Flow rate restriction simulation set-up

siw · November 11, 2015, 08:13

Hi,

Can anyone suggest a suitable set-up for modelling this scenario?

I have an internal system upstream of a centrifugal fan and when the fan is operating it pulls a volumetric flow rate of 346 litres/second through the system. My CFD model is of the system upstream of the fan only and I would use an outlet mass flow rate and a total pressure inlet (the air is drawn in from the atmosphere) to model this condition - so far so good. However, I need to include a restrictor, e.g. orifice plate, into the system so that the volumetric flow rate is reduced to 250 litres/second. There would be a corresponding pressure loss across the restrictor.

I could simply alter the outlet flow rate boundary condition to the new value of 250 L/s then let CFX calculate the restrictor pressure loss. But this would not produce the reduction in flow rate as part of the calculation because it is the restrictor which reduces the flow rate and not the boundary conditions. Since CFD works on mass conservation then specifying 346 L/s at the outlet and having CFX alter the flow rate through the model does not seem possible to me.

Thanks.

shivasluzz · November 11, 2015, 11:22

2 methods..
1. model the orifice plate also...

2. analyse the orifice plate separately and develop the del P vs mass flow characteristics..
create function in cfx-pre using that relation
create 2 domains from your single domain (by splitting at the place of orifice; no need to model orifice)
create domain interface and use that function for interface model..

ghorrocks · November 11, 2015, 17:37

The fan has a performance curve (dP vs flow rate), and the system has a performance curve (dP vs flow rate). The system will run at equilibrium when at a given flow rate the pressure rise over the fan equals the pressure loss through the system. So:

If you like pen and paper solutions, draw the two performance curves and where they intersect is the operating point.

But if you want to implement this into CFX so the model finds its own operating point - then make the inlet a CEL function of pressure versus flow rate which describes the fan performance curve and then CFX will find the equilibrium with what ever you model. If you decide to not model the oriface plate (and I recommend you do not model it if you know its effect and don't care about the details) then include its effect in the inlet boundary performance curve.

What I am describing is essentially sluzzer's option 2.

shivasluzz · November 12, 2015, 06:59

thats a good explanation... and this was the point i was getting confused for months...

basically thr r two things
1. component analysis
2. system analysis (combination of components)

often it is complicated to do system analysis in cfd...
so develop characteristics for components using cfd and do component matching to find the operating point...

system softwares like Amesim, Simulink etc., does this job for you
https://en.wikipedia.org/wiki/LMS_Imagine.Lab_Amesim
(it will be too complicated to match by hand when the variables and components are more)

there is a nasa system engineering handbook regarding component matchings... other than that so far i didnt find good resource regarding system principles.. if any body knows any good material, then please recommend (i put this question in system analysis forum and i got no response :P)

shivasluzz · February 16, 2016, 12:15

Like to bring back this old thread with a new doubt...
If there is single input and single output for a component in a system, then we can characterise that component using an expression or relation like dp vs flow rate for an orifice... But what if the component we have to characterise by expression or relation has 'Multiple input and Multiple output' ?!

Somebody please give some explanation

November 11, 2015, 08:13	Flow rate restriction simulation set-up	#1
siw Senior Member Stuart Join Date: Jul 2009 Location: Portsmouth, England Posts: 733 Rep Power: 25	Hi, Can anyone suggest a suitable set-up for modelling this scenario? I have an internal system upstream of a centrifugal fan and when the fan is operating it pulls a volumetric flow rate of 346 litres/second through the system. My CFD model is of the system upstream of the fan only and I would use an outlet mass flow rate and a total pressure inlet (the air is drawn in from the atmosphere) to model this condition - so far so good. However, I need to include a restrictor, e.g. orifice plate, into the system so that the volumetric flow rate is reduced to 250 litres/second. There would be a corresponding pressure loss across the restrictor. I could simply alter the outlet flow rate boundary condition to the new value of 250 L/s then let CFX calculate the restrictor pressure loss. But this would not produce the reduction in flow rate as part of the calculation because it is the restrictor which reduces the flow rate and not the boundary conditions. Since CFD works on mass conservation then specifying 346 L/s at the outlet and having CFX alter the flow rate through the model does not seem possible to me. Thanks.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
rotor 37 choking mass flow rate fluent simulation	sehs15	FLUENT	1	January 15, 2015 06:07
two fluid, volumetric/mass flow rate at different volume fractions	Veronique Pe	CFX	2	November 26, 2014 05:30
Mass flow rate prediction of Purge control valve using set pressure drop	enr_venkat	CFX	11	February 27, 2014 11:30
mass flow rate...	sanjar	OpenFOAM Running, Solving & CFD	1	December 2, 2013 00:09
[snappyHexMesh] determining displacement for added points	CFDnewbie147	OpenFOAM Meshing & Mesh Conversion	1	October 22, 2013 09:53

November 11, 2015, 11:22		#2
shivasluzz Senior Member sluzzer Join Date: Jul 2014 Posts: 142 Rep Power: 11	2 methods.. 1. model the orifice plate also... 2. analyse the orifice plate separately and develop the del P vs mass flow characteristics.. create function in cfx-pre using that relation create 2 domains from your single domain (by splitting at the place of orifice; no need to model orifice) create domain interface and use that function for interface model..

November 11, 2015, 17:37		#3
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	The fan has a performance curve (dP vs flow rate), and the system has a performance curve (dP vs flow rate). The system will run at equilibrium when at a given flow rate the pressure rise over the fan equals the pressure loss through the system. So: If you like pen and paper solutions, draw the two performance curves and where they intersect is the operating point. But if you want to implement this into CFX so the model finds its own operating point - then make the inlet a CEL function of pressure versus flow rate which describes the fan performance curve and then CFX will find the equilibrium with what ever you model. If you decide to not model the oriface plate (and I recommend you do not model it if you know its effect and don't care about the details) then include its effect in the inlet boundary performance curve. What I am describing is essentially sluzzer's option 2.

November 12, 2015, 06:59		#4
shivasluzz Senior Member sluzzer Join Date: Jul 2014 Posts: 142 Rep Power: 11	thats a good explanation... and this was the point i was getting confused for months... basically thr r two things 1. component analysis 2. system analysis (combination of components) often it is complicated to do system analysis in cfd... so develop characteristics for components using cfd and do component matching to find the operating point... system softwares like Amesim, Simulink etc., does this job for you https://en.wikipedia.org/wiki/LMS_Imagine.Lab_Amesim (it will be too complicated to match by hand when the variables and components are more) there is a nasa system engineering handbook regarding component matchings... other than that so far i didnt find good resource regarding system principles.. if any body knows any good material, then please recommend (i put this question in system analysis forum and i got no response :P)

February 16, 2016, 12:15		#5
shivasluzz Senior Member sluzzer Join Date: Jul 2014 Posts: 142 Rep Power: 11	Like to bring back this old thread with a new doubt... If there is single input and single output for a component in a system, then we can characterise that component using an expression or relation like dp vs flow rate for an orifice... But what if the component we have to characterise by expression or relation has 'Multiple input and Multiple output' ?! Somebody please give some explanation