Articulated Locomotive Specifications

American (Puerto Rico) Class 35 (Locobase #4761)

Tom Farin's http://www.ironhorse129.com/Prototype/Mallet/Baldwin65/PuertoRico_Mallett.jpg has the builder's card for this class and all the data comes from that source. He says three were delivered. The photograph shows a cabbage-stacker.

Baltimore & Ohio (B&O) Class DD1 (Locobase #2794)

The first standard-gauge Mallet built in the United States, this helper engine was jointly designed by the B&O's James E Muhlfield and Alco's Carl J. Mellin. Old Maud, so called after a comic-strip mule, proved an immediate success on the B&O's western Pennsylvania grades. Her scrapping came only in 1938.

Baltimore & Ohio (B&O) Class EL-1, -2, -3, -5 (Locobase #439)

Several classes beginning with 15 EL-1 (1916) and followed by 15 EL-2 (7200-7214 in 1916), 30 EL-3 (7115-7144 in 1917) and 26 EL-5 (7145-7170 in 1919-1920) -- all by Baldwin. All surviving locomotives were converted from 1927 on after the prototype demonstrated a considerable improvement in pulling power. The last of these left service in 1954.

Sagle 1964

Baltimore & Ohio (B&O) Class EL-6a (Locobase #3556)

Wiener (1930) for data.

Baltimore & Ohio (B&O) Class EL-3a (Locobase #305)

Baltimore & Ohio (B&O) Class KK-1 (Locobase #3099)

These are the only 2-6-6-2 articulateds the B&O operated other than the BR&P KK-1s (see record #295) and are almost unknown today. This engine was converted briefly to a 4-4-6-2 arrangement for passenger service, but soon was restored as a Prairie Mallet and lasted until 1953.

Baltimore & Ohio (B&O) Class KK-2 (Locobase #296)

The 7450 with the conventional stayed firebox was built to compare to the 7400 with an Emerson watertube firebox. In addition to 110 sq ft of heating surface in the combustion chamber, this engine's firebox heating surface at first included 46 sq ft of arch tubes in a total of 436 sq ft, but the appliance contribution was later revamped with 18 sq ft of arch tubes and 95 sq ft of thermic syphons for a total of 403 sq ft.

Baltimore & Ohio (B&O) Class George H. Emerson (Locobase #351)

Baltimore & Ohio (B&O) Class EM-1 (Locobase #333)

Last EM-1 retired in 1960.

Bingham & Garfield Class 100 (Locobase #6941)

The Bingham Central originally was created in 1907 to ensure that the ore from the Bingham mines would get to the Garfield smelter. In 1908 Utah Copper organized the system as the B & G and began surveying the route immediately. Construction took the next 3 years with the first trip of this mighty Mallet and 41 new cars occurring on 14 Sept 1911.

The first four engines were produced individually in June (works #49978), July (50018), and November (50483) 1911 and August 1912 (51710). Three more followed with only detail differences in March 1917 (56788 & 56965) and November 1918 (59492). A chart in the diagram book shows that the locomotives varied in the number of tubes (262, 261, 256, 264, 263, 255, and 265, respectively).

Very few 16-coupled Mallets with no auxiliary axles were built for US operation, but the B & G found them satisfactory.

Kennecott Copper Corporation bought Utah Copper (parent of the Bingham & Garfield) in 1936. In 1947, Kennecott inaugurated the new, all-electric Copperton Low Line that supplanted the steam-powered Utah Copper Division line.

Booth-Kelly Class 2 (Locobase #3644)

Wiener (1930) singles out the design, saying "They deserve far more than a mere passing mention, because they were so well thought out and designed that the type has remained unchanged until this [1930] day, except for the adoption of ...superheaters and arch tubes ...which tend to increase efficiency or to reduce maintenance costs." At some point the balloon stack was replaced by a straight pipe.

http://web3.foxinternet.net/sarek/mallets tells us that Booth Lumber #2 carried on until 1945 and was scrapped a year later in Wendling, Oregon.

Boston & Maine Class M-2-a (Locobase #6942)

The firebox heating surface includes 5 arch tubes of 60 sq ft. Builder's work numbers were 63455-63456. 801 had 5 fewer tubes, which resulted in a total evaporative heating surface of 5,147 sq ft.

In the east, about the only job for which these big all-adhesion Mallets were suited was yard switching and that's what the B & M wanted them to do. After 7 years of work at the Mechanicville hump yard, however, the B & M sold them to the Bingham & Garfield copper mining road in Utah.

Buffalo, Rochester, & Pittsburgh (B & O) Class XX (Locobase #314)

J.B. Lyon Co. New York (State), 1919 v.9 no.14 pt.1 , the BR & P was planning to buy 15 of these for just over $1 million.

Other than the Triplexes built for the Erie and the Virginian and the latter's AE 2-10-10-2s (Locobase 417), these Mallet compounds had the most total cylinder volume of any built-from-scratch locomotive.

(The Buffalo, Rochester & Pittsburgh was an independent railroad until bought out by the B&O in 1932. In B&O service, these Mallets were known as EE-2. )

Buffalo, Rochester, & Pittsburgh (B & O) Class LL (Locobase #295)

Several batches delivered as follows, as classified by the B&O and renumbered in 1932:

KK-4 7500-7504 Schenectady 1914

KK-4a 7505-7509 Brooks 1917

KK-4b 7510-7525 Brooks 1918

KK-4c 7526-7540 Brooks 1918

KK-4d 7541-7554 Brooks 1923

One was rebuilt as KK-5 in 1949 with chassis lubrication, bed casting

Retired by 1951. Sagle 1964.

Data confirmed by table in May 1916 RME.

Canadian Pacific Class R1 (Locobase #3569)

Designed for helper service over the passes of the Canadian Rockies, these were very unusual Mallets for a variety of reasons. They were the only articulateds ever operated by a Canadian railroad and they had the two cylinder sets centered under the boiler. As seen in profile, the front driver set led, then its pair of LP cylinder, then the HP cylinders and finally the driver set they drove. As Wiener (1930) points out, this reduced steam pipe length, a reasonable goal especially for compounds.

The boiler had a vertical superheater whose elements intercepted the flow of gas from the boiler to the feedwater heater forward. Externally, this showed as a box on the boiler over the front cylinder set.

Although these can't be seen as failures, the CP decided on rigid-wheelbase locomotives of greater power (culminating in the 2-10-4 Selkirks) and this 5-engine class (and one more simple-expansion variant, which see at Locobase 3570) were converted to 2-10-0 Decapods in 1917. (See Locobase 4529).

Canadian Pacific Class R1 (Locobase #3570)

Carolina, Clinchfield & Ohio Class L-1 (Locobase #1318)

Had slightly more tractive effort than the L-2s, which were USRA designs. Operated until 1948-1950.

Carolina, Clinchfield & Ohio Class L-2 (Locobase #1320)

Standard USRA design based on N&W Y-2 Mallet compound locomotive and delivered to the Clinchfield in 1919. Unlike the light articulateds, the "heavy" was considered a great success. These had 14" piston valves on all cylinders.

Ten more entered service in 1923 as L-3 (Locobase 1321). This class operated until 1950-1952.

Carolina, Clinchfield & Ohio Class L-3 (Locobase #1321)

Follow-ups to the earlier 10 built to the USRA heavy articulated design (Locobase 1320). These were essentially similar, but came from Brooks rather than Baldwin. Operated until 1951-1952.

Carolina, Clinchfield & Ohio Class E-1/E-2 (Locobase #337)

The E-1 & E-2 were virtually identical to the earlier D&H J-95s although the Clinchfield locos had a higher adhesion weight and larger piston diameter. Firebox heating surface included 78 sq ft of arch tubes. Eight E-1s were delivered in 1942-1943, four E-2 in 1947. Both classes were retired by 1954.

For the six E-3s, see the Union Pacific entry (Locobase 1345).

Central of Georgia Class ML (Locobase #7111)

These light Mallets (works #59038-59047) served as pushers on the Sylacauga grade in Alabama, but proved to be overpowered for the Central. They were traded to the Illinois Central in 1926 for some Santa Fes. That railroad, which had gained control of the C of Ga in 1907, operated them as 6000-6009 for about 10 years before scrapping them.

See the Georgia Encyclopedia online -- http://www.georgiaencyclopedia.org/nge/Article.jsp?path=/Transportation/LandTransportation/Railroads&id=h-1282 (accessed 14 November 2005) -- for more information about the Central of Georgia.

Chesapeake & Ohio (C & O) Class H-1, H-2 (Locobase #440)

Although the design represented a significant increase in power, their low drivers and small size doomed them to a relatively short life. They were all scrapped by 1935.

Wiener (1930) for data.

Chesapeake & Ohio (C & O) Class H-3/H-4/H-6 (Locobase #441)

Firebox heating surface included 25 sq ft of arch tubes. All 175 had 12"-diameter piston valves for the rear, HP cylinders and slide valves for the LP cylinders driving the forward engine set.

150 H-4s were built at Schenectady and Richmond between 1912 and 1918. They diiffered from 24 H-2s of 1911 (Brooks) by 35,000 lb of engine weight each. Individual H-4 batches had adhesion weights ranging from 358,000 to 364,900 lb.

55 H-6s, 45 built by Alco in 1920-1923 and 10 by Baldwin in 1949, rounded out this numerous class. The 25 H-3s were built for the Hocking Valley in 1917-1918 to the same design, but the boilers were pressed to 220 psi, which yielded a slightly higher tractive effort.

(20 H-5s built by Baldwin and Alco were USRA light Mallets.).

The H-2s retired in 1935, the H-4s 1930-1955, and the H-6s from 1952 to 1957.

Chesapeake & Ohio (C & O) Class H-5 (Locobase #299)

Firebox had combustion chamber and 34 sq ft of arch tubes contributed to firebox heating surface. This design was one of two articulated types that formed part of the roster of standardized designs developed for general production in WW I.

These light USRA Mallets weren't very well received. Only 30 were built, 10 by Baldwin, 20 by Alco Schenectady. The Baldwins served the Wheeling & Lake Erie for decades, while the 20 Alcos saw service on the C&O, where they "steamed well and performed satisfactorily." Yet the Chessie didn't like the engines much and picked at the design (poor ashpans, need to reduce steam pressure from 225 to 210 psi, poor cab layout). Eugene Huddleston (Trains, March 1991) contends that the C&O's requirements had outstripped the 2-6-6-2 arrangement and that contributed to the road's disaffection.

Chesapeake & Ohio (C & O) Class H-7 (Locobase #312)

Firebox had combustion chamber and boiler had Elesco K-54 feedwater heater. When delivered, boilers were pressed at 205 psi.

First 25 H-7 (1540-1565) built by Alco in 1924, second 20 (1570-1589) built by Baldwin in 1927. Baldwin H-7a credited with slightly higher weight on drivers, larger heating surface (6,580 sq ft). Cylinder horsepower for all was 4,092 hp.

Alexander (American Locomotives, 1950) notes that H7 pulled 9,500-ton trains over 113-mile run from Russell, Kentucky to Columbus, Ohio in 5 hours. Averaged 4,400 miles per month until displaced by the 2-10-4s in Ohio and the 2-6-6-6 Alleghenies in the rest of the territory. 30 of the fifty were later leased to the Union Pacific. By then the fireboxes had been fitted with 167 sq ft of thermic syphon as shown in the specs.

Chesapeake & Ohio (C & O) Class H-8 (Locobase #304)

Built in two batches: 1600-1644 in 1940-1944, 1645-1659 in 1948. Firebox had 3 syphons and arch tubes combining for 162 sq ft of heating surface as well as a huge combustion chamber. Eight similar engines with smaller boilers and superheaters went to the Virginian. Could operate 11,500-ton coal trains at speeds up to 45 mph. Probably the heaviest 12-axle locomotives ever built, with certainly the highest axle load ever put on rails. Average axle load was 84,650 lb and the first driving axle bore 86,700 lb.

Puzzle: C&O Power gives higher weight on drivers (507,900 lb for the first batch, 504,010 lb for the 1948 engines), and heavier engine weights (771,300 lb and 751,830 lb, respectively. Possible answer: Cyclopedia weights were as designed, C&O Power weights may have come from C&O drawings.

Bob Quehl (EMail:rquehl@adelphia.net) from Pittsburgh, PA offered considerable detail on the difference as well as other information ( http://www.worldrailfans.org/Forum/1998/Jan/2281650.shtmlon Feb 4 2000, 4:06) " In the December 1998 issue of Trains Magazine there is a very interesting article that suggests the first 10 H-8s delivered in December 1941 to January 1942 weighed actually weighed in the area of 775,330 lbs in "working order". This was just the engine weight including water in the boiler, 12,000 lbs of sand in the domes an a two man engine crew. This is interesting because the specifications for the order between the C&O and LIMA called for a weight of 726,000 lbs.

"Apparently it became known by some senior managers at the C&O that the engines were considerably overweight. They did not start to raise questions as there was a cover-up attempt by C&O and Lima, until the neighboring Virginian Rail Road ordered their own Alleghenies. The president of the Virginian, one Frank Beale, was a former high level person in the C&O and knew what the H-8 was supposed to weigh. When he got the specifications from Lima for the Virginian's engines, he could not understand why they weighed considerably more and one thing led to another. The Virginian had a lot of trouble getting their 8 Alleghenies delivered as the surrounding rail roads did not want that much weight on their rails ...One engineer at Lima gave the H-8's weight as high as 778,200 lbs. For the engine and tender combined weight I have seen several references in other sources around 1,200,000 lbs.

As for power, in 1943 a C&O dynamometer test reportedly gave the highest drawbar horsepower (where it really counts) of any locomotive made. They had the highest weight on drivers of any steam locomotive, and must have really pounded the rails. With 67 or 69 inch drivers (I can't remember which off hand) they would easily have been capable of 60 mile an hour or faster speeds and I have seen some photos of them hauling express box cars trains. I have also read that some of them were used during WW II to haul fast troop trains for the military."

Chicago & Alton Class K1 (Locobase #9102)

In the same year that Brooks delivered the first of its L1 Mikados to the Alton (Locobase 9100), it supplied this trio of light Mallets. They apparently borrowed the same boiler design, but lengthened the tubes by a foot and included more flues. Like the Mikes, the HP cylinders had 14" piston valves.

One was sold almost immediately to the Chesapeake & Ohio and classed as an H-3, although it was quite different from the Chessie's other H-3s. It served the C & O until the late 1920s.

The other two persisted in pusher service until the Alton was taken over by the Baltimore & Ohio. The pair then traveled east, but into storage rather than into service and, without any further service, they were scrapped in 1938.

Chicago Great Western Class H1 (Locobase #3091)

After March 1914 tests among 10 each of the CGW's Consolidations, these Mallets, and the newer Mikados over the Eastern Division between Oelwein, Iowa and Stockton, Ill, the Railway Age (30 October 1914) summarized the Prairie Mallet's performance: "It is surprising to note the relatively small amount of gross tons per mile for the Mallet locomotives with such a high percentage of gross to potential ton miles." It isn't surprising, therefore, to find these engines moving on to the Clinchfield by 1916. The CGW simply found the Mikados they were buying more suitable to their requirements.

Picked up by the Clinchfield after the Chicago Great Western found them unsuitable, they lasted only a few years. As Drury (1993) comments: "Their early retirement in 1925 is probably a good indication of their performance."

Chicago, Burlington & Quincy (CB&Q) Class T-1 (Locobase #1047)

Data from locomotive diagram published on Vernon Beck's website -- http://home.earthlink.net/~vnlbeck/T/T1t.tif (accessed 22 March 2003).

Chicago, Burlington & Quincy (CB&Q) Class T-2 (Locobase #2906)

There was more novelty. Steam from the high-pressure cylinders would be led back through 19 2" tubes gathered in the 17" diameter flue in the preheater.

All 10 would eventually be converted to oil burning and would receive Elesco feedwater heaters. Although they proved difficult to maintain, they clung to a niche on the Edgemont-Deadwood line in the Black Hills until the late 1940s.

See also data from locomotive diagram published on Vernon Beck's website -- http://home.earthlink.net/~vnlbeck/T/T2t.tif (accessed 22 March 2003).

Chicago, Burlington & Quincy (CB&Q) Class T-3 (Locobase #5476)

Drury (1993) doesn't say much about this lone Consolidation Mallet, just that it was used for hump work at Galesburg and was later converted to oil-burning and transferred to Alliance, Nebraska from which it was retired in 1934.

Columbia River Belt Line Class Skookum (Locobase #4069)

The design was rated at 200 tons up a 2 1/2% grade (1 in 40). According to the December 1967 Industrial Railway Record (No 15), two significant design limitations were the requirement to negotiate 160-ft radius curves and the light, 56-lb/yard track.

As the logging locomotive website's summary of service shows, this little Atlantic mallet covered most of the United States timber harvesting regions during its long life.

1909 - 1909: Little River Railroad #126. Townsend, TN. ironhorse129 explains that the 126 was too long for the Little River's tight curves and it was quickly returned to Baldwin

1910 - 1920: Columbia River Belt Line Railroad "Skookum." Blind Slough, OR

1920 - 1924: Carlisle-Pennell Lumber Co #7. Onalaska, WA

1924 v 1930: Transferred to Carlisle subsidiary Newaukum Valley Railroad Co #7. Onalaska, WA.

1930/1931: Leased to Mud Bay Logging Co. #7, Olympia, WA

1931 - 1933: Carlisle Lumber Company #7. Onalaska, WA.

1933 v 1955: Deep River Logging Co. #7, Deep River, WA

Disposition after logging service:

1956: Sold to Charles Morrow, dismantled and moved in 1960 to Snoqualmie, WA

1980: to Rogan Coombs, Garberville, CA.

1990: Moved to Mineral, WA for storage at Mt. Rainier Scenic Railroad

Delaware & Hudson Class H (Locobase #8952)

These were delivered as pushers and a look at the numbers shows just how dedicated they were to that service. All the weight was adhesive, the drivers were small, the boiler big. Six of the locomotives were later superheated to one standard, the rest to another; see Locobases 8953-8954.

Delaware & Hudson Class H - superheated - 1600 (Locobase #8953)

For some reason, the D & H chose two different combinations of tubes and flues when they came to superheat these big pusher Mallet Compounds. This was the more superheater-heavy result, a locomotive with considerable amounts of dry steam. It kept the small tender that was delivered in 1909.

Other than the tube arrangement, however, the designs remained essentially as they had been delivered and continued to serve into the 1940s and early 1950s.

Delaware & Hudson Class H - superheated - 1606 (Locobase #8954)

Of the two combinations of tubes and flues adopted by the D & H when they superheated these H-class pusher Mallet Compounds (Locobase 8951), this design retained more of the small tubes and introduced fewer flues. On the other hand, the tender trailed by this subclass was much bigger.

Other than the tube arrangement, however, the designs remained essentially as they had been delivered and continued to serve into the 1940s and early 1950s.

Delaware & Hudson Class H-1 (Locobase #5006)

Drury (1993) says that the Hs were 2 ex-Pittsburgh & West Virginia 2-6-6-0s that were converted to 0-6-6-0s and used as pushers. Richard E. Cooper's compilation of D&H steam power presented in Jim Shaughnessy's 1997 reprint of his 1982 Delaware & Hudson asserts the same.

Delaware & Hudson Class J-95 (Locobase #338)

Considered by many observers to be the handsomest of the Challenger type. Forty engines -- 1500-1519 purchased in 1940 (builder's numbers 69297-69316), 1520-1534 (70006-70020) over next 2 years, and 1535-1539 (74666-74670) procured in 1946. All were disposed off in 10 months from October 1952 to August 1953.

Denver & Rio Grande Western (D&RGW) Class Centipede 28 62 62/62 1/4 EEE (Locobase #6823)

In the same year that Erie was testing the Matt Shay, Baldwin prepared a proposal for the Rio Grande. The diagram, preserved in a later diagram book (D&RGW 12 - 1937 Folio 10 supplied in May 2005 by Allen Stanley from his extensive collection), shows a locomotive whose design has been toned down. The cylinders are 2" smaller in diameter, but most of the other details were repeated.

This engine's firebox heating surface, which included 108 sq ft of combustion chamber and 88 sq ft of arch tubes, almost matched that of the Erie. The grate itself was deliberately limited to 80 sq ft through a "Gaines wall" such as that used in the Yellowstones (see Locobase 331), which was expected to raise interior temperatures. In the Erie's, which originally had a similar setup, the Gaines wall was removed when the grate proved much too small. Doubtless the same fix would have been resorted to had this locomotive ever entered service.

In the event, the Erie's experience probably dampened enthusiasm and the locomotive was not built.

Denver & Rio Grande Western (D&RGW) Class L-62 (Locobase #294)

Eight Mallet compounds for helper service in Soldier Pass and Tennessee Pass. They were delivered as saturated-steam locomotives with a total of 395 2" tubes and a firebox whose heating surface included 40 sq ft of combustion chamber. Total EHS was 4,114 sq ft. As shown in the specs, however, they were later superheated. Unlike most such conversions, this class retained virtually all of the heating surface area of the saturated boiler.

Later renumbered 3300-3307. Retired in 1947-1952.

Denver & Rio Grande Western (D&RGW) Class L-95 (Locobase #435)

These Mallets preceded the 2-10-2s described in a Railway Age Gazette article of 3 August 1917. Used in Minturn-Malta (Tennessee Pass), Colorado helper service. Rio Grande's calculation of compound tractive effort yielded 96,000 lb.

The specifications refer to the later boiler (new in 1941) whose firebox heating surface included 114 sq ft in the combustion chamber and 88 sq ft in thermic syphons in both the firebox and the combustion chamber.

Denver & Rio Grande Western (D&RGW) Class L-76 (Locobase #1466)

Data from 1952 D & RGW locomotive diagram book supplied in May 2005 by Allen Stanley from his extensive collection.

Denver & Rio Grande Western (D&RGW) Class L-107 (Locobase #315)

Mallet compound with 39-in LP cylinders. Later variant of the L-95 and basically a copy of the USRA heavy articulated. Note the good boiler proportions in which the firebox heating surface included 49 sq ft of arch tubes and 82 sq ft of combustion chamber for a total of 435 sq ft. Before long, direct heating surface expanded in one of two ways. In one, 115 sq ft of syphons replaced the arch tubes; that version is shown in the specification. In the other, firebox circulators added 76 sq ft to the surface for a total of 462 sq ft.

They saw out steam, being retired in 1947-1951.

Denver & Rio Grande Western (D&RGW) Class L-131 (Locobase #316)

Firebox heating surface includes the combustion chamber (170 sq ft) and thermic syphons in both the combustion chamber and in the firebox (137 sq ft) or circulators (143 sq ft), boiler fitted with feedwater heater, valve motion limited cutoff.

Ten engines that were simple-expansion improvements on the earlier Mallet compounds. Rated at 3,300 tons on the east side of the Continental Divide (maximum grade 1.42%) and 1,400 tons on the west slope (max grade 3%). A later class of 10 (3610-3619) from Alco was designated L-132.

Denver & Rio Grande Western (D&RGW) Class L-105 (Locobase #1345)

Firebox had 200 sq ft of heating surface in three thermic syphons and arch tubes as well as 206 sq ft in the combustion chamber. This second group of five engines had smaller heating surfaces and consequently different ratios.

A third group of six (3800-3805) L-97s were leased from the UP during World War II. They were returned to the UP, which then sold them to the Clinchfield in 1947. See the UP Challenger entry.

Denver & Rio Grande Western (D&RGW) Class L-105 (Locobase #339)

Firebox had 200 sq ft of heating surface in three thermic syphons and arch tubes. These were built for dual-service operation over the Continental Divide.

Denver & Salt Lake Class 76 / L-77 (Locobase #6819)

Drury (1993) shows that these started out on the D & SL as 0-6-6-0 Mallets that acquired a leading truck in 1912 to improve their over-the-road qualities. The result was a very uncommon Mallet wheel arrangement in the US. At the same time they were fitted with mechanical stokers. A photo of 200 in 1947 (Drury, p. 162) underscores the short boiler over the engines, even if one mentally deducts the leading truck.

Still the design served its purpose for decades, retiring on the Denver & Rio Grande Western several years after that railroad bought the remaining locomotives in the class in 1947. See the as-built 2-6-6-0s also supplied to the D & SL in 1912 at Locobase 5407.

Denver & Salt Lake Class 210 / L-77 (Locobase #5407)

Drury (1993) says that these locomotives were the beneficiaries of two updates to the Mallet design the railroad was already operating -- a leading truck to make it a better road engine and a mechanical stoker.

They apparently served their purpose well because all but one went to the Denver & Rio Grande Western in 1947, from which they retired a few years later (1949-1951) as 3370-3375.

Dick Construction Co Class 144 (Locobase #5041)

Duluth, Missabe & Iron Range (DM&IR) Class M (Locobase #317)

Eight Mallet compounds virtually identical with Virginian 600. Stoking these engines was sometimes a problem even when two firemen were used. Although capable of handling 85 empties on the steep grades when fitted with mechanical stokers, the smallish boilers restricted their speed when so engaged.

Workhorse engines that were later given Schmidt superheaters, Standard mechanical stokers (in 1925), and Elesco feedwater heaters (in 1930). These engines were followed by 4 more engines fitted with superheaters from the start.

Duluth, Missabe & Iron Range (DM&IR) Class M-1/M-2 (Locobase #318)

Four Mallet compounds that followed the 8 bought in 1910 (see record # 317). One source says that 208-209 had 77-sq ft grate area, which is smaller than earlier engines and presumably was still harder to fire and that 210-211 reverted to the early 84-sq ft grate, yielding a GDF of 61,750 and a CHS/GA ratio of 78.02. The data comes from a January 1917 table in Railway Mechanical Engineer for "208" that uses the 84-sq-ft figure.

Frank A King, writing in the November 1980 Trains, confirms the 84-sq-ft figure and notes that overall length was 5 feet less because the firebox was moved 5 ft farther forward.

All were converted to simple expansion locomotives in the 1930s. (see record # 319).

Duluth, Missabe & Iron Range (DM&IR) Class M - superheated (Locobase #6653)

When the M-class Mallets (see Locobase 317) were fitted with Schmidt superheaters, Standard mechanical stokers (in 1925), and Elesco feedwater heaters (in 1930), the result was as is shown here. The firebox heating surface now included thermic syphons.

Duluth, Missabe & Iron Range (DM&IR) Class MS/etc (Locobase #319)

210 (class M-2S) was rebuilt in 1929 with virtually identical specs except for an EHS of 5,441 sq ft and CHS of 7,921 (SHS% of CHS 31.31, BDF 792, CHS/GA ratio 94.3). She was nicknamed Madame Queen.

211 was rebuilt in 1931 and had an EHS of 5621 sq ft and a CHS of 8101 sq ft (SHS% of CHS 30.61, BDF 774, CHS/GA ratio 96.44).

The 207 (class MS) was rebuilt along similar lines in 1931, but came out heavier at 444,650 lb on the drivers.

Duluth, Missabe & Iron Range (DM&IR) Class M-3/M-4 (Locobase #334)

Boiler tube and flue length (sheet to sheet) was shortened by 2 feet to 21 feet. The mixture of tubes and flues changed as the design substituted a Type E superheater for the M-137's Type A. The number of 2 1/4-in flues decreased to 82 and Baldwin installed 245 of the 3 3/4-in flues (vs. the M-137's 270 of the former and 75 5 1/2-in flues) . As a result, superheater area increased by 25%. The first 5 M-4s had Worthington feedwater heaters, the last 5 Elescos. Piston valves measured 12" in diameter.

Following contemporary practice, the M-3s had integrally cast frames and cylinders, Timken roller bearings on all driving axles, ASF roller bearings on all truck axles.

These engines performed very well from the time they entered service in May 1941, soon handling train loads 25% heavier than the earlier M-1/M-2 engines could manage. The last 10 engines were M-4s, built by Baldwin in 1943 (works #64707-64716), which used carbon steel in certain components because of a shortage of high-tech alloy steels. They had an engine weight of 699,700 lb.

During 1943-44, as many as 12 "Yellowstones" were operating on the D & RGW, the GN, and the NP under lease where they were highly regarded. In fact, the D & RGW telegraphed the DM&IR with the claim that these Yellowstones were the finest engines ever to run on the Rio Grande.

The Lake Superior Railroad Museum site -- http://www.lsrm.org/Museum/mallet.htm (visited 9 Feb 2004) -- gives some interesting consumption numbers: " When working at full power, [the locomotive] could consume some 10 to 12 tons of coal an hour and evaporate water into steam at the astounding rate of 12,000 gallons per hour. The amount of coal ...used in one hour would be enough to heat a home for two winters." (and Minnesota winters at .0.that, one supposes.)

Erie (ERR) Class L1 (AERJ 1908) (Locobase #5700)

Could the 24' tubes have been the original length but soon seen as overwhelming both the fireman and his grate without any commensurate gain in heating value? The tube count certainly suggests as much. The two earlier accounts showed 446 tubes and a total of 6,629 sq ft of heating surface. A later 1907 AERJ raised the tube count to 468, but shortened the length to 21 ft (total ehs 6,108 sq ft). In the June 1908 account of the testing, we see still fewer tubes and a lower overall heating surface area.

Known at one time as the Angus, they were the only camelback Mallets of this wheel arrangement. Drury (1993) comments that "On the job they proved mostly that it took a skilled and strong fireman to produce the power they were designed to deliver."

In 1921, they were rebuilt by Baldwin as 2-8-8-2s; see Locobase 7745.

In 2004, MTH modelers unveiled its model of the 0-8-8-0 and explained the nickname as follows:

"The L1 obtained the nickname "Angus-type" as a result of noted railroad operations writer Angus Sinclair's comments that the L1 would dry up all the country's canals and make all forms of water transportation obsolete thanks to the engine's incredible thirst. Because only three L1 locomotives were constructed, Sinclair's comments never rang true but the engine did establish the use of Mallet type engines beyond narrow gauge light duty use."

(see http://www.mth-railking.com/newsdetail.asp?artid=128, visited 11 Nov 2004).

Erie (ERR) Class P-1 (Locobase #3300)

The goal of the design was to put the maximum tractive effort into a helper engine. The result, conceived and patented by Baldwin's George R. Henderson, was a sort-of-Mallet, sort-of-tank articulated locomotive. One prototype was prepared (works #41308) that was soon joined by two others (works # 43350-43351).

The Triplex arrangement of 3 engine sets had the middle, high-pressure cylinders exhausting into equal-size low-pressure ones fore and aft. The front LP set exhausted into the normally placed stack, which maintained a draft over the grate. The rear cylinder exhausted passed through a feedwater heater and up a narrow pipe at the rear of the tank. The firebox heating surface included 108 sq ft of combustion chamber and 88 sq ft of arch tubes.

Unfortunately for such grand ambitions, the design failed in two important respects:1) it could never have generated sufficient steam as both the boiler and the grate were far too small and 2) the couplers on most freight cars couldn't stand the strain of the prodigious pulling power this design could muster.

As originally delivered, the grate measured 90 sq ft, which gave an impossible GDF of 123,379. The grate seen in the data (122 sq ft) was fitted in the second and third examples. To create the larger grate, the Gaines brick wall in the firebox was removed and the arch tube layout reduced heating surface from 88 to 74 sq ft. (See Wiener, 1930, for comparison data.) As can be seen, however, both demand factors are still far too high.

Comstock (1971) quotes an Erie fireman, Andrew Goobeck, who wrote in an October 1942 Railroad Magazine article that "We used to say that the best place for anybody to cool off on a summer day was behind the ...firedoor." Drury (1993) notes that the engines were never simpled, even for starting, because "two or three revolutions of the drivers would have exhausted the boiler."

Also, although the Matt Shay (first of the three) did start a train of 250 coal cars, the run lasted only 17 miles before the force of the pull snatched one from a coal car and stopped the train. They could be useful pushers, however, and operated on such grades as Susquehanna Hill until the late 1920s.

For Baldwin's Centipede proposal to the Denver & Rio Grande Western prepared in the same year as the Matt Shay's first runs, see Locobase 6823.

Erie (ERR) Class L1 - modified (Locobase #7745)

Locobase 5700 describes the one-of-a-kind camelback pusher locomotive class that served only the Erie. It was just too unusual and Erie likely also felt that the large boiler was simply too big and long for the Mother Hubbard layout. In 1921, they were rebuilt by Baldwin as 2-8-8-2s and fitted with mechanical stokers and superheaters. In this new arrangement, they ranked as small engines except for the generous area of their Wootten firebox. Firebox heating surface included 54 sq ft of arch tubes and 96 sq ft in the combustion chamber.

The firebox heating surface now included 96 sq ft of combustion chamber area and 54 sq ft of arch tubes. The cab was moved back to a conventional location behind the firebox. The lead truck extended wayyyyy forward of the smokebox -- Locobase wonders about the weight distribution of this design and indeed the diagram shows a relatively low axle loading of 16,130 lb.

In this form they operated until 1930.

Great Northern (GN) Class M-2 - -compound (Locobase #352)

Similar to Southern Ls engines, these M-2s were heavier, and had slightly smaller drivers, smaller boilers, lower superheat ratio, and a different assortment of tubes. Fitted with Emerson superheater of 1,777 sq ft in the tubes, which was soon removed. and a feedwater heater. The specifications show the locomotives after the Emerson superheater was removed.

Some were simpled; see the M-2 simple entry (Locobase 2800). 22 were converted in 1929-1931 to 69"-drivered 2-8-2s and designated O-7s.

Great Northern (GN) Class L-1 - 1st batch (Locobase #5354)

These first five performed as helpers in the Cascades, the next 45 went out on the road themselves. The CB & Q bought 3 identical T-1s (Locobase 1047) in 1908.

Great Northern (GN) Class L-2 (Locobase #3554)

These were smaller than the earlier L-1s (Locobase 5354), with a boiler, says Drury (1993) taken from the H-2 Pacifics. Small Mallets with Belpaire fireboxes, these engines weren't repeated as the Great Northern went for much bigger articulateds. Northern Pacific bought 6 copies in the same year (Locobase 821) .

Great Northern (GN) Class L-1 - 2nd batch (Locobase #819)

One can guess that those engines weren't quite powerful enough, so the GN tweaked the L-1 design slightly (yielding a slightly smaller boiler) and ordered the 20 more in 1908 that are shown in the current entry.

The Northern Pacific purchased the same design in the form of 16 Zs (Locobase 1046) in the same year. Western Maryland's 9 M-1s (Locobase 8321) used the same boiler and firebox, but had larger HP cylinders.

Great Northern (GN) Class N-1 (Locobase #307)

Great Northern (GN) Class L-1 -S (Locobase #8854)

At least two of the GN's L-1 2-6-6-2s (Locobase 5334) were superheated. In the process, the HP cylinders grew by 2 1/2" while the boiler's overall heating surface remained nearly the same. As these two were converted to O-6 Mikados in 1926 (Locobase 7449) as well as the rest of the class, the program must not have been very extensive.

Great Northern (GN) Class L-2 -S (Locobase #8855)

Although the GN found its 2-6-6-2s unremarkable, the railroad still superheated some of these pocket Mallets. Leaving the Belpaire firebox and the power dimensions alone, the shops performed the usual exchange of fire-tubes for flues. Most kept their HP slide valves, but three traded in for piston valves. Like the L-1, the class was rebuilt as 2-8-2s (O-5 class; Locobase 7450) in the 1920s.

Great Northern (GN) Class M-2 - simple (Locobase #2800)

One of the few classes of locomotives delivered with the odd 2-6-8-0 arrangement. Drury (1993) speculates that it was adopted as a way to get more adhesion. As delivered they were M-1 compound engines (see Locobase 352) , but they were simpled in 1921. In keeping with their unusual arrangement, the rebuild gave them front cylinders 22" in diameter, the rear ones 23 1/2". (Hence Locobase's compromise cylinder size adopted to ensure proper tractive effort calculation.) See Larry Bohn photograph (http:/www.prostar.com/web/gngoat/gn27013.htm) for an illustration of the rebuild.

Charles R. Wood, in Lines West (New York: Bonanza Books, 1967), comments about these engines (p 66): "Essentially, [they] were slow-speed locomotives and any attempt to push the speed was hard on the engine and track." Watching the side rods carve great swaths of space close to the ground led some enginemen to dub them "grass-cutters." Wood continues: "Rough -riding at any speed above 25 mph, they vibrated with a fierce clatter of gangway plates and jingle of cab gagues. As one engineer put it, 'I couldn't even keep my shoelaces tied.'"

Under another photo of the M-2 on p 67, Wood attributes the roughness to large counterweights on the last set of drivers under the cab. And the lack of trailing truck meant that a backing M-2 "...met every rail joint and switch point head on, and changed the angle of direction with a very positive lurch." Yet, Wood concedes, these engines had their place in the GN's scheme of operations:

"[W]ith all their faults, they could move tonnage - not very fast to be sure - but between Butte and Great Falls they would come slogging up the grade under a vast canopy of smoke and cinders while the front bandstand articulated to the curves and the exhaust bounced off the hills and mountainsides."

By the end of the decade, 22 had become O-7 Mikados.

Great Northern (GN) Class N-2 (Locobase #2802)

In the early 1940s they got new frames and roller bearings as well as a larger grate; at that point they became N-3s; see Locobase 6445

Great Northern (GN) Class R-1s (Locobase #320)

Firebox had combustion chamber that contributed 155 sq ft to direct heating surface, the boiler had exhaust steam injectors, and steam was distributed through 14" diameter piston valves.

The first four were built by Baldwin (works #58480-58481, 58528, 58542) and the last ten by Great Northern.

Great Northern (GN) Class R-2 (Locobase #321)

Boiler fitted with exhaust steam injector. The firebox heating surface included 178 sq ft from the combustion chamber and a surprisingly high 78 sq ft in arch tubes.

The Great Northern's essay in simple articulateds produced this immense specimen that reflected the influence of the superpower revolution of the late 1920s. Note the size of the boiler and the superheat surface. Compared to the R-1s by Baldwin, these engines had 10% more EHS, 85.4% more SHS, 8.8% lower BDF, and a 14.5% increase in indicated TE. Notice too that these engines grossed out at 1,003,530 pounds with loaded tender.

Great Northern (GN) Class N-3 (Locobase #6445)

In the early 1940s they got new frames, nickel-steel boilers, and roller bearings as well as a larger grate; at that point they became N-3s.

http://www.greatnorthernempire.net/index2.htm?GNEGNPrototypeDrawingsSteam.htm (visited 7 Jan 2005) shows all of the numbers in the GN's locomotive diagram. A note says the boiler had a nominal pressure of 275 psi, but was lowered to 265 psi for working. A comparison with the N-1's factors shows what a huge increase in power these rebuilds represented.

The timing couldn't have been better, either. The first of the rebuilds (2024) was completed on 10 February 1940, the last (2009) wrapping up almost exactly 2 years later on 12 Feb 1942. Thus the greater horsepower was available in time for the Great Northern's World War II traffic spike.

And did the N-3s ever deliver, according to Charles R. Wood, in Lines West (New York: Bonanza Books, 1967), p 109, describes them as: "Amazing engines, they hauled ore on the iron range, hustled time freights on the Kalispell Division and served in heavy duty passenger service on troop trains." On the latter service, Wood quotes Mr. J S Miller, Assistant to the Superintendent of Motive Power in 1945/46 as saying that N-3s could "...run like a deer" and hit speeds of 50-60 mph.

Virtually all of the class was retired in 1955.

Hammond Lumber Company Class (Locobase #4054)

http://web3.foxinternet.net/sarek/mallets . #5 had the shorter career, parallelling that of #6 for the first 8 years on the Hammond Lumber and Hammon & Little River Railroad. In 1937, #5 went to Crown Willamette as their #16 and soldiered on until 1959, when she was scrapped.

#6 wandered throughout the west, winding up in Flagstaff, Arizona in 1960. She subsequently went on display in two different locations.

Data from Bruce (1950). Additional data from the Arizon Hitorical Society's History of the Pioneer Museum found (18 April 2003) at http://www.infomagic.net/~ahsnad/LocomotiveHistory.htm . They note that the 1929 price was $46,000.

Interstate Class 20 (Locobase #6441)

The diagram is less informative about heating surfaces, giving only the total of evaporative and superheater combined as 8,332 sq ft. This is 737 sq ft more than the USRA standard.

Kansas City Southern (KCS) Class G-1 (Locobase #437)

Assigned to KCS's Northern Division (KC to De Queen, Ark.). Two later refitted briefly to burn pulverized coal, but reverted to oil burning in 1930. Firebox heating surface included that provided by 92 sq ft in 4 arch tubes and 127 sq ft in a combustion chamber.

Kansas City Southern (KCS) Class G-2 (Locobase #438)

Calculated compound tractive effort was 110,400 lb or, using the ARA method, 103,700 lb. BDF, GDF calculated from ARA compound TE. Low-pressure, 41-in diameter pistons contained a volume of 24.44 cu ft each.

Kansas City Southern (KCS) Class G-2S (Locobase #308)

http://home.insightbb.com/~sammy56/folio/fo43-11.jpg (August 2002) has the blueprint elevation with all the data. Although the diagram clearly shows a combustion chamber, its length isn't recorded.

Maine Central Class X (Locobase #6529)

On the MEC they served on the Crawford Notch grade in New Hampshire. Within a year they'd been converted to coal burning. Rendered redundant by the introduction of 2-8-2s, the class worked sporadically for years. The three unsuperheated engines were scrapped in 1929, while 1204, which had been superheated during repairs following an accident, lingered until 1934.

McCloud River Class (Locobase #3575)

A photograph in Westing (1966) shows the oddball result. For one thing, the two locomotives were not quite symmetrical. One had a saddle tank, the other a pannier tank with wood racks outside. On the other hand, both engines had the diamond-shaped spark-arresting stack and ornately curved sand and steam domes, the latter with the safety valve on top. Cylinders rode with the piston valve below and driven by inside valve gear. And each engine had its own bell.

According to a website that documents the McCloud -- http://www.trainweb.org/mccloudrails/LocoRosterDetails.html, accessed 8 December 2005 -- the two engines were soon separated because the tandem "...derailed constantly on the light trackage." As number #5, one of the two was sold in 1919 to Weed Lumber Company, then was scrapped after sale and use by Lystul-Watson Logging Company. #6 first went to Atkinson Construction Co. #6, then to A.D. Schader as their #6, and finally to Permanente Metals Co as #2515. Much longer lived than its twin, this engine was scrapped in 1949.

Mexican Central (NdeM) Class HR-2 (Locobase #5715)

Mexican Central followed this singleton with an order for 20 more in 1910-1911, which suggests that this was a successful prototype.

Mexican Railway Class R-1 (Fairlie) 1889 (Locobase #6957)

The first of three classes of double Fairlies on the Mexican, used to scale some of the most difficult railroad terrain in the world, according to Wiener (1930). He points out that nearly all of the graidents on the 264 miles of railway exceeded 2% and many top 4%. Moreover, it twisted and turned with 320-ft radius curves following one after another. They worked Cordova-Boca del Monte.

The FCM book shows that the first 12 came over a long period from 1889-1901. See Locobase 3634 for the 1902 batch.

Mexican Railway Class R-1 (Fairlie) 1902 (Locobase #3634)

This batch was the second series of the first Fairlie design and began delivery in 1902; see Locobase 6967 for the first batch.

Mexican Railway Class R-2 (Fairlie) (Locobase #1362)

Mexican Railway Class R-3 (Fairlie) (Locobase #1363)