Cp values larger than 1

stergiani · November 14, 2017, 09:14

Hei,

I'm running simpleFoam for air flow around buildings and I want to calculate the Cp values on the facades. The weird thing is that I get values way greater than 1, around 1.2, which seems wrong for my incompressible flow.
The inlet pressure is set at 0, so I guess my reference pressure is equal to 0.
I checked air viscosity and velocity profiles, and everything looks normal.
Any ideas on what I have missed?
Thanks

Stergiani

matejfor · November 15, 2017, 03:58

Well in incompressible flow, your pressure is not in Pascals but in Pa/rho .I do not recall the c_p expression very well but can the density influence your results?

matej

stergiani · November 15, 2017, 05:08

Hei, Matej.
Thank you for your reply.
Yes, I have checked it and still get extreme values, around 1.5
I calculate the Cp by the following equation Cp=(p-p0)/(1/2*ρ*Uref^2).
Since my pressure in inlet and outlet is set to 0, I guess that p0=0.
Could it be that my domain is small, and although it looks like the solution is converged, I get wrong results?
Thanks

Stergiani

matejfor · November 15, 2017, 05:41

Hmmm....
I guess the pressure difference is multiplied by density in your incompressible place.
As the c_p looking at the coefficient takes only the static pressure difference from the simulation, if the numbers are wrong, it must the pressure. To get a good pressure distribution on the surface you need to make sure your mesh is correct. For your RANS simulation you need to have a good prismatic layer with a proper y+ values. And good convergence.

stergiani · November 15, 2017, 08:22

Thanks, Matej!
I improved my mesh and everything looks better. The convergence is good, but my yplus is around 200. And I'm a bit confused about yplus, cause some take it very small, but others consider even up to 300. Any insight?
Sorry, for the many questions, but it's my first time working on CFD

Stergiani

matejfor · November 15, 2017, 09:04

Now you have landed on a very special ground. BL modelling is still a question in CFD. OpenFOAM is using highRe wall functions where your y+ should be around 30-300. Unless you are resolving the BL with a number of cells

Please have a look here: https://www.openfoam.com/documentati...urbulence.html

and read any book on CFD, e.g. https://www.amazon.com/Introduction-...lalasekera+cfd
or similar.

stergiani · November 15, 2017, 09:47

I have already got the book you suggested, so I'd better start studying

Thank you for all the help!

Stergiani

November 14, 2017, 09:14	Cp values larger than 1	#1
stergiani New Member stergiani Join Date: Mar 2017 Posts: 9 Rep Power: 9	Hei, I'm running simpleFoam for air flow around buildings and I want to calculate the Cp values on the facades. The weird thing is that I get values way greater than 1, around 1.2, which seems wrong for my incompressible flow. The inlet pressure is set at 0, so I guess my reference pressure is equal to 0. I checked air viscosity and velocity profiles, and everything looks normal. Any ideas on what I have missed? Thanks Stergiani

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
using chemkin	JMDag2004	OpenFOAM Pre-Processing	2	March 8, 2016 23:38
cyclic BC - modify boundary face values	marluc	OpenFOAM Programming & Development	0	May 4, 2015 17:50
Velocity values are normal but pressure values are too big	rv82	OpenFOAM Running, Solving & CFD	4	April 13, 2015 04:59
Setting patch field values equal to internal field values	leroyv	OpenFOAM Programming & Development	1	October 21, 2014 16:49
strange node values @ solid/fluid interface - help	JB	FLUENT	2	November 1, 2008 13:04

November 15, 2017, 03:58		#2
matejfor Senior Member matej forman Join Date: Mar 2009 Location: Brno, Czech Republic Posts: 182 Rep Power: 17	Well in incompressible flow, your pressure is not in Pascals but in Pa/rho .I do not recall the c_p expression very well but can the density influence your results? matej

November 15, 2017, 05:08		#3
stergiani New Member stergiani Join Date: Mar 2017 Posts: 9 Rep Power: 9	Hei, Matej. Thank you for your reply. Yes, I have checked it and still get extreme values, around 1.5 I calculate the Cp by the following equation Cp=(p-p0)/(1/2ρUref^2). Since my pressure in inlet and outlet is set to 0, I guess that p0=0. Could it be that my domain is small, and although it looks like the solution is converged, I get wrong results? Thanks Stergiani

November 15, 2017, 05:41		#4
matejfor Senior Member matej forman Join Date: Mar 2009 Location: Brno, Czech Republic Posts: 182 Rep Power: 17	Hmmm.... I guess the pressure difference is multiplied by density in your incompressible place. As the c_p looking at the coefficient takes only the static pressure difference from the simulation, if the numbers are wrong, it must the pressure. To get a good pressure distribution on the surface you need to make sure your mesh is correct. For your RANS simulation you need to have a good prismatic layer with a proper y+ values. And good convergence.

November 15, 2017, 08:22		#5
stergiani New Member stergiani Join Date: Mar 2017 Posts: 9 Rep Power: 9	Thanks, Matej! I improved my mesh and everything looks better. The convergence is good, but my yplus is around 200. And I'm a bit confused about yplus, cause some take it very small, but others consider even up to 300. Any insight? Sorry, for the many questions, but it's my first time working on CFD Stergiani

November 15, 2017, 09:04		#6
matejfor Senior Member matej forman Join Date: Mar 2009 Location: Brno, Czech Republic Posts: 182 Rep Power: 17	Now you have landed on a very special ground. BL modelling is still a question in CFD. OpenFOAM is using highRe wall functions where your y+ should be around 30-300. Unless you are resolving the BL with a number of cells Please have a look here: https://www.openfoam.com/documentati...urbulence.html and read any book on CFD, e.g. https://www.amazon.com/Introduction-...lalasekera+cfd or similar.

November 15, 2017, 09:47		#7
stergiani New Member stergiani Join Date: Mar 2017 Posts: 9 Rep Power: 9	I have already got the book you suggested, so I'd better start studying Thank you for all the help! Stergiani