heatSource for coupled simulation

Submitted by mardussa on Mon, 09/24/2018 - 06:16

Hi all,

I have been trying to implement a case similar to the packedBedTemp cfdemSolverPisoScalar example case where hot particles are cooled by a flowing continuous phase. For my case I will not know the temperature of the particles, but instead the power (in Watts) of each particle so I am thinking I will need to use the heatSource model instead of the Temperature model within heat/gran. I have read the brief discussion of it in the user manual, but I am having trouble replicating results I have gotten through commercial codes.

I have implemented the particles and heat generation as follows in my in.liggghts_run file:

# cfd coupling
fix cfd all couple/cfd couple_every 100 mpi
fix cfd2 all couple/cfd/force


# heat transfer
fix ftco all property/global thermalConductivity peratomtype 200. # lambda [W/(K*m)]
fix ftca all property/global thermalCapacity peratomtype 900. # cp [J/(kg*K)]
fix cfd3 all couple/cfd/convection T0 300.
fix heattransfer all heat/gran initial_temperature 300.


# set particle temperature for the bed
run 0
region total block INF INF INF INF INF INF units box
set region total property/atom heatSource 10

I have currently got 100 particles in my simulation, and from what I understand the definition of heatSource would result in each particle having a power of 10W. Is this incorrect? I ask this because I am not seeing any significant change in the temperature of the fluid (water). Even after a 50 second simulation there is an insignificant change in fluid temperature, whilst the temperature of the particles is increasing rapidly (temperature difference between particles and fluid is on the order of 50 degrees).

Does anyone know anything more concerning the heatSource model? Should I not be setting things such as T0 and initial_temperature?

Kind regards,
Robert

Forums:

mardussa | Mon, 09/24/2018 - 06:20

I have also commented out most of the /CFD/constant/liggghtsCommands file so that it now reads simply:

liggghtsCommandModels
(
runLiggghts
);

And does not reset the particle temperature each coupling interval as I believe the example case does.

paul | Mon, 09/24/2018 - 08:23

Have you tried outputting f_heatSource for each particle in your dumps? That way you can check whether you have set it correctly.

mardussa | Tue, 09/25/2018 - 02:51

Thank you again Paul. I included f_heatSource in my dump files, and it appears that each particle has the specified 10W heatSource value as I had hoped. I therefore don't understand why my particles are not passing on this energy to the fluid. There is a large difference between the water and the particle temperature, yet I am seeing only a 0.1 degree increase in water temperature after 100 seconds of simulated time.
Perhaps there is an error in my material properties some where. I will check these and look around for anything else that seems incorrect.
Thank you,
Robert.

mardussa | Thu, 10/04/2018 - 08:41

Just an update on my past post, it turns out the temperature gain of the fluid was realistic, I was just no paying attention to how much I had scaled the simulation down from my old results.