CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > CFX

Eulerian multiphase flow- flow direction not comply with gravity

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   April 14, 2017, 16:37
Default Eulerian multiphase flow- flow direction not comply with gravity
  #1
Member
 
Join Date: Jul 2013
Posts: 94
Rep Power: 11
mejahan is on a distinguished road
Hi All,

I am simulating a high viscous liquid flow with the air flow. The effects of the thermal convection and gravity was considered in the simulation.
However, the direction of the liquid flow is not what is expected according to the gravity force direction. I used full buoyancy model with setting air as the reference density.
I attached an image to better explanation.
I tried to change the direction of the gravity and still the same result (weird).
Re=0.1 (very high viscous flow)
We=1.5
The dimensions are very small in order of 1-5 mm.
----------------
When I increase the injection velocity (Re=1.0), I get physical results where the injected liquid stays in the lower surface.
I did the mesh and time sensitivity analysis.

Any thoughts?

Many thanks for your time in advance.
Attached Images
File Type: jpg Fluid-sim.jpg (83.8 KB, 16 views)
mejahan is offline   Reply With Quote

Old   April 15, 2017, 00:02
Default
  #2
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,144
Rep Power: 134
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Why is this result not physical? If the moving wall bends the extruded stuff to the side, then the highly viscous nature means it does not move fast enough to straighten out and it keeps bending around - which is exactly what you are seeing. So your model could well be correct.
ghorrocks is offline   Reply With Quote

Old   April 15, 2017, 03:02
Default
  #3
Member
 
Join Date: Jul 2013
Posts: 94
Rep Power: 11
mejahan is on a distinguished road
Hi Glenn,

Thanks for your reply, good to hear from you.
But in a higher viscous flow, the rate of the momentum transfer is higher, therefore, the fluid should follow the velocity of the no slip wall and move with the wall.
In more investigations, first I increased the Re=1.0 and got the expected results. Why the flow is sticking to the bed here?
Later, I neglected the buoyancy effects with the same low Re=0.1 and got a weird result. If the results of the first simulation is physical, then why I got a different results with not considering buoyancy? The effect of the buoyancy here should not be considerable in influencing the direction for the flow.
I need to mention that the velocity of the bed and injection velocity are the same. And the Temperature of the injected flow is higher than the bed.

Thanks for your time in advance.
Attached Images
File Type: jpg Fluid-sim-1.jpg (56.6 KB, 13 views)
mejahan is offline   Reply With Quote

Old   April 15, 2017, 05:09
Default
  #4
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,144
Rep Power: 134
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
I think you will find the situation is more complex than you describe. Yes, the stuff will experience shear from the moving wall. But it also experiences shear as it is deformed from the nozzle and bends. So the more viscous stuff will also resist the deformation of the extruded shape more. The shape which has the least deformation is rigid body motion, which for a curved shape means it bends around in a ring. That is what you are getting for the very viscous material. So it makes sense to me.
ghorrocks is offline   Reply With Quote

Old   April 15, 2017, 15:58
Default
  #5
Member
 
Join Date: Jul 2013
Posts: 94
Rep Power: 11
mejahan is on a distinguished road
Hi Glenn,

I would appreciate if you help me understand the problem


I know that the bending is the cause of the non-uniform shear at the wall of the extruder, and the flow experiences this non-uniform shear after it reaches the bed and dragged by the shear at the bed. But why we see completely different flow behaviors with Re=0.1 and Re=1.0?


In addition, Why the flow is completely detached from the bed in lower Re after some time? There should be a reason for non-uniform shear at the extruder, but when the flow is detached from the bed, how the flow still continues to bend?


What is the effect of buoyancy here?


Thanks
Attached Images
File Type: jpg Untitled-2.jpg (71.8 KB, 5 views)
mejahan is offline   Reply With Quote

Old   April 15, 2017, 18:07
Default
  #6
New Member
 
Artur
Join Date: Jan 2017
Posts: 11
Rep Power: 7
Pumba is on a distinguished road
Look at the way this spring on the left side is bent.
I think a very viscous material behaves the same way.

Edit: i don't know why it's detached though...
Pumba is offline   Reply With Quote

Old   April 15, 2017, 18:50
Default
  #7
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,144
Rep Power: 134
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
That's a great way of explaining it, very good.

Have a think about how the initial bend put into the steel wire stays in there and ends up going round and round - and forms a spring. We are suggesting the same thing is happening in your high viscosity case.

For this to happen the high viscosity has to be able to stop (or at least slow down) the forces which are trying to make it sag down (due to weight in gravity) or be drawn along the moving wall.
ghorrocks is offline   Reply With Quote

Old   April 16, 2017, 03:50
Default
  #8
Member
 
Join Date: Jul 2013
Posts: 94
Rep Power: 11
mejahan is on a distinguished road
The initial deformation in the solid material caused by an external shear force continue to exist but here we have a high viscous fluid with viscosity of 0.5-2 Pa.s. In addition, there should be a source for the non-uniform shear at the extruder, but when the flow is completely detached from the bed, then there would be no source for such non-uniform shear at the injection area.
mejahan is offline   Reply With Quote

Old   April 16, 2017, 03:57
Default
  #9
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,144
Rep Power: 134
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Something is pushing it to the side. That could be either physical, so something is really pushing it to the side; or it could be numerical error, so it is not real and you need to solve the equations more accurately.

So I would first closely look at the time history. Does it go to the side right from the start? Or does it hit the moving wall then move to the side? Is there something else pushing it to the side (maybe the gas phase)?

If nothing there explains it then start looking for numerical errors. Do you get it when you make the convergence tolerance tighter? Double Precision numerics? Improved mesh quality? Finer mesh? Smaller time step?

All these things have to be done by you checking your model carefully.
ghorrocks is offline   Reply With Quote

Old   April 17, 2017, 01:24
Default
  #10
Member
 
Join Date: Jul 2013
Posts: 94
Rep Power: 11
mejahan is on a distinguished road
I did the time and mesh sensitivity analysis and results are the same, but in less dense mesh and higher Courant numbers, the interface is more smeered around, still the same results. And the solver is on double precision.
The flow first hits the bed and then is deflected by the shear force on the bed, after that it starts to bend up for some reason which is confusing me, and after a while it detaches from the surface, but still continues to have a upward curled shape which is against the gravity.
We have some air flow around the liquid flow that can affect the liquid flow but since its density and viscosity is much lower those of the air, I do not expect to see a considerable change in the liquid flow. ( the bed velocity is the same as injection velocity)
After trying to change many aspects of this simulation such as changing the liquid viscosity and velocity , bed velocity and temperature, changing the buoyancy setups, etc., there are some phenomena that are difficult for me to understand:
1- Why the liquid flow is curled up against the gravity force?
2- Why it is detached from the bed?
3- What is the effect of surface tension in here?
4- What is the effect of Buoyancy in this simulation? ( I had completely different results when the buoyancy effect was not considered.
5- Why the liquid flow sticks to the bed when the injection velocity is increased? (Re=1.0)

Thank you for your time in advance.
mejahan is offline   Reply With Quote

Old   April 17, 2017, 01:53
Default
  #11
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,144
Rep Power: 134
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
We cannot answer any of these questions. For us to help you answer them you will need to post an image of your full geometry, mesh and CCL. Some excerpts from your output file would help as well.
ghorrocks is offline   Reply With Quote

Old   April 17, 2017, 04:19
Default
  #12
Member
 
Join Date: Jul 2013
Posts: 94
Rep Power: 11
mejahan is on a distinguished road
Sure, I provided an image about the mesh and geometry and the CCL file.
Please let me know if you need more information.

Look forward to hearing your opinion.

Thanks,
Attached Images
File Type: jpg Untitled3.jpg (202.2 KB, 9 views)
Attached Files
File Type: txt CCL-Expo.ccl.txt (10.4 KB, 3 views)
mejahan is offline   Reply With Quote

Reply

Tags
buoyancy, multiphase flows, viscous flow

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Multiphase Porous Media Flow - Convergence Issues VT_Bromley FLUENT 7 May 14, 2020 16:34
mesh size in Eulerian multiphase flow - liquid/solid mixture microfin FLUENT 4 December 23, 2016 13:55
How to simulate the eulerian multiphase model about particle jhlee9622 STAR-CCM+ 2 November 24, 2016 11:37
Gravity Vector Direction for One-Way Particle Track FrankS CFX 3 July 21, 2016 18:21
fluid flow fundas ram Main CFD Forum 5 June 17, 2000 21:31


All times are GMT -4. The time now is 01:49.