Ironstone and roasting

Most of Britain's iron in the period did not come from bright ore veins. It came from dull grey stone lying in beds among shale and coal, at about a third iron, that had to be burnt before it was worth carrying to a furnace. How to prepare ore for the furnace in the game is in the burden guide. This page is about what the ores were, how they lay, and what roasting did to them.

Clay ironstone as it comes out of the shale: a flattened lump with nothing of metal about it. The mineral is iron carbonate, "the principal ore from which iron is smelted in England and Scotland", found in the coal measures in "flattened spheroidal masses, varying in size from the dimensions of a small bean to pieces weighing a ton" (Overman p. 29). This one is from the Illinois coal measures, and the numbers are a museum's.
Clay ironstone as it comes out of the shale: a flattened lump with nothing of metal about it. The mineral is iron carbonate, "the principal ore from which iron is smelted in England and Scotland", found in the coal measures in "flattened spheroidal masses, varying in size from the dimensions of a small bean to pieces weighing a ton" (Overman p. 29). This one is from the Illinois coal measures, and the numbers are a museum's.
Red hematite from Cumbria, kidney ore, the other kind of ore in this page: already an oxide, heavy and rich. It forms "kidney-shaped" aggregates in the cavities of the Cumberland beds (Bauerman pp. 62-63), and the red hematites of Lancashire and Cumberland are "probably the richest ores of iron that we possess in this country" (Truran p. 7).
Red hematite from Cumbria, kidney ore, the other kind of ore in this page: already an oxide, heavy and rich. It forms "kidney-shaped" aggregates in the cavities of the Cumberland beds (Bauerman pp. 62-63), and the red hematites of Lancashire and Cumberland are "probably the richest ores of iron that we possess in this country" (Truran p. 7).
The Cleveland Ironstone in the cliff at Jet Wyke on the Yorkshire coast: the ledges are its seams, the lowest forming the rock platform on the beach. A bed of about 20 feet in all, "interstratified bands of ore, shale, and iron pyrites", of which 12 to 17 feet could be worked (Bauerman p. 78). Green and oolitic, it is the form the game's main seam takes.
The Cleveland Ironstone in the cliff at Jet Wyke on the Yorkshire coast: the ledges are its seams, the lowest forming the rock platform on the beach. A bed of about 20 feet in all, "interstratified bands of ore, shale, and iron pyrites", of which 12 to 17 feet could be worked (Bauerman p. 78). Green and oolitic, it is the form the game's main seam takes.

What the ores were

Truran sorts the ores of Britain by what comes with the iron: clay in the argillaceous ores of the coal measures, carbon in the carbonaceous ores of the same ground, lime in the calcareous ores, and silica in the red and brown hematites, the ores of the oolitic formation and the rest (Truran p. 1). The first two supplied the great works.

Pure iron carbonate is 48.22% iron, the rest being protoxide's oxygen and 37.93% carbonic acid (Bauerman p. 52). As raised it was far from pure. A rich Dudley stone was 37.7% iron and a stone nearer the Staffordshire average 29.3% (Truran p. 3); the South Wales seams averaged over 32% (Truran p. 2); Overman's figure is 30 to 33% (Overman p. 29). The name clay band or clay ironstone is the smelter's for this carbonate when it is shapeless and clayey, and black band for the kind carrying enough coaly matter to burn by itself (Bauerman p. 52).

Hematite is another matter. A selected specimen was 81.6% peroxide of iron (Truran p. 8), about 57% iron worked out, though "taking the average of a cargo as it comes from the mines the yield will not reach 50 per cent" (Truran p. 8). The Cumberland and Lancashire ore lay in "very irregular deposits" in the carboniferous limestone; near Cleator it formed a bed of 15 to 60 feet between a shale floor and a limestone roof (Bauerman p. 62).

How it lay

The game's two kinds of bed. On the left the oolitic ironstone seam: a sheet that crops out at one end and runs down at 1 in 3.3 to 48 m below its crop. Its first few blocks have weathered to limonite and loose stones lie along the crop. Below the water table each band waits for a stronger pump. On the right, planned, the coal measures: coal on fire clay with two courses of ironstone nodules in the shale between. The shares are of one body on flat ground.
The game's two kinds of bed. On the left the oolitic ironstone seam: a sheet that crops out at one end and runs down at 1 in 3.3 to 48 m below its crop. Its first few blocks have weathered to limonite and loose stones lie along the crop. Below the water table each band waits for a stronger pump. On the right, planned, the coal measures: coal on fire clay with two courses of ironstone nodules in the shale between. The shares are of one body on flat ground.

Clay ironstone comes in the coal measures, in seams of a quarter of an inch to three feet and in nodules of one inch to two feet, "lying parallel with and not unfrequently in close proximity to the coal seams" and mined in the same way as the coal (Truran p. 1). South Wales had seams adding up to twenty-one feet (Truran p. 2). Some was got by open working, known in Wales as "patching", but most came from pits, from a few yards down to a hundred and fifty fathoms (Truran p. 1). The ground that gave the iron gave the coal to smelt it.

Cleveland was different: a single thick bed of oolitic stone, usually "a dull bluish green" (Bauerman p. 78), and lean, at 29 to 32% iron raw (Bauerman pp. 212-213). Its furnaces burned hard coke from the south of Durham (Bauerman p. 212), so the coal came to the stone.

The game keeps the two shapes. The oolitic ironstone bed follows Cleveland and is a single thick sheet with no coal in it. The planned coal measures follow Wales, with coal and iron from one drift and much more waste rock for each ton. The hematite flat follows Cumberland: rich, small, in limestone, always below the water table.

Where a bed meets the ground it rusts. The first blocks of an ironstone layer below the surface are vanilla's limonite ore, an oxide already, more of it in hot, dry ground and little or none in cold, wet ground.

Why it was roasted, and how

"Whether an iron ore should be roasted, is a question which very seldom arises" (Overman p. 39). Only massive red hematite and certain magnetites were commonly exempt (Bauerman p. 106). Roasting does three things. It drives off water and carbonic acid, so the iron is concentrated into less weight. It leaves the pieces in their shape but porous, so the furnace gases can reach inside them (Bauerman p. 106). And it decomposes sulphides, sending the sulphur off (Bauerman p. 106). Carbonate is the ore it does most for.

What the fire actually changes. Cleveland stone of 29 to 32% iron comes out at 40% (Bauerman pp. 212-213), so a quarter of its weight has gone before it reaches the furnace. Argillaceous stones lost 20 to 33%, 27% on the average of 18 assays (Truran p. 21), and clamped ironstone 28 to 33% (Bauerman p. 108). The rich hematites of Lancashire and Cumberland lost about 6% (Truran p. 21), and being dense and hard to burn they were often charged raw (Truran p. 19).
What the fire actually changes. Cleveland stone of 29 to 32% iron comes out at 40% (Bauerman pp. 212-213), so a quarter of its weight has gone before it reaches the furnace. Argillaceous stones lost 20 to 33%, 27% on the average of 18 assays (Truran p. 21), and clamped ironstone 28 to 33% (Bauerman p. 108). The rich hematites of Lancashire and Cumberland lost about 6% (Truran p. 21), and being dense and hard to burn they were often charged raw (Truran p. 19).

The first way was the clamp, on open ground. A bed of coal a few inches thick, ten or twelve inches of stone, then more coal and stone to four or five feet, lit from the bottom and left for eight to twelve days (Truran pp. 15-16, Bauerman p. 107). It cost almost nothing to set up and worked poorly: the outside stones came out half burnt, a change of wind or a shower threw the heat about, and "adjacent stones are not unfrequently melted into a hard refractory mass" (Truran p. 16). It also cost more coal, 2 cwt of small and half a cwt of large to each ton, against 1 cwt of small in a kiln (Truran p. 17). Black band was calcined on its own coaly matter, with slack only at the base to start the fire (Bauerman p. 108).

The kilns at Dowlais (Truran, Plate I, p. 15): front, two sections and plan. The inside is 20 feet long, 9 feet wide at the top and 18 feet high, with a floor of two-inch cast plates, a fire-brick lining and an iron ring on the upper edge. Roasted ore is drawn at floor level through the arches, and the small openings above them set the draught.
The kilns at Dowlais (Truran, Plate I, p. 15): front, two sections and plan. The inside is 20 feet long, 9 feet wide at the top and 18 feet high, with a floor of two-inch cast plates, a fire-brick lining and an iron ring on the upper edge. Roasted ore is drawn at floor level through the arches, and the small openings above them set the draught.
Another arrangement (Truran, Plate II, p. 20): at several works the kilns stood under an elevated tram-road, so that the waggons discharged directly into the kiln.
Another arrangement (Truran, Plate II, p. 20): at several works the kilns stood under an elevated tram-road, so that the waggons discharged directly into the kiln.

A kiln was worked as a slow column. Two or three small coal fires went on the floor, then raw stone over them; a layer eight or nine inches thick with about 5% of small coal was added when the last layer reached a red heat. By the time the kiln was full the bottom was cold enough to draw (Truran p. 15). A Dowlais kiln held 70 tons of argillaceous ironstone and calcined 146 tons a week (Truran p. 16), about three and a half days in the kiln (Bauerman p. 112).

The danger was too much fire. A carbonate heated too hard "melts together with the oxide, and forms a black cinder" (Overman p. 40), and in a clamp a place burning too fast was smothered with small ore or ashes before the stones clotted together (Bauerman p. 108). Pieces had to be near one size: Truran found the time to heat a stone nearly proportional to its smallest diameter, a 12 inch piece taking 24 hours and a 2 inch piece 4 (Truran pp. 19-20). Fuel enough for the big lumps overburnt the small ones.

The oven Overman preferred (Overman p. 42, Fig. 2): an arch of coarse ore at the foot holds up the charge, leaving a space where the fuel burns apart from the ore, so the workman gives "just so much heat" as he thinks needed (Overman pp. 42-43). It cost more to run and could roast any ore coarse enough to pass the draught.
The oven Overman preferred (Overman p. 42, Fig. 2): an arch of coarse ore at the foot holds up the charge, leaving a space where the fuel burns apart from the ore, so the workman gives "just so much heat" as he thinks needed (Overman pp. 42-43). It cost more to run and could roast any ore coarse enough to pass the draught.
The Gjers kiln of the Cleveland district, as printed in 1870 (Bauerman p. 113, Fig. 1): a fire-brick shell cased in iron plates, 20 feet across at the widest, over a cast-iron floor with a central cone that sends the ore outward to the draw openings between the pillars. The ore stayed about two and a half days, on 1 ton of slack to 20 of ore (Bauerman p. 113).
The Gjers kiln of the Cleveland district, as printed in 1870 (Bauerman p. 113, Fig. 1): a fire-brick shell cased in iron plates, 20 feet across at the widest, over a cast-iron floor with a central cone that sends the ore outward to the draw openings between the pillars. The ore stayed about two and a half days, on 1 ton of slack to 20 of ore (Bauerman p. 113).
What is left at Rosedale East on the North York Moors: the arches of a row of calcining kilns built into the hillside. Rosedale's stone, a dark magnetic oolite (Bauerman p. 78), was burnt where it was mined.
What is left at Rosedale East on the North York Moors: the arches of a row of calcining kilns built into the hillside. Rosedale's stone, a dark magnetic oolite (Bauerman p. 78), was burnt where it was mined.

What the game keeps of it

Raw ironstone blooms and charges the blast furnace as it comes. The roast is optional, and the kiln is planned to come after the casting plant; see the calcining kiln. Crushing is in the ore crusher. Where the numbers go after the kiln is in ore and fuel per ton.

Period basis, and what is the game's own

  • Period: ironstone beds and nodules in shales, and hematite in limestone, each lying as the books describe them (Truran pp. 1-2, Bauerman pp. 62, 78). The game's oolitic seam, hematite flat and planned coal measures follow those three.
  • Period: iron carbonate loses its carbonic acid in the fire and becomes a richer, more porous oxide (Bauerman p. 106). The game plans the roasted item to give more metal per ore unit, 9.2 u against 8.5 u raw.
  • Period: a shaft kiln worked as a column, charged at the top with stone and small coal and drawn at the foot (Truran p. 15, Bauerman p. 112). The game plans the same kiln, and over-fired ore turns to clinker as Overman warns (Overman p. 40).
  • Period: most ore was roasted as a matter of course (Overman p. 39). The game's own: roasting is optional for every ore, because a player without a kiln must not be locked out of iron.
  • The game's own: one roasted ore unit is one raw ore unit. The real stone lost a quarter or more of its weight in the kiln; the game keeps the count and raises the yield instead.
  • The game's own: the bed dips 1 in 3.3, steeper than the real beds, whose fall the books read for this page do not give, so that a body of 160 m fits four depth bands and the pumps have something to climb. See the Cornish pumping engine and mine drainage.
  • The game's own: the base game's iron ore is an oxide in small discs deep in rock, with no seams and no carbonate. See vanilla against reality.
  • The game's own: the coal measures with their nodules and the kiln itself are planned, and the water that closes the deep bands is planned too. The oolitic ironstone and hematite flat are what the mod's data has now.

The numbers

Ore figures here are period ratios set on vanilla's scale (how we scale); a block is a metre (scale and units). An ore unit is one crushed-iron's worth of ore.

In the gamePeriod figureHow it was scaledGame value
Iron in ironstone30 to 33% (Overman p. 29), over 32% in South Wales (Truran p. 2), Cleveland 29 to 32% (Bauerman p. 212)Half of hematite's ore units, the ratio below6 ore units a lump, 7.5 a block
Iron in hematiteSpecimen about 57% worked out, cargo under 50% (Truran p. 8)Twice ironstone, from the richer specimen12 ore units a lump, 15 a block
Bed thickness, ClevelandAbout 20 ft, 12 to 17 ft workable (Bauerman p. 78)True size, thinner2 to 4 m, average 3
Seams and nodules, coal measuresSeams 1/4 in to 3 ft, nodules 1 in to 2 ft (Truran p. 1)A one-block coursePlanned 1 m courses, 6 ore units a block
Hematite bodyBed 15 to 60 ft at Cleator (Bauerman p. 62)The game's own size3 m thick, lens 64 by 48 m
DepthPits to 150 fathoms (Truran p. 1)The game's own48 m below the crop
DipNone givenThe game's own1 in 3.3
Yield of iron, raw and roastedCleveland 29 to 32% raw, 40% calcined (Bauerman pp. 212-213)The game's own recovery ladder8.5 u raw, planned 9.2 u roasted, per ore unit
Weight lost roastingArgillaceous 20 to 33% (Truran p. 21), clamped ironstone 28 to 33% (Bauerman p. 108)Not modelledNone, one ore unit for one
Kiln fuelAbout 5% small coal, 1 cwt a ton (Truran pp. 15, 17)By heat valuePlanned 1:20
Time in the kilnAbout 3.5 days (Bauerman p. 112)Shortened by the make factor 3.608Planned about 22 game hours
Kiln throughput70 t held, 146 t a week (Truran p. 16)The same factor on the planned shaftPlanned about 58 ore units a game hour

Open, and what you can tell us

The books read for this page give the thickness of the beds but not their dip, so the game's slope is its own. If you know a source that gives the dip of the Cleveland or South Wales ironstone beds, we would like to read it.

Picture credits