CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   fluid jet simulation with 1 dimensional equation (https://www.cfd-online.com/Forums/main/8191-fluid-jet-simulation-1-dimensional-equation.html)

sangwon October 8, 2004 13:37

fluid jet simulation with 1 dimensional equation
 
Hi All:

I'm in a computer graphics community and trying to simulate fluid jet. It's not in full 3D space. The axis is fixed, and need to see the change of the thickness of the jet along the axis while I perturb the velocity at the inlet position.

The requirement is 1. FAST & STABLE -- it should be realtime 2. need NOT be accurate -- if it 'looks' realistic, that's enough 3. should be able to handle the breakup without much hassle

I found a work by Jens Eggers who used 1D fluid equation to simulate drippling faucet. He derived the equation from NS equation in cylindrical coordinate. I actually got the source code from one of the authors, but it is too slow and complicated as a graphics application.

Can somebody give me a reference or advice?

Junseok Kim October 8, 2004 14:46

Re: fluid jet simulation with 1 dimensional equati
 
You may want to the website, put your mouse on the jet figure, then you will see the movie.

http://math.uci.edu/~jskim/simulations.html

Junseok

Junseok Kim October 8, 2004 14:51

Re: fluid jet simulation with 1 dimensional equati
 
sorry, I didn't read your post in detail, I missed some the calculation is slow and is not suitable to your case, what I meant is that it is not real time simulation.

Junseok

sangwon October 8, 2004 15:49

Re: fluid jet simulation with 1 dimensional equati
 
But your simulation looks really cool! Did it use full NS solver or simplified 1D solver? Do u think it would be possible to simplify your code somehow to make it very fast? ( again, accuracy doesn't matter much )


Junseok Kim October 8, 2004 16:31

Re: fluid jet simulation with 1 dimensional equati
 
The numerical code is solving full NS in axisymmetric domain (2D) and for the interface capturing I used phase-field model (Cahn-Hilliard equation), the basic idea is very similar to popular level set method. Jet simulation is complicated process, involving gravity, oscillating inflow velocity, and surface tension effects.

I am not sure how fast is fast in your case. The meaning of fastness is relative.

Junseok


Junseok Kim October 8, 2004 16:34

Re: fluid jet simulation with 1 dimensional equati
 
As you may noticed from my simulation result, after pinch-off the droplet has inverted curvature, which is 2D effect, from the 1D you can't get that one because it has multiple solution from 1D point of view.

Junseok


All times are GMT -4. The time now is 05:49.