Cp(theta) of 2D laminar circular cylinder flow

Michael_x · May 12, 2015, 15:57

Dear all,

I have a question concerning the Cp(theta) of a 2D cylinder flow at Re=150 and Ma=0.2 (see picture attached).
What I do not understand is: at theta=0 deg Cp should be 1 but in my simulation it is 1.04. Where can these 0.04 come from?
I am doing a compressible calculation (Thermal Energy) with Air Ideal Gas.

Thanks in advance,
Michael

ghorrocks · May 12, 2015, 19:23

4% error is pretty close in most cases. But if you want to be more accurate have a look at the FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

Michael_x · May 25, 2015, 10:52

Thanks for the answer.

I am using the fourth order backwards Euler scheme in time and the high resolution scheme in space so I am wondering if the error should not be smaller?
The same problem has been calculated with scientif CFD softwares, too, and people used to get errors smaller than 1% (time scheme: runge kutte 4th order accurate, space scheme: central differences 4th order accurate).

Do you guys expect such big errors using CFX (lower formal accuracy in advection scheme)?

Thanks in advance and greetings,
Michael

ghorrocks · May 25, 2015, 21:01

CFX is capable of very high accuracy - you are not limited by the software for how accurate you can go in most cases. The usual limitations of accuracy are the correct choice of physics options, development of a suitable mesh, time and convergence criteria and a computer big enough to run it. So the limitation is the ability of the user to do all these things properly, not the software.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
convergence issue flow around circular cylinder	SB123	STAR-CCM+	1	December 13, 2014 16:18
Flow over 2D Cylinder, Laminar and Turbulent	Tsr63	FLUENT	5	November 13, 2014 13:13
Particle deposition on circular cylinder in turbulent flow	Julian K.	CFX	1	October 3, 2011 18:51
asking about wake in Inviscid flow in cylinder circular	teguhtf	FLUENT	0	April 23, 2011 01:14
Modeling flow acros circular cylinder	Zhihua	Main CFD Forum	3	April 20, 2004 12:16

May 12, 2015, 19:23		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	4% error is pretty close in most cases. But if you want to be more accurate have a look at the FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

May 25, 2015, 10:52		#3
Michael_x New Member M S Join Date: Nov 2014 Posts: 4 Rep Power: 12	Thanks for the answer. I am using the fourth order backwards Euler scheme in time and the high resolution scheme in space so I am wondering if the error should not be smaller? The same problem has been calculated with scientif CFD softwares, too, and people used to get errors smaller than 1% (time scheme: runge kutte 4th order accurate, space scheme: central differences 4th order accurate). Do you guys expect such big errors using CFX (lower formal accuracy in advection scheme)? Thanks in advance and greetings, Michael

May 25, 2015, 21:01		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	CFX is capable of very high accuracy - you are not limited by the software for how accurate you can go in most cases. The usual limitations of accuracy are the correct choice of physics options, development of a suitable mesh, time and convergence criteria and a computer big enough to run it. So the limitation is the ability of the user to do all these things properly, not the software.