Cascades 501 Wreck 18 December 17

[Tadman]

Washington State did just fine, they hired the wrong guy to operate the trains. Their TOC (Amtrak) has a record of accidents lately.

I'm in Spain this week. You wouldn't believe it. I've probably seen 100 Talgos, and not one of them jumped off the track and killed anybody on account of bogie design. Somehow they've all managed to respect that -9.8m/s squared and stay on the ground.

[justalurker66]

Jumping the tracks was a speed issue. I assume you did not see any Talgos violate the designated/design speed of the tracks on which they were traveling?
Killing people is what happened after the train jumped the tracks. Fortunately the Talgos you saw were not put to the test while you were watching.

110 mph in a curve rated half that speed doesn't work in Spain. The 2013 Talgo crash that killed 80 near Santiago de Compostela, Spain, demonstrated that.

[rcthompson04]

Yes that is the fundamental question. Are the Talgos going to handle a crash just as well as whatever is being used in the US? The 2013 crash you reference seems very similar to a similar crash in 2015 where the body count was substantially lower in the US.

[EuroStar]

110 mph in a curve rated half that speed doesn't work in Spain. The 2013 Talgo crash that killed 80 near Santiago de Compostela, Spain, demonstrated that.

This is oversimplifying it too much. It is the combination of the curve/track and the equipment that the laws of physics had problems with leading to the wreck. This equipment could have made a less sharp curve at this speed. This curve probably could have been made at this speed by different equipment (according to Talgo different leading engine). The easy solution to keep the laws of physics from taking the trains off the tracks was to reduce the speed, but once we are beyond considering that, the physics has no preference for which alternative would be better. Which one is chosen to blame is somewhat arbitrary. To the extent that fixed infrastructure is orders of magnitude more expensive, the NTSB is more likely to blame the equipment as it is perceived as the lower cost solution. Overall, from social benefit this is probably the correct approach, but I certainly find Talgo's argument very appealing that were it not for the heavy leading Charger, this could have been much smaller accident or completely relegated to a footnote somewhere as an over-speed report. I do think that it will be a pity if they retire the equipment before WASDOT has secured new equipment from Siemens or someone else that is capable of similar performance.

As for using some diesels to reuse the existing Acela coaches ... people are more likely to see pigs fly.

[John_Perkowski]

Any PTC system cut out brings a 79mph top speed restriction. Many lines running PTC have not lifted their speeds beyond 80 for passenger trains yet anyways.

How much of Amtrak routing beyond the Corridor is host road FRA Class 4, 5, or 6 track? If it ain’t Class 5, it’ll never be greater than 79mph.

[justalurker66]

The easy solution to keep the laws of physics from taking the trains off the tracks was to reduce the speed, but once we are beyond considering that, the physics has no preference for which alternative would be better.

I'm not blaming Talgo for the actual derailment - either here or in Spain. But what happens afterwards is their responsibility. That is where crashworthiness comes in.

[ApproachMedium]

Any PTC system cut out brings a 79mph top speed restriction. Many lines running PTC have not lifted their speeds beyond 80 for passenger trains yet anyways.

How much of Amtrak routing beyond the Corridor is host road FRA Class 4, 5, or 6 track? If it ain’t Class 5, it’ll never be greater than 79mph.

Probably a lot more than you would think. Anywhere that IITS was installed and they WERE doing 90 was maintained for that on the old ATSF. The RF&P was signaled for 90mph but CSX i dont think maintains it for better than 80 currently (has cabs) The 79 number limitation is for no train control system, 80mph is for with a train control system. Michigan also has regular scheduled trains that run 110mph and so does Illinois. New york state has on CSX (leased to amtrak) 110mph. The number of places that have track or signals better than 80 is out there for sure, some places have the signaling and protection but the track is not maintained to class 5. Class 5 spec is not that hard for them to upgrade to if the track is signaled and protected for it.

[Tadman]

This crash resistance thing is crazy. Let's try an exercise:

1. Airbus should make their planes far stronger so when they crash out of the sky at twice the legal operating speed, the passengers survive.
2. Ford should make Explorers much stronger so when they are driven off a bridge at twice the speed limit, the passengers survive.

See how idiotic this looks? Nobody would ever suggest this. The airline industry focuses on not crashing planes. In the recent Boeing 737Max debacle, not once has anybody suggested crash integrity. Never ever ever ever. You wouldn't find a buzzfeed or Guardian reporter suggesting such, let alone reputable news or regulatory agency.

The motor industry does have crash integrity standards, but not for driving a car off a bridge at double the speed limit or into a bridge abutment at same speed.

And as for PTC on automobiles? Yeah right. Right now Tesla is very firm that a driver must monitor autopilot, not the other way around.

[mtuandrew]

And as for PTC on automobiles? Yeah right. Right now Tesla is very firm that a driver must monitor autopilot, not the other way around.

Yet every other car has an iPhone or Galaxy seemingly glued to the driver’s eyes. PTC is coming for the paved roads before long too, because drivers are far more irresponsible than engineers.

[WhartonAndNorthern]

This crash resistance thing is crazy. Let's try an exercise:

1. Airbus should make their planes far stronger so when they crash out of the sky at twice the legal operating speed, the passengers survive.
2. Ford should make Explorers much stronger so when they are driven off a bridge at twice the speed limit, the passengers survive.

See how idiotic this looks? Nobody would ever suggest this.

You and I are largely in agreement and have had discussions on this in the past. The key FRA test is that the car should not deform when a load of 800,000 lbs is applied to the ends*. Certainly some standards of buff strength are required to withstand rigors of switching and interchange (if you're delivering your cars by freight rail). This standard** doesn't cover what happens when a car is hit from the side or a car is dropped off an overpass (the passengers will probably hit the ceiling).

This video is from the FRA's own youtube channel: https://www.youtube.com/watch?v=ZxaARFvCzpE" onclick="window.open(this.href);return false; Look what happens when an FRA car hits. It mirrors what happened at Chatsworth. For years, we were told: European equipment can't run on US rails because US freight trains are much larger than European freight trains, and the FRA's nightmare scenario is a passenger-freight head-on collision. Chatsworth was the nightmare scenario and the passenger loco telescoped into the first coach.

*= actually, it's at the "buff stops" 4 to 6 feet inside the car
**= other standards do cover side sill integrity I believe

[EuroStar]

And as for PTC on automobiles? Yeah right. Right now Tesla is very firm that a driver must monitor autopilot, not the other way around.

Yet every other car has an iPhone or Galaxy seemingly glued to the driver’s eyes. PTC is coming for the paved roads before long too, because drivers are far more irresponsible than engineers.

Keep dreaming...Automotive crashes have been with us for many many decades now and the NTSB has not issued calls for PTC for any rubber wheeled vehicles. Buses would be the obvious candidate as they carry many people as half a train car, but nothing. Nada. Zero. Zip. Bus crashes kill at least as many people an year as passenger trains in the US, but they lack the spectacle and the uniqueness of it. There are even crashes where all passengers of the bus have perished (for example, after falling in a ravine), but in bus and car crashes the victims are "anonymous". They do not get shown on national TV because there are so many of them. There are more hazardous materials carrying trucks than you think, but has anyone asked for PTC for them? No. All calls tend to be for more training and more certification of the drivers. Regardless of whether the rail industry likes it or not, its "failure" rate is low enough that when a crash happens it is spectacular and ends up getting a lot of coverage on national TV generating more public pressure to reduce the "failure" rate even further. The same is true for the airline industry (just look at what happened to the 737 Max). This is not true for the car/bus/truck industry. Autonomous cars are coming, but not because of the safety (even though it is frequently touted as an important feature), but for convenience so that people can get more leisure time reading, sleeping or whatever instead of driving.

[justalurker66]

This crash resistance thing is crazy. Let's try an exercise:

Reductio ad absurdum?

Perhaps we should argue the alternative. Seat belts, air bags, crush zones, roll cages, safety glass - every crash improvement that has been made to automobiles in your lifetime should be removed. After all, none of them are needed if no one crashes their cars. One could also argue that having such features leads to complacency of drivers. They feel safe even if they drive reckless. Just as some fear that PTC will lead to complacency in the train cab.

A vehicle should be able to withstand any reasonable accident. It is reasonable to expect a train capable of traveling 110 MPH in normal service to crash at 110 MPH. When (not if) it does crash, the design should be such to lessen the impact of the wreck - not make it worse.

If you prefer to focus on crash prevention then design a system that prevents the train from operating at 110 MPH at places where it is unsafe. Or do both ... design the train to be wrecked and to prevent a wreck.

[Tadman]

Nah.

Seatbelts and airbags solved a serious problem. In 1970, you had about 50,000 deaths per year in autos. Today you have about 30,000. That delta is largely because of such safety devices.

We also pretend that 30,000 number is no big deal. Why isn't more being done about it. What's up with $15b on PTC for 10 fatalities a year, but we have a coke and a smile for 30,000 in autos? See how crazy this is?

Look at the same time frame for on-train deaths. Have there ever been over 100/year? Most years there are less than 10, many years there are zero. What problem are we solving with either PTC or crash resistance?

Back to an assertion I made about a year ago: There have been less on-train deaths this year than grade crossing fatalities this month. There have been less on-train deaths in the history of Amtrak than there have been bear attacks resulting in deaths since 1971.

[Vincent]

It's true that very few people die every year in rail accidents, but that's only because passenger rail is so rarely used by travelers when compared to autos and airplanes. Statistics show that air travel is still the safest means of getting from place to place. But in the 1950s and 1960s air travel was perceived by most people as a dangerous way of getting around. Once the airlines and the government focused on preventing accidents, people's perception of air travel changed and air travel became the dominant and preferred choice of long distance travelers.

Air travel isn't the most efficient and sustainable method of transporting large numbers of people on short and medium distance journeys, however. Passenger trains would be more efficient. Spending $15 billion on our current passenger rail system might look foolish, but if it allows the passenger rail system to expand and become more viable in more short and medium distance markets, it might be a wise investment in the long run.

[electricron]

Nah.
Look at the same time frame for on-train deaths. Have there ever been over 100/year? Most years there are less than 10, many years there are zero. What problem are we solving with either PTC or crash resistance?

PTC was implemented on freight lines too, where there are no fare paying passengers. Why? There is more to PTC than just saving the lives of passengers and crews on the trains. There could be thousands of citizens in the nearby neighborhood that could be injured by two freight trains crashing head on into each other - depending upon the freight being carried on the train. It's more than just about deaths, it is also about injuries.

And I believe you will also see data showing highways cause far more injuries than railways. As long as we have a Congress that passes laws based on "feelings", on what is technically possible instead of what is practical, we'll continue to see over-stepping laws and regulations. Welcome to the club!