Trains.com

Energy Efficiency of Passenger Trains

9572 views
43 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    October 2006
  • 1,123 posts
Posted by HarveyK400 on Monday, July 6, 2009 12:25 PM

oltmannd
....

I suspect that Amtrak does not keep good actual fueling data.  The frt RRs generally don't keep good data and Amtrak is generally in a worse position.

Engineering simulations would give good results provided you had a handle on the aerodynamic drag of the various consist configurations.  You'd have to get these from testing, preferably with a dyanmometer car or with some good wind tunnel testing on models..  Amtrak had (has?) a pretty good engineering test group, but I don't know if they've been into this.

You are right in that counting locomotives and assuming gal/hr just doesn't work.  It's not what's pulling the train that counts, it's how hard the train pulls back that determines fuel economy.

At $1.50/gallon and 79 mph, it probably doesn't matter much, but at $4/gal and 110 mph, it matters a lot.  I think there probably is a lot of room for improvment on the aerodynamic side of things.  Weight doesn't matter much unless you are constantly burning off energy to reduce speed for curves, etc., only to have to accerate back up to speed.  A mild hybrid locomotive would take care of this problem.

Now is the time to get busy.  The low hanging fruit might be to test matched train consists, replace cabbage cars with cab cars, test the effect of simple skirting and underbody aerodynamic modifications.

 

The problem is not calculating drag.  That has been tested and factored for various streamlining elements by Davis and others; and is still relevant today for Amtrak. 

I also would use the high end of rolling resistance to more accurately reflect modern deeply ballasted track structure and heavier rail.  The weight of the cabbage is significant for rolling resistance as well as for acceleration, even though this pales more in comparison to aerodynamic drag as higher speeds are reached.  A Cabbage is roughly equivalent to two coaches.

The time spent modeled for acceleration, cruising, dynamic braking, and idling will provide a pretty close estimate of fuel consumption.  Fuel consumption by throttle setting and output power cannot be calculated without access to railroad or manufacturer information.  Acceleration builds to Run 8 so as not to slip; but once track speed is attained, the power needed is less to maintain speed and depends on the track gradient.  This build-up is sufficiently congruent with the power needed and throttle position/fuel consumption (gal/hr) without exceeding the limit of adhesion.  I don't recall if fuel consumption rates have been shared for the Genesis P42.  Dynamic braking and recovering speed after a permanent restriction can be modeled with track chart or map information. 

The wild cards are the speed reductions and stops for temporary restrictions and signal indications.  With respect to irregular delays, logging fuel consumption by run might provide some insight, however, this includes non-revenue consumption for deadhead moves in and out of yards and to and from fuel racks, and idling during layovers.  Arguably, this too is part of the service.

The resistance of the train may affect the relative consumption of each unit; but efficiency comprises both the weight of the load(s) or number of passengers and the gallons of fuel consumed.  How hard the train pulls back is not a strictly accurate relationship.  If the Empire Builder weighs 2.5 times more than a Hiawatha but averages 2.5 times the number of passengers, that seems to be pretty bad for the latter.

Hybrid locomotives would seem to be worthwhile for intercity services.  That's as close as I'll go in agreeing with you.  A 3,200 hp diesel may be adequate to maintain track speed with a corresponding reduction in fuel consumption instead of one rated at 4,200 hp.

As for streamlining, this could be ignored at 79 mph; but I don't think so at 110 and above.  Tests by SNCF (Peters, RGI, Oct, 1982) indicate that as much as 56% of un-streamlined drag comes from under-floor equipment.  One suggestion was to pack equipment close together.  The lower level of Superliner cars also affords an effective improvement.  The next significant source of drag, as much as 21%, comes from the open space between the trucks of adjacent non-articulated cars.  Having a mix of locomotives, Amfleet, and Horizon cars affects the "skin friction," probably being effectively more comparable to heavyweight cars.  The suggestion for matched train sets is a good one.  Maybe surprising to some readers are the relatively small affects of front and tail design, full diaphrams, and faired trucks.

  • Member since
    October 2008
  • From: Near Burlington, WA
  • 380 posts
Posted by Maglev on Monday, July 6, 2009 11:01 AM

A comment on electricity:  on a tour of the Maui Electric Company Maalaea power station, we saw a combustion turbine which is essentially an airplane jet engine hooked to a generator.  But power plants are held to different emissions standards than airplanes, so a comparison of jets to electric powered trains must take that into consideration.  I also believe that high-altitude pollution from airplanes is a unique problem, although there is not much research to support this (there was one study conducted during plane groundings of 9-11 which showed dramatic effects...) 

Efficiency is also influenced by initial costs and impacts, as well as expandability.  Thus, for example, an under-utilized transit line might be a better investment than work on a highway which is already near capacity. 

The Trains carbon emission figures are for one person in an automobile.

And one last thought--how about electronic brakes to improve passenger train fuel efficiency?  I suspect there is much more benefit on freight trains.

"Make no little plans; they have no magic to stir men's blood." Daniel Burnham

  • Member since
    January 2001
  • From: Atlanta
  • 11,971 posts
Posted by oltmannd on Monday, July 6, 2009 10:41 AM

henry6

I wouldn't be too sure about that....railroads were always issuing instructions and admonishing firemen and engineers on how to better use coal to get the most power for the least amount of fuel.  Same with diesel.  Units that shut off when not needed, gensets, matching horsepower to use, using minimum number of units per train,  using electric power where and when available (for the most part, don't go the Conrail story here), rules of accleration, etc.  Railroads are keenly aware of thier power bills!

 I am sure of it.  They are keenly aware of how much fuel they buy and how much fuel is dispensed, but they have no idea of how much they put in each locomotive where or when. Really.  I'm not making this up.

The only exception is direct to locomotive fuel from a vendor.  They won't pay the vendor w/o a locomotive number.

Most fuel racks have a meter upstream of all the nozzles, not at each nozzle.

-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/

  • Member since
    December 2001
  • 8,156 posts
Posted by henry6 on Monday, July 6, 2009 7:46 AM

oltmannd

I suspect that Amtrak does not keep good actual fueling data.  The frt RRs generally don't keep good data and Amtrak is generally in a worse position.

.

 

I wouldn't be too sure about that....railroads were always issuing instructions and admonishing firemen and engineers on how to better use coal to get the most power for the least amount of fuel.  Same with diesel.  Units that shut off when not needed, gensets, matching horsepower to use, using minimum number of units per train,  using electric power where and when available (for the most part, don't go the Conrail story here), rules of accleration, etc.  Railroads are keenly aware of thier power bills!

RIDEWITHMEHENRY is the name for our almost monthly day of riding trains and transit in either the NYCity or Philadelphia areas including all commuter lines, Amtrak, subways, light rail and trolleys, bus and ferries when warranted. No fees, just let us know you want to join the ride and pay your fares. Ask to be on our email list or find us on FB as RIDEWITHMEHENRY (all caps) to get descriptions of each outing.

  • Member since
    January 2001
  • From: Atlanta
  • 11,971 posts
Posted by oltmannd on Monday, July 6, 2009 6:54 AM

Paul Milenkovic

One of the interesting questions is not just the system-wide average of Amtrak fuel usage, but how much fuel is used by the different services.  How much fuel is used by the Empire Builder per seat compared with the Hiawatha?  How much fuel is consumed by the heavy and aerodynamically mismatched NPCC (Cabbage Car) -- there are corridor trains with and without for comparison?  How does fuel efficiency increase if there are more cars per locomotive -- a kind of economy of scale trains are supposed to be good at?

These questions can be answered if fuel data were somehow released; they could also be estimated by doing detailed engineering calculations.  Engineering is just like accounting, only one deals with physical quantities such as mass, prime mover fuel usage at different power levels, aerodynamic drag coefficients, line profile and number of station stops, operating speeds and speed restrictions, instead of simply with money.  Counting locomotives and assuming a certain gallons-per-hour or gallons-per-mile usually doesn't do the job.

Why do we need to know this, and why does this matter?  Because if we are advocating trains, it would help our cause if we had firm numbers on current fuel usage along with increases in energy efficiency with certain changes or improvements.  Every other mode has worked hard at gains in fuel efficiency when trains have been coasting as it were on this question.

Excellent post. 

I suspect that Amtrak does not keep good actual fueling data.  The frt RRs generally don't keep good data and Amtrak is generally in a worse position.

Engineering simulations would give good results provided you had a handle on the aerodynamic drag of the various consist configurations.  You'd have to get these from testing, preferably with a dyanmometer car or with some good wind tunnel testing on models..  Amtrak had (has?) a pretty good engineering test group, but I don't know if they've been into this.

You are right in that counting locomotives and assuming gal/hr just doesn't work.  It's not what's pulling the train that counts, it's how hard the train pulls back that determines fuel economy.

At $1.50/gallon and 79 mph, it probably doesn't matter much, but at $4/gal and 110 mph, it matters a lot.  I think there probably is a lot of room for improvment on the aerodynamic side of things.  Weight doesn't matter much unless you are constantly burning off energy to reduce speed for curves, etc., only to have to accerate back up to speed.  A mild hybrid locomotive would take care of this problem.

Now is the time to get busy.  The low hanging fruit might be to test matched train consists, replace cabbage cars with cab cars, test the effect of simple skirting and underbody aerodynamic modifications.

-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/

  • Member since
    July 2004
  • 2,741 posts
Posted by Paul Milenkovic on Sunday, July 5, 2009 11:24 PM

Maglev

An article in July Trains uses information from the Bureau of Transportation Statistics to compare carbon emissions per passenger mile:

  • Transit--0.25
  • Amtrak--0.45
  • Toyota Prius--0.55
  • Aircraft--0.97
  • Average Car--1.10
  • SUV--1.57 

There are many other economic and environmental factors which may influence a determination of efficiency.  A mode of transportation that carries only a fraction of its capacity is one that can be easily expanded as population increases.

Don't know if electrically powered modes (part of Amtrak, much of rail transit) get any bonus points in these calculations of CO2, but I was under the impression that electric power was a negative with respect to CO2 on account of coal power, although the electric car could make up for it by using less overall energy than a gasoline car that has to keep an engine warmed up and turning at all times.

The most credible numbers for BTU/passenger mile, in my opinion, are on David Lawyer's Web site.  I have mentioned it before (just Google David Lawyer -- it is perfectly safe as he is not funded by the Cato Institute, has no known ties to the Concrete Lobby (he asserts that he is an environmentalist), and his site has no advertising).  His hypothesis is that the automobile was on a per passenger mile basis an energy-saving mode compared to the trains the automobile replaced, steam powered trains or electric interurbans powered by stationary power plant steam.  He asserts that the reason we are using so much more energy is that we are using many times more automobile passenger miles than the peak of rail passenger miles back in the day. 

The reason we are driving so much is that the automobile offers such flexibility and convenience that we can.  Back in the day when rail was king, people by and large didn't travel very much on average because travel was inconvenient.  Perhaps with automobiles we travel too much because automobiles are too convenient, permitting living arrangements that are wasteful sprawl that would be unworkable if travel were rationed to what trains could provide.

David Lawyer's numbers are at variance with what Trains is reporting, and his numbers suggest (different sources vary and Amtrak reporting has varied over time, with a large spike in recent times that had been corrected wtih revised statistics) that Amtrak is somewhat more energy efficient than planes and cars, but certainly not by the factor of two Trains is claiming.  I question whether a Prius is twice as efficient as an average car, especially in intercity highway driving that is the comparison to Amtrak.  If you are talking about a non-SUV car, the sales leader for many years has been the Toyota Camry, and at steady highway speeds, a Prius offers perhaps a 20 percent saving in fuel use (25 percent increase in gas mileage),  certainly not a 50 percent saving (double the gas mileage).  I average 35 MPG in a Camry at 65 MPH highway speed, which is confirmed with a computer model I have for converting car aero drag, weight, engine displacement, and gearing into gas mileage.  I defy anyone to tell me they get 70 MPG at 65 MPH in a Prius -- this does not happen unless they are feet from the back bumper of a truck.  So, the Trains people have some 'splainin' to do on their numbers.

David Lawyer also offers credible explanations why trains may not be as fuel efficient as some may think.  On one hand, trains have low rolling resistance per ton, they can benefit from the economy of scale of a large engine in a locomotive compared to the many small engines in a large number of cars, and train cars offer some aerodynamic advantage by blocking the wind of one another.  On the other hand, there are a variety of reasons why trains are heavier per seat than cars and certainly airplanes -- the vibration environment of steel wheel on steel rail contact that provides low rolling resistance requires a stouter coach structure for an equivalently smooth ride to a rubber tire vehicle, the train of coaches that lowers air resistance in turn requires greater buff and draft strength for safe operation of the individual coaches within that train, also the amenities of increased leg room, wider seats, the getting-up-and-stretching-your-legs room that is expected of rail travel along with non-revenue baggage, lounge, dining car, low-density sleeping cars, all add to weight.

The other thing about trains is that whatever amenities -- baggage, lounge, crew dorm to have on-board staff, dining car, individual room sleeping cars, leg rest coach seats -- that reduce the number of seats per train car also increase energy consumption.  There is nothing inherent about having these amenities, which to many railfans and passenger advocates are the whole essense of train travel.  What is so "1950's" about those amenities is they were made more widely available at that time in a vain attempt to prevent the loss of business to planes are cars, and here in the 21'st century we believe these amenities to be essential if not intrinsic to trains.

Load factor, or low load factor, should not be viewed as a failing of train travel but as another amenity.  If you have 50 percent load in coach, you can sprawl out across two seats for overnight sleep in some imitation of the open section, conductor permitting this -- with 100 percent load factor, you would have a seat mate.  With lower load factors, there are plenty of free seats; with high load factors, you may not get to chose a convenient time to travel.  Amtrak could increase its energy efficiency several fold by using the same seating density, absence of getting-up-and-walking-around room, no lounge car for chatting with strangers from across America, same fare structure and reservation system to occupy every seat as the airlines.  Amtrak travel would also become unpleasant as air travel, perhaps much more unpleasant because you would be occupying your cramped seat with knees jammed against someone's seat recline for longer periods of time.

One of the interesting questions is not just the system-wide average of Amtrak fuel usage, but how much fuel is used by the different services.  How much fuel is used by the Empire Builder per seat compared with the Hiawatha?  How much fuel is consumed by the heavy and aerodynamically mismatched NPCC (Cabbage Car) -- there are corridor trains with and without for comparison?  How does fuel efficiency increase if there are more cars per locomotive -- a kind of economy of scale trains are supposed to be good at?

These questions can be answered if fuel data were somehow released; they could also be estimated by doing detailed engineering calculations.  Engineering is just like accounting, only one deals with physical quantities such as mass, prime mover fuel usage at different power levels, aerodynamic drag coefficients, line profile and number of station stops, operating speeds and speed restrictions, instead of simply with money.  Counting locomotives and assuming a certain gallons-per-hour or gallons-per-mile usually doesn't do the job.

Why do we need to know this, and why does this matter?  Because if we are advocating trains, it would help our cause if we had firm numbers on current fuel usage along with increases in energy efficiency with certain changes or improvements.  Every other mode has worked hard at gains in fuel efficiency when trains have been coasting as it were on this question.

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?

  • Member since
    July 2006
  • 9,610 posts
Posted by schlimm on Saturday, July 4, 2009 4:11 PM

Of course the issue of layovers with idling engines wouldn't be an issue if we had more electrified lines using low and/or non-carbon power generation sources.

C&NW, CA&E, MILW, CGW and IC fan

  • Member since
    October 2006
  • 1,123 posts
Posted by HarveyK400 on Saturday, July 4, 2009 1:35 PM

DMUinCT

Who cares about how effcient?  

Passenger service is a SERVICE, intercity or commuter, provided by a Non-Profit government, Fed or State, or City.  

You can have the government build roads and parking garages or run passenger trains, it's the peoples choice.  Cars, trucks, airplanes mean more air polution.  It takes time and traffic jams to switch people to trains....



I could not disagree more.

If trains do not offer energy efficiency, reduced emissions, and more effective peak capacity, rip out the tracks and add expressway lanes and parking garages.  Emissions are related to energy efficiency and fuel consumption.  Buses can provide the service.

  • Member since
    December 2001
  • 8,156 posts
Posted by henry6 on Saturday, July 4, 2009 1:25 PM

DMUinCT

Who cares about how effcient?  

  Passenger service is a SERVICE, intercity or commuter, provided by a Non-Profit government, Fed or State, or City.  

 

 

But that's the determining factor of who get the money to build and expand!   And, while once the domain of private enterprise, it is now under governments controls by default: big business didn't want to deal with dealing with governments.  Could it become private and/or for profit?  Time will tell.  Railroads someday just might want the traffic whether for income or to get something else from government.

RIDEWITHMEHENRY is the name for our almost monthly day of riding trains and transit in either the NYCity or Philadelphia areas including all commuter lines, Amtrak, subways, light rail and trolleys, bus and ferries when warranted. No fees, just let us know you want to join the ride and pay your fares. Ask to be on our email list or find us on FB as RIDEWITHMEHENRY (all caps) to get descriptions of each outing.

  • Member since
    October 2006
  • 1,123 posts
Posted by HarveyK400 on Saturday, July 4, 2009 1:13 PM

The questions can be broken down into two parts: overall TRE energy efficiency, and the cost of the extended dwell time at certain stations.

Don't get bogged down in the details unless you feel compelled to verify consumption.  The easiest way to figure the approximate p-m/gal is to ask for the annual Trinity Rail Express diesel fuel consumption and compare that to the passenger-miles from estimated from ticket sales.  As with autos and buses, some energy is used for lighting and air-conditioning.

Actually, if two sets go back and forth more frequently, the high utilization mitigates the long layovers and idling between peaks of more typical commuter operations.  While consumption is higher, more passengers are served which I assume is a policy goal and cost.

Worst case is that 30% utilization of a 2-car train still comes to about 90 passengers, more than filling seats in two buses and contributing to diverting some peak travel from driving.

Albuquerque is interesting; but not entirely unique, in long-distance operations.  Occasional long station stops are built into long-distance schedules in part to allow making up for delays in route, and in part to provide smoking breaks for passengers.  Albuquerque is longer than usual to allow shopping at the Native American art and craft vendors.  The long stop is possible because of the low volume of freight traffic on this part of the route.  Considering that the train is cruising along most of the day with the notable exception of terminal layovers, enegy efficiency for transportation is as close to ideal as it gets. 

The problem of added weight for baggage, dormitory, diner, and lounge cars on long-distance trains becomes apparent.  While Superliners weigh more than single level cars, the increased capacity results in less weight per space. 

Load factors are perhaps the greatest factor in fuel efficiency.     

  • Member since
    November 2006
  • From: Southington, CT
  • 1,326 posts
Posted by DMUinCT on Saturday, July 4, 2009 1:08 PM

Who cares about how effcient?  

  Passenger service is a SERVICE, intercity or commuter, provided by a Non-Profit government, Fed or State, or City.  

  You can have the government build roads and parking garages or run passenger trains, it's the peoples choice.  Cars, trucks, airplanes mean more air polution.  It takes time and traffic jams to switch people to trains.

  A between 3:45 and 8:15 each week day, Boston Commuter Rail (MBTA) trains will make the 44 mile trip to Providence RI with local stops.  Eight, fully loaded, double deck coaches moving at 80 mph to the Rhode Island capitol.   Then back to Boston nearly empty.  (the later trains stay over in Pawtucket RI storage for the next morning runs.

Don U. TCA 73-5735

  • Member since
    October 2008
  • From: Near Burlington, WA
  • 380 posts
Posted by Maglev on Saturday, July 4, 2009 11:30 AM

An article in July Trains uses information from the Bureau of Transportation Statistics to compare carbon emissions per passenger mile:

  • Transit--0.25
  • Amtrak--0.45
  • Toyota Prius--0.55
  • Aircraft--0.97
  • Average Car--1.10
  • SUV--1.57 

There are many other economic and environmental factors which may influence a determination of efficiency.  A mode of transportation that carries only a fraction of its capacity is one that can be easily expanded as population increases.

"Make no little plans; they have no magic to stir men's blood." Daniel Burnham

  • Member since
    December 2001
  • 8,156 posts
Posted by henry6 on Saturday, July 4, 2009 8:26 AM

Sam, with your statement it become clear that 1) the inherent efficiencies of rail are there and 2) would be maximized if more people would use rail. So, my conclusion is that we should encourage the use of passenger rail to maiximize its efficiencies.   And remember, when operating a transit system, measurements are based on total service rather than individual trains or stations.  Those that are  bad enogh to be a drag on the sevice efficiency or its costs, will get weeded out, however.

RIDEWITHMEHENRY is the name for our almost monthly day of riding trains and transit in either the NYCity or Philadelphia areas including all commuter lines, Amtrak, subways, light rail and trolleys, bus and ferries when warranted. No fees, just let us know you want to join the ride and pay your fares. Ask to be on our email list or find us on FB as RIDEWITHMEHENRY (all caps) to get descriptions of each outing.

  • Member since
    July 2006
  • 9,610 posts
Posted by schlimm on Friday, July 3, 2009 10:44 PM

Sam:  With your accounting background, you're very skilled at digging through these statistical reports.  I think the answer is in this one:

Normal 0 http://transportationfortomorrow.org/final_report/

C&NW, CA&E, MILW, CGW and IC fan

  • Member since
    April 2003
  • 305,205 posts
Energy Efficiency of Passenger Trains
Posted by Anonymous on Friday, July 3, 2009 10:09 PM

In theory passenger trains are more energy efficient than cars, light trucks, and jet airplanes.  But are they in practice? 

On Wednesday I rode the Trinity Railway Express (TRE) from Dallas to Fort Worth and back.   Approximately 30 per cent of the 3:05 p.m. train seats appeared to be occupied, which squares with some numbers that I had gotten from the TRE in 2006.  The returning rush hour train appeared to have a 50 to 60 per cent load factor, which is a bit lower than the average 80 per cent rush hour load factor that I calculated from the 2006 numbers, but it is in line with the Fort Worth numbers, which are lower than the Dallas numbers. 

It is 33 rail miles from Dallas to Fort Worth.  It takes the TRE approximately 55 minutes to make the run.  The trains have two to four cars.  Each train lays over at its end point for an approximate average of 25 minutes.  Thus, 45 per cent of each run is spent idling, which clearly uses less fuel than running, but I wonder how much of the running energy efficiency is lost because of the end points dwell time.

When I was in Albuquerque in May, I popped by the transit center on several occasions to observe Numbers 3 and 4.  Number 3 is scheduled for a 50 minute service stop in Albuquerque; Number 4 is carded for a 43 minute stop.  On both days Number 3 was early by an average of 25 minutes, thus making the average dwell time 75 minutes, whilst Number 4 was early by an average of 30 minutes, thus making the average stop 73 minutes.  The trains were pulled by at least two locomotives, although on one occasion Number 3 had three locos.  Again, I wonder how much of the running energy efficiency of the trains is negated by the relatively long layovers in Albuquerque.

During FY08 the Southwest Chief had an average load factor of 63.1 per cent.  Overall, Amtrak's trains had an average load factor of 52.3 per cent, which brings me to these questions.  What criteria were used by those who claim that trains are more energy efficient than personal vehicles and airplanes to come to their conclusion?  And did they consider average load factor, dwell times, etc.?

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy