Modelling mixing tank using MRF

June 18, 2006, 03:40

Hi,

I am using MRF to model a baffled mixing tank. The boundari conditions that I used are

- no-slip, stationary wall for top, bottom and tank wall

- no-slip, moving wall for impeller and shaft

- interior for the boundary between moving and stationary zones

- 1 pair periodics for moving zones and 1 pair periodics for stationary zone

I got a very nice vector plot, and I think the direction is right. But the problem is the maximum velocity is 20% higher than the impeller tip speed and become worse when I double the cell number (using grid adaption - boundary) and using second order discretization scheme. I also found that the maximum velocity occured behind the impeller blade, not in the tip of impeller.

Could anyone please help me? Thank you

June 18, 2006, 15:20

hi,

Shouldn't you define the interface btw moving & stationary zones as "interface" b.c.? I do so for sliding mesh and mrf calculations.

is it the tangential velocity that you are talking about?

ozgur

xinboy · October 31, 2009, 08:38

I think the problem is due to settting as below
"no-slip, moving wall for impeller and shaft"
you should fix the impeller and shaft with the rotating zone!

greeneye · November 3, 2009, 06:38

i have a software that used for mixing tank if u want tell to send it for you

xinboy · November 3, 2009, 07:13

Thank you for your enthusiasm!
I expect the software you mentioned, though I'm not working this now.
could you send it to the email below?
cuishuxin1234@yahoo.com.cn

skris2009 · November 3, 2009, 23:09

I'm trying to model a washing machine in Fluent.
Geometry: circular drum, 1 inlet and outlet.
I'm trying to specify a rotating wall boundary condition. However, I'm struggling to use MRF. If anyone can explain this with an example, I would really appreciate that.

Also, I'm doing a particle tracking (injecting soap) to see how much of the soap sticks to the clothes as opposed to how much exits without interacting with the clothes. I have clothes turned on as 'sticky, escape wall' condition. However, when I do a particle track-dpm-from vel inlet. How can I find out the mass/number of the particles sticking to the clothes? The summary report only gives me total mass injected (not total mass exiting the domain.)

Any help (especially on the latter) is highly appreciated. Thanks

xinboy · November 4, 2009, 00:28

About the first question, the wall's velocity should be zero relative to adjacent cell zone if you use MRF. Otherwise, you may also use the rotating wall without MRF.
Because it needs to resolve the equation in the relative reference frame by using MRF, so the wall is stationary relative to the relative reference frame. while carrying out the rotating wall, it is actually that the sheer velocity is set the values as specified in the stationary N-S equation. Also Note that you cannot use the moving wall condition to model problems where the wall has a motion normal to itself. FLUENT will neglect any normal component of wall motion that you specify using the methods below.

About the second question, you may choose different surfaces to show the mass flow.

June 18, 2006, 03:40	Modelling mixing tank using MRF	#1
Adi Guest Posts: n/a	Hi, I am using MRF to model a baffled mixing tank. The boundari conditions that I used are - no-slip, stationary wall for top, bottom and tank wall - no-slip, moving wall for impeller and shaft - interior for the boundary between moving and stationary zones - 1 pair periodics for moving zones and 1 pair periodics for stationary zone I got a very nice vector plot, and I think the direction is right. But the problem is the maximum velocity is 20% higher than the impeller tip speed and become worse when I double the cell number (using grid adaption - boundary) and using second order discretization scheme. I also found that the maximum velocity occured behind the impeller blade, not in the tip of impeller. Could anyone please help me? Thank you

June 18, 2006, 15:20	Re: Modelling mixing tank using MRF	#2
ozgur Guest Posts: n/a	hi, Shouldn't you define the interface btw moving & stationary zones as "interface" b.c.? I do so for sliding mesh and mrf calculations. is it the tangential velocity that you are talking about? ozgur

November 3, 2009, 06:38	hello	#4
greeneye New Member reza Join Date: Nov 2009 Posts: 5 Rep Power: 16	i have a software that used for mixing tank if u want tell to send it for you

November 3, 2009, 23:09	Rotating wall BC	#6
skris2009 New Member Sowmya K Join Date: Nov 2009 Posts: 22 Rep Power: 16	I'm trying to model a washing machine in Fluent. Geometry: circular drum, 1 inlet and outlet. I'm trying to specify a rotating wall boundary condition. However, I'm struggling to use MRF. If anyone can explain this with an example, I would really appreciate that. Also, I'm doing a particle tracking (injecting soap) to see how much of the soap sticks to the clothes as opposed to how much exits without interacting with the clothes. I have clothes turned on as 'sticky, escape wall' condition. However, when I do a particle track-dpm-from vel inlet. How can I find out the mass/number of the particles sticking to the clothes? The summary report only gives me total mass injected (not total mass exiting the domain.) Any help (especially on the latter) is highly appreciated. Thanks

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Modelling of a Mixing Tank in Workbench ANSYS 12.1	shadyenany	FLUENT	12	June 9, 2017 05:53
3D mixing tank using MRF/SRF	cfd_confuse	Main CFD Forum	1	September 13, 2016 02:43
how to start for modelling n meshing a mixing tank with impeller	jeevan	ANSYS Meshing & Geometry	9	July 16, 2013 07:06
Mass conservation problem in mixing tank multiphase simulation	rockewan	FLUENT	0	April 6, 2010 12:34
3D modelling for mixing tank	boonho	FLUENT	2	March 22, 2005 04:54

October 31, 2009, 08:38		#3
xinboy New Member Shu-xin Cui Join Date: Oct 2009 Posts: 6 Rep Power: 16	I think the problem is due to settting as below "no-slip, moving wall for impeller and shaft" you should fix the impeller and shaft with the rotating zone!

November 3, 2009, 07:13		#5
xinboy New Member Shu-xin Cui Join Date: Oct 2009 Posts: 6 Rep Power: 16	Thank you for your enthusiasm! I expect the software you mentioned, though I'm not working this now. could you send it to the email below? cuishuxin1234@yahoo.com.cn

November 4, 2009, 00:28		#7
xinboy New Member Shu-xin Cui Join Date: Oct 2009 Posts: 6 Rep Power: 16	About the first question, the wall's velocity should be zero relative to adjacent cell zone if you use MRF. Otherwise, you may also use the rotating wall without MRF. Because it needs to resolve the equation in the relative reference frame by using MRF, so the wall is stationary relative to the relative reference frame. while carrying out the rotating wall, it is actually that the sheer velocity is set the values as specified in the stationary N-S equation. Also Note that you cannot use the moving wall condition to model problems where the wall has a motion normal to itself. FLUENT will neglect any normal component of wall motion that you specify using the methods below. About the second question, you may choose different surfaces to show the mass flow.