• If you could restore a steam locomotive or 10

  • Discussion of steam locomotives from all manufacturers and railroads
Discussion of steam locomotives from all manufacturers and railroads

Moderators: Typewriters, slide rules

  by Tom6921
 
Basically, if you could restore up to 10 presently preserved steam locomotives in North America to operation (not counting the ones that are in the process of being restored) which ones would you restore?

Some contenders for me are:

SP 4460 (Run it with 4449)
N&W 1218
N&W 2156 (make all the big 3 operational)
CN 6400
C&O 2-6-6-6 #1604
Maybe a second Big Boy
The C&O Yellowbelly at the B&O museum.
  by Allen Hazen
 
Sentimentally, I'd go for PRR 4483, the I1sa that used to be on the front lawn of the WABCO plant in suburban Pittsburgh and is now at a western NY state railway museum. This, in multiples, was what the steam locomotive era was about!
But practically… Once we have a locomotive in working order, we'd like to see it work! So, we ought to choose a locomotive that could pull passenger excursions, and maybe the OCCASIONAL photo freight. So: we ought to be thinking of small to medium size passenger or mixed-service types.
So, having lived in Pittsburgh when I became a railfan and so being to some degree a Pennsy fan… How about the E6 Atlantic that's at Strasbourg? (I think there's already a project to restore a G5 Ten-wheeler, so if you also restore an H-8/9/10 Consolidation, you'd have a wonderful educational opportunity to show how the same steam power plant could do different railroad work: those classes all had the same boiler. But in the interests of having something we could see operate regularly, do the passenger power first!)
  by Allen Hazen
 
John Perkowski--
I see we both have soft spots in our hearts for ten-coupled freight engines!
(Since the PRR leased some ATSF 2-10-4 for a while toward the end of steam operations, some of Madam Queen's successors probably passed some of 4483's sisters in yards near Cleveland.)
--
Not a matter of restoration, but Britain (successfully! with Tornado) and the U.S. (with the T-1 folks who want to build a new duplex) both have movements to build new steam locomotives of "extinct" classes. Do you suppose there are rail fans in France to do anything similar? Since it is, I think, widely thought that the failure to preserve 242A1 -- possibly the best steam locomotive ever built! -- was a tragedy.
  by Pneudyne
 
I suspect that with a restored 10-coupled locomotive, there might be some difficulty in finding lines over which it could be allowed to operate, given that as a group, this type was generally viewed as being hard on the track and roadbed, and on underline bridges.

A lot of that expressed concern was in respect of vertical track loading resulting from hammerblow and piston thrust. In this regard, a while back I found a most interesting article in the R&LHS Newsletter Volume 24, No. 3, Summer 2004, available on-line at: http://rlhs.org/Publications/Quarterly/PDF/nl24-3.pdf" onclick="window.open(this.href);return false;. This was “Counterbalancing 10-Coupled Power”, by Don Leach. From this one could infer that for any restored 10-coupled locomotive, it would be prudent to use the light-weight rods, disc drivers and the counterbalancing techniques that prevailed at the end of the steam era, even if the locomotive concerned was not actually built or rebuilt with these features.

The article also answered a couple of long-standing questions, which were: (1) why did the UP’s allowed axle loadings in the steam era lag behind those of many other of the larger Class I roads; and (2) why was the UP 2-10-2 about half-a-size smaller than say the SP 2-10-2, which was proximate in size to the USRA Heavy 2-10-2. (Both the UP and SP examples look as if they could have been Harriman-heritage designs.) To the first question, UP evidently restricted its axle loadings because of its desire for relatively high operating speeds; for a given axle loading, track damage increases with speed. Thus the limits were, I think, successively 59 000 lb for the 2-10-2, 4-10-2 and 4-12-2 designs, 65 000 lb for the first Challengers and 67 500 lb for the FEFs, Big Boys and “big” Challengers. The UP 2-10-2 was designed both for the 59 000 lb limit, and for higher operating speeds than was customary for this wheel arrangement. The 30-inch piston stroke, rather than the common 32-inch number, was one factor that helped along this vector.

The larger 2-10-4s tended to have very high axle loadings, perhaps towards 10% higher than the roads that ran them allowed for their heaviest 8-coupled locomotives. Insofar as these roads were using 2-10-4s for tasks for which elsewhere the 4-6-6-4 might have been chosen, one may see the need for the highest possible adhesion weight. The above relativity shows up when one compares say the C&O T-1 with the Van Sweringen 2-8-4s to which it was related. (For example, the NKP 2-8-4 was said to be 70% of a T-1, not the 80% that the relative driver count would indicate.) Still, I have wondered whether the “excess” axle loading of some 2-10-4s was the result of an effort to reduce the dynamic augment-to axle loading ratio. If the former was essentially fixed as the result of the dimensions required to transmit the desired power, then axle loading was the only available variable which could be adjusted accordingly, that is upwards.

Less seems to have been written about lateral railhead forces, but it is difficult to imagine that most 10-coupled locomotives would have been anything but unedifying in this regard. Clearly the long rigid wheelbase was a disadvantage. But since virtually all had two-wheel engine trucks, then probably quite proportion of the nosing and curving forces were taken by the leading and second sets of drivers. In hindsight, it is perhaps surprising that in later years, as higher operating speeds were required, that there was not some move to four-wheel engine trucks, and possibly also the Alco-Blunt “lever principle” in which in a set of n driving axles, the first (n-1) all had lateral motion devices with appropriately graduated compliances.

Of course, the SP and UP had the 4-10-2 wheel arrangement for their respective three-cylinder 10-coupled designs (which bore about the same relationship to each other as did their 2-10-2s), but that I think was necessitated primarily by the need to carry the extra weight of their three-cylinder front ends. Still, that experience seems to have convinced UP that four-wheel engine trucks were highly desirable for freight locomotives. (Although one could counter-argue – maybe not all that strongly - along the happenstance vector that as on the 4-10-2, it was a necessity for the 4-12-2, from which it was carried over to the first Challengers which were in some respects a “bent” version of the 4-12-2 – with the same grate area - but with a four-wheel trailing truck to carry the extra weight.) Anyway, as the 4-10-2 was a reality, it is not so difficult to imagine a 4-10-4.

As to the “lever principle”, I don’t think that it was all that widely employed, at least in full. UP, on the FEF-2 & 3 (not sure about the FEF-1), Big Boys and big Challengers, and D&H on the K-62 are the examples that readily come to mind, but not any 10-coupled designs. Perhaps the roads that operated heavy 10-coupled locomotives had in early days adjusted to probably higher intensity of track maintenance required where they operated, and simply accepted it as a “fact of life”, or as their chosen trade-off against the probably higher locomotive maintenance intensity required with articulateds.

In view of the tracking and curving difficulties, perhaps surprising is that 10-coupled locomotives were used on some Cape and metre gauge roads. Of note is that metre gauge 2-10-4s and 4-10-2s (three-cylinder) were operated in Brasil.

Returning to the main theme, would a restored AT&SF 2-10-4 be allowed to operate over the route it was originally designed for, Belen to Clovis, NM?

Cheers,
  by Allen Hazen
 
Pneudyne-- Thanks for that (typically, for you) well-written and informative post!
Remarks:
---The thought of a PRR I-1 with lightweight rods (PRR actually experimented with Aluminum side rods on one!) and disk drivers is… unnerving. Stretches the meaning of "restoration," I'd say.
---As for the problem of lateral forces from the long rigid wheelbase… At least the early I-1 were originally built with blind drivers on THREE axles: flanges only on the first and last drivers. (And that's with 62 inch drivers: for a large-drivered 2-10-4 without some sophisticated lateral-motion arrangement the potential problem would be even worse.)
---As regards the "lever" arrangement of lateral-motion devices on all but the last coupled axle: I think the New York Central Niagara was another design with this feature. (Not surprisingly, since probably the same Alco designers who worked on the UP and D&H Northerns also worked on the New York Central's.)
-----
And, just as a mental picture or a suggestion to modellers… Many large late U.S. steam locomotives had "Centipede" tenders: five axles mounted on the tender frame, with a leading two-axle truck. Sketches of the New York Central's C-1 (the proposed but unbuilt "Duplex" elaboration of the Niagara) show an eight-axle tender: like a "Centipede" with an additional single-axle trailing truck. It has always seemed to me that the most appropriate tender for a 4-10-4 would be a nine-axle "Millipede": five frame mounted and with leading AND trailing two-axle trucks.
  by Pneudyne
 
I recall reading somewhere that the “Centipede” tender was not the best during backup moves on tight curves, so that the addition of a two-wheel trailing truck, making it a “4-10-2” rather than a “4-10-0” would have been beneficial.

At least according to the 1940 October Railway Mechanical Engineer (RME) article reprinted in Train Shed Cyclopedia (TSC) #49, the “Centipede” design was worked out between the UP and GSC. The choice of a four-wheel lateral motion lead truck may well have been an expression of the established the UP preference. The controlled lateral motion allowed for the rigid frame axles and the use coil springs at the dead ends of suspension system was also aligned with then-current UP (and Alco) practice. At lower speeds freight train speeds than the UP wanted though, a two-wheel leading truck could have been satisfactory.

Re the NYC Niagara, the description in RME 1945 October as reprinted in TSC #56 suggests that it had two lateral motion devices, although the wording could be seen as being somewhat ambiguous, not totally outruling that there could have been three. Somewhat surprisingly, the diagram shows that it did not have coil springs at the dead ends. The rationale for these was that laminated springs had a rate that increased with vibration frequency, whereas coil springs did not, and so their inclusion (in series with the laminated springs) lessened the high-frequency “hardening”.

That RME article attributes the Niagara design to the desire of developing the L-4 4-8-2 to provide higher boiler capacity, rather than say “slideruling” the J-3a 4-6-4 out to a 4-8-4. But as you say, no doubt the Alco designers brought to the table much of what they had put into the UP FEF-2/3 and D&H K-62 designs. I guess one could see the Niagara as being an incrementally developed FEF-2/3, or a K-62 expanded out to about FEF-2/3 size. Interestingly, Farrell and Pearsall (in their book covering all of the North American 4-8-4s) saw it as a development of the Rock Island 4-8-4 family, and in particular of the D&H variant.

Cheers,
  by Allen Hazen
 
The "Fallen Flags" site has the New York Central's 1946 locomotive diagram book, and I thought that might give more details on the Niagara's lateral motion devices:
http://www.rr-fallenflags.org/nyc/nyc-46lb-abc.pdf" onclick="window.open(this.href);return false;

Alas, the relevant data page only says that the Niagara HAD lateral motion devices, without more detail on the the number of axles so equipped. (Sniff.) I'll see whether I can find anything more.

As for relations to earlier Alco 4-8-4 designs… The Niagara was somewhat bigger than the U.P. FEF-2/3: 100 inch boiler diameter versus 98 inch i.i.r.c.

(Thanks for your comments!)
  by Pneudyne
 
Re the UP FEF, I looked through the Kratville book on this locomotive series. There was no mention of the number of lateral motion devices on the first series, although that the FEF-2 had three such was covered, along with a good photograph of the driving wheel set showing those devices.

By deduction though, I’d say that the FEF-1 had two lateral motion devices, on the first and third driving axles. The book had two large and very detailed foldout diagrams, one for the FEF-1 and the other for the FEF-3. In the latter case, there are shown dotted small circles just fore and aft of each of the 1st, 2nd and 3rd driving axles, but not for the 4th. These I’d say represent the circles of the horizontally oriented lateral motion device coil springs. The FEF-1 diagram shows the same as being on the 1st and 3rd driving axles.

This allows us to interpret the RME article wording on the Niagara with greater precision. It stated that lateral motion devices were applied to the front and intermediate drivers. I now read this as meaning the first and third driving axles, as was done for the FEF-1. Also – and something I should have recalled before – for 8-coupled locomotives, the term “intermediate” was used for whichever of the 2nd or 3rd set was not the main or driving set.

Count and disposition of lateral motion devices appears to have varied according to railroad preference. For example, as said, the D&H K-62 of 1943 had three, on the 1st, 2nd and 3rd driver sets. But the similar Rock Island R-67 of 1944 had just one, on the leading driver set. The Lehigh Valley T-2b of 1943 (or 1944?), which was essentially to the original Rock Island R-67B design, had two, on the 1st and 4th driver sets.

Kratville included a preliminary sketch for the UP 4-8-4. It showed a 98 inch BMOD boiler, 27 x 30 inch cylinders, 77 inch drivers and 67 500 lb axle loading. The 77-inch driver size may have been a legacy from the Harriman era, having been used for the heavy Pacifics. Boiler pressure (BP) and grate area were not indicated. But the BP may not have been above 270 lbf/in². That would have given a factor of adhesion (FA) of 4.14. The FEF-1 had an FA of 4.24, the FEF-2 was 4.23. I suspect that the UP was conservative in this regard in deference to the sometimes-adverse ambient conditions it experienced on its trans-Wyoming main line. I guess that the grate area would have been somewhere in the 90s ft². One could say that the UP preliminary sketch design was proximate to the Lehigh Valley T-3 class. This had the Rock Island boiler with firebox extended from 88 to 96 ft² grate area and pressed to 275 lbf/in². Although the firebox extension was done by Baldwin, I doubt that Alco would have had any problem with using it. In fact, later it more-or-less did for the D&H K-62 and final edition of the Rock Island R-67.

But evidently the UP and/or Alco decided that more boiler capacity was needed, so that the definitive FEF-1 had 100 inch BMOD, 100 ft² grate area and 300 lbf/in² pressure, these dimensions being repeated for the FEF-2 and FEF-3. So, the FEF boiler as built was broadly comparable to that of the Niagara. The UP was evidently seeking a passenger locomotive that as well as being suitable for its mountain grades, was also capable of sustained high-speed running whilst being relatively kind to the track and roadbed in a lateral as well as a vertical sense. Thus, it probably needed to avoid the excesses of weight and cylinder size found on some other large 4-8-4s, as well as overly ambitious FA numbers. The NYC would not have needed mountain grade capability, but the other properties were generally desirable.

Cheers,
  by Allen Hazen
 
A locomotive with lateral motion devices on three axles would, I think, usually have had a different amount of "play" on each. Since the "Railroad Mechanical Engineer" article you cite (the one reprinted in the "Train Shed Cyclopedia") only gives two figures for amounts of lateral motion allowed, I think it is almost certain that the Niagara (or, to be ultracautious, the prototype, S1a, Niagara) had l.m.d. on only two driving axles-- I would think your suggestion that they were on the first and third drivers is likely. (Trying to find relevant information I noticed, in the T1 write-up in the same "Train Shed Cyclopedia" fascicle, that the Pennsylvania's rival for the Niagara also had l.m.d. on the first and third drivers.)
I could have sworn I had somewhere read that the Niagara had l.m.d. on the first three driving axles 9 (with different amounts of "play" on each). But I haven't been able to find my supposed source… and I have had very clear but erroneous memories before! For instance… I was also quite confident that the Union Pacific's 4-8-4 had boilers that were two inches smaller in maximum diameter than the Niagara's, but looking just now in the 1940 "Locomotive Cyclopedia" (Kalmbach facsimile reprint from the 1970s) I saw that the drawing of the FEF-2 was very clearly labelled as having 100" BMOD! (Which leaves a small puzzle: the S1a Niagara has significantly more evaporative heating surface and marginally more superheating surface than the FEF-2: how did they manage to fit this into a boiler with the same diameter? Is the FEF boiler more strongly coned than a Niagara's, so that the front end is smaller in diameter? This would limit the number of tubes and flues that could be fitted into it.)
  by Allen Hazen
 
O.k., one question answered. George Elwood's "Fallen Flags" rail image site has steam locomotive diagram books for BOTH the NYC and the UP. So we can compare at least a few features of the FEF and Niagara designs. Both had BMOD -- outside diameter at the widest point (third course, just in front of firebox) -- of 100 inches. But the Niagara's boil was closer to being cylindrical: INSIDE diameter at the first course was 90" on the Niagara but only 86" on the FEF. So the front tube sheet of the Union Pacific 4-8-4 was smaller than that on the New York Central's: so the cross section into which the tubes and flues had to fit was bigger on the Niagara.

In detail: the UP engine (well, FEF-2, which is the one I have data on: I think the FEF-3 was basically similar) had 50 tubes and 184 flues. The Niagara had 55 tubes, and 177 flues… but the FEF's flues were only 3 3/4" in diameter, but the Niagara's were 4". So, if they were the same length (they were close in the S1a: the S1b "production" Niagara's had a shorter combustion chamber and longer tubes/flues), the Niagara's 177 flues would be equivalent in evaporative surface area to about 199 FEF flues. Which explains the difference in heating surface area...
  by Backshophoss
 
Barry is one of many UK scrappers that have allowed some sales of retired steam locos,what is left might be beyond saving.
Most US scrap yards made fast work of cutting up steam locos,the metal was re-refined at steel mills in to cars,trucks and steel beams.
  by dan the ca railfan
 
Backshophoss wrote:Barry is one of many UK scrappers that have allowed some sales of retired steam locos,what is left might be beyond saving.
Most US scrap yards made fast work of cutting up steam locos,the metal was re-refined at steel mills in to cars,trucks and steel beams.
still how do they last that long retired?

because there is still one line in china that uses steam locomotives for freight and regular passenger service!