Actuator disk

[Anonymous]

Hi all,

I am pretty new to CFD and I made my first steps by myself using Autodesk's tutorials. I'm doing this CFD work for my master thesis.

But now I have a few questions about actuator disks.

I want to model a ducted tidal turbine, to see how the flow behaves for different positions in the duct. I'm not looking for any value of energy extraction and stuff like that. Also, I have some power and thrust curves of a turbine.

Is the actuator disk method relevant to my problem or is there something more simple to do ?

If the actuator disk is the proper method, how can I implement this into Autodesk CFD ?

Thanks a lot,

Thomas

[Jon.Wilde]

Hi,

If you just want to apply a power curve - use an internal fan material and just model a solid cylinder in CAD. This is likely the simplest solution.

Thanks,

Jon

[Anonymous]

Hi,

Thanks for your answer.

I have read that I can also use a resistance material, and assign a pressure drop to that.

What would be the most appropriate solution ?

Thanks a lot,

Thomas

[Jon.Wilde]

Hi Thomas,

Are you trying to model a fan or a resistance? If you want the volume to drive the flow from it, you will need a fan, if you just want to provide a pressure drop in a system, use a resistace.

Thanks,

Jon

[Anonymous]

Hi,

I am modeling a tidal turbine in a duct that provides a venturi effect, and I want to see how the flow behaves according to different position of the turbine in the duct. According to existing papers that I read, the resistance material associated with the pressure drop seems to be the correct tool to use.

Thanks a lot for your quick answers.

[Jon.Wilde]

Hi Thomas,

No problem at all, pleased to be able to help.

I have not read the papers you have but I would think that a resistance would demonstrate a pressure drop over it's volume and a fan would induce a pressure rise from the inlet to outlet. 

Best of luck with the analysis!

Jon

[Anonymous]

Hi Jon,

Ok so now I am confused.

What would be the difference between the two methods, as it appears to me that both can be applyed to what I am doing ?

In what case I should choose to model the turbine with a pressure drop, and in what case it would be more sensible to do it with a fan model ?

Thanks,

Thomas

[Jon.Wilde]

They are representing two different things.

The internal fan is designed to represent a fan or impeller and will drive the flow, creating swirl if you choose and also raising the pressure as it passes.

The resistance is designed to represent a filter, so this will cause a pressure drop as the fluid passes.

Hope that helps you decide.

Thanks,

Jon

[Anonymous]

Hi,

From what you explained to me, the Fan tool would be more likely useful to model the swirls produced by a turbine, whereas the resistance material would be much better to model the energy removal.

Following this, I have decided to use the resistance material, which model my pressure drop (the resistant factor is basically the drag coefficient of the studyed turbine).

Now I have convergence problems, and no idea where it comes from. I modelised the resistance disk as an extruded disk as thin as i can on inventor, and used the correction tools of CFD to suppress faces/edges. What am I doing wrong and what can I do to make my model converging ?

Thanks a lot,

Thomas

[Anonymous]

Okay I found a source of error.

For my study, I want to model a tidal turbine, its blades (my actuator disk), its support structure and a venturi duct around it.I created an assembly of these 3 components.

And from this assembly I build the "water domain" by using the simplify tool of Inventor to create a simplify part of my assembly. Then I open this part, extrude a big rectangle that will be my water domain and combine it with the simplified part to remove the shape of the simplified assembly from the water part. Then I reopen my first assembly, add the water and open the whole assembly into CFD.

But when it's opened into CFD, There is a part that appears called actuator_disk_U_turbine_body.

WOuld you have any suggestion of where it can be from, and how can I suppress it ? Is it because I build my actuator disk from a projected surface of the turbine ? Do I have to fully constrain my disk before opening the whole thing into CFD ?

Thanks a lot,

Thomas

(PS: I am sorry with all my questions but nobody can help me in my university. Also, I will credit you in my dissertation.:) )

[Jon.Wilde · Accepted solution]

No problem at all 🙂

I am certain there is a far easier way to do this in one go, we should not need to go backwards and forwards.

Are you going to have a large external domain of water around this structure yes? It does look like it.

If so, just draw this in Inventor, overlapping with your parts and launch into CFD. Done.

You can suppress the solids from the mesh in CFD - same result with zero effortt 😉

Also - do not model the resistance/turbine as a thin volume. This is asking for meshing trouble - model it roughly the thickness of the swept blade area. You will need to give this a good uniform mesh with 4-5 elements from the inlet face to the outlet. Making it thin is going to make this impossible 🙂

Some images would be really useful if you are able to share them.

Hope that helps,

Jon

[Anonymous]

Awesome !

So here is the assembly opened into CFD.

Is the suppression of solids from the mesh directly done, has it considers the junction of the water and the parts of my assembly as a single part ?

That really helps !

Thomas

[Jon.Wilde]

Hi,

Yes - much easier.

You should be able to suppress any solid parts from the mesh, leaving the resistance/turbine and water to be meshed.

Ought to work well - also extend your outlet a little, it's a bit short 😉

Thanks,

Jon

[Anonymous]

Hi,

Thanks a lot for your help and advices.

Unfortunately, I have now another problem: My computer died this morning so I reinstalled CFD and Inventor on a computer in my university.

As I don't have the administrator rights, I got some restrictions, but with an administrator we set up properly the rights ( we allowed the CFD server service in the administrative tools, as explained by the software.)

But I can't run any simulation. As soon as I click on Solve, I get a message saying that the solver has exited. I tried with different simulations and nothing change.

Can it be linked with another admnistrative right with the solver ? When I click on "more information" it doesn't help, it just send me on the Solver manager help page online.

I hope I will be able to fix my computer at the begining of next week but if not, it would be great for me to be able to work on the university's computer.

Thanks a lot again,

Thomas

[Jon.Wilde]

Hi Thomas,

Can I recommend logging questions as a new thread - otherwise things can be missed.

Try changing the properties of the CFD installation folder (similar to below) so that the 'user' account has full security rights:

Check the anti virus is not blocking any of the executables from the same folder also.

Thanks,

Jon

[Anonymous]

Ok I will do that, thanks a lot Jon !

Thomas