For the last two years I’ve been paying increasingly close attention to trailblazing work by Norfolk Southern (NSC) in the field of battery-powered locomotives. My interest was piqued in June of 2010 when Norfolk Southern hired Axion Power International (AXPW.OB) to develop a battery management system that would allow rail locomotives to run on battery power and recharge their batteries through regenerative braking. I believed the decision was positive news for Axion because nobody hires a battery manufacturer to design a BMS for somebody else’s product. My enthusiasm was tempered, however, by knowing that an earlier Norfolk Southern retrofit, the NS 999, was unveiled in September 2009 and quickly proved to be an insurmountable challenge for the AGM batteries that were used in the original design. I also knew that a technical development project for a Class I Railroad would require a couple years of work before a rational implementation decision could be made.
A key milestone was reached this week when Axion announced that NS had ordered $475,000 of PbC® batteries that will be installed in the NS 999 over the next couple months. The two companies are also moving forward on a parallel development track for a larger and more powerful long-haul locomotive that will use twice the battery power.
Now that I have a clear data point on battery costs for both the switcher and a long-haul locomotive, I can finally dig into the fundamental economics of what Norfolk Southern is trying to accomplish with this project.
The raw unsubsidized numbers are amazing!
The technical case for a battery-powered switching locomotive is easy to make because they operate in urban rail-yards assembling and disassembling trains, which means travel distances are short and charging infrastructure is both easy and cheap to install. On average, a switcher spends about 75% of its time idling and only 25% of its time working. While it’s a little dated, this 2004 graph from the Argonne National Laboratory summarizes daily fuel consumption for a GP38-2 switcher, the locomotive platform Norfolk Southern used for the NS 999.
While the exact numbers used to generate the graph are not available, my best estimate of total fuel consumption is about 340 gallons per day, or 85,000 gallons for a 250-day work year. At the current price of around $3 per gallon for off-road diesel, the annual fuel cost for a switching locomotive is $255,000. Since switching locomotives are already equipped with electric drive, dynamic brakes and power control systems, the biggest cost of converting a locomotive to battery power is pulling the fuel tanks, diesel engine and generators and replacing them with a big rack of batteries and a custom BMS to keep things in balance.
I can’t accurately estimate the power subsystem costs for a locomotive conversion, but I’m certain that it will be less than the $400,000 per MW that Sandia National Laboratories recently estimated for grid-based power subsystems that use expensive inverters for AC-DC conversions. Using a high-side estimate of $400,000 for the balance of system costs on a switcher retrofit, the cash-on-cash payback period from fuel savings alone will be about 3-1/2 years. Using a more conservative mid-range estimate of $200,000, the cash-on-cash payback period will be closer to 2-1/2 years. When you factor in collateral environmental benefits like 875 tons of annual CO2 abatement and the elimination of diesel emissions from urban train-yards, battery-powered locomotives offer compelling value and utility.
The planned long-haul locomotive, which will cost twice as much but save about 170,000 gallons of diesel fuel per year, should offer similar economic and environmental benefits.
The following table provides summary information on the locomotive fleets operated by the four largest North American Class I Railroads.
|Multiple purpose units||3,904||3,579||7,632|
In light of the economic and environmental benefits of battery-powered locomotives, it seems reasonable to assume that at least half of the switcher fleet and up to 20 percent of the freight locomotive fleet would change to battery power over the next decade if the second-generation NS 999 and the planned long-haul locomotive perform as expected. The batteries required to support such a transition would cost something on the order of $4.4 billion.
Today, the biggest question on everybody’s mind seems to be “since laboratory testing took two years how much longer will it take to test the two prototypes before implementation begins in earnest?”
To understand the likely deployment timeline, you need to understand what the laboratory testing accomplished. As I noted above, a typical switching locomotive spends 75% of its time idling. In a battery laboratory, much of the idle time can be eliminated and a company can pack the equivalent of three or four year’s use into a single testing year. It can also stress the batteries with extreme load profiles that go beyond the normal operating range.
In a situation like the Norfolk Southern project where parallel testing was conducted concurrently in three battery laboratories, a company can condense the equivalent 18 to 24 years of application experience into two years. So by the time the batteries are installed in a prototype vehicle, the battery manufacturer and the user know how they’re going to perform and the principal goal of the prototype testing is to identify any complications that were missed during the laboratory phase.
In light of Norfolk Southern’s prior experience with the NS 999 and the extraordinary amount of laboratory testing that was conducted over the last two years, I expect the next round of prototype testing to be measured in months rather than years. While I’d be pleasantly surprised to see significant implementation beyond the planned freight locomotive in 2012, the ramp rate in 2013 could be impressive because the market is so large and the economic and environmental values are so compelling.
In the November-December 2009 issue of its employee magazine BizNS, Norfolk Southern explained that the biggest challenge was developing “an energy management system that allows us to take maximum advantage of kinetic energy,” emphasized that “advances in battery technology will be the primary driver for widespread industry use of electric locomotives” and observed that it was “eying the use of lithium-ion and nickel based rechargeable batteries, as well as improved lead-acid batteries.”
Two and a half years later, there’s only one contender left standing, Axion’s PbC®.
Battery-powered locomotives are a potential billion-dollar niche market where size and weight are not mission-critical but cost and performance are. There are several comparable niche markets in automotive applications like micro-hybrids and stationary applications for power generation, distribution and use. I expect the PbC to be a formidable competitor in many of those markets.
Disclosure. Author is a former director of Axion Power International (AXPW.OB) and has a substantial long position in its common stock.