coefficient of power-Law viscosity Model in OpenFOAM

Daniel_Khazaei · February 23, 2014, 13:16

Hello foamers

I have searched a lot but I was not able to find any proper explanation about coefficient of power-Law viscosity Model.

I have used online curve fitting tools to determine constant coefficients in power-Law viscosity model based on my experimental data. These values are obtained for k and n:

k = 0.0102 [Pa.s^(n)]

n = 0.46

Now when it comes to OpenFOAM, it uses kinematic viscosity and after reading source code of power-Law viscosity model, I have found out that k has the same dimension as nu.

I have divided my k value by rho and that gives me:

k = 7.293e-6

However, when I use this value for k the solution diverges quickly. I have seen a few tutorials in which k has a very large magnitude like 2500.

Am I missing something here? maybe I did not understand the model properly.

best wishes

regards

90nash · March 23, 2014, 17:30

Hello Daniel,
Were you able to find a good explanantion for the power law model. I am also a little confused as to why k value is taken so large in some cases. If i take k value small, the solution is not converging. Please help!!

fs.chemech · July 7, 2014, 01:46

Hi friends,

Have you found a solution for this problem that you can share with me?

I have the same problem too...

Cheers,

FS

AlexaB · September 21, 2014, 04:06

Dear friends,

I have the same problem as you.

Can someone post an explanation for this problem, plz?

Thank you very much!!

wurst · January 28, 2015, 09:05

Same problem here. When I divide by density I get very high velocities...

90nash · January 28, 2015, 15:29

Hi,
I had this problem some time back. Someone suggested to initiate the flow with constant viscosity. The k value is supposed to be large (if i remember correctly). So yeah, i ran a few time steps with constant viscosity and then switched to power law model and that allowed the simulation to run smoothly. Try that!

Ananthakrishnan · April 5, 2016, 05:23

Hi

was anyone able to find the explanation of the units for k in openfoam?
From this thread
http://www.cfd-online.com/Forums/ope...del-blood.html

i understand that openfoam multiples strain rate by one sec, so k does not have n in its power but how to eliminate sec from SI units. I get the values of k experimentally in pa.s^n. I dont know how to incorporate this into openfoam

Thanks

February 23, 2014, 13:16	coefficient of power-Law viscosity Model in OpenFOAM	#1
Daniel_Khazaei Senior Member Daniel Join Date: Mar 2013 Location: Noshahr, Iran Posts: 348 Rep Power: 21	Hello foamers I have searched a lot but I was not able to find any proper explanation about coefficient of power-Law viscosity Model. I have used online curve fitting tools to determine constant coefficients in power-Law viscosity model based on my experimental data. These values are obtained for k and n: k = 0.0102 [Pa.s^(n)] n = 0.46 Now when it comes to OpenFOAM, it uses kinematic viscosity and after reading source code of power-Law viscosity model, I have found out that k has the same dimension as nu. I have divided my k value by rho and that gives me: k = 7.293e-6 However, when I use this value for k the solution diverges quickly. I have seen a few tutorials in which k has a very large magnitude like 2500. Am I missing something here? maybe I did not understand the model properly. best wishes regards

January 28, 2015, 15:29		#6
90nash Member Nadish Saini Join Date: Feb 2014 Location: Raleigh, North Carolina Posts: 39 Rep Power: 12	Hi, I had this problem some time back. Someone suggested to initiate the flow with constant viscosity. The k value is supposed to be large (if i remember correctly). So yeah, i ran a few time steps with constant viscosity and then switched to power law model and that allowed the simulation to run smoothly. Try that! wurst likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Non-Newtonian Power Law for Viscosity	John	FLUENT	16	September 12, 2015 07:18
Water subcooled boiling	Attesz	CFX	7	January 5, 2013 04:32
modelling solids viscosity in eulerian multiphase model	derkaiser	FLUENT	1	December 5, 2011 04:42
about compresive phase	James	CFX	10	September 12, 2006 04:16
Two-Phase Buoyant Flow Issue	Miguel Baritto	CFX	4	August 31, 2006 13:02

March 23, 2014, 17:30		#2
90nash Member Nadish Saini Join Date: Feb 2014 Location: Raleigh, North Carolina Posts: 39 Rep Power: 12	Hello Daniel, Were you able to find a good explanantion for the power law model. I am also a little confused as to why k value is taken so large in some cases. If i take k value small, the solution is not converging. Please help!!

July 7, 2014, 01:46		#3
fs.chemech New Member Filippo Join Date: May 2014 Posts: 27 Rep Power: 12	Hi friends, Have you found a solution for this problem that you can share with me? I have the same problem too... Cheers, FS

September 21, 2014, 04:06		#4
AlexaB New Member Alexa Join Date: Sep 2014 Posts: 5 Rep Power: 12	Dear friends, I have the same problem as you. Can someone post an explanation for this problem, plz? Thank you very much!!

January 28, 2015, 09:05		#5
wurst New Member Jonas Join Date: Sep 2014 Posts: 6 Rep Power: 12	Same problem here. When I divide by density I get very high velocities...

April 5, 2016, 05:23		#7
Ananthakrishnan Senior Member Ananthakrishnan.A.S Join Date: Feb 2012 Location: Mumbai (Bombay), India Posts: 157 Rep Power: 16	Hi was anyone able to find the explanation of the units for k in openfoam? From this thread http://www.cfd-online.com/Forums/ope...del-blood.html i understand that openfoam multiples strain rate by one sec, so k does not have n in its power but how to eliminate sec from SI units. I get the values of k experimentally in pa.s^n. I dont know how to incorporate this into openfoam Thanks