Kirk, The Cupola Furnace (1899)

Books The Cupola Furnace

Covers E. Kirk, "The Cupola Furnace: A Practical Treatise on the Construction and Management of Foundry Cupolas" (1899, 432 PDF pages). Pages below are the book's printed folios, which are the PDF page less 24. PDF pages 1, 4-6, 386, 427-429 and 432 carry no text layer (front matter, advertisements). Two tables have no usable text and were read from images: heights and capacities by diameter (p. 14) and the Buffalo blower table (p. 309); no other figure was looked at. Read in full: chapters I-VI (pp. 1-35, 44-148: construction, tuyeres, management, the melting experiments, fluxing), XI-XVII (pp. 222-262: over-iron, what a cupola melts, cost, bad melting, melters, explosions), XIX-XX (pp. 271-299: hot blast, banking, blast pipes, gates, gauges, air per ton) and XXII-XXIII (pp. 332-345). Skipped: the catalogue of patent tuyeres (pp. 36-43), the patent cupolas of chapter VII (pp. 153-205), the art of melting, scales and accounts (pp. 206-221), tin-plate scrap (pp. 226-229, except the scraps on p. 342), spark catchers after p. 263, the blower catalogue (pp. 300-331, except pp. 306-309) and the index. Subject the construction and management of the foundry cupola - proportions, tuyeres, blast, melting, fluxing, faults, ladles and spark catching.

What the book says

The furnace and its proportions (pp. 1-7, 13-14, 136, 298, 332)

  • The cupola melts a ton of iron on 172-224 lb of coke; the blast furnace spends 20-25 cwt of coke per ton of pig, the reverberatory 10-20 cwt, the pot furnace a ton (pp. 1, 332). It "can be run intermittently" and "readily be put in and out of blast" (p. 1). "The blast furnace is a cupola furnace, and is constructed upon the same general principle" (p. 136).
  • Height, bottom plate to charging door, by bore (p. 14): 18 in, 6-7 ft; 30 in, 9-12 ft; 40 in, 12-15 ft; 60 in, 16-20 ft. A 40 in cupola melts 4-8 t an hour, 8-20 t a heat. A high cupola melts more with less fuel, because the escaping heat warms the stock (p. 13); the gas that escapes unburnt lights above the stock once it sinks below the door (p. 13).
  • Side tuyeres cannot carry the blast to the centre of a cupola wider than 4.5-5 ft; above 60 in melting does not rise with size (pp. 297-298). Before modern tuyeres "two or more cupolas were run at the same time" for a large casting (p. 30).

Tuyeres (pp. 18-21, 49-57, 89)

  • "At least two tuyeres are always placed in a cupola", on opposite sides, because a single tuyere drives the heat against the far wall "and the destruction of lining at this point is very great" (p. 54). Two suffice for a small cupola; more promote bridging (p. 18). Combined tuyere area three times the blower outlet (p. 49), two to three times (p. 19).
  • "It is not the velocity ... that does the melting. It is the volume of blast" (p. 49). Small tuyeres are closed by a lump of fuel (p. 50).
  • Height: 2-4 in above the sand bottom for hand-ladle work, 6-8 in for heavy work, never above 10-12 in; jobbing shops use 18-24 in, Bessemer works 5 ft, and fuel below the tuyeres is wasted (pp. 20, 52-53).
  • The fuel in front of a tuyere is bright when the blast goes on, then "chilled and blackened by the large volume of cold blast", and still passes air; poking is needed only when slag threatens to run iron into the tuyere (p. 89).

Taps, slag hole, spout (pp. 16-18, 51-53, 65-75, 98, 339)

  • One tap hole at the bottom plate runs "all the iron a cupola is capable of melting"; two above 8 t an hour for hand ladles; six 40 lb ladles a minute is the most a spout can serve (pp. 17, 73-74).
  • The slag hole sits below the tuyeres, at the back or "at the greatest possible distance from the tap hole" (p. 18). It is needed only when a small cupola runs over 1.5 h or a large one over 2.5 h (p. 17). Slag floats on the iron; a low slag hole can stand open all heat and regulate itself (pp. 18, 51-52).
  • "Molten iron should be handled in a ladle and not in a cupola"; iron held in the cupola or run into a low flat bottom "becomes dull"; "iron can be kept hotter in a ladle than in a cupola" (pp. 20, 51-52, 65-66, 88, 341). When iron runs dull, "draw all the melted iron off at once" (p. 339).
  • The tap-hole diagnosis (p. 98): bed burned too much, the first charge comes fast and dull; bed too high, the first iron is a long time coming; too much fuel, slow and dull toward the last; iron charges too heavy, dull at the end of each charge; fuel charges too heavy, slow melting at the end of each.

The bed, the melting zone and charging (pp. 76-84, 113-131, 333-337)

  • Iron melts only in a zone 6-8 in deep in practice, about 18 in in the experiments, above the tuyeres; blast volume sets its height, not its depth (pp. 77, 120, 123-124). Bed top above the tuyeres: 14 in on anthracite, 18 in on hard Connellsville coke, 20-25 in on soft coke (p. 78).
  • Charging in alternating layers melts "about one-half greater" than mixed charging (p. 80). The first charge is 1.5-4.5 times the bed's weight on coke; charges run 6-15 lb of iron to the pound of coke (pp. 81- 82). "The rule for charging a cupola is to place three pounds of iron upon the bed to each pound of fuel placed there, and ten pounds of iron upon the charges of fuel" (p. 333).
  • Too heavy an iron charge sinks the bed until "iron melts dull, or sinks below the bottom of the zone, and melting ceases" (p. 128). Excess fuel places the iron above the zone: "Fast melting cannot be done, or hot iron melted, with too large a quantity of fuel" (pp. 128, 336-337).
  • Ratios: theory 10 lb iron per lb anthracite, 15 per lb coke; practice 6 and 8 (p. 90). Good melting: 8 on coke, 7 on anthracite, 6 on charcoal, 4 on gas-house coke; best practical 6.5-7, 5-6, 4-5, 3 (p. 343). Short heats 8, most heats 6-7, heats of days about 10 (p. 337). "Use a light blast when melting with charcoal or gas-house coke" (p. 343).
  • "Iron may be melted in almost any cupola with either coal or coke"; anthracite burns away slower in the zone and carries heavier charges (p. 130). Fuel below the tuyeres stays dull red and is not consumed (pp. 121-122, 334).

Blast (pp. 85, 115-121, 276-279, 290-296, 309, 341)

  • Full blast from the start and the same to the end; "never vary the blast in different parts of a heat" (p. 85). When the blast goes on, dust goes up the stack, then "a bluish colored gas which bursts into a bluish flame as the stock settles, changing to a yellowish hot flame as the stock sinks still lower"; kept charged to the door, the gas does not ignite (p. 85).
  • Gauge pressure mostly measures the tuyeres' restriction; a cupola at 2 oz out-melted one at 6 oz (pp. 117-118, 293). Close-packed charges raise the pressure (pp. 115, 120, 295). A large volume blew the fire out of the experimental cupola (p. 119); a strong blast hardens iron (pp. 143, 145).
  • Buffalo blower table (p. 309): 8-14 oz for cupolas from 20 to 84 in. 30,000 cu ft of air melts a ton (p. 296). The blast gate puts volume "under control of the melter" (p. 290).
  • Banking: a charged cupola with iron and slag drawn off and the tuyeres rammed shut held from 4 p.m. to 8 a.m. and melted well (p. 276); Knoeppel's 60 in cupola, banked with eleven tons in it, poured hot the next morning (pp. 277-279). A cupola charged and closed front and tuyeres "may be left to stand for hours" (pp. 340-341); the bed barely burns under a charge with the draught shut (p. 133).

Flux, slag and loss (pp. 17, 84, 135-147)

  • Limestone in small doses (5-10 lb a ton) only makes a brittle cinder and tends to bridging (pp. 84, 136). To tap slag: 25-100 lb a ton, on the iron, none on the bed or the first sixth of the heat; the slag weighs about what the limestone does (pp. 84, 137-138, 140-142). Slagging does not melt faster; it keeps the cupola in blast longer (p. 141).
  • Slag holds 4-7 % iron; the loss of iron was 3-4 % greater when the cupola was slagged (p. 138). Tin-plate scrap loses up to 25 % under a very strong blast (p. 342).

Dropping the bottom and the dump (pp. 98-102)

  • Melted iron is drawn off to the pig bed, then the props are knocked out; "a sheet of flame and dust instantly shoots out ten feet or more" (p. 99). The dump is picked for iron, milled for the rest, and its coke taken back to the scaffold; anthracite through a cupola is worthless (p. 101).

Hazards (pp. 64-65, 257-262)

  • A wet spout, bod, sand bottom, floor or mould explodes molten iron; rusted wet scrap and a water-filled casting charged whole have wrecked cupolas (pp. 257-262).

Hot blast (pp. 271-275)

  • Recuperative hot-blast cupolas saved little fuel and broke their pipes; "the only practical means" of using the waste heat is a high cupola (pp. 13, 273-275).

Ladles

Pages are printed. West, Stimpson and Bolland's Supplement are not in books/.

  • Hand ladle holds 40 lb; "Six 40-pound hand ladles are all that can be safely taken from a spout per minute" (p. 74); a cupola too slow to fill one in 8 or 10 seconds taps turn about (p. 52). Bull ladle is named once, "small bull- and hand-ladle work" (pp. 8, 11), between hand and crane; no weight. Crane ladles swing to and from the spout (p. 26).
  • Spout height above floor: hand 18-20 in, bull and hand 24-30 in, crane 3-5 ft (p. 11); hand ladle rests on a hollow pedestal 8-10 in high (p. 3). Spout 6-8 in wide, 1-2 ft long for small ladles (p. 17).
  • First few ladles warm the ladles and go to the pig bed (p. 87); gangs sized so ladles do not chill between catches (p. 88).
  • Ladles are daubed with fire clay and loam and dried with wood (pp. 231-232); a damp ladle or wet daubing explodes the iron (p. 262). Skimmers are heated red before use (p. 259). Leftover iron goes to a pig mould (p. 222).
  • No shank, crutch, bail, crew count or shell material is given.

Spark catchers and the stack top (pp. 263-270)

Printed pages; the PDF page is the printed page plus 24 (chapter XVIII, "Spark catching devices for cupolas", is PDF 287-294). Figures 55-57 not viewed; no dimension is quoted from them.

  • Caps: "Iron caps or hoods are also placed one or more feet above the top of cupola stacks to arrest sparks; but they, like the arch, only arrest the sparks in their upward flight and throw many of them down upon the foundry or scaffold roof." (p. 268). The only mention of a cap; Kirk rates it poor. No cone, angle, rise, diameter, overhang or thickness is given; the material is only "iron".
  • Gap: "one or more feet above the top" of the stack (p. 268).
  • Fire-brick arch: "a half circle fire-brick arch opened at both ends on the top of the stack, making its total length and breadth equal to the outside diameter of the stack" (p. 268); sparks escape out of its open ends.
  • Disk: "an iron disk of a few inches smaller diameter than the stack ... in the stack near the top" (pp. 268-269); not for contracted stacks with a strong blast, and the iron is soon destroyed by the heat and gas.
  • A hood is also named as tried and failing at one Albany foundry: "A number of devices, such as hoods, etc., were tried, but none of these proved effective" (p. 267).
  • The best device is no cap but an enlarged stack: "provide sufficient room in the stack for the blast to expand ... and lose its lifting force, when the sparks will fall back in the cupola" (p. 269). Stack casing the same diameter as the cupola casing, lined with four-inch fire-brick on angle iron, "when the stack is of proper height" throws no sparks (p. 269); where a high stack is not wanted, the enlarged stack of Fig. 56, larger than the cupola, "low stacks requiring to be of a larger diameter than high ones" (p. 266). No ratio is given.
  • A contracted stack "throws out a great many sparks at the top. An enlarged stack ... throws out very few" (p. 5 of the casing chapter, PDF 29).
  • Stack height above the charging door: no figure. Only "when the top of the stack is only a few feet above the charging door" draught for lighting is weak, a proper height gives enough (p. 270); stacks run "from a few feet above the foundry roof to twenty or thirty feet" (PDF 29; printed page not read off). Fig. 56 cuts the casing at the bottom of the charging door and sets the larger stack casing on a plate (below).
  • Stack plate: "an iron plate or ring placed upon the top of the cupola casing, where it is supported by the casing and cast-iron brackets riveted or bolted to it on the outside"; the outside "extends over the cupola casing from six to twelve inches" (p. 266). A 6-12 in overhang is the only overhang figure; it is the plate's, not a cap's.
  • Support of a cap or hood: not given. The Fig. 55 old brick stack stands "upon an iron plate supported by iron columns, on a level with the top of the cupola" (p. 263); the O'Keefe return flue throws a dome "across the cupola stack above the door" with the flue led down to the floor and back (pp. 267-268), the dome with a small opening so the cupola could be charged in blast.
  • Not given: a cap's diameter, form, angle, rise, thickness, bracket or leg detail, and any stack height in inches or feet above the door for a given bore.