[jmenendez]

Hello

In the CFX solver output, in the H-Energy energy balance, terms like this appear:

 +--------------------------------------------------------------------+
 |                            H-Energy-S1                             |
 +--------------------------------------------------------------------+
 Boundary         : S1 Hub                                 -5.9921E-01
 Boundary         : S1 Outlet                              -2.5992E+02
 Bnd Src/Visc Work: S1 Outlet                              -1.4649E-04
 Boundary         : S1 Shroud                              -7.2303E-01
 Bnd Src/Visc Work: S1 to R1 Side 1                         6.6730E-02
 Domain Src (Neg) : S1                                     -3.5928E+01
 Domain Src (Pos) : S1                                      3.5928E+01
 Domain Interface : R1 to S1 (Side 2)                       2.6110E+02
                                                           -----------
 Domain Imbalance :                                        -7.3914E-02

How can I calculate that term: Bnd Src/Visc Work: R1 Shroud 8.3414E-01, in CFD-Post, what function and variable should I use? I've been reading the manual but I haven't found anything to solve my doubt.

Another important question.

The term viscous dissipation is by definition always positive, why in my oulet is it negative?

This simulation has no sources, simply a fluid domain in a compressor, SST turbulence model, and non-adiabatic conditions (fixed temperature walls). The fluid is real gas. It's not a heat conjugate problem

thank you very much for any help

[ghorrocks]

Viscous dissipation is turned on by default. In most cases it adds a negligible amount of heat to the fluid - your case is an example of this. It is negative as heat added is positive. The sign of heat addition or removal is arbitrary, you just need to be consistent.

I do not know of a way of visualising the viscous work component by itself. But as it is so small is it really worth worrying about?

If you don't like the viscous work term you can turn it off. Then re-run your simulation and you will have no viscous work and the world will be much simpler.

[jmenendez]

Originally Posted by ghorrocks

Viscous dissipation is turned on by default. In most cases it adds a negligible amount of heat to the fluid - your case is an example of this. It is negative as heat added is positive. The sign of heat addition or removal is arbitrary, you just need to be consistent.

I do not know of a way of visualising the viscous work component by itself. But as it is so small is it really worth worrying about?

If you don't like the viscous work term you can turn it off. Then re-run your simulation and you will have no viscous work and the world will be much simpler.

Dear ghorrocks,

That term isn't always that insignificant. There are many publications that consider it. In my case I am simulating small-scale turbomachines with ultra high speeds (300 krpm, 15 mm inlet), and this term is capital for the energy balance. They are machines where the smallest Reynolds number involves significant losses. The above imbalance is an example of the inlet, but in the impeller I have other values.

Thank you very much for your advice, regards.

[Opaque]

Careful with the vocabulary here.

Viscous work and viscous dissipation are not the same thing. They are "siblings" , but careful they should not be confused.

Viscous work = div (tau : v)

Viscous dissipation = viscosity * || Sij ||^2

If you expand the viscous work term, you will obtain two terms: viscous dissipation (irreversible component > 0) and a second term (reversible component). The reversible component cancels out within the domain, see Domain Src always having the exact value with opposite signs. At the boundary, their value comes in, and it is NOT negligible at moving walls.

See work by Lyman,
https://www.semanticscholar.org/pape...3cd326cf81b59f

Hope the above helps

[jmenendez]

Originally Posted by Opaque

Careful with the vocabulary here.

Viscous work and viscous dissipation are not the same thing. They are "siblings" , but careful they should not be confused.

Viscous work = div (tau : v)

Viscous dissipation = viscosity * || Sij ||^2

If you expand the viscous work term, you will obtain two terms: viscous dissipation (irreversible component > 0) and a second term (reversible component). The reversible component cancels out within the domain, see Domain Src always having the exact value with opposite signs. At the boundary, their value comes in, and it is NOT negligible at moving walls.

See work by Lyman,
https://www.semanticscholar.org/pape...3cd326cf81b59f

Hope the above helps

Sorry, English is not my native language and the concepts have no equal translation.

I need to count any amount of heat in the micro-turbomachine impeller. These are not large quantities and, as you may have seen in other threads I have created, I am having a lot of trouble obtaining them. These machines have a non-adibaticity problem, and it becomes necessary to measure this term.

They are not like large turbomachines which are generally considered adiabatic (except when cooling blades, etc.)

Thank you very much again