Learning Forum

[Lukas Walther]

Hello im simulating methane combustion partially premixed (chemical equilibrium). Im analyzing the effect of diffrent pressures. One simulation is computed with 1 bar and the other one with 10 bar (mass fractions are the same). Boundary conditions are the same in bouth simulations.
According the literature the flame should be shorter under higher pressure, because the flame spread is accelarted.
Im getting the oposite....the flame under pressure is longer than the one under atmosperic pressure.
Why? ans idea?
Simulation (10 bar)Simulation (1 bar)

Heat release rate is plotted.

Thanks

 

[Lukas Walther]

PS: sorry for the scaled screenshots. bouth simulation are made with the same geometry and mesh

[Rob]

Can you replot using the same heat release scale?  It's hard to tell if there is a difference because we're not looking at the same thing. 

[Lukas Walther]

The 10bar simulation has a 10x higher poweroutput due to the increased density. if i apply the scale of the 10 bar simulation to the 1 bar simulation u can hardly see anything.

[Lukas Walther]

This is the 10 bar simulation with the same scale as the 1 bar simulation. i dont know wheter this is useful or not, because its limited to 1350 kg/m^3 s.

[Lukas Walther]

note: the hrr in this case is a named expression: Productformation rate * Density, not the default hrr of the pdf section.

[Rob]

Now compare the flame length: I can't easily do this as image size and position make it difficult.  What gas density did you use? 

[Lukas Walther]

on this screenshot it is cleary visible that the 10 bar flame is longer. the density in the 10 bar simulation is 11.4 kg/m^3. phi 0.7, 100 % CH4, 11.8 m/s inletvelocity

[Rob]

On the above it's clear that the reported range is different too. If the report is product formation * density and there's a big density difference between the two models can we use that report as a metric? 

[Lukas Walther]

yes i think it can be used as a metric

[Rob]

So, next is check convergence and the assumptions in the model. How do the flow fields compare? Flame speed may be higher at high pressure (there's a curve as I vaguely remember doing it for my PhD) but combustion also required fuel & oxident to be in the right place at the right time.