Brass and zinc

Brass bushes sit at every bearing and gland the game's machines turn, and brass is copper with zinc in it. Getting the zinc in is the hard part, because of one fact of physics: zinc boils before copper melts. Zinc and mercury also come out of their ores in the same way, as a vapour caught in a closed vessel. The retort furnace, its mercury and direct-fusion brass are planned and not yet in the game, and no guide teaches them yet. This page takes the furnace with its metals, and is about the real practice and how the game plans to follow it. Where the metals sit in the order of the game is on the metal ladder, and the crucible's other use on cementation and crucible steel.

Sphalerite, the zinc ore of the game and of Percy's "blende": a sulphide of zinc, dark and glassy. The game's zinc comes from it, roasted first to drive off the sulphur.
Sphalerite, the zinc ore of the game and of Percy's "blende": a sulphide of zinc, dark and glassy. The game's zinc comes from it, roasted first to drive off the sulphur.
Cinnabar, a sulphide of mercury, in red-black crystals on quartz. The base game mines it and makes only a red dye from it.
Cinnabar, a sulphide of mercury, in red-black crystals on quartz. The base game mines it and makes only a red dye from it.
The melting and boiling points of the three metals. Copper is still solid at 906 C, when zinc is already boiling away, and the dashed line marks where copper first turns liquid. Anything that melts copper in the open drives zinc off as a vapour. Mercury is a vapour well before that, at 350 C. This one chart is why zinc cannot be melted from its ore in an open pot, and why brass was so long made without ever melting the copper.
The melting and boiling points of the three metals. Copper is still solid at 906 C, when zinc is already boiling away, and the dashed line marks where copper first turns liquid. Anything that melts copper in the open drives zinc off as a vapour. Mercury is a vapour well before that, at 350 C. This one chart is why zinc cannot be melted from its ore in an open pot, and why brass was so long made without ever melting the copper.

Why zinc had to be caught as a vapour

Zinc's oxide gives up its metal to carbon at a red heat (Overman p. 35), but the metal leaves as vapour: a zinc ore in a furnace goes off as white fumes of its oxide (Overman p. 31), or settles as oxide on the cold charge (Johnson Principles p. 288). In an open furnace it is simply lost up the flue. So zinc works shut the ore and the carbon in a closed clay vessel and led the vapour into something cooler, where it would condense as metal. No such vessel was needed for iron, copper or lead.

The ore needed a separate step first. Calamine, the older ore, is calcined; blende, the sulphide, has to be roasted to oxide, because few sulphides give up their metal to carbon alone (Overman p. 36). Roasting blende is "one of the most troublesome" operations of the kind (Percy 1861 p. 541). It was done on a flat reverberatory bed, at Silesian works stirred every ten minutes until the sulphurous vapours ceased (Percy 1861 pp. 550, 567-568), and "it is the custom in most zinc-works" to do it on the waste heat of the distilling furnace (Percy 1861 p. 569).

The Belgian zinc furnace, Figs. 689 and 690 of Robert Hunt's supplement to Ure (Ure Supplement p. 1093), with labels added. A narrow arched chamber full of clay tubes stands over a coal grate. Each tube leans to its mouth so that slag runs to the cooler end (Percy 1861 pp. 580-581), and has a clay nozzle luted in, and a sheet-iron cone on the nozzle. The zinc vapour condenses in the nozzle, and the waste gas goes on to roast the next charge of ore. Sixty or so retorts make one furnace (Percy 1861 p. 581).
The Belgian zinc furnace, Figs. 689 and 690 of Robert Hunt's supplement to Ure (Ure Supplement p. 1093), with labels added. A narrow arched chamber full of clay tubes stands over a coal grate. Each tube leans to its mouth so that slag runs to the cooler end (Percy 1861 pp. 580-581), and has a clay nozzle luted in, and a sheet-iron cone on the nozzle. The zinc vapour condenses in the nozzle, and the waste gas goes on to roast the next charge of ore. Sixty or so retorts make one furnace (Percy 1861 p. 581).

Three ways of distilling zinc were in use. The English pot furnace of Champion's Bristol works, from about 1740 (Ingalls pp. 1-2), led the vapour downward through a hole in the bottom of a big closed pot into a tube and a chamber below (Percy 1861 pp. 550-551). The Silesian furnace used larger clay muffles, begun at Wessola in 1799 or 1800 (Ingalls p. 2). The Belgian, or Liege, method that Dony found in 1805 (Ingalls p. 2) used many small clay tubes in rows, each one mended or replaced by itself, and its furnace "remained substantially the same" for the rest of the century (Percy 1861 p. 578). At one British works in 1861 Silesian and Belgian furnaces worked side by side, around 100 tons of zinc a week between them (Percy 1861 p. 584).

The Silesian furnace and its muffles, Figs. 691 to 697 (Ure Supplement p. 1095). The clay muffles M, one drawn alone at top right, are about 3 ft 3 in long and 20 in high. They stand along an arched chamber over the fire, and their fronts carry bent pipes for the zinc to run down. A Silesian retort took about 71 lb of charge (Percy 1861 p. 569) against 13 to 27 lb in a Belgian one (Percy 1861 p. 583), but it had to dry for three months before use (Percy 1861 pp. 559-560).
The Silesian furnace and its muffles, Figs. 691 to 697 (Ure Supplement p. 1095). The clay muffles M, one drawn alone at top right, are about 3 ft 3 in long and 20 in high. They stand along an arched chamber over the fire, and their fronts carry bent pipes for the zinc to run down. A Silesian retort took about 71 lb of charge (Percy 1861 p. 569) against 13 to 27 lb in a Belgian one (Percy 1861 p. 583), but it had to dry for three months before use (Percy 1861 pp. 559-560).

A Belgian retort was charged twice a day with ground, calcined ore mixed with half its weight of coal slack (Percy 1861 pp. 583-584), and after about six hours "most of the nozzles will be filled with melted zinc", which was drawn with a little scraper (Percy 1861 p. 583). Zinc wanted a heat close to white (Percy 1861 p. 569). Cracked or perforated retorts were replaced every morning, new ones having been "previously gradually heated to redness in an adjoining kiln" (Percy 1861 p. 583).

Why brass was melted the way it was

For centuries brass was made without any metallic zinc at all. Cementation packed copper with calamine and charcoal in closed pots (Ure p. 464), held "at a temperature below the melting-point of copper", so that the copper, "being thus exposed to the action of the vapour of zinc, becomes permeated with this metal" (Percy 1861 p. 617). If the copper did melt it ran to the bottom of the pot, "and much of the zinc would then escape" (Percy 1861 p. 617). The method works because of the chart above: the heat stays where the zinc is a vapour and the copper still solid. Champion took a patent for this old process in 1779 (Percy 1861 p. 612), and it was still being worked at Llanelly in 1859 (Percy 1861 p. 612).

Once metallic zinc could be bought, the order could be turned around. James Emerson's patent of 1781 made brass "by the direct fusion of its two metallic elements" (Ure p. 464). Copper went into the crucible first and was melted, then "the zinc is gradually and cautiously added to the copper when the latter has just melted" (Percy 1861 p. 618). The copper is already as hot as it needs to be, and the zinc spends as little time as possible in it before the pot comes out; a due allowance for the unavoidable fume went to the founders (Percy 1861 p. 618), and 2.5 to 4 lb of waste on a 120 lb charge, about 2 to 3%, is one maker's figure (Ure p. 466). The furnace was a hole about 10 in square and 2 ft deep packed round the crucible with coke (Percy 1861 p. 618; Ure p. 465). A cupola would not do: brass melted in one oxidizes and is inferior to crucible brass (Kirk p. 223).

What a player pays in ore zinc for 100 units of brass at 35% zinc. Today the base game loses nothing. The plan adds the two losses: the retort is a 95% yield, direct fusion loses about 3% to fume as the period did, and the base game's old crucible mix is planned to lose 20% of its zinc. All three give the same ingot; the routes differ only in what they cost.
What a player pays in ore zinc for 100 units of brass at 35% zinc. Today the base game loses nothing. The plan adds the two losses: the retort is a 95% yield, direct fusion loses about 3% to fume as the period did, and the base game's old crucible mix is planned to lose 20% of its zinc. All three give the same ingot; the routes differ only in what they cost.
What the brass was for: a connecting rod strap with its two brasses C in place, on the chuck to be bored (Rose p. 246, Fig. 879). A bearing of this kind is two pieces of hard brass or gun metal, one of which a wedge or screws can move toward the other as the bore wears (Farey p. 471). This is the part the game's brass goes into, as the bushes turned on the machining line.
What the brass was for: a connecting rod strap with its two brasses C in place, on the chuck to be bored (Rose p. 246, Fig. 879). A bearing of this kind is two pieces of hard brass or gun metal, one of which a wedge or screws can move toward the other as the bore wears (Farey p. 471). This is the part the game's brass goes into, as the bushes turned on the machining line.

Mercury from the same kind of furnace

Mercury is easier still to drive out of its ore, since it boils at 660 F (Overman p. 433), and it was being distilled in stacked earthen pots before 1635 (Ure vol. 3, art. Mercury). Lime takes the sulphur from cinnabar, and the mercury goes off as vapour (Ure vol. 3, art. Mercury). In the Palatinate, retorts took 56 to 70 lb of cinnabar with 15 to 18 lb of quicklime, about four to one, with their receivers part filled with cold water (Phillips p. 461); by 1847 Landsberg had iron-lidded retorts with a cast-iron eduction tube into a water-filled condenser, the charge "almost totally expelled in the course of three hours" (Phillips pp. 462-463). Mercury ores fire only at a cherry red (Phillips p. 459), well under the white heat zinc wants.

Distilling quicksilver, a woodcut from Agricola in the Wellcome Collection. Bell-shaped pots stand over a fire and their spouts lead into a second row of vessels, the same idea as the retort and its receiver. Mercury vapour is a poison, and much later the men at small California retorts still wore wet sponges and were often salivated (Egleston pp. 808-813).
Distilling quicksilver, a woodcut from Agricola in the Wellcome Collection. Bell-shaped pots stand over a fire and their spouts lead into a second row of vessels, the same idea as the retort and its receiver. Mercury vapour is a poison, and much later the men at small California retorts still wore wet sponges and were often salivated (Egleston pp. 808-813).

What mercury was for, on a steam plant, is the instruments. Its weight makes a column short enough to mount. Watt read his condenser's vacuum on a glass tube about 32 in long standing in a bason of mercury, the column around 28 in against a 30 in barometer (Farey pp. 377-378), and a mercury thermometer in a metal capsule reads the hot blast (Overman pp. 433-434). A boiler-room engineer was expected to know the gauges and the thermometer (Hawkins p. 9). None of the books at hand describes the closed-air mercury gauge that the game plans for the low-pressure boiler.

One atmosphere as a column of water and of mercury. Water stands 10.33 m to the atmosphere, mercury 0.76 m (Dichmann p. 7). The game's planned barometer gauge, closed-air gauge and thermometer use mercury for the same reason Watt's did, and the column keeps its true length, 12.2 voxels.
One atmosphere as a column of water and of mercury. Water stands 10.33 m to the atmosphere, mercury 0.76 m (Dichmann p. 7). The game's planned barometer gauge, closed-air gauge and thermometer use mercury for the same reason Watt's did, and the column keeps its true length, 12.2 voxels.

Period basis, and what is the game's own

  • Period: zinc was distilled from roasted blende or calcined calamine in closed clay retorts, with carbon, and caught in cooled nozzles (Percy 1861 pp. 578-584). The game plans the Belgian furnace at small scale: four retorts over a firebox in place of sixty, with a bed that roasts the next charge on the furnace's own waste heat (Percy 1861 p. 569).
  • Period: brass was made by adding zinc to molten copper in a crucible (Percy 1861 p. 618). The game plans the same in the pot of its crucible furnace, giving the base game's brass ingot, with about 3% of the zinc lost as fume (Ure p. 466).
  • Period: the older cementation brass (Percy 1861 pp. 612-617) is left out here and belongs to a later early-game mod, as does the bronze air furnace.
  • Period: cinnabar was distilled with quicklime, about four parts to one (Phillips p. 461), in iron or earthen retorts (Ure vol. 3, art. Mercury). The game plans an iron retort and the same charge.
  • The game's own: the base game melts zinc from hammered nuggets in its crucible at 419 C, and mixes brass with no loss. The plan takes away the nugget smelt, where the zinc would really boil away, and loses 20% of the zinc in the crucible mix, both on one switch that is on by default. The 20% is the game's own, set against the period's 2 to 3%, to make the retort worth building. See vanilla against reality.
  • The game's own: one furnace distils both metals, by two heats. With the door above it open the furnace runs at a red heat, for mercury and the roast; shut, it climbs to near white, for zinc. No book at hand sets zinc and mercury retorts in one furnace.
  • The game's own: mercury's only uses are the three instruments. It is not alloyed or amalgamated. The vapour harms a player standing close to an open furnace, an estimate of the danger rather than a figure from a book.

The numbers

Everything here is planned, except the base game's copper range and the column's length, and the furnace's numbers are not yet tuned. The furnace is scaled down in count, not in size, as the scaling page describes for period ratios, and scale and units gives the unit values.

In the gamePeriod figureHow it was scaledGame value
Retorts in a furnace61 at Moresnet, 78 in a larger furnace (Percy 1861 p. 581)Cut for a furnace built and mended by handPlanned 4
Zinc chargeCoal about half the ore by weight, Moresnet (Percy 1861 p. 584)Whole itemsPlanned 7 roasted ore and 3 carbon
Zinc won11 to 12% of the zinc lost, half of it recovered (Percy 1861 pp. 584-585); 2% lost in the roast (Percy 1861 p. 568)Period yield roundedPlanned 95%: 105 u of ore zinc gives 100 u
Retort life3 or 4 of 61 replaced a day (Percy 1861 p. 583), 15 to 20 days worked outThe game's ownPlanned 6 distillations (zinc), 12 (mercury)
Heat for zinc"near white" (Percy 1861 p. 569)The game's own, degrees read from coloursPlanned 1250 C
Heat for mercuryCherry red (Phillips p. 459)The game's own, degrees read from coloursPlanned 700 C
Brass rangeAbout 2 parts copper to 1 of zinc (Percy 1861 p. 608); Muntz metal 60 to 40 (Percy 1861 pp. 619-620)The base game's range kept60 to 70% copper
Direct-fusion loss2.5 to 4 lb of 120 lb, about 2 to 3% (Ure p. 466)Period valuePlanned about 3% of the zinc
Crucible-mix lossNone givenThe game's ownPlanned 20% of the zinc
Ore zinc for 100 u of 35% brassWorked out from the rows above35 divided by the zinc keptBase game 35 u; planned 38 u direct fusion, 46 u crucible mix
Mercury charge56 to 70 lb cinnabar to 15 to 18 lb lime, about 4 to 1 (Phillips p. 461)Whole itemsPlanned 4 cinnabar and 1 quicklime
Mercury yieldNone usedThe game's own, an untuned estimatePlanned 0.4 L a charge
Column to one atmosphere760 mm of mercury (Dichmann p. 7)True length0.76 m, 12.2 voxels

Open, and what you can tell us

The books at hand do not date the Liege process exactly: Percy could not say in which year it began (Percy 1861 p. 578), and Ingalls gives Dony and 1805 (Ingalls p. 2) in a later book. No book at hand names a furnace temperature in degrees, only colours. How long a direct-fusion crucible stayed in the fire is not stated. If you know a source for any of these, or a works where zinc and mercury really were distilled in one furnace, we would like to read it.

Picture credits