[Johndm]

I have set up a water (fan type) turbine in a body of flowing water (a pipeline).
I have set up the water flow rate at 10m/s and the turbine rotation at 60rpm.
The study converged quickly but I'm not sure the rotation speed matches what would occur at 10 m/s flow.
How do I check if there is imbalance between the input (flow rate) and the output (fan rotation)?
Also, once balance is obtained, how do I get a value of the turbine torque generated?
Please see ansys workbench file attached here: https://drive.google.com/file/d/0B0a...ew?usp=sharing

[maximilianomavica]

Hi Johndm, did you solve it? How can you predict the rotation speed or rate of the model?

[ghorrocks]

The usual way to do this is:
* Do a series of steady state models covering the possible range of rotation speeds, maybe 20rpm to 100rpm.
* Extract from these simulations the torque generated by the rotor. This gives you the turbine curve, which is a function of torque versus rotation speed.
* Have a look at the system connected to the turbine. That might be a generator or it might be some other load. It will have a torque required to drive as well, and that torque might change versus rotation speed.
* Extract from the system the torque versus speed curve of the system.
* You now have two curves: the turbine curve and the system curve. The operating point will normally be where these two curves intersect. That is where the turbine torque generated equals the system torque absorbed.

As you can see, doing CFD is only part of the approach. You then need to plot it out and find the intersection point. And also note that the system it attaches to is necessary as well - which is obvious when you think about it.

[maximilianomavica]

Hi Ghorroks, what happens if you need to know the rotation speed of the propeller. Let's assume that there is no system connected.

[ghorrocks]

The same concept applies: If the system torque is zero, then the turbine will run when the turbine torque is zero. In other words, when it is running slower it will generate a positive torque which speeds it up, and when it is running over speed it will generate a negative torque which will slow it down. So the zero net torque point is where it will end up running steady state.

[maximilianomavica]

So if I applied an inlet velocity to the fluid, then make a series of simulations with different rpm and evaluated the torque for each case, where it is zero it should be the correct rpm?

[ghorrocks]

The approach I described assumes that the simulation approach you are using is appropriate and that the only factor you are trying to determine is the rotor speed for the given conditions. Providing this is the case then yes, your question is correct.