Atchison, Topeka & Santa Fe 2-6-6-2 "Mallet Mogul" Locomotives in the USA

Class 1157 (Locobase 9011)

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.

This single Prairie Mallet was assembled out of two 2-6-2 Prairies (1051 & 1125). It was fitted with a superheater, a reheater, and a chamber to supply some heat to the exhausted steam from the HP cylinders before it was directed into the LP cylinders. The diagram shows the long profile with a joint made of plates that linked the two 2-6-2 boilers as well as the surprisingly high drivers.

In 1924, the 1051 was allowed to reclaim its identity as the 1157 was broken up.

Class 1160 (Locobase 9012)

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 37, pp. 297+; and E D Worley, Iron Horses of the Santa Fe (Dallas: Southwest Railroad Historical Society, 1976), p. 274. Information about how many, when, and by whom this class was supplied from [] (accessed 6 February 2006), the Santa Fe All-Time Steam Roster maintained by Evan Werkema. (Thanks to Chris Hohl for his 6 May 2020 email reporting the tender's original coal capacity.) Works numbers were 35617-35620, 35680-35681, 35727-35730 and were produced in November 1910.

When Baldwin began supplying its Prairie Mallets to the Santa Fe, the company sent the 25 rigid-boiler engines shown in Locobase 3553. A few months later, the builder added ten more that were for some reason given lower numbers.

Like the earlier engines, this batch's high-pressure cylinders used 13" (330 mm) piston valves, the LP cylinders were fed by 15" (381 mm) piston valves. Their 69" drivers--tallest ever fitted to 2-6-6-2s, wrote E.D. Worley--reflected their Prairie service as they operated in Kansas, the Texas Panhandle, and in New Mexico. But they were a bit heavier and presented slightly different heating surface areas.

Measuring heating surface areas in these Mallets can get complicated as the boiler was subdivided into several sections and included the Jacobs-Shupert stayless firebox (see Locobase 463 for a description of this home-built furnace), two combustion chambers, a large Baldwin feed water heater, and Santa Fe-design reheater and superheater.

The firebox's heating surface area included 34 sq ft (3.16 sq m) in four water tubes. The boiler section was a typical wagon top type, but manufactured with flat sided gussets that raised the centerline section to provide 6 1/2" (165 mm) more steam space over the crown sheet.

From the firebox forward, the boiler barrel held the boiler's fire tubes, a combustion chamber of 62 1/2" (1,588 mm), the Santa Fe superheater and reheater, a second combustion chamber measuring 32" long, and Baldwin's feed water heater. The spec's description of the Santa Fe apparatus shows 340 2 1/4" (57 mm) tubes, each 58 9/16" (1,439 mm) long and welded at the back end only. "Particular attention to be paid to seams and flues in superheater to make same absolutely tight." The feed water heater also contained 340 2 1/4" tubes, each of these measuring 92" (2,337 mm) long.

Among the specification's "Hereafter" notes for later reference were:

a requirement for stronger cylinders more securely fastened to the frame

reverse shafts keyed at one end for easy removal,

moving the front truck forward to allow the LP cylinder heads to be removed

a 1 March 1912 reference to "broken frames",

the need to make pedestal caps stronger using wrot [sic] iron in response to M H Haig's letter dated 11 July 1912.

Delivered with tenders holding 12 tons (10.9 metric tons), the class later used tenders carrying more coal.

Whatever their oddities, these locomotives apparently represented an adequate solution. The frame cracking and understrength cylinders referred to in the Hereafter notes were rectified, after which, says Iron Horses of the Santa Fe, "the Prairie Mallets could pull an amazing amount of tonnage," albeit at speeds of no more than 30 mph (48 km/h) because of the build-up of back pressure between the two sets of cylinders.

Class 1170 (Locobase 3553)

Data from Baldwin Locomotive Works Specification for Engines as digitized by the DeGolyer Library of Southern Methodist University, Volume 37, p. 289; and E D Worley, Iron Horses of the Santa Fe (Dallas: Southwest Railroad Historical Society, 1976). p. 273-274.. Works numbers were 34864 in June 1910; 34971-34973 and 35000 in July 1910; 35041-35045, 35090-35093, and 35135-35137 in August; and 35173-35183 in September.

Known as "Prairie Mallets. This entry covers the 24 rigid boilers. Locobase is struck by the relatively tall drivers for a compound Mallet design. This group were unusual Mallets in having piston valves on all four cylinders: 13" (330 mm)-diameter for the HP cylinders, 15" (381 mm)-diameter for the LP cans. The Jacobs-Shupert firebox heating surface included 34 sq ft (3.16 sq m) of arch tubes.

Superheater area includes a puny 300 sq ft (27.87 sq m) for the Jacobs superheater, 655 sq ft (60.85 sq m) of "reheater". From smokebox to firebox, the various boiler components arranged themselves in this order: Feedwater heater-32" (813 mm) long combustion chamber-reheater/superheater-62.5" (1,588 mm)-combustion chamber-front tubesheet to boiler. The feedwater heater measured 68 11/16" (1,745 mm) in diameter and had 340 2 1/4" diameter tubes each measuring 7 feet 8 inches (2.34 m) long for a total surface area of 1,516 sq ft. (140.84 sq m) The superheater/reheater combination had the same number (but thicker #13 WG of the same diameter) tubes that measured 4 ft 10 9/16 inches long (1.49 m).

These operated successfully until 1927-1934. See Locobase 9012 for the 1160 class.

See Locobase 13787 for two sisters that tested two different flexible-boiler designs.

Class 1170/1158 (Locobase 13787)

Data from Baldwin Locomotive Works Specification for Engines as digitized by the DeGolyer Library of Southern Methodist University, Volume 37, p. 289. See also "Flexible Boiler on Mallet Articulated Compound on AT & SF", Railway and Locomotive Engineering, Volume 24, No 3 (March 1911), pp. 107-108; and E D Worley, Iron Horses of the Santa Fe (Dallas: Southwest Railroad Historical Society, 1976). p. 273-274. Works numbers ran 34862-34863 in June 1910.

The Atchison, Topeka, and Santa Fe's "Prairie Mallets" of 1910 included 24 built with rigid boilers; these are described in Locobase 3553. Baldwin also delivered a pair that have achieved a modest notoriety in articulated-locomotive design annals for their hinged boilers. The first version had a bellows arrangement of sixty 10-in wide rings bolted alternately. These tended to burst when cinders fouled them on curves. A later version included a ball-and-socket set-up. Each will be described more fully below.

All of the class were designed for passenger service, itself an unusual combination of large LP cylinders and relatively tall drivers. Also unusual was the use of piston valves on all four cylinders: 13"-diameter for the HP cylinders,15"-diameter for the LP cans.

This was far from the end of the novel features of this design. The Jacobs-Shupert firebox was a stayless design developed by two Santa Fe engineers; see Locobase 463. Its heating surface included 34 sq ft (3.16 sq m) of arch tubes.

This pair had larger feedwater heaters filling the boiler barrel just behind the smoke box. The installation measured 68 11/16" (1,745 mm) in diameter and had 322 2 1/4" diameter tubes each measuring 9 feet 10 inches long arranged around a single core flue 11"(279 mm) in diameter. Total feedwater heating surface area was 1,893 sq ft (175.87 sq m).

It was in the next section back that the 1170 and 1171 parted company with the 1173s and with each other. A 14-foot, 5-inch combustion chamber containing the Santa Fe-design superheater and reheater connected the feedwater heater with the boiler. Baldwin's DD1 designation went to the version with flexible ball joints that joined the two parts of the boiler. According to R & LE, "...the connection consists of two cast iron sleeves, fitted one within the other and provided with snap rings to keep the joint tight. Each sleeve forms a ball joint with a cast iron ring, which is bolted to the shell of the corresponding boiler section."

So, in addition to the engine units being able to traverse independently of each other, their associated boiler segments could articulate independently as well. R & LE describes how the two were constrained from wandering too far from each other's alignment: "To assist in holding the boiler sections in alignment, a centering device is placed on each side, on the horizontal center line of the boiler. This arrangement consists of a pair of helical springs, which are seated in cast steel brackets riveted to the shells of the front and rear boiler."

DD2 was described in the DeGolyer specs as a "Vauclain accordion flexible smoke box" and R & LE gives the details: "sixty rings of high carbon steel, having a thickness of No. 14 wire-gage. These rings are 10 ins. wide and have an outside diameter of 75 1/2 ins. They are made with a set, so that, when placed adjacent to each other, they form a series of V-shaped joints. The adjacent rings are riveted together at the inside and bolted at the outside and the connection is bolted in place between the front and rear boiler sections. The products of combustion traverse the flexible connection through a cylindrical flue 44 ins. in diameter. This flue is riveted to the rear boiler section, and prevents cinders from lodging in the crevices between the connecting rings."

There's no record of whether the pleated union could be persuaded to play musical notes during its operation.

Inside the front section's portion of the combustion chamber reposed the Jacobs superheater, a small, horizontal, 2 1/2-foot diameter drum holding 247 2 1/4" diameter tubes, each measuring 30" long. In the rear section of the ball-and-socket DD1, the reheater was 50" long; the DD2's reheater measured 39". Each had 269 2 1/4" tubes, but the superheater's tubes were welded to the tube sheet at each end because they withstood the higher temperatures of the gases as they came from the firebox. The reheater intercepted the exhaust gas from the HP cylinders; at that point, lower temperatures meant the tubes were rolled and beaded.

The squeeze-box variant 1171 suffered the most noticeable failures when cinders wedged themselves between two leaves of the bellows. As the engine entered a curve, the plates would bear down on the cinders and pop the rivets holding them together. The DD1's flexible joints, as might be imagined, had unending problems with seals.

Never really successful, the two engines were renumbered 1158-1159 and scrapped in December 1927 (1159) and July 1929 (1158). E D Worley's verdict best encapsulates a reasonable judgement on these attempts at innovation: "One can only stand in open-mouthed admiration of the vast dissipation of mechanical talent implicit in the accordion-style pleats ...Pure conjecture is quite insufficient to visualize the thousands of lost machinist and boilermaker hours that were doubtlessly wasted to keep these mechanical abortions on the road."

Class 3300 (Locobase 442)

Data from AT&SF 9 -1920 locomotive diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. See also Baldwin Locomotive Works Specification for Engines as digitized by the DeGolyer Library of Southern Methodist University, Volume 38, p.1 and p.14; and See also "The Mallet Compound", Railroad & Locomotive Engineering, Volume XXIV [24], No 11 (November 1911). p. 8. (Thanks to Chris Hohl for his 9 September 2016 email noting the original driver diameter.) 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. 37283-37284, 37289-37292 in November 1911; 37312-37314, 37320-37323, 37338-37348 in December.

This set of high-drivered articulated compound Mallets (a very rare combination indeed) continued the Santa Fe experiment in such locomotives begun by the 1910 set described in Locobase 3553 and 9012. In fact, the Baldwin spec states in its first paragraph: "The details of design are to be so prepared as to provide interchangeability of parts with locos of the present 1160 and 1170 class ...except where modification or improvement of design renders interchange impractical."

Locobase 15536 offers a detailed discussion of the lessons learned from the use of Mallets in road service from the Traveling Engineers' Associations interviews with three railroads including the Santa Fe. In the case of Santa Fe, reported the R&LE, its mechanical department adopted " a very sensible policy in trying to reduce the natural prejudice of enginemen to the Mallet compound. Every convenience that could be devised to help the enginemen ...included air operated fire doors, bell ringers, reverse levers and cylinder works [sic], also coal passers."

In road service, "weight of opinion" had the Mallet being "easier on the track than consolidation and Mikado types and ...they will go with ease into sharp curve sidings that consolidation engines will not enter."

Firebox heating surface included 32 sq ft (2.97 sq m) of arch tubes. The boiler was very similar to those of the earlier engines, but had four fewer tubes. The surface area in the specs includes the boiler tubes, firebox, and feedwater heater (1,533 sq ft/142.42). The Jacobs superheater contributed 318 sq ft (29.54 sq m) and the HP-exhaust reheater another 639 sq ft (59.37 sq m). At some later date, the Santa Fe fitted the engines with thicker tires, increasing driver diameter to 70" (1,778 mm).

Four more were completed with the ball-and-socket flexible boiler design described in Locobase 13787 (road number 1170).