Problems detected after running more than 500 iterations

JX- · February 16, 2017, 22:30

dear all,

so i have gotten thru most of the problems i faced earlier and now i have tried solving the problem. But it still leaded me to another dead end where i encountered another issue here...

anybody has any idea what went wrong? the values of Cd Cl and moment is slowly converging but out of a sudden, divergence detected in the solver, epsilon.

here is an image to summarized my topic.

any help is appreciated.

thanks,
JX-

LuckyTran · February 19, 2017, 23:42

It's not all of a sudden really. It took 500 iterations because you lowered the relaxation factors.

As with all convergence problems, check your boundary conditions to make sure you have specified correct ones, make sure you have a good initial guess, and then check your mesh quality.

The COUPLED solver is more unstable, I recommend trying the SIMPLE solver (with all default settings) to see if it still blows up. Also try using the higher-order term relaxation option in Fluent, it helps a bit for bad initial guesses.

JX- · February 21, 2017, 01:45

Quote:

Originally Posted by LuckyTran

It's not all of a sudden really. It took 500 iterations because you lowered the relaxation factors.

As with all convergence problems, check your boundary conditions to make sure you have specified correct ones, make sure you have a good initial guess, and then check your mesh quality.

The COUPLED solver is more unstable, I recommend trying the SIMPLE solver (with all default settings) to see if it still blows up. Also try using the higher-order term relaxation option in Fluent, it helps a bit for bad initial guesses.

this is what i get when i use the SIMPLE solver with all default settings while remaining the other conditions the same as before. i really wonder whats wrong...

LuckyTran · February 21, 2017, 11:31

Looks look typical divergence behavior. Divergence is much easier to detect when you don't tweak anything. =)

Go through your case setup again. Make sure boundary conditions are correct. Make sure you have a good initial guess. Make sure your mesh is okay.

Play with your initial guesses for the k and epsilon. Dbl, triple, quadruple check the specification of k and epsilon at boundaries. Usually people do not specify k and epsilon directly but use the turbulence intensity and something.

February 16, 2017, 22:30	Problems detected after running more than 500 iterations	#1
JX- New Member Jun Join Date: Oct 2016 Posts: 19 Rep Power: 9	dear all, so i have gotten thru most of the problems i faced earlier and now i have tried solving the problem. But it still leaded me to another dead end where i encountered another issue here... anybody has any idea what went wrong? the values of Cd Cl and moment is slowly converging but out of a sudden, divergence detected in the solver, epsilon. here is an image to summarized my topic. any help is appreciated. thanks, JX-

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
settlingFoam unstable?	bendel_boy	OpenFOAM Running, Solving & CFD	38	July 8, 2016 05:07
Extrusion with OpenFoam problem No. Iterations 0	Lord Kelvin	OpenFOAM Running, Solving & CFD	8	March 28, 2016 11:08
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 06:20
Upgraded from Karmic Koala 9.10 to Lucid Lynx10.04.3	bookie56	OpenFOAM Installation	8	August 13, 2011 04:03
Error while running rhoPisoFoam..	nileshjrane	OpenFOAM Running, Solving & CFD	8	August 26, 2010 12:50

February 19, 2017, 23:42		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 66	It's not all of a sudden really. It took 500 iterations because you lowered the relaxation factors. As with all convergence problems, check your boundary conditions to make sure you have specified correct ones, make sure you have a good initial guess, and then check your mesh quality. The COUPLED solver is more unstable, I recommend trying the SIMPLE solver (with all default settings) to see if it still blows up. Also try using the higher-order term relaxation option in Fluent, it helps a bit for bad initial guesses.

February 21, 2017, 11:31		#4
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 66	Looks look typical divergence behavior. Divergence is much easier to detect when you don't tweak anything. =) Go through your case setup again. Make sure boundary conditions are correct. Make sure you have a good initial guess. Make sure your mesh is okay. Play with your initial guesses for the k and epsilon. Dbl, triple, quadruple check the specification of k and epsilon at boundaries. Usually people do not specify k and epsilon directly but use the turbulence intensity and something.