> My understanding is that bulk power generation is probably more efficient than having the motive power in the locomotive itself.

Almost certainly. But this probably evaporates once you need to pull significant amount of battery.

Trains are very efficient at carrying weight, especially if they can dump power into the grid (or battery) when braking. That metal-on-metal wheel has super low rolling resistance.

> Trains are very efficient at carrying weight

Sure. But you also need a whole lot of them to provide sustained power to move a large train for a decent trip length.

> especially if they can dump power into the grid (or battery) when braking

Smaller benefit here to regen. Sure, a 10000 tonne train has a lot of kinetic energy-- maybe 3 gigajoules or something. But compare to > 50MW to keep that train moving at a decent clip. So a complete stop with 100% of the energy recaptured is only about 60 seconds of rolling resistance energy use.

I think the cost of batteries is going to be a bigger issue than the weight. Which is why hybrid (battery combined with overhead power lines) looks so attractive to me.

> Which is why hybrid (battery combined with overhead power lines) looks so attractive to me.

It may be good for some segments, but consider the number of charge cycles and charge/discharge rate, along with the peak charging currents on the overhead lines.

For, say, 10000 tonne trains, 15MW sustained is a good very optimistic target for how much power you need to put out to keep rolling (before you make this any worse with adding battery mass and before you consider e.g. slopes, braking, etc).

If your goal is a discharge rate of C/1, you're carrying 75 tonnes of batteries-so it masses about 1% more.

Then, what duty cycle are you charging at? 100% coverage from the overhead power lines means you need to bring 15MW into the train. If the goal is a big saving in the amount of overhead power lines needed, 10% coverage from the overhead power lines means you'll instead need to bring 150MW into the train during those spans (and, this is a really aggressive 10C charge rate...)

I think C/1 is too aggressive for good cycle service, 10C is waaaay too aggressive for charging-- and these are best case scenario (flat ground, etc). If you multiply the amount of batteries by a big amount, now the mass change to the train and the rolling resistance contribution to the train from batteries becomes significant).

edit: in the end, I think you're usually better off just electrifying the portions of the train route where it's easiest, and burning diesel the rest of the way, instead of trying to somehow dotted-line-overhead-electrify things and limp through with batteries.

Agree that 10% coverage is too low, but for 50% coverage charging speed falls to 1C, which is totally doable (and as you say, for diesel-electric hybrid it could start making sense at even lower percentages).

The Pareto principle means that even if battery capacity means you need 80% or 95% coverage you could still see large savings over a 100% electrification.

Remember that you can also coast. If it's just about a low bridge or short tunnel you might lower the pantograph a few 100 metres on either side of the bridge. No need to wear the battery at all for such a short stretch - even with the brakes fully applied the train takes about a mile to stop. Trains don't currently do this because they don't have sophisticated position-based pantograph control and if they had to do an emergency stop under the bridge they would be stuck, but with more advanced control computers and battery backup both those problems can be solved.

Where I live they are electrifying 150 miles of track. This means 85 bridges need to be raised, demolished or rebuilt, and it's a big part of the cost. And Denmark is pretty flat - there are no tunnels as far as I remember.

By the way, 50MW seems way high to keep the train moving. The locomotives are 3-4MW peak and I rarely see them use more than 3 or 4 of them on a train. And I would expect them to be using much less than peak when they are not accelerating or hauling up a steep grade. https://en.wikipedia.org/wiki/GE_Dash_9_Series

50MW is too high (I overestimated freight train speed); I revised to 15MW on my previous post (too low).

> The locomotives are 3-4MW peak and I rarely see them use more than 3 or 4 of them on a train.

3-4 locomotives is typical for a "small" 3000 tonne train.

When you get to the 10,000 tonne monsters, you might have 8 GE Evolutions distributed, capable of 24MW sustained.

It's also worth noting this is not really enough. If rail freight was a little faster, it could be used for more things.

> 50%, which is totally doable

The US rail freight market and scale is different from Denmark. I don't think you understand how much goods travels over track which is near nothing and very long. Especially in the Western US. E.g. Los Angeles to San Antonio is just one of many major rail route. It's through 1200 miles of largely desert with nothing around. Estimates are that it would cost $5M/track mile to electrify.

Just to note the scale difference: Banedanmark moves about 2.6 billon tonnes-km according to Wikipedia. BNSF and Union Pacific railroads-- the #1 and #2 rail freight carrier in the US-- move about 1500 billion tonnes-km.

On the other hand, it takes about 12L of diesel to pull a tonne of freight across the distance.

We'd be far better off trying to just move more freight to the existing rail network than to try and electrify it, if we're looking at things from a standpoint of reducing CO2 emissions.

>> 50%, which is totally doable

Don't misquote me, dude. Here's what I wrote:

> for 50% coverage charging speed falls to 1C, which is totally doable

I was saying that it's doable to charge batteries at a rate of 1C.

> Don't misquote me, dude. Here's what I wrote:

Sorry-- When I quote from earlier ones I lose the window and I sometimes screw up quoting.

ON THE OTHER HAND, you seemed to imply 95% electrification was plausible... so I don't think I got the spirit of your comment wrong. If 50% isn't even plausible, then it's all a bit moot, isn't it?

[Also, note that if you have 50% coverage and are charging at 1C, you're also putting 1 cycle on the battery per 2 hours].

I think you're generalizing from a small market, with much less rail in empty space, to a massive market, with lots of rail in empty space, that moves a very large chunk of goods by rail.