CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   VAWT mesh in gambit (http://www.cfd-online.com/Forums/main/16201-vawt-mesh-gambit.html)

Jonathan January 13, 2009 13:13

VAWT mesh in gambit
 
Hello,

i'm trying to do a simulation with fluent of a wind turbine with vertical axis. It's a H-Darrieus-Rotor. I've done a 2d mesh with the section of the 4 blades. They're in a circular face, set as rotating face.

The problem is that the results give a poor momentum, and I think it is because it doesn't calculate the momentum of the lift force, just the momentum of the drag force.

Can anyone help me?

Thank you very much

Bikini Girl with Machinegun January 16, 2009 01:52

Re: VAWT mesh in gambit
 
At what tipspeed ratio does it give poor results? VAWT have known problems at TSR around 1.

Does Fluent automatically include dynamic stall effects which are known to be important for VAWT. (For example, the angles of attack can change from -180 to 180 degrees at TSR=1).

Incidentally, why are you using 4 blades? High solidity means that you might be sacrificing a lot of efficiency to get high starting torque (if that's the reason for going to 4 blades instead of 3).


Jonathan January 17, 2009 13:23

Re: VAWT mesh in gambit
 
it actually gives poor results at every TSR i've tried so far.

I'm doing this as my final project for my degree, and the aim is to try different combinations with different type of blades, angle and number of blades. We have prepared a progam in Delphi that allows us to start the simulation at one TSR and when it's finished, it starts automatically another one, so that in the end we have the whole efficiency curve. And by now all the results have been very dissapointing.

About the dynamic stall effects, i don't now if Fluent automatically includes them.

Thanks for your answer. Jonathan

Bak_Flow January 19, 2009 15:00

Re: VAWT mesh in gambit
 
Hi Jonathan,

what method are you using to solve the multiple frame of reference problem...sliding mesh or MRF?

What makes you think that lift forces are not calculated? In general in aerodynamic problems the lift force is larger and easier to calculate accurately.

So, I am a bit surprised that Fluent has not calculated this appropriately?

What is your meshing strategy?

Give us some more details and maybe we can help.

Regards,

Bak_Flow

Jonathan January 22, 2009 09:18

Re: VAWT mesh in gambit
 
The way i'm doing it is the following:

there's a circular area with centre in 0,0. to this circular area i substrac the areas of the blades and set it as a sliding mesh.

there's a bigger area with a square form. in the middle of this, i substract a circle as big as the other area. i set it as a stationary area.

in this second area i set the link site as velocity inlet and the other three as pressure outup. but i think i should set the link, upper and lower sites as velocity input and only the right site as pressure output (i'm not very sure of this though)

then in the calculation the rotational speed of the sliding mesh is changed several times so i get the whole efficiency curve.

for the viscous flowing, i select k-epsylon.

the solver is set as segregatec, implicit, unsteady and 2nd order implicit.

with these settings, i've tried a savonius rotor and the results are not bad. That's why i thought that the problem appears when calculating the lift force momentum.

Hope anyone can help.

Thanks


All times are GMT -4. The time now is 00:38.