Erosion and deposition model for turbidity currents

chancy · September 3, 2015, 13:29

Dear all,

I am using ANSYS CFX to model 2D turbidity currents.
Geometry is a long flume with a length of 6 meters and for the width I used one cell for it to be 2D. The height of the flume is of 1m.
Mesh is of 9 mm.
I am using the Eulerian inhomogeneous model.
The SST (Shear Stress Turbulent) model is used to model turbulence.
An initial volumetric concentration of 2.5% is used for the mixture. For the particles I was obliged to use one diameter (mean diameter) because with these options I cannot enter a whole grain size distribution.

The current is well formed, but the only issue concerns sediment transport. In experimental tests, I have a high deposition of the current while this numerical model does not deposit...In other words, if you add a well at the end of the model instead of a wall, ALL the sediment will end up going out of the system while in real cases, a lot deposits before reaching the well.

I am quite lost how to solve this problem: many options:
1) is it the mesh size? Should it be smaller?
2) I cannot find a deposition/erosion equation used in the cfx code. Should I introduce it? If yes, is there a tutorial on how to modify CFX scripts?
3) Should I use the Lagragian model instead? But then is it possible to model turbulence very well?

I am looking forward to receiving your answer(s)!

Thank you

ghorrocks · September 3, 2015, 17:52

CFX does not have a particle deposition/scouring model. You cannot do this in CFX with either the Lagangian or Eularian models.

chancy · September 3, 2015, 18:00

Thank you for your answer ghorrocks. Is it possible to implement a few equations inside the code of CFX? Is it open for modification? If not, does FLUENT have such a model? If not, what do you recommend for such modeling?!

ghorrocks · September 3, 2015, 18:12

CFX (and Fluent) both have access to considerable flexibility so if you can define some equations which describe deposition/scouring then you might be able to implement it. But have a think about it - if it was easy to implement don't you think it would already be in there?

This question has been asked several times on the forum and nobody seems to have successfully done a deposition/scouring model with CFX. So if you want to model deposition/scouring then I would look at other CFD codes which support this. I have no idea if such a code exists as deposition/scouring is not my field of expertise.

hamide · November 8, 2015, 07:36

hi
i wanna model particle deposition in turbulent flow, and i can't understand the non-dimensional Gaussian random field to obtain the instantaneous fluctuation velocity field that affect particle movement.
would u please help me...

ghorrocks · November 8, 2015, 16:14

What don't you understand about it?

hamide · November 9, 2015, 01:22

[QUOTE=ghorrocks;572496]What don't you understand about it?[/Q

it results a especial equation! i can't understand that equation!!!!

the equation is in the doc file i attached and it's the part of one paper i read ...

ghorrocks · November 9, 2015, 05:04

Follow the references back to the original definition of the equation. That is the standard thing to do when something does not explain it adequately.

But this just appears to be a function to generate a random velocity fluctuation in space and time for the components of velocity.

hamide · November 9, 2015, 05:57

Quote:

Originally Posted by ghorrocks

Follow the references back to the original definition of the equation. That is the standard thing to do when something does not explain it adequately.

But this just appears to be a function to generate a random velocity fluctuation in space and time for the components of velocity.

thank you so much dear ghorrocks ....

September 3, 2015, 13:29	Erosion and deposition model for turbidity currents	#1
chancy New Member Join Date: Sep 2015 Posts: 2 Rep Power: 0	Dear all, I am using ANSYS CFX to model 2D turbidity currents. Geometry is a long flume with a length of 6 meters and for the width I used one cell for it to be 2D. The height of the flume is of 1m. Mesh is of 9 mm. I am using the Eulerian inhomogeneous model. The SST (Shear Stress Turbulent) model is used to model turbulence. An initial volumetric concentration of 2.5% is used for the mixture. For the particles I was obliged to use one diameter (mean diameter) because with these options I cannot enter a whole grain size distribution. The current is well formed, but the only issue concerns sediment transport. In experimental tests, I have a high deposition of the current while this numerical model does not deposit...In other words, if you add a well at the end of the model instead of a wall, ALL the sediment will end up going out of the system while in real cases, a lot deposits before reaching the well. I am quite lost how to solve this problem: many options: 1) is it the mesh size? Should it be smaller? 2) I cannot find a deposition/erosion equation used in the cfx code. Should I introduce it? If yes, is there a tutorial on how to modify CFX scripts? 3) Should I use the Lagragian model instead? But then is it possible to model turbulence very well? I am looking forward to receiving your answer(s)! Thank you

November 8, 2015, 07:36	particle deposition	#5
hamide New Member hamide hayati Join Date: Oct 2015 Posts: 28 Rep Power: 10	hi i wanna model particle deposition in turbulent flow, and i can't understand the non-dimensional Gaussian random field to obtain the instantaneous fluctuation velocity field that affect particle movement. would u please help me...

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Is it possible to model sand bed erosion in Fluent?	sharif88	Fluent Multiphase	2	June 25, 2015 13:22
Deposition and erosion process	suizo	CFX	5	June 11, 2014 18:52
Sediment transport and deposition	Chaudhry	FLUENT	0	April 10, 2013 09:41

September 3, 2015, 17:52		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	CFX does not have a particle deposition/scouring model. You cannot do this in CFX with either the Lagangian or Eularian models.

September 3, 2015, 18:00		#3
chancy New Member Join Date: Sep 2015 Posts: 2 Rep Power: 0	Thank you for your answer ghorrocks. Is it possible to implement a few equations inside the code of CFX? Is it open for modification? If not, does FLUENT have such a model? If not, what do you recommend for such modeling?!

September 3, 2015, 18:12		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	CFX (and Fluent) both have access to considerable flexibility so if you can define some equations which describe deposition/scouring then you might be able to implement it. But have a think about it - if it was easy to implement don't you think it would already be in there? This question has been asked several times on the forum and nobody seems to have successfully done a deposition/scouring model with CFX. So if you want to model deposition/scouring then I would look at other CFD codes which support this. I have no idea if such a code exists as deposition/scouring is not my field of expertise.

November 8, 2015, 16:14		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	What don't you understand about it?

November 9, 2015, 05:04		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Follow the references back to the original definition of the equation. That is the standard thing to do when something does not explain it adequately. But this just appears to be a function to generate a random velocity fluctuation in space and time for the components of velocity.