[neilduffy1024]

Hi all,

I am writing UDFs for a drying model. I have UDFs for heat transfer between the two phases (implemented as energy sources) and a temperature dependant evaporation model (implemented as a mass source). The latent heat of evaporation is accounted for in one of the energy source UDFs. All works fine up until evaporation begins to occur, when the temperature fluctuations are worse than expected.

I think this is because I don't have an energy source to bring the evaporated gas up to the appropriate temperature - from user manual, "mass sources enter with no thermal heat"

For example, for a droplet of liquid water which is evaporating. Aside from heat transfer, I think that I should have an energy sink (from gas phase) to account for the latent heat of evaporation, and an energy source (to gas phase) to bring the water vapour (mass source to gas phase) from 0 K (i.e. no thermal heat) to say 373 K. This would be the enthalpy of water vapour at 373 K (and say 1 atm) multiplied by the mass source??

If someone could please confirm, it would be much appreciated.

Many thanks,

Neil

[KristianEtienne]

Hi,

Just a thought; I have had similar issues when treating mass transfer (with the VOF model, is this your approach also?) and found that these oscillations where due to the strongly localized source terms occuring by the sharp gradient in phase fraction (to identify the interface).

If you belive that your problems are of the same nature, I would recommend you to read: "Evaporation model for interfacial flows based on a continuum-field representation of the source terms", Hardt and Wondra, JCP277, 5871-5895 (2008). The authors here present a method for smearing out the source terms, which greatly improves stability.

Good luck!

-KE

[neilduffy1024]

Hi Kristian,

Thanks for your reply. I am actually using a porous media model (the droplet just seemed like a clearer description). But I do believe that one of my problems is with source terms causing the simulation to lose stability, so your recommendation should be very helpful.

Thanks,

Neil

[wond]

Originally Posted by KristianEtienne

Hi,

Just a thought; I have had similar issues when treating mass transfer (with the VOF model, is this your approach also?) and found that these oscillations where due to the strongly localized source terms occuring by the sharp gradient in phase fraction (to identify the interface).

If you belive that your problems are of the same nature, I would recommend you to read: "Evaporation model for interfacial flows based on a continuum-field representation of the source terms", Hardt and Wondra, JCP277, 5871-5895 (2008). The authors here present a method for smearing out the source terms, which greatly improves stability.

Good luck!

-KE

Hi Kristian,

Have you encounter unresonable results of highe pressure and velocity beyond the interface within hardt' model?

Thanks!

Wond

[wond]

Originally Posted by neilduffy1024

Hi all,

I am writing UDFs for a drying model. I have UDFs for heat transfer between the two phases (implemented as energy sources) and a temperature dependant evaporation model (implemented as a mass source). The latent heat of evaporation is accounted for in one of the energy source UDFs. All works fine up until evaporation begins to occur, when the temperature fluctuations are worse than expected.

I think this is because I don't have an energy source to bring the evaporated gas up to the appropriate temperature - from user manual, "mass sources enter with no thermal heat"

For example, for a droplet of liquid water which is evaporating. Aside from heat transfer, I think that I should have an energy sink (from gas phase) to account for the latent heat of evaporation, and an energy source (to gas phase) to bring the water vapour (mass source to gas phase) from 0 K (i.e. no thermal heat) to say 373 K. This would be the enthalpy of water vapour at 373 K (and say 1 atm) multiplied by the mass source??

If someone could please confirm, it would be much appreciated.

Many thanks,

Neil

Hi, Neil

In my opinion, we should set a reference temperature (T_ref) during the simulation. The enthalpy is related to such tempurature point (T_ref). The sensible heat could be caluted by the Cp(T-T_ref) which is a part of source term.

Of cause, I'm not very sure.

Wond