[Niftyjig]

My data is polluted because the fuel was- I was desperate to get my machine to a fueling station, and I was angry and trapped. So, I grabbed anything I could find in a shed that looked flammable and poured it in.

This was about 2004.

The engine barely ran, and it turns out one of the small jugs had rust, water and other contaminants.

I did get it to run off and on, enough to go about 5 blocks (to a gas station), but each time it started was not enough to get any useful information.

After adding fuel, it was not much different until it ran for a short period, then it got interesting.

The heavy majority of the tank volume was kerosene (only had the ashtray coins). Once it started, I pressed the accelerator and it died. This is what heated it up, I believe, because it idled-- Rough, then cleared up. The second time, i was more careful. It didn't die until it was rolling down the road. After that, I played it safe. The engine would only operate in a narrow RPM band, and felt as if it had virtually unlimited torque. I wanted to make it home, though, so I didn't throw it into 5th, just to see what would happen in the name of science.

Fuel delivery was controlled by a mass air plate, with constant variable-flow injectors. This was a 79 rabbit, fitted with a GTI engine and a close ratio 5-speed box, closer than a convertible/GTi/GLi 9A. Possibly it was from a diesel. It is interesting to note that this engine and even the diesel version were based on a carbourated system which had the intake and exhaust manifolds on both sides, helping to atomize the fuel. My injectors went straight to the ports, but those ports in the head were even internally DIRECTLY adjacent to the hot exhaust.

The injectors are not designed to atomize that viscosity of fuel, though I suspect that due to their designed-in flexibility, they likely had better luck than duty-cycle varied, constant-delivery-rate injectors from a more modern digital proportional EFI system would have.

The fuel system setup meant I couldn't vary fuel delivery separate from air delivery. That would have been more useful data, and probly got me home faster.

After I parked, it would not start, a few hours later. I found ignition plugs covered in the viscous fuel, effectively fouling them and preventing any arcs. I cleaned them and it started rather easily, but was not drivable. I "determined" there was not available enough power to run on the road without garnering unwanted attention, which could result in massive fines if it was "determined" my tags were out of date and I carried no insurance.

It was likely this incident which led me to find fuel lines clogged, several years and a few fuel pumps later.

I got to this page trying to confirm a spark-ignition diesel existed, as the 1982 Audi 4000 diesel is listed as spark-ignited indirect-injection 1.5l (obviously a VW block) by automobile-catalog.com .I was there looking for gear ratios to help a smooth-shifting mod to an upcoming mogas conversion; we have found golf/4k/fox to be very strong, smooth and versatile boxes, avoiding the post-85/86 exploding diffs.

It's likely a typo, but I wouldn't swear to it. 4ks are rare enough I can't just go check a junkyard. Rabbits with the "comparable" engine were rated lower, the Audi 4k is listed as 52 on autocat.

The only really solid data I found was a military test, a230243.pdf found at dtic.mil .They outline achieving exactly what you proposed, with the caveat of fouling plugs (same issue I had). They used a "modified diesel engine", so likely were able to vary fuel input separately. They favored 12:1 compression ratios, for some reason, claiming it was somehow more efficient. The cold-climate instant-start and power benefits are clear, though. They also cited those. The document goes on to call for the design of an anti-fouler, but my interest stopped there. This is in 1990. Perhaps this is where those tubular anti-foul bushings on the rack at the auto parts store came out of, it was almost 30 years ago, now.

I have also seen designs similar to which you speak where the fuel was pre-atomized under extreme pressure. While normally used in power plants to run turbines on whatever they can get their hands on the cheapest, I have seen it done in a car. Then, you can take full advantage of the higher energy density in diesel, well beyond the normal 1/3 incidental higher MPGs vs. gasoline.

I got the idea to go to mogas (propane only, no butane from LPG) because the stock engine in a loved car had a high normal CO output (almost 1% total volumetric output). Since then, the NOx output, especially of diesels, has troubled me almost as much as the fact the issue doesn't get airplay. Incidentally, that same engine produces rather low NOx, I believe.

Either way, mo-gas will be far more economical and support an obscene amount of boost. Propane atomizes 100% into true vapor at any temperature on the globe, WITHOUT ingesting the extra air and heating it to burn nitrogen, like a diesel cycle does. I just have to pre-heat the fuel (familiar?) to atmo temp. If I want to add that high-pressure miracle stuff later, the gasoline-standard compression/ignition setup will still be ready-- I like the idea of being able to stock diesel on a site, it's not anywhere near as dangerous as the volatile or liquid-gas fuels.

There's another cycle that might be on the horizon that's compatible with diesel, though, so I'm not throwing out all those blocks.