"I haven't had a chance to be involved with a CNG program, but I did take a class on the fundamentals of internal combustion engines, by which I mean fundamentals of gasoline-fueled engines. In a gasoline engine, the fuel is stored as a liquid and passes through the fuel system to the injectors as a liquid, and the injector atomizes the liquid so that it can vaporize to be burned. The obvious difference between that and CNG is that CNG is stored as a gas, transported as a gas and metered into the engine as a gas. So the fuel storage would have to be as high-pressure gas cylinders, and the fuel lines and injectors would have to be large enough to carry enough fuel to the engine in already-vaporized form. On the good side, the problem of vaporizing liquid fuel under cold conditions, like engine start-up, disappears. CNG is already gas, so as much gas as enters the cylinder will burn (assuming there's enough oxygen). This would improve cold start emissions compared to those from gasoline. The oxygen sensor shouldn't have any trouble detecting the stoichiometric point of the CNG, so fuel control should not present any fundamental new problems, but the stoichiometric point would be different than for gasoline, so that might produce a need for some changes in calculations. Since the fueling system would be so different, I guess you wouldn't need a charcoal canister - no liquid fuel vaporizing and trying to escape. I would be thinking a lot about the safety aspects of potential tank rupture and leakage, there would need to be new diagnostics for that, and of course the gas cylinder pressure would have to be down regulated as it leaves the tank, or at least, for sure before it reaches the fuel injectors. I am sure there is lots more to it, but these are a couple of the things I can think of right away without having had a chance to really delve into it."