[eSKa]

Hello,

I´ve got problems with the calculating of the force of my fan (metro).
F= massFlow * velocity

areaAve(Velocity w)@Outlet ---40,79 m s^-1

massFlow()@Outlet ---60,07 kg s^-1

massFlow()@Outlet * areaAve(Velocity w)@Outlet ---2450,37 N

force_z()@Outlet --- 2254,37 N

I´m wondering about the difference of 200N. It would be great to know how the expression force_z()@ is defined.
There is no hint in de Ansys Post manual, or something else in the internet.

Can me pls someone help

Thanks


eska

[ghorrocks]

Why are you calculating the force on the outlet? Don't you want the force on the fan blades? So shouldn't you be calculating force_z()@WallBlades?

[eSKa]

Hi,

thanks for the answer.


I want to calculate the force of the whole fan. So I think I need it at the Outlet?
I ve got a value of a metro fan (experimentel) which I have to validate.



eska

[ghorrocks]

The fluid in the outlet does not touch the fan. So how can it apply a force to the fan? It is the fluid next to the fan which applies the force to the fan. So what is the point of measuring the force at the outlet? And what does force at the outlet mean anyway?

If you want the force on the whole fan then do force_z()@TheWholeFan. As simple as that. (where "TheWholeFan" is a wall boundary defined as the entire outer surface of the fan.)

[AtoHM]

Or are you looking for thrust?

[eSKa]

yes, I m looking for the thrust.
In den experimental test they installed a meassuring outside of the fan, to calculate the thrust(/the force?). On some pages I saw the word force as a definition for thrust.

[eSKa]

Originally Posted by ghorrocks

The fluid in the outlet does not touch the fan. So how can it apply a force to the fan? It is the fluid next to the fan which applies the force to the fan. So what is the point of measuring the force at the outlet? And what does force at the outlet mean anyway?

If you want the force on the whole fan then do force_z()@TheWholeFan. As simple as that. (where "TheWholeFan" is a wall boundary defined as the entire outer surface of the fan.)

My Outlet is at the end of the fan, but still inside.

So the expression force_z() is in my case wrong?

[Opaque]

Please review the equations you are using, keeping in mind the following

integral of ( a * b) =! integral of (a) * integral of (b)

and recall massFlow(a)@Boundary is an "integral of", similar to areaInt(b)@Boundary

[eSKa]

Im trying to explain it better:

In a company they calculated the thrust of a metro fan, with measuring equipment on the outside of the fan. The fan is standing on a frame with boxes which can measure forces. At least they got the thrust of the hole fan (during the test).

Now I have to simulate the fan, and the target is to get the same thrust.
In the Internet I saw different ways for calculating the thrust, but I have some works where they work it out with this formula:
F(thrust) = massflow * velocity(outlet) = area*density*velocity^2
The massflow is calculated by the same velocity, also at the outlet. This formula works great, I have a thrust with 2450 N, and I need 2221 N.
But now I found the expression force_z(). With this one I got 2250 N. Even better then with my formula. But which is now the right one?

After ur answer (thank you!!) I think the expression force_z() is wrong. Maybe u can give me a hint where the difference could be. It would be great


eska

[eSKa]

Originally Posted by Opaque

Please review the equations you are using, keeping in mind the following

integral of ( a * b) =! integral of (a) * integral of (b)

and recall massFlow(a)@Boundary is an "integral of", similar to areaInt(b)@Boundary

thank you, I will check this now, I m a bit confused, but I found some things in this forum.

[Opaque]

Originally Posted by eSKa

Im trying to explain it better:

...
Now I have to simulate the fan, and the target is to get the same thrust.
In the Internet I saw different ways for calculating the thrust, but I have some works where they work it out with this formula:
F(thrust) = massflow * velocity(outlet) = area*density*velocity^2
The massflow is calculated by the same velocity, also at the outlet. This formula works great, I have a thrust with 2450 N, and I need 2221 N.
But now I found the expression force_z(). With this one I got 2250 N. Even better then with my formula. But which is now the right one?

After ur answer (thank you!!) I think the expression force_z() is wrong

...

eska

Be careful concluding that a simplified version of a formula is better than the general version of the same concept.

The total force on the outlet is by definition an integral, and it cannot be decomposed in the product of the integral of the independent arguments. Can you in the lab measure the independent quantities? Sure you can but it is a lot more work (and money) to do so.

You need to measure a distribution of the quantities of interest in the cross-section, and add them up appropriately.

Which is better for your final goals? Depends on what you want to rely on.

Recall the measurements have an error (this is just on of them), and the simulation has another (truncation error and model error). You must plot them on the same graph with some estimates of their errors (error bars), and you can then start making conclusions.

Hope the above helps,

[eSKa]

thank you for your great answer!

but there is one more question:
what does this expression force_z() mean? can I take it as the thrust or not. where is the difference?

[ghorrocks]

If you use force_z() on a wall boundary it gives you the total force acting on the surface, both pressure force and shear force, in the z direction. I don't know exactly what it does if you ask for force_z() on a fluid region (eg an outlet boundary) and it is unlikely to be meaningful.

This is why I suggested you work out the force on the blade wall boundaries, not the fluid.

[eSKa]

ok, thank you, this helps.

I would like to calculate it on the blade wall, but I have only this one value for the whole fan

[Gert-Jan]

The fastest route would be to define a Surface Group in CFD-Post, put all surfaces of the blades in that group and determine the force on this Surface Group. How difficult can it be?

[Opaque]

Originally Posted by eSKa

...
what does this expression force_z() mean? can I take it as the thrust or not. where is the difference?
...

From the ANSYS CFX Reference Guide:

The force on a boundary is calculated using momentum flow data from the results file, if it is available. The result can be positive or negative, indicating the direction of the force. For non-boundary locators, an approximate force is always calculated.

CFD-Post calculates the approximate force as follows:

If the locator is a wall boundary, the force is equal to the pressure force.

For all other locators, the force is equal to the pressure force plus the mass flow force (due to the advection of momentum).

In all cases, if wall shear data exists in the results file, the viscous force is added to the calculated force.

[eSKa]

Originally Posted by Opaque

From the ANSYS CFX Reference Guide:

Wow! Thank you. Im wondering why my searches couldn t find this explanation. Thank you!!

[eSKa]

Thank you for all the helpful answers. Now I want to present you how I calculated the whole force on the fan.
Because of the experimental value(2221N - measured in a company), which is measured with a frame outside of the fan, with force measurement boxes, the formula
F(thrust)= massflow * velocity
in my eyes is not right.
So I tried it with the expression force_z() (z=flow direction).

10*(force_z()@R1 Blade )+ force_z()@R1 Hub+ force_z()@R1 Shroud+ force_z()@M Blades+ force_z()@M Hub + force_z()@M Shroud + force_z()@S1 Blades+ force_z()@S1 Hub+ force_z()@S1 Shroud+ force_z()@S2 Hub+ force_z()@S2 Shroud + force_z()@K1_wall + force_z()@K2_wall

I added my 10 blades and all wall boundaries to my expression. Im getting 2237N (2221 needed). For me its a great value, but is this calculation in your eyes right?

Would be great to see some oppinions. Thanks

[ghorrocks]

I don't know how you defined your wall boundaries and named them, so I cannot confirm your equation is correct. But if I guess what they represent it looks correct. In short the force measured by your load cell holding the entire fan is the sum of the forces acting on each of the fan elements.

So it looks good to me, and if the result is accurate to your expected value then that is even better.

[evcelica]

"Force_z()@" at non wall surface is as the literature states, pressure and momentum force:
The closest equivalent expression to make this would be
Pressure Force: areaAve(Pressure)@UserSurface
Momentum Force: areaInt(Density * Velocity w^2)@UserSurface. (slightly different than Mass flow * velocity, which is only for a uniform velocity profile)

Opague Quoted:
CFD-Post calculates the approximate force as follows: If the locator is a wall boundary, the force is equal to the pressure force.
I believe it is actually the pressure force plus the wall shear force.

So actually force_z is calculated the same everywhere: Pressure + Wall shear + Momentum. but shear is zero on a non-wall, and momentum is zero at a wall.