[Fraxiel]

Hello everyone, Im trying to simulate a flow around a fixed wing UAV, the flow is relatively slow around 22m/s and the mesh quality is good, I have a tetrahedral mesh with an orthgonal quality of 0.201. Moreover, the mesh has capture proximity and curvature and it captures the geometry of the aircraft well. Im running the pressure based solver along with coupled. All the others are set to second order. However, when I check the residuals the continuity is not converging I have no idea why. The continuity residual has reached a value of 1e+2. while the others have reached a value of 1e-7. Also the k-epsilon residuals have not reached the value of 1e-3. Anyone know whats going on here?

[LJA_Koers]

Can you help us with a little more info?

Maybe a velocity contour of a cross section and the boundary conditions you applied on the boundaries of the domain? Did you apply no-slip conditions on the far-field boundaries? Or on the UAV surface without an inflation layer?

Also, for a problem like this, I would write an expression to see what the ratio of mass inflow to outflow is to check for mass balance

The residuals are scaled to their initial values. If you use hybrid initialization, the flow is more resolved than if you would use standard initialization, and the residuals reduce less without the solution being less accurate. Still, the continuity residual shouldn't be 1e+2 though. I think that when I had my first internship experience with CFD, hybrid initialization wasn't available yet and a residual of 1e-3 was a rule of thumb for convergence, but is no longer a required value. Somebody more experienced than me can hopefully explain this a little more.

[Fraxiel]

The fluid domain is a spherical domian, so I have no inlet and no outlet its just a pressure far field, and I set it with a Velocity and a certain presssure and temperature. The aircraft has the no slip condition along its surfaces. I cannot post a velocity contour unfortunately or anything of that matter due to reasons. However the velocity contour makes sense the air speeds up on the top of the wing which results in the pressure difference and creates the lift required. The CL and CD values make sense. I wanted to check the mass flow rate and the net mass flow rate is equal to the mass flow rate through the pressure farfield. I tried it with inflation and without the inflation layers and faced the same issue. Any advice?

[Fraxiel]

Originally Posted by LJA_Koers

Can you help us with a little more info?

Maybe a velocity contour of a cross section and the boundary conditions you applied on the boundaries of the domain? Did you apply no-slip conditions on the far-field boundaries? Or on the UAV surface without an inflation layer?

Also, for a problem like this, I would write an expression to see what the ratio of mass inflow to outflow is to check for mass balance

The residuals are scaled to their initial values. If you use hybrid initialization, the flow is more resolved than if you would use standard initialization, and the residuals reduce less without the solution being less accurate. Still, the continuity residual shouldn't be 1e+2 though. I think that when I had my first internship experience with CFD, hybrid initialization wasn't available yet and a residual of 1e-3 was a rule of thumb for convergence, but is no longer a required value. Somebody more experienced than me can hopefully explain this a little more.

The fluid domain is a spherical domian, so I have no inlet and no outlet its just a pressure far field, and I set it with a Velocity and a certain presssure and temperature. The aircraft has the no slip condition along its surfaces. I cannot post a velocity contour unfortunately or anything of that matter due to reasons. However the velocity contour makes sense the air speeds up on the top of the wing which results in the pressure difference and creates the lift required. The CL and CD values make sense. I wanted to check the mass flow rate and the net mass flow rate is equal to the mass flow rate through the pressure farfield. I tried it with inflation and without the inflation layers and faced the same issue. Any advice?

[LJA_Koers]

if you applied a constant velocity, in lets say the positive x-direction, on the entire sphere then maybe that causes problems in the area in the wake of the UAV if the sphere isn't large enough.

[Fraxiel]

Originally Posted by LJA_Koers

if you applied a constant velocity, in lets say the positive x-direction, on the entire sphere then maybe that causes problems in the area in the wake of the UAV if the sphere isn't large enough.

Hmm that could be the issue as I took the length of the fusleage and multiplied it by 10, in all directions. I remember specifically that the sphere has to be larger than 15 times to properly capture the wake. Ill try that and let you know what happens!! thanks

[LJA_Koers]

Did you find a solution?

[Fraxiel]

Originally Posted by LJA_Koers

Did you find a solution?

I kept searching and found out that the continuity and residuals aren't really a criteria for if my solution makes sense or not. I finished creating a mesh independence study and found out that I have a problem with the first layer height that I fixed and I will try to see the new results next week. Appreciate the help <3 and Ill try to keep you posted