Is it possible to use LES to simulate flow past trailing edge?

jasonyuan · October 14, 2013, 10:56

sbaffini · October 15, 2013, 03:26

I'm not an expert on noise or serrated trailing edges but:

- Maybe a compressible solver could be necessary for the noise source computation

- How does the characteristic length of the sawtooth trailing edge compares with the grid spacings that will be used in the simulations? Is the wall-modeling feasible at all in this case?

just my 2 cents

jasonyuan · October 15, 2013, 03:44

Thanks for your advice！ Actually, I have no experience on CFD..I'm an undergraduate.
The proposal was written after reading several papers. I just randomly choose one wall model. I don't know whether this wall model is feasible or not.
So, what's the advantage of compressible over incompressible

sbaffini · October 15, 2013, 04:19

The way pressure is computed in compressible or incompressible flow solvers is substantially different as well as its meaning. More specifically, for incompressible solvers the computed pressure has nothing in common with the acoustic pressure.

As a consequence, incompressible solvers can't be used for acoustics.

As long as wall models are concerned, it is not the model itself to be problematic. The problem is, as you are going to use a wall model at a somehow high Reynolds number, you will end up with somehow large grid cells. The question is, what is the ratio between this grid cell size and the typical length of the serrated edge? If the Reynolds number is sufficiently high you might end-up with non-trivial grid requirements.

I repeat, i am not an expert of the specific field, i'm just guessing. I don't even know what is the level of the proposal needed to get into the graduate school you are thinking of. In my country you can go Ph.D. only after the batchelor and the master; hence, at that point, you are supposed to know things differently. But, even in that case, i ended up with a much less detailed proposal at my time (considering, however, that usually it really has low relevance in my country, in the sense that you are not supposed to really work on that).

October 14, 2013, 10:56	Is it possible to use LES to simulate flow past trailing edge?	#1
jasonyuan New Member yuanhang Join Date: Oct 2013 Posts: 3 Rep Power: 13	Is it possible to use LES to simulate flow past trailing edge? Last edited by jasonyuan; November 3, 2013 at 00:41.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Periodic Pipe Flow LES	dvolkind	CFX	8	March 21, 2020 06:30
LES transitional pipe entrance flow	hasanduz	Main CFD Forum	2	October 11, 2013 18:59
[DesignModeler] Creating a blunt trailing edge from collapsed leading edge	gmd	ANSYS Meshing & Geometry	0	March 2, 2012 10:48
Flow past a sphere	Fabio	FLUENT	23	December 18, 2009 16:32
Supersonic flow past a wedge with counter flow	Mahesh Bailakanavar	FLUENT	0	February 14, 2008 01:21

October 15, 2013, 03:26		#2
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,195 Blog Entries: 29 Rep Power: 39	I'm not an expert on noise or serrated trailing edges but: - Maybe a compressible solver could be necessary for the noise source computation - How does the characteristic length of the sawtooth trailing edge compares with the grid spacings that will be used in the simulations? Is the wall-modeling feasible at all in this case? just my 2 cents

October 15, 2013, 03:44		#3
jasonyuan New Member yuanhang Join Date: Oct 2013 Posts: 3 Rep Power: 13	Thanks for your advice！ Actually, I have no experience on CFD..I'm an undergraduate. The proposal was written after reading several papers. I just randomly choose one wall model. I don't know whether this wall model is feasible or not. So, what's the advantage of compressible over incompressible

October 15, 2013, 04:19		#4
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,195 Blog Entries: 29 Rep Power: 39	The way pressure is computed in compressible or incompressible flow solvers is substantially different as well as its meaning. More specifically, for incompressible solvers the computed pressure has nothing in common with the acoustic pressure. As a consequence, incompressible solvers can't be used for acoustics. As long as wall models are concerned, it is not the model itself to be problematic. The problem is, as you are going to use a wall model at a somehow high Reynolds number, you will end up with somehow large grid cells. The question is, what is the ratio between this grid cell size and the typical length of the serrated edge? If the Reynolds number is sufficiently high you might end-up with non-trivial grid requirements. I repeat, i am not an expert of the specific field, i'm just guessing. I don't even know what is the level of the proposal needed to get into the graduate school you are thinking of. In my country you can go Ph.D. only after the batchelor and the master; hence, at that point, you are supposed to know things differently. But, even in that case, i ended up with a much less detailed proposal at my time (considering, however, that usually it really has low relevance in my country, in the sense that you are not supposed to really work on that).