[trayana]

Greetings, everyone!

Working on a bachelor thesis, without any previous CFD modelling experience, my colleagues and I wanted to do some analysis for a propeller, since we are designing a quadcopter. We took data from APC propeller (7"x3"), and modelled it in CATIA, after which we used Xflr for generating polars.

We followed Capitao's report (https://www.tfd.chalmers.se/~hani/ku...itaoReport.pdf) closely, to try to get an estimate of the forces, and added extra calculations based on the first citation - Development of Virtual Blade Model for Modelling Helicopter Rotor Downwash in OpenFOAM. We scaled the disk to match our dimensions, and inputted our profiles and section divisions. However, despite that, we get underestimated values. Modelling the case for 10,000 RPM, in hover, we obtain the following final results:

rotorDisk output:
min/max(AOA) = 2.7406107, 12.881586
min/max(eps) = 2.1925445, 13.558032
Rotor thrust = 1.8003661
Rotor torque = 0.023655741
Rotor power = 24.772234
Effective drag = -0.41199112
Effective lift = 1.8003661
smoothSolver: Solving for Ux, Initial residual = 9.7307958e-05, Final residual = 9.715026e-06, No Iterations 6
smoothSolver: Solving for Uy, Initial residual = 9.6761527e-05, Final residual = 9.65839e-06, No Iterations 6
smoothSolver: Solving for Uz, Initial residual = 2.6749538e-05, Final residual = 2.3465639e-06, No Iterations 8
GAMG: Solving for p, Initial residual = 6.2058751e-05, Final residual = 4.4921504e-06, No Iterations 2
time step continuity errors : sum local = 3.2287854e-06, global = 3.2293843e-07, cumulative = 0.0036784382
smoothSolver: Solving for omega, Initial residual = 2.762776e-05, Final residual = 2.722353e-06, No Iterations 10
smoothSolver: Solving for k, Initial residual = 3.0377333e-05, Final residual = 2.4706647e-06, No Iterations 11
ExecutionTime = 148.88 s ClockTime = 153 s


SIMPLE solution converged in*277*iterations

Nonetheless, the file provided by APC Propellers indicates the following values: Thrust - 2.815N, Power 30.58W, Torque 0.029 Nm. Close to ours, but not enough, we were wondering if there is something that we could be missing. Our current belief is that the polars are off, but unsure how to check those. Lack of experience perhaps prevents us from using the right boundary conditions, too...
I'll send our current progress, and if anyone takes the time to provide some insight, we'll greatly appreciate it!

[nakonakata]

Hi Trayana, were you able to improve the results? We also have similar problems with DOE-RM1 and NREL Phase VI Wind Turbine validation cases where although simpleFoam has very good convergence and the mesh is structured, hexahedral and radially constructed, rotorDisk outputs are not meaningful. We also follow chalmers report.

Best,

Atakan