The heart of a blacksmith shop
Editor's note: The following is Part 2 of a two-part article on the 15 HP Reid Type A that iron sculptor and blacksmith Joel Sanderson uses to run the line shaft in his blacksmith shop. Read Part 1.
This is a 15 HP Type A running on propane with hot tube ignition. For my shop to run smoothly, the engine driving it must also run smoothly. I can expect this engine to fire every revolution and to maintain a speed with no greater than 5 RPM variation.
Hot tube theory
One of the first things I had to figure out was hot tube ignition and its principles. I don't believe this ignition system is as simple as it first appears. There are many variables affecting a hot tube's performance, and many of those vary while the engine is in operation. The three primary factors affecting the hot tube are its length, temperature and the amount of compression into the hot tube, as generated by the main cylinder. However, the temperature of the engine's water jacket and the gas-to-air ratio also significantly affect the hot tube and the engine's timing.
Each firing in the cylinder leaves burned gases in the hot tube, which fail to exhaust out of the engine. The longer the hot tube, the easier it is to compress these gases in order to expose the new charge to enough area in the hot tube to induce combustion. In other words, the longer the hot tube, the less pressure (by compression) is necessary for ignition and the earlier in the cycle that ignition will occur.
The temperature of the hot tube also affects the timing. The hotter the tube, the easier the new gas mixture ignites, requiring less heat from compression to induce combustion. This means, as the temperature is increased, the ignition will occur earlier in the stroke.
Compression generates heat. This heat, when combined with the heat from the hot tube, affects when in the cycle the ignition will occur. This is important to remember, because as the governor opens and admits more air and gas, more heat will be generated by compression, causing the ignition to occur earlier in the stroke. Compression also determines how fast a fresh mixture is forced into the hot tube. More volume in the cylinder, when compressed, means the remaining spent charge in the hot tube (from the previous firing) is forced further into the tube. Just as with a longer hot tube, where more area induces earlier ignition, with greater compression, more area of the tube is exposed to the new charge.
Knowing how a Reid's timing is affected, it is understood that - due to the workload on the engine - each engine is running with no load and the mixer valves are nearly closed. Thus, they prevent the charging cylinder from drawing in a full breath of gas and air. This partial charge is then introduced to the main cylinder, which, not having been given a full charge, will not make full compression. This mixture is still of proper ratio to fire, and if a spark were introduced it would, but hot tube ignition requires sufficient compression of the gasses into the hot tube to ignite. In order for this to happen with this partial compression at no load, the timing of the hot tube would have to be advanced to the point where it would ignite the charge under low pressure. This is done by either lengthening the hot tube or heating it to a higher temperature.
As a load is put on the engine, the mixer slides open more, allowing a fuller charge into the engine. If the engine is still timed to fire at the previous reduced pressure, a premature ignition will result, and the impulse will occur before the piston reaches its optimum position, causing irregular running and back firing.
If the reverse were done and the engine were timed to fire steadily under load (by shortening or cooling the hot tube), the engine may not have enough compression to fire each revolution when the load is removed and the governor restricts the intake. This is not a problem. On a 2-cycle engine the exhaust leaves the engine due to the expansion of gas following combustion. Unlike with a 4-cycle engine, there is no mechanical discharge of the gas. This means the mixture from the missed cycle largely remains in the cylinder. The next cycle adds another charge to it, which then makes enough compression to fire. The unburned gases from the missed cycle are not wasted, and those nice big flywheels even out the speed so the alternate firing is hardly noticed.
My 15 HP engine, timed to run evenly under load, loafs along, firing every other cycle when the line shaft is disengaged. As the clutch is thrown on and the governor opens, the engine begins to fire every revolution. The engine speed drops just 3 RPM to spin 90 feet of the line shaft.
For this engine to fire steadily under the load I am requiring from it, the hot tube needs to have an internal length of 6-3/4 inches and be heated as evenly as possible to a fairly bright orange. If the temperature of the tube drops, the engine will begin to fire on alternate revolutions. Though the engine will maintain a constant speed while four-stroking, it puts undesirable strain on the belts and their dressing.
The temperature required to achieve combustion is not the same for all fuels. Natural gas will ignite when exposed to a temperature between 900 and 1,000 degrees Fahrenheit. Propane won't ignite until it reaches between 920 and 1,120 degrees. This may not sound like a great difference, but 120 degrees is the difference between a red heat and an orange heat.
Of course, people's opinions of color vary, but when I heat by hot tube to what I consider to be bright cherry red, as called for in the Reid manual, I do not get consistent ignition. I have found my hot tube needs to be bright orange in order to produce even impulses each revolution. Interesting enough, that color difference is just about the difference in degrees between the ignition point of natural gas and propane.
Since the hot tube is, quite simply, raising the temperature of the charge until it ignites, how warm that charge is initially must be considered. An engine expected to run well at 120 degrees will require a longer and/or hotter hot tube than one working at 170 degrees. This means that a thermostat is a necessity in order to maintain a constant temperature and proper timing.
My engine is plumbed with a 160-degree Failsafe thermostat. This holds the engine between 162 and 166 degrees, depending on the workload. Most automotive thermostats are intended to be mounted in or very close to the engine block; therefore, circulation through the thermostat is not necessary for measuring the engine's temperature. Some thermostats have a small hole drilled off to one side to allow a little water to sneak by even when it's closed. This is very important if the thermostat is to be installed in a pipe even a short way from the cylinder. Without this hole the engine will overheat and the thermostat won't even know it. The thermostat I am using did not have this, so I drilled a 3/32-inch hole in it. The pump is able to squirt plenty of water through this to keep the thermostat informed of the engine's progress.
My cooling system consists of a Model T radiator (with a fan) mounted over a 250-gallon tank. The water passes through the radiator and into the top of the tank. It is then drawn off the bottom of the tank and recirculated through the engine. It takes about five hours of running for the warm water entering above to begin to be drawn back to the engine, in which time it has cooled enough that even on a hot summer day the engine does not over heat. The 250-gallon tank holds enough water that in the event of a leak in the system I should have plenty of time to notice the problem before the engine would be harmed.
Gas to air ratio
Having the proper gas to air ratio is very important to a hot tube's ignition. Too lean a mixture or too rich a mixture will affect how easily the charge ignites. A mixture that is harder to ignite will have to be forced further along the stroke until it reaches combustion temperature. My Z with spark ignition is quite a bit more forgiving, since the ignition is only going to occur when the spark is introduced. With a hot tube, the ignition will occur later or earlier, depending on the mixture's proportions.
The proper mixture may be found fairly easily by first starting with the setting described in the manual and then carefully adjusting the air slide in or out until the engine runs the best. As the mixture gets too rich the engine may slow down; as it is made too lean it will begin to run irregularly and backfire. Somewhere between these two settings the engine will be found to pick up speed and smooth out. Once this ratio is found and the engine is shut down, a small file mark should be made on the mixer stems next to the bushings for future reference. Remember, consistent gas to air ratio is essential for consistent ignition with a hot tube.
It is said that Reid governors are not very effective as speed controls - they merely keep the engine from running away. As the Reid's mixer links are configured, full movement of the governor's flyballs moves the gas slide only about 1/8-inch while the air moves 1/2-inch, giving the air far greater movement than the gas. This changes the ratio depending on the governor's position. The manual says to set the air slide 1/8-inch ahead of the gas slide (the air being open 1/8-inch when the gas slide is closed).
This means the gas to air ratio changes even more dramatically just as the slides are about to close, making a much leaner mixture. All this mixture changing affects the timing of the engine, leading to very irregular running, snorting and sneezing. Besides, such little movement does not utilize the capacity of the ports in the mixer slides and does not come close to maintaining a steady speed.
The ports in the air slide are 3/4-inch wide; those in the gas slide are 3/8-inch wide. I reconfigured the links so that with full movement of the governor's flyballs the air and the gas both fully open and close. This means the gas to air ratio remains the same throughout the governor's movement (with the gas slide always moving half the distance of the air slide), and it has enough travel to respond appropriately to varying loads. This gives the engine far more available power and far more steady speed.
To steady this added movement, I made a simple hydraulic dash pot, or damper, which clamps to the fuel valve arm and prevents any sudden movement of the mixer links. This has been especially beneficial under very light loads when the governor is really too sensitive for the load being drawn from it.
With no load change, my engine varies almost imperceptibly and loses only 5 RPM when a load is thrown on.
I must point out that for this much governor travel it is necessary to operate under pretty low gas pressure, or the engine will flood. My Reid is operating on 7-1/3 inches of water.
How it runs
When I first start this engine on a cold day (the water jacket may read 30 degrees in the winter) it is necessary for the hot tube flame and the engine's gas to be set slightly higher than their running positions. Even then the impulses will be irregular, and the engine will fire every other or even every third revolution. As the temperature of the water jacket increases and the mixture's temperature becomes timed with the hot tube, the engine starts to fire every revolution. This usually begins to happen around 100 degrees. Then I can turn the gas to the engine and the hot tube flame down to their running positions. By the time the engine reaches its 160-degree working temperature, it fires every revolution, never missing a beat. As the load from the line shaft is increased, the Reid just breathes a little deeper, continuing to fire every revolution. However, when I disengage the engine from the shop, it immediately begins to fire every other revolution and speeds up to 168 RPM from 165 RPM, but it is still a constant speed.
This is a very impressive engine. Not only is it a delight to see run, but the power it produces is remarkable. If I add up all the horsepower required to run every machine in the shop, it well exceeds 20 HP. This 15 HP Reid can pull that without even blinking.
Contact Joel Sanderson at: 425 Maple Road, Quincy, MI 49082; (517) 639-4614 or visit his website.