[Attesz]

Hi all,

I'm simulating a quadrotor RC helicopters propeller:
http://www.cfd-online.com/Forums/cfx...imulation.html

The main goal is to find the best blade shape. To get this, I have to measure thrust in CFX post.

Now, I define a plane behind the rotor, and let the CFX calculate the normal force on it. The CFD forces are smaller than measured. I know, it can caused by a lot of things, but maybe I'm wrong in force calculation.

So my question is: how to measure the thrust of the propeller?

Thanks,
Attesz

[ghorrocks]

Most people are more interested in the thrust from the rotor, not a plate underneath the rotor. Why don't you calculate the thrust on the rotor blades?

[Attesz]

Hi Glenn,

I've calculated normal force on the rotor blades, but the values are lower than on the plane measured, so more lower than real. I don't know why?

I think, CFX calculates the force on a plane from the equation massflow*velocity, which is the general equation for thrust calculation. To this equation comes the pressure difference member: area*(pressure@outlet-pressure@environment), which has a low value in my work.

Thanks for your reply,
Attesz

[ghorrocks]

So let me paraphrase you: When you put in a feature which has nothing to do with the rotors the force acting on it is closer to what you think it should be so therefore you will use this force rather than the force on the rotors which is actually what you want.

Pardon my sarcasm but I think this line of logic is not helpful. If the force you want seems wrong then you should look at the accuracy of your simulation rather than doing some fudge on the results to make it appear better than it is.

Calculate the force on the rotor using the forceX/Y/Z() function, and if the answer is wrong then you have an error in your simulation and you should fix it.

[Attesz]

Thank you Glenn,

but I didn't want to cheat, I didn't know how to measure force simply. I tought measuring that is more difficult, than measure it on the blades.
However, I'll examine the mesh, but also the measuring of the real quadrotor can be inaccurate (ground effect, the effect of rotors to each other, and also the "measuring device": a kitchen balance . it's a student team project).

Best regards,
Attesz

[armanpournasiri]

Hi all
Im working on a propeller and i have to calculate thrust in cfx but i dont know from where I am start .can you help me?

[urosgrivc]

You are probably looking at something like a rotating domain problem.
depends on your case, let say, if you have some other parts blocking the flow which could produce downforce and not lift than you need to include those surfaces too and measure force on these also.
You can probably include simetry which will aid speed, resolution and convergence to your problem.

-make a geometry of the surrounding air (you most probably do not need a propeler volume in your simulation just air that surrounds it).
-make a nice mesh, finer near the propeler surfaces maybe some inflation layers.
-rotate it
-Force_direction@propeler surface , is your ansver for force generated by the propeler.

[armanpournasiri]

hello
thank you.
You can also help me with the mesh.please?
i think my mesh is not well.but my geometry is true.
I must also say that I am studying the problem about surface piercing propeller.

[urosgrivc]

The mesh should be very fine near the wals if you have high velocity you vill find out that mesh quality and density will play a big role in geting the corect result. This case should be done in tranzient if both the rotor and the body of the quadrotor will be done in the same simulation but be prepeared for wery long computational time. Consider doing rotor simulation seperately from the body simulation

armanpournasiri this is probably your ansver.
surface piercing propeller? you mean vater surface like half the rotor outside and half inside the water?
this can be done by activating multiple fluids in a single domain and activating gravity maybe even additional surface tension model. I can only see this vorking in transient though.

[armanpournasiri]

Hello and thanks for your help.
In the case of the mesh, I must say that my mesh aspec retio is 68.52 which is less than 100 .but my skwnes is .99 that i think it do not fit.
also about On the topic that you guessed right.
propeller works between water and air .
my next question is :do i need to udf in my question? becuse i I do not know programming.

[ghorrocks]

Mesh quality requirements vary simulation to simulation. Some simulations require very strict mesh quality, some are much looser. The numbers in the output file are just guidelines.

Please show an image of what you are modelling and how you propose to model it.

[urosgrivc]

I can imagine only few basic expresions being used in the surface piercing propeller simulation.
maybe for initial conditions of water location, and for force or moment monitors.
But I think this kind of simulation is only posible to do in transient and it will probably include rotating and stationary domain.
explain more

[satishorion]

Hi,
I am working on ducted propeller problem. I have performed steady RANS simulation using ANSYS Fluent. I am trying to estimate the thrust produced using two approaches - 1. Standard force output (drag) on the prop blades
2. Momentum balance between inlet and outlet planes sufficiently far from the prop plane.

The results for different in both the cases. What should be done in this case. The mesh is pretty good with boundary layer around the walls.

PS: The mass flow rate variation at the planes are less than 3%. Although the massflow at the domain inlet and outlet are fully balanced.

Regards
Satish

[ghorrocks]

This is the CFX forum. Try the Fluent forum for details specific to Fluent.

But your question is a general CFD one of accuracy. Have a look at the FAQ: https://www.cfd-online.com/Wiki/Ansy..._inaccurate.3F

In short:
* Show that your results are fully converged
* Show that your mesh is fine enough by a mesh refinement study
* If transient, show your time step is fine enough
* Confirm your physics model is correct

A few other points: You say "The mesh is pretty good" - on what basis do you say that? Did you actually measure your mesh accuracy or are you just guessing?

And: The thrust of a ducted propeller does not come from the drag. So if this does not compare to the thrust from the inlet/outlet momentum balance then it could be because you are looking at the wrong thing.