New Rail Cars Are Harder Than Moon Landings?

by Jim Mathews / President & CEO

Amtrak says it’s unlikely we’ll see new long-distance cars entering revenue service until sometime after 2030. And I believe them. But that’s because the status quo settles for talking about new train cars in 2020 and getting new ones in 11, 12, or even 15 years’ time, not because that’s the best we can do. Why are we settling for the status quo?

I’ve talked with a lot of people about this issue in the past month or so. Amtrak executives, supplier-industry executives, designers, middle managers at component companies, and others. One thing I’m told is that Buy America Act provisions mean we have to wait until we have a fully independent U.S. manufacturing base. Another is that Amtrak is inventing an entirely new customer experience, and that there’s nothing off-the-shelf anywhere in the world that can satisfy those needs. Still others say that industry just isn’t interested in delivering equipment on the timeline all of us know we need, because Amtrak just isn’t a big enough or important enough customer.

Those are real issues. They aren’t made up. But if we all collectively decide we are unsatisfied with the status quo, we can change that. We can beef up the Buy America waiver process. We can look harder for off-the-shelf answers. And $2 billion to spend on new rolling stock, thanks to the Bipartisan Infrastructure Law, is not nothing. We don’t have to settle. If we settle, it’s because we want to settle. Or we lack the ambition to try.

Consider this short and hardly complete list of really complex, ambitious, all-new things that took less time than we’re being told we have to wait for new long-distance equipment designs to hit the rails for paying passengers:

The Empire State Building, the tallest building in the world when it opened in 1931, went from pencil sketches to occupancy in one year and six months. It was delivered 45 days AHEAD of schedule.

The Apollo Moon landing, which required inventing whole new categories of technology to do something that no humans had ever done, took less than ten years. Neil Armstrong walked on the surface of the moon eight years, one month, and 26 days after President John F. Kennedy told Americans that “we are going to the Moon...because it’s hard.”

Or how about a more recent example? Construction on the 800-mile-long Trans-Alaska Pipeline System across unforgiving Alaskan terrain between Prudhoe Bay and Valdez began in 1975, and the first barrel of oil traveled through the pipeline in the summer of 1977. That’s less than three years (if you don’t count the five years of litigation blocking its construction over environmental concerns).

Many of you probably know that I spent three decades of my career in aviation before arriving here at the Association, developing specific expertise over that time in aircraft engines. For those who don't know, what's happening in modern jet engines in many ways defies physics. To squeeze the very last scintilla of efficiency from the jet engine operating cycle, engineers needed to create advances in metallurgy, in materials, and in managing the flow of air (a working fluid) at extremely high temperatures and pressures.

Airbus wanted a new engine for its A330 widebody twin rival to Boeing’s 777 widebody, and in order to develop, build, and certify these engines, Rolls-Royce Aero Engines (and, to be fair, engine makers General Electric and Pratt & Whitney, too) had to invent new ways to deal with the tips of fan blades moving at supersonic speeds without creating supersonic shock waves inside the engine, to keep internal components from melting with internal temperatures reaching 1,700 degrees or higher, and to do it while making sure that about 95 percent of what came out the back was water vapor.

Rolls started doing this science in June 1987, and a year later started spending real money to work on it for real after Airbus started making noises that they wanted Rolls power as an available option for the A330. Just a shade less than four years later, the new Trent 700 made its first run on a test stand and the regulators certified it for flight just beyond the end of 1993.

In other words, a bit more than five years to invent new physics and metallurgy and get it approved by tough regulators for human flight.

This is not a knock on rail as an industry but...it's just train cars, for heaven’s sake. A metal box on two sets of wheels. We've been doing it for over 100 years. Has technology improved? Yes. Are there things that are possible now that weren't possible even 10 years ago? You bet. Are there groups inside Amtrak that are pushing to create an on-board experience that's a lot better than the pathetic thing we have now? Absolutely, and thank God they are.

But none of those things can explain how it can take a dozen years to get new passenger equipment on the road for long-distance passengers.

A quick perusal of the Buy America waivers U.S. regulators have granted suggests that many are being done for relatively small percentages of the overall value of the procurement where the waiver is sought. It took a year, but in 2015 the Federal Railroad Administration granted a waiver to Buy America requirements so that Amtrak could buy the replacements for its aging Acela trainsets. But the waiver only needed to apply to 6.8 percent of the cost of the 28 new Alstom Avelia trainsets, because FRA only needed to grant an exception for eight components. I’m sensing a potential benchmark...

Perhaps one way to streamline the procurement of new equipment would be to recognize that, guided by the waivers that have already been granted, the Administration could set a policy of blanket waivers up to, say, 10 percent of the value of a procurement before requiring a buyer to formally enter the waiver process. Instead, a simple notification to FRA of an intended procurement, with a certification that no more than 10 percent of the listed components are sourced outside the U.S., and an explanation of how the circumstances of that procurement would otherwise meet the existing waiver criteria, might be a way to get things going more quickly.

You could sunset the new process after seven to ten years, giving the U.S. passenger rail industrial base time to spool up and be in a position to deliver 100-percent of the content required, which is really what the letter and spirit of the Buy America Act are all about.

We also ought to embrace the realization that a mostly off-the-shelf offering would probably satisfy a large share of the use cases in U.S. passenger rail. There are plenty of sleeper designs operating in Europe. There’s no need to re-invent every detail down to the last carpet fiber or LED light fixture. Few Amtrak riders who have ever taken Austria’s Nightjet have returned declaring that Amtrak's service is better than Nightjet.

If I were President Biden, I’d declare that we’re going to put long-distance passengers on new equipment before the end of this decade, not because it is easy, but because it’s the right thing to do and because we just passed a historic infrastructure investment law to do it. And it shouldn’t be anywhere near as hard as getting to the Moon.