That paper does sound interesting. Is it available?
The part I am not getting is how the engine recovers from a misfire. If it misfires once, what changes that allows it to fire again?
If it misfires, no exhaust is left behind, and that entire stroke is used just to scavenge out the exhaust remaining from the previous firing. Next time, you are starting over again from nearly scratch. This also explains why the two-stroke firings at the same mixture are much less powerful than the 4-stroke firings - when you are just lean enough or loaded enough to firing every time, lots of exhaust it left from the previous firing, making the current firing partially pure charge and partially old exhaust. The difference in the pressure is very large, 50% difference or so.
That's also why 4-strokes (that is, valved engines that can only run in a 4) are so much more efficient, run hotter, and have extremely powerful power strokes. In a proper 4-stroke, you have an entire stroke used just to clean out the old exhaust, which does a very good job. On a 2-stroke, a fair bit of the incoming charge just shoots out the exhaust, and, you are just hoping most of the exhaust products are driven out when the exhaust port opens. At least some of it is still hanging around when the port closes, meaning it gets trapped and you are not firing a clean charge every time.
When you finally get the load or mixture low enough, it doesn't fire every time, meaning you get nearly nothing on that stroke* but you do force almost all the old exhaust products out, meaning the next one fires very strongly even though it is too rich to be ideal. Rich enough or low enough load, and it misfires on the next firing, and all that stroke does is add some scavenging.
Note that both of the run modes with 2-stroke engines end up doing things very inefficiently in terms of fuel consumption, which is why they tend to use A LOT more fuel, and, why changing the various scavenging features can have a huge effect on the power, efficiency, and running temperature. And why you nearly never see 2-strokes in motor racing aside from motorcycles, despite the fact that they are generally *much* more powerful for their displacement than 2-strokes. (typically 50-100% more, that is, *twice the power for a given displacement*).
Brett
*but not entirely nothing. After looking at Scott Bair's cylinder pressure data, it seems like the "dead" stroke on a 4-stroking 2-stroke engine actually adds a bit of energy/RPM - that is the compression stroke slows the prop down, the expansion (with no firing, mind you) speeds it up more than it was, slightly. My interpretation is that this engine is stealing some of residual heat to the unfired charge, like a Stirling engine. It's not free, because it also cools the rest of the engine and causes the next actual firing to be slightly less energetic.