CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CCM+ (http://www.cfd-online.com/Forums/star-ccm/)
-   -   How to apply motion to overset region (http://www.cfd-online.com/Forums/star-ccm/127548-how-apply-motion-overset-region.html)

Henry Arrigo December 13, 2013 15:15

How to apply motion to overset region
 
Hello
How is it possible to apply a prescribed unsteady movement to an overset region in STAR-CCM+?

(and by unsteady movement I mean translational motion which has the dimension of m not m/s, cm/s, mm/s and so forth)

Henry Arrigo December 20, 2013 14:24

anybody has any idea?

ggulgulia January 2, 2014 03:23

Hello Henry

It is possible to apply a user defined motion, either by activating the DFBI Translation, Rotation in the 'Tools' folder or by defining your own field function.

Click the link below :)

http://postimg.org/image/xx3rduv83/http://postimg.org/image/xx3rduv83/

Henry Arrigo January 2, 2014 12:18

Thank you Gulgulia
Translation, rotation tools need rotational/translational velocities to be specified, not the displacement. I am looking for something whereby I would be able to apply rotational/translational displacement to the rigid body.

dsk2 January 2, 2014 16:53

Henry,

Its convenient to be able to specify displacement as opposed to velocities in Tools->Motion, but unfortunately I haven't found a way to do that either. So, what I do is calculate the velocities that give the displacement I desire and use the corresponding velocitiy expressions. Thus I end up with having to to significant math beforehand depending on how complex the trajectory of the moving body is.

Tasha4620 January 22, 2014 11:21

Translating Motion
 
I am a final year engineering student attempting to get a simple bluff body shape to translate vertically to simulate it going over a bump in a track.

I have been reading the previous posts and on the star CCM+ software that I am using under the field functions- motions option the construction tree just says stationary and there is no option of changing it.

Can you suggest the reason or as anyone got any more information about how I could impose motion on my model.

Thanks alot

Tasha

ping January 31, 2014 07:49

Tasha4620 - this is quite simple to do - create your field function of velocity in that direction eg Yvel. then create a motion type translation. click on the .... in the translation velocity and type the field function name in the required component eg $Yvel. then set time to unsteady and apply that motion to the region.

ggulgulia February 12, 2014 02:59

Hey Tasha

Maybe you have not yet activated the Implicit Unsteady model that's why it's showing only Stationary Model. Please check your physics model.

ggulgulia February 12, 2014 03:07

Hello Henry

Sorry for the delay in reply.

You can define the motion velocity.For this you have to

1. activate the Overset motion by creating appropriate mesh and interfaces.

2. Next as I mentioned you have to select DFBI body in the motion tools.

3. Expand the Region containing the rigid body which will move, go to Boundary Condition--> Motion . In the Properties Window change it to DFBI.

4. The above step will add a DFBI Node in the design tree. There you can specify the inertia properties of the body, velocity, and release time.

I suggest define an appropriate release time so that the flow fields develop sufficiently before the DFBI is activated.

Josh August 25, 2014 23:07

Quote:

Originally Posted by dsk2 (Post 468413)
Henry,

Its convenient to be able to specify displacement as opposed to velocities in Tools->Motion, but unfortunately I haven't found a way to do that either. So, what I do is calculate the velocities that give the displacement I desire and use the corresponding velocitiy expressions. Thus I end up with having to to significant math beforehand depending on how complex the trajectory of the moving body is.

Hi folks, especially DSK,

I am having a similar issue. I have a cylinder rolling sinusoidally at the surface. The displacement equation is:

[roll motion with respect to time] = [rolling amplitude] * sin([angular frequency]*[time])

However, the Rotation Motion in CCM+ requires a rotation rate [rad/s] rather than displacement value. So, does this "significant math" you speak of entail rearranging the above equation and solving for angular frequency? Surely, there's an easier to achieve forced sinusoidal rolling motion in CCM+.

Thanks!

ping August 26, 2014 04:17

for sinusoidal motion you can easily differentiate the equation of displacement you want to get the velocity equation

you could also do this within field functions or out in excel etc and import as a table

the reason most motion is input as velocity is that it then becomes indepedant of your timestep - otherwise each time you change timestep you need to change displacement

Josh August 26, 2014 20:09

Hi ping,

[time] is the real time, not the timestep, so it should be independent of timestep, correct?

Regardless, I derived the equation and input that as the motion. Now I want to verify the rolling angle is correct in STAR-CCM+. I have created a probe point on the corner of the object and compared the maximum Y value with the initial Y value, then pulled a little Pythagorean theorem to find the roll amplitude angle. Is there an easier way to go about this so I can verify each simulation has the correct roll amplitude?

ping August 27, 2014 04:32

when doing forced motion you normally want to achieve the same displacement at a certain instance in time independant of what you set the timestep to, and a velocity input achieves this

a probe its move with the motion property enable is the best way i have found to do this


All times are GMT -4. The time now is 12:08.