It ain’t easy. Most of the vehicles with torque converter drive that could be locked up for “direct” were relatively light, and the amount of heat generated in normal use can be astounding (as can the damage to things like pump turbines if sudden loads are applied, as when slack runs in on a heavy consist). Voith built an interesting ‘take’ on locomotives with multiple effective gear ratios: they had multiple TCs with different planetary or final-drive ratios, and “simply” by filling and draining these could vary the output torque without fixed clutches or mechanical moving parts and clearances in the drivetrain. Presumably these could be fairly easily locked up for direct in any of the selectable ‘speeds’.
We had a brilliant, but comparatively short period of experimentation with ‘hydraulic’ drive in the Sixties, first with the Krauss-Maffei ‘Amerika-Loks’ and then with the “Alco-Haulics”. These appeared to founder more on maintenance considerations and general ‘orphan’ status than on any implicit unreliability in the torque-converter principle at ‘locomotive’ scale … but I invite anyone’s knowledge or distinctive competence that says otherwise.
Voith is still very much in the rail business with its triple-converter drives (and other types e.g. single converter plus multi-ratio gearbox for DMUs - ZF is active in that area as well).
In Europe, those Voith ‘hydraulic’ drives (more accurately hydro-mechanical) are very popular for low to medium power switcher and road-switcher locomotives - combined with a high-speed diesel engine, it gets you a high power-to-weight ratio loco with naturally good wheelslip resistance, since in single prime-mover applications all of the axles are mechanically locked together via the final drives (it takes a lot of complex electronics to achieve the same effect in an electric AC drive loco!).
As SP & K-M found in the 1960s, it’s possibly not the best solution for high-tractive effort/low-speed lugging for long periods, but it’s still a competitive drive system for general rail applications.
My recollection is that Voith use the term “Hydrodynamic” to differentiate their drives from the comparatively rare (in locomotoves) “Hydrostatic” drive where hydraulic pressure is used in a hydraulic motor to rotate the axle.
On my first visit to Mackay in Queensland in 1972, I climbed int
I’ve always used the term “hydrokinetic” to distinguish torque converters and fluid flywheels from hydrostatic (e.g. vane-motor, positive-displacement) drives. Here is a PDF reference that describes some of the features and differences. Using that nomenclature avoids potential confusion with using ‘hydrodynamic’ in more common contexts such as tribology and naval architecture.
As a peripheral note: the method that is used on a large railroad locomotive to lock a torque converter (or fluid coupling) ‘ought to’ be a bit more complex than a simple Maybach or dog clutch, or similar mechanical device with positive engagement. Here is a reference that describes some of the concerns involved with locking a hydrokinetic device for direct drive; I think it is fairly clear how this idea would be adapted to optimize resistance braking without compromising the reliability or integrity of the transmission (and its fluid!)
Interestingly, the characteristics of torque converters of the kind found in normal ‘automatic transmissions’ are not at all favorable to fuel-efficient operation of many designs of road-vehicle diesel engines. I wonder whether at least part of the reported economy of the ‘semi-automatic’ bus transmissions relates to the absence of torque multiplication; were these transmis