Fully developed flow in heated pipe

Sarutochi · March 29, 2019, 16:35

Hello,

I am modelling a heat flux on the side of a pipe with fully developed flow. In order to obtain fully developed flow I have created a periodic interface between the pressure outlet and velocity inlet. This works for fluid flow but as the energy equation for my model is not converging. I have come to realise that the issues lies in the fact that since the flow coming out at the pressure outlet is at a higher temperature than the flow at inlet, as the flow comes back in at the inlet, there is a sharp temperature gradient which is contrary to what the model expects thus causing an increase in residual.

My question is how can I compensate for that difference while keeping the temperature profile seen at the outlet?

Sarutochi · March 29, 2019, 20:10

Cfd-online residuals.jpg

This is an image with the residuals of my model.

LuckyTran · April 1, 2019, 14:22

Are you sure you set it up right? There is a pressure jump condition and a temperature jump condition. The pressure at the pipe outlet will be less than at the pipe inlet. The interface takes care of this pressure jump. The temperature at the outlet can be greater or less than the inlet depending on your boundary conditions but the interface also takes care of the temperature jump.

Sarutochi · April 1, 2019, 15:48

I have a heat flux on the pipe so it should increase in temperature. I have a constant heat flux walls condition in my interface and I set the temperature specification to bulk inflow temperature. Is that the correct setup?

LuckyTran · April 2, 2019, 08:50

Yes that is correct. The temperature will increase, but the temperature profile will be rescaled back to the inlet bulk temperature.

Sarutochi · April 2, 2019, 11:18

I set it to the initial bulk temperature of my fluid (in this case 293 K), that is correct right?

And if that isn't the problem, what might be causing the residual to fluctuate like it is without converging to a reasonable solution?

LuckyTran · April 4, 2019, 10:42

I can't tell you if the bulk temperature is correct or not. That's like asking me if your birthdate is correct on your ID. I don't have this information. There is a proper boundary condition and you need to set it to whatever it needs to be.

You have general convergence problems. So do regular debugging. Find the numerical oscillations, wherever they are and try to fix them.

March 29, 2019, 16:35	Fully developed flow in heated pipe	#1
Sarutochi New Member Timothy Baines Join Date: Feb 2019 Posts: 6 Rep Power: 7	Hello, I am modelling a heat flux on the side of a pipe with fully developed flow. In order to obtain fully developed flow I have created a periodic interface between the pressure outlet and velocity inlet. This works for fluid flow but as the energy equation for my model is not converging. I have come to realise that the issues lies in the fact that since the flow coming out at the pressure outlet is at a higher temperature than the flow at inlet, as the flow comes back in at the inlet, there is a sharp temperature gradient which is contrary to what the model expects thus causing an increase in residual. My question is how can I compensate for that difference while keeping the temperature profile seen at the outlet? Last edited by Sarutochi; March 29, 2019 at 20:08. Reason: Could not figure out how to upload image

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Issues on the simulation of high-speed compressible flow within turbomachinery	dowlee	OpenFOAM Running, Solving & CFD	11	August 6, 2021 06:40
Turbulence intensity in fully devloped pipe flow - Reference	Shocktrain	Main CFD Forum	0	October 14, 2010 06:58
Fully developed flow ,CEL expression	kmgraju	CFX	0	August 6, 2010 04:15
Fully Developed Flow at an Inlet	Ed Mueller	CFX	5	January 13, 2009 11:50
2D representation of pipe for fully developed flow	Nathan	FLUENT	3	November 22, 2008 10:01

March 29, 2019, 20:10		#2
Sarutochi New Member Timothy Baines Join Date: Feb 2019 Posts: 6 Rep Power: 7	Cfd-online residuals.jpg This is an image with the residuals of my model.

April 1, 2019, 14:22		#3
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 65	Are you sure you set it up right? There is a pressure jump condition and a temperature jump condition. The pressure at the pipe outlet will be less than at the pipe inlet. The interface takes care of this pressure jump. The temperature at the outlet can be greater or less than the inlet depending on your boundary conditions but the interface also takes care of the temperature jump.

April 1, 2019, 15:48		#4
Sarutochi New Member Timothy Baines Join Date: Feb 2019 Posts: 6 Rep Power: 7	I have a heat flux on the pipe so it should increase in temperature. I have a constant heat flux walls condition in my interface and I set the temperature specification to bulk inflow temperature. Is that the correct setup?

April 2, 2019, 08:50		#5
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 65	Yes that is correct. The temperature will increase, but the temperature profile will be rescaled back to the inlet bulk temperature.

April 2, 2019, 11:18		#6
Sarutochi New Member Timothy Baines Join Date: Feb 2019 Posts: 6 Rep Power: 7	I set it to the initial bulk temperature of my fluid (in this case 293 K), that is correct right? And if that isn't the problem, what might be causing the residual to fluctuate like it is without converging to a reasonable solution?

April 4, 2019, 10:42		#7
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 65	I can't tell you if the bulk temperature is correct or not. That's like asking me if your birthdate is correct on your ID. I don't have this information. There is a proper boundary condition and you need to set it to whatever it needs to be. You have general convergence problems. So do regular debugging. Find the numerical oscillations, wherever they are and try to fix them.