Gordon,
This has been an interesting thread. I too have struggled to duplicate prototypical engine operation and speeds in scale model engines. In my opinion the most important factor is a well built engine. Terry Mayhugh has set the standard for this and his work is amply displayed on this website for all to learn. It isn’t easy, but it is especially worthwhile in slow speed engines in particular.
Flywheel inertia cannot be duplicated in a slow speed prototypical working scale model engine. Somebody here once ran the calculations and determined that fact, not me, but I am a believer. However, you can come close without going too far beyond scale. In my experience the flywheel is the second most important factor to a good running slow speed engine. I am referring here to throttled engines and not H&M. You need oversized flywheels with as much of the flywheel mass as possible located at the rim of the flywheel. This is evidenced by the oversized flywheels (diameter) of electric lighting engine designs of the early 20th century both throttle governed and H&M.
I recently attended a large antique tractor show and was amazed at how slow an old two cylinder tractor engine like the Rumley Oil-Pull could idle, and how long they spun over after ignition cut off. Lot’s of inertia. These were long stroke engines built in those days for low speed torque and lugging, but they also ran very smoothly at very low idle speeds quite unlike single cylinder tractors. One of the issues I haven’t seen addressed is the volumetric efficiency (VE) of a slow speed carbureted scale model engine. In a model VE will be very low due the throttle plate restricting engine breathing on the intake stroke as well as reducing the dynamic compression on the compression stroke. This is exacerbated at no load idle operation and often evidenced in model engines by a steady 8 cycling operation that requires two intake strokes for each firing stroke. This is likely due to too much of the previous combustion gases remaining in the cylinder. One way to mitigate this condition in some model engine designs is to increase the compression ratio slightly to about 5:1 or 5.5:1. The other way that apparently works is to build a longer stroke multi-cylinder engine. This conclusion is solely from my observation of building a .9” bore x 1.4” stroke Titan two cylinder model engine that often idles smoothly at 400 RPM which is only about 200 RPM below its normal operating speed. Yes, it has heavy flywheels but not oversized. I believe this long stroke low compression design breathes a little better and longer at very low speeds and the additional more gentle power pulses of the second cylinder make this design better suited to slow speed. There is too much inherent friction in a one cylinder model engine of equivalent twin displacement.
Carburetion for slow speed model gasoline engines has always been difficult using liquid fuels. I have build a dozen or so designs without satisfaction. No one has come up with a tiny version of a good small multi-jet self compensating carburetor for gasoline model engines. Using the smallest available venturi carburetor has always worked out best in my experience, but they have all still been too large for slow speed engines. I am down to one gasoline fueled model engine currently, and all the rest are now propane fueled. Fuel atomization is no worry and it just seems to work better for me as most of my engines are constant speed. Easy starting and clean combustion chambers are great additional benefits.
In this thread George Britnell made a statement that he didn’t think that his Holt model engine running at 700 RPM would develop much power. I would invite you to view a video of my 1906 Bruce-MacBeth Model Gas Engine Electric Plant video on Youtube. I built this engine in 2008 and one day my curiosity about the power output of this engine got the best of me. I decided to do an informal load test, but with some trepidation. The engine is a 1/10th scale 4 cylinder engine, 1” bore x 1.2” stroke, 4:1 C/R, gas engine running on propane and driving an alternator that I fabricated. The alternator efficiency is unknown, but is assumed to be rather low. At the time of the load test, the output of the alternator was 100-125 vac depending on engine speed, but normal operating speed was about 600-650 RPM. I had a string of old Christmas lights with 6 watt incandescent bulbs. Their wattage rating proved accurate. I kept the voltage at 115-120v during the test by increasing engine speed as I screwed in one bulb after another. I got to 15 bulbs brightly lit at 115v but then chickened out to add any more load at the top speed of 700 RPM. I did not want to ruin a years worth of work for another bulb or two of load. The engine was still able to respond to more throttle which I tested only momentarily. The 700 RPM was a little faster than a real B-M engine rated speeds, but not by much. When you compare the model and prototype engine cylinder displacements the model engine should scale closely to the real prototype in power output. Granted this was not an accurate test, but was proof enough for me that these little engines can produce near scale power at near prototype speeds.
My engine projects have been undersquare designs simply because they are modeled after real slow speed prototype engines from the beginning of the industrial gas engine hay days. It is reasonable to believe this bore/stroke design was quite effective in providing the high torque needed with the low speed engine designs of the times and working models should confirm this.
Jeff
