4-6-2 "Pacific" Locomotives in the USA

The 4-6-2 type, or "Pacific", as the class was known, was the predominant steam passenger locomotive during the first five decades of the 20th century. Between 1902 (when the first North American locomotives of this wheel arrangement were produced) and 1930, about 6800 locomotives of the type were built for US and Canadian service. One reasonably accurate estimate of the number of steam locomotives produced for Class 1 US and their Canadian equivalents are 75,000 units. Thus, Pacifics made up about 9% of total steam locomotives built. To put this figure in perspective, a reasonable figure for US domestic Class 1 steam locomotives installed between 1931 to 1949 is 2500/2600. ¹

Everyone is aware of Pennsylvania's fleet of 425 K4s locomotives, the largest such class in the world. However, for another example of the pervasiveness of the Pacific locomotive type, consider this: as of January 1, 1946, in the New York Central diagram book, the number of J class 4-6-4 passenger locomotives was 274; the number of K class 4-6-2s was 368, although, in all fairness, 102 of the Pacifics were class K11, built as fast freight locomotives, and used for local freight to a certain extent. Forget about numbers for the moment: Of the "Trunk Lines" serving New York City, the "Anthracite Roads" and the New England lines, the New Haven and the Boston & Maine in particular, only the New York Ontario & Western did not roster Pacifics for passenger service. Of the four principal passenger-carrying railroads of the South, only one did not rely on Pacifics as their main passenger power. On the Atlantic Coast Line, the Louisville & Nashville and the Southern, Pacifics, supplemented on the Southern and the L&N by a modest fleet of 69 inch drivered 4-8-2s, and on the ACL by 12 marginally successful 4-8-4s with 80 inch drivers, the Pacifics were the passenger locomotive until the arrival of the diesel. It should be noted that Pacifics came in a wide variety of designs, not too surprising for a class built for so many different roads over such a long period of time, They were built with driver diameters of as low as 67 inches up to 80 inches, and with steam pressures from 170 up to 250 psi. Further, they were built for many services; some for fast freight, others for dual service and many more, probably the most, for express passenger service. Tractive efforts ranged from 25/26.000 lbs to over 50,000 lbs.

West of the Mississippi, there was a similar pattern, although the transcontinentals tended to begin purchasing of 4-8-2s with relatively high (73 - 74 inch) drivers in the 1920s. One interesting sidelight is that the Union Pacific and the Southern Pacific both had Pacifics on a design similar to those on the Illinois Central. This was a result of all of these roads being controlled, from 1902 to 1913 by Edward H. Harriman, who was a great believer in standardization. ²

The Pacific, as a type, is generally considered to be an enlargement of the Atlantic (4-4-2), although it also had a direct relationship to the Ten-Wheeler (4-6-0) as well. ³ To show the relationship between the three classes, and to also show how rapidly locomotive designs were developing, consider the following examples:

• The first is a 1900 4-6-0 of the LS&MS; it had 80 inch drivers, a 200 psi boiler, and a starting tractive effort of 23,800 lbs. The grate area was 36.6 square feet, with a firebox length of 120 inches and a width of 44 inches. Tube length was 15 feet, and total heating surface was 2890 square feet.
• The second example is a 4-4-2, built for the New York Central & Hudson River in 1905. This design had 79-inch drivers, 200 psi and a starting tractive effort of 23,700 lbs. Its grate area was 50 square feet, with a firebox length of 96 inches and a width of 75 inches. Tube length was 16 square feet, and total heating surface, 3306 square feet.
• The final example is ALCO's 50,000 locomotive, a prototype for many future designs, including, arguably, the K4s. This unit had 79-inch drivers, 185 psi and a starting tractive effort of 40,800 lbs. The grate area was 59.7 square feet, the firebox dimensions 114 x 75 inches. The tube length was 22 feet, and the total heating surface 4048 square feet.

Note that the LS&MS 4-6-0 had its firebox inside the drivers. Effectively, high drivered locomotives of this wheel arrangement had gone as far as they could go, for there was no way in which locomotives with such high drivers could mount their boilers on top of the fireboxes and still meet clearance requirements, and, probably more significant, have sufficient stability. A few years later, the NYC was building dual service 4-6-0s with a 55 square foot grate, but these locomotives were equipped with 69-inch drivers, and the firebox would fit over the drivers.

The NYC&HR 4-4-2 was a remarkable locomotive. All were equipped with superheaters during their career, and were very close in capability to the much larger Pennsylvania E6s Atlantic. But, the continued growth in the length of passenger trains, as well as the universal use of all steel cars, caused train weights to exceed the capacity of any two axle locomotive, even given the excellent track of the NYC.

ALCO's 50,000, which, as many readers are aware, had a long and successful career after its service as a demonstration locomotive as Erie 2509. It was one of, if not the first, example, of the "heavy" Pacific, designed and built to handle the passenger trains which were approaching 800/900 tons behind the tender, and at a time when such heavy trains were required to operate at speeds in the 60 mph range. It was the only one of the three examples to be superheated, with a superheating surface of 900 square feet. By later standards, it would have been considered "over cylindered", with 185 psi. and 27 x 28 inch cylinders. Possibly, the relatively low working pressure was a product of an attitude prior to the First World War that superheated locomotives could operate successfully with lower boiler pressure, resulting in lower fuel costs and less boiler maintenance, and, also, because of lubricating problems with high pressure superheated steam. Another possibility may have been nothing more complicated than a desire to keep a good factor of adhesion for a locomotive which, as a demonstrator, was going to operate on some railroads which did not have the sturdiest track in the United States.

Compounds and Superheaters

An inventory of main-line steam locomotives in the United States as of 1904 shows that, of 233 Pacifics in service, 85 were four cylinder compounds. The reasons for the introduction of compounds, to oversimplify, was that they were more economical on fuel, and, a factor particularly important to those railroads out West where water supplies were limited, and the water was of poor quality, compounds also used less water. Thus, the Atchison, Topeka and Santa Fe, which had compounds in many wheel arrangements, built 196 four cylinder "balanced" compound Pacifics between 1905 and 1914, although it should be pointed out that at the same time the road was also buying two cylinder simples of the same wheel arrangement. ⁴

The reason for the decline of the compound rigid frame locomotive in the United States was the general introduction of the Schmidt type superheater, which offered both fuel economy and better performance without the maintenance and construction costs of the compounds, although, in all truth, superheaters presented problems as well. So, most of the Santa Fe compound Pacifics were converted to two cylinder superheated locomotives, and enjoyed a long and successful service life. (During the mid 1920s, largely promoted by ALCO, there was considerable interest in three cylinder simple locomotives. There were a few three cylinder Pacifics built for Mexican service, but there was no series production of such Pacifics for any US or Canadian line.)

USRA Pacifics

There were two classes of 4-6-2 locomotives designed and built under Government auspices; one, the generally called "light", more correctly 4-6-2A, with 55,000 lbs weight per driving axle, and "heavy", 4-6-2B, with 60,000 lbs per driving axle. In common with all USRA road designs, these locomotives had many progressive features. For one thing, they had adequate firebox-boiler dimensions and intelligent cylinder proportions, resulting in a "100% boiler". In other words, the boiler horsepower was greater than the cylinder horsepower. They were all equipped with combustion chambers, to add to better fuel consumption and increase the firebox-heating surface while maintaining useful tube lengths. Since the USRA locomotives were built for wartime service, there were obviously more freight types constructed than passenger: thus, there were 81 USRA "light" Pacifics built as opposed to 625 55,000 lb axle loading 2-8-2s and 20 "heavy" 4-6-2s against 233 "heavy" Mikados.

The "light" 4-6-2 was a beautifully balanced locomotive. It had a large boiler, as, indeed, did all of the USRA locomotives. 73 inch drivers were ideal for a period when train speeds (and, particularly, under wartime restrictions) when 70/75 mph speeds were sufficient. 25 x 28 cylinders were a good proportion, but the locomotive design had a starting tractive effort of over 40,000 lbs, at time when relatively few Pacifics had such a high starting tractive effort. The railroads to which these locomotives were assigned were the Atlantic Coast Line, which received 45, plus 25 non-USRA follow-ups; the Baltimore & Ohio, 30 units from the USRA, 15 subsequently, and the L&N, which received 6, later ordering 20 more. One of the most interesting developments was on the ACL. This company ordered 165 further Pacifics, identical to the USRA ones, except that they had 69-inch drivers and 210 lbs boiler pressure. These Pacifics, with a starting tractive effort of 45,275 lbs. On the relatively flat route between Richmond and Savannah, these Pacifics had a freight tonnage rating of 4400 tons (unadjusted) in either direction. And, with their 69 inch drivers. they were more than capable of maintaining the moderate passenger train schedules of the 1920s and early 1930s. ⁵

The "heavy" was a somewhat more complicated design. Basically, it was a development of ALCO's 50,000, mentioned earlier, through ALCO built PRR 3395 of 1911, considered to be the precursor of the famous (deservedly so) PRR K4s class to what seems to have been a 79 inch drivered version of the K4s with a conventional firebox. The demand for Pacifics of such a capacity for tonnage and speed was rather limited; the NYC, for example, was content to compete successfully with the Pennsy with 79-inch drivers and only 30,000 or so pounds of starting tractive effort. Thus, the only railroad which received USRA 462B engines was the Erie, which got 20, and later ordered 11 more on its own. The only other railroad to order this design was the Chicago and Eastern Illinois, which required them for its part in handling Florida trains between Chicago and Evansville, Indiana. But, from the USRA heavy Pacific was developed one of the best known American steam locomotives, the Southern Railway Ps4 class. These engines, identical with the basic USRA designs except for 73-inch drivers, had a starting tractive effort of 45,000 lbs. Again, used in services which did not require maximum speeds with long trains on moderate grades, this was a type which, outside of the fact that they were one of the most attractive classes ever to run on steel rails, were outstanding performers. As was the case of all the classes enumerated above, they lasted as the main passenger power of their railroads until the diesel-electric came along to replace them. ⁶

Notes

1. These figures are from Alexander, "The American Locomotive" and Bruce, "The Steam Locomotive in America". They are derived from the figures published in two Simmons-Boardman trade magazine, the weekly "Railway Age" and the monthly, 'Railway Mechanical Engineer". There are slight differences: Bruce says there were no locomotives delivered in 1930, Alexander lists 5. Unfortunately, Canadian deliveries were not broken out in the totals after 1930. Looking in my copies of the 1941 and 1947 editions of the "Locomotive Cyclopedia", also published by Simmons - Boardman, I come up with a total of 940 locomotives built in Canada between 1931 and 1946. Since, to the best of my knowledge, non-steam locomotives were not being built in Canada then, this is a reasonable figure. Except, of course, that it almost certainly includes locomotives for export.
2. One of the most interesting studies of why certain locomotives were designed and ordered appeared in "Railroad History" 135, Fall, 1976. This was an article written by Robert L. Frey, entitled "The Biography of a Heavy Pacific". The locomotives in question were the Northern Pacific's Q-6 class, and Mr. Frey covers in great detail the reasons for the choice of this particular class, and what the perceived alternative designs were. There is an interesting sub-text to this, also. One of the rejected alternatives was a 73-inch drivered 4-8-2, roughly an enlargement of a USRA heavy Mountain. But, a few years later, this concept became a reality, but in the form of a 4-8-4, the change being to accommodate a larger grate to facilitate the use of "Rosebud" coal, an on-line fuel which seem to fit almost into the category of "semi-lignite." If the original concept had been followed, and the NP had purchased 4-8-2s, then the first 4-8-4s would have been DL&W 1501-1505, and the type name "Pocono" might have spread beyond eastern Pennsylvania and western New York.
3. At the time we are discussing, roughly 1895 through 1905, there were also numerous "Prairie" (2-6-2) types in service. These locomotives, however, are not considered to be ancestors of the Pacifics. They are best regarded as improved versions of the 2-6-0, with a larger firebox supported by the trailer truck. As such, with drivers of from 63 to 69 inch diameters, they were an excellent class for light freight and dual service purposes. Unfortunately, when, as happened on several roads, the ultimately parent of New York Central Lake Shore and Michigan Southern for one, 2-6-2s were designed for express passenger service with high drivers (up to 81 inches), stability problems developed. In the LS&MS case, Prairies had to be converted to Pacifics by the addition of a four-wheel pilot truck.
4. For reasons of stability, practically all compound locomotives built for high-speed passenger service in the United States had four cylinders. There were three types; the Vauclain, a Baldwin design with outside cylinders, driving on a common crosshead; balanced, another Baldwin concept, with low pressure cylinders outside and high pressure inside, all driving on a common wheel, the high pressure cylinders driving through two drive rods to a crank axle on the main driver. The third was the Cole, an ALCO design, also with inside and outside cylinders, but with the inside high-pressure pair driving on a crank axle on a different driver than the outside low pressure.
5. The boilers and fireboxes of the USRA designs were both large and efficient for the period. Thus, on many roads, an increase of ten pounds in psi could be readily accommodated, with a consequent enlargement of starting tractive effort. Since the USRA locomotives, in general, were marginal designs in terms of adhesive weights, this tended to make a group which were already considered slippery even more so. However, as time went on, roads compensated: thus, the magnificent J-4 2-8-2 of the Louisville & Nashville, with 200 psi boilers and 63,000 lbs starting tractive effort, were constructed with 16,000 pounds more weight on drivers than the USRA 282 B Mikes.
6. On both the L&N and the Southern, 4-8-2s were used in passenger service on heavy grades: One class on the Southern of its own design, plus USRA 482As, and USRA 482As on the L&N.

Introductory information by the late Edward G. Weinstein.

---

• Railroads that used 4-6-2 locomotives in other countries