
[Sponsors] 
June 7, 2013, 10:29 
coronary artery with stenosisturbulence model?

#1 
Senior Member
Join Date: Feb 2011
Posts: 128
Rep Power: 6 
Dear all,
I want to use a turbulence model at my simulation with a modell of a coronary artery, which is including bifurcations and stenosis. I was looking for some publications and found a lot, but I am still not sure which is the best model to choose (komega SST, Transition SST, SAS or LES?) for this situation and how to proceed correct. Did I understand it right, that I have to perform a simulation and check y+ afterwards? I have a pulsatile flow and different flow conditions, so I guess y+ will be different for the single conditions and the single phases of the flow. Which one do I have to consider? And is it a probelm, if y+<1? Has somebody experience with this problem and could give me some hints or some literature advice? This would be really nice! Thank you in advance! Lilly 

June 7, 2013, 11:13 

#2 
Super Moderator

Y+ means the height of the first cell parallel to the wall. (the input for the spacing 1 of an edge)
Y+ depends on density, speed and dynamic viscosity. Once you have these parameter, use the "y+ estimator" tool available in cfdonline. for accurate results, one has to to try a low y+ (from 1 to 30). i am not sure i answered your question. 

June 7, 2013, 11:15 

#3 
Super Moderator

may be this can help Is my mesh correct


June 11, 2013, 07:22 
thank you! but still problems with y+

#4 
Senior Member
Join Date: Feb 2011
Posts: 128
Rep Power: 6 
Hi Ali, thank you for your help! It became at least clearer to me and I didn't know about the y+ calculator before!
My problem at the moment is: I have a pulsatile velocity pattern and simulations with different velocity patterns. Therefore my y+ will be different for each phase of the pattern and each pattern. Which y+ value do I have to consider? And is it still correct, that y+ should not be smaller than 0.1 (since I got really really small values for my mesh as you can see at my pic)? I don't know whether I monitored y+ properly (see pic) since the y+ values get larger from the wall to the inner of my 3d geometry (I cut a plane (plane_xz) through my geomtry to monitor the y+ values there)? Furthermore I read that there should be at least 10 layers inside the viscous and buffer layer for "Enhanced Wall Treatment", but how do I know where these layers end? And is this enhanced wall treatment automatically activated for SSTkomega or Transitional SST? It would be really nice if somebody could answer my questions! Thank you in advance! Lilly 

June 11, 2013, 07:35 

#5 
Senior Member
Lefteris
Join Date: Oct 2011
Location: UK, Greece
Posts: 187
Rep Power: 5 
You can create a fine mesh and not use wall functions at all. Then you won't have to worry about y+.
__________________
Lefteris 

June 11, 2013, 17:36 

#6  
Super Moderator
Sijal Ahmed Memon (turboenginner@gmail.com)
Join Date: Mar 2009
Location: Islamabad Pakistan
Posts: 3,909
Blog Entries: 6
Rep Power: 38 
Quote:
Quote:
Quote:
For transition model enhanced wall treatment is meaningless As described above, that I have used 100 layers for transition model, you can see the results here some results with Klkww transition model 

June 12, 2013, 16:57 
boundary layer thickness(local?) Reynolds number

#7 
Senior Member
Join Date: Feb 2011
Posts: 128
Rep Power: 6 
Thanks a million Aeronautics El.K. and thanks a million Far!
You answers were really helpful! I was looking for this boundary layer thickness equation today and found an equation at which one need to know the Boundary Layer Length as well. One also need to know this parameter at the y+ calculator (I have ssen there was also a conversation about this at which Ali was involved some time ago). Is there an equation for this boundary layer length as well? Or how can I estimate it? I was looking for it several hours but havenīt been successful. Another thing is the (local?) Reynolds number I need for this equation: Is this the one in the cell at the maximum thickness of the boundary layer? (I have seen one can plot the cell Reynolds number at Fluent) or is it the one directly at the wall? And did I understand it correctly, that the boundary layers you are using are also adapted to the location of the geometry, that means the boundary layers are not uniform? It would be really nice if somebody could give me a hint! Thank you! Lilly 

June 12, 2013, 17:23 

#8 
Senior Member
Lefteris
Join Date: Oct 2011
Location: UK, Greece
Posts: 187
Rep Power: 5 
Lilly, pause for a minute and take a breath.
It seems that you're having many difficulties with what you're doing and you seem confused with turbulence modeling. Firstly, forget about y+. Just create a very fine mesh (for example use dy=0.00001 for the first cell) and a growth factor of 1.11.2. Secondly, do not use wall functions (no standard wall functions, no enhanced wall treatment, nothing at all). Thirdly, by "boundary layer length" you mean that you want to estimate the point of the transition from laminar to turbulent flow or you just confuse it with the turbulence length scale? Oh, something just occurred to me. Is the flow in the coronary artery turbulent, really? I mean, I know nothing about biological flows so I'm asking out of curiosity.
__________________
Lefteris 

June 13, 2013, 00:38 

#9 
Super Moderator
Sijal Ahmed Memon (turboenginner@gmail.com)
Join Date: Mar 2009
Location: Islamabad Pakistan
Posts: 3,909
Blog Entries: 6
Rep Power: 38 
Lilly :
Boundary layer thickness (length) for laminar and turbulent flows is given by two different formulae as shown http://en.wikipedia.org/wiki/Boundarylayer_thickness While the first layer (or first node or first cell) distance from the wall is given by the http://geolab.larc.nasa.gov/APPS/YPlus/ Here length is the characterstics length of your body e.g. for airfoil it is chord length, for cylinder it is dia of cylinder. Any how it is not the boundary layer thickness or length, it is the just the length of geometry under consideration. 

June 18, 2013, 05:12 

#10 
Senior Member
Join Date: Feb 2011
Posts: 128
Rep Power: 6 
Thanks a million to both of you again, Far and Aeronautics El.K.! Your answers helped me a lot!
Concerning the turbulent flow inside an artery: we have a stenosis placed inside the artery and behind this stenosis trubulences can arise (but we don't expect a turbulent flow inside a straight healthy artery at the size of artery we are considering). At the equation describing the boundary layer thickness (which I need at least to estimate the thickness of the boundary layer of my mesh). There also appears ": the distance downstream from the start of the boundary layer which should be kind of the length until the boundray layer is fully developed" (http://en.wikipedia.org/wiki/Boundarylayer_thickness). This is kind of confusing to me since I don't know this length! Or is this the diameter of the pipe again for the case of an pipe flow? Thank you for any hint! Lilly Last edited by Lilly; June 18, 2013 at 06:38. 

Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
An error has occurred in cfx5solve:  volo87  CFX  5  June 14, 2013 17:44 
Wrong calculation of nut in the kOmegaSST turbulence model  FelixL  OpenFOAM Bugs  27  March 27, 2012 09:02 
Low Reynolds kepsilon model  YJZ  ANSYS  1  August 20, 2010 13:57 
KOmega Turbulence model from wwwopenFOAMWikinet  philippose  OpenFOAM Running, Solving & CFD  30  August 4, 2010 10:26 
Fan heater model: what turbulence source to use?  andy20  CFX  7  March 3, 2008 17:42 