At well more than a quarter million a mile, as experienced by this demo project so far, this method of PTS will prove commercially non-viable; especially when cost conscious railroads finally admit that other methods of providing positive train separation are several times less expensive - and more effective anyway. But that won't happen until the deadline gets close. These last twenty years of "experimenting" on technologies already known by a vast majority of signal engineers to be non-viable; were, and are in fact, nothing but a delaying tactic. But several hundreds of millions in federal money have already been spent (wasted) anyhow.

Indeed, UP, CSX, and BNSF were not the only experimenters either. Over the last dozen years or so, several public transit agencies across the country have tried the radio ranging and CBTC technologies that these so-called PTS / PTC systems depend on. One in particular, along with its major railroad supplier, spent well over a hundred million dollars of ARPA money with nothing much to show for it but an unusable white elephant - a hundred million dollar white elephant. Besides, none have succeeded so far in actually providing failsafe "positive train separation" - can't locate trains in subways or tunnels at all - couldn't even tell which train was on what track - anywhere!. Federal lawsuits are already flying over these radio based systems - and against very important major suppliers, too.

Some of these experiments used radio ranging off GPS satellites. Others used ground based radio ranging antennas to locate the trains (especially in subways). All failed to correctly locate anything in motion to better than +/- 300 feet on the raw measurements, often actually locating the train as though it were on the street corner a half a block away, while the next measurement could put it somewhere up the other side of the tracks. These radio based methods must therefore use very fancy and sophisticated computerized statistical analysis of many measurements over much time to get any closer. Even then, with a +/- 50 foot statistical probability, none could tell which train is on what track.

JUST IMAGINE ! - controlling a train based upon where it probably is! -- slightly better than a w.a.g. guess I suppose!

Even the GPS in your car can only tell a probability of actual location to some ridiculous margin of error - the fancier the price, the smaller the expected margin or error. Note also, that the faster you go, the less accurate is the GPS output. Indeed, a friend often walks a nearby lake trail - while his GPS occasionally puts him in the lake at mere walking speed! He is amused to think something up in the sky thinks he is walking on water!

I can't even count the numerous times we've seen on TV, where some turkey turns onto a railroad track because he/she/it was blindly following a GPS based navigator that had located itself at a nearby intersection with a parallel street, rather than where it actually was - on a grade crossing! Some of these GPS errors have caused deadly accidents.

Where these GPS based systems have shined so far on the railroad, is in monitoring the condition and general location of field equipment, aiding the engineer in train vs track profile management, psuedo-enforcement of work limits and slow orders, and giving the engineer more information sooner than conventional signaling can do. Commonly, a big plus hoped for, has been the possibility of more finesse in selection of train speeds for best economy under various traffic conditions. However, neither GPS, nor radio based systems can do any of this in a safety critical way. Remember, it just will not do; to merely replace human error with machine error.

Without the ability to continuously and accurately tell which track which train is on, how can the system know which train to give which movement authority to. Even if the system merely tells the train which track the authority is good for, How is that to be enforced. Is it really positive train separation if we must rely on humans to input this data. Real Positive Train Separation will require real positive knowledge as to where the train really is. So what safety is actually to be gained with such flawed technologies. Note that conventional track based cab signaling systems have none of these problems - their only limitation is the amount of information they can communicate to the train (but they do it in a failsafe way).

Perhaps the ideal solution to gain all the possible benefits of all these technologies is to use both - each doing what each does best. But neither GPS, nor ground based radios are ever going to be enough technology alone, to actually achieve "Positive Train Separation" without a conventional track circuit based semi-automatic train control system as a foundation.

No wonder Sen. Feinstein is angry. She has been misled by certain vendors with a huge financial ax to grind - sold a bill of goods, as it were - told that GPS alone will do it. Being misled, she is confused and angry that MetroLink, which is aware of these realities, is preparing to install conventional cab signaling with semi-automatic train controls onboard, as are numerous other agencies. Some of these high tech vendors have lied to congress - she need to nail them!.

But she also needs to know that the proposed semi-automatic train controls to be installed by Metro-Link include all the functional requirements for an actually failsafe positive train separation system, such as signal aspect, speed limit, and braking enforcement, etc; that effectively stops a train before it can even run a red, or go too fast into a crossover or curve.

Indeed, a great many many systems around the world (TGV for example) have already had fully functional ATC in service for many years. She needs to find out why some vendors of these fancy new radio based systems have failed to mention this fact to congress. The fact is, that properly configured, track based automatic train controls do provide all the "positive train separation" asked for by congress, and have none of the problems of GPS or land based radio systems.

This tired old signal engineer is glad to see places like Metro-Link are finally getting on with it right now - especially after decades of obstinate American railroad opposition. Now, if the freight railroads would just get on with it too; instead of continuing to cover up their obstinate negligence, by pretending costly new technology is needed - IT AIN'T !

Yeah - I know - I'd get fired if they knew who I really was! -- But they don't !

OPB