CFD analysis of low reeynolds number airfoil

abu.bakar.ist · May 29, 2019, 04:55

hi, i am simulating low reynolds number flow over E387 airfoil. For meshing i am using ICEM.

Firstly i perform a grid study (reynolds number 200,000, alpha 4), increasing the mesh size and checking pressure distribution , lift and drag coefficient. Final grid (also attached fig 1 and 2 ) give wall y plus less than 1 (convergence 10e-4). Pressure distribution lift and drag coefficient very close to experimental.
To capture laminar separation bubble i am using SST transition model, intermittency =1 , tub intensity = 0.1% (from wind tunnel data) and Tub ratio = 10 (as default). SIMPLE coupling, 2nd order spatial discretization.
pressure distribution plot, fig 3.

Then I ran alpha sweep on the final grid. I am having problems at alpha 8 and higher. Fluent is unable to resolve the flow. (fig 4 and fig 5)

In literature, very good results have been presented using same scheme and transition model. Can some one guide me and let me know what i am doing wrong. I am not an expert in CFD, i am sorry if i miss some details ..

fig 1.bmp

fig 2.bmp

fig 3.bmp

fig 4.bmp

fig 5.bmp

AlexanderZ · May 30, 2019, 00:11

what is Alpha -> angle of attack?

start with last converged solution (let it be solution for aoa=5)
change aoa little bit (for aoa=6) and continue computation (don't initialize)
continue this iterations

best regards

abu.bakar.ist · May 30, 2019, 01:01

yes . alpha is angle of attack ..
i will try it ..
thanks for you comments

abu.bakar.ist · May 30, 2019, 01:36

i have tried it .. did work ...
results are the same.. no improvement ...

AlexanderZ · May 30, 2019, 06:52

you may try to use transient simulation with small timesteps.

Imagine a particle flow through your domain with input velocity, so that particle should went through your domain twice, for example:
your domain is 20m, velocity is 100m/s, so you have to simulate at least 0.4 sec physical time. With timestep = 0.001 -> it will be 400 time step.

Also check Cd and Cl

best regards

Fabio88 · August 30, 2019, 04:10

Try to change P-V coupling (ex. Coupled). Might help.

May 29, 2019, 04:55	CFD analysis of low reeynolds number airfoil	#1
abu.bakar.ist New Member bakar Join Date: Dec 2012 Posts: 4 Rep Power: 13	hi, i am simulating low reynolds number flow over E387 airfoil. For meshing i am using ICEM. Firstly i perform a grid study (reynolds number 200,000, alpha 4), increasing the mesh size and checking pressure distribution , lift and drag coefficient. Final grid (also attached fig 1 and 2 ) give wall y plus less than 1 (convergence 10e-4). Pressure distribution lift and drag coefficient very close to experimental. To capture laminar separation bubble i am using SST transition model, intermittency =1 , tub intensity = 0.1% (from wind tunnel data) and Tub ratio = 10 (as default). SIMPLE coupling, 2nd order spatial discretization. pressure distribution plot, fig 3. Then I ran alpha sweep on the final grid. I am having problems at alpha 8 and higher. Fluent is unable to resolve the flow. (fig 4 and fig 5) In literature, very good results have been presented using same scheme and transition model. Can some one guide me and let me know what i am doing wrong. I am not an expert in CFD, i am sorry if i miss some details .. fig 1.bmp fig 2.bmp fig 3.bmp fig 4.bmp fig 5.bmp

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Ffd_control_point_2d	feiyi	SU2	4	September 30, 2019 12:42
[mesh manipulation] Mesh Refinement	Luiz Eduardo Bittencourt Sampaio (Sampaio)	OpenFOAM Meshing & Mesh Conversion	42	January 8, 2017 12:55
Compressor Simulation using rhoPimpleDyMFoam	Jetfire	OpenFOAM Running, Solving & CFD	107	December 9, 2014 13:38
DecomposePar unequal number of shared faces	maka	OpenFOAM Pre-Processing	6	August 12, 2010 09:01
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues	michele	OpenFOAM Meshing & Mesh Conversion	2	July 15, 2005 04:15

May 30, 2019, 00:11		#2
AlexanderZ Senior Member Alexander Join Date: Apr 2013 Posts: 2,363 Rep Power: 34	what is Alpha -> angle of attack? start with last converged solution (let it be solution for aoa=5) change aoa little bit (for aoa=6) and continue computation (don't initialize) continue this iterations best regards

May 30, 2019, 01:01		#3
abu.bakar.ist New Member bakar Join Date: Dec 2012 Posts: 4 Rep Power: 13	yes . alpha is angle of attack .. i will try it .. thanks for you comments

May 30, 2019, 01:36		#4
abu.bakar.ist New Member bakar Join Date: Dec 2012 Posts: 4 Rep Power: 13	i have tried it .. did work ... results are the same.. no improvement ...

May 30, 2019, 06:52		#5
AlexanderZ Senior Member Alexander Join Date: Apr 2013 Posts: 2,363 Rep Power: 34	you may try to use transient simulation with small timesteps. Imagine a particle flow through your domain with input velocity, so that particle should went through your domain twice, for example: your domain is 20m, velocity is 100m/s, so you have to simulate at least 0.4 sec physical time. With timestep = 0.001 -> it will be 400 time step. Also check Cd and Cl best regards

August 30, 2019, 04:10		#6
Fabio88 Member Fabio Malizia Join Date: May 2010 Location: Leuven (Belgium) Posts: 51 Rep Power: 15	Try to change P-V coupling (ex. Coupled). Might help.