Stopping, holding back and diverting molten iron in an open runner

Books Blast-Furnace Construction in America, The Principles, Operation and Products of the Blast Furnace, The Cupola Furnace, The Iron-Founder, An Autobiography

Covers (printed folios unless said; the Johnson Construction pages were read by OCR, the PDF has no text layer)

  • J. E. Johnson Jr., Blast-Furnace Construction in America (1917), folio = PDF page less 17. Pp. 344-346 (PDF 361-363) read in full, p. 328-329 (pig beds, PDF 345-346), p. 347 (chills, PDF 347 and following).
  • Johnson, Principles of Blast Furnace Operation (1918), folio = PDF page less 32. Pp. 315-321 read (slag-notch bot, clay stopping, mud gun). The skimmer and dam are NOT in this book.
  • E. Kirk, The Cupola Furnace (1899), folio = PDF page less 24. Bod sticks p. 93, bod material pp. 94-95.
  • T. Bolland, The Iron Founder (1893 ed.), folio = PDF page less 14. P. 232 read.
  • H. Bessemer, Sir Henry Bessemer, F.R.S.: An Autobiography (1905), folio = PDF page less 47. Pp. 147-148.
  • F. Overman (1851): grep for dam, sluice, gate, trough, cut off found only the furnace-front dam-plate (pp. 167-195) and a chill-mould trough (p. 264); nothing on a runner gate. Not used.
  • US patent 608,143, M. Killeen, "Blast-furnace casting apparatus", filed 1897-10-19, granted 1898-07-26 (patents.google.com/patent/US608143A/en).
  • US patent 108,375, J. W. Middleton, 1870-10-18 (patents.google.com/patent/US108375). Purpose whether a slotted iron gate dropped across a runner of molten iron is period. Lines starting "Inference" or "INFERRED" are the note's own; all other lines are what the source states. Not found: any source before 1890 that describes a gate in a runner (the books are retrospective).

A. The iron runner with gate grooves (the direct answer)

  • Johnson, Construction, p. 346 (1917, describing practice after about 1890): "Following the introduction of the iron skimming trough, iron runners were substituted for the sand runner all over the cast house. These are provided with grooves into which gates are set to cut off the flow of the metal, and are grouted after each cast with a mixture of sand and clay kept hot by a steam coil, the gate grooves being filled up flush with this mixture."
  • Same page: "With good practice only a very light skull of iron is left in these iron runners when the cast is over, and the runner requires only a new coating of sand and grout to make it ready for use again."
  • Same page, the point that matters for a reusable gate: "with the iron runner it is possible to raise a gate which has once been set and allow iron to flow again down a section of trough from which it had been cut off, which it is never safe to do with sand runners because the withdrawal of the gate from the sand is extremely likely to cause a boil."
  • Same page: "These advantages have resulted in the universal introduction of iron runners where the iron is run direct to the ladles and not cast in sand beds."
  • Why an iron runner at all (p. 344): molten iron on a wet spot "cannot bore into the iron as it does into the sand in case of a boil"; the iron trough by the tap-hole "is thoroughly grouted and carefully dried because the grout prevents the iron from burning to the trough which it would do if the latter were bare". So the runner's coat is grout (sand and clay, wet-mixed), dried before the cast.
  • The name in Johnson is "gate" throughout; "gate grooves" for the slot. No thickness, width or depth of plate or groove is given.
  • Killeen patent, 1898 (granted; filed 1897): "a vertically-movable metal plate or block which fits within a transverse groove in the trough", the groove "lined or filled with sand, loam, or other earthy material in which the shutter will seat ... thus making a tight joint and effective cut-off for the metal". A second form: "the metal gate is provided with a grooved edge which fits over a rib cast in the runner transversely thereto". The patent text calls it a "shutter" in the claims' language and "gate" for the second form. No dimensions, no operating gear stated in the text (drawings not seen).

B. The skimmer, the dams and the Killeen trough (Johnson, Construction, pp. 344-346)

  • Sand runner era (to about 1890s): "a dam of sand some six to twelve or more inches high is put into the bottom of it four to eight feet from the end of the iron trough; just above or in front of this dam is set a skimmer, formerly a thick plate of iron cast in open sand in the floor of the cast house, and guided by two pairs of pins driven into the sand of the runner on each side of it."
  • Worked by hand tool: "When the cinder begins to show on the iron this skimmer is driven down with a hammer until its edge projects below the surface of the metal sufficiently to prevent the head of cinder floating on the iron from pushing under it, the cinder then rises and flows over a lateral dam, a low place in the side of the sand trough". So a plate is driven down into metal that is already flowing (a skimmer holds back the slag layer, it does not stop the iron).
  • Faults of the sand arrangement: slag got down to the beds, or "the iron itself would be forced out over the cinder dam and wasted"; a basin of iron under slag was left at the end of the cast.
  • Killeen skimmer (Edgar Thompson Works, Braddock): "an additional section of iron trough linked to the section next the furnace. It has cast with it the cinder and iron dams, a slot in which to set the skimmer (preferably made of firebrick in an iron frame to prevent its cutting away during the cast, as the old iron skimmers did), and a supplementary lateral outlet extending clear to the bottom of the trough and closed by an iron gate with a staple in the top. Before cast this gate is set to place and the joint between it and the body of the skimmer trough well protected with sand and grout."
  • End of cast: "The considerable volume of iron remaining in the trough at the end of the cast is released by withdrawing the supplementary lateral gate and allowing the iron to run into a special runner ... The cinder is prevented from following it by a hand gate dropped when the iron is all discharged and before much of the cinder can follow." A gate here is lifted to open (staple for a hook or bar) and a second hand gate is dropped across a stream that is still running. The same drop also "drains the iron trough effectively to the bottom" so the tap-hole can be reached.
  • Date of the Killeen design: patent filed 1897, granted 1898; Johnson gives "the nineties" for the growing outputs that forced it.

C. Sand beds and runners (Johnson, Construction, pp. 328-329, 347)

  • "Casting in sand beds. - The runner and beds were until about twenty years ago practically always made up in sand, since sand does not wash in a liquid with a high surface tension like a molten metal, and at the same time is porous enough to permit the free escape of the gases" (p. 328). The runner is the "sow" and the pigs lie across lateral runners (p. 328).
  • Sand "must be coarse and open-grained, practically free from clay to prevent boils" (p. 329). A boil at the end of the iron trough is the worst interruption (p. 329).
  • Withdrawing anything from sand is unsafe (p. 346, above). Johnson does not describe cutting a sand dam with a bar or shovel to start a bed; the book gives the iron running down a trough to the beds and a "special bed". Not documented in the books read.
  • Chills (p. 347): moulds "washed with a solution of clay water" and given "a coating of clay all over the molds", "virtually infusible"; a clay wash on iron is period for iron-against-metal contact.
  • Inference: stopping a sand bed meant the founder's own sand or clay thrown in; no source here says how.

D. The bot, the bod and the stopping hook (context: they stop a tap hole, not a runner)

  • Slag notch "bot" (Johnson, Principles, p. 315, 1918): "The cinder notch ... is closed by an iron plug on the end of a long iron rod, which serves as a handle. This piece, called the 'bot,' acts as a plug and also as a chill ... the furnace 'keeper' takes the bot by its long handle, puts it through the cooler and up to the base of the monkey, then gives it a quick thrust and holds it for a few seconds. The cold bot and the water-cooled walls of the monkey chill the thin skin of slag between them and instantly make a perfect joint." It is closed against slag already flowing, and left in until the next flush, "pulled out with a ring and wedge". The handle can be bent so the keeper stands to one side. Date: the water-cooled monkey and bot are late 1800s to 1900s.
  • Iron tap hole (Principles, pp. 318-320): "Until a few years ago it was the universal custom to stop the tapping hole with balls of fireclay driven in by a stopping hook operated by hand labor". The steam "mud gun" (Vaughan, Johnstown, "about 25 years ago") shoots clay in; with mechanical swinging gear (very recent in 1918) it works "even when the iron is running full force"; before that it was done "with the iron running full force" only by "furnacemen of unusual nerve" (p. 321).
  • Cupola bod (Kirk, 1899): "The bod is a plug used for closing the tap hole when it is desired to stop the flow of iron from a cupola" (p. 94). Stuff "worked a little wetter than molding sand", a handful "pressing it firmly on the end of the stick" (p. 96 about); sticks of wood "from 5 to 10 feet long", or a "combination stick" with an iron rod and "round button" (p. 93). "the stick is held against the bod in the tap hole until it sets" (p. 93). Too wet a bod makes "the iron fly"; fireclay-rich bod holds water (p. 94). A bod is made "large, long and pointed" for a long stop-in (p. 96 about). A cupola is stopped in against a running stream: the bod is pressed in while the iron flows (stated as the normal use, p. 94). The bod is cut away with a tap bar by a half rotation (p. 96 about).
  • Hot skimmers are warmed to red heat before a ladle skim, "wet skimmer made iron explode" (Kirk p. 259 about, PDF 283).

E. Ladle stoppers and gate plugs (context)

  • Bessemer (1905, describing 1856): the pan below the converter had "A fire-brick plug attached to a long handle ... fitted to a fire-brick ring or opening in the bottom of the pan" (p. 147); the casting ladle, "loamed-up", "provided with a discharge valve at the bottom, so that it can be moved from mould to mould by closing the valve" (p. 148). The PDF text has the ladle valve at p. 148; p. 149 was not found to say more.
  • Bolland (1893), p. 232: for a casting run from one ladle, "let plugs covered with loam be inserted in these gates, and the runner made around them, withdrawing them in time to let in the hot iron to the top of the mould". A loam-covered plug that is pulled while metal stands behind it: loam is the foundry coat on any iron that meets iron.
  • Middleton patent 108,375 (1870): suspended receiving vessels with an outlet closed by a "wedge-gate" and a "stop or gate ... simply a slide which covers and opens the outlet", pressing a clay stop in; vessels "of iron, lined with fire-brick or clay". A slide gate of iron for molten iron, 1870, in a casting scheme, not a runner.

Findings on the runner gate

  • Is a slotted iron gate in a runner period? Yes, sourced. Johnson 1917 p. 346: iron runners "provided with grooves into which gates are set to cut off the flow of the metal", from the 1890s when iron runners replaced sand ones; the Killeen trough (patent 1898) has a slot for the skimmer and a gate across its side outlet. Before about 1890 the books describe none (runners were sand; the stop was sand, the hammered-down skimmer plate and the plugged tap hole). Not found before the 1850s; present in the iron-runner works of 1890-1910.
  • Name and coating: "gate", the slot "gate grooves"; the patent says "shutter" or "metal plate or block". The groove is filled flush with "sand and clay" grout kept hot (steam coil) after each cast, and the groove is "lined or filled with sand, loam, or other earthy material" in which the gate seats (Killeen). The runner itself is grouted and dried. The plate's own wash is not stated for gates; Bolland's gate plugs are "covered with loam" and chills take a clay wash (INFERRED that a gate plate was loam- or clay-washed). Material of the plate: "metal"; "iron gate" (Killeen skimmer outlet); the old skimmer a "thick plate of iron cast in open sand".
  • How worked: by hand. A "hand gate dropped", a gate "withdrawn" and "raised", with "a staple in the top" to take a hook or bar. The old skimmer was "driven down with a hammer". No lever, chain or rack on a runner gate is stated. Lever-held bot gear exists only for the slag notch (Principles p. 315) and the mud gun swing (Principles p. 321).
  • Closes on flowing metal? Dropped across a running stream: yes, in the Killeen trough the hand gate is "dropped when the iron is all discharged and before much of the cinder can follow", and the skimmer is driven into flowing iron. Raised again later: stated safe in an iron runner, unsafe in sand. A gate set before the cast and "well protected with sand and grout" is the normal state (Killeen). The books say nothing about a gate closing on a full-force stream of iron; INFERRED that closing a thin plate into a stream is hand work with a little spill, and that the grooved joint is sealed by the hot grout, not by the metal.
  • What it looks like: a flat iron plate, a "thick plate" for the old skimmer; the second Killeen form has a grooved lower edge riding a rib across the runner floor. The slot is a transverse groove across the trough floor and walls, filled flush with grout after a cast; sand dam six to twelve inches high, four to eight feet below the trough end. Thickness of the plate, width and depth of the slot are not stated by any source read. Any numbers are INFERRED.
  • Stopping a runner with clay at both ends matches the pre-1890 practice only loosely; Johnson gives no clay cap on runner ends. The bot and bod are plugs on a long handle pressed into a hole, not slotted plates, and fit a tap or notch, not a runner.