HydraulicLocoTrans2

??Regarding diesel-hydraulics, I don't think that there have been any comprehensive English-language treatments of the concept and its execution over the years, although it would make a fascinating study.

The British Rail diesel-hydraulic fleet has been well covered in the literature, but there one has to step carefully through a minefield of myths, legends and the extreme partisanship associated with anything to do with that short line that ran west of London, and through Swindon!

During the 1950's, there was quite some debate in the trade press about the relative merits of the Voith triple-converter system and the Maybach Mekydro converter-plus-four-speed-gearbox system. Maybach claimed greater efficiency over the speed range (marginally true) and at part loads (true,) but the complete, albeit very short, torque interruptions during gearshifts were a significant liability, and magnified driveline vibration problems. With hindsight, I think it's fair to judge that Voith had what was ultimately the better idea.

Interestingly, Maybach and Voith had co-operated over development of the Fottinger principle in the early days, but went their own separate ways in the late 1930's after Maybach espoused the change-speed concept, while Voith wanted nothing to do with it.

There was also debate over just what constituted a hydraulic transmission. The Voith triple-converter system was unequivocally hydraulic, but the Mekydro system was sometimes described as 'hydro-mechanical.' The ORE promulgated a definition in which a hydraulic transmission was any in which ther was always a hydraulic torque multiplying element in circuit between the engine and the road wheels, regardless of what other torque multiplying devices were in series with it. Under this definition, the Mekydro qualified as a hydraulic transmission.

Transmissions like the Twin Disc system, which employed a three-stage (as distinct from triple converter) at lower speeds, but with a direct drive clutch effectively shunting the converter at higher speeds, did not qualify. Rather, they were hydro-mechanical, being hydraulic at lower speeds and mechanical at higher speeds.

Essentially mechanical, change-speed transmissions that employed fluid couplings (non torque-multiplying) in place of friction clutches -- as used on most of the first generation British DMU (diesel multiple unit railcars for passenger service -- Will) fleet, and on many smaller British shunting locomotives, domestic and export -- were viewed as being strictly mechanical. That the fluid coupling was not a hydrokinetic device did not alter anything -- there was no hydrokinetic torque multiplication involved.

This all sounds quite reasonable at first glance, but it quickly becomes unstuck upon further examination. As well as triple-converter systems, Voith also made varieties that combined two torque converters with one fluid coupling, one converter with two fluid couplings, and so on. In these cases, the fluid couplings took over the drive at intermediate and or higher speeds, the converter(s) then being drained of fluid and inoperative. But haven't we already established that a fluid coupling does not constitute a hydraulic transmission, but rather a form of mechanical drive? In fluid coupling mode, such transmissions are not fundamentally different to the Twin Dsc in clutched mode. So did Voith in fact make hydro-mechanical as well as hydraulic transmissions.....(And, after saying that, is there a guided missile on its way from Heidenheim to Sydney?) {Ed note; Voith is located in Heidenheim, Germany; Steve Palmano is located in Sydney, Australia.}

Then, in the late 1950's, SCG in the UK introduced a variant of its familiar Wilson epicyclic gearbox mechanical transmission in which a (Schneider-type) torque converter replaced the usual fluid coupling. The trade magazine "Diesel Railway Traction" described this as a true hydraulic transmission, within the meaning of the act. But, the Schneider-type converter is equipped with a free-wheeling stator, so that at higher rotational speeds it becomes a fluid coupling, with no torque multiplication. (Most passenger car automatic transmissions have this kind of converter, these days with modulated lock-up clutches --- although that said, the Borg-Warner DG of the late 1950's also had a 3rd gear lock-up clutch, nothing being really new.) So, really, any transmission with a Schneider converter is more hydro-mechanical than hydraulic.

Thus, the definition of "hydraulic transmission" is really quite blurred.

Unlike their diesel-electric counterparts, diesel-hydraulic locomotives do not have load control systems. With good converter matching, a hydraulic transmission should be able to keep engine loading within a few percent of nominal full across both the engine speed range and vehicle speed range. Voith is on record as saying that governors with controlled droop are preferable to the isochronous type. But good matching isn't always achieved, and the British Rail Class 52 ("Western") suffered from this problem, although it probably wasn't the worst of its design faults.

Thus, except for part-converter filling at very low speeds, control of diesel-hydraulic locomotives is essentially by engine speed alone, although fuel rack position may be used as an alternative, particularly for those whose transmissions include fluid couplings or lock-up clutches, and therefore whose engines must operate along the torque curve and not at preset speed. Torque converter loading of the engine follows the cubic law "propeller curve," so that where stepped control of engine speed is employed, the speed notches start off widely-spaced, but get progressively closer as top engine speed is approached.

Reverting to the British Rail diesel-hydraulic fleet, with long-distance hindsight, what was in fact regarded as about the worst of the fleet, the NBL-built Class 41 (D600 series) was potentially probably the best design.

The D600 is generally regarded as not a very good locomotive. Still, it did have some attributes, but suffered the misfortunes of (a) being "designed" and built by NBL; and (b) being equipped with what was clearly the worst of the trio of suitable German high speed engines of the period.

NBL had very limited design capabilities when it came to diesel and electric locomotives, both conceptually and in the detail aspects. It also suffered from poor standards of workmanship. Thus the D600 was hardly designed by NBL; rather, it was essentially a diesel-hydraulic adaptation of the LMS 10000/1 diesel-electric prototypes. It was supposed to be a C-C, but NBL couldn't figure out the driveline details so took the easy way out and made it an A1A-A1A, thus significantly lessening its utility.

It inherited from LMS 10000/1 a constructional form that guaranteed excessive weight. Normally with cab units, the superstructure framing is load bearing, thus allowing the use of a lighter underframe. This was not the case with 10000/1, where (I think) there was a requirement that the engine be removeable sideways due to limited overhead clearance in existing maintenance shops. So both the prototype and the derivative D600 designs had full strength underframes.

I think it fair to say that a competent builder, like Vulcan Foundry, using standard techniques of the 1950's, could have produced the D600, in A1A-A1A form, at a bit under 100 tons. In C-C form, it likely would have been no heavier than the later Class 52 (D1000 series "Western.") Bear in mind that Vulcan Foundry had already produced the prototype English Electric Deltic locomotive, which, using conventional construction techniques, weighed in at 106 tons for 3300 HP.

In its favour, the D600 design did have a very clean and uncluttered machinery layout. All major components were reasonably accessible, and not "crammed in" to small spaces, or even under the driving cabs, ans they were on the BR Class 42/43 (D800 series) and Class 52. Also, it inherited a very good basic bogie design from LMS 10000/1 that provided smooth riding at the highest speeds right from the start. (The LMS Ivatt-Fox bogie is said to have been inspired by the EMD A1A bogie used under the first E-model passenger locomotives.) The Class42/43 design, on the other hand, had a rather complex bogie design that was initially very troublesome at higher speeds, and evidently conferred a rather higher specific rolling resistance. The D600 had a much simpler, just as effective BTH control system than the overly complex BBC system used on the D800. Also, it had the Voith transmission, which in practice was better than the Maybach Mekydro.

Into the realm of speculation, had BR entrusted the D600 design and development to a competent builder who could have produced, say, a 105-ton C-C, and had it been fitted with a better choice of engine, there would have been no need for the D800 and D1000 designs. Engine power output growth would have allowed the same basic D600 to do the D1000's job. It also seems likely that the same body would have accomodated a pair of the later and larger M.A.N. 23/23 (MTU 956) engines, allowing the installed power to go above 4000 HP. (Getting such a pair into the D1000 body, if possible at all, would have been a real "shoehorning" exercise.)

Perhaps, too, had Vulcan Foundry been involved, then English Electric might have found a ready application for its stillborn U-series high-speed engine.

The D600 design, without stretching, would easily have accomodated the addition of engine-driven generators for electric train heating, whilst retaining the steam generator if required. Also, there was adequate space for the easy installation of much larger compressors as would be required for train air brakes when BR made the conversion.

Presumably, later editions would have had the Brush electro-pneumatic control system in place of the BTH system, and as this speculative exercise imputes some foresight to those involved, it would have been made MU-compatible with BR's prevailing 'Blue Star' diesel-electric control system.

As said, the M.A.N. 18/21 engine was the worst of the 1950s German high-speed trio, and the NBL license-built version successfully plumbed even greater depths of poor performance. Evidently, M.A.N. did eventually work through the problems, just before the formation of MTU, but the latter organization did not adopt it for continued production, whereas the Daimler-Benz and Maybach designs were carried forward as the 493/652 and 538 respectively. On the other hand, M.A.N. must have put its 18/21 learnings into the 23/23, because that design, and not the Maybach Mercedes Benz MC 956, became the MTU 956 series.

In practice some of the best diesel-hydraulic designs were the standard B-B export models developed by Henschel and Krupp in the early 1960's, and first used in Thailand (Henschel) and Burma (Krupp.) Quite a few were built for various railways worldwide, and I understand that some of the Thailand and Indonesian examples are still in service. In the Indonesian case, it's worth noting that this was a railway that was also familiar with the "agricultural" simplicity of the EMD G8. So that the diesel-hydraulics lasted so well suggests that their putative complexity and "delicacy" was not too much of a drawback. Steve Palmano, January 2005.

Diesel-Hydraulic Locomotives / Hydraulic Transmissions 2.

Recently, Steve Palmano sent a classic post to the World Diesel List. It contains a great deal of information about diesel-hydraulic locomotives, particularly from the Eurpoean point of view -- appropriate since that's where the vast majority were built and used. Steve has given his gracious permission to reproduce that post here. My comments follow.

An incredible piece from a very knowledgeable man! You can read more of such things about European locomotives on the World Diesel List, which is a great source of information about locomotives worldwide. It presents no mindless banter; only meaningful discussion and actual events as they happen.

American fans may not understand completely some of the references Steve makes in his fantastic dissertation, so I'll go through it quickly and just add a few brief notes.

In the middle 1950's, the Deutsche Bundesbahn (that's the West German nationalized railway) issued a set of specifications for various sizes and power ratings of diesel-hydraulic locomotives they would eventually procure to replace steam locomotives in non-electrified areas. Various locomotive and diesel engine manufacturers responded.

The diesel engine makers were Mercedes Benz, Maybach Motorenbau, and Maschinenfabrik Augaburg-Nurnberg. This latter company is usually just referred to by its initials, MAN. The Mercedes engines eventually proved to be the best in German units which began to show up in ever increasing numbers by the late 1950's and early 1960's. The Maybach engine was acceptable, but not great. The MAN engine was absolutely awful.

When the Western Region of the nationalized British Railways decided to embark upon diesel-hydraulic locomotive procurement instead of diesel-electric, this immediately removed any possibility of having the primary contractors being located in Great Britain. All the technology resided in West Germany. So, various companies acquired licenses to produce German engines and hydraulic transmissions in England. Not only were these early engines, and so not fully debugged, but the precision needed for them to be built and operated satisfactorily just was not present, as Steve mentions. Thus, the poor MAN L12V 18/21 engine, which wasn't good in German-built form, was horrendous when built by North British Locomotive (NBL.)

Steve Palmano has given all of us an excellent insight into, and great running start at leaning about, hydraulic transmissions in locomotives with his great article. I am indebted to him for it, and for his permission to reprint it here.