Max Iteration/Time Step on acoustic analysis

recep911234 · January 10, 2019, 20:52

im working on naca0012 and i modificate the profile in leading edge. I have to reduce the spl so i should get the acoustic analysis. But my computer not so good for 300k+ mesh and fluent solve very slow. i want to frequency range 10^4 so i set the time step size 5e-7 and number of time step 4000 but it took a day to solve this acoustic analysis. Fluent default settings on max iteration/timestep is 20 and i want to reduce to 3 but i didnt know how it effect on my analysis. is it so important to calculation?
thx ^^

LuckyTran · January 10, 2019, 22:26

You've already run 20 iterations per time-step so you have the convergence history. Just look at it and you'll know the answer to your question.

And is this SIMPLE/PISO/COUPLED or density based solver? If PISO/COUPLED and you satisfy the Courant number < 1 criteria, you should only be using 1 iteration per time-step.

recep911234 · January 11, 2019, 05:34

Quote:

Originally Posted by LuckyTran

You've already run 20 iterations per time-step so you have the convergence history. Just look at it and you'll know the answer to your question.

And is this SIMPLE/PISO/COUPLED or density based solver? If PISO/COUPLED and you satisfy the Courant number < 1 criteria, you should only be using 1 iteration per time-step.

I solve simple solver and im beginner on the cfd i didnt get it courant number and i confused. Should i use piso or coupled is this better option for me? Thx for answer btw

I set the max iteration per stepsize 3 and i get the solution converged message on 137.iteration. What does it means?

LuckyTran · January 11, 2019, 11:30

Stay SIMPLE since you are new. PISO requires much smaller time-steps for accuracy and if you don't know what Courant number is then there's no point going to PISO. COUPLED can act like SIMPLE/PISO depending on settings you use. So, stick with SIMPLE.

Make some point monitors and plot (or print to console) the velocity vs iteration, pressure vs iteration and/or temperature vs iteration. Run your case with 20 iteratinos per time-step. Watch those values to make sure they converge to some tolerance that you like (i.e. 3/4/5/6 significant digits). That will tell you how many iterations it takes to converge. It will also answer your question of what the effect will be. As you make the time-step smaller, the number of iterations needed will be fewer and more as your time-step is larger. Since you don't know Courant number, I have no clue how small/big your time-step is. There is a Courant number field in Fluent. You should look at it.

January 10, 2019, 20:52	Max Iteration/Time Step on acoustic analysis	#1
recep911234 New Member kaan Join Date: Nov 2018 Posts: 7 Rep Power: 8	im working on naca0012 and i modificate the profile in leading edge. I have to reduce the spl so i should get the acoustic analysis. But my computer not so good for 300k+ mesh and fluent solve very slow. i want to frequency range 10^4 so i set the time step size 5e-7 and number of time step 4000 but it took a day to solve this acoustic analysis. Fluent default settings on max iteration/timestep is 20 and i want to reduce to 3 but i didnt know how it effect on my analysis. is it so important to calculation? thx ^^

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Adapt Fluent Mesh to weirOverflow tutorial	WaterHammer1985	OpenFOAM Running, Solving & CFD	2	July 5, 2018 03:37
multiphase turbulance case floating error	harsha_kulkarni	OpenFOAM Running, Solving & CFD	3	February 18, 2016 06:06
Problem of simulating of small droplet with radius of 2mm	liguifan	OpenFOAM Running, Solving & CFD	5	June 3, 2014 03:53
dynamic Mesh is faster than MRF????	sharonyue	OpenFOAM Running, Solving & CFD	14	August 26, 2013 08:47
Upgraded from Karmic Koala 9.10 to Lucid Lynx10.04.3	bookie56	OpenFOAM Installation	8	August 13, 2011 05:03

January 10, 2019, 22:26		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,762 Rep Power: 66	You've already run 20 iterations per time-step so you have the convergence history. Just look at it and you'll know the answer to your question. And is this SIMPLE/PISO/COUPLED or density based solver? If PISO/COUPLED and you satisfy the Courant number < 1 criteria, you should only be using 1 iteration per time-step.

January 11, 2019, 11:30		#4
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,762 Rep Power: 66	Stay SIMPLE since you are new. PISO requires much smaller time-steps for accuracy and if you don't know what Courant number is then there's no point going to PISO. COUPLED can act like SIMPLE/PISO depending on settings you use. So, stick with SIMPLE. Make some point monitors and plot (or print to console) the velocity vs iteration, pressure vs iteration and/or temperature vs iteration. Run your case with 20 iteratinos per time-step. Watch those values to make sure they converge to some tolerance that you like (i.e. 3/4/5/6 significant digits). That will tell you how many iterations it takes to converge. It will also answer your question of what the effect will be. As you make the time-step smaller, the number of iterations needed will be fewer and more as your time-step is larger. Since you don't know Courant number, I have no clue how small/big your time-step is. There is a Courant number field in Fluent. You should look at it.