[noraishah@nm.gov.my]

Hi all!
Currently, I am dealing with flue gaseous which comprise of O2, H20, Co2 and N2 and enter the cooling tower (quencher) at T= 950C with operating pressure at 1 atm.
The geometry of the tower is H= 12.5m; D=2.2 m and the inlet diameter = 1.0m
The calculated density of the flue gas is =0.277kg/m3 ;gravity is enabled at y-axis, thus g= -9.85 m/s
The outlet diameter is also = 1.0m

Boundary conditions:
1-Inlet velocity = 3.5 m/s; P=0, mass fraction of flue gas: O.075, H2o=0.12; CO2= 0.14; N2=0.67
2- Interior= mixture-fluid
3-Outlet pressure = 0
4- Wall: Room temperature

I enabled the cell zone ; with operating density=0.277, pressure=1 atm and gravity = -9.81 m/s

I used recommended URF to accelerate convergence but until now, the flow is still not resolved (has not achieved convergence).
Actually, the overall idea is to introduce DPM since spray water will be injected after the flow is resolved. BUt, due to this, I have to postpone my injection model.
Thus, I do hope that you can assist me with this problem and I really mean that and I am totally lost. Thank you!



For your informations, I have sought several youtubes for months and journals regarding my problems and follow the tutorials and endless trials and still to no avail.Hope to have your kind and immediate responds regarding this.

Attach is the results of my simulation works.

[MayTheFlowBeWithYou]

Currently, the easiest thing I can think of you can try is to lower the URFs despite the recommendations and see what effect that has.

In addition: What discretization schemes are used?

[noraishah@nm.gov.my]

Thank you very much for your prompt reply. appreciate that!

All second Order Upwind, except Pressure = Body force weighted

THe URFs used are:

P=0.3, density=1, momentum=0.7, TKE=0.5, T.V=0.7, TDR=0.5, Body force = 1, species=1; Energy =1
Is it ok if I proceed with next step which is injection inside the quencher (dpm) without resolving the flow?

Thank you

[MayTheFlowBeWithYou]

Well, the only thing you'd gain on starting the injection without resolving the flow is that you could see if whatever way you use to inject actually work, but your results will obviously be incorrect.

Personally, I'd first divide all URFs by 5 or 10 to see if the convergence improves. It is a problem you'll have to solve at some point anyway.

[noraishah@nm.gov.my]

Thank you sir, but can u please elaborate on idea of dividing the URFs? I dont get it

Appreciate your respond, TQ

[MayTheFlowBeWithYou]

Lowering URFs generally increases the stability of your solution, thus it might help to get rid of the oscillations in your simulation. In most of the cases where I have had an issue like yours, lowering URFs proved helpful to me. This is why I recommend trying that.

[noraishah@nm.gov.my]

owh ok, Thank you very much. I really get your point

appreciate your prompt reply, Again, TQ

[MayTheFlowBeWithYou]

Ok, no problem, let me know how it works out, I'm curious