Rollover danger on N/W curve elevated line

[NE933]

In the sharp turn in Long Island City, between Queensboro Plaza and 39th Ave., N/W trains enter slow most of the time, but then speed up while the last few cars are still on it. This morning it happened, and I was in the last car. It swayed back and forth so much that I prayed it wouldn't derail, tip over and land on the street below. I imagine that if the last car did capsize it will pull much, if not all, the other cars with it. Such an accident could kill and / or maim hundreds. Has anyone else noticed and given much attention?

[Triaxle]

Chances of that are about zero. That line has been in service for about 100 years, with higher speeds on that curve than today. The timers do force too low a speed, and so when the train leaves the timer's block it accelerates after braking. This seems to be common, the MTA would rather grind the brakes then accelerate thousands of times per day so that a in the case where a drunk operator, complicit conductor and cover-for-mah-bro supervisors combine as in the Union Square wreck, the train will be moving too slowly for it to matter.
About half the weight of the subway car is in the truck assemblies, so the center of gravity in much lower than you'd think. That said, india gauge would be more stable. BART used that so that their trains could (theoretically) run over the Golden Gate Bridge.

You probably felt that swaying as the train was first leaning to the inside of the turn from going too slowly, then rocked on its springs as the last car cleared at normal speed, relieving the inward tilt but also creating a rocking motion.

[GirlOnTheTrain]

Ever notice those extra rails on the elevated structure? In addition to the timers slowing things down, like mentioned above, those are also there to prevent a scenario like you described. If the train derails, it's less likely to fall off the structure.

[jhdeasy]

In the sharp turn in Long Island City, between Queensboro Plaza and 39th Ave., ...... Has anyone else noticed and given much attention?

As a boy living in Astoria in the 1954 thru 1976 era, who rode the BMT Astoria line on trips to/from "the city" (Manhattan), I once had similar thoughts which made me nervous, but I eventually learned enough about railroading that I was not worried about such an accident. I think the two tracks on the curve have a certain amount of super-elevation, where the outside rail of the curve is somewhat higher than the inside rail of the curve, which accounts for the leaning on the curve. I believe the engineering term may be "cant deficiency."

I also remember a friendly Conductor who invited my attention to the dis-used BMT trackways curving between the dis-used northern (original BMT) half of the Queensboro Plaza elevated station and the Astoria line. They remained in place until about 1963, when most (but not all) of that structure and the flying junctions east of the station were scrapped. I recall noticing they had a gentler curve and they were at a lower elevation than the BMT train I was currently riding on the high curve of the former IRT track. They looked to be a "friendlier" route to anyone who was nervous about the train leaning on a sharp high curve.

http://www.nycsubway.org/perl/show?75913
http://www.nycsubway.org/perl/show?113424

[Head-end View]

Triaxle, re: your comment about BART's gauge and running over the G.G. Bridge, what does the track gauge have to do with trains being able to run over a certain bridge or not?

[ExCon90]

There was a theory at the time that a wider gauge would render the trains more resistant to being blown sideways in high winds, and the Golden Gate Bridge would be the right place to find them. Whether that much additional gauge width would make any difference I don't know. There were some positions taken by the planners that seem a little eccentric, like not having collision posts on the cars because, being computer operated, the system wouldn't have any collisions.

[Head-end View]

Yeah, such over-reliance on computers turned out to be a little over-optimistic, huh? Worked real well on WMATA a few years back........ I don't recall, have there been any collisions on BART? I only remember the bad fire they had in the Transbay Tube in 1979. Say what people will about NYC Subways signal system with trip arms being antiquated but it's a simple, effective technology that's worked pretty well for a hundred years.

[ExCon90]

I'm not sure about collisions on BART, but not long after the line to Richmond started, a train overshot the bumping posts and protruded over the street below, recalling a couple of famous examples at LAUPT and Gare Montparnasse in Paris. I think it turned out some 98-cent part failed in service.

[Head-end View]

Sorry to be off-topic; I just can't stop shaking my head sadly at the ignorance of some BART planner who didn't think collision posts were needed because on a system operated entirely by computers, that "there won't be any collisions". That has to be one of the most ignorant statements ever made in the history of railroading. And I can think of two families of deceased WMATA train operators who would have no problem disproving that theory.

[R36 Combine Coach]

The last time an elevated train derailed off the structure and onto a street was probably the CTA wreck on 2/4/77. And the entire train did not derail, only two or three cars.

[MACTRAXX]

R36:

See: http://www.chicago-l.org/articles/1977crash.html" onclick="window.open(this.href);return false;

The February 4, 1977 crash occurred after a train rear-ended one ahead at the 90 degree curve
on the northeast corner of the Loop L structure at the street corner of Lake and Wabash Streets.

Four cars derailed and fell off the L structure with two landing on their sides in the street - a third
car fell to an angle against a support column and the fourth hung over the street coupled to the
four cars behind in what was an 8 car Lake-Dan Ryan train.

The CTA would later place girders on the outsides of curves such as this one as added precaution
to keep an accident of this type from occurring again.

This was a serious accident which still is the worst in the history of the CTA.

Thankfully it has been many years since anything remotely this bad has happened on a New York
City elevated structure. I do feel that the millions of passengers safely carried over elevated lines
speaks for itself.

MACTRAXX

[Head-end View]

Interesting that the collision was caused by a train operator "keying-by" a stop-signal at restricted speed. Same cause of a couple of minor NYC collisions in the early 1970's which caused the NYC Transit Authority to ban the practice of "keying-by".

[DaveBarraza]

In the sharp turn in Long Island City, between Queensboro Plaza and 39th Ave., N/W trains enter slow most of the time, but then speed up while the last few cars are still on it. This morning it happened, and I was in the last car. It swayed back and forth so much that I prayed it wouldn't derail, tip over and land on the street below. I imagine that if the last car did capsize it will pull much, if not all, the other cars with it. Such an accident could kill and / or maim hundreds. Has anyone else noticed and given much attention?

Yes. It was given much attention by NYCTA and their contractor Parsons Brinkerhoff immediately after the Union Square wreck. A comprehensive study was performed of safe operating speeds and safe braking distances system wide.

Any scenario which would have placed a train in excess of "unbalancing speed" on a given curve would have been addressed in the first phase of the signal system enhancements installed as a result of the study. -by the end of the 1990's.

Note that "unbalancing speed" is considerably higher than "comfort speed" - exceeding "comfort speed" results in passengers potentially losing their balance or otherwise having to hold onto the bars. The faster "unbalancing speed" places the CG of the train outside the outside rail resulting in a derailment.

You're OK. You don't have to start taking the bus.