Turbo

Note: I wrote this to clarify my own understanding of how things work, and I think I have the essence captured.  I apologize in advance to those with more experience than I with these engines; I am not trying to install myself as the grand pooh-bah here.  If, by my relative lack of 2-stroke flying experience there’s some tidbit of physics I have missed, or have something wedged into my mind sideways, I certainly invite discussion.  The egoless scientist humbly seeks the truth for the benefit of us all, not the puffing up of his own self-esteem.  Good God, that’s tacky!  So here goes:
The 4-stroke engine has the luxury of a full half-rotation in which to extract the fuel’s heat energy.  For the 2-stroke, with the low-pressure, low-leverage bottom of the power stroke cut away, less time is available so the duration of the combustion event becomes more critical. 
 The time interval required for a combustion event to occur is shortest when the air/fuel ratio is close to stoichiometric, and cylinder charge pressure and temperature are high.  Too far from stoichiometric on the lean side robs heat from the combustion process, slowing the reaction, while operating on the rich side robs heat too, but also creates a host of intermediate products of the incomplete combustion, like CO, particulates, etc., expensively cooling and slowing the reaction.  High altitude or part-throttle operation lowers the charge pressure, molecular mean collision frequencies, and hence lowers the reaction rate.  In the absence of exhaust-pipe resonance effects, 2-stroke thermal efficiency has got to be maximized when full combustion of each charge occurs as far ahead of exhaust port opening as possible (without pinging, of course).  This explains why the 2-stroke works most efficiently when the engine is slowed down by load, and is why leaning back to near stoichiometric at high altitude is important, as is not over revving the engine at part throttle in cruise or descent.  Part-throttle over revving lowers charge pressure and reduces time available, causing the combustion process to finish up relatively late in the cycle, wasting fuel and raising EGT, which looks like, but is distinct from lean-of-stoichiometric operation.  This potentially delivers the thermal double-whammy of less cooling by charge expansion and combustion still ongoing as the ports open.  This thermally stresses the pistons, and favors ceramic-coating the piston tops as a protective measure.  Piston cooling only comes from the underside’s exposure to the cooler incoming charge and heat conduction through the piston skirts to the cooler cylinder walls.  The piston’s aluminum has very high thermal diffusivity, but a relatively low melting point.  Here, the EGT relative to limit serves as a guide.  Yes, these engines like to run with carb slides open!
 We’ve all heard of those model airplane 2-strokes that turn 30,000 rpm.  This works since the much smaller cylinder diameter substantially reduces the characteristic time required for the combustion event to occur (assuming the same speed for the flamefront).  So how about a greater number of smaller cylinders?  Would a different engine design help here?  As it turns out, holding engine displacement constant, characteristic combustion time varies inversely with the cube root of the number of cylinders divided by the two we have, while scrubbed cylinder area goes up by that same cube root.  So frictional losses go into this tradeoff.  Going to a 3-cylinder engine only reduces characteristic time by 13%, and it would take 16 cylinders to reduce it to ½ of its 2-cylinder value, while engine scrubbing friction would double.  There’s very little kindness there, unfortunately.  The additional complexity of more cylinders would increase cost and weight too. Oh well!  Let’s look instead at the propeller.
 If the (fixed-pitch) propeller is set up for brisk climb performance, it will be very much unloaded in cruise, and only able to absorb a fraction of the engine’s power potential.  I was surprised to find out by how much.  My prop calculations confirm that my 3-bladed 72” IVO prop, with the pitch set to absorb all 65 hp at Vy in climb will only be able to absorb about 15 hp in 85 mph cruise at 5000 rpm.  This may not be enough power to even go 85 mph, and is very likely an over-rev condition, the only cure for which is more prop pitch. (BTW I am running at a higher pitch than this!). The cleaner we make our airframes, the worse this problem becomes, as the cleaner airframe demands even less thrust power for a given speed in level flight.  Ground setting the prop to a higher pitch will obviously reduce climb performance as the engine’s max rpm drops, but thrust power available for cruise at, say 5000 rpm increases dramatically, increasing cruise speed a little (at high speeds, achievable speed varies roughly with the cube root of the thrust power.).  If you really want the best of both worlds, the answer is the IFA prop, but now you no longer legally have a light-sport category airplane.
 Another way to mitigate this disparity is to go to a smaller prop diameter or perhaps fewer blades.  This requires you to increase prop pitch in order to properly load the engine. At the higher blade pitch, the prop may be partially stalled at takeoff, although now it’s less sensitive to forward speed, and will unload less in cruise, allowing the engine to not overspeed as much.  This is due to simple geometry, as changes in blade element angle of attack corresponding to changes in forward speed are smaller at the higher blade angles and pitch.   But turning slower means less potential power from the engine.  There’s got to be an optimum in there someplace, but it’s likely different for different folks, depending on how they fly their airplane.  Hopefully the optimum is not at the sickly climb performance of a C-150, as this would totally destroy our street cred as a STOL airplane! 
 Incidentally, calculations for my 72” IVO show that going from 3 blades to 2 is impractical since this makes the prop too small, causing serious blade-element stall and horribly low efficiency when the prop is pitched to absorb 65 hp at Vy.  The alternative of cutting a prop down is risky in that you can’t glue it back together if you cut off too much.  Then there’s the sensitive prop-balance issue. I’m not too keen on this approach yet.
 At this point I will start feeling out the tradeoff between climb and cruise performance.  It occurs to me that I have a lot of climb performance to comfortably give away, but I have not yet flown my little Avid with two of us onboard.   My gut feel is that the STOL variant may not be as slow a cruiser as previously thought; it’s just that most of us are not keen on trading away much of our stellar climb performance.