Particle fate criteria

sakurabogoda · March 17, 2013, 01:19

Dear All,

I have a doubt about particle fate criteria.

I have a particle sample (0-10um) and I want to find the collection efficiency of those in a cyclone separator.

Then CFX output file shows there are particles which exceeded time, integration and distance limits.

My problem is how to know do I want to increase values of these criteria or not? Because some particles should end-up with one of these fates but some particles should leave domain. Still nothing has leaved.

Many thanks for helps..

ghorrocks · March 17, 2013, 04:44

If the particles have long residence times you might have to increases these particle termination parameters.

sakurabogoda · March 17, 2013, 23:28

Dear Glenn,
Thanks a lot for the reply.

Actually I don't know the residence time and have to find from this simulation.

Total simulation time is 10s here and I am using 12m/s inlet velocity for particles. I guess particle residence time is less than 2min, but even after 5min nothing leaves in simulations.

I can see there is a strong recirculating zone inside the cyclone.

I have done experiments but there particles comes out from exit, after 1s.

I cannot find the reason why simulations are going on like this?

Do you have any idea?

oj.bulmer · March 18, 2013, 04:14

Are you sure you are injecting enough particles to do justice to probability of both occurrences? Why not do a bit literature search? Or, perhaps the particle number independence test etc?

OJ

sakurabogoda · March 18, 2013, 04:42

Dear OJ, Thanks a lot for reply.

The particles having the range of 0-10um, so there should be particles trapped by geometry and leave the domain.

If I understand you correctly, you are asking do I know whether particles exit or not? Then I know compared to assigned particle range at least 1/2 of particles should be leave the domain. And other 1/2 should be trapped inside.

So, my problem is, still there are no leaving particles but particles terminated by exceeding time, distance and integration limit, so I have increased them.

But I cannot find why particles are not leaving the domain?

oj.bulmer · March 18, 2013, 05:31

Okay, I would pitch it in different way. We say that if we toss a coin probability of it being heads or tails is 50%. But this argument is valid only if we toss the coin enough times! If we toss it only four times and result is heads for 3 times, it is not that probability of heads is 75%, but that we have not tossed coin enough times.

I was hinting at the fact that there has to be enough number of particles entering through the inlet so that some would be trapped and some would escape. If they aren't enough, then the result may not be ideal representation.

How are you sure you have specified the appropriate number of particles in appropriate distribution (their spacing and position with respect to each other) at the inlet? If you aren't sure, doesn't it make sense to ensure this distribution as per the experimental data you have? If you can't determine the particle distribution and number, then try literature search to get some pointers that will guide you decide it.

You can't expect the same results as experiment, without same/adequate boundary conditions.

OJ

sakurabogoda · March 18, 2013, 06:27

Dear Oj,
Thanks a lot. I got your point. But I dont think it is going on my simulation. Please refer follows to get an idea on particles that I assigned.

MATERIAL: Particles
Material Group = Particle Solids
Option = Pure Substance
PROPERTIES:
Option = General Material
EQUATION OF STATE:
Density = 500 [kg m^-3]
Molar Mass = 1.0 [kg kmol^-1]
Option = Value
------------------------------------------
FLUID: particles
BOUNDARY CONDITIONS:
MASS AND MOMENTUM:
Normal Speed = 12 [m s^-1]
Option = Normal Speed
END
PARTICLE MASS FLOW RATE:
Mass Flow Rate = 0.05 [g s^-1]
END
PARTICLE POSITION:
Option = Uniform Injection
NUMBER OF POSITIONS:
Number per Unit Time = 25 [s^-1]
Option = Direct Specification

-----------------------------------------------
FLUID DEFINITION: particles
Material = Particles
Option = Material Library
MORPHOLOGY:
Option = Dispersed Particle Transport Solid
PARTICLE DIAMETER DISTRIBUTION:
Maximum Diameter = 10.08 [micron]
Mean Diameter = 3.78 [micron]
Minimum Diameter = 0.647 [micron]
Option = Normal in Diameter by Number
Standard Deviation in Diameter = 2.44 [micron]
----------------------------------------------------------

Could you pls tell me any thing wrong? I cannot find the mistake.

Thank you.

oj.bulmer · March 18, 2013, 08:59

Quote:

Number per Unit Time = 25 [s^-1]

Alright, here is the question. How did you arrive at 25 particles/s value?

Quote:

PARTICLE DIAMETER DISTRIBUTION:
Maximum Diameter = 10.08 [micron]
Mean Diameter = 3.78 [micron]
Minimum Diameter = 0.647 [micron]

How did you arrive at this particle size distribution? (I am assuming here that the particle density is right).

The answer to these questions may help you model your simulation more relevantly. From what I know, the separation efficiency curve slowly rises from smaller particles and gets flatter for larger particles with highest efficiency. Whilst, correct flow rate of particles and fluid is also relevant, since the efficiency decreases with higher flow rate.

So your particle number rate and particle size distribution should greatly influence the efficiency.

OJ

sakurabogoda · March 18, 2013, 09:31

NUMBER OF POSITIONS:
Number per Unit Time = 25 [s^-1]

This value is aasigned by me. As far as I understood, I can assign representative particles into any value? Am I wrong?

PARTICLE DIAMETER DISTRIBUTION:
Maximum Diameter = 10.08 [micron]
Mean Diameter = 3.78 [micron]
Minimum Diameter = 0.647 [micron]

This normal distribution is relevent with particle sizes which I am going to use in experiments.

Still I stuck on this and I would really greatful for advices.

oj.bulmer · March 18, 2013, 09:54

Quote:

This value is aasigned by me. As far as I understood, I can assign representative particles into any value? Am I wrong?

In your experiment, if you put a palm at your inlet for 1 second and if you observe you have collected 25 particles, then the boundary condition you assigned is right. Otherwise it is not. You can not assign "any" value and still expect your results be same as experimental results. This is the point I am getting at.

Regarding distribution, you may want to check the standard deviation. The size distribution should greatly impact the results.

If these basic things are right, then only you can move ahead and troubleshoot for other possibilities.

OJ

sakurabogoda · March 18, 2013, 12:54

Dear OJ,
Thanks a lot for reply, but I have a doubt. My understanding in particle number rate is,

Number rate= mass flow rate assigned/ mass of actual particle

Here, if I find this for mean particle diameter (3.78um), number rate is 1.13E15/s. So if I assigned 25/s, then one particle position indicate 1.13E15/25 partilce.

Am I wrong?

Really appriciate your helps..

March 17, 2013, 01:19	Particle fate criteria	#1
sakurabogoda Senior Member S.Bogoda Join Date: Jul 2012 Posts: 133 Rep Power: 13	Dear All, I have a doubt about particle fate criteria. I have a particle sample (0-10um) and I want to find the collection efficiency of those in a cyclone separator. Then CFX output file shows there are particles which exceeded time, integration and distance limits. My problem is how to know do I want to increase values of these criteria or not? Because some particles should end-up with one of these fates but some particles should leave domain. Still nothing has leaved. Many thanks for helps..

March 18, 2013, 04:14		#4
oj.bulmer Senior Member OJ Join Date: Apr 2012 Location: United Kindom Posts: 473 Rep Power: 20	Are you sure you are injecting enough particles to do justice to probability of both occurrences? Why not do a bit literature search? Or, perhaps the particle number independence test etc? OJ Last edited by oj.bulmer; March 18, 2013 at 04:38.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Particle tracking prob, urgent.	sakurabogoda	CFX	1	March 11, 2013 21:11
forced to sticking of soot particle	kmgraju	CFX	0	November 27, 2012 09:08
Particle Transport Modelling - Fate Diagnostics	leeportnoff	CFX	9	January 12, 2012 09:48
Particle fate "Integration error"	ghorrocks	CFX	5	December 13, 2011 11:03
DPM UDF particle position using the macro P_POS(p)[i]	dm2747	FLUENT	0	April 17, 2009 01:29

March 17, 2013, 04:44		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	If the particles have long residence times you might have to increases these particle termination parameters.

March 17, 2013, 23:28		#3
sakurabogoda Senior Member S.Bogoda Join Date: Jul 2012 Posts: 133 Rep Power: 13	Dear Glenn, Thanks a lot for the reply. Actually I don't know the residence time and have to find from this simulation. Total simulation time is 10s here and I am using 12m/s inlet velocity for particles. I guess particle residence time is less than 2min, but even after 5min nothing leaves in simulations. I can see there is a strong recirculating zone inside the cyclone. I have done experiments but there particles comes out from exit, after 1s. I cannot find the reason why simulations are going on like this? Do you have any idea?

March 18, 2013, 04:42		#5
sakurabogoda Senior Member S.Bogoda Join Date: Jul 2012 Posts: 133 Rep Power: 13	Dear OJ, Thanks a lot for reply. The particles having the range of 0-10um, so there should be particles trapped by geometry and leave the domain. If I understand you correctly, you are asking do I know whether particles exit or not? Then I know compared to assigned particle range at least 1/2 of particles should be leave the domain. And other 1/2 should be trapped inside. So, my problem is, still there are no leaving particles but particles terminated by exceeding time, distance and integration limit, so I have increased them. But I cannot find why particles are not leaving the domain?

March 18, 2013, 05:31		#6
oj.bulmer Senior Member OJ Join Date: Apr 2012 Location: United Kindom Posts: 473 Rep Power: 20	Okay, I would pitch it in different way. We say that if we toss a coin probability of it being heads or tails is 50%. But this argument is valid only if we toss the coin enough times! If we toss it only four times and result is heads for 3 times, it is not that probability of heads is 75%, but that we have not tossed coin enough times. I was hinting at the fact that there has to be enough number of particles entering through the inlet so that some would be trapped and some would escape. If they aren't enough, then the result may not be ideal representation. How are you sure you have specified the appropriate number of particles in appropriate distribution (their spacing and position with respect to each other) at the inlet? If you aren't sure, doesn't it make sense to ensure this distribution as per the experimental data you have? If you can't determine the particle distribution and number, then try literature search to get some pointers that will guide you decide it. You can't expect the same results as experiment, without same/adequate boundary conditions. OJ

March 18, 2013, 06:27		#7
sakurabogoda Senior Member S.Bogoda Join Date: Jul 2012 Posts: 133 Rep Power: 13	Dear Oj, Thanks a lot. I got your point. But I dont think it is going on my simulation. Please refer follows to get an idea on particles that I assigned. MATERIAL: Particles Material Group = Particle Solids Option = Pure Substance PROPERTIES: Option = General Material EQUATION OF STATE: Density = 500 [kg m^-3] Molar Mass = 1.0 [kg kmol^-1] Option = Value ------------------------------------------ FLUID: particles BOUNDARY CONDITIONS: MASS AND MOMENTUM: Normal Speed = 12 [m s^-1] Option = Normal Speed END PARTICLE MASS FLOW RATE: Mass Flow Rate = 0.05 [g s^-1] END PARTICLE POSITION: Option = Uniform Injection NUMBER OF POSITIONS: Number per Unit Time = 25 [s^-1] Option = Direct Specification ----------------------------------------------- FLUID DEFINITION: particles Material = Particles Option = Material Library MORPHOLOGY: Option = Dispersed Particle Transport Solid PARTICLE DIAMETER DISTRIBUTION: Maximum Diameter = 10.08 [micron] Mean Diameter = 3.78 [micron] Minimum Diameter = 0.647 [micron] Option = Normal in Diameter by Number Standard Deviation in Diameter = 2.44 [micron] ---------------------------------------------------------- Could you pls tell me any thing wrong? I cannot find the mistake. Thank you.

March 18, 2013, 09:31		#9
sakurabogoda Senior Member S.Bogoda Join Date: Jul 2012 Posts: 133 Rep Power: 13	NUMBER OF POSITIONS: Number per Unit Time = 25 [s^-1] This value is aasigned by me. As far as I understood, I can assign representative particles into any value? Am I wrong? PARTICLE DIAMETER DISTRIBUTION: Maximum Diameter = 10.08 [micron] Mean Diameter = 3.78 [micron] Minimum Diameter = 0.647 [micron] This normal distribution is relevent with particle sizes which I am going to use in experiments. Still I stuck on this and I would really greatful for advices.

March 18, 2013, 12:54		#11
sakurabogoda Senior Member S.Bogoda Join Date: Jul 2012 Posts: 133 Rep Power: 13	Dear OJ, Thanks a lot for reply, but I have a doubt. My understanding in particle number rate is, Number rate= mass flow rate assigned/ mass of actual particle Here, if I find this for mean particle diameter (3.78um), number rate is 1.13E15/s. So if I assigned 25/s, then one particle position indicate 1.13E15/25 partilce. Am I wrong? Really appriciate your helps..