• History of the E-unit

  • Discussion of Electro-Motive locomotive products and technology, past and present. Official web site can be found here: http://www.emdiesels.com/.
Discussion of Electro-Motive locomotive products and technology, past and present. Official web site can be found here: http://www.emdiesels.com/.

Moderator: GOLDEN-ARM

  by Allen Hazen
No argument here! (But maybe GE's North American locomotive offerings are a BIT more varied than you say: the new commuter locomotives (for Boston?) may not be assembled in a GE plant, but they haveGE working parts: between GEVO engine and the electricals, probably over two-thirds of the locomotive by value.)
  by Engineer Spike
I have heard somewhere that the twin 12-567 was an economy move. They could better justify the production of the 12 since it was also used on the SW series. Who knows it this theory holds water.
  by Desertdweller
I think the twin 12-cylinder formula was used because these 12-567 engines were being used in production in switchers, and it was easy to obtain 2,000hp by installing two engine/generator sets.

Notice that ALCO followed the same formula (DL109). Two 1,000hp switcher engines equaled one 2,000hp unit.

I think there was room in an E-unit to shoehorn in two 16-567s, but there was a bigger problem in feeding 1500hp into two traction motors. Remember the 3,000hp EMD locomotives came with the 40-series. Along with new traction motors and generators, and 645 engines.

What were EMD's options? E unit production was almost over with before turbos were applied to the 567. Theoretically, turbos could have been applied to the 567s. I don't know what a turbo 12-567 would produce. Maybe 1400-1500hp? EMD didn't have a turbo until the SD24/GP20.

If a 3,000hp E unit were made (E10?), maybe C-C trucks would be able to absorb the power. I don't understand why EMD waited so long to go turbo, and only did so after UP did it themselves with Air Research turbos on GP9s.

  by theastralcity
It also should be noted that in order to get the 3,000 hp, the British Rail Class 55 "Deltic" used a truly novel design for the prime mover:
http://en.wikipedia.org/wiki/Napier_Deltic" onclick="window.open(this.href);return false;. English Electric, I have read, thought about trying to market the Deltic to the Canadians, but nothing came of it. That locomotive has a reputation in the UK and is very celebrated, much the same way the we have nostalgic love of say the UP Turbines here. The Napier Deltic prime mover was troublesome to say the least, the paired down version in the Class 23 was even more so, and lead to that type having a very short service life. The Class 55 served well, and were only replaced by high speed trains in the 1970s, but they were top-flight brass, and got the utmost of care to help them along.

British railroading is very different than American in their view of passenger transportation. Even today, with Amtrak posting record numbers, a "busy" American station in a mid-sized city can at best hope for only about 10% of the traffic in a comparative UK one. During the period of nationalization under British Rail, labor costs were often overlooked while in the US at the same time were always seen as the bane of the railroads. As such, the Brits fielded many more high-labor locomotives than we did, and were quicker to adopt the new and novel in everyday service.

Add to this the economics of EMD at the time of the E8's creation. Say you were the staunchly-conservative Pennsylvania, and needed 6,000hp for a fast streamliner, It'd be just as easy to put an A-B-A of E8s on the front as it would be to put an A-A configuration of 3,000hp engines. In fact from the railroad's perspective that might be better. If a unit died in route it still would probably be possible to limp in with 4,000hp instead of only 3,000. Now if you're EMD, that need for 6,000hp is happier for you as well, because it means selling an extra unit which means more income too. Proven power, rugged, and easy to service and cheap gets the job done in American railroading.
  by Engineer Spike
I think railroads try to use what is least demanding to maintain. A machinist once told me that a bright high school auto shop student could repair a 567. The other designs took more time and skill to keep in shape. I have a copy of New Haven Power, and it said that the FM C Liners fell by the wayside because they took so much more time to repair than the Alco PA, or DL109.

I have seen BR class 55, and I know that they would be parked in the weeds within 5 years, if in US or Canadian service. The simplicity of the E, along with its redundancy of 2 separate power units made it a winner. This thinking is likely why EMD didn't go to turbochargers until it couldn't get by. 3000hp from a non turbo unit would have been impossible. Can you imagine a 20-645 non turbo in a GP40?
  by Allen Hazen
"This thinking is likely why EMD didn't go to turbochargers until it couldn't get by. 3000hp from a non turbo unit would have been impossible. Can you imagine a 20-645 non turbo in a GP40?"
--- I'm trying to imagine it. It would have been heavy: there's a reason why EMD didn't try to match GE's U36B with a 20-645 engined "GP-45". And, at the per-cylinder ratings of 645 engines in GP/SD-38 and switchers, it would only have had 2500 hp.

I think your general point is right, Engineer Spike. The BR Class 55 ("Deltic") is maybe a special case: a specialty locomotive of very high power/weight ratio for use on the tastes express passenger runs, for which management was willing to accept the need for special maintenance attention (as in: nursed every night by factory representatives). BR only bought about two dozen of them, as opposed to several hundred slightly heavier and slightly less powerful Class 47 and Class 50 with medium speed diesel engines. … Still, even the best British diesels seem to have needed more maintenance than a U.S. railroad would have tolerated: when a Wisconsin Central affiliate took over much of the freight service on British railways, they scrapped large numbers of "indigenous" locomotives, replacing them with Class 66 from … EMD!


Thinking about it some more… What the Deltic/Class 55 is maybe comparable to is the Turbotrain that the North East Corridor project tried to get Metroliner performance on non-electrified lines with. Compared with THAT, the Class 55 was a marvellously dependable and maintainable design!
  by Engineer Spike
My comment about the 20-645 was just to show just how if it could have been done, it would have, due to the increased simplicity of it. I know this is impossible due to weight. It would have been a fuel hog too. I'm just illustrating how the EMD thought was to use the most simple design possible.

One other point is that EMD had the mindset of an automobile manufacturer. Simple, cheap, and get the most return out of every part. The 12-567 was needed for switchers too. If the same line of engines could be used for several products, it saves cost.
  by Desertdweller
Maybe the reason EMD waited so long to use turbos was on account of concerns about reliability and maintenance. This is the reason GP38's are so popular with small railroads. A non-turbo 645 seems to be cheaper to maintain than a turbo 567. Also, in sustained low-speed operation, non-turbo units tend to last longer.

  by SSW9389
EMD pitched a C-C truck twin engine booster unit with turbocharged 12-567Ds to SP in 1961. It was to be called the RB-3600 and was EMD's response to the Krauss Maffei diesel hydraulic. The horsepower rating was 1800 per engine. See the Southern Pacific 1973 Motive Power Annual by Joe Strapac for details.

This is just to show that EMD could have placed turbocharged 12-567Ds in an E unit, but obviously it would have to have more than four traction motors to soak up the horsepower.
  by Allen Hazen
1800 hp means the rating (power per cylinder) is the same as that on the 16-567 on an SD-24. So at least the engines for a "Turbo E" would have been feasible in 1959.

Turbocharged 12-645 engines have been used a fair bit (GP-39, SD-39, and lots of export models with model numbers starting GT22 or JT22), but I don't know if the turbocharged 12-567 ever got into a locomotive. Does anyone else know offhand whether there was ever a "GT12"?

Thanks, SSW9389! I've never read the Strapac book, and didn't know about that proposal. I think there was a proposal in the 1980s -- not from EMD but from Morrison-Knudsen, who were doing lots of things with 645 engine blocks -- for a very large CC unit with twin 12-645T engines, for use in Australia (which makes me wonder what the axle loadings were supposed to be!).


As to the "Turbo E"… I don't like the idea of adding extra traction motors: it would need a new truck, and so not be as "E-ish" as I'd like! Maybe we could make do with four motors. E-units are for high-speed passenger service, and if we could arrange to have the locomotive used ONLY on trains it was suited for, maybe the 3600 hp could have been used with four motors. Minimum continuous speed of something like 50mph, but that might be tolerable if you could assume trains that would keep up a steady 80+ mph between (reasonably infrequent) stops. One needs to do some fairly heavy-duty economic and political alternate history to get it in the 1960 era, but… Suppose the CB&Q, CNW and C,M,StP&P had come to an agreement (involving various government agencies, state and federal -- as I say, heavy duty political fiction) to set up a demonstration project for air-competitive high-speed rail between Chicago and the Twin Cities: moving all freight off the Milwaukee line so it could be turned into a dedicated high-speed passenger line. Trains running long distances at over 100mph (E units were marketed with traction motor gearing allowing speeds up to 117 mph), with a very small number of intermediate stops.
… Special, twin-engine, diesel for fast trains on one route: this reminds me of something… Chicago to the Twin Cities is somewhat further than London to Edinburgh, but comparable, so… The "Turbo-E" would have been more powerful than a British Rail "Deltic" (production units introduced in 1962, but a prototype was demonstrated in the mid-1950s), but also much heavier: 20% more power, 50% more weight.
(Sorry, I seem to keep mentioning Deltics in this string!)


O.k., what if I give in and allow six motors? Well, it's not all THAT implausible: EMD did, after all, produce a 3600 hp CC passenger unit in the late 1960s: the FP45. It had trucks designed originally for freight (SD) units, but I think the Santa Fe, at least, was able to run FP45-powered trains at 90 mph. And a truss-framed carbody like an E-unit's would have allowed some weight-saving: surely desirable in express passenger service!
  by Pneudyne
EMD seemed to have been particularly conservative when it came to the power settings for the post-WWII E-units.

The E7 was rated at 2000 hp not the full 2250 hp that was possible. The E8 went to 2250 hp, but that was after the 94 hp/cylinder rating had been service-proven in the F3.

The E9 was rated at 2400 hp when the full rating for the 12-567C pair would have been 2620 hp. This probably reflected the same conservative attitude that had obtained with the E7. Also, EMD may have been reluctant to go above 2400 hp in a 4-motor locomotive with an adhesive weight of around 220 000 lb. As far as I know the E9 was never uprated to the full capacity of the 12-567C engines, nor was the 12-567D1 used. This would have provided 2650 hp basis the 1325 hp rating that applied to the RS-1325. So it looks as if 2400 hp was considered to be the upper limit for the locomotive, even where proven higher engine power was available. Traction motor capacity might have been another limiting factor in the 1950s, but perhaps less so by the early 1960s.

EMD also promoted the twin-engined nature of its E-units as a safety factor against power failure.

Whilst the 2400 hp 16-567D3, as used in the SD24, was on the face of it an attractive option for a potential successor to the E9, it certainly would have been something of a major reversal for EMD. That is, a single powerplant at a higher, as yet untried rating, with the additional risk factor of turbocharging, would be replacing two established and derated powerplants. Not only that, but the major re-engineering of the E that would have been required was probably seen as not justified given that by the late 1950s, its sales were surely in decline. Rather letting the E9 simply run out in its existing form was the logical choice.

Getting a pair of 16-cylinder engines into the E9 carbody would likely have been a “shoehorn” exercise. Possibly the combination of a12-cylinder engine forward and a 16-cylinder engine rearward would have been possible. If we assume that a C-C wheel arrangement would have been necessary or at least highly desirable above 2400 hp, then there is the issue of truck design to consider. Perhaps the EMD flexicoil C truck would have been satisfactory at reasonably high speeds, but likely something better would have been required for the very highest speeds. Thus EMD may have had to develop a C-version of its Blomberg A1A swing-bolster truck. (A C version of the Blomberg truck is not so far-fetched when one considers that the A1A design was the alleged inspiration for a widely-used series of British C trucks.). I doubt that EMD would have been too keen on buying in a swing-bolster C truck from Commonwealth.

Two engines and a C-C wheel arrangement also pose the question as how to best arrange traction motor connections. Keeping the two power systems separate throughout, and so having each with fixed 3P connections would seem logical, but whether EMD would have been happy with that in the late 1950s is open to question. That said, the G12 export model, in 1425/1310 hp form, had a fixed 4P arrangement. Two different engines, that is a “12” and a “16”, would complicate matters even further.

Returning to the single 2400 hp 16-567D3 powerplant passenger locomotive theme, then possible the FL9 would have been a feasible starting point. Devoid of its third rail DC equipment, it likely would have accommodated the turbocharged engine and its ancillaries without being overweight. One may also imagine a longer, “FXL” variant, extended forwards so that it could run on a pair of Blomberg A1A trucks.

I imagine that EMD delayed the introduction of turbocharging for as long as it could. As long as it did not need turbocharging, it could claim a simplicity advantage for its two-stroke engine over the four-stroke competitors, which were necessarily turbocharged. Not only did that disappear when it introduced turbocharging, but two-stroke engines are more difficult to turbocharge than the four-stroke type, in part because the exhaust gases are much more diluted with scavenge air. Once one reached the point where the desired specific output mandated turbocharging regardless of whether the two-stroke or four-stroke form was used, then a “clean sheet” design could well favour the four-stroke. This became more so as material and design improvements allowed four-stroke engines to approach the combustion limited piston speeds. Evidence is found in the ocean-going marine (as distinct from the North American inland marine case) medium-speed trunk piston engines. Historically both two- and four--stroke examples were found. But from the 1970s the four-stroke type dominated, and the two-stroke type dies out. For example Sulzer designed its then-new Z40 engine to be either a uniflow two-stroke of a four-stroke, and convertible between the two configurations. But very few of the two-stroke version were actually built. (Of course the very biggest crosshead-type marine engines are necessarily two-strokes; with stroke-to-bore ratios often above three, it is difficult to imagine that anything other uniflow scavenging would work.)

EMD F7 & E8 p.04,05.jpg
  by Allen Hazen
Another post with lots to think about!
One thought that comes to mind, about the "shoehorning" of two 16 cylinder engines into an E-like body (to get a 3000 hp locomotive instead of 2000, or -- in the E-9's era -- a 3500 hp unit). Six-axle passenger units in the 2000 hp range were offered by all four (EMD, Alco, Baldwin and FM) of the builders doing road locomotives in the U.S. in the 1940s and 1950s. FM and Alco (post WW II) managed single-engined units, which were significantly shorter and, at least in Alco's case, lighter than an E. Alco, before the war, and Baldwin needed two engines, and had engines that were heavier for a given output than EMD's 567. As a result, Alco's Dl-109 (two turbocharged 539 engines, similar to those used in the S-2 switcher) was longer and heaver than an E. Baldwin's engine was even heavier: their first 2000 hp A1A trucked passenger units had two orally aspirated 8-cylinder engines. (The turbocharged version of that 8-cylinder engine is what went into their 1500 and 1600 hp units, so it was Baldwin's competitive equivalent of EMD's 16-567.) As a result the Baldwin locomotive was MUCH longer and heavier! Had EMD wanted to build a 3000 hp locomotive in the late 1940s (had they, in other words, thought there was a market for such a unit), they could certainly have put two 16-567 engines into something no longer and no heavier than what Baldwin was trying to sell as an "E equivalent."

Will Davis and his brother posted (on their "Railroad Locomotives" blog) excerpts from some Baldwin promotional material from right after the war. Baldwin suggested as a possibility something looking much like their 2000 hp passenger locomotive, but rated at 3000 hp, using turbocharged instead of normally aspirated engines.
  by Allen Hazen
Davis blog post with the Baldwin proposal for a 3000 hp variant of their 2000 hp passenger unit.
http://railroadlocomotives.blogspot.ca/ ... talog.html" onclick="window.open(this.href);return false;
(Note that, the EMD with its comparatively wimpy traction motors might have gone for six if they had tried to build a 3000 hp unit in the late 1940s, Baldwin seems to have thought that four Westinghouse 370 motors would suffice.)
  by mtuandrew
I don't see that EMD offered a turbocharged 8-567, but was there ever any interest by railroads in acquiring or rebuilding E-units to 2 x turbo 8 configuration for lighter weight? (or for that matter, keeping the A1A trucks and moving to 1 turbo 16-567)
  by Pneudyne
Allen Hazen wrote:Davis blog post with the Baldwin proposal for a 3000 hp variant of their 2000 hp passenger unit.
http://railroadlocomotives.blogspot.ca/ ... talog.html" onclick="window.open(this.href);return false;
(Note that, the EMD with its comparatively wimpy traction motors might have gone for six if they had tried to build a 3000 hp unit in the late 1940s, Baldwin seems to have thought that four Westinghouse 370 motors would suffice.)
No doubt that EMD could have built such a 3000 hp unit in the late1940s had it chosen so to do, and one would reasonably expect that if it had been done, it would have been shorter and probably a bit lighter than Baldwin’s offering, and that it likely would have been of C-C wheel arrangement.

I’d guess that it would have weighed somewhere around 380 000 lb, giving an axle-loading of about 63 000 lb. That I think may have made it too heavy for ultra-high speed passenger work of the period (say much time spent above 90 mile/h), where something like a 55 000 lb axle loading might have been preferred.

So rather than an ultra-high speed passenger locomotive, the 3000 hp unit would be more in the heavy passenger class. In power terms, a single unit would be equivalent to an FP7A plus F7B combination, and by virtue of its lower weight as compared with that combination, might have had the edge when it came to acceleration and balancing speed on the flat. But the greater adhesive weight of the FP7A plus F7B combination (say 500 000 lb) would surely be preferable in the mountains, and that is why an FP7 not an E8 would be used there anyway.

Absent significant changes in thinking about axle loadings for ultra-high speed passenger work, speeds, that gets us back to the notion of using an E8 fleet east of the mountains, and an FP7 fleet in the mountains, with no obvious non-trivial niche for the 3000hp unit. One imagines that EMD’s initial tactic had the Baldwin 3000 hp proposal excited the imagination of one or more of its customers would have been to argue the merits of doing the job with an E8 and FP7 mix.

I suppose in terms of conceptualizing a twin 16-567 engined unit, one could extrapolate backwards from the DD35, paralleling the actual pathway that takes one back from the GP35 to the GP7 and F7. Doing that, though, makes my 380 000 lb weight estimate look rather optimistic, even allowing for the lighter C-C running gear as compared with D-D.

EMD F7 & E8 p.02,03.jpg