Inlet Boundary Mass Flow Error

AtoHM · October 19, 2018, 01:59

Hi all,
I've come across a strange issue. I am basically simulating water flow into a tank. The inlet is a horizontal rectangular patch. The flow direction should be applied with 45 deg inclination. So I used Cart. Vel. Components to apply it. The total inlet velocity is known from case mass flow and inlet patch area. To have it applied with 45 deg, I apply it multiplied with 1/sqrt(2) in both my desired directions.
Problem is: mass flow in result and monitor is only 1/sqrt(2) times the amount we were going for. The direction however seems to be 45 deg as we wanted.

I looked it up in the modeling guide, where it is stated, that the normal-to-boundary component of the input is ignored. I must be missing something here, but I dont get a) why it is implemented like that and b) how to get my desired mass flow at the inlet.

Any help is very appreciated.

Gert-Jan · October 19, 2018, 04:27

I don't see the problem. The massflow into your domain equals the integral of the normal velocity * density. So, if your velocity component that is normal to your boundary is reduced by a factor 1/sqrt(2), then the massflow is reduced by that factor as well. In fact, the massflow into the domain is not affected by the components perpendicular to the normal component at all. You could apply a strong swirl, or 100 m/s in the other direction....... it will not affect the amount of mass going into your domain.

So, if you want to obtain the same amount of mass on your inlet, you should not multiply by a factor of 1/sqrt(2). Just leave it as it is. You should just make the other direction equal to the normal direction. Then you have 45°.

AtoHM · October 19, 2018, 05:17

Thanks, you are right. I was a bit too focused on the vector thing and ignored your first point, that only normal component is actually going through the patch.
Still seems kind of odd to apply it twice ...

October 19, 2018, 01:59	Inlet Boundary Mass Flow Error	#1
AtoHM Senior Member M Join Date: Dec 2017 Posts: 703 Rep Power: 13	Hi all, I've come across a strange issue. I am basically simulating water flow into a tank. The inlet is a horizontal rectangular patch. The flow direction should be applied with 45 deg inclination. So I used Cart. Vel. Components to apply it. The total inlet velocity is known from case mass flow and inlet patch area. To have it applied with 45 deg, I apply it multiplied with 1/sqrt(2) in both my desired directions. Problem is: mass flow in result and monitor is only 1/sqrt(2) times the amount we were going for. The direction however seems to be 45 deg as we wanted. I looked it up in the modeling guide, where it is stated, that the normal-to-boundary component of the input is ignored. I must be missing something here, but I dont get a) why it is implemented like that and b) how to get my desired mass flow at the inlet. Any help is very appreciated.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Building OpenFOAM1.7.0 from source	ata	OpenFOAM Installation	46	March 6, 2022 14:21
Radiation in semi-transparent media with surface-to-surface model?	mpeppels	CFX	11	August 22, 2019 08:30
Wrong flow in ratating domain problem	Sanyo	CFX	17	August 15, 2015 07:20
[swak4Foam] GroovyBC the dynamic cousin of funkySetFields that lives on the suburb of the mesh	gschaider	OpenFOAM Community Contributions	300	October 29, 2014 19:00
OpenFOAM without MPI	kokizzu	OpenFOAM Installation	4	May 26, 2014 10:17

October 19, 2018, 04:27		#2
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	I don't see the problem. The massflow into your domain equals the integral of the normal velocity * density. So, if your velocity component that is normal to your boundary is reduced by a factor 1/sqrt(2), then the massflow is reduced by that factor as well. In fact, the massflow into the domain is not affected by the components perpendicular to the normal component at all. You could apply a strong swirl, or 100 m/s in the other direction....... it will not affect the amount of mass going into your domain. So, if you want to obtain the same amount of mass on your inlet, you should not multiply by a factor of 1/sqrt(2). Just leave it as it is. You should just make the other direction equal to the normal direction. Then you have 45°.

October 19, 2018, 05:17		#3
AtoHM Senior Member M Join Date: Dec 2017 Posts: 703 Rep Power: 13	Thanks, you are right. I was a bit too focused on the vector thing and ignored your first point, that only normal component is actually going through the patch. Still seems kind of odd to apply it twice ...