Hi, this is osama. Actually I am trying to perform CFD on a structure but I am getting an error. I am new to simscale. Please can someone help me out with the CFD analysis. Here is the link to my project:
Hi Osama, welcome to the forum!
Your issue is most surely related to mesh quality, as explained in this article:
I can see you are currently running a new mesh, but here go my suggestions to improve the model:
- Reduce the enclosure size. This will allow you to reduce mesh cells and spend them where they are needed.
- Optimize the close region refinement. It should include the ground and it doesn’t need to go all the way to the top of the enclosure.
- Use automatic layers on the first runs, only change to manual when your simulation is working, you are confident in the results, and only want to perform a mesh independence analysis.
I copied over your project and performed the setup, you can find it here:
Hi @ggiraldof. I was able to get the mesh after almost 3 hours from the setup which you put. However I am getting an error while running the simulation. Please have a look and do help me out please.
Hi again,
What is the error you are getting? Any messages?
Also, I can see that the mesh is too big to be run with a community account, so maybe you should back down the refinement level on the close region.
ok so what should I take the minimum edge length as
I think you can augment the current value by a factor of 5 or 10.
Your first goal is to perform a draft simulation, but bear in mind that the results are highly sensible to the cell size, thus at the end you should end up with a size similar to what you have right now to get accurate results.
Another suggestion to to cut on mesh size is to make use of the symmetry of the structure if possible, i.e., reduce the enclosure in half :
@ggiraldof I have performed CAD cleaning operations to help me out with the simulation. Please can you help me out as to how can I proceed further from here to perform mesh in order to get exactly accurate results.
The general procedure is to perform a mesh independence analysis.
This consists of starting with a mesh you consider reasonable to get results, and then iterate with a refined mesh. This is repeated until the results do not change too much between mesh refinement iterations (setting a target variation, for example 5% or less)
If you start with a very coarse/bad mesh, this could take a lot of iterations. If done right, it should happen in three or four iterations.
My simulation is giving an error. Please can anyone help me with this. I have performed the mesh but it is giving me an error before starting the simulation. Please someone help me resolve this simulation
Why did you change the meshing algorithm?
I changed the algorithm because my simulation was failing
Hi, I successfully managed to perform CFD analysis. However the force Z plot which I am getting is very much unsteady. By theoratical calculations, I must be getting a pressure force of around 23000 N. However the pressure value plot in my CFD analysis is not steady alot. Please anyone can help me out with this ? @Retsam
Well, you will reduce initial wake in simulation by setting Potential flow initialization to Yes.
After 400 steps you have stable forces already (mind that the flow will be turbulent around your raft), but due to the simulation domain (enormous) size, so called Residuals are still on convergence slope.
But main problem in your simulation is your structure mesh:
In my opinion, you should not try any simulation with such a inappropriate mesh. So please focus on creating correct mesh.
Take care,
Retsam
1 Like
oh okay. so @Retsam please can you help me out as to how can I improve my mesh by looking at the mesh settings that I have made ?
@osamarauf: If I tell your professor that from 2019 on, your ‘study time’ at SimScale was 28 minutes, he would not believe:
Joined Aug 1, '19 Read 26m
Please let me explain that you came here to learn and play with 3000 core-hours. CFD is not for everybody: it is for guys willing to learn. If I make simulation in your place, you will learn nothing.
So please start to study tutorials on meshes first:
https://www.simscale.com/docs/tutorials/
Once done, go to different documentations:
https://www.simscale.com/docs-search/?search_term=mesh+generation
I hope you should be ready in couple of weeks to restart your project. If you have more than 20 hours of Read time and you still fail with your mesh, please be back. I will be happy to help you with fine-tuning of your simulation.
Take care,
Retsam
ok @Retsam I will definitely read that. I just wanted to ask that is there a way that I can perform this simulation with a community plan account ? or do I have to go to the proffesional account.
Secondly I wanted to ask that is it the size of the bounding box and cartesian box that I have selected wrongly?
In my opinion, that kind of simulation can be performed without any problem with community plan. But without correctly crafting the mesh and simulation, you need to pay for more resources.
For aircraft / wing simulation your BMB may be correctly sized. For simple drag force calculation on blunt body it is too big.
Take care,
Retsam
Hi @Retsam. I finally managed to run the simulation. The mesh seems to be quite better than before to me. The results also seem sensible. I just wanted you to see my simulation setup and please let me know if this one is correctly done. It would be really kind of you. Here is the link
Hi @BenLewis. Basically I have translated the results from CFD into FEA . I have used wind force and the structure consists of 144 solar panels. Each solar panel weights 24.5 kg. The size of each solar panel is 2180 mm x 996 mm x 40 mm. Furthermore, the structure is made up of mild steel. The top part of the structure consists of I beams . while the bottom rectangular part consists of C-channels.
Basically I want to find out if this material is safe to design or not. I have used fixed support at the bottom side. I performed FEA and the maximum value of von mises stress that I am getting is around 244 MPA. And we know that the tensile strength of mild steel is 250 MPA. So the factor of safety is coming around 1.03 which is less than 2. So I am assuming that the structure is unsafe.
I just wanted to ask that have I set up the simulation correctly. I have given the link to my simulation below. Secondly I wanted to ask that is this the correct approach to find out wheather the structure is safe or not.
Thirdly @BenLewis , by looking at the pictures can you give your views on the design of the structure. Is it safe to design or not. If not, then what modifications should be needed in the design.
