Hello, I'm a french student. I'm workin on fluent, and I've got to fill in the Under relaxation factors box. But I don't know what is really these factors. Can somebody help me ?

Thanks !

Cyril

If somebody knows more about the discretization options too !...

Read the Chapter "Using the solver" in the users guide (Chapter 26 for fluent 6.2). Its all there.

RoM

As I told you, I'm a student, and I do not have any access to this User Guide (maybe some teacher's got it, but I don't know where to find it.) Is there any place on the internet to find it ?

Cyril

There is no documentation installed with your fluent version? How can your teacher expect you to work with fluent without any docs? The full documentation as pdf (~245MB) can be downloaded from fluents service center www.fluentusers.com . You will need a login /password, maybe your teacher can provide it. An online version can be found here

http://www.hpcu.uq.edu.au/Manuals/Fl...nt6.2.16/html/

Good Luck

RoM

Thanks! By the way... I'm student in an engineering school, specialised in aeronautics. I'm looking for an end-of-study placement... Are you working in that area?

Cyril

I am working in the field of combustion/gasification so i am probably of little help in aeronautics business. I wish you good luck to find the right job in the right place (i know its hard).

RoM

we have equations descretised in the form ap * phi_p = sum(al * phi_neigh) + su now

there are two ways one can implement the under relaxation factors, usually :

ap = ap / urf

su = su + (1-urf) * ap * phi

and calculate new phi

and another way is :

calculate new phi and

phi = phi_old + urf * (phi_new - phi_old)

in the first way, we can increase the diagonal dominance and hence we can increase the convergence. so in this way if you increase the urf too much, ap might be unnecesaryly increased and may goof up the solution altogether.

using urf means that we allow the solution to change little bit from one iteration to another, the reason is we solve for so many scalars that a high change in one variable might make the solution unstable.

fluent uses the second approach for applying urfs, and usually the default values are fine, but if you see a problem in getting convergence, you might want to go to press urf = 0.1 and mom =0.5, they mostly work well, but in some high mach flows we could have 0.7 for press and 0.5 for mom, , but usually we keep pressure urf small.

Well, I have a slihgtly different question. I am working on Coanda ejector flows in which the secondary fluid is dragged by a primary fluid ejected from a nozzle and it is flowing along a curved wall(coanda effect). I have used Coupled implicit solver and Realisable K-E (with enhanced wall treatment) and K-W models to solve the flow field. I have given very high urfs and for K-w model the residuals show lots of oscillations and haven't converged properly. When I use RK-E the residulas converged very well to 10^-4 without showing much oscillation. But the predicted values are almost 1.5 times greater than that of the experiment at the mixing section of the two flows and predicted values at completely mixed section is less than by a factor of 0.6.In this case I used low urfs (0.3). My question is since the flow has a high turbulent mixing zone where turbulent parameters play a major role, whether changing the urfs will improve the computational values or urfs are just meant for improving the stability of the solutions?