Atchison, Topeka & Santa Fe 2-10-2 "Santa Fe" Locomotives in the USA

Class 1600 (Locobase 11184)

Data from AT&SF 7 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. See also data from table in July 1904 AERJ. DeGolyer, Volumes 28, pp. 31+, and 29, pp. 6+; and E D Worley, Iron Horses of the Santa Fe (Dallas: Southwest Railroad Historical Society, 1976). More extensive information on the Neely furnace in Walter Bohnstengel, "Fuel Oil Burning Apparatus on the AT&SF Ry.", Railway Review, Volume 66, No. 24 (12 June 1920), pp. 1009. (Thanks to Chris Hohl for his 9 March 2022 correcting the balance between coal burners and oil burners.)

The 1600s were built in 1905-1907 and had works numbers as follows:

1905

July 26073, 26085, 26086, 26111, 26117

August 26152, 26166, 26183, 26229, 26252, 26259, 26291-26292, 26313, 26326, 26344, 26351-

26352

September 26386-26387, 26443, 26450, 26459-26460, 26472, 26490, 26499

October 26528-26529, 26554, 26582

November 26761, 26847, 26854, 26873, 26920, 26935, 26949

December 27046-27047

1906

March 27636-27637, 27705, 27712, 27745, 27763, 27795, 27807, 27830

April 27863, 28021, 28032

May 28273, 28280

June 28433, 28497

December 29647, 29694-29695, 29728, 29751-29752, 29781-29782, 29819, 29841-29842, 29886

1907

January 29982, 30012, 30034-30036; February 30151

This class was nearly identical to the 900s shown in Locobase 463, but put about a ton more weight on the drivers and were delivered as a mix of oil- and coal-burning locomotives. The oil-burners were 1620-1639 and 1656-1673; the coal-burners--1600-1619, 1646-1655--which pulled tenders filled with 7,000 US gallons (26,495 litres) of water and 14 tons (12.7 metric tons) of coal were the rest. Oil burners used what would be the Santa Fe-developed back-mounted Booth burner--this design was used for years on all Santa Fe oil-burning locomotives.

A copy of an 18 January 1904 letter from the Santa Fe's A Lovell to Baldwin preserved in the later spec reported that Lovell had visited and left a sketch showing how he wanted the builder to reinforce the frame. Engine 939 of the previous class had suffered a break and he wanted to forestall a similar occurence in the 1600s.

His annoyance at finding Baldwin's frame card #5068 unchanged can be imagined, particularly as engine 1604 of the present class had broken its right frame in the same places. He now requested that the front frame's top rail be deepened from 5" to 6 1/2" (127 to 165 mm), the bottom member from 4" to 4 1/2" (102 to 115 mm), and the bottom front projection of the main frame from 4 11/16" to 5 7/16" (119 to 138 mm). A handwritten note entered on the copy reported that the bottom member's depth beginning with 1657 was increased to 7 1/4" (164 mm) beginning with 1652. Locobase supposes this significant thickening resulted from Baldwin's identification of the specific weakness in the frame design.

According to the Topeka State Journal of 4 Feb 1907, preserved in the clipping collection offered by Frank Ellington on [] (visited 23 July 2004), "The firebox contains clay balls about 10 inches in diameter. The oil is turned among these and lighted. The clay balls become intensely hot, giving a much more even heat than coal. "

Locobase had not heard of this aid to combustion, and knew no more until he examined the original specs in the DeGolyer volume, which contained the admonition: "Special care taken in drilling tell-tale holes in clay balls so they will be parallel to axis." Walter Bohnstengel later the results of using the "Neely furnace", as this layout was called. The clay balls did indeed "give a desirable flame distribution, but the arrangement was not rigid enough to withstand locomotive service."

Worley commented (p. 205) that the tandem-compounds' appetite for expensive maintenance led the Santa Fe mechanical department ("tighter than Ebenzer Scrooge when it came to..a steady reduction in labor costs per locomotive service year") adopted "hard-fisted tactics" including the "heavy employment of low cost ($1.80 per day vs $3.50 per day) and high output Japanese machinists, particularly in its California based repair shops."

All were later rebuilt as two-cylinder simples with 28 x 32-in cylinder dimensions, reduced boiler pressure (to 200 lb psi), and a resulting tractive effort of 74,800 lb See Locobase 8256.

Notice that Locobase 8249 shows a set of Mikados delivered at the same with a four-cylinder-in-plane layout that shared many components with this set of Santa Fes. It's not clear who borrowed from whom, but the Mike's boiler had 62 more small tubes than did these SFs, which were delivered in much greater quantity.

Class 1674 (Locobase 8275)

Data from AT&SF 7 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. See also DeGolyer, Volume 46, pp. 188+ and E D Worley, Iron Horses of the Santa Fe (Dallas: Southwest Railroad Historical Society, 1976). Information about how many, when, and by whom this class was supplied from [] (also visited 6 February 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema. (Thanks to Chris Hohl for his 13 April 2019 email noting the original tender capacity of 12 tons.) Works numbers were:

1912

June 37955-37958, 37976-37980; July 38059-38063; August 38133-38136, 38190-38191

1913

March 39432-39436, 39505-39510

Soon after the Santa Fe took delivery of its tandem-compound 1600-class 2-10-2s, it procured these simple-expansion engine. The grate stayed the same, but the boiler was superheated from the start. Worley commented (p. 205) that their construction "commemorated the true beginning of the modern era of Santa Fe motive power policy." Shifting from Kendrick's aggressive exploration of alternate types of power and arrangements, John Purcell's 30-year adminstration as Chief Mechanical Engineer "was marked by conservative design and simple cylinders."

As delivered, they had four arch tubes that contributed 29 sq ft (2.69 sq m) to the firebox heating surface. Their boilers were set to 170 psi (11.7 bar), but soon they were reset to the 200 psi shown in the specs. In later years, weight on the drivers rose to 258,420 lb (117,217 kg) while total engine weight ballooned to 312,060 lb (141,548 kg).

A later coal tender carried 14 tons of fuel and weighed 185,400 lb (84,096 kg) loaded. Oil burners 1674-1693 trailed tenders weighing the same and carrying the same amount of water as the coal carriers, but the fuel load topped out at 3,300 US gallons (12,491 litres).

1697, 1701, 1704 were sold for scrap in late 1940. All of the others operated throughout World War Two. The rest were sent to the ferro-knacker in a steady parade of small numbers from 1949 through 1956. The last set--1683, 1688, 1692, 1694 and 1698--all were sold for scrap in June 1956.

Class 3010 (Locobase 5389)

Data from table in May 1916 RME. Data from AT&SF 9 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. Information about how many, when, and by whom this class was supplied from [] (accessed 6 Feb 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema.

These ten engines emerged as the result of disassembling the ten 2-10-10-2s that the Santa Fe tried out in 1911 (Locobase 418). The rear units had the lower numbers. Note that the conversion was to a simple-expansion design.

In their new guise, the daughter engines gave decades more service. E D Worley's extensive roster coverage included 3015 photographed without its leading truck. "Whether this alteration was effected by design or via cannibalism for spare parts is not known;", Worley commented,"however, proof is here that an 0-10-2 existed."

Retirements began in 1939-1940 and continued until 1950.

Class 3020 (Locobase 70)

Data from AT&SF 9 -1920 and 6 - 1927 locomotive diagrams supplied in May 2005 and August 2013 by Allen Stanley from his extensive Rail Data Exchange. Information about how many, when, and by whom this class was supplied from [] (accessed 6 Feb 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema. (Many thanks to Chris Hohl for his 22 September 2017 email reporting unlikely boiler pressure values for 177 entries. A Locobase macro caused the error .)

When the Santa Fe conceded defeat in its attempt to develop a 2-10-10-2 that would operate satisfactorily (Locobase 418), it broke the articulateds apart, placing the rear boiler and engines in class 3010 (Locobase 5389) and turning the low-pressure front sections into the 3020 class. burned oil and her tender carried 3,160 gallons of oil. She also had four Nicholson arch tubes that contributed 27 sq ft to the firebox heating surface. 3020 apparently was renumbered 3291.

Class 3800 (Locobase 71)

Data from AT&SF 9 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. All of Drawing Room Specification Book Number 61 of the DeGolyer Library's BaldwinRail Data Exchange is devoted to Santa Fe locomotive orders in the late 19teens and 1920s. The 3800 class specs run from page 339 to 366; see also Volume 70, pp. 19+. See also "Three Types of Locomotives for the Atchison, Topeka & Santa Fe Railway System," Railway & Locomotive Engineering, (December 1919), pp. 363-365. (Many thanks to Chris Hohl for his 31 December 2016 email and spreadsheet that led Locobase to a typo in the road numbers. His spreadsheet also underscored the bewildering variety of pressures, tube counts, weights, fuel types ultimately associated with the class. A 22 September 2017 email reported unlikely boiler pressure values for 177 entries. A Locobase macro caused the error .) )

Information about how many, when, and by whom this class was supplied from [] (accessed 6 Feb 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema.

Also see "Atchison, Topeka and Santa Fe steam engine diagrams and blueprints," Kansas Memory, a website of the Kansas Historical Society, diagram at [], last accessed 24 August 2012 for the variety of tube & flue layouts to be found in this class over the years.

Works numbers spanned eight years. The first 76 included:

1919

April 51754-51755; June 51892-51895, 51927-51930, 51952-51953, 51963, 51984-51985; July 52006-52009, 52048-52050, 52069-52071, 52094, 52116-52118

1920

August 53595, 53615; September 3684, 53751-53753; October 53828, 53868; November 53936, 54072

1921

July 54923-54925; August 54938-54943; September 55000

1923

May 56504, 56512, 56566-56570; July 56717-56722; August 56837-56844, 56896, 56995-56998

Boiler had an Elesco H-5 feed water heater and the firebox heating surface included 112 sq ft (10.4 sq m) of combustion chamber and 37 sq ft (3.45 sq m) of arch tubes; see Locobase 8263 for the later version that had a slightly modified boiler. Drury (1993) says these were the first Santa Fe Santa Fes that were not "essentially turn-of-the-century Decapods with trailing axles ..." Instead, they were derived from the contemporaneous 3160-class Mikados (Locobase 5499). Piston valves measured 15" (381 mm) in diameter.

3800-3829 and 3870-3875 were delivered as coal-burners trailing tenders weighing 280,000 lb (127,007 kg) including 20 tons (18.2 tonnes) of coal. Beginning with 3840, the class was delivered as oil-burners. After the first 50 were delivered with Hodges trailing trucks, the last 25 arrived with Deltas. Boiler pressure increased to 210 psi (14.48 bar) in 3850-3869. Chris Hohl found that two of this first version--3868-3869--were delivered with Franklin trailing truck boosters.

According to Steve Glischinski (Santa Fe Railway (Osceola, Wisc: MBI Publishing Company, 1997), p. 100), at first, the engines presented "maintenance headaches and had to be rebuilt, after which they went on to provide many years of reliable service for their namesake road." A 2 November 1924 note found after page 80 in Volume 70 from Santa Fe's Motive Power Department concerns an oversight on Eddystone's part. Investigating why a firebox door seam rivet in 3805 blew out and slightly injured a crew man, the shops found several more that were either broken in two or nearly so. When they determined that the rivet was made of steel, Santa Fe's engineers were clearly furious:

"I wish to call your attention to the fact that specifications ...specify the use of burden iron rivets in all firebox seams except the mud ring." After establishing that the firebox had not been repaired since the locomotive's delivery, H H Lanning said that it was "practically certain that the steel rivets which failed were applied at your works."

Lanning broke the paragraph to set apart his conclusion: "It should be distinctly understood that all rivets, except mud ring rivets, used in the construction of locomotive fireboxes built for the Santa Fe, are to be of burden iron, and that steel rivets are not to be used under any circumstances." [Locobase note: it was likely an error in the type of rivets used to fastern her hull plates to the frames played a major material role in the loss of RMS Titanic.]

A few of the 3800s were retrofitted with thermic syphons in a program that saw 22 1/2 sq ft (2.1 sq m) of arch tubes removed and 106.5 sq ft of thermic syphons (9.9 sq ft) installed. Firebox heating surface area blossomed to 501 sq ft (46.5 sq m). 3874-3875 were so modified (or perhaps delivered in that configuration) as were some of the earlier engines.

One was delivered as a 2-10-4; see Locobase 8262.

As late as 1950, 132 of these very satisfactory freighters remained in service and an even 100 in 1953. But by 1956, all had been retired.

Class 3876 (Locobase 8283)

Data from AT&SF 9 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. See also DeGolyer, Volume 70, pp. 140-315 and "Atchison, Topeka and Santa Fe steam engine diagrams and blueprints," Kansas Memory, a website of the Kansas Historical Society, diagram at [], last accessed 24 August 2012). (Many thanks to Chris Hohl for his 31 December 2016 email and spreadsheet that led Locobase to a typo in the road numbers. His spreadsheet also underscored the bewildering variety of pressures, tube counts, weights, fuel types ultimately associated with the class.)

Information about how many, when, and by whom this class was supplied from [] (accessed 6 Feb 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema. Works numbers were:

1924

April 57761-57765; June 57822-57825, 57830-57835

1926

May 59246-59253, 59286-59292; July 59315-59316, 59338, 59344-59348; August 59396-59397

1927

January 59763-59764, 59779-59781; February 59826-59835, 59908-59910; April 59938-59944

Beginning with the 1923 locomotives, the 3800-class Santa Fe boiler lost twenty small tubes in favor of four more large flues. Firebox heating surface still included 112 sq ft (10.4 sq m) of combustion chamber and 41 sq ft (3.8 sq m) of arch tubes. Boiler pressure increased by 10%. Piston-valve diameter remained a healthy 15"(381 mm).

All were delivered as oil-burners and all used Delta trailing trucks. Chris Hohl notes that 3883-3890 and 3916-3930 were delivered with trailing truck boosters. 3900-3914 later used larger tenders with 20,000 US gallons (75,700 litres) of water and 7,170 gallons (27,138 litres) of oil fuel.

Class 900 (Locobase 463)

Data from AT&SF 7 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange; and DeGolyer, Volume 26, pp. 32, 33, and Volume 27, p. 80. See also "Baldwin 2-10-2 for the Santa Fe", Railway and Locomotive Engineering, Volume 16, No 11 (November 1903), p. 513; and E D Worley, Iron Horses of the Santa Fe (Dallas: Southwest Railroad Historical Society, 1976). For the Jacobs-Shupert firebox, see below and "Jacobs-Shupert Sectional Firebox", Railway and Locomotive Engineering, Volume 22, No 9 (September 1909), p. 357-358. Information about how many, when, and by whom this class was supplied from [] (also visited 6 Feb 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema. (Thanks to Chris Hohl for his 6 December 2018 email noting different fuels in different batches and their tenders.

915-984 were built first, followed by 900-914. They had works numbers that show the class came in single numbers except for two pairs. The actual numbers were

1903

September 22782, 22793, 22819, 22838, 22851, 22862, 22884-22885

October 22928, 22951, 22958, 22977, 23000, 23009, 23040, 23047, 23087, 23112

November 23135, 23149, 23167, 23173, 23190, 23196, 23234, 23237, 23243, 23259, 23279

December 23287, 23293, 23303, 23318, 23342, 23374, 23393, 23424-23425, 23455, 23468

1904

January 23506, 23515, 23522, 23546, 23585, 23608, 23614, 23637, 23644, 23659

February 23679, 23691, 23708, 23724, 23730, 23740, 23753, 23764, 23787, 23798, 23803

March 23822, 23833, 23845, 23900, 23910-23911, 23959-23960, 23978

April 24061, 24068, 24073, 24096, 24101, 24105, 24110, 24119-24120, 24139, 24150

May 24178, 24187, 24233, 24241

This was a tandem-compound design in which the high-pressure and low-pressure cylinders were lined up one behind the other (HP leading). One 13" (330 mm) diameter piston valve served each of the two cylinders sets.. The class introduced the idea of a "Santa Fe" wheel arrangement that was adopted by several railroads for drag-freight service. The leading two driven axles and the pony truck were equalized together; behind them, the rear three driven axles and the trailing truck were equalized together.

Chris Hohl noted that the first 40 engines (900-939 and, for some reason, 984) were delivered as coal burners and trailed tenders carrying 8,500 US gallons (32,173 litres) of water and 10 short tons (9.1 metric tons) of coal. 940-983, on the other hand, arrived as oil burners and in place of coal held 3,300 US gallons (12,491 litres) of fuel oil.

Some of the engines were delivered with fireboxes designed by two Frisco engineers - Henry W Jacobs & Frank W Shupert -- that eliminated staybolts. (Jacobs was assistant superintendent of the road at Topeka and Shupert was a "practical foreman of boilermakers".)

Baldwin's 1912 description stands as well as any to outline its differences from a typical firebox: "The inside and outside shells of this firebox are each composed of a series of channel sections, which are bent to a horseshoe form. The usual stay-bolts are replaced by plates, which have openings cut in them to permit the free circulation of steam and water, and are riveted between the adjacent channels."

Although staybolts were a big maintenance headache for all railroads, the J-S alternative didn't prove much more durable. Designed to be extremely rigid, the structures worked themselves loose under the wracks and strains of the forces generated by such big locomotives. The quest for rigidity probably contributed to the "Hereafter" notes in the 900 spec, which recommended reinforcing the frame according to the "Company" (i.e. Santa Fe) plans and a separate injunction to "make front frame heavier".

E D Worley offered a sardonic comment on the maintenance innovation that cut tandem cylinder maintenance time by three-quarters. At first, the shops required 30-40 hours to renew the valve and cylinder packing. Then the staff attached lugs on the smokeboxes that accepted a detachable jib crane that helped cut the time to 8-9 hours. Worley noted (p. 204): "All concerned patted themselves on the back for this outstanding of triumph of mind over matter ...but they missed the point completely [as did the trade journals, he notes]--they had evolved a Rube Goldberg device that merely catered to the ineptitudes of a basically maintenance-producing cylinder design."

The 1600s were built in 1905-1907 and were virtually identical; see Locobase 11184. See also Locobase 8256 for the superheated makeovers of both classes.

Class 900/1600 - superheated (Locobase 8256)

Data from AT&SF 7 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. Information about how many, when, and by whom this class was supplied from [] (also visited 6 Feb 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema.

Locobase 463 shows the original tandem-compound design in which the high-pressure and low-pressure cylinders were lined up one behind the other (HP leading). 900s were delivered as coal burners, 1600s as oil burners.

In 1916, the railroad started simpling and superheating these two classes to a common design. Apparently they accomplished the simple-expansion makeover simply by removing the HP cylinders in front, shortening the piston rod, and bushing the LP cylinders to achieve the 28" diameter. The firebox acquired two arch tubes that contributed 13.95 sq ft (1.30 sq m) to the firebox heating surface. Two--including 985, the only engine in the class built as a simple-expansion engine (Locobase 16403) were fitted with Baker valve gear; all of the others used Walschaert gear.

The 1600s were heavier, but otherwise the two classes were identical.

Class 985 (Locobase 16403)

Data from DeGolyer, Volume 26, p.226; and ATSF 9 -1920 Locomotive Diagrams supplied in August 20013 by Allen Stanley from his exteznsive Rail Data Exchange collection. (Thanks to Chris Hohl for his 31 December 2022 email supplying the tube sheet length.) Works number was 23989 in March 1904.

Amid the dozens of tandem-compound 900-class Santa Fes that came to that railroad in 1903-1905 (Locobase 463), this lone simple-expansion locomotive appeared near the end of the run. It was ordered separately from the rest on 26 October 1903. Like others late in the tandem-compound run, it was an oil burner when delivered.

Chris Hohl supplied the tube length including tube sheets and given in the Locobase specficfications. Between the tubesheets, Baldwin measured 19 ft 10 7/8 inches.

When it was superheated in 1917 as part of the 900 class (Locobase 8256), the 985 was fitted with Baker valve gear instead the Walschaert gear used in the rest.