CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   why continuity doesn't solve when p be fixedGradient in groovyBC? (http://www.cfd-online.com/Forums/openfoam-solving/120345-why-continuity-doesnt-solve-when-p-fixedgradient-groovybc.html)

immortality July 4, 2013 16:09

why continuity doesn't solve when p be fixedGradient in groovyBC?
 
1 Attachment(s)
I have set p in outlet be:
Code:

(internalField(p)-50000)
in a period of time in start of run.
but it seems continuity doesn't solve for that as can be seen in the attached image.why?

ngj July 6, 2013 20:27

How come that you are not satisfied with 1e-18? I would say it was perfect from a continuity perspective.

Kind regards,

Niels

immortality July 7, 2013 07:40

hi Niels
in first time steps its a straight line without fluctuations.I think maybe it be an issue.

gschaider July 8, 2013 15:07

Quote:

Originally Posted by immortality (Post 438262)
hi Niels
in first time steps its a straight line without fluctuations.I think maybe it be an issue.

No. For continuity error of 1e-18 to be a problem the system has to be really tiny. If your system is really that tiny the question is whether the continuum assumption is still valid (if it isn't then of course the question is whether the standard OF-solvers are still the right thing for you)

immortality July 8, 2013 15:25

dear Bernhard.I don't want that residuals to be less than 1e-18 ar such thing.
I focus on time 0 till around time 2e-5 that the continuity is a stright horizontal line.I ask why in this period (in figure) there are not any fluctuations in residuals.is it normal?how and why?!

immortality July 8, 2013 16:20

I found out why its so!
in first time steps the residuals are of order 0f 1e-22 or 1e-21 and are very low in compare with later times(2e-5 to next)(I don't know the exact reason) then they appear as a line without fliuctuations,in next time steps continuity errors are a bit higher but are low enough as dear Benhard and dear Niels told to say continuity is satisfied.


All times are GMT -4. The time now is 05:03.