Saito has a reputation for being a docile 4-stroke engine. I have considerable experience with the YS 120 & 140 engines. A lean mixture is the usual culprit with backfires. Not enough "flywheel" (prop in my case). Use only OS-F glow plugs, Enya #3 is next choice. 5-10% nitro fuel. The addition of nitro gives a broader needle setting. This is due to the higher percentage of oxygen in the fuel. The YS engines like more nitro due to the supercharging. The engine will run best with the factory muffler in place. The back pressure helps with a consistent mixture. Cooling the engine may be a major problem. They are intended for model airplane use with lots of airflow over the cooling fins. Excessive heat leads to a lean mixture. In an earlier time castor oil was the best available lubricant for all glow fuel engines. Castor carried a lot of the heat away in the exhaust. A change in processing to extract the oil from the bean has changed that condition. It is now leached out with solvents rather than pressed out of the bean. The source I trust says the end product still appears the same but the new process does not work well in our engines. I choose to believe them. Castor also deposits a varnish layer in the engine that must be mechanically removed. Saito no longer offers a warranty on engines run on castor. Synthetic lubricants are recommended. Valve adjustment: I prefer to adjust the valves so there is just a perceptible amount of clearance when they are fully open. Wiggling the rockers with your fingers is best. As the valves seat in the clearance will get tighter. Loose valves allow hammering that will damage valve train parts. A lean run can damage a ring (and sleeve!) immediately. Turning an engine over by hand is not a reliable compression check. An engine running at full throttle that runs out of fuel nearly always dies with a backfire. How much energy in a backfire? Enough to break a crank shaft! How much RPM? 12000 is way more than anyone runs this engine in a model airplane. 7000-8000 rpm is more the norm. Something like a 14-10 or 15-10 APC prop will run in that range. I get about 9300 rpm on a YS140, 15-10 prop, and 20% nitro. Most other users feel the rpms are too high, the load too small... The 12000 number is likely published by the manufacturer as a peak horse power point. They all lie. You might find good information that is applicable to your use among the helicopter and race car modelers. They have similar cooling problems. IMHO, This is not a good fit for this engine! John Ferrell W8CCW "My Competition is not my enemy" http://DixieNC.US ----- Original Message ----- From: "Steve Smith" To: "'Microcontroller discussion list - Public.'" Sent: Tuesday, May 30, 2006 4:06 PM Subject: [OT] Model engine starting > Guys, > I know there are a few people that have some experience with glow engines > on > here so here is the dilemma . I have a Saito 100 four stroke engine that > is > being built into a model trike... Due to the arrangement of the power > train > it has a c/f clutch that's in line with the differential so I have no > access to the end of the engine for starting it. So I have modified a > traxas > ezstart so that it fits on the nose of the engine using a sprag (one way) > bearing type HSK8 this fits on the main shaft and the final gear is glued > on > to the outer shell. > > Now comes the problem... > > When the engine is dry the system works fine, When it is fuelled it > backfired and blew the guts out of the bearing (the sprags reversed) its > rating is 3.2nM and an overrunning speed difference of 12000 rpm with a > 8mm > core and 22mm outer diameter. > > How do you > 1. Stop a glow engine backfiring > 2. Calculate how much energy comes from a backfiring engine so I can > accurately size a starter bearing. > 3. Start the thing a different way (with an onboard starter) > > There is a pic for engine control here to start it adapted from a glitch > catcher > Initially wait for 0 throttle then at least 75% then 0 within 1 sec > then it will provide the following > Glow via 555 pwm for 1 sec prior to and during > 6 sec start via power fet to xx% throttle (f629) needs about 50A of > fet the starter takes nearly 30A at locked rotor using a 7.2v nicad stick. > > It will also double as a fail safe and glitch catcher but otherwise pulse > in > goes out to the throttle servo should only need a single channel (works on > paper) not yet coded. Might be useful for car and boat starting for others > I > have not seen it done before except on turbines. > > Lest have some help with this one guys I am not the best mechanic > > Steve > > > > > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist