Cavitation: Volume fraction and pressure

kandolf · June 15, 2009, 05:49

Model: I am working on simulating a vibratory cavitation test system. I would like to model 1 cycle of transducer motion (goes up and comes down - 15 microns at 20KHz) to calculate the resulting pressure changes in the fluid, liquid force exerted on the transducer e.t.c.

Since its a cylinder transducer, I only model a small section of 15 degrees of the transducer as a wall (moving mesh). The domain 0.5mm thick contains: an opening, moving wall (top), stationary wall (bottom) and 2 symmetry boundaries (sides). I have water and water vapour as materials and use Rayleigh-plesset model.

Questions:
1. I am pretty sure that this system generates cavitation, should I use minimum volume fraction: e.g 75% for water and 25% for vapour or use 1 and 0? and why?

2. I have no idea about the pressure values at the opening. How much should I specify? any good approximation? or is there any alternate way to give details for opening boundary?

3. The resulting pressures should be high about several hundered MPa (thats all I know). Is it okay to get both high negative and positive pressures?

4. I get about 200N force on the transducer for Vol.fraction 75 and 25%, is this reasonable? is there a way to calculate and compare?

I would appreciate if anyone could please help me on this. Please also let me know if you have come across any similar problem.

Many thanks,
Josef.

ghorrocks · June 15, 2009, 07:58

Some comments (I won't go so far as to call them answers):

1) I think the default minimum VF is 1e-10 and the maximum is 1.0. Don't change these values unless you have a really good reason.

2) If you don't know the pressure at the opening then you need to move your boundary further back until you have somewhere you can define it. I think you will find acoustic waves are important in this flow and they can travel a long way up pipes and ducts. I reckon you need to take it back to a reservoir/atmosphere. If you don't have a reservoir/atmosphere then I reckon your ducting may affect the acoustic waves and therefore may affect the result - just something to keep in mind.

3) High positive pressures are possible but pressures below zero absolute are not. Sometimes a model using incompressible fluids gives pressures below absolute zero - when you use a cavitation model the negative pressures should not be present and you should have cavitated regions.

4) I have no idea. It obviously depends on your design. You need to do the normal mesh, convergence, time step, boundary proximity and physics checks before you can have confidence in the answer.

Glenn Horrocks

kandolf · June 16, 2009, 02:40

Glenn, Thanks a lot for your feedback. I will leave the volume fraction to default. The trouble is with the opening – the system is a small beaker like structure with the transducer immersed in, I will have to re-think to position the opening in the model. Pressure values I get are okay (few MPa), but I am struggling to identify the right one without properly defining the opening. Much appreciated for your help!

June 15, 2009, 05:49	Cavitation: Volume fraction and pressure	#1
kandolf New Member josef kandolf Join Date: Jun 2009 Posts: 2 Rep Power: 0	Model: I am working on simulating a vibratory cavitation test system. I would like to model 1 cycle of transducer motion (goes up and comes down - 15 microns at 20KHz) to calculate the resulting pressure changes in the fluid, liquid force exerted on the transducer e.t.c. Since its a cylinder transducer, I only model a small section of 15 degrees of the transducer as a wall (moving mesh). The domain 0.5mm thick contains: an opening, moving wall (top), stationary wall (bottom) and 2 symmetry boundaries (sides). I have water and water vapour as materials and use Rayleigh-plesset model. Questions: 1. I am pretty sure that this system generates cavitation, should I use minimum volume fraction: e.g 75% for water and 25% for vapour or use 1 and 0? and why? 2. I have no idea about the pressure values at the opening. How much should I specify? any good approximation? or is there any alternate way to give details for opening boundary? 3. The resulting pressures should be high about several hundered MPa (thats all I know). Is it okay to get both high negative and positive pressures? 4. I get about 200N force on the transducer for Vol.fraction 75 and 25%, is this reasonable? is there a way to calculate and compare? I would appreciate if anyone could please help me on this. Please also let me know if you have come across any similar problem. Many thanks, Josef.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Error message: Insufficient Catalogue Size	Paresh Jain	CFX	33	August 16, 2024 06:09
Simulation of a single bubble with a VOF-method	Suzzn	CFX	21	January 29, 2018 01:58
air bubble is disappear increasing time using vof	xujjun	CFX	9	June 9, 2009 08:59
How to apply negtive pressure to outlet	bioman66	CFX	5	June 3, 2006 02:40
Volume vs Quantity vs Pressure	L Jones	Main CFD Forum	1	March 1, 2003 07:16

June 15, 2009, 07:58		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Some comments (I won't go so far as to call them answers): 1) I think the default minimum VF is 1e-10 and the maximum is 1.0. Don't change these values unless you have a really good reason. 2) If you don't know the pressure at the opening then you need to move your boundary further back until you have somewhere you can define it. I think you will find acoustic waves are important in this flow and they can travel a long way up pipes and ducts. I reckon you need to take it back to a reservoir/atmosphere. If you don't have a reservoir/atmosphere then I reckon your ducting may affect the acoustic waves and therefore may affect the result - just something to keep in mind. 3) High positive pressures are possible but pressures below zero absolute are not. Sometimes a model using incompressible fluids gives pressures below absolute zero - when you use a cavitation model the negative pressures should not be present and you should have cavitated regions. 4) I have no idea. It obviously depends on your design. You need to do the normal mesh, convergence, time step, boundary proximity and physics checks before you can have confidence in the answer. Glenn Horrocks

June 16, 2009, 02:40		#3
kandolf New Member josef kandolf Join Date: Jun 2009 Posts: 2 Rep Power: 0	Glenn, Thanks a lot for your feedback. I will leave the volume fraction to default. The trouble is with the opening – the system is a small beaker like structure with the transducer immersed in, I will have to re-think to position the opening in the model. Pressure values I get are okay (few MPa), but I am struggling to identify the right one without properly defining the opening. Much appreciated for your help!