How to calculate effect of force on single joint 2 body system

quarkz · February 29, 2016, 23:17

Hi,

I am trying to simulate the effect of a varying point force on a single joint 2 body system, as shown in the attached image.

http://pasteboard.co/1VOPbkXy.jpg

There are 2 bodies and the thin plate is massless. It will flap and generate force in the fluid. I have no problem with the CFD part.

But how do I calculate the effect of the instantaneous force on the 2 bodies? I am looking into rigid body dynamics and multibodies dynamics. Is there a good source which deals with this exact problem?

Also, may I know if there's any Fortran code available to solve this type of system? Hope I don't have to write a new subroutine if possible.

Thanks.

pivalse · March 1, 2016, 09:09

The CFD portion of your calculation returns the pressure distribution of the fluid acting on the surface of your multi-body system.
The integrals of the pressure over the surface for each single-body element returns a force and a moment.
If your multi-body system is composed of rigid single-body elements, you apply the force and the moment to the center of mass of each.
If the single-body elements are not rigid, then this does not work anymore and you will have to use FSI (Fluid Structure Interaction), whereby CFD is coupled with a FEA code that calculates the deformation of the body at each time step.

Did I understand your question correctly?

quarkz · March 2, 2016, 00:00

Ya thanks pivalse, that's correct. They can be assumed rigid.

Body 1 and 2 are connected by a link joint and they can rotate about one another. Also later on I need to add more bodies.

Do I form 2 eqns for each body for solve them as a linear set of equations simultaneously?

Thanks.

pivalse · March 2, 2016, 09:32

The mechanical system you are describing is made of 2 bodies that are linked. In 2D (as it seems to be the case) each body has 3 degrees of freedom (2 translations and 1 rotation). The link between them represents a "lock" on one or more degrees of freedom (depending on the type of link).

In your case your total degrees of freedom are:
2 bodies x 3 DOG - 1 link = 5 unknowns

This means you will need 5 equations that relate force (or moment) to motion.

pivalse · March 2, 2016, 15:44

I forgot to ask if your system is tethered to any reference system or if it is freely floating.

quarkz · March 10, 2016, 23:29

Hi pivalse, it's freely rotating. So does it fit what you've described?

Thanks

pivalse · March 12, 2016, 19:43

Yes it does, but let me better define the terms even though I imagine you already know all this.
If you have one rigid body floating in a 2D space, it has three degrees of freedom (U_x,U_y, and Omega_z). This means that if the body is subjected to an unbalanced force, it will accelerate indefinitely in the direction of the force vector and it will start rotating if the force vector does not go through the center of gravity of the body.
For objects floating like the one you described, there is no obvious frame of reference such as, for example, the ground. If it was a single rigid body, you would use its center of gravity as the reference system and impose two or three boundary conditions such as a pinned point (U_x=U_y=0) or an encastre (U_x=U_y=Omega_z=0).
The effects of the integral of the surface forces acting on the body would result in a force vector and a moment that are balanced by the reaction force and moment at the center of gravity.
In the case of a multi-body system, the choice of the reference system is not so obvious and depends on what you are trying to calculate.
I hope this adds to clarity.

February 29, 2016, 23:17	How to calculate effect of force on single joint 2 body system	#1
quarkz Senior Member TWB Join Date: Mar 2009 Posts: 401 Rep Power: 19	Hi, I am trying to simulate the effect of a varying point force on a single joint 2 body system, as shown in the attached image. http://pasteboard.co/1VOPbkXy.jpg There are 2 bodies and the thin plate is massless. It will flap and generate force in the fluid. I have no problem with the CFD part. But how do I calculate the effect of the instantaneous force on the 2 bodies? I am looking into rigid body dynamics and multibodies dynamics. Is there a good source which deals with this exact problem? Also, may I know if there's any Fortran code available to solve this type of system? Hope I don't have to write a new subroutine if possible. Thanks.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
force and rigid body	armin najarian	CFX	1	August 20, 2015 13:46
rigid body force	armin najarian	CFX	5	August 2, 2015 07:31
Force can not converge	colopolo	CFX	13	October 4, 2011 22:03
Calculate Drag force for flat plate	vsun	FLUENT	0	October 3, 2010 07:56
Problems with SUPG body force term	FEM question	Main CFD Forum	0	January 21, 2006 17:51

March 1, 2016, 09:09		#2
pivalse New Member Join Date: Apr 2014 Posts: 11 Rep Power: 12	The CFD portion of your calculation returns the pressure distribution of the fluid acting on the surface of your multi-body system. The integrals of the pressure over the surface for each single-body element returns a force and a moment. If your multi-body system is composed of rigid single-body elements, you apply the force and the moment to the center of mass of each. If the single-body elements are not rigid, then this does not work anymore and you will have to use FSI (Fluid Structure Interaction), whereby CFD is coupled with a FEA code that calculates the deformation of the body at each time step. Did I understand your question correctly?

March 2, 2016, 00:00		#3
quarkz Senior Member TWB Join Date: Mar 2009 Posts: 401 Rep Power: 19	Ya thanks pivalse, that's correct. They can be assumed rigid. Body 1 and 2 are connected by a link joint and they can rotate about one another. Also later on I need to add more bodies. Do I form 2 eqns for each body for solve them as a linear set of equations simultaneously? Thanks.

March 2, 2016, 09:32		#4
pivalse New Member Join Date: Apr 2014 Posts: 11 Rep Power: 12	The mechanical system you are describing is made of 2 bodies that are linked. In 2D (as it seems to be the case) each body has 3 degrees of freedom (2 translations and 1 rotation). The link between them represents a "lock" on one or more degrees of freedom (depending on the type of link). In your case your total degrees of freedom are: 2 bodies x 3 DOG - 1 link = 5 unknowns This means you will need 5 equations that relate force (or moment) to motion.

March 2, 2016, 15:44		#5
pivalse New Member Join Date: Apr 2014 Posts: 11 Rep Power: 12	I forgot to ask if your system is tethered to any reference system or if it is freely floating.

March 10, 2016, 23:29		#6
quarkz Senior Member TWB Join Date: Mar 2009 Posts: 401 Rep Power: 19	Hi pivalse, it's freely rotating. So does it fit what you've described? Thanks

March 12, 2016, 19:43		#7
pivalse New Member Join Date: Apr 2014 Posts: 11 Rep Power: 12	Yes it does, but let me better define the terms even though I imagine you already know all this. If you have one rigid body floating in a 2D space, it has three degrees of freedom (U_x,U_y, and Omega_z). This means that if the body is subjected to an unbalanced force, it will accelerate indefinitely in the direction of the force vector and it will start rotating if the force vector does not go through the center of gravity of the body. For objects floating like the one you described, there is no obvious frame of reference such as, for example, the ground. If it was a single rigid body, you would use its center of gravity as the reference system and impose two or three boundary conditions such as a pinned point (U_x=U_y=0) or an encastre (U_x=U_y=Omega_z=0). The effects of the integral of the surface forces acting on the body would result in a force vector and a moment that are balanced by the reaction force and moment at the center of gravity. In the case of a multi-body system, the choice of the reference system is not so obvious and depends on what you are trying to calculate. I hope this adds to clarity.