Negative value in scalarTransportFoam

waiter120 · June 12, 2015, 04:01

Hello Foamusers.
I have made my own solver for complex electro-thermal problem with convective term.
This solver is based on scalarTransportFoam. And it use velocity filed that doesn’t change with time.

The main problem is with this convection term for temperature. If I use mesh with small element size and big time step, I get negative local value of temperature or very high temperature that is completely unphysical.
To deal with this I use small time steps and it gives a good results. But main task is to model very long period of time (200-500 hours). So calculation with time step of 1-10 seconds will be to long.

Can anyone suggest me the way other than work with mesh size or time step to break through this problem.
Thanks.

chegdan · June 12, 2015, 11:00

This was actually one of those head scratching questions i had while in graduate school a fe years ago. Have a look at:

http://www.cfd-online.com/Forums/ope...tml#post280210

The problem I had was negative Concentrations on a tracer i was injecting to look at flow patterns

and dead zones. My tracer response curve was laughable (negative values) but this reformulation worked for me. I followed up in that post with where and why....so i hope that helps. Good luck

pnovo · May 5, 2017, 07:13

Hi,
I went through the whole (linked) thread and made the suggested modification and still get negative values in my results. I am testing the concentration profile inside a channel with two inlets, one outlet. First I find the steady-state U field using simpleFoam. Then I use the converged solution as an initial condition in scalarTransportFoam. At the inlets (equal flow rate) I inject 0 and 1 concentrations. After a few iterations ('have been decreasing the controlDict time steps systematically and now I am even doing sub-milli second steps) I can find concentration values to the power of 6 and beyond (also both and negative values). Wonder if you have additional hints to solve this issue?
Regards

tesorieri · May 31, 2021, 13:16

I had a very similar problem with unbounded values using scalarTransportFoam with a fixed velocity field from a previous simpleFoam simulation.

I had two simpleFoam results from a low speed field and a high speed (5 times higher) simulations. The scalarTransportSimulation gave good results for the low speed case, but the high speed always gave unbounded results. I've tried reducing the timestep to a very small number and using diffusive schemes. Nothing worked until I recompiled scalarTransportFoam following the cited thread.

June 12, 2015, 04:01	Negative value in scalarTransportFoam	#1
waiter120 New Member Join Date: Sep 2011 Posts: 15 Rep Power: 15	Hello Foamusers. I have made my own solver for complex electro-thermal problem with convective term. This solver is based on scalarTransportFoam. And it use velocity filed that doesn’t change with time. The main problem is with this convection term for temperature. If I use mesh with small element size and big time step, I get negative local value of temperature or very high temperature that is completely unphysical. To deal with this I use small time steps and it gives a good results. But main task is to model very long period of time (200-500 hours). So calculation with time step of 1-10 seconds will be to long. Can anyone suggest me the way other than work with mesh size or time step to break through this problem. Thanks.

June 12, 2015, 11:00		#2
chegdan Senior Member Daniel P. Combest Join Date: Mar 2009 Location: St. Louis, USA Posts: 621 Rep Power: 0	This was actually one of those head scratching questions i had while in graduate school a fe years ago. Have a look at: http://www.cfd-online.com/Forums/ope...tml#post280210 The problem I had was negative Concentrations on a tracer i was injecting to look at flow patterns and dead zones. My tracer response curve was laughable (negative values) but this reformulation worked for me. I followed up in that post with where and why....so i hope that helps. Good luck Bahram likes this.

May 5, 2017, 07:13	Still geting negative values	#3
pnovo New Member novo Join Date: Jun 2015 Posts: 11 Rep Power: 11	Hi, I went through the whole (linked) thread and made the suggested modification and still get negative values in my results. I am testing the concentration profile inside a channel with two inlets, one outlet. First I find the steady-state U field using simpleFoam. Then I use the converged solution as an initial condition in scalarTransportFoam. At the inlets (equal flow rate) I inject 0 and 1 concentrations. After a few iterations ('have been decreasing the controlDict time steps systematically and now I am even doing sub-milli second steps) I can find concentration values to the power of 6 and beyond (also both and negative values). Wonder if you have additional hints to solve this issue? Regards

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Other] mesh airfoil NACA0012	anand_30	OpenFOAM Meshing & Mesh Conversion	13	March 7, 2022 18:22
[blockMesh] non-orthogonal faces and incorrect orientation?	nennbs	OpenFOAM Meshing & Mesh Conversion	7	April 17, 2013 06:42
[blockMesh] error message with modeling a cube with a hold at the center	hsingtzu	OpenFOAM Meshing & Mesh Conversion	2	March 14, 2012 10:56
[blockMesh] BlockMesh FOAM warning	gaottino	OpenFOAM Meshing & Mesh Conversion	7	July 19, 2010 15:11
[blockMesh] Axisymmetrical mesh	Rasmus Gjesing (Gjesing)	OpenFOAM Meshing & Mesh Conversion	10	April 2, 2007 15:00

May 31, 2021, 13:16		#4
tesorieri New Member hernan Join Date: Mar 2013 Posts: 8 Rep Power: 13	I had a very similar problem with unbounded values using scalarTransportFoam with a fixed velocity field from a previous simpleFoam simulation. I had two simpleFoam results from a low speed field and a high speed (5 times higher) simulations. The scalarTransportSimulation gave good results for the low speed case, but the high speed always gave unbounded results. I've tried reducing the timestep to a very small number and using diffusive schemes. Nothing worked until I recompiled scalarTransportFoam following the cited thread.