Wilson, Treatise on Steam Boilers (1873), Fairbairn, Useful Information for Engineers (1860), Courtney, Boilermaker's Assistant (1880) and Hawkins, Maxims and Instructions for the Boiler Room (1897-1903): boilers, plate, safety valves, injectors and pipes
Books A Treatise on Steam Boilers, Useful Information for Engineers, The Boilermaker's Assistant, Maxims and Instructions for the Boiler Room
Sources
- Robert Wilson (chief engineer to the Manchester Steam Users' Association), A Treatise on Steam Boilers: their Strength,
Construction, and Economical Working (London, Lockwood, 1873), archive.org
atreatiseonstea02wilsgoog. No text layer; read through OCR and the key pages checked on the page image. Pages are printed folios; PDF page less 15 near the front, less 13 around pp. 270-297, less 12 from p. 300. The scan of pp. 298-299 (one PDF leaf, 311) is cut away and unreadable; text there (Lancashire boiler example, power rules) is not used. - William Fairbairn, Useful Information for Engineers, Second Series (London, Longman, 1860), archive.org
usefulinformati01fairgoog. No text layer; OCR plus page images. Printed folio is PDF page less 29 for Lecture VIII (pp. 299-314). - John Courtney, The Boilermaker's Assistant (Weale's series, revised by D. Kinnear Clark, London, Crosby Lockwood, 1880),
archive.org
boilermakersass02courgoog. No text layer; OCR plus page images. Printed folio is PDF page less 10. - Hawkins, Maxims and Instructions for the Boiler Room (Theo. Audel and Co., New York, copyrighted 1897, 1898, 1903),
archive.org
maximsinstructio00hawk. Text layer (rough OCR). Printed folio is PDF page less 4. American practice of the 1890s, steel plate; the other three are British, wrought-iron plate, 1860-1880. Purpose real figures for boiler evaporation, Cornish proportions, working pressure and plate rules, safety valve, injector, cast pipe, efficiency and pipe loss. Facts are the books'; INFERRED marks my own arithmetic or reading. Units converted here (1 lb = 0.4536 kg, 1 ft = 0.3048 m, 1 in = 25.4 mm, 1 sq ft = 0.0929 m2, 1 psi = 6.895 kPa, 1 atm = 101.325 kPa = 14.7 psi, long ton per sq in = 15.44 MPa); the books give no metric. Every pressure in the books is a gauge pressure (above the atmosphere) unless stated, so the atm column is gauge atm.
0. What the books do not give
- Injector steam-to-water ratio: not in Hawkins pp. 206-211 or any other book here. Section 7 works one out (INFERRED).
- Heat loss per sq ft of bare or clothed steam pipe (Btu per sq ft per hour): not given. Hawkins gives only a relative value table of coverings (wool felt = 1.000) and one loss figure for a mile of main (section 9).
- A working pressure for cast-iron pipe: none. The books give cast-iron's tensile strength (Wilson, about 6 tons per sq in) and the boiler of vertical cast pipes worked at 80-90 lb, nothing for a pipe line (section 8).
- A Cornish boiler's heating surface, water space or steam space for a small mill boiler as a table: not given. Wilson works one Cornish example (p. 300) giving shell, flue, length and grate only; Courtney gives rules from nominal hp. Section 2 derives the rest (INFERRED).
- Cornish working pressure by decade as a table: only scattered statements (section 3).
- Wilson's Lancashire example and his rule for power from heating surface and grate (pp. 298-299) are lost to the cut-away scan.
- Steam density at pressure from these books: Wilson prints a saturated-steam table (pp. 302-307) and Fairbairn discusses steam density (pp. 306-313); neither was transcribed here, since the steam-engines note carries steam tables.
1. Evaporation: per sq ft of heating surface, per lb of coal, per horse power
| Figure | Original | SI | Book, page | Quote |
|---|---|---|---|---|
| Average maximum evaporation per sq ft of heating surface | 21 lb of water an hour, "one cubic foot of water evaporated by three square feet" | 102.5 kg/m2 h | Wilson 290 (PDF 303) | "We may take the average maximum evaporative effect of a square foot of heating surface at 21 lbs. of water per hour, or one cubic foot of water evaporated by three square feet of surface." |
| Same, ordinary tubular and externally fired boilers | 3 to 7 lb an hour (1 cu ft from 21 to 9 sq ft), 20 lb per sq ft of furnace surface to nil at the chimney end | 14.6 to 34 kg/m2 h; furnace 98 | Wilson 290 | "in ordinary tubular and externally fired boilers from 3 to 7 lbs., or 1 cubic foot from 21 to 9 square feet of heating surface, ranging from 20 lbs. per square foot of furnace surface to a few ounces or nil per square foot at the point where the gases quit the boiler" |
| Same, locomotive (highest average) | 13.5 lb an hour | 65.9 kg/m2 h | Wilson 290 | "the highest average value for the whole surface in the boiler is 13.5 lbs. of water from one foot of surface" |
| Cornish and Lancashire with water tubes (Galloway) | 8 lb water from 60 F per lb coal; 1 cu ft an hour from 16 sq ft; reckoned 8 sq ft per hp, 10 or 12 more common | 3.9 lb/sq ft h = 19 kg/m2 h (INFERRED: 62.4/16); 0.74 to 1.1 m2 per hp | Wilson 300 | "they have evaporated 8 lbs. of water from 60 deg. per 1 lb. of coal, and 1 cubic foot of water per hour from 16 square feet of heating surface, and may be reckoned at 8 square feet per H. P.; but 10 or 12 square feet per H. P. are more commonly required" |
| Water per nominal hp | 1 cu ft of water an hour | 28.3 kg/h per hp | Courtney 85 | "One cubic foot of water evaporated per hour is equivalent to one nominal horse-power." |
| Heating surface and grate per nominal hp, any boiler | 12 sq ft heating surface and 3/4 sq ft grate (fraction printed small, reads 3/4) | 1.11 m2 and 0.07 m2 | Courtney 87 | "Twelve square feet of heating-surface, and 3/4 square foot of fire-grate, per nominal horse-power, are very good proportions." |
| Boiler hp (trade standard) | 30 lb of feed water at 100 F into steam in 1 hour at 70 lb gauge | 13.6 kg/h | Hawkins 234 | "a horse power as applied to steam boilers is thirty (30) pounds of feed water at a temperature of 100 degrees Fahr. converted into steam in 1 hour at 70 pounds gauge pressure" |
| Heating surface per hp, horizontal tubular | 15 sq ft | 1.39 m2 | Hawkins 235 | "Find the square feet of heating surface in the shell, heads and tubes, and divide by 15 for the nominal horse power." |
| Same, boiler maker's table | "15 square feet has been allowed for each horse power" | 1.39 m2 | Hawkins 81 | "In estimating the horse power by means of the above table, 15 square feet has been allowed for each horse power" |
| Example tubular 54 in x 16 ft | 800 sq ft heating surface, 25 sq ft grate: 75 hp, 100 hp in good practice | 74 m2, 2.3 m2; 54 in = 1.37 m, 16 ft = 4.88 m | Hawkins 236 | "An ordinary 54 x 16' boiler with forty 3" tubes, 25 sq. ft. of grate surface and 800 sq. ft. of heating surface, in a general way is a 75 h. p. boiler, but good practice will get from it 100 h. p." |
| Water per lb of coal, theory | 14.62 lb coal; 12.83 lb "average coal" | 14.6, 12.8 kg per kg | Hawkins 18, 231 | "one lb. coal has an average evaporative capacity of 14.62 lbs. of water" ; "about 12.83 pounds of water" |
| Water per lb of coal, in practice | "little over half" of theory | about 7 to 8 kg per kg | Hawkins 18 | "In practice but little over half of these results are attained" |
| Water per lb of coal, Cornish worked example | 7.25 lb from 60 F (rate 12 lb coal per sq ft of grate an hour) | 7.25 kg per kg; 87 lb per sq ft grate h = 425 kg/m2 h | Wilson 300 | "taking the evaporation at 7.25 lbs. from 60 deg. per lb. of coal, we get 12 x 7.25 = 87 lbs. of water evaporated per square foot of grate per hour" |
Wilson's table of boilers in use, p. 296 (read on the page image). Heating surface per lb of fuel burned an hour, furnace efficiency, lb of water at 212 F per lb of coal ("total heat from 1 lb. of coal being taken at 14 lbs. of water evaporated from 212 deg."), lb of steam at 60 lb pressure from feed at 62 F. Right-hand column is the last. The "per sq ft per hour" column is mine (INFERRED: water per lb coal divided by sq ft per lb coal per hour; kg/m2 h = lb/sq ft h x 4.882).
| Sq ft per lb coal an hour | Efficiency | Water per lb coal, 212 F | Steam at 60 lb per lb coal | Water per sq ft an hour (INFERRED) | kg/m2 h |
|---|---|---|---|---|---|
| 0.50 (very small) | 0.40 | 5.6 | 4.7 | 11.2 | 55 |
| 0.75 (Cornish, Lancashire, egg-ended, multitubular) | 0.48 | 6.72 | 5.7 | 8.96 | 44 |
| 1.00 | 0.53 | 7.42 | 6.3 | 7.42 | 36 |
| 1.25 | 0.56 | 7.48 | 6.6 | 5.98 | 29 |
| 1.50 | 0.60 | 8.4 | 7.1 | 5.6 | 27 |
| 1.75 | 0.62 | 8.68 | 7.4 | 4.96 | 24 |
| 2.00 | 0.64 | 8.96 | 7.6 | 4.48 | 22 |
| 3.00 (water-tube and cellular) | 0.69 | 9.66 | 8.1 | 3.2 | 16 |
| 4.00 | 0.71 | 9.94 | 8.4 | 2.5 | 12 |
| 5.00 | 0.72 | 10.0 | 8.5 | 2.0 | 10 |
| 6.00 | 0.73 | 10.2 | 8.6 | 1.7 | 8 |
Wilson p. 296 on the table: "With a clean boiler, good coal, skilful firing, and introducing the feed water at a high temperature, the quantities in column four may be increased by from 10 to 30 per cent.; and on the other hand, with a dirty boiler and unskilful attendance they may be diminished from 65 to 20 per cent." (as printed). Feed water at 250 F or more from an economiser "with a corresponding saving in fuel".
Rate of combustion on the grate (lb of coal per sq ft of grate an hour):
| Boiler | Original | SI | Book, page | Quote |
|---|---|---|---|---|
| Cornish, lowest | 4 | 20 kg/m2 h | Wilson 272 | "Lowest rate of combustion in Cornish boilers 4" |
| Cornish, usual | 10 (Wilson's worked Cornish uses 12) | 49 (59) kg/m2 h | Wilson 272, 300 | "Usual rate in Cornish boilers 10" ; "The usual rate of combustion in Cornish boilers is about 12 lbs. of coal per square foot of grate area" |
| Externally and internally fired factory boilers | 10 to 18 | 49 to 88 | Wilson 272 | "Usual rate in externally fired and internally fired Factory boilers 10 to 18" |
| Marine | 14 to 26 | 68 to 127 | Wilson 272 | "Usual rate in Marine boilers 14 to 26" |
| Maximum, semi-bituminous coal | about 40; economy falls fast above 30 | 195; 146 | Wilson 272 | "the evaporative economy decreases rapidly with a combustion exceeding 30 lbs." |
| Coal and coke on one grate | at least 14.5 lb coal, 9.5 lb coke (OCR reads 14 1/2 and 9 1/2) | 71, 46 kg/m2 h | Hawkins 29 | "about 9 1/2 lbs. of coke should be burned per hour on each square foot of grate area, while at least 14 1/2 lbs. of coal can be burned upon the same space" |
| Grate to heating surface | not less than 1 : 40, may be 1 : 50 | - | Hawkins 175 | "The actual ratio of grate to heating surface should not in any case be less than 1 to 40, and may with advantage, in many cases, be 1 to 50." |
Note: the Cornish ratio 1 : 40 conflicts with Wilson's own Cornish (grate 12 3/4 sq ft; section 2 gives about 18 to 20 : 1) and Hawkins' own Lancashire (850 : 33 = 25.8 : 1). Hawkins' rule is for the tubular boilers of his day.
2. Cornish and Lancashire proportions
| Figure | Original | SI | Book, page | Quote |
|---|---|---|---|---|
| Cornish form | Cylindrical shell with flat ends and one internal furnace tube "a little greater than half" the shell diameter; gases run through the tube, return by two side flues, then the bottom flue to the chimney | - | Hawkins 54 | "a second cylinder B, having usually a diameter a little greater than half that of the boiler shell" |
| Cornish worked example (for a 16 in x 24 in cylinder, 60 rev/min, cut off at 1/4, 60 lb) | steam 5026 cu ft an hour, +25 per cent = 6282 cu ft = 1106 lb of water an hour at 60 lb; grate 12 3/4 sq ft (5 ft long x 2 ft 7 in); furnace tube about 2 ft 9 in; 6 in bottom water space, 2 ft 3 in crown to shell crown: shell 5 ft 6 in diameter, length four diameters, 22 ft | steam 142 m3/h; water 502 kg/h = 0.139 kg/s; grate 1.18 m2 (1.52 m x 0.79 m); tube 0.84 m; shell 1.68 m x 6.71 m | Wilson 300 | "we have a boiler 5' 6" diameter; and taking the length at four times the diameter, we shall have 22 feet as the length" |
| The same duty as a Cornish or a Lancashire | for a grate of about 20 sq ft Wilson advises a Lancashire of 7 ft diameter with two 2 ft 9 in tubes (length cut off in the OCR) instead of a Cornish 6 ft 0 in diameter with a 3 ft 6 in furnace tube and a grate 6 ft 0 in x 3 ft 4 in | Cornish 1.83 m diameter, tube 1.07 m, grate 1.83 m x 1.02 m | Wilson 300-301 | "instead of making the grate 6' 0" x 3' 4", and using a Cornish boiler 6' 0" diameter, with 3' 6" furnace tube" |
| Nominal hp of that boiler by rule | length x diameter / 6: 22 x 5.5 / 6 = 20.2 hp (INFERRED) | 20 hp; water 17.7 cu ft/h = 501 kg/h agrees with Wilson's 1106 lb | Courtney 85 | "multiply the length of the boiler by the diameter, in feet, and divide by 6: the quotient is the nominal horse-power" |
| Heating surface of that boiler | 12 sq ft per hp = about 212 sq ft; or Courtney's square-root rule S(sq yd) = hp^2 / grate = 26 sq yd = 235 sq ft (INFERRED) | 20 to 22 m2 | Courtney 85-86, 87 | "Or, 1 square yard of heating surface per nominal horse-power." ; "In tubes and flues the effective surface measured on the circumference is 1 1/2 times the diameter." |
| Capacity | 1 cubic yard of boiler capacity per nominal hp; steam room about 8 x the engine cylinder | 0.76 m3 per hp (INFERRED) ; 20 hp gives about 15 m3 | Courtney 86 | "Capacity of Boiler.--One cubic yard of boiler capacity per nominal horse-power." ; "Steam-room should be about eight times the contents of the cylinder of the engine supplied with steam by the boiler." |
| Grate area per hp | 1 sq ft of grate per nominal hp from the square-root rule; 3/4 sq ft in the general rule | 0.09 / 0.07 m2 | Courtney 86-87 | "Or, 1 square foot of fire-grate per nominal horse-power." |
| Mill boiler, Lancashire, 1860 | about 30 ft long, 7 ft diameter, two internal flues about 2 ft 8 in diameter, shell and flues of 3/8 in plate | 9.14 m x 2.13 m, flues 0.81 m | Fairbairn 301 | "about 30 feet in length, 7 feet in diameter, with two internal flues a a, about 2 feet 8 inches in diameter. These flues as well as the external shell are generally composed of 3/8 inch plates" |
| Lancashire steam space | 7 ft x 30 ft: steam space about 240 cu ft | 6.8 m3 | Wilson 148 | "a Lancashire boiler 7 feet by 30 feet, the steam space would be about 240 cubic feet" |
| Lancashire, 1890s | usual length 27 ft (21 ft short); furnace tubes at least 33 in; shell not under 7 ft; flat ends | 8.23 m (6.4 m); 0.84 m; 2.13 m | Hawkins 58 | "The usual length of these boilers is 27 feet, though they are occasionally made as short as 21 feet. The minimum diameter of the furnaces is 33 inches, and in order to contain these comfortably the diameter of the boiler should not be less than 7 feet." |
| Lancashire data of Fig. 18 | 75 lb; 27 ft x 7 ft; heating surface 450 (furnace and flues) + 30 (Galloway tubes) + 370 (external flues) = 850 sq ft; furnace 33 in; grate 33 sq ft; weight about 15 1/2 tons | 0.52 MPa, 5.1 atm; 79 m2; 3.07 m2; 15.7 t | Hawkins 60 | "Steam pressure, 75 lbs. per sq. inch. Length, 27 feet. Diameter, 7 feet ... Grate area, 33 sq. feet." (plate thicknesses unreadable in the OCR) |
| Boiler setting with external flues | gases go out along the bottom, return up the side flues | - | Hawkins 59 | "three turns of about 27 feet each" |
| Wigan trials 1868 | Galloway no better than plain Lancashire; over 9 lb water from 100 F per lb coal | 9 kg/kg | Wilson 294 | "a high rate of evaporative efficiency was maintained, being above 9 lb. of water from 100 deg. per lb. of coal" |
INFERRED from the rows above: a small mill Cornish of the 1860s-70s is a 5.5 ft x 22 ft shell with a 2.75 ft furnace tube, 12 sq ft of grate, about 215-235 sq ft of heating surface, passing 500 kg of water an hour (about 0.14 kg/s) at 60 lb (4.1 atm gauge). That is 20 nominal hp. A boiler a third that size scales on the same rules: 7 nominal hp is length x diameter = 42 (e.g. 4.5 ft x 9.3 ft), 4 sq ft of grate to 1 sq ft per hp, about 84 sq ft of heating surface, 7 cu ft (200 kg) of water an hour.
Flue strength (why the tube is half the shell and why long flues carry rings). Fairbairn p. 301: the 7 ft boiler above bursts at 303.5 lb "which of course ought at the same time to be the collapsing pressure of the flues"; "the flues would give way with about 100 lbs. pressure per square inch, or the boiler would be destroyed by collapse at one-third of the bursting pressure of the shell". Collapse law p. 304 (P in lb per sq in, L ft, D in; Wilson prints the same constant, p. 313): "P = 806,300 x t^2.19 / (L D)". Fairbairn p. 305 proposes angle-iron or T-iron rings; a flue sized for collapse at 450 lb: "For a collapsing pressure of 450 lbs. the safe working pressure would be 75 lbs. per square inch." Wilson pp. 18-19: a Cornish of 3/8 in plate with a tube half the shell is equally strong in shell and tube at 11.5 ft between rings (INFERRED from his formula; his text: "the boiler in this case should not exceed 11 1/2 feet in length to be equally strong in shell and tube").
3. Working pressures by period
| Period and boiler | Original | SI (gauge) | Book, page | Quote |
|---|---|---|---|---|
| Watt's time, wagon boiler | seldom over 10 or 12 lb | 0.07-0.08 MPa; 0.7-0.8 atm | Fairbairn 299 | "the pressure in the boiler seldom exceeded 10 or 12 lbs. per inch" |
| Fairbairn's 1860 view | "from 10 lbs. to 50 lbs., and even in some cases to 150 lbs." | 0.07 to 0.34 MPa (1.0 to 3.4 atm), some to 1.0 MPa | Fairbairn 1-2 (PDF 26-27) | "has caused a general increase of the working pressure from 10 lbs. to 50 lbs., and even in some cases to 150 lbs. on the square inch" |
| Mill beam engine, 1850s (comparison for Fairbairn's Lecture IV) | 20 lb in the boiler, 100 nominal hp (300 actual), cylinder 50 in x 7 ft | 0.14 MPa; 1.4 atm | Fairbairn 233 | "Pressure of steam in boiler . . . 20 lbs. per square inch." |
| Fairbairn's flue rule | safe working 75 lb | 0.52 MPa; 5.1 atm | Fairbairn 305 | see section 2 |
| Factory Cornish, 1873 (Wilson's worked example) | 60 lb | 0.41 MPa; 4.1 atm | Wilson 300 | "working at 60 lbs. pressure" (the engine of the example; the boiler it feeds is the 5 ft 6 in Cornish of section 2) |
| Cast-pipe sectional boiler | 80 or 90 lb | 0.55-0.62 MPa; 5.4-6.1 atm | Wilson 27 | "vertical cast-iron pipes 4 inches or 6 inches in diameter are at present employed and worked at pressures as high as 80 lbs. or 90 lbs." |
| Lancashire, 1890s | 75 lb | 0.52 MPa; 5.1 atm | Hawkins 60 | see section 2 |
| Marine, older | up to 35 lb, flat plates | 0.24 MPa; 2.4 atm | Hawkins 60 | "up to 35 lbs. per square inch, is usually made of flat plates stayed together" |
| Marine, medium high | 60 to 150 lb | 0.41 to 1.03 MPa | Hawkins 60 | "constructed for steam of 60 to 150 lbs. per square inch" |
| Single-cylinder cut-off engine ceiling | 80 lb ("not good economy to carry pressure higher") | 0.55 MPa; 5.4 atm | Hawkins 235 | "it is not good economy to carry pressure higher than eighty pounds in single cylinder automatic cut off engines" |
| Compound engine | 80 to 120 lb | 0.55-0.83 MPa | Hawkins 235 | "For pressure between eighty and one hundred and twenty pounds, the compound engine gives the best results" |
No table by decade exists in these four books. Cornish-and-Lancashire pressures run 10-12 lb (1800), 20 lb (1850s mill engine), 50 lb the general rise by 1860, 60 lb usual 1873, 75 lb 1890s (INFERRED stringing of the rows above).
4. Plate strength, riveted seam efficiency, working-pressure rule
| Figure | Original | SI | Book, page | Quote |
|---|---|---|---|---|
| Boiler plate tensile strength, best | 24 tons per sq in; "Best Best" 21 tons; Yorkshire 25; Staffordshire crown 20; Glasgow best best 24 | 371, 324, 386, 309, 371 MPa | Wilson 37 (PDF 52), Courtney 92 | "All plates of the very best quality having a longitudinal tenacity of 24 tons per square inch" ; "Best Yorkshire plate, 25 tons. Crown, or best Staffordshire plate, 20 tons." |
| Fairbairn's plate means | Yorkshire 22.8 to 25.8 tons; Staffordshire 19.6 (along) to 21.0 (across); mean 22.5 and 23.0 | 347 and 355 MPa mean | Fairbairn 113 (PDF 140) | "Mean ... 22.519 23.037" tons per square inch, along and across the fibre |
| Plate strength vs temperature | not impaired up to 340 F, probably to 500-600 F; "20 1/2 tons" | 0 to 170 C | Fairbairn 108 | "the iron is in no degree injured by a temperature ranging from zero up to 340 deg." |
| Wilson standard plate | 21 tons per sq in along the fibre; 10 to 15 per cent lower across; add 20 per cent if plate proved 25 tons | 324 MPa | Wilson 73 | "Taking 21 tons per square inch as our standard strength for plates along the fibre" |
| Riveted seam strength relative to solid plate | double riveted 70, single riveted 56 (entire plate 100); on a continuous long seam a single-riveted shell 40 per cent | 0.70, 0.56; 14.7 and 11.5 tons per sq in | Wilson 73 | "Entire plate 100. Double-riveted joint 70. Single ,, ,, 56. These proportions of strength are usually employed, and were originally deduced by Sir W. Fairbairn from his experiments." |
| Same in Courtney | double 0.7, single 0.56 | same | Courtney 92 | "Multiply the tensile strength of the plate by 0.7; the product is the strength of the joints in double-riveted boilers. For single-riveted joints multiply by 0.56" |
| Working strain vs joint strength | "not more than 1/5" (the fraction prints as a small one-fifth; the printed rules' constants fit 1/5 exactly, section 4 below) | factor of safety 5 | Courtney 92 | "The working strain should not be more than 1/5 of the strength of the joint." |
| Wilson's factors | working 6 and testing 3 "usually"; 5 "or even somewhat less" for new boilers examined every 6-12 months; 6 or 8 for boilers not well known; test limit 1/3 of joint strength; old boilers 1/4 | 6, 3 | Wilson 213, 216 | "usually 6 and 3, would give the working and testing pressures respectively" ; "a factor of safety of 5 or even somewhat less, and for those whose condition is not well known a factor of 6 or 8" |
| Elastic limit of plate | 10 tons per sq in net section, "about one half the breaking weight" | 154 MPa | Wilson 214-215 | "the limit of elasticity for boiler plates may be safely taken at 10 tons per square inch of nett section" |
| Wilson's bursting rule | P = c x T / D, P in lb per sq in, T in sixteenths of an inch, D in quarter feet; c = 1097 single riveted, 1372 double riveted wrought iron (1728 and 2156 steel) | - | Wilson 310 | "P = T/D x c ... 1097 for single riveting, 1372 for double riveting, Wrought iron" |
| Courtney's working pressure (iron) | P = 2 t x 7,800 / D (Yorkshire, double riveted); 6,200 single; Staffordshire 6,200 double, 5,000 single; ordinary iron 3,700 double, 3,000 single; t and D in inches, P in lb per sq in | - | Courtney 93 | "Multiply twice the thickness of the plate in inches by 7,800 for double-riveted joints, or by 6,200 for single-riveted joints, and divide by the diameter of the boiler in inches." |
| Courtney's plate thickness | thickness = P x D / 15,600 (Yorkshire double), 12,400 single; Staffordshire 12,400 / 10,000; ordinary 7,400 / 6,000 | - | Courtney 93 | "Multiply the given pressure by the given diameter in inches, and divide by 15,600, for the thickness in inches ... of plate double-riveted; or, divide by 12,400 for plate single-riveted." |
| Hawkins (US inspectors' rule, steel) | pressure = (1/6 x lowest tensile strength x plate thickness) / radius, single riveted; add 20 per cent for double riveting; hydraulic test 150 lb to 100 lb of working | factor of safety 6 | Hawkins 117 | "Multiply one-sixth of the lowest tensile strength found stamped on any plate in the cylindrical shell by the thickness ... and divide by the radius or half diameter ... for single riveting, to which add twenty per centum for double riveting." |
| Steel plate (Hawkins) | tensile 55,000-65,000 lb per sq in; elongation 20 to 25 per cent | 379-448 MPa | Hawkins 90 | "Tensile strength, 55,000 to 65,000 pounds per square inch." |
| Thinnest plate | 1/4 in | 6.35 mm | Hawkins 94 | "The thinnest plate which should be used in a boiler is one-fourth of an inch" |
| Ring seam load | a circle seam has half the strain of a long seam | - | Hawkins 93 | "a circle seam has just one-half the strain to carry as a longitudinal seam" |
| Wilson on collapse of a flue (factor) | blowing-off pressure factor not under 4 against the collapsing pressure, 6 in new boilers whose pressure is later raised | 4 to 6 | Wilson 313 | "This should in no case be less than 4, and in new boilers, in which the pressure is so often after a time increased, a factor of not less than 6 should always be allowed." |
INFERRED worked checks:
- The Courtney constants are the 5-factor rule: 25 tons x 2240 x 0.7 / 5 = 7,840, printed 7,800; with the 0.56 single seam 25 x 2240 x 0.56 / 5 = 6,272, printed 6,200; Staffordshire 20 tons gives 6,272 and 5,018, printed 6,200 and 5,000; ordinary iron 3,700 and 3,000 fit about 12 tons.
- Boiler plate 5/16 in, shell 6 ft, double riveted, Wilson's formula: 5 / 24 x 1372 = 286 lb (1.97 MPa, 19.4 atm) to burst; Courtney's Yorkshire rule gives a working pressure of 2 x 0.3125 x 7800 / 72 = 68 lb (0.47 MPa, 4.6 atm); Staffordshire 54 lb (0.37 MPa, 3.7 atm); ordinary iron 32 lb (0.22 MPa, 2.2 atm); Wilson at factor 6 gives 48 lb (0.33 MPa, 3.2 atm). Single-riveted: Wilson burst 229 lb, factor 6 gives 38 lb (0.26 MPa, 2.6 atm).
5. Safety valve
| Figure | Original | SI | Book, page | Quote |
|---|---|---|---|---|
| US steamboat law, lever valve | 1 sq in of valve per 2 sq ft of grate | 6.45 cm2 of valve per 0.186 m2 of grate | Hawkins 189 | "Lever safety valves to be attached to marine boilers shall have an area of not less than one square inch to two square feet of the grate surface" |
| Spring-loaded valve with lift, same law | 1 sq in per 3 sq ft of grate; lift not under 1/8 of the opening's diameter; seat 45 degrees | 6.45 cm2 per 0.279 m2 of grate | Hawkins 189-190 | "an area of not less than one square inch to three square feet of grate surface of the boiler ... a distance of not less than that equal to one-eighth the diameter of the valve opening" |
| "Pop" valve table | 2 in, 2 1/2 in, 3 in, 4 in, 5 in, 6 in for 9.42, 14.73, 21.20, 37.69, 58.90, 84.82 sq ft of grate | 51, 64, 76, 102, 127, 152 mm for 0.875, 1.37, 1.97, 3.50, 5.47, 7.88 m2 | Hawkins 190 | "2 inch 'Pop' Valve for 9.42 square feet of grate surface." (the 2 1/2 in row reads "3i" in the OCR and is 14.73 / 3 = 4.91 sq in, which is 2 1/2 in; INFERRED, every row is exactly 3 sq ft of grate per sq in of valve) |
| Rankine's rule | valve area, sq in = 0.006 x lb of water evaporated an hour | 0.006 sq in per lb/h = 1.32 mm2 per kg/h | Hawkins 190 | "Multiply the number of pounds of water evaporated per hour by .006, and the product will be the area in square inches of the valve." |
| US Navy rule | 0.005 per lb of water an hour | 1.1 mm2 per kg/h | Hawkins 190 | "Multiply the number of pounds of water evaporated per hour by .005" |
| Philadelphia (and French) rule | area, sq in = 22.5 x grate sq ft / (working pressure lb + 8.62) | - | Hawkins 190 | "Multiply the area of grate in square feet by the number 22.5. Add the number 8.62 to the pressure allowed per square inch. Divide (1) by (2)" |
| Largest valve | 4 in diameter; beyond that use more valves | 102 mm | Hawkins 191; Wilson 121 | "The maximum desirable diameter for safety valves is four inches" ; "it is better to increase the number than the size of the valves when the diameter is as much as five inches" |
| No stop valve between boiler and safety valve | - | - | Hawkins 191 | "There should not be any stop valve between the boiler and safety valve." |
| Dead-weight valve load | 4 in valve at 100 lb needs over 1,200 lb of weight | 102 mm, 0.69 MPa: 5.5 kN | Hawkins 189 | "a four-inch valve, intended to blow off at 100 lbs. per square inch would require weight of over 1,200 lbs." (12.57 sq in x 100 = 1,257 lb; INFERRED check) |
| Lever valve rule | load = (ball x lever length / valve-to-fulcrum distance + valve and lever weight) / area; worked: 8 sq in, 120 lb | - | Hawkins 192; Courtney 88-90 | "to the quotient add the effective weight of the valve and lever, and divide the sum by the steam pressure." |
| Valve rise above blow-off | disc valve with wide seat: 12 to 20 lb above a 60 or 70 lb setting | +0.08 to +0.14 MPa | Wilson 122 | "as much as from 12 to 20 lbs. above the initial blowing-off pressure of 60 lbs. or 70 lbs." |
INFERRED safety-valve sizing for the 20-hp Cornish of section 2: 12 sq ft of grate at 1 sq in per 3 sq ft = 4 sq in, a 2.3 in valve; at 1 sq in per 2 sq ft, 6 sq in, a 2.8 in valve; Rankine's rule on 1,106 lb of water an hour, 6.6 sq in, 2.9 in; Philadelphia rule at 60 lb, 22.5 x 12 / 68.6 = 3.9 sq in, 2.2 in. The four rules agree on a 2 1/4 to 3 in valve.
6. Water and steam stored
| Figure | Original | SI | Book, page | Quote |
|---|---|---|---|---|
| Steam cu ft per cu ft of water, 60 lb | 1 cu ft of steam at 60 lb is 353 times more bulky than the water it came from | 353 | Wilson 300 | "at 60 lbs. pressure 1 cubic foot of steam is 353 times more bulky than the water from which it is raised" |
| Steam produced by 1 cu ft of boiler water on release | 60 lb: nearly 36 times 1 cu ft of 60 lb steam (about 162 cu ft at atmosphere) | - | Wilson 148 | "1 cubic foot of water in the boiler will produce nearly 36 times that amount" |
| Steam space | about 100 x the average volume of steam generated per second (engine volume and revolutions set it) | - | Hawkins 89 | "it may be made a hundred times as great as the average volume of steam generated per second" |
| Heat in steam | total heat of steam at 60 lb 1207 Btu per lb from 0 F; evaporating from 60 F needs 1147 | 2.81 MJ per kg | Wilson 307 | "the steam will have a temperature of 307 deg., and a total heat of 1207" (from the printed table; OCR, not image-checked) |
7. Pump and injector (Hawkins pp. 206-211)
| Figure | Original | SI | Page | Quote |
|---|---|---|---|---|
| Two feeds on every boiler, one an injector | - | - | 206 | "every boiler ought to have two feeds in order to avoid accidents when one of them gets out of order, and one of these should be an injector." |
| Steam jet speed | about 1,700 ft per second into the atmosphere at 60 lb | 518 m/s | 206 | "The rate of speed at which steam - taking it at an average boiler pressure of sixty pounds - travels when discharged into the atmosphere, is about 1,700 feet per second." |
| Condensation | the bulk of the steam falls about 1,000 times; speed is kept | 1/1000 | 206 | "its bulk will be reduced say 1,000 times, but its velocity remains practically undiminished" |
| Heat given up by each lb of steam | about 900 thermal units; 778 ft-lb per unit | 900 Btu = 950 kJ per lb = 2.09 MJ/kg | 210 | "The energy so given up by each pound of steam amounts to about 900 thermal units, each of which is equivalent to a mechanical force of 778 foot pounds." |
| Thermal efficiency | nearly 100 per cent; mechanical efficiency "very low" | ~1.0 | 208 | "As a thermodynamical machine, the injector is nearly perfect ... its thermal efficiency should be in every case nearly 100 per cent." |
| Lift | "will raise water about 25 feet" from a well or tank below the boiler | 7.6 m | 211 | "will raise water about 25 feet if required, from a well or tank below the boiler level" (checked on the page image) |
| Exhaust injector | needs exhaust warmed above 212 F; carries cylinder oil into the boiler | - | 209-210 | "such injectors work well when a little live steam is admitted into the exhaust sufficient to heat it above a temperature of 212" |
| Feed temperature | high steam pressure needs cold water; hot feed stops it; throttle the steam to a lower pressure | - | 210 | "for high pressure of steam good results can only be obtained with cold water" |
| Not economical | "The injector is not an economical device, but it is simple and convenient" | - | 211 | quoted |
| Deposit | lime stops the jets; clean with 1 part muriatic acid to 9-12 parts water | - | 211 | "use a solution of one part muriatic acid to from nine to twelve parts of water" |
Injector steam to water ratio: not given. INFERRED heat balance (steam 60 lb gauge total heat about 1,177 Btu, water in at 60 F, delivery at an assumed 150 F): about 1,177 - 118 = 1,059 Btu of heat per lb of steam to heat 90 F rise per lb of water, so about 11.8 lb of water per lb of steam. The delivery temperature is my assumption, not the book's. Hawkins' own 900 Btu per lb figure gives 10 lb of water per lb of steam at the same rise.
8. Cast-iron pipe
| Figure | Original | SI | Page | Quote |
|---|---|---|---|---|
| Cast iron tensile strength | about 6 tons per sq in | 93 MPa | Wilson 25-26 | "The tensile strength of cast iron being about 6 tons" (OCR from a cut page; figure partly read) |
| Cast iron for steam pipes at full boiler pressure | cast iron is "almost invariably used, except with marine boilers, for pipes to carry the steam to the engine cylinders at the full boiler pressure" | - | Wilson 27 | "cast iron is almost invariably used, except with marine boilers, for pipes to carry the steam to the engine cylinders at the full boiler pressure" |
| Why not for the boiler | brittle and treacherous: small hidden flaw breaks it without warning | - | Wilson 26 | "a very slight hidden or surface defect, in an unsound casting, is sometimes sufficient to lead to a sudden explosive fracture" (the page is clipped; read on the image) |
| Cast-pipe sectional boilers | 4 in or 6 in vertical pipes, worked at 80 or 90 lb | 100-150 mm; 5.4-6.1 atm | Wilson 27 | see section 3 |
| Flange joint, old | iron borings caulked in the hub and spigot or a ring gasket of wrought iron wrapped in yarn dipped in red and white lead; flanges later faced and turned with rubber, copper, paper or cement gaskets | - | Hawkins 248-249 | "The cast-iron hub and spigot joint, Fig. 115, caulked with iron borings, is probably the oldest kind of joint." ; "the facing or turning up of the flanges and the use of a gasket of rubber, copper, paper or cement" |
| Hawkins on cast pipe use | wrought-iron welded pipe has taken over heating and small piping | - | Hawkins 249 | "The introduction of wrought iron welded pipes has diminished the use of cast-iron pipes for many purposes" |
A working pressure for cast pipe is not documented. INFERRED from Wilson's cast-iron strength alone: a 4 in bore, 3/8 in wall cast pipe at 6 tons per sq in bursts at 2 t S / D = 2 x 0.375 x 13,440 / 4 = 2,520 lb (17.4 MPa); at Wilson's factor 6, 420 lb (2.9 MPa, 28.6 atm). The wall is therefore not what limits a cast pipe; the flange bolts, the joint and casting flaws are, and no book here gives a figure for them.
9. Efficiency and loss
| Figure | Original | SI | Book, page | Quote |
|---|---|---|---|---|
| Furnace-and-boiler efficiency of boilers in use | 0.40 (very small) to 0.64 (Cornish and Lancashire at 2 sq ft per lb coal an hour); 0.69-0.73 water-tube; Cornish and Lancashire at 1 sq ft per lb: 0.53 | 40-73 per cent | Wilson 296 | "Efficiency of furnace" column of the table in section 1 |
| Efficiency formula | E' = E (S / (S + A F)) x B, B = 4/5 for chimney draught ("20 per cent" miscellaneous loss) | - | Wilson 295 | "B is a fractional multiplier to allow for miscellaneous losses of heat, which, for chimney draught, is here taken at 20 per cent." |
| Hawkins | "little over half" of theory | 50-55 per cent | Hawkins 18 | quoted above |
| Hawkins, heat actually used | heat usable is the difference of fire and water temperature: furnace at 975 F gives "one-half", at 2,600 F "three-fourths" (fractions OCR-garbled) | - | Hawkins 175 | "the grate should be of such a size as to have the fire burn rapidly" |
| Scale | 1/16 in scale 13 per cent more fuel; 1/8 in 33 per cent; 1/2 in 60 per cent | - | Hawkins 143 | "1-16 inch of scale causes a loss of 13 per cent. of fuel" |
| Setting | the setting alone varies fuel use by as much as one-fifth | 20 per cent | Hawkins 236 | "will vary the quantity of fuel used by as much as one-fifth" |
| Feed-water heater size | 10 sq ft of heating surface in the heater per hp, to raise the temperature of the feed | 0.93 m2 per hp | Hawkins 199, 201 | "Ten square feet of heating surface in the feed water heater also represents one horse power." |
| Steam main loss | 5 per cent of the main's capacity lost by condensation in one mile; 75 lb carried 5 miles of mains falls 2 lb at a point half a mile out (OCR "two", as printed) | 1.6 km; 0.52 MPa | Hawkins 259 | "Interesting experiments show that the loss by condensation in carrying steam one mile is 5 per cent. of the capacity of the main" |
| Coverings, relative value | wool felt 1.000, mineral wool 0.832, sawdust 0.680, charcoal 0.633, asbestos 0.363, coal ashes 0.345, coke in lumps 0.277, undivided air space 0.136 (Emery) | - | Hawkins 270 | "Table of Relative Value of Non-Conductors ... Wool Felt 1.000 ... Asbestos .363" |
| Covering recipe | 1 in of sawdust and starch paste "will produce the same effect as that of the most costly non-conductors"; 1/5 in coats; felt chars at high pressure | 25 mm | Hawkins 269-270 | "a thickness of 1 inch will produce the same effect as that of the most costly non-conductors" |
| Heat loss per sq ft of pipe | none | - | - | not documented |
Hawkins' mile-of-main figure is the only number for a steam line's loss. It is about 5 per cent per mile of capacity, so the loss in a short mill run is a few tenths of a percent a minute of steam. (INFERRED reading; the book gives no time basis and does not say whether the main was clothed.)
10. What the books do not give
- Injector steam to water ratio (Hawkins 206-211 gives speed, heat, lift of 25 ft and a 60 lb example, no ratio); the INFERRED balance in section 7 is mine and rests on an assumed delivery temperature.
- Heat loss from bare or clothed pipe in Btu per sq ft per hour or W per m2 K; Hawkins' relative values and the 5-per-cent-a-mile line are all.
- A cast-iron pipe's working or bursting pressure; Wilson gives the material's strength only and a sectional boiler of cast pipe.
- A Cornish boiler's heating surface and water volume as printed figures; derived in section 2 from Courtney's and Wilson's rules.
- A working-pressure table by decade; the points are in section 3.
- The Lancashire example and power rule on Wilson pp. 298-299 (page torn from the scan).
- Stack and draught effects on evaporation; Wilson's chapter XIII-XIV on combustion and draught were not read here.
- Farey's steam-pipe rule (p. 590); it is in the steam-engines note.
- Steam density tables at pressure (Wilson pp. 302-307 hold a table; Fairbairn pp. 306-313 argue density), not transcribed in these notes.
11. Disagreements between the books
- Water per lb of coal in a Cornish: Wilson's examples 7.25 lb from 60 F (1873, Welsh or semi-bituminous), Hawkins' theoretical 14.62 and "little over half" in practice (1897-1903, anthracite and bituminous US coals). They agree on practice (7.3 against about 7.5 to 8 by Hawkins' half). Hawkins is newer.
- Factor of safety: Courtney (1880) 5; Wilson (1873) 5 to 8 with 6 usual; Hawkins (1897) 6 with a 1.5 hydraulic test. Wilson's own working rule for boilers examined at intervals is "5 or even somewhat less".
- Heating surface per hp: Courtney 12 (and 1 sq yd = 9 sq ft per hp from his square-root rule); Wilson 8 to 12; Hawkins 15. The newer book's hp is a 30 lb steam hp at 70 lb gauge against a 1 cu ft (62 lb) nominal hp, so the two are not the same unit (INFERRED reading).
- Grate to heating surface: Hawkins 1 : 40 to 1 : 50 against Wilson's worked Cornish near 1 : 18 to 1 : 20. Hawkins is for tubular boilers; his own Lancashire is 1 : 26.