divergence of u calculation from the result

sourav0921 · July 29, 2019, 02:25

Hi,
I am using pisofoam solver to solve the incompressible turbulent channel flow problem. To saisfy the continuity I calculated the divergence of U using the command duv (phi). It comes in order of 10^-3 to 10^-6. Absolute tolerence is in the order of 10^-6. But when I calculated manually the divergence of U for a cell based on the result data of U it comes in order of 10^-2. Can anyone please tell me cause of this error?

gridley2 · August 1, 2019, 20:22

The reason is likely that fvc::div() does not have the units you may think. It just sums gradients at cell faces time face areas, IIRC. So the real divergence is obtained by dividing that quantity by cell volume. (or maybe multiply?)

July 29, 2019, 02:25	divergence of u calculation from the result	#1
sourav0921 New Member SOURAV GANGULI Join Date: Dec 2018 Posts: 11 Rep Power: 7	Hi, I am using pisofoam solver to solve the incompressible turbulent channel flow problem. To saisfy the continuity I calculated the divergence of U using the command duv (phi). It comes in order of 10^-3 to 10^-6. Absolute tolerence is in the order of 10^-6. But when I calculated manually the divergence of U for a cell based on the result data of U it comes in order of 10^-2. Can anyone please tell me cause of this error?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Divergence detected in AMG solver: k,epsilon,Increasing relaxation sweep?	BHOLU	FLUENT	2	September 25, 2020 02:24
Average study of RANS results	serrar	FLUENT	6	April 26, 2019 12:23
Divergence problem for species transport model	MY	FLUENT	3	January 11, 2014 04:46
what reason for AGARD445.6 flutter Calculation is divergence (MPCCI+FLUENT+ANSYS)	franckgong	Main CFD Forum	0	February 10, 2012 10:07
How to improve the calculation result for the f	H.P.LIU	Phoenics	1	December 8, 2000 11:28

August 1, 2019, 20:22		#2
gridley2 New Member Gavin Ridley Join Date: Jan 2019 Location: Tennessee, USA Posts: 25 Rep Power: 7	The reason is likely that fvc::div() does not have the units you may think. It just sums gradients at cell faces time face areas, IIRC. So the real divergence is obtained by dividing that quantity by cell volume. (or maybe multiply?)