How much power is generated by dynamic braking?

[dldance]

QUOTE: Originally posted by arbfbe

I have checked the power meter logs on the Oakway SD60s on the BNSF several times. The three highest use positions are Run 8, Idle and Dynamic braking but I don't remember the particular order. Idle was way up there. Throttle 1-7 were negligible.

Flywheels have one major problem on a locomotive. They do not like to change directioins which may cause serious problems going around curves. Flywheels also work the best when they have a larger diameter and relatively speaking locomotive carbodies are sort of narrow for the weights involved.

The newest diesels can produce far more dynamic braking horsepower than locomotive horsepower. Those traction motors have way more capacity than the diesels that feed them.

Sounds like the way my teenage sons drive - ma***he throttle - slam on the brakes - and idle to watch the girls.

On a more serious note, the Active Power flywheel energy storage system encloses the flywheel in an evacuated chamber. This allows them to use a smaller flywheel at much higher RPM. Thus the directional issues should be minimized. The flywheel uses composite materials and is designed so the the first failure mode is delamination. Thus in the rare event that it does self destruct - it turns into something more like a feather duster as the layers separate - rather than shrapnel.

I know that there has been research on use of flywheels in city transit in Europe but do not the the current status. That leads me to believe that the initial applications were not economically successful. Still it sounds like an interesting area for potential research.

dd

[Anonymous]

Composite flywheels in vacuum are a fantasic technology that almost works.

the problem isn't the rotors or the motors, or even the vacuum. it's the bearings. supporting 20 lbs rotating at 250kRPM is hard enough without vibration and shock to deal with. magnetic levitation bearing technology sshows promise, but are particularly sensitive to orientation changes.


if anyone has ever played with a conventional steel gyroscope toy, you may have encountered a mode where a sharp jolt or several well-timed nudges can get the whole assembly to wobble violently around its center of mass at something near the fundamental rotational speed of the rotor. (try it sometime with a bike wheel)
when a power-storage flywheel gets into that mode, the forces are usually sufficient to destroy the rotor. As Dldlance points out, the composite sort usually don't get far outside their containers, but there's still all that energy being released as heat, noise and vibration-- something akin to dynamite exploding inside the can; it happens about that fast-- a memorable experience, according to a co-worker. Even though the rotor is contained, the energy has to go somewhere. . . I'm not saying the problem is unsolvable, just that it hasn't been solved for mobile applications yet.

Wire hybrid: Neat idea, and workable with todays' technologies, but (as MWH likes to explain) capital intensive. (Though once the initial investment is in place, the incremental cost to add trains is attractive. Any government policy makers in the house? Here's a place where throwing money at a problem might do some good.)
It also moves a portion of the emissions produced moving goods over hills to large, fixed places where they can be more effectively mitigated, and allows the USA's HUGE stocks of coal to be directly used as a primary power source to transportation.

Another advantage (from a railroad point of view) is all those diesels that are idling to stay warm could earn money pumping power into the national grid at the same time; or, as an alternate, they could be shut down and use grid power to say warm and restart-- no more waiting for jumper cables when a battery goes south; just raise the pan.

[carnej1]

The FRA and the US DOE have had a program to develop a flywheel energy storage unit for locomotive use for a number of years now. Last I read this was proceeding forward and the flywheel system had been built and tested in stationary applications. It's supposed to be mounted in an old Bombardier LRC body and mated to a Bombardier HST 5,000 HP "Jettrain" locomotive. I've also read that several freight railroads have studied the concept for use with conventional diesel engines...

[dldance]

QUOTE: Originally posted by carnej1

The FRA and the US DOE have had a program to develop a flywheel energy storage unit for locomotive use for a number of years now. Last I read this was proceeding forward and the flywheel system had been built and tested in stationary applications. It's supposed to be mounted in an old Bombardier LRC body and mated to a Bombardier HST 5,000 HP "Jettrain" locomotive. I've also read that several freight railroads have studied the concept for use with conventional diesel engines...

Stationary flywheels work well - we have a client with several in service in Europe and are looking at additional installations. But as crazytechie says - getting them to work reliably in transport service is another thing. I will watch those experiments with great interest (from a distance.)

dd

[ckape]

crazytechie: I hadn't considered sending the electricity to the power grid before. You could set up a caternary system that only receives power from the locomotives, just on stretches with heavy dynamic brake use, and where your locomotives will be idling for extended periods. Transmitting the electricity from the braking-heavy areas shouldn't be much of a problem with modern voltage converters and whatnot. I'm not really sure how the power grid would handle spikes in power like that, so I don't really know if that would be an issue.

This way you don't have to worry about scheduling to make sure that you've always got a train going uphill whenever you've got another locomotive using the brakes.

[Anonymous]

CKape-

that's the beauty of being grid-connected: when a locomotive goes into dynamic and starts pumping power into the system, all the other gensets' load-sensing regulators back off a smidge. The national grid has several TW installed, and runs close to capacity. Even figuring a dozen trains each with 4 or 5 locomotives each dumping 0.75MW into the grid, that adds up to 42MW-- about what a single modern Aeroderivitive natural-gas fueled peaker set puts out; peanuts to the grid as a whole.

the big question is what an unpredictable +/- 42MW, (envelope) semi-chaotic swing would do to *local* line conditions; It might be smart to insulate the rail grid from the consumer grid with a series of storage stations; small pump/generate hydro, superconducting loops or a flywheel farm could absorb the shocks both outgoing and incoming. At the same time you could frequency convert so rail could run on a more convienient frequency. (DC, 25Hz, 400 Hz have all been suggested as possible substitutes)

Out here in Kalifornia, we're trying to corner the market on those Aeroderivitive plants to keep from having blackouts again. There really are a lot of those going in, and they're designed to be fast response-- something like 90 seconds from 'secured' to full output, and even faster to regulate to local line conditions. So even without "buffer" stations between the RR and the consumer, I think the grid would be OK, assuming proper frequency stabilization.

[dldance]

OK – so I ran some numbers on using dynamic braking to generate power to the grid. I used the grade from Soldier Summit down to Thistle Junction in UT (ex DRGW) as an example. That grade drops about 2000 feet of elevation in about 27 track miles for an average grade of -1.4%. It has good access to the grid with power generating stations at Helper and Price.

If we make the following assumptions:
- 8 trains per day
- 4 locomotives per train (2 lead + 2DPU is common practice for the climb from Helper to Soldier Summit)
- Average downhill speed 25 mph (a train covers that stretch in about 1 hour)
- Max braking power of about 2.5 MW per locomotive (per Randy’s corrected numbers)
- No efficiency losses
- cost of catenary about $1 million per mile ($27 million)
- average revenue $0.4 per KWhr

That stretch would generate about $1.1 million per year in gross power revenue. Assuming no maintenance costs and neglected the cost of adapting the locomotives, payback is almost 25 years.

dd

[Anonymous]

Ok, since we're in scribble-on-envelope mode here, i spliced in a few of my own numbers:
(and using your simplifiying assumption of 100% electrical efficiency)

-off-road Diesel Fuel, Delivered to the property: $1.00/ Gallon
-prime movers produce roughly 14hp/hr per gallon in notch 8 (that's a number out of the air based on aircraft numbers that are 15 years old; modify to suit)
-8 trains down implies 8 trains up the same grade, same time at max power (might be a bit of a stretch, but we're playing here. Up time seems like it would be longer than down.)
-neglecting costs of converting the locomotives (might be reasonable as the locomotives could be used on many 'wire-hybrid' grades)
-assuming free maintenance of wires and connections (emphatically NOT valid, but again, we're playing with numbers here)
-50% power gets recycled in moving trains up-grade, also 'billed' at $0.04/KwH

I come up with just shy of $2k/day in avoided cost of diesel, which, when added to your numbers gives an 'accelerated' payback of 15 years or so.

On the face of it, I agree: that's not a terribly tempting investment, even at today's low interest rates. (which i see we've neglected)

I predict, however, that the rate of increase in the cost of grid power will be lower than the rate of increase of diesel fuel; particularly with the available sources of low-sulphur crude shrinking and the EPA already thinking about Tier III. So that improves the return on investment a bit.
Then some improvement in traffic control might be feasable, based on someone's "micro-block" suggestion, and might get more trains over the hill (up and down) which would bring the payoff horizon closer still.

the question becomes one of "when does the payoff horizon get close enough?" I submit that we (as a nation) are going to be increasingly in competition with China and the pacific rim nations for all mineral resources, not just oil; and that the longer we put off major upgrades to our infrastructure, the more expensive (in real GDP) we're going to find those upgrades.

[dldance]

The hill that I chose is interesting because some of the coal mines are at the top of the hill. So over the stretch of track we are looking at the major traffic pattern is empties up and load down.

I agree with your assessment, crazytechie, this is just close enough to reality to be worth further consideration. The wire maintenance costs may be a killer in the long-term but for the first ten years, they ought to be reasonable. Snow might be an issue here.

dd

[daveklepper]

Intelligent thinking. Goes along with upgrading the power companies transmission lines and better electric power for the future. Tunnels would not have to enlarged if DC 750 volt center Lional style third rail was used in the tunnels, a pefectly feasible option. I'd stick with the same 25,000 or 20,000 volt 60-cyle AC for catenary even if the power company's transmission lines are dc, for reasons of safety and easy voltage boost or cut. But the two prime candidates for electrification in my view are Omaha - Ogden and Harrisburg - Pittsburgh. Traffic density.

[Anonymous]

Empties up, loads down probably biases the equation to a faster payoff; if we get half the 'down' power back as 'up' power, and only require (say) 75% power to get up, that starts looking pretty good, no?

Snow does complicate the picture some, (ice on wire, Flash-over on the insulators, etc.) though traction effects and increased drag, etc would not change markedly for a wire hybrid scenario than it would in a single-power situation.
But, cold weather also gives some more options for energy savings: units stationary on the grade ( for whatever reason) could make money (to offset the cost of idling to stay warm) or conserve fuel (depending on electrical engine heaters) by being connected to the grid.


Dave- why DC on the center rail? adds complexity if the 'native' power of the line is different than the tunnel power. Complexity = cost.
Elsewhere I have elucidated some of the problems with DC as a motive power; in particular, the problem of sustained arcing would be a problem in the dusty, grimy, oily, possibly damp environment between the rails in a tunnel. Not that AC would automatically be better. . .no matter what sort of power it is, high voltage and oily dirt are a bad combination. (not to mention dragging equipment!) I think we'd be better off with an outside third rail.
But maybe i'm overlooking something?


a note on personnel safety:
Especially in areas where someone might inadvertantly get across the line, AC is probably safer than DC. (despite what Thomas Edison wanted us to believe. . . DC makes muscles clench ; AC causes spasm, tending to throw the victim clear. Of course at 750+V AC or DC, my (personal) biggest worry is that I WOULD survive- minus major parts of my anatomy. I've seen it. Big craters and burns, skin grafts that don't take. . .it's not pretty.)

[arbfbe]

Yes, but if we just had a refrigerator sized cold fusion reactor in each locomotive then we could just waste all the dynamic braking energy with no effect on the overall ecomony.......

Alan

[daveklepper]

DC ONLY in tunnels and approache/transition zone and only because it is extremely expensive to say raise clearances in the Moffat Tunnel to put in catenary. Center third rail, Lionel style, can give greater current carrying capacity the side third rail, won't bother special freight cars that might not clear side third rail, and would be powered only when trains are about to enter, are in, and leaving the tunnel. The transition zones would have heated conrete roadbeds to insure no ice buildup on the center third rail and would be as short as feasible. Low voltage dc has less insulation problems than ac (remember that the peak voltage of ac is 1.4 times the rms equivalent power voltage), and is easily designed into compatible locomotives that are a diesel serving as the road slug for the electric in electric territory and the electric serving as the road slug for the diesel when outside the electrification. Snow and ice affecting caternary is a problem that has faced electric railroads for years, but they manage through frequent service, by having two pans (or trolley poles) or more collecting power, and by special sleet and ice cutters on the pantographs. The North Shore interurban continued to provide good service during some really terrible snow storms that closed highways complete for days, and they did it by continuing to run trains back and forth all night. You are absolutely right about third rail not be appropriate for a mountain electrication, but it is the appropriate technology for avoiding having increase tunnel clearances only. I'd use for the Moffat and for other long tunnels, and raise clearances on short tunnels, looking at the cost benefit figures in each case. The voltage has to be much lower in a third rail installation, even a center third rail, then catenary because of the insulation situtation. I think the highest voltage, dc, ever used on a side third rail was one interurban at 1500 volts, and I'm not sure it stayed that way. 500, 600, and 750 is typical in modern rapid transit lines today.

[dldance]

In mountain railroading - with combined diesel/electric operations - why electricfy the tunnels at all. With distributed power on long heavy trains, the lead engines could retract their pans while the midtrain/training power continues to push. I know that wouldn't work for long tunnels like the Moffet but it would work well on the shorter tunnels that are more common - such as the Thistle tunnels on the Soldier Summet grade.

dd

[arbfbe]

Based on the MILWAUKEE Road electrification you are making way too much of the problems of ice on the trolley than actual practice has shown. Pans and wires work quite well. It is easier to just divert the water away from the wire.

When the MILW was investigating the upgrade of their electrification from 3300 VDC to 10K VAC the added insulation required for bridges and tunnels was in the 2" to 3" range for air insulation. Something like High Density Polyethelene plastic sheeting would likely solve the problem with a 3/4" thickness applied to tunnel roofs.

While it would be nice to capture the electrical energy lost to heat in dynamic braking it does not seem fuel costs are yet high enough to cause a capital investment necessary to recover it. Electrification, especially for short runs will likely be the last option the railroads will apply. Batteries, compressed air and flywheels seem the most likely.

[daveklepper]

Electrification in the long run will come when the power companies need the railroad rights of way for transmission lines, and when the USA get serious about energh independence, and when oil prices continue to rise and advanced nuclear power reduces electrical energy costs.

[Anonymous]

the rule of thumb for air insulation is 1" = 10KV, in clean environments (like offices or labs)

the roof of a tunnel with diesel-powered coal trains rumbling through it at all hours doesn't strike me as the cleanest environment. . . So let's cut that back to 3" = 10KV. So, if we take a "standard" transmission line voltage of 25kVac, apply the RMS correction (root 2), and we're looking at 1 foot clearance from the tunnel roof, and a further clearance of 1 foot from the doublestacks. . . Which probably have the occasional corona point (a deep scratch will do) and all sorts of other crap on top of them. (dirty snow comes to mind) So add another couple of inches for safety,
Hey, presto, Dave is right- third rail is the cheap and easy way thru tunnels. That, or limit 'Wire Hybrid" to a smaller loading guage. (not the best way to gain wide accepance)

So we can look at the poly. Plastics attract dirt. (just ask any model railroader about dirt and plastic wheels) Dirt may insulate at 12Vdc, but I'm here to tell you that it conducts pretty well at 30kV. . . . So you'd need a LOT of poly to increase the leakage path to the point where flashover would be minimal even with the dirt. Then you have to worry about all the hot, carbon-laden gas coming out of the exhaust stacks of the diesels and melting the poly, or worse.

Which brings up a point I hadn't thought of: Diesel soot (carbon and traces of sulfuric acid) on the caternary insulators and wire. . . I wonder what the solution was in the steam days when you had steam under the wire?

Still sounds cheaper to do the third-rail thing . . . or follow dldances's theory and bull through on diesel power alone and put the pan back up on the other side. Cheaper still except for the odd crew that doesn't get the pan down fast enough. Or go to the opposite extreme and prohibit diesels from anything other than idle in the tunnel, and run on the wire. The pull-the-pan option seems cheapest to me. . . No fiddly high voltage troubles at all.

My goal here is to try and think out a (relatively) low-first cost way to improve the energy efficiency of railroading. . . I think I'm talking myself out of the Diesel-Wire hybrid idea.

(come to that, railroading is pretty fuel efficient except for the idling part. maybe that's the thing to go after, not the Dynamic brakes. . .EMD and GE seem to think so.)

[arbfbe]

techie,

Added insulation as in more than was already there for the DC system.

It's OK there are a lot of words in this thread for sure.

Alan

[modorney]

On a BART train, about 90 percent of the braking is dynamic braking. Friction brakes cut in around 4 mph or less.

Usually, the dynamic braking feeds a big resistor grid, underneath the car. However, if there is another train that's in your power circuit, then the other train can use your electricity. However, that usage is pretty local, a train coming down the hill to Orinda can't help a train climbing the hill after Pittsburg.