Can't accurately simulate non-Newtonian fluid?

Aran55336 · October 5, 2023, 17:02

Hi, I am trying to simulate a stirred tank with an anchor impeller using DNS in Lethe (although I'm not sure the program will be relevant to the solution). I have been able to simulate a different non Newtonian fluid (cmc) with power law coefficients K=0.339 n= 0.87134. However, the other fluid I am trying to simulate is really really viscous, K=5 n=0.05. The program also gives me an option to enter a shear rate minimum, which I have left at the default 1e-3 for the program. I have been managed to get around 20 seconds of simulation time with the first fluid, with only 33000 cells, with the impeller rotating at 1rps, timestep 0.005. With the second fluid, I have been unable to get a stable simulation with 1 million cells timestep 0.001. I have only managed to have it be somewhat stable with over 15 million cells and 1/100th of the timestep of the first one. Is there anything else I can do? I have a suspicion it might be to do with the viscosity at low shear rates but not completely sure.

arjun · October 7, 2023, 03:54

the viscosity on lower shear rates is definitely is a problem. Is there any way for you to limit it?

October 5, 2023, 17:02	Can't accurately simulate non-Newtonian fluid?	#1
Aran55336 New Member Aran Join Date: Jul 2023 Posts: 5 Rep Power: 3	Hi, I am trying to simulate a stirred tank with an anchor impeller using DNS in Lethe (although I'm not sure the program will be relevant to the solution). I have been able to simulate a different non Newtonian fluid (cmc) with power law coefficients K=0.339 n= 0.87134. However, the other fluid I am trying to simulate is really really viscous, K=5 n=0.05. The program also gives me an option to enter a shear rate minimum, which I have left at the default 1e-3 for the program. I have been managed to get around 20 seconds of simulation time with the first fluid, with only 33000 cells, with the impeller rotating at 1rps, timestep 0.005. With the second fluid, I have been unable to get a stable simulation with 1 million cells timestep 0.001. I have only managed to have it be somewhat stable with over 15 million cells and 1/100th of the timestep of the first one. Is there anything else I can do? I have a suspicion it might be to do with the viscosity at low shear rates but not completely sure.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Artificial wall error in fluent	aiden_1995	Main CFD Forum	0	January 9, 2023 15:06
How to simulate spherical solid particle immersed in fluid	CFDapplication	OpenFOAM Running, Solving & CFD	0	May 10, 2016 17:55
How to simulate a fluid flow with solid particles mixed with them??	pankajbhatter9	CFX	7	November 4, 2013 19:19
How to choose the mean diameter value for dispersed fluid?	creddy_trddc	CFX	1	October 30, 2011 05:30
Non newtonian fluid	Jay	FLUENT	0	January 4, 2005 22:56

October 7, 2023, 03:54		#2
arjun Senior Member Arjun Join Date: Mar 2009 Location: Nurenberg, Germany Posts: 1,286 Rep Power: 34	the viscosity on lower shear rates is definitely is a problem. Is there any way for you to limit it?