Use periodic boundary conditions with DES - Urgent!

HeKhalil · November 22, 2012, 05:40

Hye guys,

This might be an elementary question but i still not sure about its answer.

I am simulating a flame-vortex interaction in a premixed swirl stabilized combustor. Basically, i try to evaluate the effect of vortex shedding on the flame using fluent.
I would like to use DES in my simulation.
Due to the enormous computational cost i only considered quarter section of the combustor while applying rotational periodic boundary conditions in the span-wise direction.
My question is, am i ok with applying the periodic boundary conditions while using DES or i shall consider the complete combustor knowing that the geometry is symmetrical and the flow enters the domain perfectly premixed.

your help is highly appreciated!

julien.decharentenay · November 22, 2012, 19:05

Hi Hesham,

The use of periodicity will limit/constraint the spanwise vortices frequency that can be captured by the simulation. Although it might be acceptable in some cases, I would not recommend it.

From a practical perspective, my suggestion would be to compare a simulation with a complete combustor and a simulation with rotational periodic boundary conditions (it could be half, quarter, or third) and assess the impact of the periodicity assumptions.

Good luck. Julien

FMDenaro · November 22, 2012, 19:10

Simulation of turbulence in portion of a domain with cyclic conditions must be limited to RANS formulation

Far · November 22, 2012, 20:46

see this http://cfd.mace.manchester.ac.uk/des..._MStrelets.pdf

HeKhalil · November 23, 2012, 12:31

Thanks guys, but i have a question for you julien, the problem of mine is to capture the longitudinal instability frequency due to flame vortex interaction as a result of vortex shedding to the flame. this frequency is equal to the acoustic quarter wave mode of the combustor. so i am interested only in the longitudinal frequency of vortex shedding (stream wise direction), can't this allow me to use periodicity? or the span wise frequency might affect that?

Also, i cannot do the simulation twice! one for the complete one and one for the quarter section and compare because of the enormous time consumed in each. i use 3 million cell resolution in only quarter section which takes along 5 days to complete one integral time (time of flow to enter and exit the domain once).

By the way, the combustor diameter is 110 mm. i use 220 cell in the cross-stream direction and 60 cells in span wise direction. this grid is totally uniform and structural grid with spatial step of 0.5 mm in both stream wise and cross stream directions. the grid as i said is more coarse in span wise direction to reduce computational cost. this gives me y+ of 50 near the wall and the effective DES length scale (by fluent) is positive every where except near the wall. am i ok with this grid? i know that the grid shall be refined near the shear layer down stream the dump plane of the combustor but i couldn't do this due to meshing difficulties so is this grid is ok?

November 22, 2012, 05:40	Use periodic boundary conditions with DES - Urgent!	#1
HeKhalil New Member Hesham Khalil Join Date: Nov 2012 Posts: 9 Rep Power: 14	Hye guys, This might be an elementary question but i still not sure about its answer. I am simulating a flame-vortex interaction in a premixed swirl stabilized combustor. Basically, i try to evaluate the effect of vortex shedding on the flame using fluent. I would like to use DES in my simulation. Due to the enormous computational cost i only considered quarter section of the combustor while applying rotational periodic boundary conditions in the span-wise direction. My question is, am i ok with applying the periodic boundary conditions while using DES or i shall consider the complete combustor knowing that the geometry is symmetrical and the flow enters the domain perfectly premixed. your help is highly appreciated!

November 22, 2012, 19:05		#2
julien.decharentenay Senior Member Julien de Charentenay Join Date: Jun 2009 Location: Australia Posts: 231 Rep Power: 18	Hi Hesham, The use of periodicity will limit/constraint the spanwise vortices frequency that can be captured by the simulation. Although it might be acceptable in some cases, I would not recommend it. From a practical perspective, my suggestion would be to compare a simulation with a complete combustor and a simulation with rotational periodic boundary conditions (it could be half, quarter, or third) and assess the impact of the periodicity assumptions. Good luck. Julien __________________ --- Julien de Charentenay

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
mesh file for flow over a circular cylinder	Ardalan	Main CFD Forum	7	December 15, 2020 14:06
Domain Imbalance	HMR	CFX	5	October 10, 2016 06:57
periodic boundary conditions fro pressure	Salem	Main CFD Forum	21	April 10, 2013 01:44
Periodic boundary conditions in 3D Eulerian granular flow simulations	dsm	FLUENT	4	March 2, 2012 20:04
Problem with using periodic boundary conditions	Sun	FLUENT	0	January 14, 2011 10:47

November 22, 2012, 19:10		#3
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,882 Rep Power: 73	Simulation of turbulence in portion of a domain with cyclic conditions must be limited to RANS formulation

November 22, 2012, 20:46		#4
Far Senior Member Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,558 Blog Entries: 6 Rep Power: 54	see this http://cfd.mace.manchester.ac.uk/des..._MStrelets.pdf

November 23, 2012, 12:31		#5
HeKhalil New Member Hesham Khalil Join Date: Nov 2012 Posts: 9 Rep Power: 14	Thanks guys, but i have a question for you julien, the problem of mine is to capture the longitudinal instability frequency due to flame vortex interaction as a result of vortex shedding to the flame. this frequency is equal to the acoustic quarter wave mode of the combustor. so i am interested only in the longitudinal frequency of vortex shedding (stream wise direction), can't this allow me to use periodicity? or the span wise frequency might affect that? Also, i cannot do the simulation twice! one for the complete one and one for the quarter section and compare because of the enormous time consumed in each. i use 3 million cell resolution in only quarter section which takes along 5 days to complete one integral time (time of flow to enter and exit the domain once). By the way, the combustor diameter is 110 mm. i use 220 cell in the cross-stream direction and 60 cells in span wise direction. this grid is totally uniform and structural grid with spatial step of 0.5 mm in both stream wise and cross stream directions. the grid as i said is more coarse in span wise direction to reduce computational cost. this gives me y+ of 50 near the wall and the effective DES length scale (by fluent) is positive every where except near the wall. am i ok with this grid? i know that the grid shall be refined near the shear layer down stream the dump plane of the combustor but i couldn't do this due to meshing difficulties so is this grid is ok?