Boundary conditions

MarcelloAsciolla · April 7, 2018, 08:37

Good morning,
I am trying a simulation of a fan in a duct, so I have an inlet and an outlet.

The main goal of the simulation is to calculate the mass flow rate at the outlet. The only things that I know are RPM of the fan, the fluid (air), the pressure (static+dynamic) at 1 atm and temperature (25°C) and I am not sure about the boundary conditions.

I noticed that I can select these ones.
1)For inlet:
normal speed (unknown)
cartesian/cylindrical velocity components (unknown)
mass flow rate (unknown)
total pressure (<---- I choosed this one)

2)For outlet:
average static pressure (not sure)
normal speed (unknown)
cartesian/cylindrical velocity components (unknown)
mass flow rate (unknown)
static pressure (not sure)

3)For opening:
opening pres. and dyrn. (not sure, what is "dyrn"?)
static pres. and dyrn. (not sure, what is "dyrn"?)
entrainment (not sure)
cartesian/cylindrical velocity components (unknown)

I know the difference between "outlet" and "opening" surface and I expect that at the end of the duct the fluid's behavior is like the outlet one, but why can not I select the sum of static pressure and dynamic pressure there?
What is the correct boundary condition in my case at the end of the duct?

Gert-Jan · April 7, 2018, 08:58

You should set Total pressure at inlet and static pressure (=0, relative to the reference pressure, which is usually 1 atm) at outlet.

When using an opening, you have to set pressure and direction (not dyrn). But I would not recommend that.

MarcelloAsciolla · April 7, 2018, 09:10

Quote:

Originally Posted by Gert-Jan

You should set Total pressure at inlet and static pressure (=0, relative to the reference pressure, which is usually 1 atm) at outlet.

When using an opening, you have to set pressure and direction (not dyrn). But I would not recommend that.

Thank you for you answer, but why should I set 0 atm at the outlet for the static pressure?
In general I do not expect that the whole pressure is converted in a dynamic contribution...

Gert-Jan · April 7, 2018, 09:21

No indeed. This static pressure is just to get the calculation running.
In the postprocessing, you should calculate the total pressure at the outlet and use the difference with total pressure at the inlet to determine the performance of your fan.

MarcelloAsciolla · April 7, 2018, 10:06

So is this an attempt value?
Thank you very much.

Do you know also how does this work exactly? Because it looks a bit strange and it is a thing that I have never seen, but I have to admit that I do not know much about the numerical implementation of Navier-Stokes, I am still studying them. As I studied on the books the boundary value is always fixed.
I mean that in this way I am not giving a value on the boundary, so I imagine that it is implemented in the software an iterative process and at the start it will consider it as an attempt value and after it will consider the boundary value as an unknow and corrects it.

Let me ask another question too, what does "dyrn." stand for in CFX? I tried to give a look on the documentation, but I did not find its definition.

Gert-Jan · April 8, 2018, 09:34

There is nothing strange. You just set a fan rotational speed and a pressure difference. Then the CFD software calculates the torque and the flow. From the flow, average velocities at in- and outlet can be detemined, allowing you to determine dynamic and static pressures at all boundaries.

In my first post I already explained what 'dyrn' is. You made a typo since it is not 'dyrn' but it is 'dirn' and this is an abbreviation of direction. So when using this, you have to specify pressure and incoming direction of the air.

MarcelloAsciolla · April 8, 2018, 14:11

My apologies. Before this thread I thought that it was about the dynamic pressure, for this reason I did this mistake.
Thank you very much.

So that 0 atm pressure is not the ambient pressure, but another contribution caused by other factors, right?
So, for istance, if I have a problem without a difference of pressure, the duct is fixed and the fluid is moved only by the fan, is it correct to solve it with 0 atm at the outlet/opening?

Let me ask another question: if the duct is not infinite and has a finite dimension (think about a hairdryer, a fan, an air conditioning system), is correct to solve the problem only inside the duct where is the fan? Do I have to consider also the domain around it? A part of the room for example.
Consider that the study case is about the mass flow rate at the outlet and everything out of the duct is not important.

Sorry for all these questions, but it is the first time that I am using CFX and it is also the first time that I am facing these problems.

Gert-Jan · April 8, 2018, 15:52

Your reference pressure is 1 atm. That is the base. Every pressure that you define on whatever boundary is relative to this reference pressure. In otherwords, your absolute pressure is your reference pressure +/- the static pressures which varies through your domain from inlet to outlet.

I would say your outlet is only important if you want to convert dynamic pressure into static pressure. Inlet can be very important if you have sharp edges leading to a vena contracta.

April 7, 2018, 08:37	Boundary conditions	#1
MarcelloAsciolla Member Marcello Asciolla Join Date: Apr 2018 Location: Italy Posts: 43 Rep Power: 8	Good morning, I am trying a simulation of a fan in a duct, so I have an inlet and an outlet. The main goal of the simulation is to calculate the mass flow rate at the outlet. The only things that I know are RPM of the fan, the fluid (air), the pressure (static+dynamic) at 1 atm and temperature (25°C) and I am not sure about the boundary conditions. I noticed that I can select these ones. 1)For inlet: normal speed (unknown) cartesian/cylindrical velocity components (unknown) mass flow rate (unknown) total pressure (<---- I choosed this one) 2)For outlet: average static pressure (not sure) normal speed (unknown) cartesian/cylindrical velocity components (unknown) mass flow rate (unknown) static pressure (not sure) 3)For opening: opening pres. and dyrn. (not sure, what is "dyrn"?) static pres. and dyrn. (not sure, what is "dyrn"?) entrainment (not sure) cartesian/cylindrical velocity components (unknown) I know the difference between "outlet" and "opening" surface and I expect that at the end of the duct the fluid's behavior is like the outlet one, but why can not I select the sum of static pressure and dynamic pressure there? What is the correct boundary condition in my case at the end of the duct?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
sliding mesh problem in CFX	Saima	CFX	46	September 11, 2021 08:38
Radiation in semi-transparent media with surface-to-surface model?	mpeppels	CFX	11	August 22, 2019 08:30
Basic Nozzle-Expander Design	karmavatar	CFX	20	March 20, 2016 09:44
GETVAR Error in Multiband Monte Carlo Radiation Simulation with Directional Source	silvan	CFX	3	June 16, 2014 10:49
Question about heat transfer coefficient setting for CFX	Anna Tian	CFX	1	June 16, 2013 07:28

April 7, 2018, 08:58		#2
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	You should set Total pressure at inlet and static pressure (=0, relative to the reference pressure, which is usually 1 atm) at outlet. When using an opening, you have to set pressure and direction (not dyrn). But I would not recommend that.

April 7, 2018, 09:21		#4
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	No indeed. This static pressure is just to get the calculation running. In the postprocessing, you should calculate the total pressure at the outlet and use the difference with total pressure at the inlet to determine the performance of your fan.

April 7, 2018, 10:06		#5
MarcelloAsciolla Member Marcello Asciolla Join Date: Apr 2018 Location: Italy Posts: 43 Rep Power: 8	So is this an attempt value? Thank you very much. Do you know also how does this work exactly? Because it looks a bit strange and it is a thing that I have never seen, but I have to admit that I do not know much about the numerical implementation of Navier-Stokes, I am still studying them. As I studied on the books the boundary value is always fixed. I mean that in this way I am not giving a value on the boundary, so I imagine that it is implemented in the software an iterative process and at the start it will consider it as an attempt value and after it will consider the boundary value as an unknow and corrects it. Let me ask another question too, what does "dyrn." stand for in CFX? I tried to give a look on the documentation, but I did not find its definition.

April 8, 2018, 09:34		#6
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	There is nothing strange. You just set a fan rotational speed and a pressure difference. Then the CFD software calculates the torque and the flow. From the flow, average velocities at in- and outlet can be detemined, allowing you to determine dynamic and static pressures at all boundaries. In my first post I already explained what 'dyrn' is. You made a typo since it is not 'dyrn' but it is 'dirn' and this is an abbreviation of direction. So when using this, you have to specify pressure and incoming direction of the air.

April 8, 2018, 14:11		#7
MarcelloAsciolla Member Marcello Asciolla Join Date: Apr 2018 Location: Italy Posts: 43 Rep Power: 8	My apologies. Before this thread I thought that it was about the dynamic pressure, for this reason I did this mistake. Thank you very much. So that 0 atm pressure is not the ambient pressure, but another contribution caused by other factors, right? So, for istance, if I have a problem without a difference of pressure, the duct is fixed and the fluid is moved only by the fan, is it correct to solve it with 0 atm at the outlet/opening? Let me ask another question: if the duct is not infinite and has a finite dimension (think about a hairdryer, a fan, an air conditioning system), is correct to solve the problem only inside the duct where is the fan? Do I have to consider also the domain around it? A part of the room for example. Consider that the study case is about the mass flow rate at the outlet and everything out of the duct is not important. Sorry for all these questions, but it is the first time that I am using CFX and it is also the first time that I am facing these problems.

April 8, 2018, 15:52		#8
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	Your reference pressure is 1 atm. That is the base. Every pressure that you define on whatever boundary is relative to this reference pressure. In otherwords, your absolute pressure is your reference pressure +/- the static pressures which varies through your domain from inlet to outlet. I would say your outlet is only important if you want to convert dynamic pressure into static pressure. Inlet can be very important if you have sharp edges leading to a vena contracta.