[Cafard]

hi，all
i knew how to set CO2 gas and CO2 liquid
but now i have to simulate a cycle
every state is supercritical CO2
i don't know how to set material....
if someone know how to set or where i can download database
please help me
THX all~

[evcelica]

Just input the properties, what do you need? Supercritical CO2 properties? I use NIST RefProp for all fluid properties.

[Cafard]

thank you,evcelica
i really need supercritical CO2 properties
but i don't know what is NIST RefProp
could it direct input to CFX material
if it is,how could i get it?
THX

[evcelica]

Use Real Gas option:

Molar Mass: 44.01kg/kmol
Critical Temp: 304.13 K
Critical Pressure: 1070 psia
Critical Density: 467.6 kg/m^3
Acentric factor: 0.22394

Fourth Order Coefficients: (Range: 304.13K to 600K)
CPo/RCO2 = 4.28465E-09*T^3 - 1.00172E-05*T^2 + 0.010372394*T + 2.149248005
a1 = + 2.149248005
a2 = + 0.010372394 [T^-1]
a3 = - 1.00172E-05 [T^-2]
a4 = + 4.28465E-09 [T^-3]
a5 = 0 T^-4

Use default kinetic theory options for viscosity and conductivity.

*No Warranty Expressed or Implied*

[Cafard]

thank you,evcelica

the process i do is :
1.) import library data
2.) choose redkw.ccl
3.) import real gas combustion > CO2
4.) change Fourth Order Coefficients (as what you say)

does it CO2 material can simulate supercritical state?
(does the CO2 material > basic setting > thermodynamic state > gas is right? or change gas to liquid?)

i have done some easy case to prove the setting is right
but when i compare the simulate result with NIST RefProp deta
the error is 20% at least
(even i use real gas combustion>CO2 default setting)

[evcelica]

You can leave it as gas, or don't specify anything. It will give you properties for the supercritical state as far as I'm aware. I just did a simulation using this method for helium, and it came out in agreement with other published data.

I didn't import any .ccl data, I just made my own material, but it sounds like it should come out with the same results.

How are you comparing it to RefProp data?

[Cafard]

in CFX-Post i use probe to find one point state
then i key the pressure and temperature to find density (RefProp)

i try to create a new material
and follow the process as you say
but the CFX-Pre warring that i need Thermal Conductivity and Viscosity
then i turn on Transpot Properties (Viscosity and Conductivity)
and use the same as real gas combustion>CO2 set
i also set max./min. temp. and pressure to simulate work condition
but the result still wrong


i have searched on internet for days
i found some method that import a excel data or RGP file to create new material
but i don't how to do it or create RGP file

[khoopes]

I have done simulations with supercritical CO2 in CFX using an RGP table generated by REFPROP. It is not a straight forward process, I would recommend contacting ANSYS support and they can walk you through it.

[Cafard]

hi ， khoopes
thanks for your recommend
but I still want to know how to simulate S-CO2 state by using "real gas combusion > CO2" data

I have studied for a week,I found some parameter can be seted (advance option or expert parameter),does it have to set?

[khoopes]

You use refprop to generate a table and then CFX uses the table during the computation. I would contact ANSYS support and they can set you up with the refprop table generator.

[turbo]

If you move around all state changes in the cycle, you need to use the RGP file instead of R-K fits.

[Dinar]

I hope this will be helpful:
https://www.reddit.com/r/CFD/comment...al_properties/

[-Maxim-]

guys... this thread is 2 years old...

[ibraheem.215]

Dear
You can generate the properties of any fluid through REFPROP like viscosity, density, thermal conductivity etc. But this software is not available online. I am simulating the same problem by using supercritical water but i am facing convergence issues. If you guyz know about this. My channel is 2*2 rod bundle with constant heat flux and water at 25MPa

[aero.senthilkumar]

Dear all,
I am trying to simulate SCO2 flow through convergent divergent duct and 11 bar and 340K with Fluent. When I used ideal gas EOS, solution getting converged. But when I try to use with other EOS(RK, PR, NIST), I am getting divergence after some iterations.

My case details are:-
Steady flow
Density based solver
Please reply. Thanks in advance.

Regards,
Senthilkumar

[alirezame]

Due to low compressibility factor of CO2 near the critical point and inside the supercritical dome, ideal gas assumption decreases the accuracy of the result significantly. plz tell more information and what have you done to prevent divergence? (time step, discritization, look-up table range and size, initial solution, boundary conditions, ...)

[aero.senthilkumar]

Dear alirezame,
Here is the details of my settings.

Timescale I set as "Auto",
Discretization as "Upwind" for initial iterations then "High Resolution",
Look-up table range "217K-1100K" ; "0.1 Pa - 100e5 Pa" with data points of 800.
For real gas simulation initial solution is "Results with Ideal gas EOS"
My domain simple C-D Nozzle, Inlet as Inlet BC with total pressure of 11 bar and temperature of 320K ( I tried with initial temperature of 500K, but when I tried to decrease slowly to 320 , I facing error "retrun with code-1" in cfx)
Outlet BC is "supersonic"

Please correct me where I am going wrong.
Your reply will be much helpful.
Thanks.

Originally Posted by alirezame

Due to low compressibility factor of CO2 near the critical point and inside the supercritical dome, ideal gas assumption decreases the accuracy of the result significantly. plz tell more information and what have you done to prevent divergence? (time step, discritization, look-up table range and size, initial solution, boundary conditions, ...)

[alirezame]

Hello,

I would recommend to not use super sonic boundary condition in first simulation. use outlet pressure first and after getting convergence change the boundary condition to super sonic.

[alex-bishop]

Originally Posted by evcelica

Just input the properties, what do you need? Supercritical CO2 properties? I use NIST RefProp for all fluid properties.

hello! can i ask you a question about how to generate a table of SCO2 by NIST? i want to import this table into CFX.
if you have some more direct method please tell me too. for example setting in CFX-pre directly.

[Fresher]

[QUOTE=evcelica;495712]Use Real Gas option:

Molar Mass: 44.01kg/kmol
Critical Temp: 304.13 K
Critical Pressure: 1070 psia
Critical Density: 467.6 kg/m^3
Acentric factor: 0.22394

Fourth Order Coefficients: (Range: 304.13K to 600K)
CPo/RCO2 = 4.28465E-09*T^3 - 1.00172E-05*T^2 + 0.010372394*T + 2.149248005
a1 = + 2.149248005
a2 = + 0.010372394 [T^-1]
a3 = - 1.00172E-05 [T^-2]
a4 = + 4.28465E-09 [T^-3]
a5 = 0 T^-4

Use default kinetic theory options for viscosity and conductivity.


Hello,evcelica~
Thanks to your advice.I've succeeded in computing with these settings you'd given.But I still wonder how the expression "CPo/RCO2 = 4.28465E-09*T^3 - 1.00172E-05*T^2 + 0.010372394*T + 2.149248005"came from.
If anyone there know that,please help me.I'll appreciate a lot.