HOW A MOPED CAN RUN ON PROPANE GAS

This is a report covering the conversion of an ordinary gasoline-powered moped to one that runs on propane from an on-board pressure vessel. The system at present is very simple, and has relatively few parts. The difficulty in this project was not in getting the propane to the engine itself, but rather in actually running and controlling the rpm of the engine.

Beginning in the combustion chamber, compression ratios are typically higher by design. It is really the compression pressure which is the more important measurement in relationship to the RPM range in which peak pressure is available.

The most important aspect of the ignition system is the higher voltage potentials required to avoid oxide buildup on plug electrodes. Typically, at least 28,000 volts should be delivered to the plugs to achieve continuous good spark quality. As in 4-stroke engine powered automobiles, knocking can sometimes be heard. This is a symptom of premature detonation (see diagram below) - a condition in which the flame front is compressed and the remainder of the unburned fuel charge spontaneously combusts. This is extremely detrimental to engine components.

Also the timing must be altered slightly. Total spark lead (advance before TDC) may be up to 30 degrees less than a factory specification for gasoline operation due to the faster combustion of propane gas.

Propane's gaseous nature means that oil deposited on the cylinder wall is not washed away with the liquid fuel mixture as in a gaseous engine. Nor do the rings provide the same sealing capability that they do with liquid fuels. Consequently, excessive oil gets into the compression chamber. Two types of rings can solve this problem, either a molybdenum-filled or cast iron top ring, which seat more readily in worn cylinders.
The experimental engine used was a one-cylinder, two-cycle, 50cc engine (widely used on mopeds and small scooters) running on ordinary unleaded gasoline and two-cycle oil mixture, provided by an oil-injection pump. The idle speed of the engine is 1800 RPM (plus or minus 200). The bore is 41.0 mm and the stroke is 37.4 mm. The original compression ratio is 6.8:1, lower than the characteristic 9:1 compression of a typical propane engine.

 The transmission uses a V-belt for primary reduction and the moped can go (in perfect running condition) up to 30 MPH.

 The carburetor is a standard float-type, with a choke mechanism for cold starting. The choke moves a needle pin and plunger in the carburetor casing, which significantly reduces the air in the mixture.

Figure 1: the engine complete with the propane cylinder

The ignition system is a capacitive discharge type, using points and a condenser to control spark timing.

 The first modification I made was to remove the carburetor completely. I did this because the carburetor could provide no mixture control of the propane-to-air mixture without significant modification.

 Figure 2: the new fuel delivery system

 Propane is stored in a standard cylinder, the type used for many portable camp stoves. Maximum flow is controlled by a needle valve in the cylinder connector, which the throttle cable is linked to. This connector was adapted from a propane torch head. The propane is then carried through the steel-braided tubing to a T-connector. The other end of the T-connector is screwed into a manifold plate which I machined from cold-rolled steel in the machine shop. The right angle tap has tubing going to the air filter.

 With this setup, there is no direct control of the propane-to-air mixture. There may be some control of the mixture by the size of the propane and air tubing used. Specifically, the 1 / 2_" tubing used to carry the propane may be allowing an excessive flow rate of propane compared to the flow rate of air. When the engine was first run on propane, there was no tubing or ducting of any kind. I simply held the propane tank connector near the manifold (with the plate unattached), opened the valve, and cranked the engine. It started and the engine speed followed the changes I made using the needle valve.

 In order to start the thing I used a variable speed, reversible drill attached to the flywheel nut. This required the removal of the plastic fairings and the air fan (a plastic piece with small blades bolted onto the flywheel). The no-load speed of the drill was 2500 RPM, and this provided the required torque and higher cranking speeds to start a cold propane engine.

Once started, the propane engine seemed to run smoothly and with regular power. Small variations in the needle valve led to dramatic changes in engine speed. However, most of the time any change in the needle valve would cause the engine to stall out. The engine did not sputter, or even sound like it was trying to stay running; it just sped up slightly and then behaved as though it was no longer receiving any spark. I know that this was not the case, and that the spark plug was providing good spark.
 

I believe that what happened was that as the flow of propane increased, so did the pressure in the T-Valve, and consequently either less air or less propane was taken into the mixture. A possible cause of the problem is the orientation of the air intake in the T-valve. As indicated in the diagram below, the air is coming in at right angles to the propane. There will be a sufficient pressure differential (lower pressure) between points 1 and 2 as long the velocity of the propane is greater than the velocity of the incoming air. Once the air velocity reaches a certain point, the differential is not as great, and therefore air intake is slightly reduced. This leads to a cascading type of effect, because once the engine slows down, the propane flow remains constant while the air intake is lowered, and the the mixture ratio is altered.

Figure 3: mixture principle

There are two other potential problems with the current design, causing the engine to run erratically:

• Presence of air and a sensing agent in the gas. Commercially available propane canisters have a certain amount of air dissolved in the cylinder. Also in the cylinder is a sensing agent, also used in natural gas. Since both natural gas and propane are odorless, the sensing agent provides for detection of the gas through the sense of smell. This compound has a lower boiling point than propane, and remains dissolved in propane down to very low ambient temperatures.
• A dead or weak battery left connected to the electrical system may have been draining current away from coil and spark plug, thereby decreasing spark size. Good and strong spark as mentioned earlier is very important for proper ignition of propane.

   I also found that with the cooling fan removed to allow access to the flywheel nut, the engine heats up very quickly, particularly since most of the testing is done with the moped parked on its stand.

  The present engine had oil injection; a must it seems as propane gas cannot carry the oil necessary to lubricate the cylinder wall. It should be rather easy to come up with a design that will use fuel injectors and an electronic circuit to control mixture and throttle speed.

  Who says that there is no life left in 2-stroke moped engines?