[gigili206]

hi everybody
I recently simulated a wind turbine via cfx but the resulting streamlines seems to be wrong for they just rotate in rotating region and they stop rotating as they enter the stationary zone.
I defined two regions stationary and rotating with frozen rotor interface between them.
does anybody knows whats wrong with my simulation?
thanks

[gigili206]

nobody knows what should i do? I'm really stuck I cannot find whats the problem

[ghorrocks]

FAQ: http://www.cfd-online.com/Wiki/Ansys...f_reference.3F

[gigili206]

Originally Posted by ghorrocks

FAQ: http://www.cfd-online.com/Wiki/Ansys...f_reference.3F

thank you so much. sorry I asked a FAQ.

[gigili206]

dear all
as I read in FAQ I used 'velocity in stn frame' for demonstrating streamlines in my turbine and I got this:


while I was supposed to get a thing like this:


it seems my turbine doesn't rotate at all. I expected to get the wake behind the turbine. should i run transient instead of steady state to get the wake behind the turbine? are my streamlines correct in the above picture for a rotating turbine?

[gigili206]

does anybody knows haw can i get a wake demonstration like the below picture from my simulation?



is such a thing obtainable from an steady state simulation with frozen rotor approach?

[ghorrocks]

As you say, the frozen rotor approach will definitely not be able to capture flows like what you show. Frozen rotor is, in effect, a steady state simulation with the rotation taken into account at the rotor position shown. You can get the flows you show using a transient simulation but they are very expensive. I am no expert on the other options and another frame change option may be more appropriate. I would read the documentation on multiple frames of reference carefully to see what the appropriate model will be.

[gigili206]

Thanks a lot ghorrocks
I'll check the cfx documentations to find an appropriate substitution way like multiple frames of reference as you said in order to avoid transient simulation. but if i couldn't find any way i think i should run a transient simulation.

[gigili206]

dear all
I continued my simulation with MFR but when i want to get the shaft torque its always negative.
this is the expression which i used:

torque_z()@blade surface

is this the right way to get the the power by multiplying the shaft torque calculated by above expression by rotor speed?
and also why the torque is negative (i mean by RH rule its CW while the rotor rotates CCW)?

[ghorrocks]

If you are miles from the correct operating point then the torque can be negative. Are you sure you are at the correct operating point?

Alternately your simulation could be very inaccurate: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[gigili206]

i know that it might be a silly question but as you said my simulation might be too far from operating point i got serious misgiving about the configuration which i should use. i mean the direction of wind and also the pitch angle. if we assume the rotor rotates on -y direction in all the below pictures which one is the right wind direction and pitch angle (I've exaggerated the pith angle value)
1

2

3

4


sorry that my questions are so silly

[ghorrocks]

1 and 3 might generate positive power, 2 and 4 will definitely generate negative power. For 1 and 3 to generate positive power you will also need to make sure you are near the operating point. If you are miles away (ie rotor too fast or wind too slow) 1 and 3 will generate negative power as well.

And don't forget that because your rotation is in the negative direction the torque for positive power will be reported as a negative torque about the y axis. So the positive power comes from torque x omega where both torque and omega are negative so the power is positive.

[smarika]

i could get pathlines like yours. i showed the pathlines in turbine blades but its just like in the attached picture. i dont know where is m mistake.larger_domain_vel20_m500rpm.jpg

pathlines of velocity.jpg

velocity countor.jpg

Originally Posted by gigili206

dear all
as I read in FAQ I used 'velocity in stn frame' for demonstrating streamlines in my turbine and I got this:


while I was supposed to get a thing like this:


it seems my turbine doesn't rotate at all. I expected to get the wake behind the turbine. should i run transient instead of steady state to get the wake behind the turbine? are my streamlines correct in the above picture for a rotating turbine?

[smarika]

hi ghorrocks. can u tell me if my meshed turbine is correct or not. i have tried to used mrf.

Originally Posted by ghorrocks

If you are miles from the correct operating point then the torque can be negative. Are you sure you are at the correct operating point?

Alternately your simulation could be very inaccurate: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[ghorrocks]

You appear to be using Fluent. This is the CFX forum. Try the fluent forum.

[smarika]

i have used ansys fluent.
i created the mesh in gambit, and in ansys fluent i have used the followings:
1.pressure based, absolute, steady
2.k-epsilon mode
3. two cell zones. one that includes the turbine and other the domain
the zone including fan is mrf and rotating in -500 rpm in -y direction
4. turbine is moving wall, also rotating in relative to adj. cell zone at -500 rpm
5. solution method is simple schemed.
6. calculation initialized from velocity inlet (10m/s in y direction)

i can not understand why my pathlines are only around the turbine. not in behind them . what parameters should i set to get better results..

Originally Posted by ghorrocks

You appear to be using Fluent. This is the CFX forum. Try the fluent forum.

pathlines of velocity.jpg

larger_domain_vel20_m500rpm.jpg

vel_mag.jpg

vel pathlines.jpg

mesh.jpg

[ghorrocks]

I repeat - this is the CFX forum. Try the fluent forum for questions on Fluent.

But I can see you have not run the simulation for long enough to establish the flow. You need to run it longer.

[miquiita]

Hi how did you do your interface of the rotor and the domain of the rotor? or did you do the interface of the rotor domain and the bigger cylinder domain?

So thankful for answer.

Michelle