HAWT 3D Simulation Fluent

  • Last Post 22 April 2020
sid10mahesh posted this 18 April 2020

Hi guys,

I'am new to this kind of simulation. I have done a simulation of laminar flow past a scaled model of HAWT having a reynold's number of 180 using sliding mesh approach. Could guys please tell me where iam wrong because iam getting a very very low value of Cd,Cl and Cm(Close to zero).

I will tell you the list of options that i have chosen in the fluent.

Mesh Interface:  It has found out the contact on its own and since it coincides there i have given a matching interface condition.

Cell zone conditions

  • Rotor- mesh motion with axis of rotation
  • Stator- default an axis of rotation will be coming i have set it to 0

Boundary Conditions:

  • Velocity inlet with 10m/s
  • Outlet as pressure boundary conditon(atmospheric pressure)
  • Top and side walls with zero shear stress boundary condition(unconfined flow)
  • Bottom wall with no slip boundary conditions
  • Interface surrounding the blades and blades as moving wall with desired rotational velocity.
  • Rest everything as wall with no slip conditions.

Reference values

  • I have set the reference value to be calculated from inlet wall with inlet area
  • Reference length which is thickness of the enclosure etc

Solution Methods:

  • SimpleC- is only working rest all are giving floating point exception as error after very few iterations itself.

Initialization: Hybrid Initialization


Tell me what part iam missing?

Order By: Standard | Newest | Votes
Kremella posted this 21 April 2020


How deep is your convergence? Could you please provide your residual plot?

Also, are you specifying the correct rotational velocity? Have you defined the axis of rotation correctly?

What drag and lift values do you expect from this scaled model?

Thank you.





sid10mahesh posted this 21 April 2020

Im doing my simulation on a scaled model(200:1). For the orginal model the rotational speed is found to be 54rpm. So im using the same value here.

Yes the turbine is rotating when i do the velocity contour animation. Its working correctly, the rotational axis is correctly defined.

Why iam  I getting floating point exception error for certain solution methods when everything looks fine?

I should get a more drag or lift value compared to this right? Its very small!!!

Also can you tell me how to get the Cp vs lambda curve for this setup since it has a varying airfoil profile?

Please help me!

Kremella posted this 21 April 2020

Yeah, it doesn't look like your problem is converging well. What is your mesh quality? Is the orthogonal quality > 0.1?

Also, have you tried using the Coupled solver with the psuedo-transient setting to converge your problem?

Thank you.


sid10mahesh posted this 21 April 2020

Thank you Karthik for your valuable suggestion.

Yes its found to be less than 0.1. I shall re-mesh the body and will get back to you.

No i haven't tried out Coupled solver. But im doing a transient simulation. Do you mean that i should go for steady flow and then go fro psuedo transient?

How to get the Cp value for this varying airfoil profile?


Kremella posted this 22 April 2020


Yes, please improve your mesh. Please make sure that your starting mesh has a minimum orthogonal quality measure > 0.1 - 0.15 and your maximum skewness measure < 0.85 - 0.9.

I did not realize that you were running a transient simulation. Please ignore my comment about using pseudo-transient. This is for a steady solution. In your case, your continuity residual does not seem like it is converging every time step. Please reduce your time-step such that your CFL criterion is met. Please make sure you are running a sufficient number of iterations to ensure that the simulation is converging every time-step. At the end of each time-step, your set convergence criteria must be met and the simulation should converge.

Thank you.