Learning Forum

[montenegro]

Hello.

I'm modelling a 2D channel with a non-newtonian fluid flow and I need to describe its viscosity through an equation that depends on the shear rate ( η=f(gamma_dot) ). I am doing this through an UDF. How can I access the shear rate? Should I use the C_STRAIN_RATE(c, t) macro or the C_DUDY(c, t) (in this case, the absolute value of it) one?

Thank you.

Best regards

[Rob]

If you search for C_STRAIN_RATE on here you'll find one or two helpful answers.

[montenegro]

Hello.

Thank you for the reply.
I asked the question because I've seen some discussions where it is told that the C_STRAIN_RATE(c,t) macro is the shear rate and other where it is said that it is indeed the strain rate. I also read in the Fluent User's Guide that the strain rate is the shear rate.
However, I ran my calculations (with C_STRAIN_RATE(c,t) macro in my viscosity equation) and checked if the results made sense. The value of the shear rate should be zero all over a centerline in my channel and it is not when I pull the results for the "Strain Rate" in the "Derivatives" top down menu. The "dX-velocity/dy" is, though, hence my confusion.
Moreover, I used the C_DUDY(c,t) macro in my UDF (needed to use abs(C_DUDY(c,t)), because Fluent returns it with negative values in the lower side of my channel) and I could not achieve convergence.

Can you tell if there is something wrong with my procedure?

Thank you Best regards.

[Karthik Remella]

Hello Just to add to 's answer - you can use Fluent expressions directly without using UDF. You should find the parameter you are looking for under 'Derivatives'.
Karthik

[montenegro]

Hi.

Yes, I know I can set the viscosity equation via the 'custom field functions' option, but the problem persists. I am still doubtful regarding which parameter should I use, the "Strain rate" (C_STRAIN_RATE_MAG(c, t)) or the "dX-velocity/dy" (C_DUDY(c,t)) and if I use the latter one, my solution doesn't converge.

Thank you.
Best regards.

[Rob]

Can you show the plot of shear rate that you're looking at?
As an aside, I use C_STRAIN_RATE_MAG(c, t) in viscosity UDFs, and I don't think I've seen anyone using (or discussing) C_DUDY(c,t) the same purpose.

[montenegro]

Hello.

I'll upload the pictures. So, if I plot the contour for the strain rate it it is noticeable that in the center of the channel the value is very low, circa 0.26 1/s, but not zero.
This can be confirmed by plotting the same contour over a horizontal (xx) line I drew exactly in the middle of the channel.

I then plotted the contour for the dx-velocity/dy over this same centerline and you can see it is exactly zero, as the shear rate is supposed to be.
And I will send you the contour of the dx-velocity/dy over the channel as a confirmation that it is being calculated.

Thank you.
Best regards.

[Rob]

Turn off node values & replot. You may also want to have a look at xy plots rather than contours.