Milwaukee Road 2-8-2 "Mikado" Locomotives in the USA

Class L1 (Locobase 1192)

Data from MILWRD 1930ca locomotive diagram book supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange. See also Jim Scribbins, "Chicago, Milwaukee, St Paul & Pacific (Milwaukee Road) in George H Drury, Guide to North American Steam Locomotives (rev ed.) (Waukesha, WI: Kalmbach Books, 2015), p.134.

These were, according to Jim Scribbins, "basically enlarged Prairies [2-6-2].". 65 sq ft (6.04 sq m) of direct heating surface came from the combustion chamber. On the one hand, being delivered with a saturated boiler puts them in the first period of Mikado development, but rolling on 63" drivers and adding a combustion chamber point toward futher 2- 8- 2 developments.

The shops later fitted some of the class with superheaters.

Only four L1s went to the scrapper before the US entered World War II. Ten of the remaning 16 remained in service in to the 1950s.

---

Class L2 (Locobase 1191)

Data from Milwrd 1930ca Locomotive Diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange.

According to Jim Scribbins writing in Drury (1993), this Mikado design suited the Chicago, Milwaukee, St Paul & Pacific's requirements very well. The Milwaukee shops built 40 in their shops and bought 180 (155 from Brooks in 1912 and 25 from Schenectady in 1914).

After World War I, Baldwin supplied 200 more in 1920-1923 to slightly different designs; see Locobases 15196 and 15197.

Locobase doesn't know why the class began retiring as early as the 1930s, but can tell that some carried on to the end of steam.

---

Class L2-a (Locobase 15196)

Data from DeGolyer Library, Vol 67, pp. 199. Works numbers were:

1920

August 53596;

September 53624-53625, 53658-53660, 53678-53680, 53708-53711, 53754-53756

October 53819-53821, 53839-53843, 53867, 53913-53917

November 53937-53939, 53970-53973, 54000, 54036-54044, 54093-54097, 54099, 54120, 54125-

54127, 54144-54151

December 54152-54160, 54192-54194, 54197

1921

January 54287-54289, 54315-54320, 54348-54359

February 54542

When the Milwaukee decided to increase its Mikado motive power stock six years after building its own and buying from Alco (Locobase 1191), it went to Eddystone. The design had the same power dimensions and weighed about the same. Piston valves still measured 14" (356 mm) in diameter and the firebox still had 26 sq ft (2.4 sq m) of arch tubes. Also the boiler's distribution of tubes and flues was unchanged.

But the valves were actuated by Baker-Pilliod's radial gear, the grate area grew a bit due a slightly longer grate, the combustion chamber gained four inches (102 mm) and increased its contribution to heating surface area to 65 sq ft (6.05 sq m). It was delivered with a Duplex automatic stoker.

A "Hereafter" note from 24 April 1924 stating that that "No welding whatever will be permitted on forgings" suggests some in-service incidents that ended badly.

Well-suited to the Milwaukee's demands, as were all of the L2s, all but a few L2-as served into the 1950s. The 8280 was withdrawn in July 1935 and 8289 was retired in March 1944. Of the rest, several were retired in late 1949, the bulk in 1950-1953, and the last few in December 1954.

---

Class L2-b (Locobase 15197)

Data from DeGolyer Library, Vol 67, pp. 211. Works numbers were:

1922

May 55383-55387

July 55541-55542

September 55605-55616, 55663-55668

1923

May 56518-56523, 56577-56586

June 56686-56693

July 56759-56765, 56831-56836

August 56907-56911, 56984-56991

September 57062-57066, 57082-57083, 57171-57181

October 57249-57252, 57265-57365

Locobase had detected a relative insufficiency in superheater heating surface area compared to many other Mikados of the period and apparently so did the Milwaukee some 90 years earlier. The last L2s they procured came from Baldwin (as had the L2a shown in Locobase 15196), but these had boiler changes aimed at remedying that shortcoming.

Like all of the Milwaukee's L2s, this class of -bs satisfied Milwaukee Road requirements for decades and all operated into the 1950s. The first to retire was withdrawn in January 1950, the last nineteen were withdrawn as a group six years later in January 1956.

---

Class L3, L3-a, b (USRA) (Locobase 16522)

Data from MILWRD 1945 Locomotive Diagrams supplied in May 2005 by Allen Stanley from his extensive Rail Data Exchange collection. See also Jim Scribbins, "Chicago, Milwaukee, St Paul & Pacific (Milwaukee Road)" in George H Drury, Guide to North American Steam Locomotives, Revised Edition (Waukesha, WI: Kalmbach Books, 2015), p. 134.

(Thanks to John Werth for his 24 July 2021 email noting the Wilson feed water heater.) for the Wilson system: "Milwaukee's S2's: Any Milwaukee Steam Experts Out Ther[e?]" thread begun by rcall30160 [Bob Callahan] on 4 June 2010 and Larry Doyle's response on 5 June 2010 on the trainorders forum at [],2203733 and Larry Doyle, "Wilson Feedwater Heater", posted 14 June 2019 at [],4810774, last accessed 24 July 2021.

Alco's Brooks and Schenectady Works divided evenly the production of these USRA Heavy Mikados (Locobase 41). Brooks's works numbers were 59740-59776 in September 1918, 59777-59789 in October. Schenectady delivered its 50 as works numbers 61042-61046, 61148-61167 in March 1919, 61168 in May, 61175 in June; 61170-61172 , 61174, 61176, 61179 in July; and 61169, 61173, 61177-61178, and 61183-61192 in August.

The 100 Heavy Mikados, most assigned to a single railroad, proved well-suited to the Milwaukee's motive power needs. Jim Scribbins, writing in Drury 1993, 2015, commented that they were "the most powerful road engines until the 4-8-4s arrived." He add that they proved "particularly successful on the main line between Minneapolis and Harlowton [Mont.], on the routes to the Indiana coal fields, and in iron-ore service in upper Michigan."

Firebox heating surface area included 29 sq ft (2.69 sq m) in arch tubes.

Fourteen locomotives fitted with Franklin trailing-truck boosters fell into subclass L-3a: 8605, 8608, 8610-8611, 8614, 8616, 8620-8621, 8631, 8639, 8645, 8650, 8653, 8656-8657, 8660, 8674, 8679, 8691. (300-314 and 354, 359-360, 363 in the 1938 renumbering).

Nine of these had been fitted with Wilson feed water heaters by the 1930s: 8608, 8614, 8616, 8620, 8639, 8645, 8691. Two more boosterless locomotives--8671, 8685--also had feed water heaters. Wilson's 1929 patent filing (US1901216) describes the original design. Larry Doyle's 2010 response in trainorders (cited above) reported the installation lay in "in a hot well in the tender, with a large [6" (152 mm) diameter] exhaust steam pipe running on the left side of the engine from the cylinders to the tender". Baldwin Magazine's report described the Wilson Locomotive Water Conditioner in similar terms.

Doyle returned to the Wilson feed water heater in the trainorders forum in 2019 with a full quote of an article by former Milwaukee Road engineer Bill Wilkerson (It too is cited above in the second trainorders entry.). Wilkerson's detailed discussion of the weaknesses in the design as it interacted with operating conditions is too long to quote again in full in this entry. But a summary of his comments should include the following:

A thermostat valve needed to work properly all the time and scale laid on the valve by hard water in Montana and the Dakotas usually resulted in heating all of the tender's water to boiling. Hand valves that could control the steam "were all located in the most inaccessible place on the engine." To operate that valve on L3s, the fireman had to crawl out on the running board on the left side, "then lay down and reach down to get the valve."

That wasn't the end of the mischief. Boiling water coming forward to be let into the boiler used an impeller pump. If the pump became "steam-bound", it would let the steam turbine "over-speed and lock-out."

Trainorders comments speculated on whether firemen just pretended the system was broken and bypassed it somehow. Doyle added later: "I would suspect, from Wilkersons description, that these things were probably very noisy in operation, as well.

"I'd think it would be about like closing the overflow valve on a lifting injector and fully turning on the steam to it. To those who haven't experienced that, it's sorta like boiling a large potato kettle on a kitchen stove, but several hundred times louder. The whole tender shakes."

( Locobase found a later patent (US1884051A) described a boiler-mounted "pre-heater". It consisted of a bundle of tubes within a closed unit. As feed water entered the tube bundle, it was heated by exhaust steam. Locobase can't be sure if this was meant to be an added component or a replacement for the tender system. Nor can he say if any of these preheaters were produced.)

The L3s' backbones were one-piece cast steel beds from General Steel Castings Corporation. Boxpok drivers turned on Timken roller bearings, as did the front truck's wheels; rear truck and tender wheels used American Steel Foundries roller bearings.

By 1945, boilers on all of the engines were pressed to 200 psi (13,8 bar) which increased tractive effort to 62,949 lb (280.0 kN; 28,553 kg) . Fitting thicker tires increased nominal driver diameter on some locomotives to 64", slightly reducing tractive effort to 61,965 lb (275,63 kN; .28,107 kg).

Except for 336 (ex-8629), the entire stud served untl the late 1940s into the mid-1950s.