# Supercritical flow in a duct_pressure blowup at initial time step

 Register Blogs Members List Search Today's Posts Mark Forums Read

March 7, 2017, 11:49
Supercritical flow in a duct_pressure blowup at initial time step
#1
New Member

Mahdi Nabil
Join Date: Sep 2015
Posts: 9
Rep Power: 10
Hi everyone,

I am trying to simulate supercritical CO2 turbulent flow and heat transfer in a duct using buoyantPimpleFOAM. So far, I have been using const fluid properties and it was working fine.
Currently, I am using a new EOS in which rho and psi (compressiblity) are both functions of pressure and temperature (since we are in a supercritical region. In other words,
rho=rho(p,T)
psi=drho/dp = psi(p,T)
And, I am sure about the psi formulation and implementation in the code which is the exact analytical differentiation of rho wrt p. However, when I use psi(p,T) it makes the pressure to be unstable at the very first time step and the code blows up. I've printed the output:

diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 1
DILUPBiCG: Solving for Ux, Initial residual = 1, Final residual = 5.15665e-08, No Iterations 2
DILUPBiCG: Solving for Uy, Initial residual = 1, Final residual = 4.17258e-07, No Iterations 3
DILUPBiCG: Solving for Uz, Initial residual = 1, Final residual = 3.51138e-07, No Iterations 3
DILUPBiCG: Solving for h, Initial residual = 0.999991, Final residual = 1.05686e-07, No Iterations 4
GAMG: Solving for p_rgh, Initial residual = 0.999569, Final residual = 4.66604e+56, No Iterations 1000
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors : sum local = 123.314, global = -15.475, cumulative = -15.475
************************************************** ********
**** Max Pressure = 1.50284e+61 **** Min Pressure = -8.4671e+60
**** Max Temperature = 313.004 **** Min Temperature = 312.996
************************************************** ********

I have attached my p, T, and U boundary conditions in the 0 folder. Does anyone have any idea if I am making any mistake in the boundary conditions? These boundary conditions work perfectly for a constant psi value, but blows up the pressure for psi(p,T).

Attached Files
 0_P_U_T.zip (2.2 KB, 15 views)

 June 19, 2018, 07:24 #2 New Member   Join Date: Mar 2013 Posts: 24 Rep Power: 13 Hi, have you solved your problems? I have done a same case with real equation of state, as well as new model for viscosity and thermal properties. I simulated supercritical flow of nitrogen. My initial residual for pressure kept a high value around 0.95 and never went down.

 June 19, 2018, 09:34 #3 New Member   Mahdi Nabil Join Date: Sep 2015 Posts: 9 Rep Power: 10 Hi Slanth I solved this problem with a trick. Since the pressure drop in my cases are in the order of 100 or 200 Pa, I used a new modified version of equation of state (PengRobinsonGasPConst) which takes an average pressure inside the channel and avoids taking into account the minor fluctuations in the pressure field when it tries to calculate rho. I do the same thing with other thermophysical properties (cp, h, mu, and Prandtl).

 Tags blowing up, boundary condition, buoyantpimplefoam, pressure, turbulent flow