ALCO delivered the first of 12 of these new locomotives in December of 1926, with the balance arriving early in 1927. Designated Class A, and assigned road numbers 2600 through 2611, they had 73" drivers, 28 x 30 cylinders, a boiler pressure of 210 psi, a tractive effort of 57,500 lbs and weighed 426,000 pounds. Later, the boiler pressure was raised to 240 psi which increased the tractive effort to 65,700 pounds. These were the first of this new wheel arrangement and thus were to be the namesake, with the name "Northern Pacific" selected which was very quickly shortened to just "Northern".
In 1930, the Timken demonstrator came to the NP. While being tested, it suffered crown sheet damage. The NP bought it, repaired it and put it on the roster as number 2626.
Ten Class A-2s (road numbers 2650 through 2659) came from the Baldwin Locomotive Works in 1934. These heavy "Northerns" had 77" drivers, 28 x 31 cylinders, a boiler pressure of 260 psi, a tractive effort of 69,800 lbs and a weight of 489,400 pounds.
In 1938, eight Class A-3s (road numbers 2660 through 2667) were delivered from Baldwin. The specifications for these locomotives were identical to the Class A-2s except that each weighed 2,400 pounds more.
The eight Class A-4s (road numbers 2670 through 2677) were delivered by Baldwin in 1941, and had the same basic specifications as the A-3s, but were different in appearance, with 14 wheel centipede tenders and vestibule cabs, and they were 1000 pounds heavier.
The last of the 4-8-4s purchased by the NP were the Class A-5s, which were built by Baldwin in 1943. This group of 10 (road numbers 2680 through 2689) were built to the same specifications as the Class A-4s, but turned out to be 16,200 lbs heavier. They were among the heaviest of all the Northern type locomotives ever built, at 508,500 pounds only the Santa Fe Class 2900s weighed more.
There are no surviving Northern Pacific Northerns.
|Class||Road Numbers||Yr. Built||Builder||Locomotive Wt.|
|A-1||2626 (ex Timken 1111)||1930||ALCO||477,500 lbs|
Valve motion had limited cutoff. These were the very first 4-8-4s to be bought and go into service, that event occurring in 1926. One significant factor in the development of the big-firebox version of the 4-8-2 Mountain was the predominance of low-calorie "Rosebud" coal in the NP's region.
According to Richard Drury (1993), tests on trains while trailing a dynamometer car revealed two main truths: "The engines produced more horsepower when they were worked hard, and they were slightly under-boilered." Drury also notes notes one unusual design feature - the outside cradle frame carrying the trailing truck, a feature shared with the Canadian National and the Chicago & North Western Hs (Locobase 252).
The power was evident in several service areas, according to Drury, who cites the substitution of 1 A for 2 Pacifics on passenger trains between Glendive, North Dakota and Jamestown, Mont and less-frequent call for helpers on the Livingston-Missoula stretch. They also could run longer without an engine change.
There were break-in glitches, Drury notes, including "...initial difficulties with driver bearings that resulted in broken axles." So big and powerful a locomotive would find its full voice when fitted with roller bearings, as Timken would demonstrate in 1930; see Locobase 930.
Most eventually converted to oil-firing and gave up their booster engines in the late 1940s.
Similar to the earlier Alco-built 4-8-4s (Locobases 272 and 930) with the important addition of Timken roller bearings on the axles. The prototype was given the number 1111 and known as the "Four Aces." She was later taken into NP service as the 2626 (Locobase 930). The A-2 was ordered from Baldwin and introduced disc driving wheels. The A-2s had a cast-steel locomotive bed with integral cylinders. 12" (305 mm) piston valves with 8 1/4" (210 mm) travel supplied the cylinders with steam.
Baldwin's report sheds a great deal of detailed light on this class. Like all NP Northerns, the A-2s burned Rosebud coal, a fuel with 22-28% moisture content and 7-9% percent ash. Compared to more typical locomotive coal, which were rated at 11,000-15,000 BTU/pound, Rosebud generated 8,750 BTU. Such low-grade brown stuff was laid by a Standard modified type B stoker on a large grate. The firebox was joined to a very long combustion chamber to extract every last BTU.
Five of the locomotives had a Worthington preheater and the latter five had the Wilson feedwater conditioner. According to his patent filing of 10 October 1929, Lyndon Wilson's conditioner was mounted in the tender instead of on the boiler because it was designed to preheat, treat, and clean the water at the same time. (Patent 1,901,216 was awarded on 14 March 1933.) It used a separate hot-water tank in which the impurities dropped to the bottom and the clean, treated, and preheated water was drawn off the top to supply the boiler. 2655 to 2659 still had the Wilson conditioner in 1944.
Baldwin's report naturally proclaimed that the A-2s were "doing fine work in heavy passenger service." Six of them then operated on the 664 miles (1,069 km) between Jamestown, NDak and Livingston, Mont over the Fargo and Yellowstone Divisions. In summer months, train loads averaged between 12 and 19 cars over ruling grades of 1.1%. Mileage averaged 9,000 (14,490 km) during the summer months and a little less in the winter.
The other four A-2s wrangled trains between Livingston and Missoula, Mont, some 240 miles (386 km). This profile provided a sterner test of 2.2% out of Butte and Helena and 1.8% between Livingston and West End. On the steeper grades, such trains needed a helper.
See Locobases 932 (A-3), 933 (A-4), and 274 (A-5) for later Northern Pacific 4-8-4s.
NB: The direct heating surface (including the firebox heating surface) is an estimate calculated by subtracting the calculated tube heating surface from the reported total evaporative heating surface.
As suggested by the 1937 account of the A-2s success on the road (Locobase 272), this class was virtually identical except for slightly larger flues and a few more tubes. 14" (356 mm) piston valve travel was 8 inches (203 mm). The Wilson feedwater conditioner was not installed in the tenders of these engines.
The class was divided between the NP (8) and the Spokane, Portland & Seattle (3).
This octet continued the basic Northern Pacific Northern line with a slight adjustment in heating surface area due to an increase in the number of tubes and and a reduction of the flue count because of the 1/4" (6.3 mm) greater diameter. Firebox heating surface, which is not shown on the 1949 diagram from which the data are taken, included 99.5 sq ft (9.25 sq m) in six circulators. 14" (356 mm) piston valve travel was 8 inches (203 mm).
Like the others, this set had Timken roller bearings. Unlike the earlier engines, though, the A-4s (and A-5s) had vestibule cabs that offered more protection against Northern Tier winters.
Firebox heating surface included 99.7 sq ft (9.25 sq m) in circulators (6) and arch tubes. Heavier engines with larger cylinders and higher drivers. The low-calorie "Rosebud" coal must have streamed through the stoker at a prodigious rate, considering that the grate demand factor is nearly as high as other engines burning higher-grade coal. These engines had a very high adhesive weight.
This class was booked to run 999 miles without an engine change, a record for a coal-burning locomotive.
http://www.steamlocomotive.com and later conducted a correspondence with Locobase. See also the detailed account in Allen Merta, Wig-Wag (published by Eastern Iowa Division Mid-Continent Region / NMRA), Vol 3, No 9 (September 2007), pp. 3-11 at http://eid.mcor-nmra.org/PDFs/WigWag200709.pdf . This engine demonstrated Timken's roller bearings while bearing the number 1111. (Its nickname was "The Four Aces.") Timken's builder's card shows a different distribution of weights among the axles for the 235-psi boiler and for the 250-psi vessel. Locobase uses the latter in its specifications.
Allen Merta summarizes some of the specifics of the roller-bearing installation: "The roller bearings on all driving axles surrounded the axle a full 360 degrees. This took up the piston thrust in all directions. Driver pounding was reduced to a minimum. Bearing overheating was eliminated. The roller bearings were completely immersed in oil on all wheels. The oil needed to be changed only 2 to 4 times per year." Other than the roller bearings on all axles (a big exception), the locomotive was a typical big Alco. In addition to the Walschaert valve gear and Type E superheater, the 1111 was fitted with a Worthington feed water heater, a tender-mounted stoker, a Franklin booster, and a ALCO Type G power reverse gear.
Alexander (American Locomotives, 1950, p. 178) describes its tour of 88,992 miles on fourteen railroads including turns on elite passenger varnish such as the New Haven's Merchants Limited and the C&O's Sportsman. "On the Pennsylvania it handled twelve passenger cars up the Allegheny mountain grade without a helper and even saved three minutes on the standard schedule." Merta comments on the assiduity of most of the trial railroads: "Twelve of the 14 railroads seriously tested the Four Aces. In freight service, the locomotive made 328 freight runs pulling an average of 83+ cars per trip, an average speed 29.8 mph. On the Chesapeake & Ohio, Four Aces started and pulled a 132 car coal train weighing 9,864 tons. In passenger service, Four Aces made 227 runs with an average of almost 11 cars per trip at an average speed of 41.2 mph."
Most often remembered from this tour were the publicity shots in which a few men (or fetchingly clad young women) pulled the 1111 along a track to demonstrate the silky smooth, low-friction qualities of the roller-bearing installation.
Former employee Daniel Simon forwarded a photo of the 4 Aces and wrote at http://www.railpictures.net/viewphoto.php?id=307329&nseq=0#remarks (last accessed 9 June 2012) and quotes the Wikipedia entry for the 1111 as saying, "A total of 52 different parts manufacturers agreed to supply their parts for the locomotive 'on account' until the locomotive operated over 100,000 miles (161,000 km).
The NP's number for this engine was 2626. Ironically, Drury (1993) reports, the Northern Pacific wasn't particularly persuaded by the tests because the 1111's firebox wasn't designed to burn the railroad's low-grade coal. It stayed on the NP largely because it had suffered crown sheet damage while on the railroad (the last of the 14 to trial it) and Timken didn't want it back.
Merta reports: "The [crown sheet] damage occurred when locomotive crews allowed it to run low on water near Auburn, Washington." He also contends that the NP set up a adversely biased comparison between its A-class 2607, which run only 3,465 miles since fitted with new driving boxes, and the 1111, which had seen little maintenance in the previous 91,780 miles.
The NP purchased the engine on 8 February 1933.
After repairs, the NP found 1111/2626 "durable and inexpensive to operate." A later Northern Pacific diagram shows slightly different distribution for the boiler's evaporative and superheater heating surface areas. EHS is given as 5,061 sq ft while the superheater is credited with 2,157 sq ft (most likely due a difference in how one handled Type E superheater areas from the Timken calculations.)
Adhesion weight came out closer to the 235-psi figure, even though NP operated the A-1 at 250 psi. Adhesion is given as 244,900 lb (111,985 kg) while the total engine weight comes in 12 tons lighter at 399,000 lb (180,984 kg). The A-1 pulled Trains 1 and 2 between Seattle and Yakima, Wash at first. Despite a less than enthusiastic reception, the NP could only agree that Timken's claims had been borne out, says Merta: "When the first complete "general shopping" of 2626 was done in 1934, the driving boxes were disassembled. The roller bearings were in excellent condition and could have returned to service, but
the decision was made to replace them because the locomotive had 280,000 miles on it."
By 1935, the 2626 ran the 656-mile (1,056-km) stretch between Seattle and Missoula, Mont. Throughout its career, the 2626 trailed the same tender, which Timken's T.V.Buckwalter conceded was too small.
2626's last run came on 4 August 1957 when it hauled a train from Seattle to Cle Elum, Wash (about 80 miles away) and returned.
|Specifications by Steve Llanso|
|Railroad||Northern Pacific (NP)||Northern Pacific (NP)||Northern Pacific (NP)||Northern Pacific (NP)||Northern Pacific (NP)||Northern Pacific (NP)|
|Road Numbers||2601-2612||2650-2659||2660-2667||2670-2677||2680-2689||1111 / 2626|
|Locomotive Length and Weight|
|Ratio of driving wheelbase to overall engine wheebase||0.43||0.43||0.43||0.45||0.43||0.42|
|Overall Wheelbase (engine & tender)||90'||96.25'||96.25'||96.75'||97.50'||88'|
|Axle Loading (Maximum Weight per Axle)||63000 lbs||73800 lbs||74000 lbs||74000 lbs|
|Weight on Drivers||251000 lbs||294000 lbs||294000 lbs||294000 lbs||295000 lbs||256000 lbs|
|Engine Weight||418000 lbs||489400 lbs||491800 lbs||502500 lbs||508500 lbs||423500 lbs|
|Tender Light Weight||313000 lbs||387600 lbs||386050 lbs||443300 lbs||443500 lbs||299000 lbs|
|Total Engine and Tender Weight||731000 lbs||877000 lbs||877850 lbs||945800 lbs||952000 lbs||722500 lbs|
|Tender Water Capacity||15000 gals||20000 gals||20000 gals||25000 gals||25000 gals||14550 gals|
|Tender Fuel Capacity (oil/coal)||24 tons||27 tons||27 tons||27 tons||27 tons||21 tons|
|Minimum weight of rail (calculated) on which locomotive could run||105 lb/yard||123 lb/yard||123 lb/yard||123 lb/yard||123 lb/yard||107 lb/yard|
|Geometry Relating to Tractive Effort|
|Boiler Pressure||210 psi||250 psi||260 psi||260 psi||260 psi||250 psi|
|Cylinders (dia x stroke)||28" x 30"||28" x 31"||28" x 31"||28" x 31"||28" x 31"||27" x 30"|
|Tractive Effort||57511 lbs||67073 lbs||69756 lbs||69756 lbs||69756 lbs||63663 lbs|
|Factor of Adhesion (Weight on Drivers/Tractive Effort)||4.36||4.38||4.21||4.21||4.23||4.02|
|Firebox Area||485 sq. ft||512 sq. ft||490 sq. ft||589.70 sq. ft||589.70 sq. ft||483 sq. ft|
|Grate Area||115 sq. ft||115 sq. ft||115 sq. ft||115 sq. ft||115 sq. ft||88.30 sq. ft|
|Evaporative Heating Surface||4884 sq. ft||4964 sq. ft||4703 sq. ft||4673 sq. ft||4672 sq. ft||5120 sq. ft|
|Superheating Surface||1992 sq. ft||2174 sq. ft||2095 sq. ft||1930 sq. ft||1930 sq. ft||2100 sq. ft|
|Combined Heating Surface||6876 sq. ft||7138 sq. ft||6798 sq. ft||6603 sq. ft||6602 sq. ft||7220 sq. ft|
|Evaporative Heating Surface/Cylinder Volume||228.43||224.69||212.87||211.52||211.47||257.54|
|Computations Relating to Power Output (More Information)|
|Robert LeMassena's Power Computation||24150||28750||29900||29900||29900||22075|
|Same as above plus superheater percentage||31154||37375||39169||38571||38571||28477|
|Same as above but substitute firebox area for grate area||131387||166400||166894||197785||197785||155768|