[jmenendez]

Hi all,

I'm trying know how walls of impeller can influence 'turbo-variables' like pressure ratio, etc.

I have made the whole machine: diffuser, impeller. CFX and turbogrid. Not Fluent.

The boundary condition over walls is Heat transfer: temperature. This temperature is fixed for each simulation.

But I dont really know how I should estimate the heat. I thought with CFD-post function: intArea(Wall Heat Flux)@impeller walls but it gives me a little strange thermal power.

Does anyone have any ideas on the best way to estimate the thermal influence of the walls on the fluid?

Thank you very much!

[ghorrocks]

The first thing to do is you need to define exactly what you are looking for. What do you mean by "thermal influence"? The average temperature rise in the fluid passing through? The peak temperature in the fluid? The peak temperature in the rotor? Some other function of temperature, volume and/or time?

[jmenendez]

Originally Posted by ghorrocks

The first thing to do is you need to define exactly what you are looking for. What do you mean by "thermal influence"? The average temperature rise in the fluid passing through? The peak temperature in the fluid? The peak temperature in the rotor? Some other function of temperature, volume and/or time?

First of all, thank you very much for your help, I appreciate it.

In the first stage I studied the turbocharger in steady state, SST turbulence model and boundaries with adiabatic condition (shroud, hub, blades, inlet and volute, all wall surfaces). Approximately 4 million elements, residual RMS less than 10^-5,no imbalances, and isentropic efficiency, power, etc. were good and as expected.

Now I want to improve the model using non-adiabatic boundaries, so I changed BC from adiabatic to temperature, but the results are strange... For example, the isentropic efficiency has risen from from 70% to 95% (datasheet around 65%) for T = 20º (similar to ambient conditions).




In the post process (for instance, with intArea(Wall Heat Flux)@impeller walls) I have calculated the heat fluxes on the surfaces and they have unexpected trends.

So I don't know if I am defining the problem well. The mesh is very accurate: yplus around 1, good metrics, etc. In summary, I am trying to find the differences between two models: adiabatic and non-adiabatic when I change the wall temperature.

Thanks in advance, any suggestion or idea will be welcome, I am somewhat blocked.