Science: Energy: Gravitational
Railroad Downgrade Power Generator   (+5)  [vote for, against]
What goes up, must roll down

Since I'm going to be using the terms a lot, let me begin by offering some railroad-to-English translation:

'downgrade' = 'hill with train going down it'

'unit' = 'x number of locomotive(s) coupled together and under power'

'string' = 'x number of cars coupled together'

'train' = 'locomotive(s) coupled to string of cars'

When a train crests a ridge and starts down the other side, the unit has hundreds or even thousands of tons coming down behind it. It manages this weight by braking, of course, using one or both of two systems: 1) -friction brakes, in which cast steel shoes (on the locos and some especially heavy cars, like tankers) or composite rubber shoes (on the majority of car types) are applied directly to the steel wheels via impulse air (the system is complicated to explain and irrelevant to this idea), or 2) - dynamic braking, in which (to put it very simply) the deisel-electric motors that drive each axle on the unit are "reversed," turning them into dynamos that slow the train by building up massive E-M resistance (again, a gross over-simplification), incidentally charging the unit's batteries but bleeding off most of the energy they produce by converting it to heat and blowing it out the top of the locos.

This is wasted energy.

I propose that, on trains scheduled to traverse significant downgrades, a special car be coupled to the end. As the balance of the train shifts over the hill, this car deploys a pair of tailhook-style catch devices (activated by a light impulse on air-brake system, but not enough to set the brakes), latching onto receivers mounted on fairleads on each side of the track. As the train descends, the tailhook car pulls cables attached to generator cars running on parallel tracks. These are simply weighted flatcars fitted with the same dynamic motivators that propell a locomotive; the generator cars are coupled into strings calculated to offset the weight of the descending train, and they roll up the hill as the train rolls down. The dynamic braking is done by the generator cars instead of the unit, but instead of the power thus generated being converted into waste heat, it is sent to transformers via thick cables and converted to AC power, then fed directly into the power grid. At the bottom of the down grade, a simple trip mechanism unhooks the receiver cables and resets the tailhook car.

After the train is uncoupled from the cables and goes clickety-clacking off on its merry way, the generator cars roll back down the hill, using their motivators as freewheeling dynamos until it's time to slow down, when they once again employ dynamic braking-- generating electricity the whole time.

Going by some rough calculations done completely in my head based on information half-learned several years ago, I figure this would work best on 3.5 downgrades (because under that, drivers tend to use just friction brakes, even though they're not a-sposta, don't ask me why) of at least 1 1/4 miles in length. There are over 1,000 such downgrades in the US and Canada on key routes alone, not counting Class C roads and locals.

Setting this up would obviously require some considerable initial investment, which is anathema to the operators of railroads; perhaps the electric companies who stand to reap power from the installations could pony up a little dough, or (dare I say it) even the government. The US is a little tight right now, sure, but I hear the Canadians have wads of cash to blow right now, what with just having sold Vancouver Island to the Borg Collective. As for the physical infrastructure, all of the bits and pieces already exist, and specialty cars are not hard to build (my work partner and I built quite a few unique cars during my time with the railroad). The maintainance costs would be relatively low (by RR standards). I'm not sure how much power it could generate, but I am reasonably confident it would pay for itself in short order. I generally operate under the principle that anything is better than nothing

As an afterthought, I just realized that the "pusher" units stationed at long/steep up-grades could also use the generator cars to descend back to their sidings, meaning two round-trips for the generators with every train that passes through.
-- Alterother, Jul 31 2011 [zeno, Jul 31 2011]

Solid gravity storage slides [pashute, Aug 15 2011]

Demand Response http://en.wikipedia...iki/Demand_response
electricity prices link [Loris, Aug 15 2011]

It might be simpler and cheaper to install an electric third rail or overhead cable, and have the train dump its braking power into that. Of course, this would be equivalent to electrifying the steepest sections of the line.

Usually, when you have a supply of mechanical power at A and a desire for electrical power at B, you put a generator at A and run an electrical wire from A to B.

You are running a cable from A to C, putting a generator at C, then running an electrical wire to B.
-- spidermother, Jul 31 2011

I thought of the simpler version first, simply having the descending train power a fixed generator like a big wind-up toy, but then I realized the more complicated version could generate power twice, going up and coming down. I don't have the math/electrical engineering skills to figure out if this would be worth the trouble, but I thought I'd run it up the flagpole.
-- Alterother, Jul 31 2011

// install an electric third rail or overhead cable, and have the train dump its braking power into that //

This would mean modifying every loco on the line. Specialty cars are ridiculously cheap by comparison.
-- Alterother, Jul 31 2011

For diesel-electric locomotives the only major modification required would be a pickup (one per train). Or just provide one electric slug with a pickup per train. Disclaimer: I haven't worked on railways; please feel free to shoot down my suggestions as required.
-- spidermother, Jul 31 2011

Your suggestions aren't bad; they're certainly a lot simpler than my idea, but then most people's are. One point that comes from RR experience is that locos are pretty tightly packed critters to begin with, and considering that they are the lifeblood of the roads, the companies that own them are very leery of modifications, even 'kits.' I worked in the car shop, with the deisel shop next door, but I ate lunch with the deisel guys and got to listen to them bitch about this kit or that kit (usually upgrades required by the FRA that I didn't really understand). Meanwhile my partner and I were quietly plugging away at a triple-flat independent-gimbal-mount system designed for transporting 180' wind turbine blades (just as an example of one of many unique cars we built), working from blueprints we drew up ourselves, sometimes on the wall with soapstone. So I hope you'll understand if I tend to come at it from the carknocker's point of view (which, if you ask any carknocker, you will quickly learn is the only point of view worth taking).
-- Alterother, Jul 31 2011

Then again, there's no reason why we couldn't do your idea _and_ mine; it's all powered by the same gravity, after all!

A side benefit of having an external device to do the braking on big hills (I have no idea whether this has been tried in the modern RR, but they used to do similar things in the infancy of railroading) is that it saves w&t on the friction brakes, which have to be replaced frequently, and on the motivators, which have to be pulled out and rebuilt periodically (and that is a job and a half, entailing dismantling the entire underside of the loco). Pulling a car off the line to swap out a brake beam because some driver laid into the air too hard and burned up the shoes was a constant source of expense and frustration for us.
-- Alterother, Jul 31 2011

// You are running a cable from A to C, putting a generator at C, then running an electrical wire to B. //

You're right, that's too complicated; we could just put the transformer on the lead generator car, which would replace some of the big concrete blocks normally used for weight and would take a step out of the process, making for a slightly more efficient transfer of energy.
-- Alterother, Jul 31 2011

I don't have the bandwidth to watch videos right now; can somebody describe [zeno]'s link for me?
-- Alterother, Jul 31 2011

The trains are long, but not of uniform length or load. On key routes there are usually only two or three a day, and they try to keep them going the same direction if they can. One other thing about the railroad business to take into account: trains only make profits when they're moving. A fully-loaded train that's just sitting there waiting for the next one to come along not only isn't making money, but it's actually costing the road money, because the units are idling (16 cylinder deisels with more displacement than your hot tub don't just start on the turn of a key) and most of the cars don't belong to the road, they're leased. The faster a train gets where it's going, the faster the road gets paid for delivery, and the less the train costs to run.
-- Alterother, Aug 01 2011

Build the track in a giant loop, and run one train as long as the entire track.
-- spidermother, Aug 01 2011

[bigsleep] You work for a railroad, you pick these things up. I'd rather know something more interesting, but until my brain gets around to writing over it, I'm stuck with a bunch of railroad trivia. I have this problem- I like to learn things, just for the sake of learning. So, instead of just learning only what applies to me, I watch and listen and absorb everything I can. Only later do I sometimes realize it's useless boring shite that's only useful for arguing with [8th of 7].

[spider] Fancy sitting in your car at a grade crossing, waiting for a continuous train to pass by? Or can you handle taking the Boston-SanFran Loop to work every morning?
-- Alterother, Aug 01 2011

Agreed. At least my area of professional knowledge (steel, and the welding of it) changes very slowly, and grudgingly at that. Also, it is highly unlikely to become obsolete, no matter what anyone says.

Unlike railroads, unfortunately. If they don't change to keep up with the times (which they won't, trust me), our most efficient form of land freight transport will wither and die. I was laid off from the car shop a few years ago, and was offered a job with a competing road. I chose to move on instead. The rival road folded up a few months later.
-- Alterother, Aug 01 2011

You realize that modern trains are electric drive with turbine generators powering them, right? The only "mod" for this would be a few cars full of batteries, that is a LOT of energy.
-- AutoMcDonough, Aug 02 2011

The poster mentions diesel-electric locomotives, and batteries (turbine-electric is not exactly common). This is meant to be an alternative; more batteries might not be a good solution unless they become quite a lot better and cheaper.
-- spidermother, Aug 02 2011

Are you proposing that for an appropriate incline you set up a parallel section of tracks, and on that you have a load of old rolling-stock, weighted down with ballast? You then use your miles of cable to drag them up the hill when the train goes down.

Some thoughts:

a) a speciality car adds scheduling difficulties and drag throughout the whole trip, and presumably adds significantly to the initial cost.
b) You could keep the generator stationary, and power via the cable.
c) Store the energy to release on super-peak, when it's most valuable
d) Make all ballast cars the same weight. Then it's easy to select an appropriate number to act as counterweight.
e) add a passenger car as well and sell rides up & down the hill if there are villages at the top & bottom.

Edit: Oh, I see now your replies in the comments - or at least I understand what you're saying now.
You're not going to get any more power by putting the generator on (either) train. At best you could make it regenerate power on every incline (with no cables or extra track required), and then push from the back on the subsequent section. Hybrid cars do this already, and it does save the brakes. But then you need batteries, and lots of them.
You could have a hybrid system with a small generator/battery store on the train and a cable-powered generator for the big slopes.


Incidentally, what is a 3.5 downgrade? I doubt it's 3.5 to 1!
This isn't completely academic: to work out the energy available we need to know the height difference over the slope and also the mass going down it.

What would be more fun would be to redesign the track & wheels to take any achievable speed, then just coast down and hurtle onward unpowered as far as possible. :-)
-- Loris, Aug 03 2011

The grade system is kind of arcane, but it goes by percentage of a 45-degree slope, 45 being a %100 grade.
-- Alterother, Aug 08 2011

Ok, so a 3.5 grade slope is 3.5*45 degrees; about 1.6 degrees. The height difference is about 27.5 meters per km.
1.25 miles is about 2 km.
So the height difference is about 55 meters.

Potential energy is U=mgh where m is mass in kilograms, g is the Earth's gravitational constant 9.8 and h is the height in meters, and the result is in Joules.

So for a 1000 tonne train, the PE available is 1000,000*9.8*55=539,000,000 J, or 539 MJ.

If we assume 50% efficiency, that's 270 MJ. Electricity is generally measured in Watts (1 Watt=1Joule per second). Consumers tend to be charged in kW hours; 1 kH hour = 3600 kJ. One train under these circumstances generating 75kW-hour, or 0.075MW-hour.

If we're looking at selling power when it's most profitable, I found this:
//... in Ontario between August and September 2006, wholesale prices (in Canadian Dollars) paid to producers ranged from a peak of $318 per MW·h to a minimum of - (negative) $3.10 per MW·h ...// (source - see link).

I suppose that the peak payout varies somewhat, and we'll need to regenerate capacity before running out of ballast railtrucks. So let us take $100 per MW-hour. One train would yield $7.5.

Fuel & electricity prices might have gone up a bit, but I think this ball-park figure explains why such systems arn't common-place.

Now, if we could run, say, 50 such trains down the hill per day (one per half-hour on average), then maybe the system would be worthwhile.
Having a longer and/or steeper slope would be beneficial. What's the railway with the biggest drop, and how much traffic does it get?

If the system were adjusted to haul weights of some description (water, gravel, whatever), then one could have a larger capacity and wait for the highest electricity prices. And also, haul stuff yourself when it goes negative.
-- Loris, Aug 15 2011

Grade is percent by rise over run, not by angle. A 45 degree slope would be 100 meters rise in 100 meters run, hence 100% grade. A 3.5% grade would rise 3.5 meters per 100 meters of travel.

I am quite uncomfortable with [Alterother]'s assertion that //There are over 1,000 such downgrades in the US and Canada on key routes alone//. That seems awful stiff for a heavy-rail mainline. I can't say with certainty that he's wrong - but I'm pretty sure there's none here in Utah. And I would think that we'd be a logical spot for some, since the UP's 4000 series ("Big Boy", 4-8-8-4) steam locomotives were designed for nothing other than dragging 3600-ton consists from Ogden, Utah to Evanston, Wyoming (about 65 miles with a 1.14% "ruling" grade) without having to stick on a "helper" engine.
-- lurch, Aug 15 2011

//Grade is percent by rise over run, not by angle. A 45 degree slope would be 100 meters rise in 100 meters run, hence 100% grade. A 3.5% grade would rise 3.5 meters per 100 meters of travel.//

Ah... well, it won't make much difference to the numbers, anyway.

For this system to work efficiently, what we actually want is the longest, steepest slope we can find (up to maybe about 30 degrees - sorry, 66 grade). With a bit of finessing, one 'could' arrange for a large proportion of that drop in one place, then use smaller locomotives for the journey.
-- Loris, Aug 15 2011

I guess when they explained grade ratings to me, I got the dumb f#€%in' carknocker veresion. I admit it never occurred to me to verify that. My assertion of the number of 'significant grades' in North America comes from an article in a road mag I read in the foreman's office maybe four years ago. I was pretty much just trying to point out that there are a lot of places where this could be set up.

Most locomotives can't climb anything steeper than an 11 grade, if memory serves.
-- Alterother, Aug 15 2011

My Dad's retired from being a CivE doing rail layouts, and I can state that a 3.5 grade would be on the high side, and avoided if it all possible (I know it's not always, in the Alps or Rockies). The general target is as low as possible, with 1-2% being typical.

There's also the problem with this idea that the places where they can't avoid doing the steepest grades are also the places where it would be the hardest to build a parallel track section, since those lines are often hugging the edge of a mountain or similar.

I'd say run a third rail (easiest, not safest) or an overhead, use an over-sized helper engine to get up one side, and give it the connection to the power lines for the downhill side.
-- MechE, Jun 03 2013

Er...yes, a bit embarrassing, that. Since I posted this idea I have corrected my miseducation regarding grade calculation.

I like your idea, [Mech].
-- Alterother, Jun 03 2013

random, halfbakery