Non interger scalar field power - Chemistry law

Nathanael3 · June 26, 2014, 14:17

Hi Foamers,

I would like to code a chemistry problem. A field named C is consumed over time following random chemistry law:

$\frac{dC}{dt} = k C^{0.41}$

The problem is that the exponent is not an integer.
In the case where the chemistry is:

$\frac{dC}{dt} = k' C$

I use the following code:

Quote:

solve
(
fvm::ddt(C)
==
fvm::Sp(k, C)
);

It is really convenient because I solve it implicitly.

Would it be possible to do the same with the "0.41 exponent" ?

Thanks,
Nathanael.

nimasam · June 29, 2014, 03:39

nope its not possible, you should make it linear

forexample some thing like this:

Code:

solve
                        (
                                    fvm::ddt(C)    
                    ==        
                                    fvm::Sp(k/pow(C,0.51), C)
                        );

Nathanael3 · June 29, 2014, 10:56

Hi nimasam,

Thank you for your answer. I was affraid that it would not be possible.

The problem with the suggested code is that it crashes if C = 0 somewhere in the scalarfield.

Thanks for your help,
Nathanaël.

nimasam · June 29, 2014, 13:12

you can add a very small number forexample:

Quote:

1/(pow(c,0.51)+1e-08)

June 29, 2014, 03:39		#2
nimasam Senior Member Nima Samkhaniani Join Date: Sep 2009 Location: Tehran, Iran Posts: 1,266 Blog Entries: 1 Rep Power: 24	nope its not possible, you should make it linear forexample some thing like this: Code: solve ( fvm::ddt(C) == fvm::Sp(k/pow(C,0.51), C) ); __________________ My Personal Website (http://nimasamkhaniani.ir/) Telegram channel (https://t.me/cfd_foam)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 06:20
One-Seventh power law in turbulent pipe flows	selvam2487	CFX	3	September 9, 2013 17:46
UDF power law	dimsum	Fluent UDF and Scheme Programming	0	June 4, 2013 02:26
Star CCM+ issue (VOF and Non newtonian power law)	Sam	Siemens	3	January 23, 2009 07:39
from a vecor field to get a gradient of scalar	dusky.he	Main CFD Forum	4	March 30, 2007 09:08

June 29, 2014, 10:56		#3
Nathanael3 New Member Join Date: Apr 2013 Posts: 20 Rep Power: 13	Hi nimasam, Thank you for your answer. I was affraid that it would not be possible. The problem with the suggested code is that it crashes if C = 0 somewhere in the scalarfield. Thanks for your help, Nathanaël.