Micro Steam Engine

I’ve got what Amazon sells as Sunnytech Mini Hot Live Steam Engine Model Education Toy DIY (BJ001)

Some very simple assembly is required.

non running micro steam engine

The micro steam engine appears to be machined out of some brass-like copper alloy. The quality of work is acceptable, there’s no obvious burrs or other wonkiness.

The base of the structure is an alcohol lamp fitted with a wick to allow alcohol vapor to burn with atmospheric oxygen. I burned Everclear, nearly pure ethanol you can purchase from a liquor store if you’re 21 or older.

The boiler has a volume of a little more than 0.55 in3, but you can only fill it up 2/3 or halfway full, otherwise liquid water gets forced into the cylinder. Any water drops that fall on the boiler cover do boil away, so the entire boiler is above 204° F. I’m doing this at 4800 feet altitude.

My big beef is the micro steam engine vibrates itself off the “pillars”. In the image above, I have some zip ties holding the “pillars” together, preventing vibrating the pillars out of the “tray” (item #15 in the schematic).

Schematic View

steam engine exploded schematic

The exploded schematic of the micro steam engine has a few oddities.

  • Multiple fonts
  • Item 8, “Subject”. hmmmm…
  • Item 13 is not a “bolt”, it’s a “crank pin”, a 19th or 20th century piece of mechanical assembly jargon
  • Item 15 “tray”. I’m baffled by this choice of word, “flange” or “ring” would fit better
  • Item 18, “Lamp holder”. No. A “wick” is not a “lamp”.

It’s a “wobbler” or oscillating cylinder steam engine. There’s no “valve gear” controlling steam inlet and outlet like on steam locomotives Instead, the cylinder “wobbles” back and forth around a trunnion pin. The pin is purple in the diagram below, an exploded schematic view of the cylinder block, flywheel, crank and other parts of the steam engine.

close-up image of “instructions” that shows the steam inlet and outlet

The steam from the boiler comes out of a hole in the port block (item #8 on schematic), which I’ve helpfully circled and labeled in blue as “inlet”. There must be a matching hole or slot on the face of the cylinder block on the size of the cylinder, pointed at by a green arrow. When the cylinder block wobbles to the other extreme, the outlet hole (circled and labeled “outlet” in red) matches up to the same hole or slot on the rectangular block, and the expanded, cooled steam gets squirted out via item #5, the exhaust pipe. The expanded, cooled steam does have some velocity leaving the exhaust pipe: it seems like it is actually forced out.

The little adjusting nut (item 3 in schematic) has a nylon locking feature. This engine really vibrates when it’s running.

Piston and Cylinder

close up of piston extracted from engine

The piston head is 0.235 inches in diameter, and 0.167 inches long.

close up of cylinder block extracted from engine

That’s the face of the cylinder block that oscillates against the port block. You can see the 0.078 inch diameter trunnion and the single cylinder port.

The cylinder port is 0.058 inches in diameter, measured with a largest diameter drill bit shank the hole accommodates.

There’s 0.19 inches on center between the trunnion and the cylinder port

I measure a bore of 0.235 inches, which is exactly what I measure the piston head’s diameter. My calipers must be off a little, because the piston is not a close fit in the cylinder, much less an interference fit.

I eyeballed some measurements with the aid of a little steel ruler.

  • stroke 0.375 inches
  • crank pin offset 0.19 inches, this is a less accurate measurement

The crank pin offset of 0.19 inches would lead to a stroke of 0.38 inches, a little off from what I measured.

I make the swept volume of the cylinder to be 0.02 in3.

If I measure the flywheel rotational velocity with a Neiko brand digital tachometer, I get 3000 - 3400 RPM.

Assuming a half-full boiler, 0.3 in3 of liquid water, and a factor of 1700 when that water turns into steam at Standard Temperature and Pressure:

Number of strokes on a half-full boiler = 0.3*1700/0.02 = 25,500 strokes.

At 3200 RPM, one stroke per revolution, it should run for 25,500/3200 = 8 minutes on a half boiler of water. The actual running time is around 2 minutes. There’s probably enough leaks, low-quality steam and inefficiency to account for this. I’ll also admit to just barely passing my required thermodynamics class summer of 1980, so maybe I didn’t calculate correctly.

Port Block

port block face, showing inlet, outlet ports and hole for trunnion

I make the inlet and outlet ports to be 0.046 inches diameter, based on the diameter of the shank of the largest drill bit that fits in the hole. I measure the inlet and outlet ports to be 0.199 inches on center from the trunnion hole, and 0.089 inches apart. From these I calculate that the cylinder port (which is a little bigger than the inlet or outlet ports) matches the inlet and outlet ports at 10° of cylinder angular excursion.

Cylinder Angular Excursion

Here’s the micro steam engine with the piston at top dead center:

engine at top dead center

Here’s the micro steam engine with the piston at bottom dead center:

engine at bottom dead center

Here’s the micro steam engine with the cylinder at maximum angular excursion (more or less):

engine at max angular excursion

What is the maximum angle away from the axle-trunnion line?

Crank, cylinder and piston geometry

idealized micro-steam engine geometry

In the above idealized geometry of the crank, cylinder and piston geometry, the crank arm is blue, and the combination of the cylinder block and piston is red. Note that the piston slides in the cylinder, the red line is of variable length.

  • θ - crank angle, 0 is top dead center, π radians (180°) is bottom dead center.
  • α - cylinder angle, 0 at top dead center and bottom dead center, positive on power stroke, negative on exhaust stroke
  • h - distance from flywheel axle center to trunnion center
  • c - radius of the circle described by the crank pin’s motion, the crank arm length
  • x - distance from crank pin center to trunnion center. Includes piston rod and cylinder block.

x = √(c2 sin2 θ + (h - c cos θ)2)

α = cos-1((h - c cos θ)/x)

The adjusting nut (#4 on the schematic) and spring (#3) are on the end of the trunnion pin. The axle and the trunnion pin are 0.505 inches apart on center.

That’s h = 0.505, c = 0.1875 for the idealized geometry diagram above.

crank angle to wobble angle and stroke

Above, the relationship between crank angle, cylinder angular excursion, and stroke of the micro steam engine. The purple line is the distance between the crank pin’s center, and the trunnion pin’s center, roughly the piston’s stroke plus a constant. The piston’s top dead center is at 0° crank angle, bottom dead center is at 180° crank angle. The maximum wobble angle is 0.354 radians, or 20°, at crank angles of ±1.230 radians, ±70.5°

To get the piston to cycle, you have to admit steam into the cylinder somewhere between top dead center and maybe a quarter stroke, so that you can get boiler pressure and/or steam expansion to push the piston.

Ideally, you’d open an exit value just after the piston’s bottom dead center, so the flywheel doesn’t have to re-compress any steam on the exhaust stroke. The exhaust steam squirts out of the exhaust pipe rapidly, some re-compression gets done.

Steam admission and exhaust

Referring to the port block picture above, the inlet and outlet ports are 0.046 inches in diameter, and 0.089 inches apart on center. That leaves 0.043 inches of solid port block between inlet and outlet port. The cylinder port, on the other hand, is 0.058 inches in diameter. It would seem that at top and bottom dead center, there’s a direct path from boiler to the atmosphere from the inlet port, through the cylinder, and out the outlet port. I suspect my measurements aren’t that accurate. I also measure that the inlet and outlet ports are a bit closer to the trunnion center than the cylinder port is, the ports don’t line up to vent steam out of the boiler at top and bottom dead center.

The measurements mean that as soon as the cylinder rotates past top dead center, the cylinder port begins to line up with the inlet port, admitting steam to the cylinder. As soon as the cylinder rotates off bottom dead center, the cylinder port begins to line up with the outlet port, admitting steam to the cylinder.

If the 0.058 inch diameter cylinder port, and the 0.046 inch diameter inlet port are beginning to overlap, the distance between port centers is 0.052 inches. I’ll assume both ports are on a 0.190 inch radius from the trunnion. The two port enters would subtend a 15.7° arc from the trunnion center. The inlet and outlet ports are 10° off the trunnion-axle line, and the cylinder has a maximum excursion of 20°, The cylinder would have to rotate 10+15.7 = 25.7° to cause the cylinder port to swing through and past the inlet and outlet ports. The cylinder port begins to overlap the inlet port as the cylinder rotates past top dead center, and begins to overlap the outlet port just as the cylinder rotates past bottom dead center. I think this means the engine runs entirely from boiler pressure. The steam does little or no expansion in the cylinder, because the cylinder is open to the boiler almost all the way through the piston’s power stroke.

The purpose of the flywheel is to rotate the cylinder a little past bottom dead center, so the steam starts to leave the cylinder, then carry the piston to just past top dead center, where more steam can be admitted. I can’t quite fathom why the flywheel has a hole. The crank has a counterbalance, which isn’t 180° opposite the flywheel hole.

Shape of angular excursion versus crank angle curve

Mechanisms are weird: the cylinder angle isn’t a sinusoidal wave.

wobble angle compared to sine wave

I’m not sure the slightly deformed sine wave is an advantage. It looks to me like it’s harder to line up the single inlet/outlet on the cylinder to the inlet and outlet holes on the port block (item #8 in schematic) and get some work out of the piston.

Sound Spectrum

I had an Android sound spectrum app, Spectroid, listen to the running micro steam engine.

sound spectrum of running micro-steam engine

I’ve labeled some peaks

  • A 50 Hz
  • B 100 Hz
  • C about 120 Hz
  • D 200 Hz
  • And of course the highest peak, 249 Hz

If the flywheel spins at 3000 revolutions per minute, that’s 3000/60 = 50 revolutions per second. 50 times per second is once per revolution, maybe a steam puff as the cylinder port swings into the outlet port. The 100 Hz is twice a revolution, maybe the flywheel or crank bumping, perhaps the piston hitting bottom dead center and top dead center, or the cylinder rotation changing direction. The 200 Hz is from something that happens 4 times a revolution, perhaps a combination of cylinder rotation direction changing and piston changing direction, which don’t happen at 90° of crank rotation apart, but close.

I do not understand the 249 Hz noise. That’s 5 times a crank rotation at 3000 RPM flywheel speed. It’s also about twice as fast as the (roughly) 120 Hz noise, peak C above.

Operational notes

  • It’s a mild fire hazard
  • It really vibrates. Be prepared to have the boiler come out of the “tray”.
  • You don’t need a whole lot of wick showing, maybe a quarter of an inch.
  • Only fill the boiler about half way
  • You can hear the water come to a boil, and it may spit a little hot water out of the cylinder-to-port block interface or the exhaust pipe before it’s making enough steam to run the cylinder/crank/flywheel
  • Set the cylinder at piston top dead center, when you hear the water boil, give the flywheel a spin
  • if it’s running too fast, blow out the flame, let it cool and loosen the nut a little. The spring keeps the cylinder block and port block clamped together. If the spring is less compressed, some steam escapes between them rather than getting admitted into the cylinder
  • The engine will run until the boiler is dry, and then stop abruptly
  • The entire boiler/port block/cylinder/flywheel assembly gets very hot during operation, and takes a while to cool after you put out the alcohol lamp