[AlanSC]

Hello, I am modelling a magnetorheological fluid in CFX as a bingham fluid with a yield stress that depends on a magnetic field that I have already solved. I imported the magnetic field into the model, but I cannot make the yield stress dependent on it, because CFX doesn´t allow material properties to depend on variables.

I read that that can be done with a CFX Expression Language, but I don't have a closed expression for it, just the value of the magnetic field intensity on each coordinate of my model.

Is there something I can do in order to correctly model a field dependent yield stress??

Thank you in advance

Best regards,

Alan

[ckleanth]

I think that if using the standard non newtonian fluid model cfx wont let you link dynamic viscosity with a variable.

If however you specify your own equation for dynamic viscosity you will need to link shear strain rate with viscosity and theoreticaly it is possible to rewrite write the non newtonian equation models in cel. However it is also likely that the solver uses different numerics if a non newtonian model is chosen and therefore you need to raise a support call with ansys to verify this.

please let us know of the result

[AlanSC]

Thanks ckleanth, you were right, it is possible to specify a shear rate dependent dynamic viscosity, but you must be very careful with it, and probably the numerics are different for non-newtonan models, because the results that I obtained were not quite what I expected. I'll let you know when I have a response from support team.

[ghorrocks]

Yes, you have to be careful with non-Newtonian fluids. There are a whole bunch of higher-order terms which arise in the Navier Stokes equations when viscosity varies and I suspect that many of these terms are not included in CFX as the default non-Newtonian models provided in CFX don't need them. But that means if you modify the non-Newtonian models some of these terms may be important but CFX does not have them in by default.

Proceed with care - Do lots of validation and verification.

[belgacem]

Hi Friend
We want to simulate Non newtonian fluid by using ANSYS CFX and I do not know how I will add the Bingham model especilly where the yield stress depends on the pressure (Mohr-Coulomb criterio). The reason why we used the Mohr-Coulomb criterion is that we wanted to describe both frictional materials (like sand) and cohesive materials (like clay). As you know, there are two typical soil parameters, the internal friction angle and the cohesion, in the Mohr-Coulomb criterion. We can describe various flow behaviors of soils by adjusting the parameters. Also, the normal stress is included in the Mohr-Coulomb criterion, because the yield shear strength of the frictional materials depends on the normal stress. In our model, the pressure is used in place of the normal stress. More details can be found in: Estimating the impact force generated by granular flow on a rigid obstruction (Equation 1,2 and 3).

[ghorrocks]

CFX already supports the Bingham model. You do not need to add it. But I do not think CFX supports variable yield stress - try it, if it gives an error then it is not supported.

[gbrajtm]

Hi alan,
I am also working on MR damper, i have written CEL for the Bingham model. But i am facing in specifying it to annular gap. I need to specify the annular gap with Non newtonian fluid(Bingham) while it is under movement. Do you know how specify it this only for annular cells where magnetic effect is occurring.

thanks in advance

[ghorrocks]

Haven't we been discussing this in your other thread? http://www.cfd-online.com/Forums/cfx...nt-piston.html

[gbrajtm]

Yea, i just found this thread which is similar to our discussion, so i thought he might have some other hints.