Calculating cumulative work by turbulent eddy dissipation along streamlines

FrankS · January 21, 2015, 14:08

Hi All,
I am running simulations of a machine called a repulper which is like a giant blender used to break up waste paper for recycling. A repulper is just a large tank with a high-speed rotor mounted in the bottom. Waste paper is dumped in to the machine and the rotor circulates the suspension of water and waste paper and breaks up the waste paper.

I have done a lot of experiments with a real repulper and I have come up with a simple analytical model which allows me to predict my experimental results by knowing only the dimensions of the repulper, the material toughness of the paper being broken down, and the power input to the system. Although my model is accurate, in it's simple form, it doesn't take in to account rotor and tank design details such as baffle design, or number of rotor vanes. But, the complex form could be incorporated in to a CFD sim and then I can quantify these things.

I have been running simulations using CFX and they too are working well. I have the machine split in to 2 domains, one encompassing the rotor, and the other the tank/vat. I have found that the k-w SST is working really well for this. Water is my working fluid. I have been running both Frozen Rotor and sliding mesh sims. Both are accurate.

I want to apply my model to the CFD sim. Basically, I need the residence time of fluid passing through each domain and the energy from turbulent eddy dissipation at the rotor. I have used a transport equation source to get histograms of the residence time in each domain and this is working well.

What I want to get is a histogram of the residence energy from eddy dissipation in the rotor domain. I would like to integrate turbulent eddy dissipation with time along streamlines to do this but I am a bit foggy on how to implement this in the solver.

Sorry for the length of this. Any help would be greatly appreciated.
Thanks
Frank

ghorrocks · January 21, 2015, 17:00

Is your model of the real pulper an empirical model, or an analytical one? Empirical models are based on experimental measurements, analytical models are based on the underlying theoretical model.

This sounds like elementary post processing. Generate a single streamline in a streamline object, then use the calculator to integrate the k value along the streamline. To get a histogram move the streamline seed point around. And use a session file script to automate this so you can do it for lots of streamlines.

FrankS · February 3, 2015, 14:38

Hi ghorrocks. Thanks for the reply. I manged to implement your method and use a transport equation source as well.

My model is an analytical model that just happens to predict my experimental results for a wide range of conditions with an R^2 of 0.93.

January 21, 2015, 14:08	Calculating cumulative work by turbulent eddy dissipation along streamlines	#1
FrankS New Member Frank Join Date: Oct 2013 Posts: 26 Rep Power: 12	Hi All, I am running simulations of a machine called a repulper which is like a giant blender used to break up waste paper for recycling. A repulper is just a large tank with a high-speed rotor mounted in the bottom. Waste paper is dumped in to the machine and the rotor circulates the suspension of water and waste paper and breaks up the waste paper. I have done a lot of experiments with a real repulper and I have come up with a simple analytical model which allows me to predict my experimental results by knowing only the dimensions of the repulper, the material toughness of the paper being broken down, and the power input to the system. Although my model is accurate, in it's simple form, it doesn't take in to account rotor and tank design details such as baffle design, or number of rotor vanes. But, the complex form could be incorporated in to a CFD sim and then I can quantify these things. I have been running simulations using CFX and they too are working well. I have the machine split in to 2 domains, one encompassing the rotor, and the other the tank/vat. I have found that the k-w SST is working really well for this. Water is my working fluid. I have been running both Frozen Rotor and sliding mesh sims. Both are accurate. I want to apply my model to the CFD sim. Basically, I need the residence time of fluid passing through each domain and the energy from turbulent eddy dissipation at the rotor. I have used a transport equation source to get histograms of the residence time in each domain and this is working well. What I want to get is a histogram of the residence energy from eddy dissipation in the rotor domain. I would like to integrate turbulent eddy dissipation with time along streamlines to do this but I am a bit foggy on how to implement this in the solver. Sorry for the length of this. Any help would be greatly appreciated. Thanks Frank

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Dissabling turbulent dissipation	lindner	OpenFOAM	3	February 7, 2021 13:51
calculating work in sixDoFRigidBodyMotion	ashkan	OpenFOAM Programming & Development	0	July 14, 2014 23:04
How to use Eddy Dissipation rate only When using Finite rate/Eddy Dissipation model	Dziyad	FLUENT	2	February 19, 2014 17:25
How to write k and epsilon before the abnormal end	xiuying	OpenFOAM Running, Solving & CFD	8	August 27, 2013 15:33
Pressure work term in turbulent K.E. equation	lost.identity	Main CFD Forum	0	March 8, 2011 12:21

January 21, 2015, 17:00		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,716 Rep Power: 143	Is your model of the real pulper an empirical model, or an analytical one? Empirical models are based on experimental measurements, analytical models are based on the underlying theoretical model. This sounds like elementary post processing. Generate a single streamline in a streamline object, then use the calculator to integrate the k value along the streamline. To get a histogram move the streamline seed point around. And use a session file script to automate this so you can do it for lots of streamlines.

February 3, 2015, 14:38		#3
FrankS New Member Frank Join Date: Oct 2013 Posts: 26 Rep Power: 12	Hi ghorrocks. Thanks for the reply. I manged to implement your method and use a transport equation source as well. My model is an analytical model that just happens to predict my experimental results for a wide range of conditions with an R^2 of 0.93.