Great Northern 2-6-2 "Prairie" Locomotives in the USA

Class J-1 / J-2 (Locobase 5388)

Data from table in the June 1907 AERJ and GN 1916 Locomotive Diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange collection. See also DeGolyer, Volume 29, pp. 8+, Volume 30, p. 10+; and "New Locomotives for the Great Northern," Railroad Gazette, Volume XLI [41], No 17 (26 October 1906), p. 371.Works numbers were:

1906

27792-27793, 27808-27809, 27816-27818, 27829 in March;

27867-27872, 27878-27879, 27919-27921, 27945-27947, 27955-27956, 27973-27975 in April;

28418-28419, 28428-28430, 28452. 28496 in June;

28522, 28527-28528, 28548, 28637-28638, 28648, 28654, 28671 in July;

28687, 28704, 28756-28757, 28782-28783. 28851 in August.

1907

30080, 30093-30095, 30126, 30134-30138, 30171-30175, 30201-30202, 30236-30239, 30287, 30293, 30312 in February;

30332-30335, 30393-30397, 30440, 30467-30468, 30489, 30502, 30513, 30527-30528 in March; 30560-30561, 30568-30569, 30580, 30600-30601, 30654-30658, 30672, 30717-30718, 30733, 30735-30738 in April;

30756-30759, 30778-30780, 30796, 30812-30813, 30838-30840, 30849, 30861-30862, 30879-30882, 30898-30900, 30937, 30957, 30967-30968 in May;

30999-31000, 31005-31007, 31035, 31082, 31100, 31117-31120 in June

These were Prairie freighters to move fast freights across relatively flat profiles. They featured a long Belpaire firebox, low drivers, and a long wheelbase; their heating surface areas were identical to those of the Atlantics bought in the same year (Locobase 6945). Possibly because the front axle "filled in" the gap under the boiler left by the layout of the two axles under the Atlantic's boiler, its axle loading was considerably lower at 44,000 lb (19,558 kg) than the other two axles.

In addition to the 50 in the 1906 order (manufactured from March to August), the GN bought another 100 J-2s a year later.

Baldwin produced that latter batch in February-June 1907. Baldwin's specs showed a slightly smaller grate area (53.4 sq ft/4.96 sq m), but GN 1916 diagram and table gives exactly the same dimensions as the J-1.

1549 was completed by Baldwin with a Schmidt superheater; see Locobase 12959; another was designed to take Baldwin's smokebox superheater, but Locobase can't confirm

Unlike Santa Fe's experience, however, the Great Northern wasn't really satisfied with these 2-6-2s. Their use in fast freight brought to light the wheel arrangement's instability at speed. Most were relegated to branch-line service.

25 were fitted with superheaters; these are shown in Locobase 8853. Many of the others were converted to oil-burning and trailed tenders carrying 4,500 US gallons (17,034 litres) of oil instead of coal. These wer 1506, 1532, 1523, 1517, 1528, 1519, 1525, 1641, 1643, 1598, 1619, 1553, 1616, 1634, 1612, 1570.

Seventeen of these engines made their way to the Spokane, Portland & Seattle in 1908 as their F-1 class, road numbers 450-466. The GN road numbers were 1555, 1565, 1575, 1577, 1581, 1583, 1587, 1593, 1683, 1603, 1589, 1595, 1597, 1600-1601, 1566, and 1549, respectively.

Fifteen were converted to H-6 class 4-6-2s in 1923.

Class J-1-S, J-2 -S (Locobase 7492)

Data from GN 1 - 1929 Loco Diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange.

Beginning in the 1920s, the GN took the J-1 and J-2 Prairies (Locobase 5388) in hand and upgraded them. They were unusually successful in substituting heating surface in the flues and in the 30 superheater elements for the lost heating surface in the 122 small tubes they deleted - the total heating surface nearly matches that of the saturated boiler. Also, the railroad increased the size of the cylinders while cutting the boiler pressure, thus losing little calculated tractive effort while stressing the boiler less. Several--1514, 1524, 1613, 1624, 1637, 1573-- were converted to oil-burning, substituting 4,500 US gallons of oil (17,034 litres) for coal fuel.

Although the conversions at first retained the motion's slide valves, the class were later fitted with Heron 8 1/2 (216 mm) diameter piston valves. Locobase notes that the road's general foreman James C Heron earlier submitted a patent for an "appartus" that allowed the turning the drive wheels while setting the valves.

The J-1s converted were 1501, 1514, 1524, 1531, 1537, 1539.

J-2s were 1550-1551, 1554, 1556, 1559, 1573, 1582, 1590, 1594, 1596, 1613, 1615, 1624, 1622, 1626, 1631, 1637, 1649.

Class J-3-S (Locobase 8853)

Data from GN 1916 Locomotive Diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. See also Data from Baldwin Locomotive Works Specification for Engines as digitized by the DeGolyer Library of Southern Methodist University Volume 29, p 46. and "Superheating," Railroad Age Gazette, Volume XLV, No 2d (24 June 1908), p.321. Works number was 28851 in August 1906.

Locobase 5388 describes the large class of J-1/J-2 Prairies with saturated-steam boilers supplied to the Great Northern over a short period. By the time this class entered service, German and Canadian reports about the benefits of superheating led several American railroads to explore the devices. Of particular interest was the smoke-tube superheater developed by Wilhelm Schmidt, although the Canadian Pacific used a variation called the Vaughan-Horsey.

The GN ordered a single J-1 (and a single freight engine) that retained the Belpaire firebox, but featured a Schmidt superheater. 1549 was fitted with 12" piston valves and adopted a Baldwin variation of the Walschaert valve gear. The Baldwin specs from 1906 suggested the 1549 developed tons more tractive effort than did the original J-1s, yet the TP figure supplied on the same page is 26,600 lb. The boiler pressure is given as 210 psi, but the 1916 figure probably reflected practice very soon after delivery. Also, Baldwin's calculation of superheater area in 1906 came up with 566 sq ft.

Two years later, the RAG reported on the results of trials: "In passenger service a test on the Kalispell division showed a saving of 13 per cent in water and 14% per cent in coal per car mile, while in freight service on the Wlllmar division the saving was 30% per cent in water and 28% per cent in coal per ton mile, the coal figures being 137% for the simple and 98 pounds for the superheaters per 1,000 ton miles, both very satisfactory figures for prairie type engines in freight service on an undulating road. The company also reports a comparison for nine months between a superheater freight engine and a similar simple engine, showing 137 pounds of coal per 1,000 ton miles for the superheater against 171 for the simple, and a cost for repairs of 4 cents per mile against 3.87 cents, a reduction in the coal consumption of 20 per cent with practically the same cost of repairs.

Originally designated J-3, it was redesignated J-1-S (probably when the railway decided it would be a lone engine). In July 1925, the GN sold 1549 to the Spokane, Portland & Seattle as their F-1 (466). As the F-1, the engine ran until May 1948.

Class Q1 (Locobase 8859)

Data from GN 1916 Locomotive Diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. (Thanks to Chris Hohl for a 24 May 2019 email that suggested the revised wheel arrangement symbology and for correcting or adding the right wheelbase, tender weight, and compound cylinder dimensions.)

A true curiosity, this locomotive had a leading truck, three driven axles, and a trailing axle under the cab. A Baldwin Prairie (Locobase 5388), right? especially with the 69" drivers and 22" x 30" cylinders? But then under the tender was a pair of low-pressure cylinders, another set of three drivers and a trailing truck. So it was an articulated compound Mallet, right? Well, no, because after all it was the rear engine that pivoted, not the front. And it isn't really a tank engine, although it had drivers under the water and coal bunkers.

Other locomotives had tender boosters, but none of those supplied anything like the power that this compound system offered. The diagram shows clearly that the "booster" was in fact the power system from another Prairie minus the leading truck. Its four axles replaced the four axles of the original tender (although probably only the rearmost axle had any lateral play).

Beyond that Locobase has no information about this experimental locomotive.