Monday, February 13, 2012

Unorganized thoughts and sources on electrification

Initially gathered these for a post I've since dropped

Prices and Costs in the Railway Sector is a handy quick overview of very rough estimation for non-American rail costs. It's in 2000 Euros, which traded at parity with the dollar, so a simple inflation adjustment is all that's needed to bring costs up to today.

SNCF practice is to use a 60MVA substation every 40-60km and 10-15MVA autotransformers every 10-15km. These are the priciest elements of electrification, but not terribly so. Road overpass reconstructions to create clearance for electrification accommodating double stacked freight is almost certainly going to create major costs in excess of the actual electrification.

Catenary is sufficiently cheap (~$400,000 per track-kilometer) that any argument based upon the notion of delaying electrification because later improvements would require removal and reinstalling catenary ought to be ignored.

Purchasing ALP-45DPs without major modifications might be only 7.8 million per locomotive. Admittedly a high estimate compared to major freight locomotive purchases, but more affordable than their initial cost.

Amtrak's FY reports report a fuel consumption of 2.3 gallons per train-mile, but cost of diesel fuel is accelerating quickly. While Amtrak’s average for the year was $2.76 per gallon (for a total cost of $6.348 per train-mile), by September 2011 that had sharply risen and they were paying $4.38 per gallon ($10.07 per mile). Metrolink has currently secured fuel at a price of $3.25 per gallon and budgets at $3.40 per gallon in the current fiscal year ($7.48 and $7.82 per train-mile). Meanwhile Amtrak's 5 Year Financial Plan estimates a 6% annual increase in the cost of diesel with a price of 3.91 by FY2016 (page 45). I think this is probably on the optimistic side.

The British government sometime ago made available a study with a variety of actual fuel and electricity consumption figures for British trains. Of course they've redone their website since then and broken all the old links to it, but trawling around found it once more. 500 tons is close to the weight of a 6-7 car train set with American cars. Given LA Metro's price of 12 cents per kilowatt/hour, using electrical consumption figures from the Class 90 (22.62 kwh/mile), electrification with locomotives would drop fuel costs by two-thirds. EMU electrical consumption figures vary widely, down to as low as 5.7 kilowatt hours per mile, but also often being as high as a locomotive hauled consist. Off-handedly I want to suggest that third rail electrification may be more energy efficient than overhead.

Eyeballing the acceleration given in this presentation suggests a possible 20% improvement by use of electric locomotives over diesel. That said, I did not account for any difference in power capabilities.


  1. Sheesh, is Amtrak buying diesel at retail? I know I paid between $4.09 and $4.49/gal frequently this past year to fuel my Jetta TDI, but contract prices should be on the order of $0.50 to $1.00 cheaper.

  2. Overhead wiring is far more energy efficient with third rail. This is especially true of 25kV 50/60Hz systems once you crack the rather annoying technical challenge of having reliable regenerative braking systems on 25kV. (DC systems managed this fifty or more years ago, as you know with the Milwaukee Road experience).

    At that point the only significant losses in a traction supply system are the resistive losses which scale by current and thus are tiny for 25kV compared to ~750V third rail.
    Also, as your American Freight railroads have maximum speeds of under ~75-90mph over the majority of the network, non passenger lines could be electrified with simple tram derived fixed termination catenary systems if not full blown trolley wire.
    That cuts the installation and maintanence costs further.

  3. Instead of replacing bridge installations, freight railroads would probably prefer undercutting the roadbed to accommodate catenary installations inasmuch as is possible. It is an interesting technical challenge, but apparently cheaper and less disruptive than trying to rebuild road overpasses.

    BTW Norfolk Southern double-stacks clear the catenary at SEPTA's Norristown Transportation Center here in Philly. So it is certainly feasible.