Sisyphus’s train set | Hacker News Reader

Sisyphus’s train set

economist.com

49 pointsTomte17 comments

Efficiency numbers, please? Is there such a thing as a motor or generator that is more than 65% efficient? 0.65 * 0.65 is 0.42. Then add mechanical losses, and I^2*R losses in the power lines to, from, and along the train lines. Commutation losses for the pantographs (assuming a typical electric loco). Is off peak electricity really 4x cheaper than peak? Otherwise the losses kill the benefits. The article is short on math, which is sad, because the only meaningful question here is one that simple arithmetic answers.

Motors and generators tend to be way more 65% efficient. Random internet search mentions 70-98%. If they were not, pumped hydro would not be useful either.

Linked article is short on details, but the link wiredfool gave has more details including a video with numbers (2nd video 02:00). They claim >93% (Combined M&E) one-way efficiency. I don't know what one-way is, but 0.93^2 = 86%, which would make the loss noticeable, but I would still consider it good enough for storing solar/wind.

The problem is that lifting things up stores and consumes so little energy. You can lift yourself to the summit of Mount Everest for the calories in a single large meal, if you just consider the vertical climb.

But you got the horizontal to consider. With a train the horizontal is reduced due the tracks.

Yes, energy loss in the tracks (or pipes) affects both charging and discharging, so using rails is definitely better than some alternatives. But the problem is that even if the transport process is lossless, the gravitational potential energy stored at the top of the hill is not that much in the grand scheme of things. In this sense it would be better if the force of gravitational attraction were stronger. Maybe we should move to Jupiter?

Yep. Although I'm always surprised at how much energy humans consume. A 1200 kcal meal eaten quickly, for example an apple with 180g peanut butter (which I often consume in a few minutes or so) is the equivalent of "charging" my body at the rate of almost 21 kilowatts!

For comparison, that's the power consumption equivalent of running 40 household washing machines, a ceiling fan, a router, printer, bright LED lighting, a fridge+freezer, a large color TV, ten smartphones, a laptop, an electric heating blanket, two table fans, two smaller TVs, an electric shaver, a scanner, twenty more smartphones, three more table fans to keep them cool, two fountains or other water features, a small vacuum cleaner and a partridge [0] in a pear tree.

[0] Potts, G. R. 1986. The partridge: pesticides, predation, and conservation. (Daily food consumption of a partridge ranges from 71 kcal/day at 15C to 155 kcal/day at -15°C, which works out to be roughly 3.5-7.5 watts, or roughly the power consumed by a few smartphones.)

Is that the total energy the body uses from those things or the energy contained in them ?

Contained in them. It's shocking to see the difference. I saw this video the other day which is relevant. An Olympic cyclist powering a toaster to toast some bread.

https://youtu.be/S4O5voOCqAQ

True. But that's a bit like unplugging a harddrive from one computer and plugging it into your computer, and claiming that you just had a data transfer rate of petabytes per second, and then claiming that that's the equivalent of the bandwidth of all US citizens and corporations combined.

Hmm, not really. First of all, unplugging my laptop's 256gb hard drive and plugging it into another laptop, in ten seconds, would be equal to 25.6 gb/sec, or 0.0000256 petabytes per second. Even if you're doing the swap with desktop computers that are right next to each other, the largest hard drive [0] you can buy, and really fast switching, you would get a transfer rate of 0.003 petabytes/second. The US currently consumes an absolutely enormous amount of data, more than 23 exabytes/month [1]. (I looked up these numbers mostly because I was curious, and I gotta say 23 exabytes a month surprised me.)

Secondly, the power that the human body can store in just a few minutes is pretty cool. You can think of a person who eats 2300 calories a day as a 2700 watt-hour battery. Such a battery would be pretty powerful: it holds the equivalent of roughly 300 smartphone charges. You can "charge" the human body in just a few minutes of eating calorie-dense food like nuts or peanut butter or fatty food. But to charge a 2700 watt-hour battery, comparable to the human body, you'd need to draw the maximum power you could (from a wall in a house with standard wiring) - for about two hours.

Sure, I agree that the human body isn't really consuming the power until it gets fully digested. But the reality is that if I'm in a rush, I can wolf down an apple+190g peanut butter and be done 4 minutes later with 1200 calories in my body. Which is pretty useful in real life, at least for me. If my body instead "charged" from a wall outlet (which can usually deliver a lot of power, 1440W or so continuously) it would take an hour to provide me with 1200 calories. Sure, people often do eat slowly. But there have been many times in my life when there's not enough time for whatever reason, and taking an hour to eat a meal would be unacceptable. And it's times like that when I really appreciate that I can eat something fast, and "charge" my body at the rate of tens of kilowatts.

[0] http://www.pcworld.com/article/3096292/storage/seagates-10tb...

[1] http://www.cisco.com/c/m/en_us/solutions/service-provider/vn...

Yes, gravity is a very weak force compared to e.g. the Coulomb force.

I wonder how they get clean power out of the trains. At least in normal operation the power lines of the train network are terribly noisy. I guess there'll be some signifikant losses in cleaning that up.

There was a vox article 6 months ago that didn't generate much discussion that has a little more context, and a video rendering of the larger scale storage system: http://www.vox.com/2016/4/28/11524958/energy-storage-rail

Another proposed solution for buffering wind and solar energy is https://www.ecn.nl/news/item/energy-train-de-revolutie-voor-...: a train going round in a 5km diameter circular vacuum tunnel doing up to 2000km/h.

That hasn't been built yet (it seems more or less at the 'back of an envelope' stage to me) and would be more expensive to build, but it promises to store quite a bit more energy (20 GWh vs 12.5 MWh), so it might be cheaper per kWh.

I also think it can be built using existing technology and will be safe enough if the tunnel is built underground (derailments would be catastrophic)

One problem is that this only works at scale. Halve the diameter of the ring, and you likely have to halve the top speed of the train, decreasing capacity by 75%.

Is this like a very heavy flywheel?

In a sense: yes. Heavy, large, and all its mass is near its rim. For a flywheel, it's not _that_ fast, though; only about 125 revolutions/hour.

They got rid of the forces acting on the center of a flywheel that tear apart large fast flywheels by getting rid of the axle and using a rail on the outside of the wheel to counteract the centrifugal forces.

I would guess they want to use maglev as a rail system, as a conventional track may break down too soon at 2000 km/h. On the other hand, it is 'only' 15 km of rails in a controlled environment, and they can use enormous wheels, so it may be possible to lay conventional track a lot smoother than outside.

That reminds me of a more 'out there' energy storage mechanism I've looked at a couple times: Hyperloop

If you've got a large Hyperloop network with hundreds or thousands of ~40,000 kg vehicles shooting through it at 700 mph, each of those is 400 kWh of stored energy.

Of course, it'd be crazy hard to actually use because demand for travel on the network isn't gonna be constant, but it's an entertaining thought.

Reply on news.ycombinator.com