Signaling complicates positive train control

A Union Pacific train heads eastbound on BNSF Railway tracks in 2013 underneath a signal bridge and past a positive train control installation in California's Cajon Pass. TRAINS: Steve Sweeney

Let's talk about signals.

Why do railroads even have them? Why are they different? And what does that mean for positive train control?

For starters ... signals control trains.

You knew that, of course.

So let's state the obvious: A railroad with only one train, like many small shortline or industrial railroads (or the typical toy train under a Christmas tree), does not need signals.

When signals come in handy is when railroads want to run two or more trains along one stretch of track. Train crews that obey signals keep trains from crashing, while also enabling managers to push through trains efficiently — freight, passenger, or a combination of both.

Back in 2016, I had the privilege of receiving permission to attend a roughly two-day University of Wisconsin-Madison Engineering Professional Development course on signaling and positive train control hosted at Southeastern Pennsylvania Transportation Authority's headquarters. 

(The course is still offered, check the university website for details.)

The instructors, Jim Hoelscher and Bob MacDonald, and their course book laid out similar fundamentals on literally the first page of instruction — after the introductions and outline slides.

How railroads arrive at those goals is some combination of practicality, money, and compromise. So the second question "Why are they different?" is answered differently for every railroad. Cash-strapped transit agencies might have to make-do with equipment that's been in place for decades — or with technology that is no longer available. Or that's what I took away from the course and in a separate conversation with the instructors.

Intuitively, long stretches of railroad that have directional traffic and don't include complicated interlockings and passing sidings might just need a simpler system. In signaling, that makes freight and commuter railroads fundamentally different.

Those answers bring us to the last question: What does this all mean for positive train control?

In short, I take this to mean that there is nearly infinite complication and new compromises to make for women and men attempting to design a safe system for controlling trains that works over top of and seamlessly with old technology, new technology, the most expensive, versions from the lowest bidder on a government contract, and the systems designed for railroads that don't exist anymore.

This is all a preface to the fact that in little more than six months, U.S. railroads must be compliant with a Congressional mandate to have operational train control technology — Dec. 31, 2020. With few exceptions, the railroads appear to be doing it.