>>1991483Freight traffic is almost entirely loco-hauled as it has been proven time and time again that it is best to move tonnage at moderate speeds and loaded to the brim. There were some attempts of making freight EMUs, but all were unsuccessful. Which, of course, makes sense - as it is better to have a strong lead vehicle and lightweight ( relatively ) freight cars. You both get flexibility and only once maintenance heavy vehicle.
For passenger trains - except for some few exceptions - lightweight, distributed powered MUs are superior.
1. Locomotive does not take space at a station. EU stations are, increasingly, built for 200 or 400m long trains and it is best to have all that 400m available for boarding.
2. Distributed power allows for higher acceleration. In a highly trafficked commuter system, this might save you 5 seconds at every start/stop, but that adds up quickly if you have 40 stops on a line, or need to stop every mile or so.
3. Maintenance costs depend on speed and axle load - so you want your 200-300km/h trains to be as lightweight as possible. For this reason, french LGVs are limited to 17t and Japanese go as low as, if I recall, 12t on Shinkansen lines. This is expressed in infrastructure access costs, where it is relevant. In EU running an emu with 12 ton axle load will be cheaper than running a train of equivalent weight, but with a 22,5t axle load loco at the front.
4. Plain physics - while in fright trains you can get a 4:1 net-to-tare ratio, it is much worse for passenger trains. This is especially bad for shorter trains - as two double-decker car train weighs just a smidge more than a locomotive pulling them. You can get the same capacity of an emu with half the weight.
...
