Freight train derailment caught on live stream - Pennsylvania, USA

[DelW]

Yes. I've visited that area twice, and not seen any mid-train helpers, but rear (and sometimes front as well) manned helpers were commonplace on the heavier trains.

Anyone know if the train in question had helpers on the rear or not?

There was a drone video on YouTube (which I can't now find), taken soon after the event, which showed two engines on the rear. As far as I could tell, they looked like the usual manned helper set of two SD70ACEs back to back. I assume that they were the engines that removed everything downhill of the derailed cars quite early on in the clean up operations.

 

[ac6000cw]

There was a drone video on YouTube (which I can't now find), taken soon after the event, which showed two engines on the rear. As far as I could tell, they looked like the usual manned helper set of two SD70ACEs back to back. I assume that they were the engines that removed everything downhill of the derailed cars quite early on in the clean up operations.

This video maybe?:

The description on that video outlines what I also suspect caused the string-lining (unintentional emergency brake application in the rear part of the train, which stops quickly, with the front part still moving due to momentum and the lead locos still pulling hard until the brake application propagates all the way along):

In the afternoon of August 11, 2026, Norfolk Southern 33A departed Altoona, PA with a long cut of steel slabs on the head end of the train. While the train was organized relatively correctly, with all the weight at the front and empties on the rear, it was disproportionate. As it rounded the Horseshoe Curve, at approximately 3:23 PM, the emergency brakes applied, sending a shockwave through the train. The empty, rear half of the train stopped almost immediately, while the heavy steel kept rolling. This short chain reaction ended up string lining 12 cars through the apex of the curve.

I was trackside once (at the summit of CSX's ex-B&O Sandpatch Grade) when a train went into emergency braking - I can still recall the sound of it propagating along the train as a series of load 'Pssts' as each car vented the brake line - to help the application propagate as fast as possible - followed by the noise of the heavy braking and slack run-in, then everything went quiet. Unfortunately when it happened the train was traversing a crossover between the two main tracks, so it stopped blocking them both for over an hour... see my video (from 19 years ago), from the 2:12 point:

Re. the CTE stuff, this is an interesting thread on Trainorders.com (in UP-speak, note the use of SD40-2 equivalent powered axle counts e.g. a C44AC counts as 12 equivalent axles) - https://www.trainorders.com/discussion/read.php?1,1708016

(UP C44AC = GE AC4400CW, C45AC = ES44AC, not mentioned but I think my namesake would be a C60AC on UP)

 

[najaB]

I can still recall the sound of it propagating along the train as a series of load 'Pssts' as each car vented the brake line - to help the application propagate as fast as possible - followed by the noise of the heavy braking and slack run-in, then everything went quiet.

This delay could easily be mitigated by electronically activated brakes, but US railways have declined to move to this system as I suppose it's cheaper to pick up the occasional derailed train instead.

 

[Taunton]

This delay could easily be mitigated by electronically activated brakes, but US railways have declined to move to this system as I suppose it's cheaper to pick up the occasional derailed train instead.

This makes it sound like buffoon railway management can't be bothered with something "easily" achieved. In fact, with 2 million freight cars readily interchanged nationwide (and into Canada/Mexico) you would need to fit these electronics to all of them - that freight derailed at Horseshoe was not a single commodity unit train, but assembled from all types of car. Keeping that lot adequately maintained and reliably coupled is a considerable challenge.

To contrast, the early BR Southern Region built many 4-SUB EMUs with straight air brakes, immediately followed by many 4-EPB, same body style but with electro-pneumatic brakes, now wired between units. The new ongoing standard, a considerable advance, yet it was never worthwhile to convert the almost-new SUB units to this style.

 

[najaB]

This makes it sound like buffoon railway management can't be bothered with something "easily" achieved. In fact, with 2 million freight cars readily interchanged nationwide (and into Canada/Mexico) you would need to fit these electronics to all of them - that freight derailed at Horseshoe was not a single commodity unit train, but assembled from all types of car. Keeping that lot adequately maintained and reliably coupled is a considerable challenge.

Had they started twenty years ago a significant number of the fleet could already have been fitted. And I stand to be corrected, but I don't believe that every car in the consist would need to be fitted. There would still be a continuous air hose, it just means that the brake application doesn't have to make its way all the way down the train but could start from multiple places, non-fitted cars would continue to work as they currently do.

 

[DelW]

This video maybe?:

The description on that video outlines what I also suspect caused the string-lining (unintentional emergency brake application in the rear part of the train, which stops quickly, with the front part still moving due to momentum and the lead locos still pulling hard until the brake application propagates all the way along):

Yes, that looks like the one. It's quite blurry when zoomed in on the rear locos, but they definitely look like EMD style radiators, so I assumed it was a manned helper pair.

I agree on the likely sequence, the audio from the VRF cam suggests the air being dumped before any mechanical noises start.

 

[Gag Halfrunt]

Had they started twenty years ago a significant number of the fleet could already have been fitted. And I stand to be corrected, but I don't believe that every car in the consist would need to be fitted. There would still be a continuous air hose, it just means that the brake application doesn't have to make its way all the way down the train but could start from multiple places, non-fitted cars would continue to work as they currently do.

You would need to bridge the gaps created by non-equipped cars, because they will not have the trainline cable that carries the braking commands.

Electronically controlled pneumatic brakes - Wikipedia

en.wikipedia.org

 

[AdamWW]

You would need to bridge the gaps created by non-equipped cars, because they will not have the trainline cable that carries the braking commands.

Electronically controlled pneumatic brakes - Wikipedia

en.wikipedia.org

Presumably one could design a system in which an equipped car detecting an air brake application then transmits an electrical braking signal to other connected cars which are so equipped, reducing overall propagation time.

 

[najaB]

You would need to bridge the gaps created by non-equipped cars, because they will not have the trainline cable that carries the braking commands.

From the article you linked:

When first developed, electric control ECP brakes needed a number of wires along the train to control solenoids on each wagon to release the brakes, and were not considered economic for freight. This has changed with the introduction of electronic controls, allowing data to be transmitted by two-conductor wire or radio from the locomotive to a microprocessor on each car, where locally powered valves hold the desired pressure in each brake cylinder

Since there's still a continuous brake pipe, the system would fail safe since the brakes would still apply in the 'traditional' way as the pressure drops along the train.

 

[AdamWW]

From the article you linked:

Interesting. I was wondering about radio control.

 

[ac6000cw]

Yes, that looks like the one. It's quite blurry when zoomed in on the rear locos, but they definitely look like EMD style radiators, so I assumed it was a manned helper pair.

They're definitely both EMDs - I think the one nearest the train is an SD70ACU (rebuilds of ex-UP SD9043MACs, 80ft long, with three radiator fans) - and the other maybe an SD70ACe (74ft long, with two radiator fans). Enlarged screen captures from 'Rudy's Rail Adventures' YT video attached.

== Doublepost prevention - post automatically merged: 16 Aug 2026 ==

This delay could easily be mitigated by electronically activated brakes, but US railways have declined to move to this system as I suppose it's cheaper to pick up the occasional derailed train instead.

Had they started twenty years ago a significant number of the fleet could already have been fitted. And I stand to be corrected, but I don't believe that every car in the consist would need to be fitted. There would still be a continuous air hose, it just means that the brake application doesn't have to make its way all the way down the train but could start from multiple places, non-fitted cars would continue to work as they currently do.

I think the best chance there's been to roll out ECP braking on a large scale was in the aftermath of the 2013 Lac-Mégantic disaster in Canada (and some other oil-train derailments in the US). IIRC, the US federal safety regulators proposed making ECP braking mandatory on crude oil trains, but after much 'discussion' that proposal was dropped, I think in part because the railroads were in the midst of spending billions on implementing PTC at the time.

 

[zwk500]

I think the best chance there's been to roll out ECP braking on a large scale was in the aftermath of the 2013 Lac-Mégantic disaster in Canada (and some other oil-train derailments in the US). IIRC, the US federal safety regulators proposed making ECP braking mandatory on crude oil trains, but after much 'discussion' that proposal was dropped, I think in part because the railroads were in the midst of spending billions on implementing PTC at the time.

According to this DOT Factsheet: https://www.phmsa.dot.gov/sites/phm...6536/ecp-brakes-ria-fact-sheet-dec-4-2017.pdf, ECP brakes are required on trains transporting flammable liquids. It estimated combined costs of fitting the brakes to be between $375-491m, for benefits of $131-197m, interestingly.

There's also this DOT Slide deck from 2006: https://r2.ieee.org/dcnova-vts/wp-content/uploads/sites/30/2020/01/060817_ECP_Brakes.pdf

Slide 12 is perhaps the most interesting one:

Why hasn’t this been done already?

• ECP brakes are a major capital investment (on the order of $6 billion for all locomotives and cars.
• The majority of costs will fall on car owners (most cars are privately owed by shippers or leasing companies); however, the majority of benefits will flow to the railroad
• Moving from conventional to ECP brakes will be logistically difficult, and small railroads will face significant costs downstrea
• All North American freight railroads will eventually need to convert.

This slide deck estimated costs of $40k per loco and $4k per car, giving a 3-year payback time for a unit train. Compared to the later factsheet, the ratio of costs and benefits seems to have fallen off somewhat!

 

[ac6000cw]

According to this DOT Factsheet: https://www.phmsa.dot.gov/sites/phm...6536/ecp-brakes-ria-fact-sheet-dec-4-2017.pdf, ECP brakes are required on trains transporting flammable liquids. It estimated combined costs of fitting the brakes to be between $375-491m, for benefits of $131-197m, interestingly.

According to - https://www.phmsa.dot.gov/news/phms...ndate-after-ria-finds-costs-outweigh-benefits and https://www.federalregister.gov/doc...olled-pneumatic-brake-system-requirements-for - that rule was repealed/rescinded before it became mandatory.

I wasn't aware when I wrote post #41 of the 2015 'Fixing America’s Surface Transportation (FAST) Act' which introduced the ECP braking mandate for 'unit trains transporting flammable liquid(s)' - thanks for the info .

Personally, I agree with najaB's comment above - if the railroads had started fitting ECP braking to unit trains (e.g. coal, grain and oil) 20 years ago, which generally run in a merry-go-round fashion with relatively captive equipment, it would slowly create a sizeable pool of ECP-fitted equipment and lots of real-world experience with it (costs, operational and maintenance). A bit like the situation, decades ago, when dynamic braking on new locos was an extra-cost option, but now it's standard equipment, resulting in dynamic braking being the norm for many braking purposes, and having sufficient dynamic braking available often being a requirement for trains descending long, steep gradients. (The payback of course was the ability to operate longer, heavier trains plus savings on friction brake maintenance, even more so after radio-controlled locos made it possible to distribute the braking forces along the train).

 

[AdamWW]

Safety comes at a cost, which I suspect the US is well aware of and accepts a certain amount of accidents in exchange for a certain level of freight charges - a bit like we do for road transport.

I've recently been watching Youtube videos of US freight trains.

It's very clear that they take a different approach towards safety compared to the UK, e.g. staff riding on the outside of freight cars, climbing onto locomotives on the move and in one case jogging over the ballast alongside a train ready to pull the uncoupling lever to let a car roll into position while the rest of the train stops.

 

[DelW]

I believe it's mandatory when propelling to have a crewman standing on the leading wagon, in radio contact with the loco driver (who may be a couple of miles away if it's a long train reversing). They often seem to be just standing on steps or on part of the wagon frame, with apparently no harness or other safety equipment. It does look highly dangerous to British eyes.

 

[ac6000cw]

They often seem to be just standing on steps or on part of the wagon frame, with apparently no harness or other safety equipment. It does look highly dangerous to British eyes.

In some places where they have to regularly propel stuff a long way, they put a 'shoving platform' on the rear (usually an ex-caboose) for the conductor/brakeman to ride on. I think I saw one in use in Florida once (on a phosphate train).

It's very clear that they take a different approach towards safety compared to the UK, e.g. staff riding on the outside of freight cars, climbing onto locomotives on the move and in one case jogging over the ballast alongside a train ready to pull the uncoupling lever to let a car roll into position while the rest of the train stops.

AFAIK, on most railroads climbing onto moving equipment is banned these days (which is different from saying it doesn't happen - old habits die hard).

 

[AdamWW]

AFAIK, on most railroads climbing onto moving equipment is banned these days (which is different from saying it doesn't happen - old habits die hard).

If crews are comfortable doing it when they know they are being filmed by a YouTuber, I imagine it can't be all that much frowned upon.

 

[najaB]

If crews are comfortable doing it when they know they are being filmed by a YouTuber, I imagine it can't be all that much frowned upon.

To be fair to them, very few YouTubers wear the uniform so the person with a camera/phone out could just be a regular railfan.

 

[zwk500]

To be fair to them, very few YouTubers wear the uniform so the person with a camera/phone out could just be a regular railfan.

A regular railfan is just as likely to put any video they film up on youtube as a professional youtuber. Photographs less, but nowadays with phones who can tell whats still photography and what's filming, especially from a distance.

 

[ac6000cw]

Photographs less, but nowadays with phones who can tell whats still photography and what's filming, especially from a distance.

Not just phones - for 15 years or so, I've recorded all my (amateur) video using various mirrorless hybrid stills/video cameras, which look the same as a digital 'stills' cameras.

Personally, if I'd happened to have filmed someone doing something they shouldn't, I certainly wouldn't make it public on YouTube.

(On one trip to the USA, I did have someone come up and ask if was a private detective, I assume because I was sitting in my car pointing a camera out of the window - at a train, of course )

If crews are comfortable doing it when they know they are being filmed by a YouTuber, I imagine it can't be all that much frowned upon.

Just out of interest, which YouTube video is it?

 

[Ailsa_Craig]

Yes. I've visited that area twice, and not seen any mid-train helpers, but rear (and sometimes front as well) manned helpers were commonplace on the heavier trains.

Anyone know if the train in question had helpers on the rear or not?


They don't ignore it - there's rules about placement of empty cars in train consists, particularly for high centre-of-gravity cars like those empty centre-beam lumber (cut timber) cars. It's certainly not a new problem e.g. decades ago Santa Fe (ATSF) spent a lot of money re-engineering the line over the summit of Cajon Pass in California, mainly because they were having too many 'stringlining' incidents around the orginal sharp curve at the summit, as trains had got heavier.

For anyone not familiar with the Horseshoe Curve area, this some of my video from nineteen years ago, taken at the same place (the viewing area) as the Virtual Railfan video - the helper locos back then were pairs of EMD SD40-2/SD40E, nowadays AFAIK pairs of big GEs.

Not quite, after more or less dispensing with the former UP SD70ACUs they have set up some SD70ACE units for helper service

 

[Meole]

3rd derailment there in 26

 

[ac6000cw]

Not quite, after more or less dispensing with the former UP SD70ACUs they have set up some SD70ACE units for helper service

Thanks for the info. I'm reasonably confident that the derailed train in this thread had an SD70ACU as one of the helpers (see post #41), so they hadn't gone completely at that point.

3rd derailment there in 26

Were they similar types of derailments?

 

[Ailsa_Craig]

Thanks for the info. I'm reasonably confident that the derailed train in this thread had an SD70ACU as one of the helpers (see post #41), so they hadn't gone completely at that point.


Were they similar types of derailments?

I was at Cassandra just over a week ago and saw just one SD70ACU in service, the rest of the helpers were SD70ACE units. There was a line of SD70ACUs at Altoona, presumably in storage. Apparently the rebuild wasn't to as high a standard as their CPKC counterparts, NS sold off around 60 of the 100 or so units. A few years ago I encountered a storage line of 20 ACUs at Kenova WV, en route to Race land KY. Nothing much that goes there comes out alive

 

[ac6000cw]

I was at Cassandra just over a week ago and saw just one SD70ACU in service, the rest of the helpers were SD70ACE units.

Just got back from a trip to the same area - majority of helpers were SD70ACe, but I did see at least three different SD70ACUs on helper duty. Highlight was probably two helper sets literally side by side on two trains climbing the west slope - the noise of 4 x SD70ACe in Run 8 was something else...