Dear all,

I want to simulate propane/air premixed flame propagating in a curved duct.

I creat a 4-step reaction for it. And it is ignited by a source point.

But I find the reaction doesnt happen, i.e., the flame doesnt spread.

Who can help me!!!! Many thanks!

Could be a lot of things going on here. Some things to check: make sure your extinction temperature is set low enough for combustion to occur; try increasing the temperature of/amount of gas being injected by your ignitor point.

Barring that I would double check your reaction setup.

This may sound really silly, but I had a problem when I first implemented a reaction (it wasn't flame but liquid), however no matter what I did with the reaction rates it just did not react I couldn't work it out. This was until a friend of mine pointed out that I still had the reaction or combustion option set to none in the fluid models tab in the domain. If I remember correctly I set it to finite rate chemistry and it worked after that.

Hope this helps

Thanks all.

Maybe the post is a little longer.

The reactions I used in the simulation are as follows:

REACTION:Propane Air 4step

Additional Materials List = C3H8,O2,CO,CO2,H2O,H2,N2

Option = Multi Step

Reaction Description = Propane Air Four Step

Reactions List = step 2,step 3,step 4,step1 END REACTION:step 2

Option = Single Step

FORWARD REACTION RATE:

Option = Arrhenius

Pre Exponential Factor = 1.67E7[mol^-1 cm^3 s^-1]

Temperature Exponent = 0

REACTION ACTIVATION:

Activation Energy = 125615.47 [J mol^-1]

Option = Activation Energy

END

END

PRODUCTS:

Materials List = CO,H2

Option = Child Materials

CHILD MATERIAL:CO

Option = Stoichiometric

Stoichiometric Coefficient = 3

END

CHILD MATERIAL:H2

Option = Stoichiometric

Stoichiometric Coefficient = 7

END

END

REACTANTS:

Materials List = C3H8,H2O

Option = Child Materials

CHILD MATERIAL:C3H8

Option = Stoichiometric

Reaction Order = 1.0

Stoichiometric Coefficient = 1.0

END

CHILD MATERIAL:H2O

Option = Stoichiometric

Reaction Order = 1.0

Stoichiometric Coefficient = 3

END

END END REACTION:step 3

Option = Single Step

FORWARD REACTION RATE:

Option = Arrhenius

Pre Exponential Factor = 7.94E13[m^2.250 s^-1 K mol^-0.750]

Temperature Exponent = -1

REACTION ACTIVATION:

Activation Energy = 167478.97 [J mol^-1]

Option = Activation Energy

END

END

PRODUCTS:

Materials List = H2O

Option = Child Materials

CHILD MATERIAL:H2O

Option = Stoichiometric

Stoichiometric Coefficient = 1.0

END

END

REACTANTS:

Materials List = H2,O2

Option = Child Materials

CHILD MATERIAL:H2

Option = Stoichiometric

Reaction Order = 0.25

Stoichiometric Coefficient = 1.0

END

CHILD MATERIAL:O2

Option = Stoichiometric

Reaction Order = 1.5

Stoichiometric Coefficient = 0.5

END

END END REACTION:step 4

Option = Single Step

FORWARD REACTION RATE:

Option = Arrhenius

Pre Exponential Factor = 1.538E8[mol^-1 cm^3 s^-1]

Temperature Exponent = 0

REACTION ACTIVATION:

Activation Energy = 83743.64 [J mol^-1]

Option = Activation Energy

END

END

PRODUCTS:

Materials List = CO2,H2

Option = Child Materials

CHILD MATERIAL:CO2

Option = Stoichiometric

Stoichiometric Coefficient = 1.0

END

CHILD MATERIAL:H2

Option = Stoichiometric

Stoichiometric Coefficient = 1.0

END

END

REACTANTS:

Materials List = CO,H2O

Option = Child Materials

CHILD MATERIAL:CO

Option = Stoichiometric

Reaction Order = 1.0

Stoichiometric Coefficient = 1.0

END

CHILD MATERIAL:H2O

Option = Stoichiometric

Reaction Order = 1.0

Stoichiometric Coefficient = 1.0

END

END END REACTION:step1

Option = Single Step

FORWARD REACTION RATE:

Option = Arrhenius

Pre Exponential Factor = 3.49E10[mol^-0.75 cm^2.25 s^-1]

Temperature Exponent = 0

REACTION ACTIVATION:

Activation Energy = 125615.47 [J mol^-1]

Option = Activation Energy

END

END

PRODUCTS:

Materials List = CO,H2

Option = Child Materials

CHILD MATERIAL:CO

Option = Stoichiometric

Stoichiometric Coefficient = 3

END

CHILD MATERIAL:H2

Option = Stoichiometric

Stoichiometric Coefficient = 4

END

END

REACTANTS:

Materials List = C3H8,O2

Option = Child Materials

CHILD MATERIAL:C3H8

Option = Stoichiometric

Reaction Order = 0.5

Stoichiometric Coefficient = 1.0

END

CHILD MATERIAL:O2

Option = Stoichiometric

Reaction Order = 1.25

Stoichiometric Coefficient = 1.5

END

The unit for the Pre Exponential Factor in literature is kg.m-1.s-1. I dont know how to transform it to suit for the cfx.

And when the solver starts, I get the information: Enthalpy per [mol] of reaction at reference conditions (Pressure= 1.01325E+05, Temperature= 2.98150E+02)

step1 = 2.2776E+05 step 4 = 4.1159E+04 step 3 = 2.4185E+05 step 2 = -4.9779E+05(I dont why this step's enthalpy is negative.)

The ignition I define is just like this: spark = (0.005[kg m^2s^-2]/sparktime)*step(t/1.0[s])*step((sparktime-t)/1.0[s]) sparktime = 1.0*(10^-3)[s] Are there any other better ways to get the premixed gas ignited?

The way I define my ignitor is is via a sourcepoint with total mass flow on the order of 1% of the total domain flow. I inject only combustion products and set the temperature very high, usually to what I expect the temperature to be if combustion was taking place. I give it some small velocity component in the axial direction, usually only .01 m/s. Depending on your turbulence model you'll have to define some other values but I usually use SST with k = 210.655 [m^2 s^-2] and omega = 277.3 [s^-1] (based off a jet-in-crossflow study).

I also set up a serires of monitor points for temperature downstream of the ignitor in order to judge whether or not I'm combusting. I'll place one just aft of the ignitor, and then a series of points at different axial locations downstream, usually three points or so at each axial location that are located at different y coordinates so I can get an idea of the temperature profile.

Once I've determined that combustion is taking place I shut the ignitor off by setting its massflow rate to zero. It's important to shut the ignitor off and let the model run out before considering the results converged.

This method has worked consistently well for me, at least with the 23-step finite rate model I use.

Something I just noticed on your spark equation is that it's based off time, so unless you use a sufficiently small timestep you will have problems igniting. That goes for the ignition method described above as well. I usually use a physical timestep of .0001 [s] for my combustion CFD. If your's is significantly higher I might try rerunning it with the smaller timestep before trying this method.

John