|
[Sponsors] |
March 9, 2009, 12:44 |
Hi everybody,
maybe there i
|
#1 |
Member
Luca M.
Join Date: Mar 2009
Location: Luzern, Switzerland
Posts: 59
Rep Power: 17 |
Hi everybody,
maybe there is a severe bug in the adjustPhi.C. Shouldn't be the line 57 and 134 (check of processor type) corrected to: if (!isType<processorfvspatchscalarfield> (phip)) instead of if (!isType<processorfvpatchscalarfield>(phip))? (Sorry for the missing capital letters it should be processorFvsPatchScalarField ) With such implementation (the uncorrected one) the proc patch are not recognized: so adjustableMassOut and massIn are wrongly evaluated. Moreover the massCorr factor scales processors mass exchange while processor boundaries should not be used to correct mass global imbalance. The error is present also in the 1.4.1 version Regards Lu & Co |
|
March 9, 2009, 12:58 |
Hi everybody,
(There is a p
|
#2 |
Member
Luca M.
Join Date: Mar 2009
Location: Luzern, Switzerland
Posts: 59
Rep Power: 17 |
Hi everybody,
(There is a problem in the visualization, I tried in this way) maybe there is a severe bug in the adjustPhi.C. Shouldn't be the line 57 and 134 (check of processor type) corrected to: if (!isType< processorFvspatchscalarfield> (phip)) instead of if (!isType< processorFvpatchscalarfield>(phip))? (Sorry for the missing capital letters it should be processorFvsPatchScalarField ) With such implementation (the uncorrected one) the proc patch are not recognized: so adjustableMassOut and massIn are wrongly evaluated. Moreover the massCorr factor scales processors mass exchange while processor boundaries should not be used to correct mass global imbalance. The error is present also in the 1.4.1 version Regards Lu & Co |
|
March 9, 2009, 12:59 |
I believe you are right, thank
|
#3 |
Senior Member
Join Date: Mar 2009
Posts: 854
Rep Power: 22 |
I believe you are right, thanks for finding this error. Do you get the correct behavior after correcting the code?
H |
|
March 9, 2009, 13:04 |
Hi everybody,
Sorry for the
|
#4 |
Member
Luca M.
Join Date: Mar 2009
Location: Luzern, Switzerland
Posts: 59
Rep Power: 17 |
Hi everybody,
Sorry for the spam but there is a problem in email visualization of the post, you can see the bug post for the correct spell directly on the web site Regards Lu & Co |
|
March 9, 2009, 13:14 |
Hi Henry,
debugging the cod
|
#5 |
Member
Luca M.
Join Date: Mar 2009
Location: Luzern, Switzerland
Posts: 59
Rep Power: 17 |
Hi Henry,
debugging the code now the mass flux in parallel calculation are right and processor patch are correctly skipped in the boundary loop thus the mass correction apply only to non processor patches. Luca |
|
March 9, 2009, 15:15 |
Thanks for the info. I have p
|
#6 |
Senior Member
Join Date: Mar 2009
Posts: 854
Rep Power: 22 |
Thanks for the info. I have pushed the correction into our 1.5.x git repository.
H |
|
|
|