Watt beam engine

A Watt engine is a cylinder that works on steam and a cold vessel set apart from it, joined by a beam that rocks over a stone wall or an iron column. How to build and run one in the game is in the steam engines guide, and its parts are in the catalogue. This page is about the real one and how the game's follows it.

The Smethwick Engine at Thinktank, Birmingham, described on its file as the oldest engine still in working order. The timber beam and its chains are overhead; the cylinder, wrapped and bolted to its lid, stands below on the engine floor.
The Smethwick Engine at Thinktank, Birmingham, described on its file as the oldest engine still in working order. The timber beam and its chains are overhead; the cylinder, wrapped and bolted to its lid, stands below on the engine floor.
Farey's section of a Watt single engine for draining mines, 1788 (Farey Plate X, described from p. 353). The boiler C sits over its grate B and fire A in an outbuilding of its own. Steam comes along the pipe a to the cylinder E. Below E the condenser F stands in a cistern. The great lever L, "a single beam of oak" (Farey p. 355), rocks on the wall V, and chains K on its arch heads pull the piston rod n. The left-hand chains carry the rods of the pumps in the mine shaft.
Farey's section of a Watt single engine for draining mines, 1788 (Farey Plate X, described from p. 353). The boiler C sits over its grate B and fire A in an outbuilding of its own. Steam comes along the pipe a to the cylinder E. Below E the condenser F stands in a cistern. The great lever L, "a single beam of oak" (Farey p. 355), rocks on the wall V, and chains K on its arch heads pull the piston rod n. The left-hand chains carry the rods of the pumps in the mine shaft.
The game's Watt engine in its planned drawing: the oak beam on its timber gallows, the cylinder at one end and the shaft end at the other, with the condenser and air pump low down by the cylinder., iso
The game's Watt engine in its planned drawing: the oak beam on its timber gallows, the cylinder at one end and the shaft end at the other, with the condenser and air pump low down by the cylinder.

The problem it solved

Newcomen's engine made its vacuum by spraying cold water into the working cylinder. The vacuum was good and the waste was enormous, because the same iron then had to be heated again to take the next stroke of steam. Watt named the loss himself: the fuel it took to bring "the whole cylinder, piston, and adjacent parts from the coldness of water to the heat of steam, no fewer than from 15 to 20 times in a minute" (Thurston History p. 88).

His answer was to make the two jobs happen in two places. The patent of 1769 says that the cylinder "must, during the whole time that the engine is at work, be kept as hot as the steam", and that the steam "is to be condensed in vessels distinct from the steam-vessel or cylinder" (Thurston History pp. 99-100). Everything else on the machine follows from that split.

Boulton and Watt's double-acting engine of 1784, the form the game's engine belongs to (Thurston History p. 119, Fig. 31), with its parts named over the plate. Steam from the boiler comes through the throttle valve to the cylinder. When the piston has gone its stroke the exhaust goes down the pipe to the condenser, where a jet of cold water from the injection handle turns it back to water and leaves a vacuum. The air pump empties the condenser and its rod also carries the pins that throw the valves. The beam passes the work across to the connecting rod and the flywheel. This engraving shows a plain crank.
Boulton and Watt's double-acting engine of 1784, the form the game's engine belongs to (Thurston History p. 119, Fig. 31), with its parts named over the plate. Steam from the boiler comes through the throttle valve to the cylinder. When the piston has gone its stroke the exhaust goes down the pipe to the condenser, where a jet of cold water from the injection handle turns it back to water and leaves a vacuum. The air pump empties the condenser and its rod also carries the pins that throw the valves. The beam passes the work across to the connecting rod and the flywheel. This engraving shows a plain crank.

A separate cold vessel brings its own needs. Water and air gather in it and spoil the vacuum, so it needs a pump to clear them: the air pump, "wrought by the engines themselves" as the patent has it (Thurston History pp. 90, 100). Once the cylinder must stay hot, cold air can no longer follow the piston down, so Watt closed the top with a cover and let the rod slide through a stuffing box, wrapped the cylinder in a steam jacket, and greased the piston with oil and tallow instead of a layer of water (Thurston History p. 90).

The first engines were single-acting. Steam pressed the piston down, an equilibrium valve then let the steam pass under the piston, and the weight of the pump rods at the far end of the beam drew it back up (Thurston History p. 99, Farey p. 354). For draining a mine that is all that is needed, and a chain is good enough, because a chain only ever has to pull.

Turning a shaft

Mills wanted rotation. Watt knew the crank from the common lathe, and made a model of two cylinders on one axis, but did not patent it. By Watt's own account a workman passed the idea to people around a rival's engine, and James Pickard took out a patent on the crank in 1780 (Farey pp. 409, 423-424). Without troubling to dispute it, Watt "contrived other means of effecting the same object" (Farey p. 424), and the one his firm built was the sun and planet.

Watt's governor on the Boulton and Watt engine of 1788 in the Science Museum, London. The spindle is driven from the engine; as the balls swing out they lift the collar, and a lever carries the movement to the throttle valve. The same regulator was already used on water-wheels and windmills, and Watt's step was applying it to steam (Thurston History p. 116).
Watt's governor on the Boulton and Watt engine of 1788 in the Science Museum, London. The spindle is driven from the engine; as the balls swing out they lift the collar, and a lever carries the movement to the throttle valve. The same regulator was already used on water-wheels and windmills, and Watt's step was applying it to steam (Thurston History p. 116).
The sun and planet on a Boulton and Watt engine of 1788. The planet wheel is fixed to the foot of the connecting rod and cannot turn on itself, so as the rod goes up and down it carries the sun wheel round. It turns the flywheel "twice, during each complete stroke of the engine" (Farey p. 450), so the shaft runs at twice the strokes.
The sun and planet on a Boulton and Watt engine of 1788. The planet wheel is fixed to the foot of the connecting rod and cannot turn on itself, so as the rod goes up and down it carries the sun wheel round. It turns the flywheel "twice, during each complete stroke of the engine" (Farey p. 450), so the shaft runs at twice the strokes.

To turn a shaft evenly the engine had to push on both strokes, so steam was let in above and below the piston by turns, four valves now doing the work in pairs (Thurston History pp. 118-119, Farey p. 451). A chain cannot push. Thurston calls the sectors and chains or racks used until then a device whose "rudeness of design" was "only equalled by its insecurity" (Thurston History p. 112).

The complete parallel motion, patented in 1784, from Farey's woodcut (Farey p. 431), with its joints named. The end of the beam swings on an arc, and the top of a straight rod cannot follow an arc. Link B-E hangs from the beam and carries the rod at E. The counteracting lever C-D, swinging about a fixed point, pulls the lower joint the opposite way, and the two curves cancel, leaving E on "a right line" (Farey p. 432). Farey says it "has since been universally adopted" (Farey p. 431).
The complete parallel motion, patented in 1784, from Farey's woodcut (Farey p. 431), with its joints named. The end of the beam swings on an arc, and the top of a straight rod cannot follow an arc. Link B-E hangs from the beam and carries the rod at E. The counteracting lever C-D, swinging about a fixed point, pulls the lower joint the opposite way, and the two curves cancel, leaving E on "a right line" (Farey p. 432). Farey says it "has since been universally adopted" (Farey p. 431).

A linkage works through pins alone, pushing and pulling. Of the inventions in the specification of 1784, Farey names only the parallel motion and the locomotive engine as having come into extensive use (Farey p. 434).

With power going to machinery the speed mattered, and a load could change at any moment. The throttle valve, a disc in the steam pipe turned edge-on or across, set the steam by hand or by the governor (Thurston History pp. 114-115, Farey p. 458). Fig. 31 shows the governor's rod going to it at T.

How it was kept and counted

Starting was a procedure, not a switch. The fire was raised, air was blown out of the cylinder and condenser with steam through a blow valve, and the engine waited until the injection and a barometer showed a vacuum forming; only then, with the flywheel turned by hand about an eighth of a turn, was the injection opened and the valves set going (Farey pp. 459-460). The condition of the condenser was read on a barometer gauge, a glass tube in mercury joined to the condenser, near 30 inches at perfect vacuum (Thurston History p. 117).

Boulton and Watt were paid in part out of what their engines saved, usually a third of the value of the fuel, which meant an honest tally of strokes was a business matter (Thurston History p. 114). Watt's answer was a counter on the beam, a train of dials moved by a pendulum as the beam rocked, locked and opened once a year (Thurston History p. 114).

Very little pressure was used. An engine well loaded had a mean exhaustion of about 9 lb a square inch below the atmosphere during the stroke, and the boiler stood only a few pounds above it (Farey pp. 486-487). Most of the work was done by the atmosphere on one side of the piston and the condenser on the other. Where a later engine has a thicker boiler, this one has a cold vessel. The ratings page follows what that meant for horse-power and coal.

Period basis, and what is the game's own

  • Period: the separate condenser and air pump, and a jet of cold water to make the vacuum (Thurston History pp. 88-90). The game's planned drawing has them, and the injection cock is planned as a handle with five marks, from shut to fully open.
  • Period: the barometer gauge and a stroke counter on the beam (Thurston History pp. 114, 117; Farey pp. 377-378, 520-521). Both are planned on the game's engine.
  • Period: the parallel motion and sun and planet (Thurston History p. 105, Farey pp. 431, 450). The game's planned drawing carries the parallel motion, and the generator is planetary, the shaft turning twice for each double stroke.
  • The game's own: there is no throttle or governor. The steam is set at the pressure valve in the line, and a shut injection is the planned way to stop the work.
  • The game's own: the engine drives one attached sub-machine, a generator, a feed pump or a blowing engine, where a real engine drove whatever was geared or rodded to it.
  • The game's own: today it runs from 2 to 4 atm through a plain on-off gate, a long way above Watt's pressures, and draws the same steam throughout. The planned rating removes the gate and takes the pressure down to Watt's.
  • The game's own: pairing this engine with a Cornish boiler is a compression of the period. Boilers of Watt's time were wagon boilers (Fairbairn Useful p. 299); see the Cornish boiler page.
  • The game's own: wear and a burst cylinder from over-pressure, after 60 s over the band, is the game's rule, with a wrench and parts to mend it.

The numbers

Engines keep period figures rather than the fuel factor of the iron furnaces (see the scaling page), at the drawn size (scale and units). "Planned" is the rating in design; the last rows are the game today. The ratings page sets it against the other engines.

In the gamePeriod figureHow it was scaledGame value
Bore30 in, a makers' "30 horse" (Farey p. 575)The cylinder as drawnPlanned 750 mm (29.5 in)
StrokeThe stroke of a 10 horse (Farey p. 575)The drawingPlanned 1.26 m (4.14 ft)
Speed17.5 double strokes a minute (Fairbairn Mills II p. 101)Period figure; the sun and planet doubles it at the shaft (Farey p. 450)Planned 17.5 double strokes, 35 rpm
Nominal powerd^2 v / 6050, at 6.944 lb a sq in (Farey p. 575)The rule on the drawn sizePlanned 20.9 hp
Mean pressure, full working12.7 lb a sq in, the vacuum included (Farey p. 486)The rule on the drawn sizePlanned 87.6 kPa, 28.5 kW indicated
Friction2 lb a sq in (Farey p. 486)Period figurePlanned 13.8 kPa, the engine's standing torque
Useful powerHalf as much more than nominal (Farey pp. 486, 576)10.7 lb a sq in over the area and speedPlanned 24.0 kW, 32.2 hp
Working pressure2 lb in the cylinder, 4 to 4 1/2 lb at the boiler (Farey pp. 486-487, 576)Period figurePlanned 2 psi; nothing more above it
Break pressureAbove 12 lb, the wagon boiler (Fairbairn Useful p. 299)Period figurePlanned 0.82 atm
Steam33 cu ft a minute a nominal hp (Farey p. 439)Volume per stroke at 2 lb gaugePlanned 0.214 kg/s, 772 kg/h
Injection water22 to 23 lb a lb of steam (Rankine pp. 391-392); 0.6 cu ft a minute a nominal hp (Farey pp. 593-594)Period figuresPlanned 4.8 to 6.5 kg/s
Runs atNone; the gate is the game's ownGame valueToday 2.0 atm and above
Breaks aboveNone; the game's ownGame valueToday 4.0 atm, after 60 s
Steam and powerNone; the game's ownGame valuesToday 30 L/s for 0.3 of power

Picture credits