[zineb elkhaldi]

Hi All
I want to simulate fluid flow around a complete helicopter which is connected to its rotors (i mean rotate and translate movments)using ANSYS CFX. So i would like to aske how introduce regid body conditions. I used 3 domains seperated by interfaces, 3 regid bodies, and 3 sub-domains: One is contained fueslage wich is stationary, the second wich contained the main rotor (rotate about axe "y"), and the last contained tail rotor wich rotate about axe "x". The problem is that due to the interfaces.which condition must be used?

[zineb elkhaldi]

Each time I recieved this message
| ERROR #002100012 has occurred in subroutine cVolSec. |
| Message: |
| A negative ELEMENT volume has been detected. This is a fatal |
| error and execution will be terminated. The location of the first |
| negative volume is reported below. |
| Volume : -0.2675E-05 |
| Location : ( -0.75393E-01, -0.10408E+00, 0.19123E+01)

[mrkmrk]

Hi

This error indicates some of your computational domain elements are highly distorted. In order to obviate this problem, at first you have to see what is happening to your mesh. To do this, I suggest to re-run the simulation by a smaller time step size and store results in every time step. After that the solution is diverged, investigate the results of last time step to find the place with the highest amount of elements distortion. Then try to improve the mesh quality in that place and re-run the simulation by a smaller time step size. If this doesn't work, then try to use different values for the elements smoothing parameters.

[ghorrocks]

First of all - this question is unrelated to the thread it was originally posted on. Please start a new thread for a new question. I have moved this to a new thread.

Regarding your question: There is an FAQ on this: https://www.cfd-online.com/Wiki/Ansy..._went_wrong.3F

[urosgrivc]

Dont use rigid body for a helicopter.
Do you understand vhat it can be used for?
What are you triing to simulate?.. as it might be apropriate but i highly doubt.

If you want transverse motion and rotor rotation best thing to do is probably
a rotating domain for the rotor and put everithing in a ˇwind tunelˇ like scenario. but there are problems as this will only work properly in transient because of the conection betveen rotating and stationary domain. so the best thing to do is make a steady state first than inicialize transient with it.
Hope you have some large cpu power to back this up properly.

[zineb elkhaldi]

I try to simulate autorotation movement of an helicopter.

[ghorrocks]

Can this be simulated by modelling the helicopter in a range of speeds and rotor speeds to establish a performance map for the system, then do a simple ODE solver to model the helicopter motion using the performance map? This is a far simpler way of doing it, if it is possible.

[AtoHM]

Autorotation, from what I believe to know, refers to the state where the engines are shut down (power outage, ..) and the helicopter is "falling". Since air is still flowing through the rotor, it still produces some lift and I think it is still possible to do a bumpy but "save" landing.

The wake of the fuselage to the rotor is dependend on the angle of declination of the helicopter, which is also part of the solution I think, so I am not sure if neglecting it would be accurate enough.

[urosgrivc]

I think you could predict the torque which is caused on the rotor by the falling helicopter (maybe at variable pitch angles asveal).
But this is wery complex as infinite scenarios are posible.
you need to somehov simplify it.
And inetria of the engine and rotor probably plays a big roal asveal.
as energy stored from falling is in the form of inertia and gets used up wrihgt before landing.

[ghorrocks]

The wake of the fuselage to the rotor is dependend on the angle of declination of the helicopter, which is also part of the solution I think

I never said you had to ignore the fuselage. If you think the fuselage contributes then include it in the performance map simulations.

Can I ask why you are trying to model autorotation in a helicopter? What are you trying to learn from the simulation?

[AtoHM]

Yes you didnt sorry. Your suggestion is probably the best way to go for this.

The point of the simulation, I think, is that autorotation is a stable unique flight regime. Assuming no changing wind conditions, the helicopter should go into a constant decline and rotor speed should become constant and all forces and moments balance each other out. I think he wants to determine exactly these conditions - falling speed, rotor speed, declination angle. Correct me if I am wrong.

[ghorrocks]

Do you already have performance data for the blades? If not, can you estimate it from the aerofoil type and public domain data on aerofoil performance?

[zineb elkhaldi]

Thank you so much for your responses and your suggestion. I need to learn or to know how to balance the weight of the helicopter or to decrease the lift and thrust

[ghorrocks]

This is not really a suitable model for CFD. In fact if you just want to match up the mass to lift then this is a simple force balance. No need for CFD.

[AtoHM]

I wrote a multibody-simulation a while ago for a multirotor drone. It involved solving this kinds of equations. I created a matlab script to solve the ODE system. This however needs some knowledge of MBS. The aerodynamic model I used was based on the books by Seddon http://onlinelibrary.wiley.com/book/.../9781119994114 and Leishmann https://books.google.de/books/about/...4C&redir_esc=y

I think both contain sections about autorotation and how it can be determined. As Glenn suggests, this is rather a load-balance problem than for CFD. However, it would be interesting to validate results from those analytical models using CFD for example getting drag coefficients for the body and blades depending on flow direction und enhancing accuracy.