Steam hammer

A steam hammer is a heavy iron block hung on the piston rod of an upside-down steam cylinder, with a man at a valve deciding how far it rises and how hard it comes down. Its parts in the game are in the catalogue. This page is about the real one, what a heavier hammer adds, and how the game's follows it. The game's hammer is planned: what stands in the game today is the placed machine, with no work behind it yet.

A Nasmyth steam hammer of about 1850 at the Science Museum, London. Two cast standards carry the cylinder on top; the piston rod comes down through the lower cover to the block, which runs in guides between the standards; the block's replaceable face is dovetailed in over the anvil. Everything above the anvil exists to lift that one block and let it fall true.
A Nasmyth steam hammer of about 1850 at the Science Museum, London. Two cast standards carry the cylinder on top; the piston rod comes down through the lower cover to the block, which runs in guides between the standards; the block's replaceable face is dovetailed in over the anvil. Everything above the anvil exists to lift that one block and let it fall true.
Nasmyth's own picture of a hammer at work, published in his autobiography of 1883. A gang on the porter bar swings a mass out of the furnace, others tend the crane chains, with a great shaft on the anvil and heavy forgings on the floor. Work of this size is what the machine was invented for.
Nasmyth's own picture of a hammer at work, published in his autobiography of 1883. A gang on the porter bar swings a mass out of the furnace, others tend the crane chains, with a great shaft on the anvil and heavy forgings on the floor. Work of this size is what the machine was invented for.
The game's hammer in its planned drawing: 15 cwt double-acting, after the Atlas Works hammer that Percy drew, with two standards, the cylinder above, the block in its guides and the anvil standing clear in its own pit., iso
The game's hammer in its planned drawing: 15 cwt double-acting, after the Atlas Works hammer that Percy drew, with two standards, the cylinder above, the block in its guides and the anvil standing clear in its own pit.

The problem it was built for

Until then the heavy hammers of a forge were helves and tilts: a beam on a pivot, raised by cams and dropped (see the belly helve). In November 1839 the engineer of the Great Britain wrote to James Nasmyth that there was not a forge hammer in England or Scotland powerful enough to forge the 30 in paddle shaft of the ship's engines, and asked whether he dared use cast iron (Nasmyth p. 239).

The helve and the steam hammer over the same thick piece. A helve's head rises a fixed height on an arc, so a thick forging takes up the fall the head needs and the blow comes to nothing: the hammer was "gagged" (Nasmyth pp. 239-240). A block lifted straight up by steam clears whatever is on the anvil. A helve's fall is "invariable"; the steam hammer's is set by hand (Bauerman p. 292).
The helve and the steam hammer over the same thick piece. A helve's head rises a fixed height on an arc, so a thick forging takes up the fall the head needs and the blow comes to nothing: the hammer was "gagged" (Nasmyth pp. 239-240). A block lifted straight up by steam clears whatever is on the anvil. A helve's fall is "invariable"; the steam hammer's is set by hand (Bauerman p. 292).

Nasmyth put the failure down to "want of compass, of range and fall", and to the clear space for the fall being almost entirely taken up by the work: when the forging most needed a heavy blow it got next to none (Nasmyth pp. 239-240). A lever can only be made to rise so far. The answer was to give up the lever.

The first drawing, dated 24 November 1839, as Nasmyth reproduced it (Nasmyth p. 241). Left, the machine from the side with a man at the valve lever and a round shaft on the anvil; right, the same from the front, with "5 ton anvil block" written in the base. Everything is already here: an inverted cylinder, a block on the end of the rod, standards astride the anvil. In the corner, below a tilt hammer, the sketch of "the old Helve Hammer".
The first drawing, dated 24 November 1839, as Nasmyth reproduced it (Nasmyth p. 241). Left, the machine from the side with a man at the valve lever and a round shaft on the anvil; right, the same from the front, with "5 ton anvil block" written in the base. Everything is already here: an inverted cylinder, a block on the end of the rod, standards astride the anvil. In the corner, below a tilt hammer, the sketch of "the old Helve Hammer".

He wrote that the drawing already held all the essential elements of the invention (Nasmyth p. 242). The ship went over to the screw and the shaft was never forged, and no firm would then buy a hammer in a depressed trade (Nasmyth p. 243). The first one built was at Le Creusot, copied from his sketch while he was away; on a visit in 1842 he was shown a crank forged under it, the eye punched and drifted smooth, and told it was his own hammer that had made it (Nasmyth p. 246). He took out a patent that year and built one of 30 cwt with a clear fall of four feet (Nasmyth pp. 247-248).

Steam had been at a hammer before Nasmyth. James Watt's of 1784 (Thurston History p. 123, Fig. 34): the cylinder works a beam, and the beam by a rod works a hammer-helve. Watt had proposed one for Wilkinson's forge as early as 1777 (Thurston History p. 111). Steam replaced the water wheel here, and the hammer stayed a lever. Nasmyth's change was to put the block on the rod.
Steam had been at a hammer before Nasmyth. James Watt's of 1784 (Thurston History p. 123, Fig. 34): the cylinder works a beam, and the beam by a rod works a hammer-helve. Watt had proposed one for Wilkinson's forge as early as 1777 (Thurston History p. 111). Steam replaced the water wheel here, and the hammer stayed a lever. Nasmyth's change was to put the block on the rod.

What the valve gave the man at it

The machine is, as Bauerman puts it, an inverted vertical high-pressure engine standing on two cast-iron standards, the piston rod joined directly to the block (Bauerman p. 291). In the first form steam only lifted, and the blow was the fall alone. In 1843 Nasmyth let steam in above the piston as well, "to aid gravitation" (Nasmyth p. 249), and this double-acting or "top steam" form became almost universal (Percy p. 697).

An 8-ton double-acting hammer by Thwaites and Carbutt of Bradford (Bauerman p. 293, Fig. 29), labelled from his text. The man on platform f holds two handles: lever c works the slide valve and handle d admits steam. Tappet a on the rising block strikes bent lever b, which moves the valve and opens the exhaust, so the top of the stroke is set by the machine and each blow by the man. Bauerman's text gives the exhaust as e; on the plate the arrow is lettered g.
An 8-ton double-acting hammer by Thwaites and Carbutt of Bradford (Bauerman p. 293, Fig. 29), labelled from his text. The man on platform f holds two handles: lever c works the slide valve and handle d admits steam. Tappet a on the rising block strikes bent lever b, which moves the valve and opens the exhaust, so the top of the stroke is set by the machine and each blow by the man. Bauerman's text gives the exhaust as e; on the plate the arrow is lettered g.

Because the valve can check the exhaust or hold steam under the piston, the block can be stopped anywhere in its stroke (Bauerman p. 292). Nasmyth's boast was that it could crack an egg in a wine glass and that the next blow "would shake the parish" (Nasmyth p. 249), and that its attendant could "think in blows" (Nasmyth pp. 242, 249). The early self-acting gear was dropped again, in favour of the hand (Nasmyth p. 249).

It mattered at the forge. A ball that would crumble under a helve could be coaxed into a sound bloom with gentle blows (Percy p. 695). Short light strokes closed it first, and the fall lengthened as it hardened (Bauerman p. 292). Whole masses could be welded and punched, so that an anchor was made in one piece where the "bit by bit" system had left flaws (Nasmyth p. 250). And with that control the die work of the Birmingham gun trade, stamping hot iron into a shaped die, grew to the largest class of forge work (Nasmyth pp. 250-251).

The hammer did not win everywhere. A helve costing a fifth or sixth as much did as well on ordinary bar iron, where every bloom wants the same blow (Truran p. 231), and the small steam hammers of 1850 were too slow for drawing bar, at 80 to 100 strokes a minute where 150 and up was wanted (Overman p. 337).

What holds it up, and what more weight adds

A hammer strikes its anvil as hard as it strikes the work, so the anvil stands apart from the frame.

The hammer the game draws: Nasmyth's 15 cwt double-acting at the Atlas Works, Sheffield (Percy p. 696, Fig. 176). The standards bolt to a sole-plate on the surface; the anvil passes through it with 1/16 in clear and stands on sixteen upright baulks of timber, in a pit rammed hard down to the subsoil. 5 1/2 tons of anvil to a block of three-quarters of a ton (Percy p. 698). The anvil stands clear of the ground that carries the frame (Bauerman p. 294).
The hammer the game draws: Nasmyth's 15 cwt double-acting at the Atlas Works, Sheffield (Percy p. 696, Fig. 176). The standards bolt to a sole-plate on the surface; the anvil passes through it with 1/16 in clear and stands on sixteen upright baulks of timber, in a pit rammed hard down to the subsoil. 5 1/2 tons of anvil to a block of three-quarters of a ton (Percy p. 698). The anvil stands clear of the ground that carries the frame (Bauerman p. 294).

Sizes followed the work. Below 12 or 15 cwt, for heavy smithing, one standard could overhang the base and leave three sides of the anvil free (Bauerman p. 294). Blooms in a puddling forge took 30 to 60 cwt, and one of 50 cwt served twelve furnaces (Bauerman p. 292). Armour piles, marine cranks and steelworks took 5 to 50 tons, usually single-acting above 12 or 15 tons (Bauerman p. 292). Weight brought fall and foundations with it: Krupp's 50-ton hammer lifted ten feet and stood over a 185-ton anvil on a bed of cast iron blocks (Bauerman p. 295).

The heavy end could dispense with the anvil altogether. Ramsbottom's duplex hammer (Bauerman p. 295, Fig. 30): two blocks a and a' on a short level railway, drawn together by the piston b below through rods c, strike an upright ingot on carrier f from both sides at once. In the newest form each block weighed 30 tons (Bauerman p. 296).
The heavy end could dispense with the anvil altogether. Ramsbottom's duplex hammer (Bauerman p. 295, Fig. 30): two blocks a and a' on a short level railway, drawn together by the piston b below through rods c, strike an upright ingot on carrier f from both sides at once. In the newest form each block weighed 30 tons (Bauerman p. 296).

The game's two hammers sit at two points on this ladder. The planned iron-age hammer is the 15 cwt smithing size. The steel-age hammer of 3.78 t, also planned, is a forging hammer for cogging ingots, which were hammered smooth and swaged down for the first groove of the mill (Bauerman p. 378). Heavier work is what the extra tonnage buys.

Period basis, and what is the game's own

  • Period: an inverted cylinder on two standards, block on the rod, steam on both sides of the piston, the anvil separate on upright timber (Percy p. 698, Bauerman pp. 291-294). The game plans exactly this machine, at true size.
  • Period: the man at the valve set each blow (Bauerman p. 292). In the game the force is set a notch at a time, and pushes a blow through one to three layers of the work. The game's own reduction of a continuous lift to three settings.
  • Period: the striking faces were removable and changed for the job (Bauerman p. 296). The game's set, plain face, flatter and hot set choose how wide the blow is, 1 x 1, 3 x 3 and 5 x 5 voxels or a cut line.
  • Period: stamping hot iron in dies on the anvil (Nasmyth pp. 250-251). The game plans dies that stamp iron tool heads in batches, the flash coming back as bits.
  • The game's own: the base game has no power hammer but its helve, which works only certain items and picks the voxels itself. See vanilla against reality. The steam hammer works any smithing recipe at the voxel the player picks, with the player holding the work in tongs.
  • The game's own: work must be hot, and cold work rings and does not move, as at the hand anvil.
  • The game's own: a ball's soundness is not modelled, so the coaxing of a crumbling ball has no counterpart. The hammer is not a shingling machine in the game: the belly helve does that job, and the shingling page says how.
  • The game's own: steam comes from a Cornish boiler whose feed needs an engine's pump, so the hammer stands in a working plant.

The numbers

Machine sizes follow the period hammer rounded to whole voxels (see scale and units and the scaling page). All are planned; no steam hammer process runs today.

In the gamePeriod figureHow it was scaledGame value
Falling mass15 cwt (Percy p. 698)Period figurePlanned 762 kg
Stroke40 in, 1.016 m (Percy p. 698)Whole voxels at true sizePlanned 1.0 m
BoreNot given for this hammer; 28 in on the 8-ton (Bauerman p. 292)The game's own: scaled by the square root of the mass to 8.6 in, then the nearest cast cylinderPlanned 9.8 in
AnvilAbout 5 1/2 tons, on 16 baulks of about 84 cu ft (Percy p. 698)Cast iron at true sizePlanned 5.6 t
Working pressureNone given hereThe boiler and cast main the game hasPlanned 3.6 atm gauge; below 1.7 atm the block will not lift
Blow, full strokeNone givenWeight x fall, plus steam on the top face at full pressurePlanned 25.4 kJ: 7.5 fall, 17.9 top steam. The game's belly helve, 3.21 kJ
Steam a blowNone givenBoth ends swept at working pressurePlanned 0.078, 0.156, 0.234 kg at force 1, 2, 3
Blows a minute80 to 100 for small hammers (Overman p. 337)Low end takenPlanned 80 at most
On one Cornish boilerNone givenBoiler's steam against the hammer's usePlanned full force 49 a minute at usual firing, 68 fired open; force 1 holds 80

Open, and what you can tell us

The books disagree about the date. Nasmyth dates his sketch to 1839 and his patent to 1842 (Nasmyth pp. 241, 247-248), Percy names the patent of 9 June 1842 (Percy p. 695), while Fairbairn gives 1833 (Fairbairn Iron p. 133) and Swank 1838 (Swank p. 63). How fast a small hammer really ran is also unsettled: Overman's 80 to 100 strokes a minute (Overman p. 337) is a generation before Fairbairn's light hammers at 250 (Fairbairn Iron p. 136). The bore of the Atlas Works hammer is not in Percy, so if you know a maker's table for hammers of this size, we would like to read it.

Picture credits