Class 6400/D12 (Locobase 10957)

Data from []; and "JGR Class 6400" in Wikipedia at . Works numbers were 26143-26127 in Se[tember 1902.

Thoroughly American in appearance, the 6400s served the IGR?JGR for close to 30 years. They were "mainly used in the Hamamatsu region", reported Wikipedia. Their dimensions lent themselves to work on Tokaido Main Line "hauling special express trains between Numazu and Nagoya" [south of Tokyo]. In their final years, 6400s hauled trains in the Kinki region San'in regions among others.

---

Class 6700 (Locobase 3801)

Data from Hiroshi Takagi, both from his website ([]) and personal communication August 2000. His indispensable assistance is gratefully acknowledged. See also Saito Akira, "Overview of Japa's modern steam locomotives", originally published in Japan's Rail Magazine in May 2000, archived at []; and Tsutsumi Ichiro, "A Short History on Training Railway Engineers in Meiji Japan", Japanese Railway & Transport Review No. 54 (December 2009), pp. 34-40; both last accessed 13 January 2019.

Akira Saito wrote that this design was the first domestically mass-produced Japanese locomotive. It also showed the growing influence of Prussian locomotive design. Indeed, says Saito, "it was a miniaturized P4." The design is credited to Ota Kichimatsu, who first cut his teeth as an assistant for Francis Trevithick. When he moved to Kawasaki Shipyard Company's Rolling Stock Manufacturing Division, he worked with Shima Yasujiro and Asakura Kiichi to create a producible version that included the first use of Walschaert's valve gear in an engine pulling a separate tender.

---

Class 6750 (Locobase 3803)

Data from Hiroshi Takagi, both from his website ([]) and personal communication August 2000. His indispensable assistance is gratefully acknowledged. See also Saito Akira, "Overview of Japa's modern steam locomotives", originally published in Japan's Rail Magazine in May 2000, archived at []; and Tsutsumi Ichiro, "A Short History on Training Railway Engineers in Meiji Japan", Japanese Railway & Transport Review No. 54 (December 2009), pp. 34-40; both last accessed 13 January 2019.

This design, also worked out by Ota Kichimatsu at Kawasaki Shipyard Company, was essentially the 6700 (Locobase 3801) with a superheater. Tutsumi notes that this sextet was succeeded immediately by the 6760 class described in Locobase 3804.

---

Class 6760 (Locobase 3804)

Data from Hiroshi Takagi, both from his website ([]) and personal communication August 2000. His indispensable assistance is gratefully acknowledged. See also "Locomotive Practice on the Japanese Railways", Railway Engineer, Volume 44, No 2 (February 1923), pp. 68-70.

Saito Akira, "Overview of Japa's modern steam locomotives", originally published in Japan's Rail Magazine in May 2000, archived at []; and Tsutsumi Ichiro, "A Short History on Training Railway Engineers in Meiji Japan", Japanese Railway & Transport Review No. 54 (December 2009), pp. 34-40; both last accessed 13 January 2019.

Note that this differs from the immediately preceding 6750 class (Locobase 3803) in the tube and flue layout. The tubes are smaller in diameter and more numerous, as are the flues. Takagi explains that most Japanese design avoided packing the boilers with tubes and flues. Although this meant they steamed less readily, they were thought to be stronger. Boiler pressure in this class later increased slightly to 185 psi, (12.76 bar) adhesion weight rose to 62,920 lb (28,540 kg), and engine weight went to 103,132 lb (46,780 kg).

The class, noted the 1923 RE report, were "generally assigned to local passenger trains and express trains of moderate weight."

---

Class AF /D6/ 5500/ 5630 (Locobase 11042)

Data from [] and the Japanese-language blogging site [] and []. See also Hirota Naotaka, Steam Locomotives of Japan (Tokyo and Palo Alto: Kodansha International Ltd, 1972), p. 94.

Beyer, Peacock, and Co and Neilson and Company each supplied six locomotives. BP's works numbers were 3597-3602, road numbers 142-147 (later redesignated 5500 by the JNR), while Neison's works numbers were 4900-4905, accounting for 174-179 (later redesignated 5630).

Hirota commented that the 5500s were expensive to the point of limiting their Japanese customers to government railways. One example of the refinements he describes were cylinders, "set in a slightly canted position, suffered remarkably little from friction and were used without replacement during the entire life of the locomotive." He added that the class's 60 year career benefited from the start from Dubs's production of cylinder steel with a small amount of sulphur.

---

Class AJ (Locobase 11025)

Data from Richard Francis Trevithick , "Particulars of 4 WC Passenger Engine for the Imperial Govt Railways of Japan", 1 September 1895, presented to Institution of Mechanical Engineers, 31 March 1896, archived on the Institution of Mechanical Engineeers at []-locomotive?, last accessed 24 June 2023. Additional information from [] and the Japanese-language blogging site []

---

Class AN/D9/6200 (Locobase 10351)

Data from "Locomotive for the Imperial Railways of Japan," The Locomotive Magazine, Volume IX [9] (11 July 1903), p. 23. See also "JGR Class 6200" in Steam Locomotives, Locomotive Wiki at [], last accessed 21 January 2024.. Dubs & Co works numbers were 3958-3975 in 1900; 4039, 4166-4171 in 1901 (7 for Manchuria). Neilson & Co works numbers were 5633-5662 in 1899, 5721-5732 in 1900.

Smaller than earlier Dubs machines sent to Japan, this locomotive had inclined cylinders and a widespread set of coupled axles. Locomotive Wiki said that the JGR placed several variants in class AN. This entry shows the 1899-1900 deliveries by the two Glasgow builders.

Some of these express passenger locomotives later received superheaters and formed JGR Class 6250. 49 6200s and 6270s (see Locobase 10174) shucked the tender in favor of side tanks and a trailing truck and emerged as Class 1070 4-4-2T locomotives.

---

Class AN/D9/6270 (Locobase 10174)

Data from "New Locomotives for Imperial Railways of Japan," The Locomotive Magazine, Volume VI [6] (February 1901), p. 29; and "American Locomotives in Japan", Locomotive Magazine, Volume XXIV [24], No. 307 (15 March 1918), p. 39. Works numbers were 3958-3974, 4039 in 1900, 4168-4171 in 1901.

The firebox in this "rather expensive" passenger engine used Drummond's design of water tubes, which contributed 72.2 sq ft (6.71 sq m). Probably because of loading gauge limitations, the cylinders had quite a rake for the time so they could located above the bogie. The latter's axles were stabilized by prominent outside longitudinal leaf springs

The later report noted that most Japanese locomotives at the time used the six-wheel tender shown. Its leading axle was fixed, but the latter two rode in a bogie that swiveled.

---

Class AO/5700/D10 (Locobase 3233)

Data from "Schenectady Locomotives for Japan", Railroad Gazette, Volume 30, No 2 (14 January 1898), p.30. Works numbers were 4633-4642 in 1897.

The Railroad Gazette report shows an engine thoroughly American in every respect, although smaller than most North American eight-wheelers. Among the sub-gaugers of the time, however, this was a relatively large passenger locomotive. The Baldwins of the same year (Locobase 10099) had taller drivers, but smaller boilers.

---

Class AP/D11/5160 (Locobase 9619)

Data from Catalogue Descriptive of Simple and Compound Locomotives built by Brooks Locomotive Works, Dunkirk, NY (Buffalo, NY: Matthew-Northrup Company, 1899), pp. 222-223. See also "Eight Wheel Passenger Engine for Japan", Railway and Locomotive Engineering, Volume 11, No 3 (March 1898), pp. 124-125; and See also Hirota Naotaka, Steam Locomotives of Japan (Tokyo and Palo Alto: Kodansha International Ltd, 1972), p. 94. Works numbers were 2878-2883 in December 1897; 2884-2897 in January 1898.Builder info from B.Rumary, 25 Kingscombe, Gurney Slade, Radstock, BA3 4TH, ENGLAND and Jeremy Lambert as supplied by Allen Stanley in March 2004.

Classic turn-into-the-20th-Century North American Eight-wheeler shrunk to 3/4 size for use on Japan's young Cape gauge railway network. Hirota wrote that these were used primarily in the Maibara district of the relatively new Tokaido Line. Notable for their "spring devices", they were "the best of Brooks' products to be imported."

See Locobases 3233 and 10099 for Schenectady and Baldwin engines produced to similar specifications.

---

Class B50 (Locobase 3802)

Data from Hiroshi Takagi, both from his website ([]) and personal communication August 2000. His indispensable assistance is gratefully acknowledged. See also A Profile of Japanese Steam Locomotives website put up in 2007 by "Toki S." at [], last accessed 26 September 2015.

According to Takagi, this class consisted of rebuilt 6700-class engines to which superheating was added. The engines worked mostly in the west until the late 1950s. The last of the class was retired at the JNR Suita works

Note that from 1931 on heating surface areas on Japanese locomotives were calculated on the fire side of the tubes and flues. This followed German practice. A direct comparison with surface areas of North American and British locomotives requires adding approximately 6-10% to the figures shown.

---

Class L/5700/D10 (Locobase 21162)

Data from "Schenectady Locomotives for Japan", Railroad Gazette, Volume 30, No 2 (14 January 1898), p.30; and "JGR Class 5700", in Steam Locomotive, Locomotive Wiki at [], last accessed 15 January 2024. Works numbers were 5773-5775 in February 1901,, 6123-6127 in December, and (once Schenectady became the centerpiece of the American Locomotive Company in 1901) 40407-40411 in May 1906.

Locobase 3233 presents the earlier batches of these Schenectady American-type locomotives.exported by the dozen starting in 1897.

The Baldwins of the same year (Locobase 10099) had taller drivers, but smaller boilers.

---

Class PQ/5130 (Locobase 3037)

Data from Ahrons (1927); and "Locomotive for the Imperial Government Railways-Japan", Railway Signalling (later Railway Engineering), Volume II [2] , p. 241. Works numbers were 2099-3005 in 1876.

Among the first of the main-line sized engines in Japanese service. RS's brief report included positive review of both the locomotive and its crews. About the locomotive, RS wrote that their performance had "given thorough satisfaction." Trailing a load of 32 vehicle, "either wagons or carriages or mixed", an engine would consume 16 lb a mile [4.5 kg/km) of coal.

"We are informed that the native drivers work these engines with remarkable care and regularity, the trains being very punctual. They run about 100 miles [161 km] a day."

---

Class RT // 5300/5400 (Locobase 11040)

Data from [] and the Japanese-language blogging site [] and []

According to the picasaweb photo. two different builders supplied locomotives:

Beyer, Peacock, and Co produced engines with road numbers 77, 78, 79, 80, 81, 82, 83, 84, 86, 88, 90, 92. (Locobase assumes that missing numbers in the diagram had been scrappedby that time). These later were class 5300

Neilson and Co added 117, 119, 120, 121, 122, 123. (Locobase thinks it likely that 118 ws also part of this class. Although built to the same specifications as the Beyer, Peacocks, they were grouped in class 5400 by the JNR.

Train loading was 208 tons up a 1% at 20 mph.

---

Class S / 5490 (Locobase 11044)

Data from [] and the Japanese-language blogging site [] and []

This pair of Beyer Peacocks was rebuilt at the Kobe works of the IGR.

Train loading was 208 tons up a 1% at 20 mph.

---

Class Yo/5800 (Locobase 12770)

Data from Baldwin Locomotive Works Specification for Engines as digitized by the DeGolyer Library of Southern Methodist University, Volume 27, p. 225; and "JGR Class 5800" in Steam Locomotive-Locomotive Wiki at [], last accessed 15 January 2024. Works numbers were 25158-25160 in February 1905.

Locomotive Wiki reports that these three Philadelphian products essentially the Class L (later JGR Class 5700) engines provided by Alco's Schenectady Works (Locobase 21162). Indeed, Baldwin's specification page included a "Hereafter" comment referring to the railway's desire for Philadelphia to follow the specs shown on Schenectady's 75 prints. Changes to "unimportant details" would be permitted as long as they did not affect interchangeability with the rest of the class. Perhaps surprisingly, Baldwin's spec shows the company had permission to strengthen Schenectady-designed components as long as the work preserved interchangeability and didn't push weights beyond specified limiits.

Placed in Class 5800 when most railways were nationalized as the Japanese Government Railways, the trio remained in service on Japan's northernrmost main island until 1925.

---

Class unknown (Locobase 10099)

Data from "Locomotive for the Japanese Rys," The Locomotive & Railway Carriage & Wagon Review, Vol III [3] (August 1898), p. 126. See also DeGolyer, Volume 21, p. 31. Works numbers were 15402-15419.

1897 saw a variety of US Eight-wheeler locomotives heading to Japan to enlarge their passenger-engine studs. Compared to the Schenectadies of the same year (Locobase 3233), these had somewhat smaller boilers, but taller drivers that marked them as purely passenger in intent.

The engines rolled on 61 lb/yard (30.6 kg/metre) rail on lines offering minimum curve radii of 300 ft.

They were retired in 1925-1928.

NB: Firebox heating surface area was derived by subtracting the calculated tube heating surface area from the given evaporative heating surface area.

---

Class unknown (Locobase 10295)

Data from "British Locomotives in Japan," The Locomotive Magazine, Volume VIII [8] (21 March 1903), p. 199. Boiler pressure is an estimate.

These were passenger engines, but not for express service. The whole profile is workaday, but some of the dimensions, such as the grate area, seem unusually high. Locobase hasn't been able to identify the Japanese class ID.