Deadly Cargo Still Rides the Rails | Hacker News Reader

Deadly Cargo Still Rides the Rails

nytimes.com

75 pointspseudolus71 comments

And yet when a pipeline was proposed it was quickly shot down [1].

[1] https://en.wikipedia.org/wiki/Energy_East

ADDENDUM: The amount of rail accidents in Canada is shockingly high. Multiple per day [2].

[2] http://www.bst-tsb.gc.ca/eng/stats/rail/2018/sser-ssro-2018....

And yet there are lots of excellent reasons why. There's nothing safe about pipelines. In Canada alone there over a hundred pipeline leaks or explosions every year [1,2].

[1] https://equiterre.org/en/news/new-report-oil-pipeline-safety...

[2] https://apps2.neb-one.gc.ca/pipeline-incidents/

How many do you think is the appropriate number? Surely not zero, since this can only be approached in the limit of infinite expenditures. So what principle would you use to tell?

> Surely not zero, since this can only be approached in the limit of infinite expenditures.

What proof do you have of the lower bound here?

There's no physical device with zero failures over an unbounded time. Whether it's human errors or material degradation, things will fail at some point. Materials also have production faults which can make them break on the first usage. Everything has a non-zero chance of failing.

The "infinite expenditures" part comes in because you can introduce redundancies and fail-safes, but those introduce complexity / more work / more materials, which have extra costs. Then you need to spend time/money making sure the whole system works together / testing failure modes. But that may also fail, etc.

There's at least one (idealistic) method to reduce the number of pipelines leaks and explosions to zero. This will also eliminate the risk of another Lac Megantic disaster, and reduce emissions to zero! It's just not an immediate-term proposition.

This issue is too theoretical to contribute to deciding what should be acceptable.

I will open the bidding by suggesting that Canada's frequency of dangerous pipeline failures should not exceed its frequency of airliner crashes.

I assumed the parent referred to the zero lower bound. A realistic lower bound discussion is more complex/interesting of course.

I meant to segue into a realistic lower bound, not a zero bound, though I definitely wasn't clear about it.

> There's nothing safe about pipelines

They're safer than trains. The problem is new pipelines usually enable new extraction. If we could decouple those two, i.e. shift transport of existing extraction from trains to pipelines without creating lots of new extraction, that would be a net benefit to all.

Pipelines are different than trains.

- They can leak much more in an unit time than trains... hundreds of thousands of gallons, and regulatory enforcement on leaks in the US is very weak because of corporate capture.

- They are built usually by legally-stealing land (easements, eminent domain) from people along the way.

- If they leak on people's land, they don't have much recourse.

- All the safety tech and building codes in the world don't matter if the operators don't really care about leaks because the risks of fines and lawsuits is so comparatively small.

There are hundreds of oil leaks per year, and they can be absolutely huge, but because they're usually in the middle of "nowhere," they rarely make the mainstream news.

False dichotomy.

Let’s limit extraction instead and invest instead into clean infrastructure like power lines.

> Let’s limit extraction instead

Just blocking pipelines without restricting extraction virtually guarantees these outcomes. Restricting extraction while shifting transport to pipelines looks like a better solution than just blocking pipelines.

Bad publicity for fossil fuels is good news for the climate. Any new infrastructure for fossil fuels creates investors with an interest in keeping extraction going.

Incentives (carrots) and enforcement (sticks) are the only things most big polluters will respect. And that will require strong political will, leadership and outreach relationships to accomplish, one region at a time.

Let's render extraction unnecessary by reducing demand.

Meanwhile, use transport that minimizes environmental damage.

>Meanwhile, use transport that minimizes environmental damage.

You're gonna be hard pressed to find a means of transport that's more environmentally friendly than steel wheels on steel rails. Friction is a bitch.

To be more accurate you would say "restrict consumption" because that is what you are doing here. How do you plan to do that? If you did so by restricting extraction directly, as you seem to imply it would cause oil prices to jump dramatically. In fact, since oil is has extremely inelastic demand, this would be a massive windfall to the oil industry. The only other way to do it is rationing, which I'm sure would go over about as well as the massive price increase.

I'd say more like "redirect/rewire consumption". Oil demand is considered inelastic because we have minimal mass transit options in most North American cities. So, increase mass-transit and HVDC links combined with funded solar+wind sources and EV chargers to covert the fuel-demand into overall energy demand.

Also, the oil being consumed here is headed out to the global market. It's not being earmarked for local use. It's effectively fungible with global supply so the decrease here will simply have a small global price increase.

Actually, power lines aren't necessarily clean because they have significant losses, and would cause more emissions if they carried FF-generated electricity thousands of miles (losses as high as 60%). A better strategy is generating as close to consumption as possible, e.g., solar/wind/etc. right next to consumption to minimize losses.

Do you have a cite for the 60% loss? Perusing Google shows 8-15% average, so that's a hell of an outlier.

Gas engines on the other hand, have a maximum carnot efficiency of about 40% (that's 60%+ power loss there), which doesn't include extracting/transporting/storing as well as the kind of unplanned of losses this article talks about.

You can't eliminate pipe leaks, etc. But you can compartmentalize to minimize the damage a leak does, much like dividing a ship into watertight compartments.

The hydraulic system on a Boeing jetliner is compartmentalized so that a leak in one of the systems doesn't drain away all the hydraulic fluid in that system - only the actuators downstream of the leak get affected.

On the contrary, you can eliminate pipe leaks easily - just many areas have fairly lax integrity management requirements for operators, and those that don't only apply to new lines - very rarely does legislation cover preexisting pipelines.

New pipelines (especially ones considered DOT regulated in the US, overseen by PHMSA) typically are installed with automatic block valves and flow level leak detection systems. Those combined with a sensible internal and external corrosion control program can push the life of a pipeline without incident out past the useful life of the pipeline. But retrofitting these is expensive and most pipeline operators won't do it without regulatory or other political pressure.

> you can eliminate pipe leaks easily

Famous last words. A better design is to assume leaks occur, and design to minimize the damage. (This is how airliners are designed.)

> with automatic block valves

That's compartmentalization, not elimination.

Another method is to use two smaller pipes instead of one larger one - it reduces the scale of any leak.

  use two smaller pipes instead of one larger one - it reduces the scale of any leak.

... while doubling the exposure to failure in the first place.

This actually ends up increasing your likelihood of failure, as smaller parallel lines mean higher operational costs for smart pigs and other integrity assessment tools making internal issues harder to detect before failure, and increased risk for external failure due to cathodic protection system failures due to interference and shielding.

Of course it doubles the risk of failure, but each failure does half the damage.

Which sounds like a terrible trade-off, harm isn't linear here. Two leaks, each of 1000 barrels, will badly damage two different valleys. One leak of 2000 barrels would be better.

And cost isn't linear either, two half-capacity pipes will probably be almost twice the price. You could spend that on more steel for the large pipe instead.

Actually, I'm not so sure about doubling the risk. Failures may appear random, but they aren't. A failure may indicate a particular defect that can be corrected in the rest of the system before another one fails.

Er... I don't think you've completely taken into account what you're proposing. Due to the square-cube law, splitting a pipeline into two smaller pipelines means that in order to have the same total cross-sectional area you would need sqrt(2) times as much surface area. Not only is the surface area that is subject to corrosion increased, I would imagine the pipe wall thickness would probably decrease so corrosion would compromise the integrity of the pipeline even sooner. It's also double the number of welds that need to be made, double the length of pipeline to maintain, and as to the notion that it would do half as much damage, I doubt this claim. Why would it actually substantially decrease the damage to the environment? It's not like pipeline leaks completely sever the pipeline in half and just start inundating the surrounding area with a flood of oil. It's usually a crack or pinhole leak, so the smaller cross-sectional area of the pipe probably won't do much good in terms of decreasing the volume of oil leaking out.

With proper monitoring and a well maintained internal corrosion control and external threat control systems you can stretch out the timeline for non "acts of god" (ie natural disasters, terrorist attacks, etc) to greater than the life of the structure. The methods for internal and external control are well understood and practiced to varying degrees of effectiveness.

Elimination of the hazards is impractical, so minimization of risk exposure and frequency is the best that can practically be done.

All the hubbub about stopping new pipeline construction means that more incident prone methods are used - be it trucking, rail transportation, or use of 50+ year old pipelines. The new lines are better engineered, understood, and have more stringent operational controls in place. I've worked on pipelines from the 1940's that are still safely operated, and lines from the 2010's that have already self excavated. These things can be controlled with the right engineering, organizational, and governmental oversight.

Compartmentalizing risk as acts of god is a rather pointless exercise. If you can’t eliminate risk you need to be intellectually honest and say why that risk is acceptable. The Midwest for example gets large earthquakes in seemingly random locations, that’s a known issue and needs to be accounted for in a risk assessment.

As to other transport methods being less safe that also comes down to building safe systems. Car gas tanks for example are designed to survive crashes without spilling their contents. You can also build tanker trucks/trains to appropriate standards.

It seems you're not familiar with the history of US pipelines. It doesn't matter what theoretical technologies or regulations exist, because oil extraction industries get away with leaking hundreds of thousands of gallons of oil with little or no punishment or recompense, including on the farmlands they steal from people to run their pipelines across.

See the thousands of dots? Each is a spill.

https://projects.propublica.org/pipelines/

https://www.theguardian.com/us-news/2017/may/10/dakota-acces...

https://theintercept.com/2018/01/09/dakota-access-pipeline-l...

https://www.cnn.com/2017/11/16/us/keystone-pipeline-leak/ind...

https://en.wikipedia.org/wiki/List_of_pipeline_accidents_in_...

Out of curiosity, did you intentionally choose examples reminiscent of the Titanic and 737MAX?

First off, my first real job was designing the flight critical stab trim system for the 757. I have an enduring interest in airplane flight control systems, and follow airliner crashes and the resulting fixes. The hydraulic changes preventing leak propagation resulted from the Sioux City crash where the compressor disk in the tail engine came off and sawed through each of the triply redundant hydraulic lines, resulting in total loss of the hydraulics. The 737MAX troubles have nothing to with the hydraulic system.

The Titanic disaster did popularize the notion of watertight compartments, but these are well known and used in warships. Warships are very hard to sink because of it. The Bismarck is suspected to have sunk only because the captain flooded it to prevent its capture.

Submarines use compartmentalization to minimize flooding risk, and surely the space station must, too.

Pipes: bad.

Trains: worse.

Clean(er) energy generation and chemical manufacturing: mucho mucho better.

One of the largest peacetime evacuations (>200,000 people) in North America occurred in 1979 when a train carrying toxic materials derailed in Mississauga, Ontario (about 30km from Toronto) [0]. Yet, even with the benefit of the knowledge of that barely averted disaster, trains were and are still barreling through and in proximity of cities.

[0] https://en.wikipedia.org/wiki/1979_Mississauga_train_derailm...

In nearly every case the city is there only because the railroad was there first. Obviously reasonable and pertinent safety measures should be taken ("Safety First", etc) but I think it's pretty unreasonable to ask the railroads to move their 100+-year-old routes because people built a city around it.

And even for the cases of older cities that predate rail, what exactly would be the point of a railway that doesn't go anywhere?

Seems like a world seen through north american suburbs... "Why did you have to build the rail so close to where people live?!" Question answers itself.

I don't see why it matters that the railroad came first. It's just a piece of infrastructure, not a native people. This happens all the time where a main road now goes through a town that has developed. A bypass is built so it doesn't have to go through the town anymore.

The Transportation Board of Canadian has a great video w/ animations of the incident:

https://www.youtube.com/watch?v=wVMNspPc8Zc

I assumed the brakes were applied unless air pressure was available. Same as truck / trailer brakes. In the article it references air pressure is required for the brakes to be applied - is there a reason it can't be the other way around?

hand brakes were applied, but only to the first 5 lead cars only. the engineer also used air brake, but it lost its pneumatic pressure after the engine got shut off.

"Though exhausted after the journey, Mr. Harding took the laborious step of setting the mechanical hand brakes — a train’s version of an automobile parking brake — on the five lead locomotives, an equipment car and an empty boxcar.

That was the first big mistake. Investigators calculated that Mr. Harding should have also secured the hand brakes on 18 to 26 of the tank cars before retiring to his hotel."

New trains are like that. https://en.wikipedia.org/wiki/Railway_air_brake

If the technology has been in widespread use since 1869, why would locomotives presumably made some time in the latter half of the 20th century not incorporate it?

https://en.wikipedia.org/wiki/Railway_air_brake#Limitations

This was George Westinghouse's big invention from the 1860s: https://en.wikipedia.org/wiki/Railway_air_brake

It'd be great if you could use a spring to apply the brakes and an air cylinder to hold the brakes open. But the force needed to apply brakes is too much for springs, so it uses an air cylinder to apply the force. There's a complicated system where under normal conditions, pressure in the hose running the length of the train is fed into a reservoir in each car. When the pressure in the line drops (either a disconnect or the engineer hitting the brakes), a valve connects the brake cylinders to the reservoir and the air pressure applies the brakes.

So yes, if you leave some isolated rail cars sitting there with the air brakes applied, the pressure will slowly leak away over hours or days and the car will start rolling.

Apparently before air brakes dudes would run around on top of the moving train (even in rain/snow) manually cranking the brake on each car whenever the train needed to stop.

https://en.wikipedia.org/wiki/Brakeman

One of the advantages of the british vacuum brake system was that it passively failed into a 'brakes on' state because of atmospheric air pressure.

However, this was about the ONLY major advantage it had, and decades of pro-westinghouse propaganda against it blew the significance of all of the disadvantages* up to be greater than the advantage of fail-safe, and we switched to the westinghouse system.

* One of the biggest claimed disadvantages is the fact that vacuum brakes are less effective as you increase altitude, this is a HUGE disadvantage to operating vacuum brakes in the US where a large amount of rail traffic is at altitude, but in the UK where our highest peaks are 3000' and we don't generally run railways to the peaks of those hills, it's almost a non-issue.

Air pressure is required for the brakes to be applied, insofar as if you bleed all the air off the cars their brakes will release. However, air is stored in reservoirs on the cars, and once charged when the air pressure in the brake pipe that goes down the train is reduced below the reservoir pressure the brakes apply (by using air stored in the local reservoirs). As such, the brake pipe is simultaneously an air charging system and a signaling system. It's also fail-safe for normal operation since a train that breaks in two will come to a stop.

Passively-applied brakes would break a ton of standard railroad operations which depend on cars rolling free, such as in "hump" classification yards which use in-track retarders to control speed, and in other situations such as "kicking" cars into tied-down cuts.

In Philadelphia I used to take walks down by some train tracks and I'd often be only a few feet away from stopped or slow moving trains. I always wondered what sort of security is there on trains? What would stop someone from attaching some sort of dangerous cargo to the train and just letting it roll into the next big city?

Based on the amount of graffiti I see on trains ... I'd guess nothing.

Nothing other than the agents at your door, I guess.

Attaching cargo as in adding another car or tanker? That would be pretty difficult - those movements are planned. Sticking a bomb on a tanker, on the other hand, would probably be much easier, although you'd have to know beforehand what tankers are full, and I suspect that's more difficult.

But there's always going to be a level of insecurity around trains. A train yard can't be fully fenced off for obvious reasons. I know here in Oslo there was a teenager killed last year after he and a couple of others snuck into a yard one night and went into a maintenance tunnel and touched a high voltage line. They've rebuilt the fence they got in through, and made it higher, but someone determined to get in there will just go further to where the tracks cross a road.

It is not so simple to "attach" cargo to a train. The train crew knows how many cars the train consists of and there are periodic "defect detectors" that among other things report the number of axles to the crew. The crew also has a manifest, and eventually they would notice if something was out of place. Individual train cars also have radio tags that are used to monitor their status as they move:

https://en.wikipedia.org/wiki/Automatic_equipment_identifica...

Thanks for sharing this.

I live right next to a rail yard in Atlanta, partially because of the novelty. I was already worried about pollution, but now I have this scenario to keep me up at night (as unlikely as it may be).

Is there a place to look up what hazardous materials a train line can (or does) carry? The line is CSX, if that matters.

If you have a good enough view, you should get some idea from the diamond labels.

I'd be curious to know the causes of the other recent runaway train incidents. They demurred on whether adding a second operator to each train would help, but I feel like you could answer that question by looking at those recent cases, no?

Having a second operator in this case would've meant twice as many hands to operate hand brakes, and ostensibly less operator fatigue to begin with.

https://en.m.wikipedia.org/wiki/CSX_8888_incident Is the only other one I know about; the engineer got off to change a turnout and was unable to reboard as the train accelerated away.

However, the engineer "inadvertently failed to complete the selection process", meaning that the train's engine was set to accelerate, not to brake.

The fact that braking could be confused with accelerating, especially on a vehicle which can effectively only travel in two directions, sounds more like a big UI failure to me.

There's a film about CSX 8888:

https://en.wikipedia.org/wiki/Unstoppable_(2010_film)

And two people to ensure procedure happens.

You know the joke: Go fishing with a Baptist, and they’ll drink all your beer.

Go fishing with two Baptists, and they’ll watch you drink your own beer.

And now Quebec also approved transporting aircraft fuel thru the downtown of Montreal.

https://mtlcityweblog.com/2019/07/15/trains-carrying-kerosen...

People just don’t learn ...

...or the alternatives were less safe (trucks) or too expensive (realigning the tracks).

I would be interested to see a risk analysis calculating the cost of loss of life and property damage in the event of a worst case scenario for each proposed solution (trucks, realigning tracks, or drilling a pipeline).

After the San Bernardino train disaster a maintence tech for the Calnev pipeline told me the pipeline was out of service for 10-12 weeks. During that time tanker trucks were used to carry gasoline and diesel from Los Angels to Las Vegas. There were six major accidents involving tankers trucks.

https://en.wikipedia.org/wiki/San_Bernardino_train_disaster https://en.wikipedia.org/wiki/Calnev_Pipeline

Thought this was about Rust and Ruby

Lessons learned report:

https://youtu.be/larpYRCvZoU

Am I the only one that came here thinking the article is about Rust’s package manager & Ruby’s web framework?

I read these articles and I get depressed. Not about the 47 people, or the dangerous trains chugging their way around the country. I get depressed that in Baltimore, about this many people are killed every month. I get depressed that in Chicago someone is murdered every 17 hours. That 35,000 people die from cars every year, and handguns. But we should really stop those trains from causing an accident, because one time is too many.

Can't we work to stop all of these at once? Society is multithreaded.

No it isn't. Society is like AWS. It contains many organizations, which each contain many accounts, which contain many services, which contain many instances, any one of which may be multithreaded, or not. "Society" does not actively work to solve all problems at the same time, just like every single piece of computation inside AWS is not necessarily multithreaded.

Reply on news.ycombinator.com