Straight through cyclone with particles

rogerhoffmann · May 21, 2012, 15:59

Hi,

I am running some cases of a straight through cyclone, with particles. I am running it steady state, air ideal gas, k-epsilon, with Uniform in Diameter by Mass distribution (see below).

It all seemed to work fine for a series of calculations, but when I was changing some variables like, temperature (increased some degrees), reduced density of the particle material, then I began to obtain results with about 30% - 50% of the particles EXCEEDED INTEGRATION LIMIT, or DISTANCE or TIME. I tried many times to increase this limits, but all I had was huge computational time, and the results did not improve much.

My questions:
1 - Do this particles that are being terminated affect the results, or they can be just ignored?
2 - I understand that this particles are being trapped in some region of the flow. How to discover where is this happening? Could this be a mesh issue?
3 - Is it possible to know which particles are being trapped, where they originate, or their diameter?
4 - I used the Injection at Face Centres at first, but when I wanted to increase the number of particles using uniform injection (to improve the results), I also had lot of particles being terminated. Why? It seems that only using Injection at Face Centres and certain conditions I can run without particles being terminated.

Any tips, suggestions, could help.
Thank you,
Roger

FLUID: Particulas
BOUNDARY CONDITIONS:
MASS AND MOMENTUM:
Option = Zero Slip Velocity
END
PARTICLE DIAMETER DISTRIBUTION:
Maximum Diameter = 1 [mm]
Minimum Diameter = 0.54 [micron]
Option = Uniform in Diameter by Mass
END
PARTICLE MASS FLOW RATE:
Mass Flow Rate = 6.94e-4 [kg s^-1]
END
PARTICLE POSITION:
Option = Injection at Face Centres
END
END

ghorrocks · May 21, 2012, 19:42

I suspect this means that as you changed the conditions then you entered a regime where the particles just go round and round and never exit the domain. This means the particle tracks get very long, computational load goes up and it terminates the tracks at the limits you define (and gives the warning message).

Your questions:
1) If the particles never leave then they might build up into a bed or something. You need to consider whether this is possible.
2) Look at the particle tracks. That should show what they are doing.
3) Yes, all this is available in the post processor.
4) Again, look for why they are terminated. Generally it is because they end up in a vortex and never leave.

May 21, 2012, 15:59	Straight through cyclone with particles	#1
rogerhoffmann New Member Roger Hoffmann Join Date: Apr 2011 Posts: 3 Rep Power: 15	Hi, I am running some cases of a straight through cyclone, with particles. I am running it steady state, air ideal gas, k-epsilon, with Uniform in Diameter by Mass distribution (see below). It all seemed to work fine for a series of calculations, but when I was changing some variables like, temperature (increased some degrees), reduced density of the particle material, then I began to obtain results with about 30% - 50% of the particles EXCEEDED INTEGRATION LIMIT, or DISTANCE or TIME. I tried many times to increase this limits, but all I had was huge computational time, and the results did not improve much. My questions: 1 - Do this particles that are being terminated affect the results, or they can be just ignored? 2 - I understand that this particles are being trapped in some region of the flow. How to discover where is this happening? Could this be a mesh issue? 3 - Is it possible to know which particles are being trapped, where they originate, or their diameter? 4 - I used the Injection at Face Centres at first, but when I wanted to increase the number of particles using uniform injection (to improve the results), I also had lot of particles being terminated. Why? It seems that only using Injection at Face Centres and certain conditions I can run without particles being terminated. Any tips, suggestions, could help. Thank you, Roger FLUID: Particulas BOUNDARY CONDITIONS: MASS AND MOMENTUM: Option = Zero Slip Velocity END PARTICLE DIAMETER DISTRIBUTION: Maximum Diameter = 1 [mm] Minimum Diameter = 0.54 [micron] Option = Uniform in Diameter by Mass END PARTICLE MASS FLOW RATE: Mass Flow Rate = 6.94e-4 [kg s^-1] END PARTICLE POSITION: Option = Injection at Face Centres END END

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
how to determine the number of particles injected.	welch	FLUENT	2	January 18, 2024 05:08
[DPM problem] infinite particles loop in domain (twin cyclone - double cyclone))	ghost82	FLUENT	6	May 19, 2020 15:19
trying to simulate two-phase jet flow with particles in surface injection	ajkratos	FLUENT	5	March 3, 2015 22:33
Cyclone DPM Cylindrical Particles	Mohsin	FLUENT	0	August 3, 2010 04:23
Modelling Industrial cyclone behaviour	Günther Hasse	Main CFD Forum	3	October 12, 1999 20:34

May 21, 2012, 19:42		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	I suspect this means that as you changed the conditions then you entered a regime where the particles just go round and round and never exit the domain. This means the particle tracks get very long, computational load goes up and it terminates the tracks at the limits you define (and gives the warning message). Your questions: 1) If the particles never leave then they might build up into a bed or something. You need to consider whether this is possible. 2) Look at the particle tracks. That should show what they are doing. 3) Yes, all this is available in the post processor. 4) Again, look for why they are terminated. Generally it is because they end up in a vortex and never leave.