# Wall Temperature for Subcooled Boiling

Member

Dear all,

I have done my simulation for the case of subcooled boiling until 40k iterations with the URF is 1 for energy. I have coarse mesh for around 350k elements and also I made another model that insert outer diameter of the domain which increases the mesh into 630k elements.

The reason why I stopped the simulation until 40k iterations is because the mass imbalance for both case I have reached below 1% and try to get the data from the wall temperature for calculating the quality of the case.

Nevertheless, the wall temperature showed something strange value, because when the wall temperature is higher than saturation temperature it means subcooled boiling occurred according to the ANSYS FLUENT help which means the bulk temperature is still below than saturation temperature.

However, the bulk temperature is higher than the saturation temperature which gives me a problem to calculate the quality of the volume fraction itself. Do you have any recommendation what should I do to obtain a reasonable value for the wall temperature and bulk temperature? My assumptions are on the mesh refinement or I need to continue my running until certain iterations while changing some URF.

Thank you in advance,

Luthfi Ady Farizan Haryoko

## Comments

• GermanyForum Coordinator
edited February 2019

Why does it give you a problem? If the bulk is superheated then it would evaporate.

For RPI please use coarse meshes towards the wall (Yplus>30) for stability and correlations reasons.

• GermanyForum Coordinator
edited February 2019

Subcooled boiling would ideally refer to the inflow conditions. Which flavor of RPI are you using?

• Member
edited February 2019

Because I would like to validate my case into the experimental investigation, and the value is far away from approaching the exact same value with the experimental case. Also, I found out that my wall temperature is slightly different according to the total length. Based on my literature review, the wall temperature increases when the length of the coil increases as well. For example, I have my wall temperature in 5.1 m length is around 547.44K meanwhile when I predict the value for the length of 5.4 m from the inlet is decreasing into 540.76K.

Do you have any suggestion regarding this issue?

Yes, I haven't changed my mesh and it's still around >30 for the Yplus value.

Thank you for your consideration,

Luthfi Ady Farizan Haryoko

• Member
edited February 2019

For RPI, I choose the first option which is RPI wall boiling model with the setting as described below:

Viscous: Realizable k-e, Standard Wall Fn, Mixture

Phase Interaction:

Drag: Ishii

Lift: Tomiyama

Wall Lubrication: Antal et al

Turb. Dispersion: Burns et al

Turb. Interaction: Sato

Heat: Ranz-Marshall

Mass: Boiling with the correction model (Fixed Yplus value=250)

Surface Tension: specified

IA: ia-symmetric

Best Regards,

Luthfi Ady Farizan Haryoko

• GermanyForum Coordinator
edited February 2019
I cannot give any solution as I do not know your case and whether it is appropriate to use RIP. If you deviate from nucleate boiling you will need to use CHF extension.

Physical interpretation might be that you have a good wall evaporation leading to a decrease in the temperature.

You need to run your case further iterations and monitor some RIP characteristic.
• Member
edited February 2019

My case is a helical tube that has an inner pipe diameter 14.5mm with the wall thickness is 3.5mm. For this case, I'm using the outer wall diameter so my outer dia is 21.5mm (14.5+wall thickness). the total length of the coil is around 8m.

My B.C is for inlet I put velocity on 600 kg/m2s with a temp 473.15 K. To obtain fully-developed flow, first I'm running my case with single-phase then the outlet velocity profile will be taken and put it into the inlet value for running my boiling model.

I followed each step that you suggested to me on the previous discussion like ramp heat flux for every 2000 iterations. I set the first heat flux is around 70% of reported heat flux (e.g: the reported heat flux is 150kW so the initial heat flux is 105kW)

My saturation value is 534.55 K for system pressure 4.8 MPa. For the solver setting, I used Coupled for pressure-velocity coupling and placed the first order for each of discretization. For the controls, I vary the setting from 0.3-0.5 for every URF except energy that I will increase step by step from 0.6 until 1. I'm running my case using pseudo-transient with the timescale factor 0.1.

Oh I see, is that because of the vapor blanket that covers up the wall so the temperature will slightly decrease? Due to your physical interpretation, it seems to make sense for me, meanwhile, I have found another wall temperature that has good value when the length is increased from the inlet point. Any help is appreciated.

For your information, I have run the case until 45k iterations, do you think I need to go further to gain a possible solution?

Thank you in advance,

Best Regards,

Luthfi Ady Farizan Haryoko

• GermanyForum Coordinator
edited February 2019

Use another flavor of RPI as the basic RPI is not suitable for superheated vapor and close or post DNB regimes. It assumes zero resistance and convection to the gas phase.

• Member
edited February 2019

Dear Amine,

For your reference, I put my vapor temperature as my saturation temperature, so ideally it would refer to the basic RPI modeling that I used to solve my case. The purpose of this case is to investigate the subcooled boiling phenomenon inside the helical tube.

So basically, I did not take any calculation for vapor temperature on the case, I just put fixed at the saturation temperature.

Highly appreciated for your kindly response.

Best Regards,

Luthfi Ady Farizan Haryoko

• GermanyForum Coordinator
edited February 2019
Try CHF kind of RIP you probably out of the nucleate boiling regime.
• Member
edited February 2019

Understood Sir, I will update to you soon for my case.

Best Regards,

Luthfi Ady Farizan Haryoko

• Member
edited March 2019

Dear Amine,

How you to get ideas that my case above must be solved using CHF condition? Is that any specific explanation that you could shared to me regarding this problem?

Because based on my understanding, my case above is related with subcooled boiling, so I need some strong justifications on why I should go to CHF condition?

Your support is highly appreciated,

Thank you,

Best Regards,

Luthfi Ady Farizan Haryoko

• GermanyForum Coordinator
edited March 2019

CHF extension does take into account for the nucleate boiling regime, prior and post DNB regime. So it is more general than the old fashion RPI which is based on the weak assumption of no convection to the gas phase from wall (might be valid only in case of small superheat : isolated vapor bubbles or few bubble columns)

Sign In or Register to comment.