The train that powers its station | Hacker News Reader

The train that powers its station

bbc.com

60 pointsSlashmanX42 comments

The tube already does regenerative braking. This trial is about better regenerative braking. Currently, the trains return power to the track when braking, but if the track is already at capacity the remainder is dumped via resistors on-board the train. If I understand correctly, this new substation can sense when this is about to happen, take that extra current and feed it through an inverter to supply it back to the power grid.

Some more details are here: http://eandt.theiet.org/news/2015/sep/tube-brake-energy.cfm

I don't fully understand the physics behind it, but how exactly does can the track be 'at capacity'? Isn't the track just fed from the power grid, so if the demand on energy on the track is lower, won't it just take less energy from the grid?

It's like the voltage regulator on your car. The car's alternator is pumping out power and the regulator is topping up the car's 12v system (actually 13.9). The regulator doesn't want to create an over-voltage situation and so dumps any extra power to a resistor, turning it into heat. On an electric train, instead of a battery the regulator is pushing power back to the third rail. Once the rails max voltage is reached, when it is at capacity, everything else is turned into heat via resistors on the train.

The London Underground uses 630 VDC traction power. Power distribution is 11KV/22KV 50Hz, converted to DC at substations every few stations along the right of way.[1] 630V is too low a voltage for long distance transmission without big losses, so distribution is at a higher voltage. AC to DC conversion is with rectifiers, which are one-way; they can't make AC from DC. With inverters, that's possible. Inverters are basically big switching power supplies. This is just an inverter installation at a substation. It's probably bi-directional, converting AC to DC or DC to AC as required.

The whole transit system isn't one big DC circuit; it's in sections of a mile or so. Regenerative braking can only power trains in the same section, unless traction power can be up-converted back into the AC system. If more trains in a section are braking than accelerating, the excess power has to be dumped somewhere, usually into big iron resistance grids that waste it as heat. Using inverters eliminates that energy dump.

[1] http://www.trainweb.org/tubeprune/tractioncurr.htm

I found this confusing:

  For each day of the five-week test, the system captured a full megawatt hour — enough to power more than 100 homes for a year.

This doesn't add up. I'd like to believe what they mean to say is "enough to power more than 100 homes", but the average UK home uses about 4.6 megawatt-hours per year, or 12.6 kWh per day. This is, more accurately "enough to power less than 80 homes each day".

Perhaps that's per-train, and I don't doubt that this is a significant energy (and heat) savings. But this is a surprising inaccuracy from the BBC.

The whole energy-vs-power confusion seems problematic. This is a grid-connected system; there are no batteries. Thus it seems average power would be a more helpful measure than energy, when talking about train-brakes and households both. If one insists on discussing energy, why not quote it in joules or BTUs or whatever? One rarely sees average power used in these contexts however, so there must be something about it that the power companies don't like.

(Mega)watt hour is a common unit of energy in electrical power because it's convenient. (For example, my power company reports my power usage in kWh (~100 kWh for my family currently)) A 60-watt lightbulb will use 60 watt hours over an hour.

>why not quote it in joules or BTUs or whatever?

Retail electricity is sold by the kWh, consequently that is the unit that most readers will be familiar with. Of the readers who prefer to use strictly Si units in their daily lives, those people, aside from frequently converting to/from various units, are aware or can quicky find that 1kWh is 3.6MJ. For anyone who wants to use BTUs, well, it's a free country, I suppose.

At least in the US, electrical energy is generally measured in kWh. Joules would be annoying to the reader, even though it's simpler.

I think the constant confusion just comes about from journalists trying to write emotionally compelling articles. Saying that all 5 days recouped enough energy to power a single home for a year sounds pretty unimpressive. So the interviewee extrapolates out to if the system ran for the whole year or otherwise were scaled up. But having no technical background (and thus little respect for technical details), the writer edits their phrasing to be much simpler. Treating "power" and "energy" as synonyms, we end up with gross misstatements like in the article.

(And then of course the fact that such gross inaccuracies get past the editors and actually published shows you just how detail oriented the modern news org actually is. Just think of that any time you read any thing!)

Strange, I though all trains had that.

There is this little anecdote about regeneration in trains: there is a mine ore train next to the Swedish/Norwegian border that goes downhill to the harbor full of ore and uphill empty, even with the efficiency factor, it's still sending quite some power to the general electric grid.

Do you have a link with more info? It would be a nice story to submit.

https://en.wikipedia.org/wiki/Iore - mentioned in passing in an earlier thread at https://news.ycombinator.com/item?id=8732383 .

I might well have read it here or on reddit, I don't remember.

I guess it's ok to propagate a factoid as long as it's true.

I got bit by this once: a factoid is an anecdote that people share thinking it is true, but it is not. From reading your comment, I think you just meant "... ok to propagate a fact as long as it's true.". Just a friendly heads-up :)

https://en.wikipedia.org/wiki/Factoid

Your Wikipedia link then continues 'the term has evolved from its original meaning, in common usage, and has assumed other meanings, particularly being used to describe a brief or trivial item of news or information."

We no longer use factoid in that meaning. Even your clarification here is not really how Mailer defined it; as "facts which have no existence before appearing in a magazine or newspaper, creations which are not so much lies as a product to manipulate emotion in the Silent Majority". (See http://www.etymonline.com/index.php?term=factoid ).

It's on the same path that "computer" (someone who computes), "awesome" ("profoundly reverential") and many other words have been on.

And the same path of "couldn't care less" versus "could care less."

Does this apply to legal language?

I have no idea. How do you mean?

I was wondering what happens when the popular meaning of a word or phrase diverges from a law which uses that phrase for a different meaning or scope.

What happens is, things get complicated.

For a related example, consider the phrase "we settled the debt without prejudice" That sounds like it means there are no bad feelings, but it actually means that there may be future attempts to change the terms of the negotiation.

Nearly all railway vehicles use brake energy. Older trams wasted the energy with heating resistors. Newer ones feed back energy back to the grid if possible. Otherwise banks of supercaps are charged to allow going short distances without power (like old town where contact wire is not desirable).

http://www.bombardier.com/en/transportation/products-service...

>Nearly all railway vehicles use brake energy.

Do you mean nearly all new ones, or nearly all in service today?

> Spread across the entire Tube system, that technology would result in savings of £6m, about $9m.

Ok, but my understanding is that doing anything across the entire tube system' would no doubt cost trillions. Repainting all the guard rails would probably cost more than 6mil. So the proposition won't save a dime. It might be environmentally friendly and might be a good idea for new stations/trains, but the money would be better spent on other projects.

In short: Does spending a million on this regenerative breaking produce more electricity than installing a million's worth of solar panels on the roof? Which saves the most carbon per pound spent?

Presumably you would only add this to the train cars not the track. Also electric motors can generally also act as generators so the additional costs on new cars would probably be minimal.

PS: Cars only weigh 85,000 lb or so which means you only need to store something like 10kwh depending on top speed.

I'm only going from the OP. The savings they describe seem minimal (6mil) whereas they are dealing with a notoriously expensive tube system. Carbon savings are great, but one should always ask if that same money might result in greater carbon reductions if spent elsewhere.

London Underground would, I expect, /invest/ $6M in reducing heat in the tunnels. I can't find costs, but there have been other projects to do this.

In the summer, the temperature can rise to uncomfortable levels (over 30°C).

Maybe that's a net figure. Certainly I trust TfL wouldn't be doing it if it wasn't worthwhile - they're generally very well-managed.

Is it impractical to have the braking spin up a flywheel which then gets used to start the train moving when it departs? With the energies involved, I can only imagine that would end up being a monster flywheel with size, safety, and reliability concerns. And the energy from the flywheel would also be wasted overcoming its own inertia.

> In the first experiment of its kind, the London Underground has installed an inverter system that converts braking energy into power.

http://www.septa.org/sustain/blog/2011/07-06.html SEPTA has been doing something similar (but using the electricity more generally than for just stations) since 2011. I'm sure SEPTA hasn't been leading the way in this type of hardware either.

Not only that, but the London Underground already uses regenerative braking. Here's the press release from 2013 from when Alstrom got the contract to put in the system that is now being tested, http://www.alstom.com/press-centre/2013/5/alstom-to-supply-g... :

> Alstom was awarded a contract by UK Power Networks Services to supply its innovative Harmonic and Energy Saving Optimiser (HESOP) energy recovery system for the Victoria Line of the London Underground. ...

> London Underground already makes good use of regenerative braking but, by adding HESOP to the power supply arrangements, the residual energy that is currently wasted in braking resistors can be made use of – this will help prevent tunnel temperatures rising.

The central line can get quite toasty in the summer, especially at peak travel times. I'd be interested in an estimate of how much of that heat is a direct result of the current friction braking system. Will the stations get noticeably cooler if induction braking is adopted? "Enough [electricity] to power more than 100 homes for a year" coming out of a single station over just five days sounds like a ton of energy, all of which is currently being dissipated as heat.

I've always wondered if there was some way to allow buildings to prewarm water for boilers using subway heat.

It'd be an interesting experience without all the extra heat, especially in the winter. (Now we just need to go full circle and use that saved electric to heat the station :-p)

The article does allude to that.

> Additionally, regenerative braking doesn’t produce the heat that conventional friction braking does, so the tunnels themselves stay cooler, requiring less energy expenditure on climate control (and keeping Tube riders happier in the process).

Yes, but came here with the same question, "how much of the heat is the breaking"

First, note that the Underground already uses regenerative braking. This article is about a new generation of braking technology.

The Gizmodo article at http://gizmodo.com/london-underground-is-trialling-regenerat... is better than this BBC article. (And the similarities help show which parts come from the press announcement.) It also mentions a £6m/$9m per anum power savings across the entire system.

Further, the article at The Memo https://www.thememo.com/2015/09/25/regenerative-braking-tech... says "If rolled out, Transport for London (TfL) estimates that the technology could save £6m every year in energy costs, or 5% of its total energy bill."

The Wikipedia article at https://en.wikipedia.org/wiki/London_Underground_cooling says 'About 80% of the heat comes from the operation of the trains, 15% from other equipment, and 5% from people'.

Putting it all together, the extra savings (5% of 600 GW-years) is equivalent to about 1/2 of the heat generated by the passengers.

That's about all can figure out on the topic.

That statement in the article didn't make any sense to me. By my understanding it's a violation of the second law of thermodynamics. You can't take a process that creates waste heat and convert that into energy with less entropy, electricity in this case, and have less total resultant waste heat. Any physicists care to chime in?

Sure you can. As long as overall entropy increases, you're AOK.

Imagine a thought experiment not too dissimilar. You have a bike, that you pedal up to speed. You then let yourself slow to a stop naturally. All the energy you put in goes to heat.

Then you do the same again, pedal the bike up to speed, but this time you click a dynamo (attached to a battery) into place and let yourself stop naturally. This time some of the energy is converted to heat, and some into energy in the battery.

It the same principle of hydro electric power. You can let water just move down some tunnels from high to low gravitational potential energy (GPE -> water velocity), or you can put some turbines in the way.

Idea: Why are the resistors on the train? Why not make them stationary? Regulator-resistors attached to the rails could be used in fixed locations to heat water. Perhaps they could dump energy into capacitors to feed back onto the rail when needed. The trains would dump energy into the rail, raising the voltage and triggering the stationary regulators. They might not need to be anywhere near the busy stations.

Perhaps they should just put the stations on the top of an incline, then as the train enters the station the momentum would be converted to potential energy and back again as it leaves the station.

This is already done — originally (in the 19th century) so that seeping water didn't pool in the stations.

That was quite common in the days when locomotives were much weaker than they are now.

The Victoria line - the newest - is built with humpbacked stations exactly like you describe (where practical), and they save something like 16% on energy. But it's not really economically feasible to do that with the older tunnelled lines.

Reply on news.ycombinator.com