CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   TVD schemes for acoustics (http://www.cfd-online.com/Forums/main/15737-tvd-schemes-acoustics.html)

Shuo September 16, 2008 20:32

TVD schemes for acoustics
 
Hi

Is it logical to use TVD schemes for compressible acoustic simulations even though it is inherently an artificial viscosity method? If not what other methods are there?

Shuo

versi September 17, 2008 04:25

Re: TVD schemes for acoustics
 
not very good because of too much diffusivity of TVD scheme in acoustics. Compact schemes, DRP schemes are better.

shuo September 17, 2008 04:32

Re: TVD schemes for acoustics
 
How about WENo/ENO type schemes?

Brian September 17, 2008 10:44

Re: TVD schemes for acoustics
 
what is Compact schemes and DRP schemes ? please explain !

Harish September 17, 2008 14:26

Re: TVD schemes for acoustics
 
WENO/ENO type schemes are fine. Also you can look at ADER schemes. If you do not expect to observe shock structures in your flow field, the compact and drp schemes are much faster compared to the ENO schemes. Also compact and drp schemes can also be applied for flows with shock structures by adding filtering ( see papers of visbal) or dissipation ( see papers of Tam). The other option is to use upwind compact ( paper of zhong) and upwind DRP schemes ( CAA benchmarks) or consider other classes of high order schemes ( high-order mccormack etc., search for the review paper by Kurbatskii) which include in-build dissipation. You should have a look at the CAA benchmark workshops for the variety of high-order schemes that you can employ for your computations. You need high-order BC with high-order schemes to avoid degrading your solution especially near walls.

Axel Rohde September 19, 2008 21:47

Re: TVD schemes for acoustics
 
A colleague of mine and I published a paper on this a few years back, where we compared my TVD acoustic solution to his aeroacoustic solution for M = 0.5 cylinder flow. Here the link to the PDF: www.microcfd.com/papers.htm


All times are GMT -4. The time now is 22:45.