Ten90 Solutions LLC

First.....Finally.  After two and one-half years NTSB has finally adopted and approved its report on grade crossing collision at Commerce St. on Metro-North Railroad's Harlem line. 

The accident was more than horrific, if there's such a thing as more than horrific.  The driver of an SUV, possibly oblivious to the flashing lights and descending gate of the grade crossing protection, got out of her vehicle to inspect the area where the gate had struck her vehicle, got back in her vehicle and drove directly into the path of train 659.  The resulting impact drove the SUV into the third rail, with the frame of the auto "guiding" the rail into the undercarriage of the  lead car and into the passenger area.  The SUV driver and 5 passengers on the train died.  

The grade crossing protection worked properly. The brake, signal, horn, systems on the train worked properly.  The locomotive engineer operated the train properly.  And... for what it's worth, this is not an accident that could have been prevented by the installation of PTC as required in 49CFR 236 Subpart I.  

In anticipation of Tuesday's discussion and approval of the investigaton and its findings, certain media outlets were reporting that the NTSB had determined that the specific configuration of the third rail on Metro-North, where the contact shoes "underrun" the third rail rather than maintain contact with the top of the third rail as is the case on the LIRR and the NYC subway contributed to the severity of the accident, allowing the third rail to "ride up" when struck by the car.  

The NTSB refuted that speculation directly during its review of the accident:  

(Acting chairman) Sumwalt: 

I read in the media yesterday, last night, that the overruning versus the underruning configuration of this rail may have contributed to this accident. Do you find any truth to that?

(Subject matter expert, 3rd rail) Ruben Payan:  

We did not find anything to do with that.  We have seen both kinds of overruning and underrunning rail.  The issue we found was with the gaps and the tapered rail at grade crossing.

And so what was the issue with the gap and the "tapered rail"-- end approach, or nose piece, or transition rail--that the SMEs found? According to the list of findings, the third rail system:

13....was not constructed to fail in a controlled manner or reak away when subjected to undersirable overloaded conditions such as those involved in this accident.  

14. The continued use of the Metro-North Railroad's current third rail system (which lacks controlled failure mechanisms) may increase the severity of railcar damage and serious injuries when accidents occur at or near grade crossings.

15. The presence of third rail systems at or near highway-railroad grade crossings on commuter railroads and rail transit properties could increase the severity of highway railroad grade crossings accidents. 

The findings are based on the fact that approximately 350 feet of  the third rail pierced the floor of the first car, entered the passenger area and caused the death of the passengers, not by electrocution, but by impact.  

Acting chairman Sumwalt characterized the end-approach (nose piece) of the third rail as a "spear."  The nose piece is a transition section of rail to guide the contact shoes smoothly back under the third rail when the train traverses a gap in the continuity of the power rail.  "Spear?" Let's look at some numbers. 

A single M7 EMU vehicle weighs approximately 127,000 pounds.  Empty.  For comparison's sake, an M1A1 Abrams tank weighs about 130,000 pounds.  Train 659 on the date of the accident consisted of 8 M7s.  With its normal passenger load, the train weighs more than 8 M1A1 tanks.

Top speed of the Abrams tank is 45 mph.  Top speed of an M7 is 100 mph.  The speed of train 659 at the point of collision with the SUV was 51 mph. The train with the emergency brakes applied, pushing the SUV, traveled some 665 feet from the point of impact.  

Now, does anyone with an ounce of experience in the physics of collision think, for a single second, that a) designing and installing a non-conducting nose-piece, approximately 6 feet in length that breaks away under impact from the following sections of contact rail, is going to mitigate the energy that 8 M7s or 10 (adjusting for the square of the higher M7 velocity)  M1A1 Abrams tanks exert on the following sections of third rail and prevent the following 340 feet from driving through the SUV and into the M7 car body?  

Does anyone seriously think that absent the nose piece the remaining power rail is not going to follow the path that it followed in this case?  

Much was made of a computer simulation that supposedly showed that the splice bars and the bolts used to attach the splice bars to connect the sections of power rail could undergo high degrees of plastic deformation, in both lateral and vertical planes without failing. Much was made of the fact that the third rail splice bars functioned to keep the rail in line and "feeding" directly into the M7 car body; that the splice bars did not fail when subjected to "overload" lateral or vertical forces.

Yet, at no point in the presentation did the NTSB experts quantify exactly what forces the splice bars did encounter  in either lateral or vertical directions during their simulation of the accident, nor how those forces compare to the forces that the splice bars and bolts must absorb during normal rail operations to safely provide continuous power transfer to the the trains current collecting contact shoes. 

We do NOT know what forces the third rail assembly experienced during the accident.  We do not know how those forces compare to the forces the assemblies must be able to absorb.  We do not know what forces would cause the splice bars and bolts to fail in a controlled manner and still provide for effective power transmission and safe train operations.  

The subject matter experts, in their exposition did not attempt to quantify the lateral or vertical forces experienced nor if in fact those forces were truly beyond what the third rail might experience under rail operating conditions.

Those are the critical quantities we must know if we are seriously going to consider a redesign of third rail; if the NTSB seriously expects railroads to comply with its recommendation that we "conduct a risk assessment for those highway-railroad grade crossings" "that have third rail systems at or near those grade crossings."

Railroads do not operate on even, clean, smooth surfaces, and if they did, the very movement of the trains would deform that geometry and create significant lateral and vertical forces in the track, and in the third rail which is attached to the track, and in the contact surface, and angle between the train's contact shoes and the third rail.  The rail must be designed to absorb all those forces, all those forces as they accumulate over time and still function. 

Are railroad supposed to design 3rd rail so up to 1000 feet, 12 passenger car lengths,  of 3rd rail on either side of a highway grade crossing will "fail controllably" at forces that may not exceed the  lateral and vertical forces transmitted during ordinary, non-emergency loading due to regular train operations?    

The NTSB speculates that the nose piece acted as a "spear," penetrating the undercarriage of the lead M7, yet....yet in its own exposition, NTSB acknowledges that the third rail penetrated the M7 in two (2) separate locations, so in at least one of those locations something other than the nose piece first pierced the undercarriage.

Here's something that isn't speculation:

Preventing  future accidents and future tragic deaths at grade crossings depends upon eliminating grade crossings; installing quarter-point gates; enforcing fines and penalties against vehicles and drivers violating crossing warnings by installing cameras (which the NTSB failed to recommend to anyone) at the crossing to capture license plate numbers of scofflaws.  

We can debate and perform  "risk assessments" forever.  We can design  "controlled failures" and the like, and that might keep some of us busy, and might save lives.  It might not.  Safe railroading, however,  demands more than "might"  or "might not;" better than "on the one hand, while on the other hand."  Safe railroading means we don't speculate.  Safe railroading means we know; we know the numbers, the quantities, the forces, so we can implement real solutions.

David Schanoes

July 26, 2017