HI I have Non-Newtonian fluids property (viscosity as function of shear rate) but have no idea how to input it into Flow-3D material property. So can anyone advice on this?

If the strain rate is within the upper range during the whole simulation, meaning the stain rate is very high, the carreau equation presented in FLOW3D manual may be simplified as power law if mu_inf happens to be zero. Basically, if you look at the carreau equation, mu_inf = 0; E_T = 1; #given the temperature is constant lambda_00 = 0; lambda_0 = 0; lambda_2 = 0;

From there, you take "Ln" on both sides and it will turn into a linear equation, then you should be able to find the exponential coefficient (1-n)/2 and mu_0.

If the strain rate in the simulation covers not just the upper range, you may send the data to support@flow3d.com and see what else could be done.

Hi HC for my case the shear rate is cover entire range, so it is not a linear graph. Noticed that Flow-3d didn't user define function to enable this. anyway will check with support team on this.

Ben,

FLOW-3D has a fairly general hence complex function for the viscosity vs. strain rate (it's in the Theory Manual). As HC was saying, you'd first need to pick up which terms in it to use, then define the coefficients for these terms. Some coefficients can be determined from the asymptotic behavior at zero and infinite strain rates, and the rest from the fitting. Obviously, the more terms you use the harder it will be to do the fitting.

One can also introduce a user-function unto the provided FORTRAN function, but this will require programming and compiling.

In this year's release the user will be able to define the non-Newtonian viscosity tabular input.

Michael

Hi Michael It good know about the improvement made on non-Newtonian input, do u have any idea when the new release will be available?

We are shooting for the end of summer.

[wendywu]

Originally Posted by michael barkhudarov
;93241

We are shooting for the end of summer.

HI,
I am wondering if viscoplastic model could be input when you release the tabular input? Thank you.

Wendy

[MuxaB]

I don't know. What does the viscosity depend on in this case?

[wendywu]

I will use the constitutive model according to Bas Jan Emile van Rens's thesis: finte element simulation of the aluminum extrusion process shape prediction for complex profiles. the constitutive model is as attatchment.
Thank you.

Wendy