Streaking (air under hull)

Loulou · September 8, 2014, 09:51

Hi everybody,

I'm a Fine Marine user and I often get some streaking troubles on hull. I try a lot of settings and didn't find a process to avoid it.
The bows of the studying boats are not in the water so the contact between solid and fluid is under the hull in a flatter area than a traditional bow.

I would like to avoid using unsteady computation to save time. streaking appear for speed higher than 5m/s.
I try a lot of things :
- Use quasi-static in order to fix the box until t = 1.5 * acceleration time
- decrease the time step
- Make a refinement box on the bow to define the Free surface to LOA/1000
- Change Y+ (from 100 to 300)
-I use acceleration ramp as tacc= 2*Lref/Vref, I try to increase it
-Use adaptive grid refinement
-Increase the diffusion to 5

I will be really happy to have new ideas and to hear from your experiences.

Thank you

laurent98 · September 8, 2014, 20:02

hi Loulou,
the only way i found to get rid of this "numerical ventilation" is to play with the pressure threshold . after one computation you look at the pressure on hull, where you want water (for exemple 800...) then you set the parameter (Uw?) to -800
hope it's help?
that was working very well for me.

Loulou · September 9, 2014, 10:35

Hi Laurent,

thank you for your answer, I tried on light model without Viscous Layer and the streaking disappear. Do you still apply the same coefficient or you always run a first calculation to see if there is streaking and use the correction after that. Because in my case, 70% of my computation have streaking, so it can take a long time to test each computation before having the coefficient?

What about using the coefficient range [0;1]?

Thanks

laurent98 · September 9, 2014, 16:40

hi,
it a numerical problem because of the viscous layers.
fist you must add the streaking corrector CIStreakCorrection. if streaking stay then use uwCoefCorrection.
Uw depend on the shape of hull, but mostly of the speed, so after some test you will have a good idea of the right value. by looking at the pressure on hull of the computation you already make you have the threshold. by using the automatic way, i had a lot of wrong computations because FM add water where it was not realistic...
cheers LL

September 8, 2014, 09:51	Streaking (air under hull)	#1
Loulou New Member Join Date: Jan 2014 Posts: 3 Rep Power: 12	Hi everybody, I'm a Fine Marine user and I often get some streaking troubles on hull. I try a lot of settings and didn't find a process to avoid it. The bows of the studying boats are not in the water so the contact between solid and fluid is under the hull in a flatter area than a traditional bow. I would like to avoid using unsteady computation to save time. streaking appear for speed higher than 5m/s. I try a lot of things : - Use quasi-static in order to fix the box until t = 1.5 * acceleration time - decrease the time step - Make a refinement box on the bow to define the Free surface to LOA/1000 - Change Y+ (from 100 to 300) -I use acceleration ramp as tacc= 2*Lref/Vref, I try to increase it -Use adaptive grid refinement -Increase the diffusion to 5 I will be really happy to have new ideas and to hear from your experiences. Thank you

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
air under hull, alpha.water=1 threshold pressure	laurent98	OpenFOAM Programming & Development	2	September 8, 2014 12:47
Radiation interface	hinca	CFX	15	January 26, 2014 18:11
air under hull	laurent98	Fidelity CFD	4	February 18, 2013 10:27
air bubble is disappear increasing time using vof	xujjun	CFX	9	June 9, 2009 08:59
Help!!!problems in air jet into water bath	kim	FLUENT	4	June 9, 2003 08:04

September 8, 2014, 20:02		#2
laurent98 Member laurentL Join Date: Oct 2011 Location: new caledonia Posts: 73 Rep Power: 15	hi Loulou, the only way i found to get rid of this "numerical ventilation" is to play with the pressure threshold . after one computation you look at the pressure on hull, where you want water (for exemple 800...) then you set the parameter (Uw?) to -800 hope it's help? that was working very well for me.

September 9, 2014, 10:35		#3
Loulou New Member Join Date: Jan 2014 Posts: 3 Rep Power: 12	Hi Laurent, thank you for your answer, I tried on light model without Viscous Layer and the streaking disappear. Do you still apply the same coefficient or you always run a first calculation to see if there is streaking and use the correction after that. Because in my case, 70% of my computation have streaking, so it can take a long time to test each computation before having the coefficient? What about using the coefficient range [0;1]? Thanks

September 9, 2014, 16:40		#4
laurent98 Member laurentL Join Date: Oct 2011 Location: new caledonia Posts: 73 Rep Power: 15	hi, it a numerical problem because of the viscous layers. fist you must add the streaking corrector CIStreakCorrection. if streaking stay then use uwCoefCorrection. Uw depend on the shape of hull, but mostly of the speed, so after some test you will have a good idea of the right value. by looking at the pressure on hull of the computation you already make you have the threshold. by using the automatic way, i had a lot of wrong computations because FM add water where it was not realistic... cheers LL