Modeling a spark for hydrogen-air combustion

[H G]

Hi,

I wrote a code to model an ICE with premixed air-hydrogen and spark ignition using a mass flow controller with plasma (as suggested in an older post). At stoich, combustion doesn’t start unless I raise radical mass and temperature to very high levels and tweak composition.

When it does ignite on forced/wrong parameters, the later cycles stay at very high temperature (don't fall below 2000K) even though the HRR and pressure curves show no combustion.

Could someone help me spot the error? Code attached. Thanks!

[Ray Speth]

Hi,

I'm not sure that a single-zone homogeneous reactor is quite up to the task of modeling an SI engine cycle, which is fundamentally controlled by a flame propagating from a localized ignition source. Using this model, you are forced to apply the spark over the entire cylinder, where as you note, it requires a large perturbation to the state.

In later cycles, I think your model is in fact showing combustion occurring, just not when you expect it. Because the temperature of the end gas is so high, the fresh reactants are burning as they are drawn into the cylinder, and hence the temperature never really drops. You can see this in the heat release peak that occurs roughly between 0 and 180 degrees, while the inlet valve is open.

As a simple model of an SI engine, you'd be better off just representing the combustion process using chemical equilibrium, and ditch the radical igniter idea. You could stop the simulation at the point where you want the spark to fire, equilibrate the cylinder constants at constant U and V, synchronize the reactor state, reinitialize the reactor network, and continue.

Regards,

Ray

[H G]

Thank you!

[H G]

Even if I ditch the spark igniter, let's say just combustion ignition of hydrogen with higher intake air temp, I'm confused on two things:

(1) why the model would show a lambda of 1.2 (max NOx) having similar NOx emissions than a lean lambda of 2.5 (should be near 0)? Max temperature is lower for the lean charge but NOx is higher. Is this due to the higher fraction of nitrogen for the lean case?

(2) why would mass flow out be larger than mass flow in? This happens even if the inlet pressure to the intake valve is the same as the outlet pressure of the exhaust valve.

 Attached is the same code modified to show the outputs (toggle line 3 and 4 in the "Cantera Set-Up" block for question 1; the mass flow is the output of the simulation integrated over time, shown in the outputs of the "Performance" block)

[Januario Hossi]

Hello,

I believe that your work aligns with my current interests in improving hydrogen and hydrogen carrier fuel combustion using ICE.

I would appreciate the opportunity to connect with you, if possible, to collaborate and exchange ideas, including Cantera.

My contact information is as follows:

hossij...@gmail.com  

+244 947 008399 (WhatsApp)

Thank you for connecting.

Best regards,  

Januario

--
You received this message because you are subscribed to the Google Groups "Cantera Users' Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cantera-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cantera-users/5bb719bb-04a8-4d38-b126-3d860aeef0f8n%40googlegroups.com.