress on a component why should there be no grid “inside the component”? I think there are some examples out there that have the whole geometry meshed. Maybe @Ali_Arafat can explain a bit more.
This particular simulation only involves external flow analysis , so only the surface of the blades are meshed. The surface geometry/mesh is kept rigid and non-deformable. There are stresses on the component due to the fluid but the component is not allowed to deform because of these stresses.
The simulations were the solid also deforms due to fluid stresses are 2 way coupled Fluid Structure Interaction (FSI) type of analysis ( FSIPRO3D Wing Body Pylon Nacelle Model 2 Way Coupled Aeroelastic Vibration - YouTube ). Generally for FSI the " inside of the component" is meshed to calculate the solid deformation and internal stresses.
Here the case is simple, so a mesh inside the component is not required.
But, FSI in the pipeline and would be available on SimScale in the future.
As you might have already noticed and as mentioned by Milad in Topic " Step-by-Step Tutorial: Homework of Session 1 - Level 1", the platform was updated yesterday. The updated added a New Feature of “Material Library” to the simulation setup under ‘Model’.
). Until then, make sure you have made the changes as shown. Ok