Locomotive Cabs, and Crew Safety in Collisions

Paul,

The MK1500D is a road/switcher that we and HB&T purchased in 1996.

The first 25, 9600 thru 9624 went to PTRA and the last 10 went to HB&T.

When HB&T was dissolved into its parent companies, the BNSF ended up with them.

They used theirs as yard transfer locomotives; we use them as yard switchers and road power.

I like them; they kick well and will stop a heavy switch cut quite fast when we are kicking cars.

And we have had little if any real issue from the Cat power.

MK Rail is starting a planned 15 year rebuild program on them, shopping them here on property for major engine and electrical rebuilds two at a time.

And it would appear the body style is popular, the GenSet and Green Goat folks copied it, and EMD had MK built a small fleet of 1500 and 2000Hp units that were leased to UP.

Overall, quite a good showing, and as we have no plans to replace ours, I think they will work another 15 to 20 years here before any serious though to scrapping them happens.

Ed - thanks for the response.  Interesting details - with about 375 unit-years of experience/ usage on them altogether by you folks now, you ought to know their good and bad points well.  The visibility over that stubby low short hood with its gentle slope downwards ought to be about as good as from an end-cab switcher, I would think ?  I'll have to find one to watch someday - none anywheres near around here, as far as I know.  Link to a 2007 broadside photo of one by Steve Schmollinger:

http://i.pbase.com/g3/04/933904/2/120142739.H3GBvOLT.jpg 

And so back to the subject of this thread:  Do the MK 1500D's have collision posts ?  Can you get a photo of them as well ? 

Actually, for all locomotives, there should be 2 sets of collision posts - 1 in the short hood, and 1 at the far end of the long hood.  Maybe those would be easier to get a photo of - or not ?

Thanks again, Ed.

- Paul North.       

Paul,

I certainly do agree that an assessment of whether or not the posts failed depends on what they are intended to accomplish.  They may not have been intended to protect from collisions with railroad equipment.  I did read some material suggesting that collision posts were intended to protect from railroad equipment collisions as well as grade crossing collisions, but that was buried in an ocean of information, so I cannot conclude how it might apply to this case.  It would be interesting if a description of performance intent were found to supplement this structural specification that you posted:

     (4) The outside end of each locomotive shall be provided with two
main vertical members, one at each side of the diaphragm opening; each
main member shall have an ultimate shear value of not less than 300,000
pounds at a point even with the top of the underframe member to which it
is attached. The attachment of these members at bottom shall be
sufficient to develop their full shear value. If reinforcement is used
to provide the shear value, the reinforcement shall have full value for
a distance of 18 inches up from the underframe connection and then taper
to a point approximately 30 inches above the underframe connection. 

In reading the above specification, I am struck by its vagueness.   Basically, it requires 300,000 lbs. of shear strength where the posts connect to the main frame.  Yet, without stipulating the height of the post, it is meaningless for practical purposes.  The shear strength of that connection would be just the beginning of defining what the post is intended to accomplish.  According to that specification, the “post” could be a couple inches high as long as it, and its weld, met the shear strength requirement.   

The post’s failure to stand the pressure of the collision was not a failure of that shear specification.  Nothing sheared.  The post bent over from inadequate column strength.  Either the post bent, or it acted like a lever to bend the frame to which it was attached. 

The shear strength of the connection of the post to the frame almost seems beside the point.  The point is the amount of force the post can take at any point on it without yielding by bending for any reason.  The post can withstand the least amount of force at its very top, and withstand more as it is applied more toward the base.  So the logical specification would be the minimum horizontal force that the post can withstand within say 2” of the top without yielding in any way.  Anything that strikes the post has the potential to apply force to the very top tip, so that has to be the point at which the minimum performance strength is specified.

It almost seems to me that the specification is a rather flawed product of a lot of push pull between the locomotive manufacturer, the railroad industry, and a collection of regulating authorities.  So you end up with something that gives the illusion of solving a problem, but is relatively meaningless.  Then the locomotive builders execute their best interpretation of what they think is called for by the sketchy specification. 

I can see how the railroads would be reluctant to cross over into the issues of protecting locomotives from crashes other than grade crossing collisions.  That would be a wide-open field with a mighty big vehicle for safety system visionaries to play with.  The potential to add cost through mandated protections would be enormous.  And the railroads probably feel that following the rules is the best crash protection from colliding with railroad equipment.  But since they have no control over the matter of colliding with vehicles at grade crossings, they may welcome crash protection for that purpose.     

Bucyrus - Yeah - you 'get it' - but I thought you would, anyway.  

You and I could probably sit down and in 15 minutes draft a workable specification that would remove all that vagueness, achieve a practical order-of-magnitude improvement on the safety to be provided to the crew (without attempting to achieve 'perfect safety'), and still be affordable and within the weight limits for most freight locomotives.   

One minor point of difference: It seems to me that the 300,000 lb. shear strength requirement for each post is far larger than the load that could reasonably be expected to be developed or needed as a result of any collision with a highway vehicle - say, an 80,000 lb. GVW truck, even allowing for 'impact' - and as compared to the 65,000 lb. objects in the Appendix E crashworthiness specifications.  Therefore, I believe that was an early attempt - the regulation appears to date from 1980 - to provide crew protection against collision with other full-size railroad equipment. 

Also, permit me to offer a clarification to one part of what you posted in the last para: "And the railroads probably feel that following the [operating*] rules is the best crash protection from colliding with railroad equipment." 

*As distinguished from the rules for locomotive safety in construction and features, etc., that I posted above. 

One reason I suggested above that a better specification be developed and implemented by either the crew's union(s), the railroad industry (including locomotive manufacturers), and/ or a railroad's own Mechanical Dept. staff and officers, was to avoid the complications of involving the federal agency, for some of the same reasons that you state.  It's really not a 'public' safety issue - no one is more directly involved than the on-board crews, such as our good friends Ed, Jeff, and Zug here (not intending to leave anyone out) -  nor hugely complicated technically.  The improvement could start to be implemented with a railroad's next major purchase order for new locomotives, and/ or the next renegotiation of the nationwide contract with the operating crafts, if it really matters to either of them.  There's some precedent for that kind of thing being imposed on the industry by outside pressures, too.   

As such - being solely a matter of private contract - perfection need not be achieved to satisfy the apparent governmental and public need for "absolute, perfect, safety" in its regulations, actions, and approvals (as Railway Man has alluded to from time to time in various signal-related matters) - "good enough for now" would allow a big improvement to occur quickly, and avoid the valid concerns that you mention above.   Meanwhile, any FRA action would likely be ham-strung with studies, comments, proposed rules, comments, revisions, litigation, etc., etc.  Someone here - coborn35, as I recall - commented that he hoped the FRA wouldn't impose some kind of 'knee-jerk' Emergency Order in response to this.  As is often the case, the best way for those in the industry to keep that from happening is to act pro-actively and pre-empt such a response by self-governing and implementing self-imposed "do it yourself" solutions, so the FRA doesn't have to. 

- Paul North. 

Rather than try to construct an almost unyielding post to ward off the impact of a crash with a heavy object, perhaps constructing protective "noses" that collapse "accordion-style" to absorb the impact might work better?  That strategy is used successfully in some vehicles.

Good point - i.e.,  a 'featherbed slowing' vs. a 'brick wall' stop, etc.  Actually, a combination of the two approaches often works best, as long as the "collapse area" is not "occupant survival space".  The collapsing action: 1) slows down the shock of the sudden stop on the crew - no seatbelts, remember - so they don't have a 'crash test dummy' type 2nd collision with the interior of the cab; and 2) attenuates the impact load and spreads it out - both in duration, so the 'peak' is not as high - and over the multiple stiff and strong framing members that can then absorb and dissipate or resist that load and impact energy, etc. 

It's a much more complicated analysis - and hence to write a specification for - but again, do-able.  The FRA's crashworthiness specificatons against the 65,000 lb. objects at 30 MPH or 50 MPH, etc. essentially requires such a method; this would be similar, but for a much larger impact load.   

One problem is then to control where the crumpling occurs to disspiate and absorb that energy - i.e., we want that to occur someplace other than the occupied compartment.  Someone mentioned above that the early EMD cab units and Geeps supposedly had 'weak points' designed into their frames at a location just behind the cab, so that upon impact they would bend/ break there instead - there are a few 'broadside' angle photos around which show how that did occur, but I don't have any handy at the moment.  I looked into that a little bit after my distant in-law relative told me he was working on such a problem, but I was not persuaded that was in fact the design - some sources said it was essentially an 'urban legend', and others pointed out that every structure has a weakest point (or link) where it will fail first, and for those EMD units it just happened to be a little behind the cabs, due to how those frames were constructed.  I'll see if I can retireve any of that research, or links, documents, etc. - no promises, though.  Any help in that regard will be appreciated ! 

- Paul North. 

It will be interesting to see a reconstruction showing the individual events of this collision.  Obviously one or more cars rode up and over the lead locomotive.  Eventually, they fell off and landed alongside the train.  However, there is no proof that all of the cars (I count three) alongside the train actually rode up and over the locomotive.  One or two of them may have simply been displaced directly to the side by the force of the collision.

Looking at the photos, it looks like the fourth photo shows the first car that was displaced by the collision.  This may have been the last car of the work train, and it may have either rode up the engine, and fell off of the top; or it may have just been displaced sideways and tipped to the side in the clear.    

Particularly interesting are the two cars nearly side by side lying alongside the train.  If these rode up and over the locomotive, they must have telescoped by one overriding the other at some point prior to landing where they did.    If these two cars went over the engine one by one without first telescoping together, they would not be lying side by side on the ground.  Although, I suppose it is possible that one car rode up the engine, and the second car rode up over the first as the continued over the top of the engine.

Looking at the photos it is not possible to tell the order of events.  The cars have been shuffled like cards.  Just ahead of the locomotive, the flatcars are stacked three high.  It seems rather surprising that throughout the progression of this collision, the cars persisted in overriding each other and overriding the locomotive rather than jackknifing at some point. 

I would call this a flatcar telescope wreck.  Telescoping was the dreaded outcome of passenger train collisions in the pre-1900 era.  Collision impacts would simply cause one car to burst through the end of the next car, and run right into its interior.  Two cars suddenly occupy the space of one like the collapsing of a star watcher’s telescope.  Of course the people riding in the car that receives the intruding car are subject to the most unfathomable horrors.  Telescope wrecks became a thing of the past as safety innovations led the way into the 1900s.  

In the sense of these historic telescoping wrecks, flat cars cannot telescope.  One flat car cannot get inside of another.  But flatcars can overrun each other easier than any other form of rolling stock.

For instance - from a post by RdHseRat on Weds. July 28, 2004 5:40 PM at: http://www.railroad.net/forums/viewtopic.php?f=6&t=4299&start=15 

edblysard

The collision post did what they were designed to do, they lifted the rear car up and over the cab, but as was pointed out, it was not the rear car, but one several deep from the rear end which ended up shearing the cab...note the "deck of cards" position of the remaining flat cars, that much mass/weight pushing on the cab and the collision post simply caused one of the cars to penetrate the cab.

Had these been any other type of car, say a box car or covered hopper, even a tank car, the post would have shoved the car either to the side or over the cab.

Note in several photos the post stayed put, but the frame just ahead of the nose bent back.

Looking at the photos posted by edblysard , it seems as if the crash posts did what they were intended to do.( literally, to create a  'ram effect' over the cab). It would seem that their welds to the locomotive fram held long enough to cause an upward deflection of the over-ridfing cars.  The unknown seems (to me) to be the effect that the impact on the overhead bridge beams had in creating a downward force on the locomotive cab. I can only guess that the bridge beams did have an amplifying effect on the downward force. Looking at the drawbar on the engine, it seems to be pointing downward as if as was mentioned (I think it was Carl, a sort of 'pole-vault' effect to lift up the cars onto the engine. The unknown would seem to be, if the accident scene was directly underneath the bridge, or was the wreckage moved down the track to enable the clearance of the wreck(?).

Paul_D_North_Jr

One minor point of difference: It seems to me that the 300,000 lb. shear strength requirement for each post is far larger than the load that could reasonably be expected to be developed or needed as a result of any collision with a highway vehicle - say, an 80,000 lb. GVW truck, even allowing for 'impact' - and as compared to the 65,000 lb. objects in the Appendix E crashworthiness specifications.  Therefore, I believe that was an early attempt - the regulation appears to date from 1980 - to provide crew protection against collision with other full-size railroad equipment. 

Paul,

In thinking about this a little more, I agree that a collision with a highway vehicle is not often likely to produce 300,000 lbs. (or double that for the two posts) of force, so I can see why you might conclude that the posts are intended for collision with railroad equipment.  However, the 300,000 lb. figure is not the actual required resistance of the collision post.  It is only the required shear strength of the post’s connection to the frame.  

I am not sure what they have actually executed in the design of the posts to fill in the missing information of the specification.  I assume that they have met the 300,000 lb. spec. for shear at the base.  They may even have achieved around 300,000 lbs. of bend resistance right up to the top, even though the specs. do not address that aspect of the design.  Nevertheless, it may have seemed like a reasonable way for the builder to interpret the spirit of the regulatory specification for the posts.

If the posts are that strong, that would certainly seem like overkill for just automobiles, but probably not overkill for heavy trucks, especially trucks carrying pre-cast concrete components, structural steel, steel plate or coils, or blocks quarried stone, for instance.  On the other hand, I would say it is way less than what would be required to resist the force of a collision with railroad equipment.  The 300,000 lb. figure for shear strength at the base would be more than enough for collisions with locomotives or rolling stock.  But I think 300,000 lbs. of bend resistance four or five feet up on a free standing post would way less than what is needed. 

   A little off topic, but I'm just curious:   Is there a requirement for similar crash protection for control cars used in push-pull commuter service?

Paul and Bucyrus...

Was on a SD70M today, decided to nose around the nose if you will, and discovered a rather welcome thing.

The collision post on these guys are huge....located in the same place as on the GE, they are incorporated into the structure of the nose...from inside the nose door they appear to be the sides of the compartment itself, about 3 feet deep at the base.

Once you do a little looking, you realize the "walls" of the hall in the nose are solid steel, about 2 to 3 inches thick, as I said, about 3 feet deep and 6 feet tall.

The sides of the nose cone and the front of the nose are welded to the post, although slabs would be a better description.

The entire structure, slabs included is welded to the locomotive bed, and the post or slabs continue down and are welded to the frame.

To give a comparative example, you could cut 3 or 4 sets of the GE post out of one side of the posts in the EMD.

As the photo posted earlier in the thread of the UP locomotive involved in the Metro Link wreck shows, they can take a huge impact.

Didn't have my camera with me at the time or I would have got a few photos.

 will try to get some later.

Thanks much, Ed.    Verrryyy interesting !  It's good to have someone who can get on the inside and observe and report so clearly.  But I wonder what would happen if those same impacts to which the SD70M's above were subjected had hit a GE with those collision posts instead ??? 

Here are copies/ quotes from a couple of my posts yesterday to another thread on the recent BNSF collision - "Fatal Rear-end Collision Reported on BNSF", at: http://cs.trains.com/TRCCS/forums/t/190729.aspx?PageIndex=3 - which belong better here instead: 

Link to NTSB press release posted by beaulieu:  http://www.ntsb.gov/Pressrel/2011/110503.html 

Paul_D_North_Jr

  [snipped]  Notably, the press release also indicates that the NTSB is going to look into the crashworthiness of the locomotive as well:   

"Finally, the NTSB has formed a work group which will examine the crash performance of the leading locomotive where crew members were fatally injured."

Thanks for posting that link.   

- Paul North.

Paul_D_North_Jr

  Link to FRA document that I just found - "Locomotive Crashworthiness Final Rule" (31 pages, approx. 1.79 MB in size) at:

http://www.fra.dot.gov/Downloads/counsel/LocoCrashworthiness062806.pdf

There's an interesting discussion on pages 4 - 11 inclusive of this 'PDF" format version.  Notably, this sccenario apparently occurred in an August 23, 1996 collision at Phoenixville, PA [ConRail, though it does not seem to have been the subject of either a NTSB or FRA accident report or investigation - PDN], per "Table 1. - Collision Scenario, Collision Mode, and Accident Representative of Scenario" on page 6 of 31 of this PDF version - "3a. Overtaking collision, locomotive to flat car . . . loading of window frame structure.  However, footnote 5 on the same page states that while the Report contained "one [scenario in which] an overriding freight car impacts the window structure during a rear-end collision", and "The Working Group initially considered" that scenario, it was not used for the crashworthiness evaluation.  [emphasis added - PDN]  Stated another way - they didn't look at this possibility in detail.     

Bucyrus, and others who are still interested in this level of detail: There is also an discussion of collision post strength requirements from the bottom right corner of page 20 through the 1st column of page 22.  Apparently AAR Standard S-580-2005 requires each collision post to be able to resist 750,000 lbs. at the point of attachment to the underframe, and 500,000 lbs. at 30" above the frame.  It seems that the Working Group considered further increasing the required strength at the latter point to 800,000 lbs., but "The Working Group found it more desirable to have the collision posts fail before the underframe does, thereby reducing the possibility of override due to either the formation of a ramp caused by underframe deformation or catapulting."  [emphasis added - PDN]      

 In short, "they blew it" - twice, even.  Perhaps that's why the NTSB is going to look at this further. 

- Paul North. 

Bucyrus had some worthwhile further comments on this after my posts over on that other thread, but I want to find some other links before I address/ respond to them.

Thanks again to both of you. 

- Paul North.   

My observations are the the bridge became a wild card in the operation of the collision posts....the posts were doing what they were designed to until the bridge limited the upward movement of the flat cars and crushed them down on the cab structure.

Paul ; I don't have permission to see the photo. It is forbidden.  

                                                         Jim

Jim - What or which photo ?  I'm not understanding - everything I linked to is either a webpage or a document.  Which one is giving you that message ? 

- Paul North. 

edblysard

Paul and Bucyrus...

Was on a SD70M today, decided to nose around the nose if you will, and discovered a rather welcome thing.

The collision post on these guys are huge....located in the same place as on the GE, they are incorporated into the structure of the nose...from inside the nose door they appear to be the sides of the compartment itself, about 3 feet deep at the base.

Once you do a little looking, you realize the "walls" of the hall in the nose are solid steel, about 2 to 3 inches thick, as I said, about 3 feet deep and 6 feet tall.

The sides of the nose cone and the front of the nose are welded to the post, although slabs would be a better description.

The entire structure, slabs included is welded to the locomotive bed, and the post or slabs continue down and are welded to the frame.

To give a comparative example, you could cut 3 or 4 sets of the GE post out of one side of the posts in the EMD.

As the photo posted earlier in the thread of the UP locomotive involved in the Metro Link wreck shows, they can take a huge impact.

Didn't have my camera with me at the time or I would have got a few photos.

 will try to get some later.

Ed,

It sounds like there is a minimum required collsion post system specified by regulating agencies, but the locomotive builders are free to exceed the minimum.  From your information, it sounds like EMD has exceeded the minimum further than GE.  

A local railfan has confirmed the damage to the highway overpass which is now closed pending repairs.

A BNSF MOW employee has provided more information about the MOW train.

Counting from the back, and counting the Herzog machine as #1 and the Scorpion flat as #2. The cars piled up on the engine are #2, #7, #8, and #9.  So 4 cars and the Herzog machine were kicked out of line dissipating some of the energy. Part of the Scorpion cars ramp was torn off and is laying back near the bridge. BTW the Herzog machine was at the rear of the MOW train but can't be considered a flatcar, so the first impact was spread over at least a somewhat greater area.

Anymore news about this wreck? I have pondered greatly about this wreck and one can not rule out all possible causes. Crew fatigue very possible, briefly destructed going over maybe some paper work or one crew member was going through their work bag and the other might have been looking down at the controls trying to figure out something... Reason for no emergency brake application could have been SHOCK... There is just so many possible reasons why they never made an emergency brake application, why they never slowed down, why they never did this and that. I don't think the event recorder will help to tell the full story, like now it showed power was reduced, speed 23mph and no one hit the red lever.

I feel like something like a cab voice recorder is in order here, just like planes it will help at least to answer some questions to why, was the crew aware to begin with. As for the cabs they can only do so much but when the ultimate crash occurs... Just my 2 cents.