[CeesH]

Dear all,

I am currently modeling a system with an external pumped cooling loop. I have no need to model the loop itself explicitly, so it's included as an inlet/outlet system with a forced flowrate. I add massless particles to my system to represent micro-organisms living inside the tank, which have to be tracked for long timespans. These particles can physically enter the cooling loop, and should not leave the system. Hence, I'm aiming to re-inject the particles that leave the outlet at the inlet of the system, to allow for this prolonged tracking without having to resolve the flow-loop completely.

However, I have trouble finding a proper solution to this. Conceptually, I recognize the following possibilities [with issue in brackets]:

DPM_BC at outlet, X particles leave at time t--> surface injection or group injection of X particles at the inlet [but I can not dynamically control the number of streams X in either case]

DPM_BC at outlet, re-inject particles via a file [but this requires dynamic updating of the file to adapt the number of particles, and seems to be a rather slow option computationally as such]

Change TP_POS(tp) and TP_VEL(tp) for particles close to the outlet, to "teleport" them to the inlet [but this simply doesn't seem to work - the position is not changeable directly]


Do any of you have any (conceptual) solutions to this problem?

Best regards,
Cees

[ullaspk1990]

Originally Posted by CeesH

Dear all,

I am currently modeling a system with an external pumped cooling loop. I have no need to model the loop itself explicitly, so it's included as an inlet/outlet system with a forced flowrate. I add massless particles to my system to represent micro-organisms living inside the tank, which have to be tracked for long timespans. These particles can physically enter the cooling loop, and should not leave the system. Hence, I'm aiming to re-inject the particles that leave the outlet at the inlet of the system, to allow for this prolonged tracking without having to resolve the flow-loop completely.

However, I have trouble finding a proper solution to this. Conceptually, I recognize the following possibilities [with issue in brackets]:

DPM_BC at outlet, X particles leave at time t--> surface injection or group injection of X particles at the inlet [but I can not dynamically control the number of streams X in either case]

DPM_BC at outlet, re-inject particles via a file [but this requires dynamic updating of the file to adapt the number of particles, and seems to be a rather slow option computationally as such]

Change TP_POS(tp) and TP_VEL(tp) for particles close to the outlet, to "teleport" them to the inlet [but this simply doesn't seem to work - the position is not changeable directly]


Do any of you have any (conceptual) solutions to this problem?

Best regards,
Cees

Did you find an answer to this. I am also trying to get an idea regarding the same. It would be helpful if anyone can suggest some methods. Thanks in advance

[vinerm]

It depends on whether the simulation is steady or transient. If it is steady, then it does not really make sense to transfer particles from the outlet to the inlet. If it is transient, then it does make sense because number of particles will increase slowly. However, since the particles are massless, they do not add any value. So, 100 particles or 1000 particles are all the same. Except that you will be able to see 1000 particle tracks.

[ullaspk1990]

Originally Posted by vinerm

It depends on whether the simulation is steady or transient. If it is steady, then it does not really make sense to transfer particles from the outlet to the inlet. If it is transient, then it does make sense because number of particles will increase slowly. However, since the particles are massless, they do not add any value. So, 100 particles or 1000 particles are all the same. Except that you will be able to see 1000 particle tracks.

Thank you for the reply. I am trying to do unsteady simulation and I want to inject a particular number of particles and then I will maintain the amount of particles constant but I need to re-circulate the particles escaped from the outlet. The re-circulation of escaped particles from the outlet is the problem I am facing now.

[vinerm]

Why do you want particles to recirculate? It's the objective of doing this that defines the approach to be used.

[ullaspk1990]

Originally Posted by vinerm

Why do you want particles to recirculate? It's the objective of doing this that defines the approach to be used.

I am actually working on air purification. So I am passing the air with particles through a filter and the remaining particles which escaped through the outlet need to be recirculated. Thanks in advance

[vinerm]

You can use UDF, DEFINE_DPM_BC. It acts on those particles that hit a particular boundary where the function is hooked. Within this UDF, you can change the positions of all the particles to their initial injection position. Look at

https://www.afs.enea.it/project/nept...udf/node66.htm

[ullaspk1990]

Originally Posted by vinerm

You can use UDF, DEFINE_DPM_BC. It acts on those particles that hit a particular boundary where the function is hooked. Within this UDF, you can change the positions of all the particles to their initial injection position. Look at

https://www.afs.enea.it/project/nept...udf/node66.htm

Thank you very much for your help. I will try this way.