hi all:

I write a code for compressible flow with Godunov scheme. The speed of the flow is almost 100 m/s. I want to know that: Can I used the code in incomprebbile flow or the low speed flow?

It depends how low you want to go, the stability of the scheme and if the scheme includes stabilising terms that are invalid at low Mach numbers. All compressible codes that are not designed to switch to a low Mach number form will become invalid/fail at some point as the Mach number is reduced.

Since the incompressible equations are different to the compressible equations no compressible code can solve incompressible flow.

Dear Ricklee,

In your case since the speed is nearly one-third of the sonic speed (~ 330m/s), the Mach number is around 0.3, and there should not be much difficulty in getting your results. But as Andy rightly pointed out, this does not mean that you can go as low as you like, there are restrictions placed on the code because you are solving compressible flow. However, if your interests lie in solving very low speed flow (say 0.25 mach and below), and always be in a compressible framework, then you could try out the pseudo-compressibilty method. The idea is basically to convert the incompressible equations into compressible-like form by addition of a pseudo parameter and then the ideas developed in compressible flows can be directly applied.

Hope this helps Regards,

Ganesh

The type of the pde changes with change of compressibility and you might have some problems with using a compressible code for an incompressible problem.

-H

As pointed out in previous comments, you can some type of low-Mach number preconditioning or pseudo-compressibility technique to allow solving compressible flow equations at very low Mach numbers. One such technique is that given in the paper

J. Computational Phys, vol 5, pp.207-223, 1993

Y-H Choi and C. L. Merkle

and various later papers by Merkle and his students. If you look at the above paper, they show stability and flow results for cases with Mach numbers as low as 1e-05 without convergence penalties and similar accuracy compared to high Mach number runs. I have also personally run cases with such low Mach numbers (down to 1e-08) using Merkle type preconditioning. The flow may not be exactly incompressible (ie, density will vary), but density variations in the 5th or 6th significant figure for very low Mach number flows is incompressible enough for me

Dear all:

I read some information about preconditioning. My code was

use approximaiton Godunov method. so it doesn't need to

iternation. Can I use the preconditioning in my code?

Besides,I test my code in very low mach number and I

found it also could run the reasonable answer. I confuse

about the compressible to incompressible.