[huramentzefix]

My car drives with either petrol or propane at the momentary condition & ECU.
The gas injection is accomplished by an aftermarket ECU called KME Nevo which I have just stuck in series with the original injectors.
When I switch to has then the KME ECU get's the signal from the original Petrol ECU through the injector wires and adapts the signal to gas.
The Propane stoich is 15.5 so the original ECU with the narrowband will always keep it around 14.7 which is a bit rich and should be good.

When I switch to speedunio, should I keep the system as it is and have the map switch when the gas is activated or should I have arduino do the entire gas injection?
Right now the KME ECU calculates the gas density based on themperature and pressure.

The stoichiometric air/fuel ratio for CNG is 17-to-1 while that for propane is 15.5-to-1 compared to 14.7-to-1 for gasoline. You can see that it takes more air to maintain the right air/fuel ratio in a modified engine.

CNG, LNG and propane all burn slower than gasoline so they typically require more initial spark advance and a faster advance curve than a gasoline engine. Total timing advance, however, should usually not exceed 30 degrees.

CNG is a very high-octane fuel with a research octane rating of 130 and pump octane rating of 120 or higher, so it can handle a higher static compression ratio than gasoline in a spark ignition engine. In a diesel engine, however, the static compression ratio usually has to be reduced if the engine will be run on CNG exclusively. The ideal compression ratio is in the 12.5- to 12.7- to-1 range.

Propane is also a high-octane fuel with a research rating of 110 octane and a pump octane rating of 104. Propane typically works best with a static compression ratio of around 10- to 10.5-to-1.

sorry for the grey on the light grey, and grey on white it's not my choice

I will install an eaton supercharger from a mercedes which has a magnetic clutch. I want it only to be engaged when on propane. Thanks to the excellent properties of propane I should be able to give the head-gasket some workout?

It would be nice to be able to fully implement the propane into the arduino, how can that be done?

[huramentzefix]

the grey on light grey seems to be when editing and is actually admirably black then when sent!
thank you!

[PSIG]

… When I switch to speedunio, should I keep the system as it is and have the map switch when the gas is activated or should I have arduino do the entire gas injection?

If the gasoline ECM regulates to stoichiometric (14.7:1), then it should also regulate to stoichiometric for another fuel (propane 15.5:1), as it is also stoichiometric. The sensor does not know what fuel is used.

The sensor reads in Lambda (stoich 1.0) and multiplies by a factor, based on the fuel. So stoich * gasoline 14.7 is reported AFR 14.7. If reading propane at stoich (L1.0) but the factor is gasoline, it will report 14.7 when propane is at 15.5. Still stoichiometric.

This is also true for other AFRs, so gasoline 13.2:1 (10% rich) will also read 13.2 when propane is 10% rich (Lambda 0.9 * 14.7 factor). While gasoline may have best power at 10% rich, another fuel may have best power at 15% rich, or other value. This is why pro tuners prefer to use Lambda instead of AFR, as the reading is correct no matter what fuel or blend is used.

Using AFR is popular only because most inexpensive AFR gauges read that, but Lambda is preferred as it is simple for any fuel. I hope that makes sense.

[huramentzefix]

thanks a lot for the explanation, that makes perfect sense now.

[PSIG]

Glad my explanation helped. To highlight the simplicity of Lambda; if you are 10% rich of stoich for propane, and want to change to 12% rich, you have to calculate from 15.5:1 to 10% less (15.5 - (15.5 * .90) = 13.95) , then again for 12% less, etc.

But if you use Lambda, it's 1.0 stoich, 0.9 at 10% rich, or 0.88 at 12% rich. Just enter .88 and you're done without a calculator. Plus it works the same for gas, propane Diesel, whatever.